repository.edgehill.ac.uk Sanders 23103400... · Web viewPrimary Health Care Research &...

Get Healthy Get Active: Prevention is better than care

GEORGE JON SANDERS

A thesis submitted in partial fulfilment of the requirements of Edge Hill University for the

degree of Doctor of Philosophy.

July 2018

1

Statements

The presented research programme evaluated the effectiveness of Sport England’s

Get Healthy Get Active physical activity intervention on older adults’ physical activity

levels. The research presented within this thesis including project design, data

collection, and data analyses was funded by Sefton Metropolitan Borough Council

and was conducted solely at Edge Hill University.

2

Acknowledgements

This thesis is dedicated to my father, mother and brother whom have provided me

with the necessary life skills and mental fortitude to complete this thesis. Your love

and support motivates me more than I could ever describe. I couldn’t have done it

without you.

I would like to thank my Director of Studies, Professor Stuart Fairclough for his

continued support and guidance throughout my PhD. I would also like to thank

Professor Brenda Roe and Dr Axel Kaehne for their insightful advice, support and

feedback provided every step of the way. The knowledge and expertise from this

supervisory team has enabled me to continually learn, improve and progress

throughout this project. It has been a fantastic experience and this is in part down to

this incredible supervisory team.

Thank you also to the Active Lifestyles team at Sefton Metropolitan Borough Council

whom were instrumental in the research process and provided me with full access to

the Get Healthy Get Active sessions. Thank you for being infinitely patient with the

endless ideas and subsequent measures thrust upon participants throughout the

sessions. You kept me grounded throughout this project and completion of this

thesis would simply not have been possible without your full support.

3

A massive thank you also goes out to all the participants and Edge hill University

students who gave their time to participate in the project. I must also thank Dr Andy

Sparks, Dr Whitney Curry, Professor Zoe Knowles, and Dr Lynne Boddy for their

expertise and input throughout the project. Michael and Sarah, it was brilliant going

through this process with you both and I can only hope that future colleagues are as

patient as you were with the constant smell of coffee, chicken and rice that lingered

throughout our office for the past three years.

In conclusion, what a fantastic experience this has been!

4

Abstract

Low levels of physical activity (PA) and high levels of sedentary behaviour (SB) among

older adults, carry considerable burdens to physical (e.g., premature mortality,

chronic diseases and all-cause dementia risk) and psychosocial (e.g., self-perceived

quality of life, wellbeing and self-efficacy for exercise) health. Numerous large scale

interventions designed to engage physically inactive older adults have shown the

potential that interventions guided by theoretical frameworks, consider

implementation at scale across levels of the socioecological model and are designed,

implemented and delivered in close partnership with stakeholders can have among

this population. This thesis aimed to investigate the effectiveness of Sport England’s

Get Healthy Get Active (GHGA) PA intervention. GHGA was delivered by Sefton

Metropolitan Borough Council (SMBC) and was designed to engage inactive older

adults in PA for at least once a week for 30 minutes.

The purpose of Chapter 3 was to elicit subjective views of older adults about

perceived facilitators and barriers to PA participation and to inform the design,

delivery and recruitment strategies of Sport England’s GHGA PA intervention.

Analyses revealed time of day, cost and social support to be key predictors in

promoting PA. Sessions that avoid taking place in the early morning or late

afternoon, are free of charge, and promote social interaction were also significant

predictors of older adults’ PA participation. Wrist- and hip-based accelerometers are

now common in assessing PA in population-based studies, however no raw

acceleration cutpoints for moderate-to-vigorous PA (MVPA) and SB exist for older

5

adults. Chapter 4 was the first to test a laboratory-based protocol using activities

representative of older adults’ PA behaviours, to generate behaviourally valid,

population specific wrist- and hip-based raw acceleration cutpoints for SB and MVPA

in older adults. These cut-points were subsequently applied within Chapter 5, along

with self-reported measures of SB, PA and health indicators, to investigate time

spent in MVPA and SB, and measures of quality of life (QoL), self-rated health (SRH),

self-assessment of physical fitness (SAPF), and self-efficacy for exercise (SEE).

Chapter 6 addressed the main objective of the thesis which was to assess the

effectiveness of the GHGA PA intervention on PA levels among inactive older adults ≥

65 years of age. The GHGA PA intervention was effective in increasing QoL, SRH,

SAPF, and SEE scores over time after adjustment for covariates. However, the

intervention was ineffective in both reducing time spent in SB and increasing time

spent in MVPA. As a measure of intervention fidelity, Chapter 7 evaluated whether

the GHGA multi-component PA intervention was implemented as intended. Results

from both deliverer interviews and session observations revealed that a high degree

of intervention fidelity was maintained throughout the GHGA PA sessions within five

core domains including: Study Design, Provider Training, Intervention Delivery,

Intervention Receipt and Enactment.

This thesis contributes to the understanding of feasible and acceptable PA strategies

in older adults. Future research is needed to establish whole system-oriented multi-

component community-based interventions that are effective at increasing PA levels

in older adults.

6

Contents

Acknowledgements 3

Abstract 5

List of Tables 11

List of Figures 12

List of Abbreviations 13

Chapter 1. Introduction 15

1.1. The Research Problem 16

1.2. Conceptual Framework 19

1.3. Organisation of Thesis 23

1.4. Original contribution to knowledge 24

1.5. Aims and objectives 25

Chapter 2. Literature Review 29

2.1. Guidelines 30

2.2. The Health Benefits of Physical Activity Participation 33

2.3. Physical Activity Levels 36

2.4. Sedentary Behaviour 38

2.5. Physical Activity Measurement 43

2.6. Correlates of Physical Activity and Sedentary Behaviour 50

2.7. Community-Based Physical Activity Interventions 53

2.8. Community-Based Intervention Process Evaluation 57

2.9. Summary of literature 60

Thesis Study Map 62

7

Chapter 3 (Study 1). Using formative research with older adults to inform a

community physical activity programme: Get Healthy, Get Active. 63

3.1. Introduction 64

3.2. Methods 65

3.3. Findings and Discussion 70

3.4. Strengths and Limitations 82

3.5. Conclusions 84

Thesis Study Map 86

Chapter 4 (Study 2). Evaluation of wrist and hip sedentary behaviour and

moderate-to-vigorous physical activity raw acceleration cutpoints in older adults.

88


4.2. Methods 91

4.3. Results 98

4.4. Discussion 101

4.5. Conclusion 106

Thesis Study Map 107

Chapter 5 (Study 3). Physical activity, sedentary behaviour, perceived health and

fitness, and psychosocial wellbeing among community-dwelling older adults. 109


5.2. Methods 112

5.3. Results 121

5.4. Discussion 129


8

5.6. Conclusions 134


Chapter 6 (Study 4). A pragmatic evaluation of the Get Healthy Get Active physical

activity programme for community-dwelling older adults. 139


6.2. Methods 143

6.3. Results 153

6.4. Discussion 158




Chapter 7 (Study 5). Intervention fidelity of the Get Healthy Get Active physical

activity programme for community-dwelling older adults. 172


7.2. Methods 177

7.3. Results 185

7.4. Discussion 197




Chapter 8. Synthesis of Findings, Recommendations and Conclusions 210

8.1. Synthesis of Findings 211


8.3. Recommendations 224

9


References 230

Appendices 275

Appendix 1. Ethical Approval 276

Appendix 2. Accelerometer Instructions 290

Appendix 3. Associated Publications 293

10

List of Tables

Table 4.1. Description of the sixteen structured activities. 93

Table 4.2. Study sample characteristics. 99

Table 4.3. Mean (SD) accelerometer output from GA and AG (mg) during each activity performed by older adults.

99

Table 4.4. Calibration cutpoints and cross-validation % agreement, kappa (k) and se and sp. 101

Table 5.1. Descriptive characteristics of the participants. 122

Table 5.2. GENEActiv wrist-worn accelerometer data descriptives. 123

Table 5.3. Self-reported physical activity and psychosocial outcome measures. 125

Table 5.4. GENEActiv SB and physical activity outcomes. 127

Table 6.1. Exercise typical of a GHGA session. 145

Table 6.2. Descriptive baseline characteristics of the participants. 155

Table 6.3. Unadjusted self-reported physical activity and psychosocial outcome measures. 156

Table 6.4. Crude multilevel model analyses of the outcome measures at three, six and 12-months follow-up. 158

Table 6.5. Adjusted multilevel model analyses of the outcome measures at three, six and 12-months follow-up. 159

Table 6.6. Significant intervention subgroup interactions. 160

Table 7.1. Frequency counts and descriptives. 198

11

List of Figures

Figure 1.1. Precede-Proceed model of health programme design, implementation, and evaluation (Green & Kreuter, 2005).

21

Figure 3.1. Predisposing correlates of physical activity participation among older adults. n = Individual mentions per person (multiple mentions not included); Fn = Focus group number; Pn = Participant number. 73

Figure 3.2. Enabling correlates of physical activity participation among older adults. n = Individual mentions per person (multiple mentions not included); Fn = Focus group number; Pn = Participant number. 77

Figure 3.3. Reinforcing correlates of physical activity participation among older adults. n = Individual mentions per person (multiple mentions not included); Fn = Focus group number; Pn = Participant number.

80

Figure 6.1. Flow of participants through the study. 146

12

List of abbreviations

6MWT Six-minute Treadmill Walk Test

AG ActiGraph

ANCOVA Analysis of Covariance

BCC Behaviour Change Consortium

BMI Body mass index

CDC Centers for Disease Control and Prevention

CONSORT Consolidated Standards of Reporting Trials

EIT Evidence Integration Triangle

EE Energy Expenditure

ENMO Euclidean norm minus one

GA GENEActiv

GHGA Get Healthy Get Active

IMD Indices of Multiple Deprivation

LPA Light physical activity

MANCOVA Multivariate Analysis of Covariance

MET Metabolic equivalent

MPA Moderate physical activity

MVPA Moderate to vigorous physical activity

13

NIH National Institute of Health

NHANES National Health and Nutrition Examination Survey

NHS National Health Service

ONS Office for National Statistics

OR Odds Ratio

PA Physical activity

PAR-Q Physical Activity Readiness Questionnaire

RCT Randomised Controlled Trial

RMR Resting Metabolic Rate

RR Relative Risk

RTM Regression to the Mean

SB Sedentary behaviour

SD Standard deviation

SEF Standard Evaluation Framework

SES Socioeconomic status

US United States

UK United Kingdom

VPA Vigorous physical activity

WHO World Health Organisation

14

Chapter 1. Introduction

15

1.1. The Research Problem

The physical and psychological health benefits of PA are widely acknowledged

(Vahlberg, Cederholm, Lindmark, Zetterberg, & Hellström, 2017). PA is defined as

any bodily movement produced by skeletal muscles that results in energy

expenditure (EE) (Caspersen, Powell and Christenson, 1985). The term encompasses

exercise, sport, recreation, walking, active transport (e.g., cycling, running,

skateboarding), occupational activity, and domestic activity such as gardening and

cleaning (Caspersen et al., 1985). Research has explored health benefits in relation to

PA intensities including light-PA (LPA), moderate-PA (MPA), moderate-to-vigorous PA

(MVPA), and vigorous-PA (VPA) (Barone Gibbs et al., 2017; Biswas et al., 2015;

McPhee et al., 2016; Windle, Hughes, Linck, Russell, & Woods, 2010). Older adults

are said to typically engage in LPA (Ku, Fox, Liao, Sun, & Chen, 2016) consisting of

activities including carrying light objects, walking slowly, and housework (e.g.,

washing up and hoovering) (Public Health England, 2017). Opportunities for older

adults to be physically active exist in many different settings and contexts such as at

home, at recreation facilities, through active commuting, and within local community

spaces (Milligan et al., 2015; Gardiner, Geldenhuys and Gott, 2016).

The chronological age of 65 years is the accepted definition of an older person in the

United Kingdom (Age UK, 2018). Guidelines issued by the United Kingdom (UK) Chief

Medical Officers and the United States (US) Surgeon Generals recommend that older

adults (≥65 years) engage in at least 150 minutes of MPA (or 75 minutes of VPA) per

week in bouts of at least 10 minutes, with muscle-strengthening and balance

16

activities included on at least two of those days (Department of Health, 2011a;

Centers for Disease Control and Prevention (CDC), 2015). Despite overwhelming

evidence outlining the benefits of regular PA on both physical (Zhu et al., 2017) and

psychosocial (Devereux-Fitzgerald, Powell, Dewhurst, & French, 2016; Franco et al.,

2015; Greaney, Lees, Blissmer, Riebe, & Clark, 2016) determinants of health at older

ages (Lehne & Bolte, 2017), less than 12% of older adults globally perform PA on a

daily basis (CDC, 2016). Objective summaries of PA levels among older adults show

that only 15 per cent of males and ten percent of females within the UK, and 9.5% of

males and 7% of females within the US meet the recommended PA guidelines

(Tucker, Welk & Beyler, 2011; Jefferis et al., 2014). Large scale cohort studies have

shown that PA further declines with increasing age, among females, those of lower

socioeconomic status (SES), and among individuals with lower levels of perceived

health status and self-efficacy (Lehne & Bolte, 2017; Murtagh et al., 2015; Smith,

Gardner, Fisher, & Hamer, 2015). Given that current PA guidelines are the same for

both adults (18-64 years) and older adults (≥65 years), such high levels of inactivity

suggests that current PA guidelines may be too demanding for the latter population

(Booth & Hawley, 2015).

Accumulating evidence suggests that prolonged and continuous bouts of SB (defined

as waking behaviours in a sitting, reclining or lying posture with EE ≤1.5 metabolic

equivalents (MET) (Tremblay et al., 2017) have similar physical (e.g., premature

mortality, chronic diseases and all-cause dementia risk) and psychosocial (e.g., self-

perceived QoL, wellbeing and SEE) risk factors to those associated with physical

17

inactivity (Wilmot et al., 2012; Edwards & Loprinzi, 2016; Falck, Davis & Liu-Ambrose,

2016; Kim, Im & Choi, 2016). In fact, SB is now an identifiable risk factor independent

of other PA behaviours (Tremblay et al., 2017). Spending on average 80% of their

time in a seated posture, and with 67% sedentary for more than eight and a half

hours per day (Shaw et al., 2017), older adults are the most sedentary segment of

society and seldom engage in MVPA (Chastin et al., 2017).

PA is a complex behaviour influenced by various individual and environmental

factors (Devereux-Fitzgerald et al., 2016; Phoenix & Tulle, 2017). Identification of

modifiable correlates and a comprehensive understanding of the influence of these

factors over time on older adults’ PA are imperative in enabling policymakers and

healthcare professionals to develop and implement successful interventions

(Banerjee et al., 2015; Devereux-Fitzgerald et al., 2016; Greaney et al., 2016).

Intervention research in the field of PA in this population has primarily focused on

pre- and post-intervention measurements and less on longer term follow-up

measurements after intervention completion (McMahon et al., 2017). Follow-up

measures post-intervention are critical for understanding implementation

sustainability and maintenance patterns (McMahon et al., 2017). To improve

population health, efficacious PA interventions in controlled research settings must

be scaled up to reach broader populations across multiple settings (Milat et al.,

2016). A recent PA intervention (Choose to Move) designed to engage physically

inactive (e.g., not meeting current PA guidelines) older adults has shown the

potential for interventions that are designed, implemented and delivered in close

18

partnership with stakeholders and across multiple settings (McKay, Sims-Gould,

Nettlefold, Hoy, & Bauman, 2017).

1.2. Conceptual Framework

As previously discussed, PA is complex and many PA correlates exist among older

adults (Devereux-Fitzgerald et al., 2016; Phoenix & Tulle, 2017). Consequently,

behaviour change is complex to achieve and maintain. Theoretical frameworks are

vital in explaining and predicting health behaviour, and theory-based interventions

are more efficacious than atheoretical approaches at changing many health

behaviours, including PA (Plotnikoff et al., 2014). According to Michie and colleagues

(2007), theories can specify causal relations between potential correlates and proffer

implications for designing interventions to promote health. Previous literature now

includes many cross-sectional associations and longitudinal relationships between

demographic, biological, psychological, social environmental, and physical

environmental variables (commonly referred to as correlates) (Bauman et al., 2002;

Bauman et al., 2012; Bryan et al., 2007; Trost et al., 2002). Such findings have

emphasised the fact that regular PA yields numerous health benefits among all age

groups (Haskell et al., 2007). However, the current evidence base examining the

effects of PA among older adults is inconsistent and at times, contradictory

(Livingston et al., 2014) and thus, research should move beyond individual factors

and instead adopt multilevel socio-ecological models of health behaviour change

(Plotnikoff et al., 2014). Given that various factors can be influential to

19

implementation, the use of multi-level socio-ecological frameworks to design PA

promotion strategies are warranted (Plotnikoff et al., 2014).

Within health behaviour and promotion research the use of a socio-ecological model

as a theoretical underpinning has been widely adopted since it was originally

proposed (McLeroy, Bibeau, Steckler, & Glanz, 1988; Simplican, Leader, Kosciulek, &

Leahy, 2015). Levels of influence within the socio-ecological model include the

intrapersonal, interpersonal, organisational, community, environmental, and policy

levels of behaviour, whilst incorporating physical and psychological influences in an

attempt to better predict PA behaviour and ensure a supportive social and

community environment (Kerr et al., 2012; Sallis, Owen & Fisher, 2008). Indeed, this

combination of psychosocial and environmental factors is significantly related to

older adults’ PA (Carlson et al., 2012). A review of the literature indicates that

interventions targeting PA determinants at different levels of the socio-ecological

model, including the social and organisational/built environment levels, have the

highest potential to increase overall PA in older adults (Plotnikoff et al., 2014). For

encouraging older adults to be physically active and less sedentary, socio-ecological

models such as the PRECEDE-PROCEED model of health programme design,

implementation, and evaluation (Green & Kreuter, 2005; Figure 1.1), are well suited

for studying PA because participation occurs in specific locations such as leisure

centres and church halls. Ecological models direct attention towards the

characteristics of locations including the broader political and environmental factors

which either facilitate or hinder participation (Sallis et al., 2006). Framing an

20

intervention within a socio-ecological model has been favourably received previously

(Haggis et al., 2013; Kerr et al., 2012) and consequently, such a model is adopted in

the current thesis in order to ascertain the predictive power of a PA intervention in

positively affecting a broad range of intrapersonal, interpersonal, organisational,

community, environmental, and policy level contexts of PA behaviour among older

adults.

Figure 1.1. Precede-Proceed model of health programme design, implementation, and evaluation (Green & Kreuter, 2005).

An approach that holds relevance for PA intervention design, and promises further

understanding of the processes leading to sustained motivation and optimal

functioning/well-being in promoting PA is the PRECEDE-PROCEED model of health

programme design, implementation, and evaluation (Green & Kreuter, 2005). This

model provides a comprehensive structured assessment of health and health needs,

through the design and implementation of health promotion programmes to meet

21

those emerging needs. The PRECEDE-PROCEED model has been considered to be

among the 10 best planning models on usefulness for research and practice (Linnan

et al., 2005) and could therefore potentially increase the sustainability of a PA

intervention among older adults. The PRECEDE-PROCEED model (Green & Kreuter,

2005) is a social-ecological framework which allows enabling, predisposing and

reinforcing factors associated with PA to be acknowledged. The model systematically

considers the social and situational circumstances of a target group, relevant

epidemiological data, environmental and behavioural (lifestyle) factors, and factors

that influence these behaviours and the environments in which they occur (Green &

Kreuter, 2005). The PRECEDE phase represents the process that precedes the

intervention and is an acronym for predisposing, reinforcing and enabling constructs

(Atun et al., 2010; Yeo et al., 2007; Yuan et al., 2010). Predisposing factors are the

motivators for PA participation and include the knowledge, attitudes and beliefs that

motivate behaviour prior or during the intervention (Yuan et al., 2010). Enabling

factors include skill development and access to resources that facilitate change and

participation in the intervention (Yuan et al., 2010). Reinforcing factors are the

positive and negative factors that result as a consequence of behaviour change and

includes the social support and rewards and/or incentives for the PA behaviour

(Atun et al., 2010; Yuan et al., 2010). Manipulation of any one of these factors has

been found to result in behaviour change that is sustainable over time (Green &

Kreuter, 1999; Yuan et al., 2010). The PROCEED (Policy, Regulatory and

Organisational Constructs in Educational and Environmental Development) phase

aids in the implementation and evaluation of programmes. The last step

accommodates intervention planning based on available resources and potential

22

barriers. There are nine key phases in the model, five for assessment, one for

implementation, and three related to evaluation. The PRECEDE-PROCEED model

allows for participation of participants in the evaluation process so that they can

determine their behaviour and health outcomes by voluntary active involvement

(Green, Kreuter, Deeds, & Partidge, 1980). By involving the target population to

assess their own needs and barriers, the participants' compliance to a tailored

intervention programme is more likely to be successful and sustainable (Cole &

Horacek, 2009; Lean, Lara & Hill, 2007). Predisposing, enabling and reinforcing

factors will be based on relevant components of the ecological model of behaviour

change (Stokols, 1992). The model has been used extensively in health promotion

planning and evaluation among older adults (Banerjee et al., 2015; Gagliardi,

Faulkner, Ciliska, & Hicks, 2015; Jancey et al., 2008) and other populations

(Makintosh et al., 2011; Emdadi, Hazavehie, Soltanian, Bashirian, & Heidari

Moghadam, 2015; Susan, Mallan, Callaway, Daniels, & Nicholson, 2017) and thus,

was appropriate for adoption.

1.3. Organisation of Thesis

The central theme of the thesis is increasing PA levels of inactive community-

dwelling older adults aged ≥65 years. A review of the literature is provided in

Chapter 2. The key topics addressed are the impact of PA and SB on health, older

adults’ PA and SB levels, correlates of PA and SB, measurement of PA and SB, and

the effects of interventions on older adults’ PA levels and SB. The review critiques

the current literature, and highlights gaps, which provide a rationale for the current

23

research. Chapter 3 presents a formative study: Using formative research with older

adults to inform a community PA programme: Get Healthy, Get Active. Wrist-based

accelerometers are now common in assessing PA in population-based studies, but no

raw acceleration cutpoints for moderate-to-vigorous PA and SB exist for older adults.

This issue is addressed in Chapter 4. Older adults’ PA, SB and health indicators prior

to exposure to the GHGA intervention are reported in Chapter 5. Chapter 6

evaluated the impact of the GHGA intervention on older adults’ PA levels, SB, and

health indicators. As a measure of intervention fidelity, Chapter 7 evaluated whether

the GHGA multi-component PA intervention was implemented as intended. The final

chapter (Chapter 8) synthesises the key findings of the thesis and discusses the

overall strengths and limitations of the research programme. Recommendations for

future research and practice and conclusions are then presented in the final section

of Chapter 8.

1.4. Original contribution to knowledge

Original contributions to knowledge will be made through the design,

implementation and evaluation of a bespoke PA intervention aimed at increasing PA

among inactive community-dwelling older adults. Through the adoption of focus

groups, Chapter 3 will provide valuable insights into current knowledge and attitudes

towards PA, as well as perceived barriers, facilitators and opportunities for PA

participation among older adults living in the community. Chapter 4 will be the first

to test a laboratory-based protocol using intermittent activities representative of

older adults’ PA behaviours to generate behaviourally valid, population specific

24

wrist- and hip-based raw acceleration cutpoints for SB and MVPA in older adults.

These novel cut-points will be applied within Chapter 5, along with self-reported

measures of SB, PA and health indicators, to investigate time spent in MVPA and SB,

and measures of QoL, SRH, SAPF, and SEE. Chapter 6 will assess the effectiveness of

a PA intervention in increasing inactive community-dwelling older adults’ PA levels

and psychosocial health statuses through self-reported measures. Chapter 7 will add

to the limited intervention fidelity literature by evaluating whether a PA intervention

was implemented as intended. Decisions throughout the programme of work will be

informed by the PRECEDE-PROCEED model of health programme design,

implementation, and evaluation (Green & Kreuter, 2005) in order to ascertain the

predictive power of a PA intervention in positively affecting a broad range of

intrapersonal, interpersonal, organisational, community, environmental, and policy

level contexts of PA behaviour among older adults.

1.5. Aims and Objectives

The main aim of the thesis is to assess the effectiveness and implementation of Sport

England’s Get Healthy Get Active (GHGA) intervention on inactive community-

dwelling older adults’ PA levels. A detailed description of the intervention can be

found in Chapter 6, but a brief summary is provided here. The GHGA intervention

was aimed at engaging inactive community-dwelling older adults in PA at least once

a week for 30 minutes, via a PA intervention. The project was funded by Sport

England and delivered by SMBC. Findings from a comprehensive meta-analysis

suggest that interventions designed to increase PA behaviour among adults aged 65

25

and older can be effective (Chase, 2015). Increased effectiveness has been observed

in interventions which are guided by theoretical frameworks, consider

implementation at scale across levels of the socioecological model and are designed,

implemented and delivered in close partnership with stakeholders (Harris et al.,

2015; Harris et al., 2018; Sink et al., 2015). Resultantly, if proven to be effective, it is

proposed that the GHGA PA intervention will lead to bigger, sustainable, national

level research projects whose results would have policy ramifications and inform the

thought and practice of professionals in PA, social work and care settings.

Five studies were conducted to address the following objectives:

Study 1 objectives.

1. To explore current knowledge and attitudes towards physical activity, as well

as perceived barriers, facilitators and opportunities for physical activity

participation among older adults living in the community.

2. To use these data to subsequently inform the design, delivery and

recruitment strategies of Sport England’s national GHGA initiative.

Study 2 objectives.

3. To test a laboratory-based protocol to generate behaviourally valid,

population specific wrist- and hip-based raw acceleration cutpoints for

sedentary behaviour and moderate-to-vigorous physical activity in older

adults.

26

4. To apply these cut-points to subsequently analyse physical activity data for

Sport England’s GHGA physical activity intervention.

Study 3 objectives.

5. To investigate gender, age, and socio-economic status differences in older

adults’ sedentary behaviour, physical activity and self-reported health

indicators.

6. To examine associations between sedentary behaviour and physical activity

with self-reported health indicators.

Study 4 objective.

7. To evaluate the effectiveness of Sport England’s GHGA physical activity

intervention on older adults physical activity, sedentary behaviour and self-

reported health indicators.

Study 5 objectives.

8. To evaluate whether the GHGA multi-component intervention was

implemented as intended.

9. To evaluate sustainability of the GHGA multi-component intervention in

terms of its feasibility and acceptability of being implemented and

incorporated in the long-term.

27

These objectives are aligned to the following Research Questions:

1. What are the current knowledge and attitudes towards PA among older adults

living in the community, as well as perceived barriers, facilitators and opportunities

for PA participation? (Objective 1).

2. What are the most appropriate wrist- and hip-worn raw acceleration cutpoints for

SB and MVPA activity in the GHGA sample of older adults? (Objective 3).

3. Are there any gender, age, and socio-economic status differences in older adults’

SB, PA and self-reported health indicators? (Objective 5).

4. What are the associations between SB and PA with self-reported health

indicators? (Objective 6).

5. Is Sport England’s GHGA PA intervention effective in increasing community-

dwelling older adults PA levels? (Objective 7).

6. Was the GHGA PA intervention implemented as intended? (Objective 8).

28

Chapter 2. Literature Review

29

2.1. Guidelines

2.1.1. Physical Activity Guidelines

PA can be classified and measured by the intensity continuum of effort required,

ranging from LPA, MPA and VPA (Butte, Ekelund, & Westerterp, 2012). A fact sheet

incorporating information from the four home countries (Department of Health in

England, the Scottish Office, the Welsh Office and the Department of Health in

Northern Ireland) provides information on experiencing PA at differing intensities,

and associated example activities for each intensity for older adults ≥65 years

(Department of Health, 2011a). LPA includes activities that take little effort and

cause older adults to breathe a little harder than normal such as carrying light things,

walking slowly and housework (e.g., washing up and hoovering). Moderate intensity

activities will cause older adults to breathe harder and their hearts to beat faster,

but they should still be able to carry on a conversation. Examples of moderate

intensity activities include brisk walking, climbing stairs and gardening. Whilst the

effects of vigorous activities are similar to those of moderate activities, these

30

activities will cause older adults to breathe much harder and their hearts to beat

rapidly, making it more difficult to carry on a conversation. Examples of vigorous

intensity activities include heavy lifting, aerobics and fast cycling (Department of

Health, 2011a). Activities to strengthen muscles and balance include body weight

exercises or activities such as dancing, chair aerobics, Tai Chi, and Yoga (Department

of Health, 2011a).

Combating physical inactivity requires an emphasis on encouraging older adults to

achieve PA guidelines (Edwards & Loprinzi, 2016). These are to engage in at least 150

minutes of MPA (or 75 minutes of VPA) per week in bouts of at least 10 minutes,

with muscle-strengthening and balance activities included on at least two of those

days (CDC, 2015; Department of Health, 2011a; Lillo, Palomo-Vélez, Fuentes, &

Palomo, 2015; Wullems, Verschueren, Degens, Morse, & Onambélé, 2016).

However, recent research notes that the same physical and psychosocial health

benefits associated with 10 minute bouts of MVPA can be achieved through total

accumulated 1 second bouts of MVPA in older adults (Jefferis et al., 2016; Sparling,

Howard, Dunstan, & Owen, 2015). Consequently, revised PA guidelines soon to be

published in the US (Office of Disease Prevention and Health Promotion, 2018) now

recognise that any amount of time spent in MVPA counts toward meeting PA

recommendations. It is also acknowledged that some older adults might not be

capable of meeting these recommendations due to poor functional ability or health.

For these older adults, it is recommended that they should complete as much PA as

they can do. In other words, even though they might not meet current PA guidelines,

31

there are still health benefits related to PA at lower levels (Warburton & Bredin,

2016). In fact, it has been suggested that two sessions per week of light-to-

moderate intensity PA each of a minimum of 45 minutes duration are optimal for

improving self-reported physical and psychosocial outcomes in older adults (Windle

et al., 2010). Although less than the recommended PA guidelines, fewer sessions of a

lower intensity are more realistic for encouraging long-term adherence to PA in older

adults regardless of gender, age and SES-group status (Kuosmanen et al., 2016). In

addition, older adults with poor mobility are advised to conduct PA that will improve

their balance and prevent falls on at least 3 days a week (World Health Organisation

(WHO), 2011). Chair-based PA interventions have been shown to be effective at

increasing balance (Lewis, Peiris & Shields, 2017), mobility (Kato, Islam, Koizumi,

Rogers, & Takeshima, 2018; Oestergaard et al., 2018) and reducing falls incidence in

frail older adults (Furtado et al., 2016). PA guidelines are now well established across

all age-groups globally (Hallal, Andersen, Bull, Guthold, Haskell, & Ekelund, 2012).

2.1.2. Sedentary Behaviour Guidelines

Compared to the research area of PA, research on SB is a relatively new scientific

field and consequently, countries on a global scale have only just started to provide

recommendations on SB for health, either by incorporating them into their PA

guidelines or by issuing specific SB guidelines (Leitzmann, Jochem & Schmid, 2017).

Existing SB recommendations mainly target children and young people with SB

guidelines for adults and older adults either absent (e.g., Australia, Austria, Belgium,

Canada, Ireland and the US) or extremely vague and simply focusing on reducing

32

prolonged periods of SB (e.g., Germany, New Zealand, Japan, and the UK). Prolonged

periods of SB are defined as ≥8 hours/day (Copeland, Clarke & Dogra, 2015). UK SB

guidelines state that older adults should minimise the amount of time spent being

sedentary (sitting) for extended periods (Department of Health, 2011b). The high

prevalence of SB and its public health significance in older adults warrants further

research in order to obtain more specific national and international

recommendations on SB for public health in this population (Leitzmann et al., 2017).

2.2. The Health Benefits of Physical Activity Participation

Performing sufficient PA is a primary modifiable determinant of health (Birkel et al.,

2015) and recent research has identified PA to be an integral contributor to a healthy

lifestyle which can provide both short- and long-term health benefits (Vahlberg et

al., 2017). Specifically, PA has the potential to benefit an array of physical (Zhu et al.,

2017) and psychosocial (Devereux-Fitzgerald et al., 2016; Franco et al., 2015;

Greaney et al., 2016) determinants of health in older adults. Favourable relationships

are evident between PA and health indicators including incident cardiovascular

disease (e.g., coronary heart disease, heart disease and stroke) (Li & Siegrist, 2012;

Lim et al., 2017), hypertension (Shaltout et al., 2017), osteoporotic fractures (de Kam

et al., 2009), depression (Potter, Ellard, Rees, & Thorogood, 2011), QoL (Potter et al.,

2011), and wellbeing (Wu et al., 2015). Frailty is a common condition among the

older population (Landi et al., 2010) and is described as a biological status in which

resistance to stressors including the immune, endocrine, musculoskeletal and

33

nervous system are reduced (Walston et al., 2006). Frailty leads to a state of high

vulnerability to adverse health outcomes and is associated with worsening of

physical functioning and falls, and higher rates of admissions to hospital, co-

morbidity and mortality (Landi et al., 2010). The relationship between PA, MVPA,

and frailty has been well explored (Mañas, del Pozo-Cruz, García-García, Guadalupe-

Grau, & Ara, 2018) and a negative association between MVPA and frailty among

older adults is now well established (Blodgett, Theou, Kirkland, Andreou, &

Rockwood, 2015; Peterson et al., 2009).

There is also evidence to suggest that among older adults PA can slow down the

progression of cognitive impairment (Sofi et al., 2011) and reduce symptoms of

dementia, delay its progression, and even prevent its occurrence (Forbes et al., 2015;

Lautenschlager, Cox and Kurz, 2010; Middleton & Yaffe, 2009). A recent meta-

analysis by Blondell, Hammersley-Mather & Veerman (2014) found significant

negative associations between PA and both cognitive decline and dementia (overall

effects of relative risk (RR) 0.65, 95% CI 0.55-0.76 and RR 0.86, 95% CI 0.76-0.97,

respectively). Results also showed that higher levels of PA (more than one 30 minute

session of MPA per week), versus lower levels of PA, were associated with a 14%

reduction in the risk of dementia (RR 0.86, 95% CI 0.76-0.97). Further support for

such findings is provided by Elwood et al. (2013) who reported PA to be a greater

predictor of cognitive impairment (Odds Ratio (OR) 0.64 95% CI 0.41, 0.92; P<0.04)

and dementia (OR 0.41 95% CI 0.22, 0.77; P<0.005), than for any other identified

cardiovascular risk or lifestyle factor including body mass index, eating fruits and

34

vegetables, smoking, and alcohol consumption. In fact, it has been estimated that 3

million cases of dementia could be averted globally, with a 10-25% shift in modifiable

risk factors including; cardiovascular risk factors such as hypertension, diabetes, the

metabolic syndrome, obesity and smoking (Cyarto et al., 2012; Molinuevo, Valls-

Pedret & Rami, 2010; WHO, 2012), and lifestyle factors such as PA (Erickson,

Weinstein & Lopez, 2012; Plassman et al., 2010). Such findings suggest a significant

and consistent protection for all levels of PA against cognitive decline in older adults

(Sofi et al., 2011). This is important given that the prevalence and financial

implications of dementia are such that small reductions in cognitive decline may

have a large impact on healthcare costs and overall individual burden (Forbes et al.,

2015).

Given the recommended PA guidelines much research has concentrated on the

effects of MVPA (Barone Gibbs et al., 2017). Specifically looking at studies conducted

in the UK, objectively-assessed MVPA has been related to various health indicators in

older adults, with low levels of MVPA associated with a greater likelihood of a

diagnosis of chronic illnesses and all-cause mortality (Fox et al., 2015), poorer

physical well-being (Withall et al., 2014) and number of falls (Barone Gibbs et al.,

2017; Simmonds et al., 2014). Lower levels of MVPA are also associated with

decreased SRH (Kuosmanen et al., 2016; Ramires et al., 2017), QoL (Lok, Lok &

Canbaz, 2017) and SEE (Dionigi, 2007; French et al., 2015).

35

PA guidelines relate to MVPA and most evidence shows positive associations

between MVPA and health (Mañas et al., 2018). There has also been an emergence

of interest in the health benefits of LPA, owing to the development of accelerometry

techniques in epidemiological studies (Lee & Shiroma, 2014; Shephard & Tudor-

Locke, 2016). Even though several studies have confirmed the potential health

benefits of LPA (van Baal, Hoogendoorn and Fischer, 2016; McMahon et al., 2017),

studies have often been cross-sectional and based upon self-reported LPA

(Autenrieth et al., 2011; Huerta et al., 2016). Objective assessment can record more

detailed and accurate patterns of personal daily activity (Jefferis et al., 2016;

Shephard & Tudor-Locke, 2016). A recent systematic review by Amagasa et al. (2018)

assessed whether objectively measured LPA was associated with health outcomes

after adjustment for MVPA. Results from 24 cross-sectional and 6 longitudinal

studies revealed that LPA was inversely associated with all-cause mortality risk and

associated favourably with the cardiometabolic risk factors of waist circumference,

triglyceride levels, insulin, and presence of metabolic syndrome. Some evidence of

the benefits of LPA on cognitive function (Johnson et al., 2016) and psychosocial

well-being has also been reported (Thraen-Borowski, Trentham-Dietz, Edwards,

Koltyn, & Colbert, 2013). These findings are important considering that many older

adults actually prefer LPA over MVPA, as LPA may be more achievable and

appropriate for this age-group (McMahon et al. 2017). Although current global PA

guidelines recommend only MVPA, promoting LPA may confer additional health

benefits and consequently, further longitudinal randomised controlled trials (RCTs)

are required to establish causality between LPA and physical and psychosocial health

outcomes.

36

2.3. Physical Activity Levels

Despite the recognised evidence of health benefits associated with regular PA,

objective measures show that compliance with the recommended PA guidelines is

poor in the general population and even worse in older adults (Troiano et al., 2008).

Results across 122 countries revealed that overall prevalence of adults reporting

three or more days per week of MVPA was only 31.4% (Hallal et al., 2012). Marked

differences were detected across regions including Africa (38.0%), Americas (24.6%),

Eastern Mediterranean (43.2%), Europe (25.4%), South-East Asia (43.2%), and

Western Pacific (35.3%) (Hallal et al., 2012). Objectively assessed levels of MVPA

have also been observed in older adults on a global scale with findings indicating

even poorer results. In the US, results from a large scale study among 3459 older

adults revealed that only 2.5% of participants met PA guidelines (Berkemeyer et al.,

2016). Similarly, a study of American older adults by Loprinzi (2013) found

participants engaged in objectively-assessed MVPA for only 10.0 minutes/day. A

study of 971 Brazilian older adults by Ramires et al. (2017) reported that men and

women engaged in only 40.5 and 22.5 minutes/week of MVPA in ≥10 minute bouts

(Ramires et al., 2017). A Swedish study by Hagströmer, Troiano, Sjöström, and

Berrigan (2010) reported that among 217 older adults participants spent 130

minutes/week engaged in MVPA. However, results are confounded by differing

methods of PA measurement (e.g., differing montiors such as the GENEActiv (GA)

and ActiGraph (AG) GT3X+ and GT9X) and monitor placement (e.g., wrist- versus hip-

worn accelerometer placement).

37

A national survey by the National Health Service (NHS) (2015) showed that in the UK,

only 20% of men and 17% of women aged 65 to 74 years meet recommended PA

guidelines. This contrasts with 49 % of men and 35 % of women aged 25 to 34 years

(NHS, 2015). More recent UK-based studies objectively assessing PA in older adults

have reported similar findings (Harris et al., 2018; Withall et al., 2014). Withall et al.

(2014) reported that older adults spent 93 minutes of their waking hours/week

engaged in MVPA. Similarly, Harris et al. (2018) found older adults engaged in 94

minutes/week of MVPA in ≥10 minute bouts. It is widely acknowledged that

increasing age leads to decreased time spent in PA (Arnardottir et al., 2013;

Berkemeyer et al., 2016; Harvey, Chastin & Skelton, 2014; Martinez-Gomez et al.,

2017; Wullems et al., 2016). When comparing those between 66-69 years old with

those aged 80 years and older engagement in MVPA decreased from 16.2

minutes/day to 10.7 minutes/day (Buman et al., 2010). In fact, engagement in MVPA

has been found to steadily decrease after retirement age (Martinez-Gomez et al.,

2017; Strain et al., 2016). Gender differences with time spent in LPA have also been

reported (Amagasa et al., 2017; Ramires et al., 2017). Ramires et al. (2017) reported

older men to spend an average of 127.6 minutes/day enaged in objectively

measured total accumulated LPA, whilst older women spent 136.2 minutes/day

(Ramires et al., 2017). Amagasa et al. (2017) reported even greater gender

differences with older males enagaging in an average of 263.1 min/day of objectively

measured total accumulated LPA, whilst females accumulated 365.3 minutes/day.

Time spent in LPA has also been shown to vary dependent upon age, with males and

females aged 80 years or more spending on average 45 and 65 minutes/day less in

38

objectively measured total accumulated LPA when compared to those aged 60 to 69

years old, respectively (Ramires et al., 2017). Overall, older adults on a global scale

are not meeting current PA guidelines and thus, it is important to understand which

factors are affecting PA participation so that successful interventions can be

developed and implemented.

2.4. Sedentary Behaviour

2.4.1. Prevalence of Sedentary Behaviour

Successful aging is a big concern in western societies. Globally, the older adult

population has dramatically increased worldwide in the last two decades, and it is

estimated that by 2015 the older population will represent approximately 22% of the

world’s population (Scully, 2012). In the last decade, SB has emerged as a new risk

factor for health (Chastin et al., 2017; Mañas et al., 2017). SB is characterised as any

waking behaviours in a sitting, reclining or lying posture with EE ≤1.5 METs (Tremblay

et al., 2017). Typical SBs among older adults are television viewing, reading and

sitting time (Pate, O'neill & Lobelo, 2008). Compared with other age groups, older

adults are the most sedentary segment of society (Chastin et al., 2017; Shaw et al.,

2017). Findings from studies in the US and Europe have reported that older adults

spend approximately 80% of their awake time engaged in SB which represents eight

to 12 hours/day (Chastin et al., 2017). Similarly, Hallal et al. (2012) conducted a

global assessment in more than 60 countries and found that the elderly had the

highest prevalence of reporting a minimum of four hours of sitting time daily.

Similarly, Davis et al. (2014) found a substantial amount of time in objectively-

39

assessed SB (71.3%) compared to other objectively-assessed PA categories (e.g., LPA:

9.0%, MVPA: 1.5%) in older adults. When objective data from a number of studies

are weighted and pooled, older adults spend a mean of 9.4 hours/day (ranging from

8.5 to 10.7 hours/day) engaged in SB (Harvey et al., 2014). From the available

studies, UK and US older adults record the highest levels of SB at approximately 11

hours/day (Bann et al., 2015; Hamer, Kivimaki & Steptoe, 2012; Sartini et al., 2015;

Withall et al., 2014).

2.4.2. Sedentary Behaviour Health Outcomes

Given the high prevalence of time spent in SB among older adults, identifying health

outcomes of objectively assessed SB in this population seems to be crucial in the

promotion of successful aging (Mañas et al., 2017). SB is an identifiable risk factor

affecting physical (e.g., premature mortality, chronic diseases and all-cause dementia

risk) and psychosocial (e.g., self-perceived QoL, wellbeing and self-efficacy)

determinants of health (Edwards & Loprinzi, 2016; Falck et al., 2016; Lewis et al.,

2017) independent of PA (Tremblay et al., 2017). Older adults who report sitting less

tend to age more successfully, report better QoL, have less dizziness, and have better

balance (Balboa-Castillo, Leon-Munoz, Graciani, Rodriguez-Artalejo, & Guallar-

Castillon, 2011; Dogra & Stathokostas, 2012; Van Uffelen et al., 2012). A recent

systematic review of objectively measured SB reported associations with health

outcomes relating to physical performance, frailty and mortality (Mañas et al., 2017).

Negative associations between SB and physical performance, regardless of MVPA

has also been reported (Rosenberg et al., 2015). Likewise, Fleig et al., (2016) and

40

Cooper, Simmons, Kuh, Brage, & Cooper (2015) found a negative association

between time spent in SB and the Timed Up and Go physical performance test

(Ikezoe, Asakawa, Shima, Kishibuchi, & Ichihashi, 2013; Podsiadlo & Richardson,

1991). Similarly, a recent study by Rosenberg et al. (2015) showed that objectively

measured SB is associated with worse physical function measured using the Short

Physical Performance Battery (Guralnik et al., 1994), balance task scores, 400 m walk

time, chair stand time, and gait speed.

Increased time spent in objectively measured SB has also been negatively associated

with physical function and frailty in older adults, regardless of participation in MVPA

(Gennuso, Thraen-Borowski, Gangnon, & Colbert, 2016; Song et al., 2015). Despite

all the potential benefits of PA in relation to frailty, frail older adults spend up to 10

hours (84.9%) of their daily time engaged in SBs (Jansen et al., 2015). Evidence

indicates that physically inactive individuals who have lower levels of functional

disability (Tremblay, Kho, Tricco, & Duggan, 2010), and those individuals who have

high levels of SB are more likely to be frail (Peterson et al., 2009). A British cohort

study by Cooper et al. (2015) found that even in young old age (60–64 years), time

spent sedentary is associated with frailty, lower grip strength and lower timed up

and go speed. Examination of large health survey data and objective monitoring

suggests those most sedentary older adults have higher levels of frailty, high activity

of daily living disability, and have higher healthcare usage (Blodgett et al., 2015).

Findings have also reported that objectively measured time spent in SB, after

controlling for MVPA, is related to metabolic syndrome (Bankoski et al., 2011),

41

cancer (Lynch et al., 2011) and mortality (Koster et al., 2012). It is also worth

emphasising that, even when people are physically active, prolonged sedentary

periods can still have a negative impact on health (Biswas et al., 2015). These

findings highlight the need to separate SB from insufficient MVPA patterns and for SB

to be identified as a modifiable risk factor independent of PA (Mañas et al., 2017).

Heightened amounts of time engaged in SB is also an independent risk factor for

psychosocial health (Biswas et al., 2015a; de Rezende, Rey-López, Matsudo, & do

Carmo Luiz, 2014; Withall et al., 2014). Psychosocial health includes psychological

and social psychological outcomes, interlinked with socioeconomic factors

(Leitzmann et al., 2017). It is broadly defined as the mental (e.g., values, attitudes,

beliefs), social (e.g., interacting with others, social support), and emotional (e.g., self-

esteem, mood and anxiety) dimensions of what it means to be healthy (Biddle,

1995). Systematic reviews have shown that sitting, television time and screen time

have all been associated with lower psychological well-being and depression

(Rhodes, Mark & Temmel, 2012), mood disorder and sense of belonging to

community (Dogra & Stathokostas, 2012). Prolonged periods of sitting are associated

with depression and social isolation (de Rezende et al., 2014). While PA has been

positively related to QoL, higher levels of SB have been associated with poorer QoL

(Balboa-Castillo et al., 2011; Meneguci, Sasaki, Santos, Scatena & Damião, 2015). The

combination of increased PA and decreased SB is related to better QoL compared to

those who were are active and more sedentary (Bampton, Johnson & Vallance, 2015;

Hart, 2016). Furthermore, Buman et al. (2010) demonstrated that sedentary time

42

was negatively associated with psychosocial well-being (β -0.03; 95% CI −0.05 - -

0.01); p < 0.001). SB has also been viewed as a positive determinant of anxiety and

depression in older adults (Fernandez-Alonso, Muñoz-García, & Touche, 2016;

Teychenne, Costigan, & Parker, 2015). Anxiety is defined as being “excessively

fearful, anxious, or avoidant of perceived threats in the environment (e.g., social

situations or unfamiliar locations) or internal to oneself (e.g., unusual bodily

sensations)” (Craske & Stein, 2016, p1). Anxiety symptoms have been reported to be

prevalent in 3.2% to 15.4% in older adults (Wolitzky Taylor, Castriotta, Lenze,‐

Stanley, & Craske, 2010). Given that anxiety has been associated with health-related

factors in older adults, further research exploring the associations between anxiety

and PA as well as SB are warranted. SB is extremely prevalent in community-

dwelling older adults and the poor long-term health outcomes of those engaged in

prolonged periods of SB are clear and independent of PA (Copeland et al., 2017).

More work identifying the most efficient methods of reducing time spent in SB is

needed in the older population (Dogra et al., 2017).

2.5. Physical Activity Measurement

2.5.1. Self-Report Measures of Physical Activity

Parallel with the concern to promote and increase PA levels among older adults, how

to most accurately measure PA has drawn the attention of researchers (Lewis et al.,

2017). As described in the behavioural epidemiological framework (Sallis, Owen &

Fotheringham, 2000), accurate measurements of SB and PA are needed to detect

potential correlates; identify relationships between such behaviours and associated

43

health outcomes; and evaluate the efficacy of intervention strategies (Lewis et al.,

2017). SB and PA levels have traditionally been measured via subjective self-report

questionnaires in older adults (Kowalski, Rhodes, Naylor, Tuokko, & MacDonald,

2012). Questionnaire-based methods of data collection have been adopted as they

are relatively cheap to conduct and have the potential to reach a large number of

participants (Aguilar-Farías, Brown, Olds, & Peeters, 2015; Celis-Morales et al., 2012;

Chastin, Culhane, & Dall, 2014; Healy et al., 2011). Several PA questionnaires have

been used and validated for older adults including the Physical Activity Scale for the

Elderly (PASE; Washburn, Smith, Jette, & Janney, 1993; Mudrak, Stochl, Slepicka, &

Elavsky, 2016) and the International Physical Activity Questionnaire for the Elderly

(IPAQ-E; Hurtig-Wennlöf, Hagströmer & Olsson, 2010), which is the questionnaire

used in the work presented in this thesis as required by the funder. The IPAQ-E is

based on the short version of the IPAQ (www.ipaq.ki.se) and assesses time spent

sitting, walking in bouts of 10 minutes or more, MPA in bouts of 10 minutes or more,

and VPA during the previous 7 days. The categorical outcome from IPAQ-E assigns

the participants into one of three PA categories (e.g., low, moderate, or high-PA).

The IPAQ-E provides favourable levels of both direct and indirect levels of criterion

validity for sitting (Spearman r = 0.28, P <0.05), walking (Spearman r = 0.35, P <0.01),

MPA (Spearman r = 0.40, P <0.01), and VPA (Spearman r = 0.37, P <0.01) (Hurtig-

Wennlöf et al., 2010). However, varying levels of test-retest reliability (intraclass

correlation ranging from 0.30 to 0.82) have also been reported (Tomioka et al.,

2011).

44

Even though these self-report measures are tailored to and validated for older

adults, the ubiquitous presence of total accumulated and sporadic PA in older adults

makes it difficult to recall in questionnaire surveys (Washburn, 2000), though such

behaviours may be of particular importance, especially for older adults who tend to

perform shorter duration exercises (Amagasa et al., 2017; Jefferis et al., 2016;

Sparling et al., 2015). A study by van Uffelen, Heesch, Hill, & Brown (2011) also

showed that older adults had difficulties remembering the frequency of sitting and

the scope of sitting activities. Consequently, recall bias is a probability (Barnett, van

den Hoek, Barnett, & Cerin, 2016) given evidence that such methods of data

collection can lead to underestimations of SB (Aguilar-Farías et al., 2015; Chastin &

Granat, 2010; Harvey et al., 2014) and overestimations of time spent engaged in LPA,

MPA and VPA (Tucker et al., 2011).

2.5.2. Objective Measures of Physical Activity

To overcome the associated limitations of self-reported questionnaires, the adoption

of research-grade objective measures including pedometers and accelerometers

have been adopted in older adults (Amagasa et al., 2017; Harris et al., 2018; Ramires

et al., 2017). Pedometers are mechanical counters that record the number of steps

in response to a vertical acceleration of the body (Hensley, Ainsworth & Ansorge,

1993). These devices are lightweight, portable, low cost, and are based on horizontal

hip movement inherent in the swing phase of a step in humans (Ewald, McEvoy &

Attia, 2010). Accelerometers are motion sensors that are sensitive to changes in

acceleration of the body in one or all three axes and are able to provide a more

45

direct measurement of the frequency, intensity, and duration of the movements

related to the activity performed (Doherty et al., 2017).

Pedometer-based interventions have been found to increase PA levels both in the

short- (Hobbs et al., 2013) and long-term (Harris et al., 2018). Consequently,

pedometers continue to be adopted in large-scale community-based PA studies

among older adults given their low cost and non-invasive nature (Harris et al., 2018;

Kerr et al., 2018). Pedometers do however have major disadvantages. They are not

sensitive to engagement in SBs, isometric exercise, activities involving the arms, and

they are also not resistant to water (Kerr et al., 2018). Furthermore, pedometers

have been found to underestimate walking at low speeds and overestimate walking

at higher speeds (Husted & Llewellyn, 2017). Consequnelty, accelerometer usage is

now common among all age-groups (Gorman et al., 2014; Healy et al., 2011) and

accelerometers have been tested within large population based surveillance systems

in a number of developed countries (Hallal et al., 2012). Accelerometers are

particularly appropriate for assessing PA in older adults as these devices require no

input from the participant over the data collection period, and superior wearer

compliance has been demonstrated among older adults when compared to younger

age groups (Doherty et al., 2017). Consequntly, they can provide accurate ways of

estimating the frequency, duration and intensity of both SB and PA (Mathie, Coster,

Lovell, & Celler, 2004; Prince et al., 2008) and are preferable when measuring older

adults’ SB and PA (Murphy, 2009). A development in accelerometer-based SB and PA

research is the move toward raw acceleration signal processing. This advance in

46

accelerometer-based PA monitoring, which has traditionally used accelerometer

output reduced to dimensionless activity “counts” per user-specified period of time

or epoch (Fairclough et al., 2016) is likely to provide greater methodological

transparency in post-data collection analytical processes and improve comparability

of data between different accelerometer models (Hildebrand, Van Hees, Hansen, &

Ekelund, 2014). Devices such as the GA (Activinsights, Cambs, United Kingdom) and

AG GT3X+ and GT9X (ActiGraph, Pensacola, FL) are capable of collecting and

recording raw unfiltered accelerations, which can then be subject to researcher-

driven data processing procedures (Welk, McClain & Ainsworth, 2012).

One of the main decisions to be made by researchers using either raw acceleration

or count-based outcomes is monitor placement location (de Almeida Mendes, da

Silva, Ramires Reichert, Martins, & Tomasi, 2017). The hip has been the conventional

attachment site for accelerometers because of its proximity to the centre of mass

(Troiano, McClain, Brychta, & Chen, 2014; Van Hees et al., 2011). However, recent

accelerometer studies have suggested that the wrist may be a preferable attachment

site as it can more accurately capture the arm motions of non-ambulatory based

activities such as household chores (Evenson et al., 2015; Landry, Falck, Beets, & Liu-

Ambrose, 2015), and is less influenced by atypical gait patterns and walking speed

variability, which are both commonly observed in older adults (Ko, Jerome,

Simonsick, Studenski, & Ferrucci, 2018). Wrist-worn accelerometers have

demonstrated excellent validity against energy expenditure as the criterion measure,

and in comparison to hip-worn monitors (Esliger et al., 2011). Furthermore, wrist-

47

and hip-worn accelerometers have demonstrated comparable free-living MVPA

classification accuracy (Hargens, et al., 2017). Superior wearer compliance has also

been reported for accelerometers worn on the wrist versus the hip in large

population-based studies including the National Health and Nutrition Examination

Survey (NHANES), Dallas Heart Study, and the UK Biobank project adopting wrist-

worn accelerometer protocols (Doherty et al., 2017; Lakoski & Kozlitina, 2014;

Troiano et al., 2014). Specifically, wrist-worn data from the 2011 to 2012 cycle of

NHANES showed that 70–80% of participants provided at least six days of data with

at least 18 hours of wear. This contrasts with 40–70% of participants who provided

at least six days of hip-worn accelerometer data with at least 10 hours of wear in the

2003 to 2004 cycle of NHANES (Troiano et al., 2014). Considering the superior wear

compliance associated with wrist-worn devices in older adults (Doherty et al., 2017),

this attachment site may be the most suitable location during free-living protocols.

However, a limitation of both wrist- and hip-worn accelerometers in studies

involving older adults (Copeland & Esliger, 2009; Taylor et al., 2014) is that most

commonly used cutpoints applied to data to classify PA intensity have been

calibrated for younger adults (Falck, Davis, & Liu-Ambrose, 2016). Specifically,

existing raw acceleration cutpoints include SB cutpoints for GA wrist-worn (46 mg)

and AG hip-worn (47 mg) accelerometers (Hildebrand et al., 2016), and MVPA

cutpoints for GA wrist-worn accelerometers in adults (93 mg; Hildebrand et al.,

2014) and older adults (100 mg; Menai et al., 2017), and AG hip-worn

accelerometers in adults (69.1 mg; Hildebrand et al., 2014). As the energy

48

expenditure associated with a given MET activity intensity threshold is lower in older

adults compared to younger adults (Hall, Howe, Rana, Martin, & Morey, 2013), using

accelerometer cutpoints developed in young adults would likely result in an

underestimation of time spent in MVPA (Barnett et al., 2016). Given the shift

towards objective accelerometry in the measurement of PA, future research should

adopt research grade accelerometers such as the GA and AG triaxial accelerometers

to develop older adult specific wrist- and hip-worn acceleration cutpoints for SB and

MVPA.

Despite objective accelerometry providing a seemingly more accurate report of SB

and PA, there are also limitations. Although accelerometers provide objective

measures, accelerometer site placement (e.g., wrist and hip) affects reliability as

during certain types of activities they cannot accurately detect postural information

(e.g., standing versus sitting) or capture some types of PA (e.g., bicycling), which may

influence estimations of SB and PA and cause some misclassification of time spent in

SB and PA intensity (Shephard & Tudor-Locke, 2016). Emerging techniques including

pattern recognition and machine learning have been found to outperform traditional

cut-off point based algorithms through being robust for individual's physiological and

non-physiological characteristics, more accurate and showing acceptable accuracies

for all activity intensities among older adults (Wullems et al., 2017). Additional

advantages of pattern recognition is that it does not require subgroup-specific

calibrations and/or specific accelerometer body part positioning, is capable of

recognising actual human activities, and works independent of accelerometer

49

brand/settings (Wullems et al., 2017). However, a disadvantage of pattern

recognition research is that pattern data is not directly transferable (Wullems et al.,

2017). For example, algorithms obtained from accelerometer counts at the wrist

would not be directly transferable to raw acceleration data obtained from a differing

accelerometer device. Consequently, large-scale studies are now needed to further

explore the applicability of pattern recognition techniques in distinguishing SB and

differing types and intensities of PA in older adults.

The adoption of wearable devices to monitor personal PA levels has also dramatically

increased through smartphones, wrist- or body-worn devices, and mobile apps. Such

devices offer opportunities for increasing PA (Harris et al., 2018) and new, easy ways

to measure steps (Marshall et al., 2009). Small short-term studies in adults and older

adults have demonstrated that mobile PA apps can increase PA self-monitoring and

engagement in regular PA (Turner-McGrievy et al., 2013; Cadmus-Bertram, Marcus,

Patterson, Parker, & Morey, 2015) and that body-worn consumer fitness trackers

(e.g., Fitbit) can increase time spent in MVPA (Cadmus-Bertram et al., 2015).

However, despite new PA monitoring opportunities, it is important not to ignore

robust, evidence on effective and cost-effective pedometer- and accelerometer-

based interventions (Harris et al., 2018). At present, there is an absence of an

instrument that meets all the advantages desired and thus, adopting both self-report

and objective measures is recommended in order to provide information about the

intensity and duration, as well as the specific types of activities that are engaged in

(Healy et al., 2011; Skender et al., 2016).

50

2.6. Correlates of Physical Activity and Sedentary Behaviour

Factors that are associated with participation in PA are typically referred to as the

study of PA determinants or correlates (Chastin et al., 2017). Correlates will be used

from this point on, as many correlates may not be true determinants, as studies

often show associations yet are unable to conclude causality (National Institute for

Health and Clinical Excellence, 2007). PA is a complex behaviour, influenced by a

number of correlates, which affect the frequency, intensity, duration and type of

older adults’ activity (Sallis & Patrick, 1994). Identifying correlates of PA and SB and

in particular those that are modifiable are imperative in developing successful

interventions (Chastin et al., 2015). Socio-ecological models suggest that behaviours

such as PA have multiple levels of influences, often including intrapersonal (e.g.,

biological and psychological), interpersonal (e.g., social and cultural), organisational,

community, physical environmental, and policy (Sallis et al., 2008). This perspective

proposes that understanding the multiple and interacting determinants of health

behaviours is essential when attempting to change behaviour (Sallis et al., 2008).

Intrapersonal correlates have been the research focus of a majority of studies to

date, either by quantifying motivators or limiters, or through interventions that

attempt to change these intrapersonal correlates (Li et al., 2005). Intrapersonal

correlates of PA include age, health or fitness status, intention to exercise, outcome

expectations, perceived behavioural control, self-efficacy, and perceived fitness

(Choi, Lee, Lee, Kang, & Choi, 2017). Interpersonal correlates relate to dimensions

51

such as how the participant interacts with their friends and family, their level of

social engagement and social cohesion, and the ambience of the social setting

(Berger-Schmitt, 2000). Also important are the support systems available through

general practitioners, community workers, and wellness centre staff. The

environmental level includes the built environment, such as access to local

recreational areas, facilities, and neighbourhood improvements to support activity,

such as footpaths and bike trails. Specifically, intrapersonal including biological (e.g.,

younger age and male sex), psychosocial (e.g., favourable health status, increased

self-determination for PA, perceived greater autonomy support, increased self-

efficacy for exercise, and higher levels of both self-determined motivation and

psychological need satisfaction), and demographic (e.g., higher education) correlates

are all related to time spent in PA among community-dwelling older adults (Bauman

et al., 2012; Chad & Reeder, 2005; Fisher et al., 2018; Hall & McAuley, 2010;

Murtagh et al., 2015; Ng, Ntoumanis & Thøgersen Ntoumani, 2014; Teixeira,‐

Carraça, Markland, Silva, & Ryan, 2012; Thøgersen-Ntoumani, Cumming, Ntoumanis,

& Nikitaras, 2012). In an earlier review of longitudinal studies Koeneman¸

Verheijden, Chinapaw, & Hopman-Rock (2011) also reported that the biological

correlates of general physical functioning and absence of disease were associated

with PA participation. Self-efficacy has consistently been evaluated as the clearest

correlate to PA (Bauman et al., 2012; Choi et al., 2017). According to Bandura’s Social

Cognitive Theory (Bandura, 1986), self-efficacy functions both directly and indirectly

with outcome expectations and other constructs and consequently has a role as a

mediating factor of social support in health behavior (Duncan & McAuley, 1993;

McNeill, Wyrwich, Brownson, Clark, & Kreuter, 2006). Interpersonal correlates of PA

52

have revolved around social support, with increased social support from friends and

family associated with increased levels of PA (Murtagh et al., 2015). Environmental

correlates of PA have included population density, crime rate, geographical location,

perceived neighbourhood safety, perceptions of a PA-conducive physical

environment (e.g., benches available throughout the community), and SES (Murtagh

et al., 2015).

Among possible correlates of SB assessed, age is most frequently associated with

sedentary time (Chastin et al., 2015). In general, age has been positively related to

SB, whether self-reported or measured objectively, and across different countries or

regions (Chastin et al., 2015). Education is also a consistent correlate of SB, with an

inverse association in European populations but not in studies from Asia, suggesting

a possible cultural factor. Health status (Ekelund, Brage, Besson, Sharp, & Wareham,

2008), body mass index (Van Der Berg, 2014), social isolation (Chastin, Fitzpatrick,

Andrews, & DiCroce, 2014), transportation options (Godfrey, Lord, Mathers, Burn, &

Rochester, 2014), SES (Barnett, Van Sluijs, Ogilvie, Wareham, 2014), and the

presence of cultural facilities in the environment and perceived neighborhood safety

are further correlates of SB among community-dwelling older adults (Van

Cauwenberg et al., 2014).

2.7. Community-Based Physical Activity Interventions

53

The growing evidence base surrounding the positive effects of increased PA in older

adults has led to increased implementation of community-based PA interventions

(Bauman, Merom, Bull, Buchner, & Singh, 2016). Such interventions have the

potential to reduce age-related morbidity and declines in activities of daily living,

maintain muscle strength and mass, improve QoL, and thus reduce the primary and

total health care costs associated with SB and physical inactivity among this

population (Bauman et al., 2016). A number of strategies to enhance PA levels in

community-dwelling older adults have been explored in the literature with

interventions delivered in a variety of settings, such as at home, in the community,

and in primary care facilities (Chase, 2015; Neidrick, Fick, & Loeb, 2012. Reviews of

the literature have indicated that interventions can be equally effective in increasing

PA levels regardless of delivery setting (Chase, 2015; Zubala et al., 2017). In terms of

intervention specific components, effective interventions typically utilise

behavioural, motivational and/or cognitive-type components as opposed to health

education or instruction alone (McCluskey & Lovarini, 2005). Furthermore, providing

information on consequences of behaviour, instructing where and when to practice,

and providing ongoing individualised feedback on progress are promising strategies

for PA promotion in this population (Zubala et al., 2017). However, considerations

towards costing and sustainability are needed with such interventions, for example

with expensive equipment and the use of external facilitators to implement the

sessions.

54

In terms of effectiveness, findings from a comprehensive meta-analysis suggest that

interventions designed to increase PA behaviour among adults 65 and older are

effective (Chase, 2015). Specifically, the overall mean effect size for two-group

posttest comparisons was 0.18 (p <.001), equivalent to a difference of 620 more

steps per day or 73 more minutes of PA per week for treatment over control groups

(Chase, 2015). Similar findings were demonstrated in prior meta-analyses studying

younger populations (Dishman & Buckworth, 1996) and healthy adults (Conn,

Hafdahl & Mehr, 2011). There are indications that purely cognitive strategies and

behavioural change techniques (BCTs) might be less suitable for older adults than

motivators more meaningful to them, including social and environmental support,

and enjoyment coming from being physically active. Consequently, a whole system-

oriented multi-component approach is required that is tailored to meet the needs of

older adults and aligned with social, individual and environmental factors (Zubala et

al., 2017).

The Lifestyle Interventions and Independence for Elders (LIFE) pilot study is an

example of an effective whole system-oriented multi-component approach

conducted in comparison with a health-education group (Sink et al., 2015). This

intervention involved a structured, moderate-intensity PA programme that included

walking, strength, flexibility, and balance training. It is recommended that

community-based programmes within this population should consider centre- and

home-based delivery settings in combination rather than isolation (Bauman et al.,

2016). The LIFE intervention involved participants attending two centre-based PA

55

sessions per week, and also performing homebased PA three to four times per week.

An important aspect of this intervention was the flexibility of session content

allowing for increased difficulty of exercises as participants progressed which serves

to allow better tailoring of the intervention to individual needs and the local context

(Lawton, Mceachan, Jackson, West, & Conner, 2014). The ultimate goal was for

participants to progress towards achieving 30 minutes of walking at moderate

intensity, 10 minutes of primarily lower-extremity strength training with ankle

weights, and 10 minutes of balance training and large muscle group flexibility

exercises per week (Sink et al., 2015). The health-education group attended weekly

60 to 90 minute sessions of interactive and didactic presentations, facilitator

demonstrations, guest speakers, or field trips. Sessions included approximately 10

minutes of group discussion and interaction and 5 to 10 minutes of upper extremity

stretching and flexibility exercises (Sink et al., 2015). The intervention resulted in

increases in self-reported PA level from baseline to 24-months (mean difference,

130.4 minutes/week [95% CI, 116.7 to 144.1 minutes/week]) compared with the

health education group (mean difference, 30.5 minutes/week [95% CI, 18.9 to 42.1

minutes/week]; P <.001) (Sink et al., 2015).

Brisk walking in older adults can increase step-counts and MVPA in ≥10 minute

bouts. The Pedometer Accelerometer Consultation Evaluation (PACE)-Lift Cluster RCT

assessed whether a primary care nurse delivered whole-system oriented multi-

component intervention increased objectively measured step-counts and MVPA at

three-, 12-months and four years post-baseline (Harris et al., 2015; Harris et al.,

56

2018). Intervention participants (n=150; control n=148) received four primary care

nurse PA consultations over 3 months, incorporating behaviour change techniques,

pedometer step-count and accelerometer PA intensity feedback, and an individual

PA diary and plan. The addition of individualised support was important as this is

reported to be a key motivator toward PA behaviour change in older adults (Brown

et al., 2015). Results at 3-months revealed that both average daily step-counts and

weekly MVPA in ≥10 minute bouts were significantly higher in the intervention than

control group: by 1,037 (95% CI 513–1,560) steps/day and 63 (95% CI 40–87)

minutes/week, respectively. At 12-months corresponding differences were 609 (95%

CI 104–1,115) steps/day and 40 (95% CI 17–63) minutes/week (Harris et al., 2015).

At 4-years post-baseline versus control results revealed sustained intervention

effects resulting in: +407 (95% CI: −177±992, p =0.17) steps/day, and +32 (95% CI:

5±60, p =0.02) minutes/week MVPA in bouts in the intervention compared to the

control group, respectively. The PACE-LIFT study shows the potential that whole-

system oriented multi-component interventions can have in community-dwelling

older adults. Strategies to implement such interventions may be dependent on the

intervention delivery and its resources. Now more than ever, policy and community

priorities should focus on raising awareness of relationships between PA and health

in older adults, as well as providing better facilities and sustainable PA programmes

in the community which are individually tailored, provide personalised activity goals,

and are delivered according to the evidence-based needs of older adults (Franco et

al., 2015). To maintain intervention effects, gradual transition to less-intensive

programmes along with some type of remote supervision is recommended to avoid

relapse (Bauman et al., 2016). Better methods are needed to set behavioural goals,

57

increase self-monitoring, and provide feedback using new technologies in real time

in the home or community setting (Bauman et al., 2016).

2.8. Community-Based Intervention Process Evaluation

Although the different multi-component interventions discussed in this chapter were

able to have a positive effect on PA levels, research has showed that multi-

component interventions may not always be successful at increasing PA in older

adults (Olanrewaju, Kelly, Cowan, Brayne, & Lafortune, 2016; Richards, Hillsdon,

Thorogood, & Foster, 2013; Zubala et al., 2017). Multi-component interventions can

be difficult to successfully implement and consequently, an accurate interpretation

of either positive or negative outcomes is dependent on having an understanding of

which aspects of an intervention was delivered and how so that they can be further

integrated into ‘real world’ community settings (Bellg et al., 2004; Oakley et al.,

2006; Craig et al., 2008). A large-scale community-based PA intervention study found

that facilitators delivered only around 44% of the specified intervention techniques

across four key sessions (Hardeman et al., 2008). It is recommended, therefore, that

process evaluations of intervention fidelity become an integral part of the conduct

and evaluation of all health behaviour intervention research (Castillo, Wang, Daye,

Shum, & March, 2017). Fidelity is the degree to which an intervention is

implemented as intended by its developers and ensures that the intervention

maintains its intended effects (Carroll et al., 2007). Whether community-based multi-

component interventions succeed at positively impacting PA levels or not, it is

important to understand how they have been implemented in practice, so that the

58

potential for long-term implementation and scaling up to inform policy and practice

of professionals in PA, social work, and care settings can be assessed. This

assessment of implementation is crucial to ensure intervention results are truly

attributable to the programme (internal validity) and that the results are

generalisable to other study populations (external validity) (Frank, Coviak, Healy,

Belza, & Casado, 2008).

The process evaluation of interventions is now advocated by the Standard Evaluation

Framework for PA interventions (SEF), which deems it to be an essential part of

designing and testing multi-component interventions (National Obesity Observatory,

2012). However, approximately only one-third of PA intervention studies report on

process evaluation (Antikainen & Ellis, 2011). This is concerning given that public

health impact is dependent on the extent to which efficacious PA interventions are

disseminated with fidelity into ‘real world’ settings, maintained, and institutionalised

(Lewis et al., 2017). If an intervention is not implemented as directed and no effect is

found, then one cannot be sure whether this is due to lack of efficacy of the

intervention or simply that it has not been implemented correctly (Hasson, 2010).

Despite recommendations for process evaluation research, recent research outlines

that there is considerable heterogeneity and variability in the conceptualisation and

measurement of intervention fidelity in the quality of measurement of delivery

fidelity in interventions promoting PA (Breckon, Johnston & Hutchison, 2008;

Lambert et al., 2017; Quested, Ntoumanis, Thøgersen-Ntoumani, Hagger, Hancox,

59

2017). Consequently, research needs to move toward adopting a common process

evaluation framework which outlines the core variety of fidelity factors consistent in

the literature that affect treatment integrity and treatment differentiation of PA

interventions in community-dwelling older adults (Borrelli, 2011; Calsyn, 2000;

Moncher & Prinz, 1991; Pérez, Van der Stuyft, del Carmen Zabala, Castro, & Lefèvre,

2015). Comprehensive treatment fidelity frameworks specifically developed to

provide guidance for the assessment, enhancement and monitoring of fidelity for

tailored health behaviour interventions include the reach, effectiveness, adoption,

implementation, and maintenance (RE-AIM) framework (Glasgow, Vogt & Boles,

1999), and the National Institutes of Health’s (NIH) Behaviour Change Consortium

(BCC) framework which is adopted in the current thesis (Bellg et al., 2004). The BCC

framework conceptualises fidelity across five core domains: Study Design, Provider

Training, Intervention Delivery, Intervention Receipt and Enactment. Study design is

concerned with whether a study adequately tests its hypotheses in relation to its

underlying theoretical and clinical processes. Provider training involves standardising

training between providers and ensuring they are trained to clear criteria and

monitored over time. Intervention delivery involves assessing and monitoring

differentiation (e.g., differences between the intervention and any comparison

treatments), competency (e.g., skills set of provider), and adherence (e.g., delivery of

intended components). Intervention Receipt refers to whether the intervention was

understood and received by participants and enactment refers to intervention

sustainability and in particular, whether participants used intervention related skills

in day to day settings (Bellg et al., 2004; Borrelli, 2011). Assessing all these elements

enables more accurate inferences to be made about programme effectiveness and

60

any implications for wider roll out/implementation (Dane & Schneider, 1998). The

model has been previously adopted among health behaviour interventions (Chiang,

Seman, Belza, & Tsai, 2008; Nes, van Dulmen, Brembo, & Eide, 2018) and provides a

set of guidelines for translating research into practice and improving the successful

implementation of interventions into ‘real world’ settings (Demiris, Parker Oliver,

Capurro, & Wittenberg-Lyles, 2014).

2.9. Summary of literature

The literature review has highlighted the importance of increasing PA and reducing

SB in community-dwelling older adults, and the potential benefits that this can have

on physical and psychosocial health outcomes. It is also outlined that older adults

both within the UK and on a global level are not meeting current PA guidelines and

thus, interventions to increase PA levels are warranted. The literature suggests that

intervention environment is key for PA promotion in older adults, and that

community-based interventions which are multi-component and target not only

intrapersonal, but also contextual factors such as interpersonal, environmental and

policy level correlates according to socio-ecological models of health behaviour hold

the most promise for positively effecting PA levels. Furthermore, interventions which

are low cost and do not warrant specialist equipment are necessary in order for

them to be sustainable. The literature also highlighted that both self-reported

questionnaires and objective accelerometers are valid methodologies for evaluating

PA levels and SB in older adults and subsequently the effectiveness of interventions.

Moreover, raw data in particular should be analysed. Finally, it is important that

61

interventions are studied in terms of their implementation and fidelity, as this

process evaluation research can improve understanding of how interventions have

been implemented in practice, and subsequently improve the successful

implementation of interventions into ‘real world’ settings.

62

Thesis Study Map

The thesis study map is presented at the beginning and end of each research study

chapter to illustrate the objectives and key findings from the five studies presented

in this programme of work. The thesis study map introduces the next study and

provides a concise summary of the completed study.

Study Objectives and Key Findings

Study 1. Using formative research with older

adults to inform a community physical activity

programme: Get Healthy, Get Active.

Objectives

To explore current knowledge and attitudes

towards physical activity, as well as perceived

barriers, facilitators and opportunities for physical

activity participation among older adults living in

the community.

Use these data to subsequently inform the design,

delivery and recruitment strategies of Sport

England’s national Get Healthy, Get Active

initiative.

Study 2. Evaluation of wrist and hip sedentary behaviour and moderate-to-vigorous physical activity raw

acceleration cutpoints in older adults.

Study 3. Physical activity, sedentary behaviour, perceived health and fitness, and psychosocial wellbeing

among community-dwelling older adults.

Study 4. A pragmatic evaluation of the Get Healthy Get Active physical activity programme for community-

dwelling older adults.

Study 5. Implementation fidelity of the Get Healthy Get Active physical activity programme for community-


63

Chapter 3

Study 1: Using formative research with older adults to inform a community physical activity


64

This study has been published in the Journal of Primary Health Care Research &

Development and can be found in Appendix 3.

Sanders, G. J., Roe, B., Knowles, Z. R., Kaehne, A., & Fairclough, S. J. (2018). Using

formative research with older adults to inform a community physical activity

programme: Get Healthy, Get Active. Journal of Primary Health Care Research &

Development. 1-10. doi: 10.1017/S1463423618000373

3.1. Introduction

In the UK there are over 11 million older adults aged 65 years and over who make up

18 per cent of the population (UK Office for National Statistics (ONS), 2017). Aligning

with the US and other developed countries (United Nations, 2015) this proportion is

projected to increase to at least 24 per cent by 2039 (UK ONS, 2017). Although

prolongation of life remains an important public health goal, of even greater

significance is that extended life should involve preservation of the capacity to live

independently and function well (Rejeski et al., 2013). The purpose of this formative

descriptive study was to explore current knowledge and attitudes towards PA, as

well as perceived barriers, facilitators and opportunities for PA participation among

older adults living in the community. The findings were used to inform the design,

delivery, and recruitment strategies of an ongoing three-year community PA

intervention project, GHGA, which forms part of Sport England’s national GHGA

programme (Sport England, 2012).

Aligned to objectives 1 and 2 of the thesis, the purpose of this formative study was

to 1. Explore current knowledge and attitudes towards PA, as well as the perceived

65

barriers, facilitators and opportunities for PA participation among older adults living

in the community who had agreed to take part in an ongoing PA programme; and 2.

Use this data to inform the design, delivery and recruitment strategies of an ongoing

community PA intervention programme, as well as international PA interventions

among this population. Given the purpose and objectives outlined, the Evidence

Integration Triangle (EIT) (Glasgow, Green, Taylor, & Stange, 2012) was adopted as

the overarching theoretical framework. Through the prompt identification of success

and failures across individual-focused and patient–provider interventions, as well as

health systems and policy-level change initiatives, the framework allows for the

exploration of the three main evidence-based components of intervention

program/policy, implementation processes, and measures of progress. Hence, this

framework enabled a steep learning cycle through an initial 12 week pilot GHGA

programme delivered by the Metropolitan Borough Council within the chosen local

authority. Results and analysis from this pilot were fed back to Sport England as the

funder, as well as deliverers and participants in order to assess, evaluate and

promptly inform adapted future iterations of the GHGA programme.

3.2. Methods

3.2.1. Participants and procedures

A descriptive formative study was undertaken from March to June 2016. Participants

were recruited from one local authority in North West England recognised as having

the highest percentage of inactive older adults (80%) compared to the UK national

average, and the highest national health costs associated with physical inactivity

66

(Active People Survey, 2014; Sport England’s Local Profile Tool, 2015). The first

author facilitated six, mixed-gender focus groups. Representative of the uptake of

participants within the target GHGA initiative, a homogenous purposive sample of 28

community-dwelling white, British older adults (five male) participated in five of the

focus groups, with an additional convenience pragmatic sub-sample of six

participants (three male) recruited from an assisted living retirement home,

participating in the sixth focus group. In total, 34 older adults (eight male), aged 65

to 90 years (mean age of participants =78, SD=7 years), participated across the six

sessions. Four focus groups involved a group size of six to ten participants, and two

involved three participants (mean focus group size=6 participants, SD=5). Previous

focus groups in PA studies have been conducted effectively with as many as 12

(Moran et al., 2015), and as few as four (Schneider et al., 2016) participants. Focus

groups took place in two church halls, an assisted living retirement home lounge,

and a theatre. All locations were free from background noise, and participants could

be overlooked but not overheard. The inclusion criterion set out by Sport England as

funders of the GHGA programme were that participants must be 65 years of age or

over, reside within one local authority in North West England, and could provide

written informed consent to participate.

GHGA is an ongoing three-year project which seeks to increase the number of

inactive older adults participating in PA at least once a week for 30 minutes, via a 12

week PA intervention delivered by the Metropolitan Borough Council within the

assigned local authority. Participants due to participate in GHGA received a covering

letter, participant information sheet, and consent form. Prior to the commencement

67

of the study, institutional ethical approval was received (#SPA-REC-2015-329) and

written informed consent was obtained for all participants prior to participation. All

focus groups utilised the PRECEDE stage of the PRECEDE-PROCEDE model (Green &

Kreuter, 2005) within their design allowing for the exploration of predisposing,

enabling and reinforcing correlates of PA participation. To maximise the interaction

between participants, focus group questions were reviewed by the project team for

appropriateness of question ordering and flow. Subsequent minor additions were

made to questions on social isolation and PA advertisement. The semi-structured

discussion guide included open ended questions structured to prompt discussion

with equal chance for participants to contribute (Stewart & Shamdasani, 2014).

Focus groups were led by a trained facilitator and with an observer/note taker also

present. Questions addressed knowledge, attitudes and beliefs towards PA as well as

views on barriers and opportunities for PA participation. An example question from a

section exploring barriers to PA was: “Can you tell me about what stops you from

participating in physical activity?” Questions therefore demonstrated aspects of face

validity as they were transparent and relevant to both the topic and target

population (French et al., 2015).

3.2.2. Data Coding and Analysis

Focus groups lasted between 20 and 45 minutes (mean focus group length =29

minutes, SD=12), were audio recorded, and later transcribed verbatim, resulting in

66 pages of raw transcription data with Arial font, size 12 and double-spaced.

Verbatim transcripts were read and re-read to allow familiarisation of the data and

68

then imported into the QSR NVivo 11 software package (QSR International Pty Ltd.,

Doncaster, Victoria, Australia, 2017).

Previous research within this population has adopted analytical procedures including

thematic analysis (Van Dyck et al., 2017), content analysis (Middelweerd, Mollee,

van der Wal, Brug, & te Velde, 2014) and used specialist qualitative data analysis

packages, such as NVivo (Warmoth et al., 2016). In supporting new methodologies

and data representation within qualitative research (Orr & Phoenix, 2015), the

current study followed the pen profiling protocol. The pen profile approach has been

used in recent child PA research (Mackintosh et al., 2011; Boddy et al., 2012;

Knowles et al., 2013; Noonan et al., 2016b) and presents findings from content

analysis via a diagram of composite key emerging themes. In summary, data were

initially analysed deductively via content analysis (Braun & Clarke, 2006), using the

PRECEDE component of the PRECEDE-PROCEED model (Green & Kreuter, 2005) as a

thematic framework which reflects the underlying study purpose. Inductive analysis

then allowed for emerging themes to be created beyond the pre-defined categories.

Data were then organised schematically to assist with interpretation of the themes

(Aggio et al., 2016). As akin to more traditional qualitative research, verbatim

quotations were subsequently used to expand the pen profiles, provide context, and

verify participant responses. Previous studies have demonstrated this method’s

applicability in representing analysis outcomes within PA research (Mackintosh et al.,

2011; Boddy et al., 2012; Knowles et al., 2013; Noonan et al., 2016a) making it

accessible to researchers who have an affinity with both quantitative and qualitative

69

backgrounds (Knowles et al., 2013; Noonan et al., 2016a). Recent findings suggest

that the discrepancy between objective isolation and felt loneliness may be

associated with undesirable health outcomes such as cognitive dysfunction.

Three pen profiles were developed to display themes within the data aligned to the

PRECEDE component of the PRECEDE-PROCEED model (Green & Kreuter, 2005).

Quotations were labelled by focus group number (Fn) and subsequent participant

number (Pn) within that focus group. Characterising traits of this protocol include

details of frequency counts and extracts of verbatim quotes to provide context to

the themes. A minimum threshold for theme inclusion was based upon comparable

participant numbers within previous research adopting a pen profiling approach

(Boddy et al., 2012; Noonan et al., 2016a) and hence, was set as ≥n = 6, with n

representing individual mentions per participant. However, multiple ‘mentions’ by

the same participant were only counted once. Methodological rigour was

demonstrated through a process of triangular consensus (Hawley- Hague et al.,

2016) between the authors. This offered transparency, credibility, and

trustworthiness of the results, as the data were critically reviewed using a reverse

tracking process from pen profiles to verbatim transcripts, providing alternative

interpretations of the data (Smith & Caddick, 2012). The process was repeated

through cross verification and discussion until subsequent agreement on data

themes in relation to verbatim extracts was reached (Aggio et al., 2016).

70

3.3. Findings and Discussion

3.3.1 Predisposing Correlates

Figure 3.1 displays the predisposing correlates of PA participation. In agreement with

previous research (Gray, Murphy, Gallagher, & Simpson, 2015; Kosteli, Williams &

Cumming, 2016), the most highly cited theme of motivation (n=29) was perceived to

be both a facilitator (n=15) and barrier (n=14) to PA participation throughout. Some

participants were proactive in seeking out opportunities for PA.

“I’m a lung cancer survivor and I just ran a mile last month and I raised £550.” (Focus group number (Fn) 1, Participant number (Pn) 2, Line 76).

Contrastingly, others expressed disinterest in PA altogether believing that they

would not derive any health benefit.

“I’ve pushed these [PA] classes to lots and lots of friends and they still ignore it, they will not come to anything like this.” (Fn1, Pn3, Lines 98-100).

Participants also reported laziness or apathy to prevent participation.

“It’s [lack of PA] apathy, just apathy, people can’t be bothered.” (Fn4, Pn3, Line 43).

The importance of pre-intervention intrinsic motivation (e.g., participating for

enjoyment) among older adults is key for both initial adoption and maintenance of

PA participation (Gray et al., 2015). Hence, future interventions could promote

intrinsic motivation for PA through the adoption of socioemotional selectivity theory

71

(Carstensen, Isaacowitz & Charles, 1999). Recent findings support this theory’s

notion that motivation for PA is more effectively promoted when paired with

positive messages about the benefits of PA rather than with negative messages

about the risks of inactivity (Notthoff et al., 2016).

The theme of age (n=20) was identified as a key barrier (n=13) to PA participation

throughout.

“They [older adults] get to a certain age and just give up.” (Fn1, Pn7, Line 110).

Social norms and cultural misconceptions often influence not only the type of PA in

which older adults engage, but whether they participate at all (Greaney et al., 2016).

Moreover, participants noted that lifestyle (n=20) often affects individual views

regarding ageing stereotypes, and therefore PA participation. Some participants felt

that physically active older adults were more likely to be habituated to PA

engagement over many years.

“Well if you’ve kept healthy, kept fit all your life, you can keep doing it.” (Fn1, Pn4,

Line 83).

Conversely, it was felt that inactive older adults were reluctant to start exercising.

“You see the ones who haven’t been doing it [PA] are not going to be able to start and do it now.” (Fn2, Pn1, Lines 121-122).

72

Previous research has also reported prior PA behaviours (e.g., being sedentary or

active) to be key correlates affecting older adults’ current PA participation levels

(Franco et al., 2015). Additionally, ageing is associated with a decrease in the size of

social networks and hence, older adults are at increased risks of isolation (Devereux-

Fitzgerald et al., 2016; Greaney et al., 2016). Corroborating with prior research

(Greaney et al., 2016), participants throughout perceived isolation (n=15) to be a key

barrier (n=14) to PA participation.

“It’s so easy to get trapped inside and not go out. People sit in front of the television from the moment they wake up to when they go to bed.” (Fn6, Pn5, Lines 79-81).

Isolation is associated with decreased social and psychological wellbeing (Owen et

al., 2010; Milligan et al., 2015) and increased SB among older adults (Nicholson,

2012). Certain targeted intervention strategies can reduce isolation by providing an

opportunity for older adults from differing socio-economic areas to take part in PA

within local community spaces (e.g., parks, leisure centres and churches), that

promote social networking by encouraging camaraderie, adaptability, and

productive engagement, without the pressure to perform (Milligan et al., 2015;

Gardiner et al., 2016). Given that SB is an independent and modifiable behavioural

target for interventions (Lewis et al., 2017), opportunities to replace SB with health-

enhancing behaviours such as moderate-to-vigorous PA (Prince, Saunders, Gresty, &

Reid, 2014), light PA (McMahon et al., 2017; Phoenix & Tulle, 2017) and standing

(Healy et al., 2015) should be promoted. However, none of the participants in the

current study noted negative health effects of prolonged sitting, or the importance

of breaks in sedentary time. Previous research has noted that older adults are not

73

yet familiar with the concept of SB and hence, are not motivated to reduce such

behaviours (Van Dyck et al., 2017). Hence, it is first crucial to increase knowledge

about the negative health consequences of SB independent from PA among both

older adults and other populations (Van Dyck et al., 2017).

Participants also emphasised the importance of having a wide range of choice and

opportunities for PA (n=22), and in general their perceptions of community provision

were positive (n=16).

“Yes it’s quite a good place [the local authority where the study took place]. There are a lot of different physical activity sessions to try.” (Fn2, Pn1, Lines 133-135).

However, in line with recent research (Baert et al., 2016; Träff, Cedersund and Nord,

2017), key barriers noted by the participants within the assisted living group included

a lack of advertisement regarding PA opportunities, and few opportunities to take

part in PA within the assisted living facility itself.

“It’s hard to know what is on if you don’t read the noticeboards and to be honest most of us have even stopped looking at that [noticeboard] because there is never anything on it.” (Fn3, Pn3, Lines 49-51).

Further research into the most effective advertisement strategies to engage older

adults in assisted living facilities is warranted (Hildebrand and Neufeld, 2009).

Regardless of living status, participants noted a strong preference not to engage with

online and/or social media channels for advertising and awareness-raising.

74

“A lot of people our age don’t like that technology stuff at all. I would not know where to start.” (Fn5, Pn2, Lines 331-332).

These results suggest educational strategies outlining the potential benefits of

technology in aiding PA participation are needed (Bird et al., 2015). This is especially

salient given that recent research has shown technology-based interventions to have

good adherence and provide a sustainable means of reducing SB and promoting PA

participation among older adults (Garcia et al., 2016; Skjæret et al., 2016).

Figure 3.1. Predisposing correlates of physical activity participation among older adults. n = Individual mentions per person (multiple mentions not included); Fn = Focus group number; Pn = Participant number.

3.3.2 Enabling Correlates

Figure 3.2 displays the enabling correlates of PA participation. Consistent with

previous research findings (Franco et al., 2015; Borodulin et al., 2016), cost (n=21)

75

was perceived to be a key barrier (n=12) to PA participation exclusively among the

community-dwelling participants who were either unable, or unwilling to pay the

perceived high costs associated with both attending and travelling to such

programmes.

“Money is the big bug bear [barrier to PA participation] isn’t it.” (Fn2, Pn5, Line 406).

Examples of competing programmes were also noted, with free and lower cost

programmes taking precedence over the more expensive.

“We like it [a local chair-based PA programme] because it’s free.” (Fn4, P3, Line 392).

Thus, to effectively increase PA participation within this population, health-

promotion strategies should go further than merely educating and raising awareness

about potential health benefits, and should also advocate for the provision of low-

cost, and easy reachable PA opportunities regardless of financial status (Petrescu-

Prahova, Belza, Kohn, & Miyawaki, 2015; Borodulin et al., 2016). It is worth noting

that for the participants recruited from the assisted living retirement home, any PA

sessions delivered were included within the cost of the overall living fee, and hence

lack of financial resources was rejected as a potential barrier for PA participation

(Baert et al., 2016).

Participants’ views on the theme of location (n=11) centered on neighbourhood

safety. Declining health and physical impairments associated with ageing increase

76

the time spent in ones’ neighbourhood and thus, neighbourhood environmental

factors such as, PA provision, proximity, traffic volume, and overall neighbourhood

safety are considered to be important correlates affecting older adults’ PA

participation (Greaney et al., 2016). Perceived neighbourhood safety was identified

as a barrier (n=7) to PA participation exclusively among the community-dwelling

older adults.

“You wouldn’t go out on your own at night around here.” (Fn1, Pn5, Line 203).

Participants from the assisted living retirement home did not view neighbourhood

safety to be either a barrier to or facilitator of PA. This neighbourhood environment

was perhaps viewed as the norm and therefore they did not associate safety

concerns so acutely (Moran et al., 2015). This association could have also affected

results obtained for the theme time/day of the week as such participants did not

recognise this to be a barrier to PA participation either.

“Time of day wouldn’t make much difference [to PA participation]. To be fair you aren’t doing much at the weekend so day of the week isn’t going to make much difference [to PA participation] either.” (Fn3, Pn1, Lines 403-405).

Conversely, community-dwelling participants reported time/day of the week to be a

barrier (n=15), with early morning or early evening sessions identified as reducing PA

participation, especially during the winter months when daylight hours are more

limited. These findings could have been further amplified by the neighbourhood

77

safety concerns also identified by this group (Hoppmann et al., 2015; Prins & van

Lenthe, 2015).

The theme of transportation (n=14) has been extensively reported to be both a

barrier and facilitator to PA participation among older adults (Bouma, van Wilgen &

Dijkstra, 2015; Haselwandter et al., 2015; Kosteli et al., 2016; Van Dyck et al., 2017).

Within the current study transportation was identified as a barrier (n=10) restricting

access to PA sessions regardless of living status.

“I would like to go to the baths [swimming pool] but it’s difficult to get there and back so I just don’t bother.” (Fn4, Pn5, Lines 302-303).

Transport is especially important for those lacking the ability to be more

independently mobile as it allows individuals to bridge larger distances than they

could by walking alone (Van Cauwenberg et al., 2016). Thus, lack of access to a car

and inadequate availability, frequency and reliability of affordable public transport

are all associated with decreased PA participation (Newitt, Barnett & Crowe, 2016).

Additionally, being dependent upon others (e.g., family, friends and peers) for

transportation has been identified as a barrier to PA participation within this

population (Baert et al., 2015). This was also noted in the current study.

“I think the worst thing is having to rely on somebody else to take you [to a PA session] as anything can happen in your own life let alone somebody else’s.” (Fn5, Pn2, Lines 266-267).

78

Prior research suggests the promotion of walking for transportation to PA sessions

among physically independent older adults (Chudyk et al., 2017). However, given the

neighbourhood safety concerns noted by participants, and the varying levels of

functional ability among this population, further research examining access to PA

sessions including walking facilities (e.g., path and crossing quality), traffic safety,

and safety from crime is warranted (Van Cauwenberg et al., 2016).

Figure 3.2. Enabling correlates of physical activity participation among older adults. n = Individual mentions per person (multiple mentions not included); Fn = Focus group number; Pn = Participant number.

3.3.3 Reinforcing Correlates

Figure 3.3 displays the reinforcing correlates of PA participation. Peer support is

associated with PA adherence in older adults (Brown et al., 2015), and was identified

as a key theme (n=18) and subsequent facilitator (n=13) to PA participation in the

current study.

79

“I’ve got to know everybody now and I’m used to you all. I feel more comfortable and I don’t feel anxious or anything.” (Fn3, Pn6, Lines 354-355).

Unsurprisingly, in light of the above several participants reported peers to be a

barrier to PA participation (n=5) because of an unwillingness to attend other PA

sessions due to anxieties about meeting new people.

“I wouldn’t like to go somewhere else as I wouldn’t like to walk in on a crowd of new people.” (Fn3, Pn6, Lines 366-367).

Although group-based activities offer older adults the chance to gain a sense of

belonging, enjoyment and establish friendships, designing sustainable exit routes in

order to retain the provision of group activities which continue to facilitate, build

and retain social bonds post-intervention should be considered by PA programmers

and policymakers (Wu et al., 2015).

In line with recent research (Devereux-Fitzgerald et al., 2016; Smith et al., 2017),

family members were identified as being both barriers (n=2) and facilitators (n=4) to

PA participation. Specifically, a barrier often reported is overprotectiveness, in which

family members may not allow older adults to participate in PA out of concern for

their safety or health (Greaney et al., 2016). Participants among the community-

dwelling groups also noted this.

“My sons in for a shock that we’re coming to this as he’s like, ‘no long walks, no boat rides’, he goes ‘you’re past it.” (Fn6, Pn2, Lines 441-442).

80

Such results suggest a need to educate family members on the importance and

benefits of PA among older adults. Educational resources such as the older adults PA

guidelines infographics for the, UK (Reid & Foster, 2016), Canada (Canadian Society

for Exercise Physiology, 2016), Australia (Australian Government Department of

Health and Ageing, 2013), New Zealand (Ministry of Health, 2013), and the US (CDC,

2008) are appropriate tools advocating for older adults to be active safely, and can

be understood by family members plus health care providers. Furthermore, the

adoption of local/national mass media messages may be a cost effective educational

solution at a time when there is a growing ageing population (United Nations, 2015;

UK ONS 2017). However, given the resistance to technology-based PA noted in the

current study, further educational strategies promoting enjoyable, easy-to-use

technology within a family environment are needed for community-dwelling older

adults (Bird et al., 2015). Participants within the assisted living group did not

perceive family members to be either barriers or facilitators to PA participation and

thus, further research is needed to identify approaches to involve family members as

additional facilitators of PA participation within this group.

Participants viewed the theme of perceived health benefits (n=23) to be both a

facilitator (n=14) and barrier (n=9) to PA participation regardless of living status.

Participants were knowledgeable regarding the potential benefits of PA for their

physical health.

“It [PA] loosens all your limbs up.” (Fn2, Pn2, Line 123).

81

Participants also noted the potential benefits of PA for their psychological health.

“The wellbeing [from PA participation] makes you feel better.” (Fn1, Pn3, Line 49).

Despite the irrefutable evidence demonstrating the benefits of PA among older

adults (CDC, 2015; Reid & Foster, 2017; WHO, 2017), participants also noted health

to be a potential barrier (n=14) to PA participation due to doubts about their

capabilities, or fear of causing themselves harm, particularly if they were unfamiliar

with it.

“People have to be sure they can come to PA sessions because my sister had a heart attack… and she can’t do a lot of these exercises.” (Fn1, Pn5, Lines 177-178).

To overcome such perceptions, educational strategies at a population level should

focus on communicating the role of PA in gaining health benefits for all as well as

how well-designed PA programmes can aid in the management of common

comorbidities specific to this age group (Gillespie et al., 2012; Hamer, Lavoie &

Bacon, 2013).

82

Figure 3.3. Reinforcing correlates of physical activity participation among older adults. n = Individual mentions per person (multiple mentions not included); Fn = Focus group number; Pn = Participant number.

Taken together with the findings of recent qualitative studies examining correlates of

PA participation among older adults living in both assisted living (Baert et al., 2016;

Träff et al., 2017) and community-dwelling older adults (Fisher et al., 2018; Phoenix

& Tulle, 2017), results from this formative research study have been used to inform

the design, delivery and recruitment strategies of an ongoing community PA

intervention project. Specifically, changes implemented to programme design have

included the introduction of, increased intervention duration from six to 12-weeks,

maintenance sessions post-initial 12-week intervention, tea and coffee after each

session to promote social interaction, and a reduction of early morning and late

afternoon sessions. Changes to programme delivery have included the introduction

of, participant choice in session activities, videoing participants at week one and

week 12 to show participants their progression, and signposting participants to other

83

local PA programmes. Finally, changes implemented to recruitment strategies have

included, improved relationships with general practitioners to enable them to refer

participants onto the programme, leafleting in church halls and charity shops, and

deliverers attending and subsequently advertising the programme at several Older

Peoples' Forums. Such methods could also be adopted throughout similar

community PA programmes elsewhere in order to increase programme fidelity,

representativeness and effectiveness.

3.4. Strengths and Limitations

Methodological strengths include the exploration of consensus and associated

discussion through the focus groups and subsequent analysis process which allowed

insight into the predisposing, enabling and reinforcing correlates of PA participation

among older adults. Consistency of themes, data credibility, transferability, and

dependability were achieved through the triangulation consensus of data between

authors and methods. While this study reiterates important insights into the

perceived barriers, facilitators and opportunities for PA participation among both

community-dwelling and assisted living older adults, value outside of this to the

wider research community may be limited due to programme funding which only

allowed for formative research strategies to recruit participants who had agreed to

take part in an ongoing PA programme. Consequently, sampling bias is a potential

issue as it could be assumed that a high proportion of the participants were already

inclined to be and/or currently physically active given the positive predisposing

comments with regard to motivation towards PA and current lifestyle choices

(Costello et al., 2011). This is especially important given that motivators and barriers

84

toward regular PA vary among currently active and inactive adults across the age

range (Costello et al., 2011; Hoare et al., 2017). Considering that less than 10% of

older adults (≥ 65 years of age) meet the recommended PA guidelines (Jefferis et al.,

2014), future research should seek to identify barriers and facilitators among larger

sample sizes of currently inactive older adults living within both the community and

assisted living facilities.

Additionally, a small convenience pragmatic sub-sample of participants from one

assisted living facility were recruited and hence results cannot be considered

representative. Furthermore, men tend to decrease participation in leisure-time PA

as they get older; whereas this dose-response is not seen among women (Amagasa

et al., 2017). Consequently, there is the possibility of gender bias given the higher

number of female participants recruited. However, the sample size, participants’

ages and gender distribution are comparable to those reported in two recent studies

examining barriers and facilitators to PA participation among older adults (Baert et

al., 2015; Moran et al., 2015). Within these two studies the total number of

participants was 15 (five male) and 40 (13 male) and the mean age of the

respondents was 74 years, and 84 years, respectively. This compares to a total

number of thirty-four participants (eight male) with a mean age of 78 years in the

current study. Nevertheless, as well as exploring correlates of PA participation in

relation to gender, functional status and age differences between the young-old (60-

69 years), old-old (70-79 years), and oldest-old (80+ years) (Heo et al., 2017), future

research should obtain additional participant characteristic data prior to the

intervention including, participants’ current sedentary time and PA levels, history of

85

PA, family history of PA, ethnicity, employment status, and educational

achievements as such have been shown to potentially affect the perceived barriers

and facilitators to PA participation among older adults (Greaney et al., 2016; Keadle

et al., 2016).

3.5. Conclusions

Older adults acknowledged the benefits of PA, not only for health but also those

relating to socialising, enjoyment, relaxation, and physical and psychological

wellbeing. The themes of opportunities and awareness for PA participation, cost,

transport, location and season/weather varied dependent upon living status. These

findings suggest current living status to be a separate correlate of PA participation

among older adults. This data can be used to further strengthen the design, delivery

and recruitment strategies of both the target GHGA PA intervention programme and

international PA intervention programmes among older adults. Future interventions

should consider educational strategies to communicate the role of PA in gaining

health benefits for all, reducing SB, and countering the negative implicit attitudes

that may undermine PA within this population. Given the small sample of

participants in the current study, further comparative research exploring the barriers

and facilitators between assisted living and community-dwelling, and active and

inactive older adults on both national and international levels is warranted.

86

Thesis Study Map





Objectives





the community.




initiative.

Key Findings:

Older adults acknowledged the benefits of

physical activity, not only for health but also those

relating to socialising, enjoyment, relaxation, and

87

physical and psychological wellbeing regardless of

socioeconomic status.

The themes of opportunities and awareness for

physical activity participation, cost, transport,

location and season/weather varied between

assisted living and community-dwelling older

adults.

Study 2. Evaluation of wrist and hip sedentary

behaviour and moderate-to-vigorous physical

activity raw acceleration cutpoints in older

adults.

Objectives

To test a laboratory-based protocol to generate

behaviourally valid, population specific wrist- and

hip-based raw acceleration cutpoints for

sedentary behaviour and moderate-to-vigorous

physical activity in older adults.

Apply these cut-points to subsequently analyse

physical activity data for Sport England’s Get

Healthy Get Active physical activity intervention.

Study 3. Physical activity, sedentary behaviour, perceived health and fitness, and psychosocial wellbeing

among community-dwelling older adults.




dwelling older adults

88

Chapter 4

Study 2: Evaluation of wrist and hip sedentary



adults.

89

This study is currently under peer-review in the Journal of Sports Sciences.

Sanders, G. J., Boddy, L.M., Sparks, A. S., Curry, W.B., Roe, B., Kaehne, A., &

Fairclough, S. J. (In Review). Evaluation of wrist and hip sedentary behaviour and

moderate-to-vigorous physical activity raw acceleration cutpoints in older adults.

4.1. Introduction

Chapter 3 (Study 1) established that older adults acknowledge the benefits of PA,

not only for health but also those relating to socialising, enjoyment, relaxation, and

physical and psychological wellbeing. Findings were also supportive of previous

research which has highlighted current living status (assisted living versus

community-dwelling) to be a separate correlate of PA participation among older

adults (Baert et al., 2016; Träff et al., 2017). As described in the behavioural

epidemiological framework (Sallis, et al., 2000), accurate measurements of SB and

PA are needed to detect potential correlates; identify relationships between such

90

behaviours and associated health outcomes; and evaluate the efficacy of

intervention strategies (Lewis et al., 2017). Accelerometers are particularly

appropriate for assessing PA in older adults as these devices require no input from

the participant over the data collection period, and superior wearer compliance has

been demonstrated when compared to younger age groups (Doherty et al., 2017).

Objective measurement methods such as accelerometry also eliminate self-report

questionnaire bias related to subjective recall of past events which is an ability that

can decline with ageing (Barnett et al., 2016). Consequently, accelerometry is now

commonly adopted for monitoring older adults’ SB and PA levels (Mañas et al., 2017;

Oguma et al., 2017; Wullems et al., 2017; Zhu et al., 2017).

A further development in accelerometer-based SB and PA research is the move

toward raw acceleration signal processing. This advance in accelerometer-based PA

monitoring, which has traditionally used accelerometer output reduced to

dimensionless activity “counts” per user-specified period of time or epoch

(Fairclough et al., 2016) is likely to provide greater methodological transparency in

post-data collection analytical processes and improve comparability of data between

different accelerometer models (Hildebrand et al., 2014). Devices such as the GA

(Activinsights, Cambs, United Kingdom) and AG GT3X+ and GT9X (ActiGraph,

Pensacola, FL) are capable of collecting and recording raw unfiltered accelerations,

which can then be subject to researcher-driven data processing procedures (Welk et

al., 2012). Interunit reliability is acceptable for both brands (Esliger et al., 2011;

Santos-Lozano et al., 2012). Although time spent in SB and light, moderate and/or

91

vigorous intensity PA can be quantified from raw acceleration data (Matthew, 2005),

currently, no raw acceleration cutpoints for SB and MVPA exist for older adults.

Therefore, in line with objective 3 of the thesis, this study aimed to determine

laboratory-based wrist-worn GA and hip-worn AG GT3X+ raw acceleration cutpoints

for SB and MVPA in older adults. The obtained cutpoints will then be adopted to aid

in answering objectives 5 and 6.

4.2. Methods

4.2.1. Study Population

A homogenous purposive sample of 34 community-dwelling white, British older

adults (10 male), aged 60 to 86 years (mean number of participants =70, SD=8 years)

were recruited through leaflets distributed at local fitness centers/gyms and

community centers, as well as through word of mouth referrals. A sample size of 30

participants was targeted so as to be comparable with recent calibration studies in

older adults (Landry et al., 2015; Wullems et al., 2017). Individuals interested in

participating in the study were pre-screened for inclusion criteria which set out that

participants must be (1) ≥60 years of age and be physically cleared for exercise using

the modified Physical Activity Readiness Questionnaire (Modified PAR-Q; Cardinal,

Esters & Cardinal, 1996; Cardinal & Cardinal, 2000), (2) have the ability to walk

briskly on a treadmill without assistance, and (3) not be taking any medications that

would influence EE or their ability to perform ambulatory activity. Participants were

excluded if they had a medical condition precluding them from exercise, were unable

to wear a portable indirect calorimeter during testing, or had limited mobility such

92

that they could not walk on a treadmill independently. Prior to the commencement

of the study, institutional ethical approval was received (SPA-REC-2016-343) and all

participants provided written informed consent prior to their inclusion.

4.2.2. Anthropometrics

Participants’ body mass (Seca mechanical scales, Birmingham, UK) and standing

height (Holtain Limited stadiometer, Crymych, UK) were measured in light clothing

without shoes. Resting blood pressure was measured upon arrival and immediately

prior to commencement of the accelerometer calibration protocol using a Boso

Medicus Prestige blood pressure monitor (Boso Bosch + Sohn, Germany).

4.2.3. Study Protocol

Participants completed two separate laboratory data collection visits (separated by

one week) at the university site. Visit 1 was an initial familiarization of the protocol

structure, equipment, and laboratory. Participants also provided written informed

consent and completed anthropometric measures, and a six-minute treadmill walk

test (6MWT; ATS Committee on Proficiency Standards for Clinical Pulmonary

Function Laboratories, 2002) to establish maximal walking speed. Participants then

completed a laboratory-based protocol consisting of six sedentary activities, six

stationary activities and four physical activities of varying intensity (see Table 4.1 for

protocol description). All activities were performed in a standardised order.

Participants were provided with standardised instructions from a predetermined

script delivered by the first author prior to beginning each activity. The activity

93

protocol only was repeated at visit 2. Visit 1 lasted ∼80 minutes whilst visit 2 lasted

∼60 minutes as participants did not need to repeat the familiarization components

and 6MWT.

Table 4.1. Description of the sixteen structured activities.

Activity Description of activity

Lying down

Reclined

Reclined

Sitting

Sitting

Sitting

Standing

Standing

Standing

Washing

Shopping

Mopping

Stepping

Walking

Walking

Walking

Lying in supine position awake, with arms at the side, avoiding bodily movement.

Reclined in supine position reading newspaper turning the page every 20 seconds.

Reclined in supine position using a mobile phone.

Sitting in a chair with hands on knees, avoiding bodily movement.

Sitting in a chair reading newspaper turning the page every 20 seconds.

Sitting in a chair using a mobile phone.

Standing upright, with arms at the side, avoiding bodily movement.

Standing upright reading newspaper turning the page every 20 seconds.

Standing upright using a mobile phone.

Standing upright at a sink washing (20 seconds) and drying (20 seconds) five items.

Standing placing five items in cupboard, then remove with the opposite hand.

Standing upright dry mopping the floor (marked out area).

Stepping up and down at 69 beats per minute on a 23cm high step.

Walking on a treadmill at 65% maximal speed individually calibrated from the 6MWT.



6MWT, six-minute walk test.

94

Sedentary, stationary and light physical activities were performed for three minutes

each, whilst the stepping and walking activities representing MVPA were performed

for two minutes each. All activities were separated by a 1 minute transition period.

To ensure consistency across measurement sessions, the first author was trained

and led on all aspects of the measurement protocol. The start and end of each

activity was timed by the first author with a stopwatch (Fastime, Leicestershire, UK).

All instruments were synchronised to the same clock to ensure that criterion

measurements for a given period of activity were matched with time-stamped

accelerometer data for precisely the same duration of the activity. This allowed for

appropriate data comparisons to be made across all recording devices.

4.2.4. Accelerometers

Participants wore GA (ActivInsights Ltd., Kimbolton, Cambridgeshire, UK) and AG

GT3X+ (ActiGraph, Pensacola, FL) triaxial accelerometers on the non-dominant wrist

and left hip, respectively. Both monitors were set to collect raw triaxial accelerations

at 60 Hz. Participants also wore an activPAL (PAL Technologies Ltd., Glasgow, UK)

accelerometer on the left anterior thigh as the criterion measure for SB (Chastin,

Culhane & Dall, 2014; Varela Mato, Yates, Stensel, Biddle, & Clemes, 2017). The

activPAL uses proprietary algorithms to classify activity into periods spent sitting,

standing and stepping. acitvPAL data were collected at a sampling frequency of 20

Hz. Participants wore the same monitors (matched by serial number) for both visits.

A total of six different GA and AG, and four different activPAL monitors were used

throughout the study. All devices were used and calibrated as per the manufacturer

95

instructions and initialised approximately 10 minutes before the start of each

session.

Immediately after testing, the activity monitors were removed and the data

downloaded to a single, secured computer. The GA data were downloaded using GA

PC software version 2.9 and saved in raw format as binary files, whilst the AG data

were downloaded using ActiLife version 3.13.3, and saved in raw format as .gt3x files

and converted to time-stamped .csv files to facilitate raw data processing. ActivPAL

data were downloaded using activPAL3 version 7.2.32, saved as .datx files and

converted to .csv “Event” files for processing. Signal processing of raw GA .bin files

and raw AG .csv files was completed offline using R-package GGIR version 1.5

(https://cran.r-project.org/web/packages/GGIR/) (van Hees et al., 2013). This R-

package facilitates data cleaning and the extraction of user-defined acceleration

levels, which can then be set to reflect the intensity thresholds as derived in this

study. Concurrent with previous studies (Hildebrand et al., 2014; Fairclough et al.,

2016; Menai et al., 2017; Rowlands, Yates, Davies, Khunti, & Edwardson, 2016), the

Euclidean Norm Minus One (ENMO) (van Hees et al., 2013b) was adopted to quantify

acceleration relative to gravity (1 mg = 0.00981 m/s -2), after which negative values

were rounded to zero. Raw data were further reduced by averaging the ENMO

values over 1 second epochs. All resulting values are expressed in milli (10 -3) gravity-

based acceleration units (mg), where 1g = 9.81 m/s2. Although the ENMO metric can

be sensitive to poor calibration (van Hees, et al., 2013b), GGIR autocalibrates the raw

triaxial accelerometer signal in order to reduce such calibration error (van Hees et

96

al., 2014). Where insufficient non-movement periods were available for auto-

calibration, back-up calibration coefficients derived from free-living data collected

with the same accelerometer units were used (Rowlands, Mirkes, Yates, Clemes,

Davies, Khunti, & Edwardson, 2017). The activPAL “Event” files provided the exact

time in seconds that posture change occurred and each second was classified as

sedentary, standing, or stepping. These files were then expanded using an Excel

formula to obtain second-by-second data, with each second subsequently classified

as either sedentary or not sedentary. These second-by-second activPAL files were

synchronised with the 1 s ENMO values from GA and AG. To exclude any transitional

movements, the middle two minutes of data from each three minute activity were

extracted and subsequently utilised for analysis. The full two minutes for each of the

stepping and walking activities were used.

4.2.5. Energy Expenditure

As the criterion measure for MVPA, oxygen consumption (VO2; ml·kg·min-1) was

measured with a portable indirect calorimetry system (MetaMax 3B-R2, CORTEX

Biophysik GmbH, Leipzig, Germany). The Metamax interface and breathing mask

(7600 Series V2, Hans Rudolph, Kansas) were set up and fitted as per the

manufacturer instructions. VO2 was measured using breath-by-breath mode and in

order to match time periods across devices, data were stored in second-by-second

intervals. These measurements were used to determine EE, which was then used to

classify activity intensity in METs. Resting EE was measured at visit 2 during the first

three lying/recumbent activities of the protocol. The observed mean resting EE was

2.89 ml·kg·min-1, which was used to define 1 MET. This value is comparable with

97

previous calibration studies in older adults (Barnett et al., 2016; Evenson et al., 2015;

Sergi et al., 2010; Siervo et al., 2014), and is consistent with the expected decrease in

resting metabolic rate (RMR) associated with ageing (Byrne, Hills, Hunter, Weinsier,

& Schutz, 2005; Kwan, Woo & Kwok, 2004). MVPA was defined as an intensity of 3

METs and above (e.g., ≥8.68 ml·kg·min-1) (Shephard, 2011).

4.2.5. Energy Expenditure

4.2.5.1. Cutpoint Calibration

A randomly counter-balanced sample of 12 female and five male participants from

visits 1 and 2 provided the calibration data. Descriptive statistics for all devices were

calculated for each activity in the protocol. The activPAL sedentary events and 3 MET

VO2 values were used as the criterion reference standards for SB and MVPA,

respectively. SB and MVPA were each coded as either 0 or 1, where 1 represented

the behaviour occurring and 0 represented the behaviour not occurring. Receiver

Operating Characteristic (ROC) curve analyses (Jago, Zakeri, Baranowski, & Watson,

2007) were used to determine SB and MVPA cutpoints. Area under the curve (AUC)

was calculated for each analysis as a measure of diagnostic accuracy. AUC values of;

≥ 0.90 are considered excellent, 0.80–0.89 good, 0.70–0.79 fair, and < 0.70 poor

(Metz, 1978). For each device two different pairs of cutpoints were generated by

analyzing combinations of sensitivity (Se) and specificity (Sp) on the ROC curves. Our

aim was to determine a cutpoint that accurately captured SB and MVPA (Se) whilst

limiting misclassification of SB and MVPA (Sp). Two approaches were adopted to

achieve this. Firstly, ENMO values that maximised both Se and Sp (Youden index)

98

(Perkins & Schisterman, 2006) were identified as one set of cutpoints (SB Youden and

MVPAYouden). The Youden index can however result in low positive predicted values

(Evenson et al., 2015) and it is recommended that researchers consider the relative

importance of Se and Sp (Welk, Laurson, Eisenmann, & Cureton, 2011), and

implications of the selected cutpoints on the biobehavioural impact on the outcome

variables (Mackintosh, Fairclough, Stratton, & Ridgers, 2012). Secondly, cutpoints

were determined that emphasised Se over Sp for SB cutpoints (SBSe) to minimise the

likelihood of classifying SB as PA, with Sp emphasised over Se for MVPA cutpoints

(MVPASp) to reduce the likelihood of misclassifying light PA as MVPA. Both cutpoints

reflected recommendations that the lower Se or Sp values should be ≥60% (Lugade,

Fortune, Morrow, & Kaufman, 2014). This prioritization approach minimises the risk

of individuals being misclassified in the target behaviour and is common in

accelerometer calibration (Landry et al., 2015; Mackintosh et al., 2012; Nero, Wallén,

Franzén, Ståhle, & Hagströmer, 2015) and fitness standards research (Welk et al.,

2011).

4.2.5.2. Cross-validation

To be consistent with good practice guidelines suggested by Welk (2005), SBYouden and

MVPAYouden, and SBSe and MVPASp cutpoints were cross-validated. The remaining 12

female and five male participants from visits 1 and 2 not included in the calibration

sample provided the cross-validation data. Two-by-two (2x2) contingency tables

were used to check classification agreement. The criterion measure and ENMO data

were first categorised into sedentary/not sedentary and MVPA/not MVPA binary

99

codes. Computed Se and Sp, Cohen’s kappa coefficients, and percentage agreement

between classifications were also assessed. Statistical analyses were performed

using IBM SPSS Statistics, version 24 (IBM, Armonk, NY), with the level of statistical

significance set at p ≤ 0.05.

4.3. Results

4.3.1. Participant characteristics

Among the 34 participants who completed the study, the mean (SD) height, weight,

and body mass index were 164.3 (1) cm, 71.7 (17.5) kg, and 26 (4.7) kg -1·m-2,

respectively. Mean (SD) maximal walking speed and blood pressure were 4.3 (1.5)

km·h-1, and 146/85 (22/11) mmHg, respectively. Further sample characteristics are

presented in Table 4.2. The mean (SD) accelerometer output from GA and AG

accelerometers (mg) are provided in Table 4.3 for each of phase of the laboratory

protocol.

Table 4.2. Study sample characteristics.

Sample Characteristics(n = 34)

Age (years)

Body Mass (kg)

Body Height (cm)

69.6 (8.0)

71.7 (17.5)

164.3 (1)

BMI (kg-1·m-2)

Blood Pressure (mmHg )

Maximal Walking Speed (km·h-1)

26.3 (4.7)

146/85 (22/11)

4.3 (1.5)

100

Values represent arithmetic mean (SD). SD, standard deviation; BMI, body mass index; mmHg, millimeters of mercury.

Table 4.3. Mean (SD) accelerometer output from GA and AG (mg) during each activity performed by older adults.

Lying Sitting Standing Household Stepping 65% Walk 75% Walk 85% Walk

GA Wrist

AG Hip

5.41 (1.47)

2.32 (2.15)

6.92 (5.61)

4.07 (2.10)

8.97 (4.34)

6.97 (1.90)

68.92 (18.30)

15.71 (4.50)

98.71 (78.21)

108.20 (19.22)

112.63 (88.10)

86.10 (19.80)

127.01 (86.45)

88.94 (31.50)

166.32 (97.40)

105.40 (34.90)

4.3.2. ROC Curve Analysis

ROC curve analysis revealed an AUC for the GA of 0.88 (95% CI: 0.87-0.88; P < 0.001)

for SB and 0.88 (95% CI: 0.87-0.88; P < 0.001) for MVPA. For the AG the AUC for SB

was 0.90 (95% CI: 0.90-0.91; P < 0.001), and 0.94 (95% CI: 0.94-0.95; P < 0.001) for

MVPA.

4.3.3. Cutpoint generation

The GA cutpoints which maximised both Se and Sp using the Youden Index were

SBYouden: 20 mg (Se = 94%, Sp = 72%) and MVPAYouden: 32 mg (Se = 88%, Sp = 77%). AG

cutpoints were 6 mg (Se = 85%, Sp = 82%) for SBYouden and 19 mg (Se = 86%, Sp = 92%)

for MVPAYouden. The cutpoints optimizing Se and Sp for GA were 57 mg (Se = 99%, Sp =

60%) for SBSe and 104 mg (Se = 60%, Sp = 89%) for MVPASp. Respective AG cutpoints

101

for SBSe and MVPASp were 15 mg (Se = 98%, Sp = 60%) and 69 mg (Se = 60%, Sp =

99%). Table 4.4 displays all cutpoints generated for both the GA and AG

accelerometers.

4.3.4. Cross-validation

The classification agreement, sensitivity, specificity and kappa coefficients between

calibration and cross-validation data for SB and MVPA cutpoints are shown in Table

4.4. GA SBYouden (Se = 47%, Sp = 92%) and MVPAYouden (Se = 76%, Sp = 76%) cutpoints

demonstrated moderate percentage agreement (73.1–76.2%) and moderate kappa

scores (0.42–0.52). AG SBYouden (Se = 47%, Sp = 92%) and MVPAYouden (Se = 76%, Sp =

76%) cutpoints demonstrated high percentage agreement (83.3–87.3%) and

moderate to substantial kappa scores (0.59–0.75). Comparatively, lower percentage

agreement and kappa scores were observed for both GA SBSe and MVPASp (67.2-

68.9%, k = 0.38-0.36), and AG SBSe and MVPASp (73.2-80.4%, k = 0.46-0.60) cutpoints,

respectively.

Table 4.4. Calibration cutpoints and cross-validation % agreement, kappa (k) and se and sp.

102

GA,.GENEActiv; AG, Actigraph; MVPA, moderate-to-vigorous physical activity; Se, sensitivity; Sp, specificity.

4.4. Discussion

This is the first study to determine GA wrist- and AG hip-worn raw acceleration

cutpoints for SB and MVPA in older adults. ROC curve analyses revealed that wrist

GA and hip AG accelerometer raw acceleration cutpoints provide good and excellent

discriminations of SB and MVPA, respectively. Results indicated that the SBYouden and

MVPAYouden cutpoints of 20 mg and 32 mg for GA, and 6 mg and 19 mg for AG yielded

the greatest classification accuracy. Such cutpoints are low compared to existing

ENMO adult SB cutpoints for GA wrist- (45.8 mg) and AG hip-worn (47.4 mg)

accelerometers (Hildebrand et al., 2016), and MVPA cutpoints for GA wrist-worn

accelerometers in adults (93 mg; Hildebrand et al., 2014) and older adults (100 mg;

Menai et al., 2017), and AG hip-worn accelerometers in adults (69.1 mg; Hildebrand

103

Cutpoint (mg)Cross-Validation

% Agreement

Cross-Validation

Kappa

Cross-Validation

Se (%)

Cross-Validation

Sp (%)

GA

SBYouden

MVPAYouden

≤ 20

≥ 32

73.1

76.2

0.42

0.52

47

76

92

76

AG

SBYouden

MVPAYouden

GA

SBSe

MVPASp

AG

SBSe

MVPASp

≤ 6

≥ 19

≤ 57

≥ 104

≤ 15

≥ 69

83.3

87.3

67.2

68.9

73.2

80.4

0.59

0.75

0.38

0.36

0.46

0.60

62

86

43

81

48

94

93

89

99

65

97

74

et al., 2014). Consequently, the SBYouden and MVPAYouden cutpoints may underestimate

SB and overestimate light PA when applied in free-living environments, resulting in

participants being falsely classified as being physically active when they are more

likely to be sedentary.

The alternative SBSe and MVPASp cutpoints were more comparable to values reported

previously (Hildebrand et al., 2014; Hildebrand et al., 2016; Menai et al., 2017). The

Se values of 99% for GA SBSe and 98% for AG SBSe cutpoints ensure that almost all

older adults who are sedentary have ENMO values below the established cutpoints,

and therefore have a very low risk of being misclassified as being physically active.

The de-emphasis on Sp (e.g., % of older adults correctly identified as not being

sedentary) acknowledges the risk that a proportion (up to 40%) of older adults who

are physically active may be classified as sedentary. However, SB is an identifiable

risk factor affecting physical (e.g., premature mortality, chronic diseases and all-

cause dementia risk) and psychosocial (e.g., self-perceived QoL, wellbeing and self-

efficacy) determinants of health (Edwards & Loprinzi, 2016; Falck et al., 2016; Lewis,

Napolitano, Buman, Williams, & Nigg, 2017) independent of PA (Tremblay et al.,

2017). Such misclassification is likely to be beneficial if already physically active older

adults were offered further opportunities to take part in interventions to reduce SB

and increase PA levels (Chastin et al., 2017). Conversely, the higher Sp values for the

GA MVPASp cutpoint (89%) and AG MVPASp cutpoint (99%) ensures that older adults

not engaged in MVPA (e.g., ENMO values below the established cutpoints) are not

falsely classified as being in MVPA and are correctly identified and targeted for PA-

104

promoting interventions (Lyons, Swartz, Lewis, Martinez, & Jennings, 2017). The de-

emphasis on Se suggests that up to 40% of older adults who are not in MVPA could

have ENMO values above this cutpoint (due to the lower true positive rate, relative

to the true negative rate (Sp)), and therefore their behaviour could be misclassified

as MVPA (Nero et al., 2015). However, it is more likely to be harmful for an older

adult to be wrongly classified as active rather than asking active older adults to take

part in additional MVPA. Hence, the goal of the MVPA cutpoint to identify older

adults who may have increased health risks due to being below this cutpoint (by

favouring Sp over Se) appears to be justified (Nero et al., 2015; Welk, Going,

Morrow, & Meredith, 2011).

Given that acceleration magnitudes are significantly lower for the AG GT3X+ relative

to the GA (John, Sasaki, Staudenmayer, Mavilia, & Freedson, 2013; Rowlands et al.,

2015; Rowlands et al., 2016), the higher wrist-worn cutpoints relative to the hip-

worn cutpoints were consistent with those observed previously (Rowlands et al.,

2015; Stiles, Griew & Rowlands, 2013). Our protocol was comparable to previous

calibration studies implemented in controlled settings (de Almeida Mendes et al.,

2017). However, laboratory calibration protocols rely on small deliberate increases in

PA intensities and movement patterns within a limited period of time, compared to

free-living activities over extended periods (van Hees, Golubic, Ekelund, & Brage,

2013a). Such protocols cannot fully reflect daily SB and PA patterns, and this may

limit the accuracy of the SB and MVPA thresholds obtained for wrist- and hip-worn

devices when they are applied in free-living environments (Van Hees et al., 2013a).

105

One of the main decisions to be made by researchers using either raw acceleration

or count-based outcomes is monitor placement location (de Almeida Mendes et al.,

2017). After comparing SB and PA estimates from wrist- and hip-worn monitors with

EE, Rosenberger et al. (2013) concluded that SB and MVPA classification accuracy

was superior for the hip-worn devices. Our cross-validation results support this due

to the superior percent agreement and kappa scores for the hip-worn AG over the

wrist-worn GA in classifying both SB and MVPA. However, our results also

demonstrate that wrist-worn accelerometers can provide accurate estimates of SB

and MVPA and the subsequent cutpoints performed reasonably well at

discriminating both SB and MVPA (Troiano et al., 2014). Indeed, a recent systematic

review of raw acceleration calibration studies reported no evidence of meaningful

differences in the accuracy of wrist- and hip-worn accelerometers (de Almeida

Mendes et al., 2017). Considering the superior wear compliance associated with

wrist-worn devices (Doherty et al., 2017), this attachment site may be the most

suitable location during free-living protocols. Consequently, the wrist-worn GA SBSe

and MVPASp cutpoints will be adopted in Chapter 5 (Study 3) to aid in answering

objectives 5 and 6 of the thesis.

Several strengths and limitations should be noted when interpreting the results of

this study. A main strength was the use of raw acceleration data from two commonly

used devices positioned at wrist and hip wear sites. These cutpoints will be of utility

to researchers using the raw data capabilities of the GA and current AG

106

accelerometers to study SB and PA in older adults. Rigorous best practice analytical

procedures were also adopted to calibrate and cross-validate the cutpoints (Welk,

2005), which were specific to adults aged 60 years and over. Resting EE was directly

measured to allow a sample-specific interpretation of 3 METs as the MVPA

threshold, and used a validated separate criterion measure for SB (Kim, Barry &

Kang, 2015). There were also a number of limitations. The sample may not have

been representative of the wider older adult population in respect of their fitness

status and motivation to engage in PA, as recruitment included a convenience

sample of healthy older adults who answered advertisements and showed an

interest in the study representing a broad age range. Furthermore, specific older

adult populations (e.g., those with chronic diseases and impaired mobility) may

require different SB and PA cutpoints (Landry et al., 2015) that reflect differences in

RMR and energy cost during ambulatory PA across this age group (Miller, Strath,

Swartz, & Cashin, 2010). Moreover, activities that replicate everyday movements

and tasks performed by older adults were incorporated. However, it is recognised

that the laboratory setting limits the ecological validity of the resultant data

(Hildebrand et al., 2016). Lastly, cross-validation was performed using the same

laboratory protocol rather than using data collected from a free-living or a simulated

free-living protocol (Welk, 2005). It was felt that the challenges associated with

having the participants wear the gas analysis system for an extended period in free-

living situations were too great to warrant taking this approach. Consequently, the

cutpoints obtained should be further cross-validated with independent samples,

ideally from other settings and within free-living environments (Welk, 2005).

107

4.5. Conclusion

In conclusion, cutpoints varied dependent upon attachment site, with the wrist-worn

GA cutpoints higher than those for the hip-worn AG. The identified GA and AG SBSe

and MVPASp cutpoints can enable researchers to classify older adults as engaging in

SB or not engaging in MVPA with an acceptable degree of confidence. Further cross-

validation research is needed to test the utility of these cutpoints in independent

samples within free-living environments.

Thesis Study Map





Objectives





the community.




initiative.

Key Findings:


108









adults.




adults.

Objectives









Key Findings

When optimizing Sensitivity for sedentary

behaviour and Specificity for moderate-to-

vigorous physical activity, wrist-worn GENEActiv

accelerometer cutpoints of 57 mg and 104 mg

were generated for sedentary behaviour and

moderate-to-vigorous physical activity,

respectively.

For the hip-worn ActiGraph GT3X+ the cutpoints

were 15 mg and 69 mg for sedentary behaviour

and moderate-to-vigorous physical activity,

respectively.

The resultant cutpoints can enable researchers to

classify older adults as engaging in sedentary

behaviour or not engaging in moderate-to-

vigorous physical activity with an acceptable

degree of confidence.

Study 3. Physical activity, sedentary

behaviour, perceived health and fitness, and

psychosocial wellbeing among community-

Objectives

To investigate gender, age, and socio-economic

status differences in older adults’ sedentary

109

dwelling older adults. behaviour, physical activity and self-reported

health indicators.

To examine associations between sedentary

behaviour and physical activity with self-reported

health indicators.




dwelling older adults

Chapter 5

Study 3: Physical activity, sedentary behaviour, perceived health and fitness, and psychosocial

wellbeing among community-dwelling older adults.

110

5.1. Introduction

Chapter 4 (Study 2) determined laboratory-based wrist-worn GA and hip-worn AG

GT3X+ raw acceleration cutpoints for SB and MVPA in older adults. The current study

progresses the work of Chapter 4 (Study 2) by implementing the wrist-worn GA

cutpoints within free-living environments.

Performing sufficient PA is a primary modifiable determinant of health (Birkel et al.,

2015) and recent research has demonstrated its potential to benefit an array of

physical (Zhu et al., 2017) and psychosocial (Devereux-Fitzgerald et al., 2016; Franco

et al., 2015; Greaney et al., 2016) determinants of health in older adults. There is

111

also growing public health interest in the amount of time spent in SB; defined as

waking behaviours in a sitting, reclining or lying posture with EE ≤1.5 METs (Tremblay

et al., 2017). Sociodemographic attributes including gender (Greaney et al., 2016),

age (Heo, Chun, Kim, Ryu, & Lee, 2017) and SES (Gray et al., 2015) have been shown

to affect both SB and PA levels in older adults. Accumulating evidence suggests that

prolonged episodes of SB have similar physical (e.g., premature mortality, chronic

diseases and all-cause dementia risk) and psychosocial (e.g., QoL and SEE) risk

factors to that of physical inactivity (Edwards & Loprinzi, 2016; Falck et al., 2016; Kim

et al., 2016; Lewis et al., 2017). Consequently, physical inactivity in combination with

prolonged periods of SB further compound negative physical (Haywood et al., 2018)

and psychosocial (Biswas et al., 2015; Pulsford, Stamatakis, Britton, Brunner, &

Hillsdon, 2015) health outcomes. The high levels of SB and low levels of MVPA in this

population are concerning given their negative associations with self-reported health

(Beyer, Wolff, Warner, Schüz, & Wurm, 2015), fitness (Kuosmanen et al., 2016) and

psychosocial outcomes such as QoL and SEE (French, Olander, Chisholm, &

McSharry, 2014; Greaney et al., 2016; Kim et al., 2016; Olson et al. 2016).

Interventions should focus with not only the single effect of either SB or PA level, but

with the balance of both (ten Brinke et al., 2015). To achieve this requires accurate

and reliable measurement of SB and PA patterns over time (Greaney et al., 2016).

Health comprises not only physical, but also psychological and social components,

and it is therefore important to take self-rated measurements into consideration

when evaluating health status (Kuosmanen et al., 2016). Self-report questionnaires

concerning SRH and SAPF are suitable ways of obtaining important information

112

about a person’s overall health and functional status (Kuosmanen et al., 2016) and

are strongly correlated with objective assessments of health and fitness (Meng, Xie &

Zhang, 2014; Wu et al., 2013). Among older adults both concepts have been found to

be influenced by sociodemographic attributes including gender, age, and SES (Bamia

et al., 2017; Meyer, Castro-Schilo & Aguilar-Gaxiola, 2014), as well as being positively

associated with PA level (Beyer et al., 2015; Haywood et al., 2018), and negatively

associated with SB (Haywood et al., 2018). Specifically, those who are physically

active, male, younger-old (60 to 69 years), and of higher SES are more likely to report

favourable ratings of health and physical fitness (Bamia et al., 2017; Kuosmanen et

al., 2016). Further research in more sociodemographically diverse older populations

is warranted to improve understanding of the relationship between gender, age, and

SES and, self-reported physical and psychosocial outcomes, as well as the

independent factors affecting them such as SB and PA levels (Kuosmanen et al.,

2016). This study examined SB and PA levels assessed by self-report and

accelerometry. The aims related to thesis objectives 5 and 6 were firstly to;

investigate gender, age, and socio-economic status differences in older adults’ SB, PA

and self-reported health indicators, and secondly to examine associations between

SB and PA with self-reported health indicators.

5.2. Methods


This cross-sectional study was undertaken between January 2016 to December 2017

as the baseline phase of Sport England’s GHGA PA programme. GHGA was a three-

113

year project aimed at engaging inactive older adults in PA at least once a week for 30

minutes, via a 12 week PA intervention. The project was funded by Sport England

and delivered by Sefton Borough Council. A full outline of the GHGA programme is

presented in Chapter 6.

A homogenous purposive sample of 380 older adults were approached whilst

attending the GHGA sessions throughout Sefton Borough in north-west England.

Sefton was appropriate for participant recruitment because compared to the UK

national average it is recognised as having a higher percentage of male and female

older adults over 65 (30% of the total population), and over 85 (6% of the total

population), and the highest percentage of inactive over 65 year olds (80%) (ONS,

2017). Furthermore, Sefton has the highest National Health Service costs associated

with physical inactivity (Sport England, 2014; Sport England, 2015). Sefton is also

characterised by large differences in SES according to the Indices of Multiple

Deprivation (IMD) (Public Health England, 2017). A total of 318 older adults

consented to take part (83.7% recruitment rate), with 207 older adults (164 female;

43 male; 54.5% participation rate), aged 65 to 102 years (mean age of participants

=77.8, SD =7.7) meeting the inclusion criteria determined by Sport England as

funders of the GHGA programme. These criteria required participants to reside

within Sefton Borough, be ≥65 years of age, be without physical and/or intellectual

disabilities which prevented written informed consent being provided, and be

physically inactive as indicated by the Single Item Physical Activity Question

(Donaldson, 2004). The Single Item Physical Activity Question was a mandatory Sport

114

England screening tool for the GHGA programme to determine participants’ activity

levels. This question asks participants,

“In the past week, on how many days have you done a total of 30 minutes or more PA, which was enough to raise your breathing rate? This may include sport, exercise and brisk walking or cycling for recreation or to get to and from places, but should not include housework or PA that may be part of your job.”

Only individuals who answered “0 days” (inactive) were deemed to have met this

inclusion criterion. The Single Item Physical Activity Question has demonstrated

sound psychometric properties (Milton, Bull & Buamn, 2010) including good test-

retest reliability agreement (kappa = 0.63, 95% confidence interval = 0.54 to 0.72),

and modest concurrent validity (r = 0.53) against the Global Physical Activity

Questionnaire (Armstrong & Bull, 2006).

Participants invited to participate in the programme received a covering letter,

participant information sheet, and consent form. Before the study commenced,

institutional ethical approval was received (#SPA-REC-2015-329) and written

informed consent was obtained for all participants prior to participation.

Participation was voluntary with no incentives provided.

5.2.2. Primary Outcome Measures

5.2.2.1. International Physical Activity Questionnaire for the Elderly

To assess SB and PA levels, the IPAQ-E (Hurtig-Wennlöf et al., 2010) was adopted as

required by the funder. IPAQ-E is based on the short version of the IPAQ

115

(www.ipaq.ki.se) and assesses time spent sitting, walking in bouts of 10 minutes or

more (labelled as Walk10min herein), MPA in bouts of 10 minutes or more (labelled as

MPA10min herein), and VPA in bouts of 10 minutes or more (labelled as VPA10min herein)

during the previous 7-days. The categorical outcome from IPAQ-E assigns the

participants into one of three PA categories (e.g., low, moderate, or high-PA). The

IPAQ-E provides favourable levels of both direct and indirect levels of criterion

validity for sitting (Spearman r = 0.28, P < 0.05), Walk10min (Spearman r = 0.35, P <

0.01), MPA10min (Spearman r = 0.40, P < 0.01), and VPA10min (Spearman r = 0.37, P <

0.01) (Hurtig-Wennlöf et al., 2010). However, varying levels of test-retest reliability

(intraclass correlation ranging from 0.30 to 0.82) have also been reported (Tomioka

et al., 2011).

5.2.2.1. Wrist-based Accelerometer (GA)

A sample of 101 participants (75 female; 26 male), aged 65 to 90 years (mean age

=77, SD=7.1) wore a triaxial GA accelerometer (ActivInsights Ltd., Kimbolton,

Cambridgeshire, United Kingdom) on their non-dominant wrist for seven days, 24

hours per day. Objective measures of PA such as accelerometers are commonly

adopted as methods of monitoring older adults’ SB and PA levels (Evenson et al.,

2015; Mañas et al., 2017; Parsons et al. 2017; Thornton et al. 2017; Wullems et al.,

2017). Accelerometers are particularly appropriate for assessing PA in older adults as

these devices require no input from the participant over the data collection period,

and superior wearer compliance has been demonstrated when compared to younger

age groups (Doherty et al., 2017). Given the complexity of SB and PA constructs, and

the increasing adoption of accelerometers within surveillance, epidemiology, clinical,

116

and intervention research, it is recommended that research in this population adopts

both objective and self-report measures to investigate and provide evidence for

participants’ SB and PA. Adopting both methods enables researchers to more fully

characterise individuals SB and PA patterns and identify potential intervention

targets for increasing activity (Shiroma, Schrack & Harris, 2018).

The GA is a water-proof device which measures raw accelerations expressed in

gravitational equivalent units at a range of -8 g and 8 g. Acceleration values are

digitised by a 12-bit analog-to-digital converter and the devices were set to record

accelerations at a frequency of 60 Hz. The GA accelerometer has demonstrated

excellent technical reliability as well as criterion and concurrent validity in adult

populations (Esliger et al., 2011). Participants were provided with a leaflet outlining

how and when to wear the devices prior to participation. The first author was

responsible for attaching and collecting the accelerometers at a time and location

most suitable to the participant. A log sheet was provided for each participant to

record any times that the device was removed and subsequently replaced. GA data

were downloaded using GA PC software version 2.9 and saved in raw format as

binary files. Signal processing of raw GENEActiv.bin files was completed offline using

R-package GGIR version 1.5 (https://cran.r-project.org/web/packages/GGIR/) (van

Hees et al., 2013). This R-package facilitates data cleaning and the extraction of user-

defined acceleration levels, which can then be set to reflect the intensity levels.

Concurrent with previous studies (Hildebrand et al., 2014; Fairclough et al., 2016;

Menai et al., 2017; Rowlands et al., 2016), the Euclidean Norm Minus One (ENMO)

metric (e.g., signal vector magnitude of the three axes; g = √x2 + y2 + z2 -1 g) (van

117

Hees et al., 2013) was adopted to quantify acceleration relative to gravity, after

which negative values were rounded to zero. Raw data were further reduced by

averaging the ENMO values expressed as over 1 s epochs. All resulting values were

expressed in milli (10-3) gravity-based acceleration units (mg), where 1g = 9.81 m·s2.

Although the ENMO metric can be sensitive to poor calibration (van Hees et al.,

2013), GGIR autocalibrates the raw triaxial accelerometer signal in order to reduce

such calibration error (van Hees et al., 2013). A valid day was defined as at least 10

hours wear-time during waking hours, with at least 3 valid days required for

inclusion in the analysis. Waking hours are not significantly different between the

young-old, middle-old and old-old (Valenti, Bonomi & Westerterp, 2017) and were

defined as being between 07:00 and 23:00. For all participants with usable data,

mean daily time spent in total accumulated 1 second bouts of SB and MVPA were

established by applying ENMO cutpoints of ≤57 mg (SB/light-PA) and ≥104 mg

(MVPA) generated from a previously conducted calibration study in older adults as

part of the GHGA project (unpublished data). Given that both US (CDC, 2015) and UK

(Department of Health, 2011a) derived PA guidelines recommend that over a week,

older adults should accumulate up to at least 150 minutes of MVPA10min, daily mean

MVPA derived only from bouts of at least 10 minutes above the MVPA cutpoint were

also reported. Total accumulated 1 second bouts of SB and MVPA, MVPA10min, and

mean accelerations (mg·day-1) were the main outcomes in the GA analyses.

5.2.3. Secondary Outcome Measures

5.2.3.1. Self-Assessment of Physical Fitness

118

To assess self-perceived fitness, the SAPF (Weening-Dijksterhuis, de Greef, Krijnen, &

van der Schans, 2012) was adopted. The questionnaire asks three questions

including: How do you rate your strength?; How do you rate your aerobic

endurance?; and How do you rate your balance? The SAPF uses a rating method from

0 (indicating the lowest rating) to 10 (indicating the highest rating) for each of the

three items. Sound psychometric properties have been demonstrated for the SAPF

scale among frail older adults with acceptable internal consistency (Cronbach alpha

=0.71) (Weening-Dijksterhuis et al., 2012), one week test-retest validity (0.70), and

moderate concurrent validity against the Groningen Fitness Test for the Elderly

(Lemmink, 1996).

5.2.3.2. Kemp Quality of Life Scale

To assess QoL, the single item Kemp Quality of Life Scale (Kemp & Ettelson, 2001)

was adopted. Following the question: Thinking about the good and bad things that

make up your quality of life, how would you rate the quality of your life as a whole? ,

respondents are asked to tick the box next to the answer that best describes their

QoL on a 7-item scale anchored by “so good, it could not be better” (score of 1) to

“so bad, it could not be worse” (score of 7). Although there is no published

information about the distribution or error of data obtained using this scale, the

scale is interval level, can provide data for parametric and non-parametric analysis,

and has been adopted previously in older adult populations (Siebens, Tsukerman,

Adkins, Kahan, & Kemp, 2015; Roe et al., 2011).

5.2.3.3. Self-Rated Health Questionnaire

119

The SRH (Sargent-Cox, Anstey & Luszcz, 2010) is a 3-item questionnaire assessing

global, age-comparative, and self-comparative health status (HS). Global HS is

measured via the question: How would you rate your overall health at the present

time?. Participants answer on a five point scale (1 = excellent; 5 = poor). Age-

comparative HS is measured on a three-point scale via the question: Would you say

your health is: (1) better, (2) about the same, or (3) worse than most people your

age?. Self-comparative HS is also measured on a three point scale via the question:

Is your health now… (1) better, (2) about the same, or (3) not as good as it was 12

months ago?. Previous research has shown good prognostic validity for mortality,

and high internal reliability (α =.75) (Ferraro & Wilkinson, 2013).

5.2.3.3. Self-Efficacy for Exercise Scale

The SEE (Resnick & Jenkins, 2000) is an 11-item scale designed to assess confidence

to continue exercising in the face of perceived barriers. The SEE scale consists of 11

situations that might affect participation in exercise. Items are rated from 0 (not

confident) to 10 (very confident). The mean score of numerical ratings from each

response indicates the strength of efficacy expectations. The SEE scale has

demonstrated excellent internal consistency (Cronbach α =0.92) and significant

predictive validity of exercise activity in older adults when controlled for age and

gender (F = 78.8; p < 0.05) (Resnick & Jenkins, 2000; Resnick, Luisi, Vogel, &

Junaleepa, 2004).

All measures were administered by the first author to each participant whilst

attending the GHGA sessions. Measures took no longer than 30 minutes to complete

120

per participant. Regular breaks were incorporated if needed, to reduce fatigue.

Additional visits were also arranged if needed, to aid completion of a full dataset due

to participant time constraints. All participants recorded their age, gender and home

postal code. Ages were categorised as 65 to 69 years, 70 to 79 years, and ≥80 years.

Postal codes were used to estimate SES by generating Indices of IMD (Department

for Communities and Local Government, 2015) using an online conversion tool

(http://imd-by-postcode.opendatacommunities.org/). The IMD is a UK government

metric used to rank area-level deprivation within and between different

communities. The IMD scores rank each super output area in England from 1 (most

deprived area) to 32,844 (least deprived area). Super output areas were then split

into 3 equal groups to represent low (1 to 10,947), middle (10,948 to 21,895), and

high-SES (21,896 to 32,844). Questionnaire data were input manually into Microsoft

Excel 2013 and checked for missing data. All 207 participants who consented to take

part were included in the final analytical sample. Confidentiality and data storage

procedures were adhered to as is set out in Edge Hill University’s research data

management (Edge Hill University, 2017) and code of practice for the conduct of

research guidelines (Edge Hill University, 2017). All participant data was anonymised

and coded to prevent identification, and securely stored using password-protected

files on the Edge Hill University computing network. The server hosting the files was

backed up every four hours and nightly to tape, to ensure data attrition was

minimised. Only research team members had access to the anonymised data. This

was shared between the team strictly for the purposes of research.

5.2.4. Statistical Analysis

121

Primary outcome measures were self-reported IPAQ outcomes and GA wrist-based

accelerometer-derived SB and MVPA. To address objective 5, multivariate analysis of

covariance (MANCOVA) assessed differences between the IPAQ outcomes and the

independent variables of gender, age category (e.g., 65 to 69, 70 to 79 and ≥80

years), and SES-group (e.g., low-SES, middle-SES and high-SES). Season of data

collection (e.g., spring/summer and autumn/winter) was included as a covariate.

Analysis of covariance (ANCOVA) assessed differences in accelerometer outcomes

between the independent variables. Accelerometer wear time (min‧d-1) was included

as a covariate for SB only, with season of data collection (spring/summer and

autumn/winter) included as a covariate. The secondary outcomes were QoL, SRH,

SAPF and SEE. ANCOVAs examined gender, age category, and SES-group differences

in each secondary outcome and season of data collection was included as a covariate

in each analysis (Prins & van Lenthe, 2015). To address objective 6, stepwise multiple

regression analyses were used to examine associations between secondary

outcomes with each of the primary outcomes. Effects were considered significant at

the p < 0.05 level. All data analyses were performed using IBM SPSS Statistics for

Windows version 22.0 (IBM Corp, Armonk, NY).

5.3. Results

Among the 318 older adults who consented to participate in the study, 111 (97

female; 14 male) were identified as being “active” as per the single item PA

screening measure (Donaldson, 2004). There was no missing data among the

remaining participants. The resulting final analytical sample consisted of 207

participants (mean age= 77.8, SD =7.7; 65.1% compliance rate). Among the sample of

122

111 participants wearing a GA accelerometer, 10 participants (eight female; two

male) did not satisfy the wear time criteria and so were excluded from the analysis.

The resulting final analytical sample consisted of 101 participants (mean age= 77, SD

=7.1 years; 91% compliance rate). Characteristics of the questionnaire- and

accelerometer-wearing samples were comparable in terms of gender (male = 20.3%;

23.7% men), age (65 to 69 years =15.5%; 18.8%, 70 to 79 years =44.7%; 45.5%, ≥80

years =39.8%; 36.6%) and SES-group (low-SES =35.9%; 31.6%, middle-SES =24.3%;

21.8%, high-SES =39.8%; 46.5%) splits, respectively. Missing data among the

accelerometer-wearing sample was accrued through a variety of reasons including

skin irritation (three participants), concern over device impact on general health

(two participants) and forgetting to wear the monitor after removing it before

bedtime (five participants). There were no significant differences for any of the

measured variables between participants included in the analyses and those

excluded. Descriptive statistics are shown in Tables 5.1 and 5.2 for the questionnaire-

and accelerometer-wearing samples, respectively.

Table 5.1. Descriptive characteristics of the participants

Self-report data Total sample(n =207)

Age years (SD)

Gender n (%)

Female

Male

Age Category n (%)

60-69 yrs

77.8 (7.7)

165 (79.7)

42 (20.3)

32 (15.5)

123

70-79 yrs

80+ yrs

92 (44.7)

83 (39.8)

Season of data collection n (%)

Autumn/Winter

Spring/Summer

SES Group n (%)

Low-SES

Mid-SES

High-SES

86 (41.7)

121 (58.3)

74 (35.9)

51 (24.3)

82 (39.8)

SD = Standard Deviation; SES = Socioeconomic status.

Table 5.2. GENEActiv wrist-worn accelerometer data descriptives.

GENEActiv data Total Sample (n =101)

Age (years)

Gender n (mean age ± SD)

Female

Male

Age Category n (%)

60-69 yrs

77.0 (7.1)

77 (76.2)

24 (23.8)

18 (17.8)

124

70-79 yrs

80+ yrs

46 (45.5)

37 (36.6)


Autumn/Winter

Spring/Summer

SES Group n (%)

Low-SES

Mid-SES

High-SES

49 (48.5)

52 (51.5)

32 (31.6)

22 (21.8)

47 (46.5)

SD = Standard Deviation; SES = Socioeconomic status.

5.3.1. Primary Outcomes

Table 5.3 shows time spent in self-reported sitting, PA behaviours, as well as

secondary outcome data for the questionnaire sample. No significant differences

between gender, age category and SES-group were present. Men spent more time

sitting (3073.1 vs 2835.3 min‧w-1) and engaged in more MPA10min (151.9 vs 116.2 min‧

w-1) than women, respectively. Women spent more time Walk10min (333.5 vs. 235.8

min‧w-1) than men and reported spending more time in MVPA10min per day than men

(66.4 vs 48.1 min‧d-1). There was no significant difference between older adults

achieving PA guidelines (57% of women; 47.6% of men) and those not achieving PA

guidelines.

125

Total sample(n =207)

Female (n =165)

Male (n =42)

Young-old (n =32)

Middle-old (n =93)

Old-old (n =82)

Low-SES (n =75)

Mid-SES (n =50)

High-SES (n =82)

Sitting (min‧w-1) 2883.7 (1137.2)

2835.3 (1118.0)

3073.1 (1204.3)

2677.2 (1028.4)

3048.8 (1188.3)

2793 (11110.9)

2951.8 (1118.5)

2728.3 (1075.7)

2932.3 (1195.7)

Walk10min (min‧w-1) 313.5 (483.5) 333.5 (512) 235.8 (345.1) 257.8 (375.1) 335.7 (473.0) 316.5 (534.2) 368.8 (639.2) 274.7 (315.7) 293.1 (396.7)

MPA10min (min‧w-1)

MVPA10min (min‧d-1)

Meeting Guidelines n (%)

YES

NO

123.5 (334.9)

62.7 (92.2)

114 (55.3)

92 (44.7)

116.2 (326.8)

66.4 (95.4)

94 (57)

71 (43)

151.9 (367.4)

48.1 (77.7)

20 (47.6)

22 (52.4)

85.6 (171.8)

31.9 (33.3)

6 (18.8)

26 (81)

131.6 (362)

68.3 (94.7)

15 (16.3)

77 (83.7)

127.6 (350.9)

68.3 (102.8)

14 (16.9)

69 (83.1)

84.9 (224.1)

66.6 (105)

57 (77)

17 (23)

86 (223.2)

46.2 (67)

9 (17.6)

42 (82.4)

180.2 (449.9)

69.2 (92.9)

15 (18.3)

67 (81.7)

Quality of Life 3.2 (.77) 3.2 (.77) 3.4 (.78) 3 (.78) 3.2 (.80) 3.2 (.75) 3.3 (.76) 3 (.71) 3.1 (.80)

Self-Assessment of Physical Fitness

9.8 (3.48) 9.4 (3.4) 11.3 (4) 9.8 (3.7) 9.6 (3.6) 10.1 (3.2) 8.8 (3.5) 10.7 (3.2) 10.2 (3.4)

Self-Rated Health 8.1 (1.5) 8.2 (1.4) 8.3 (1.2) 8.3 (1.4) 8.2 (1.3) 8.2 (1.4) 8.2 (1.3) 8.1 (1.3) 8.2 (1.4)

Self-Efficacy for Exercise

29.8 (10.1) 28.8 (10) 33.6 (9.56) 29.7 (11.1) 28.7 (10.1) 30.9 (9.7) 28.2 (9.4) 30.7 (10.7) 30.6 (10.3)

Table 5.3. Self-reported physical activity and psychosocial outcome measures.

*.= Significant difference; Data are mean (SD) unless otherwise stated; Young-old = 65-69 years; Middle-old = 70-79 years; Old-old = 80+ years; Walk 10min = Walking in ≥ 10-minute bouts; MPA10min = Moderate-physical activity in ≥ 10-minute bouts; MVPA10min = Moderate-to-vigorous physical activity in ≥ 10-minute bouts. Note. Lower score = more favourable result for Quality of Life and Self-Rated Health.

127

Table 5.4 shows time spent in SB and PA for the accelerometer-wearing sample.

Mean total wear time was 1244.7 min‧d-1 in men and 1222.5 min‧d-1 in women. Time

spent in SB (771.7 vs. 773.9 min‧d-1), LPA (113.7 vs. 113.3 min‧d-1), MVPA (74.7 vs.

72.8 min‧d-1), MVPA10min (7.8 vs 8.4 min‧d-1) and mean acceleration (26.3 vs 25.4 mg‧d-

1) did not differ between women and men, respectively. Overall, there was no

significant difference in the proportion of older adults recorded as achieving PA

guidelines from total accumulated 1 second bouts of MVPA (89.6% of women, 79.2%

men) and MVPA10min (11.7% of women, 16.7% of men).

No significant gender, age category, and SES-group differences were observed

between IPAQ outcomes and accelerometer-derived SB and PA.

128

Table 5.4. GENEActiv SB and physical activity outcomes.

Total sample(n =101)

Female (n =77)

Male (n =24)

Young-old (n =18)

Middle-old (n =46)

Old-old (n =37)

Low-SES (n =32)

Mid-SES (n =22)

High-SES (n =47)

Total wear time (min‧d-1)

Mean acceleration (mg‧d-1)

SB (min‧d-1)

1227.8 (142.6)

26.1 (7.9)

772.0 (69.0)

1222.5 (138.8)

26.3 (7.5)

771.7 (69.7)

1244.7 (155.9)

25.4 (9.3)

773.9 (67.9)

1206.7 (194.6)

24.5 (7.7)

773 (52.5)

1256.7 (114.7)

24.6 (7.8)

779.1 (68.6)

1202.1 (141.9)

28.6 (7.7)

763.3 (76.8)

1227.9 (143.4)

23.1 (7.4)

763.4 (62.2)

1187.8 (175.3)

29.1 (7.4)

784.9 (72.7)

1246 (123.2)

26.7 (7.9)

772.3 (72)

LPA (min‧d-1)

MVPA (min‧d-1)

113.5 (36.2)

74.2 (39.4)

113.7 (37.5)

74.7 (39.3)

113.3 (32.3)

72.8 (40.6)

118 (35.6)

69 (27.5)

110.4 (35.9)

70.5 (38.6)

115.3 (37.6)

81.4 (44.8)

118.8 (34.2)

77.8 (37.1)

104.5 (36.7)

70.6 (42.1)

114.3 (37.3)

73.4 (40.3)

6.2 (8.1)

41 (87.2)

6 (12.8)

4 (8.5)

43 (91.5)

MVPA10min (min‧d-1)

Meet MVPA Guidelines n (%)

YES

NO

Meet MVPA10min n (%)

YES

NO

7.7 (9.1)

88 (87.1)

13 (12.9)

13 (12.9)

88 (87.1)

7.8 (9.3)

69(89.6)

8 (10.4)

9 (11.7)

68 (88.3)

8.4 (8.9)

19 (79.2)

5 (20.8)

4 (16.7)

20 (83.3)

3.1 (4.4)

17 (94.4)

1 (5.6)

0 (0)

18 (100)

9.9 (9.8)

40 (87)

6 (13)

8 (17.4)

38 (82.6)

7.9 (9.4)

31 (83.8)

6 (16.2)

5 (13.5)

32 (86.5)

9.9 (10.4)

29 (90.6)

3 (9.4)

5 (15.6)

27 (84.4)

9 (9.2)

18 (81.8)

4 (18.2)

4 (18.2)

18 (81.8)

*.= Significant difference; Data are mean (SD) unless otherwise stated; Young-old = 65-69 years; Middle-old = 70-79 years; Old-old = 80+ years; SB = Sedentary behaviour; LPA = Light physical activity; MVPA = moderate-to-vigorous physical activity; MVPA10min = Moderate-to-vigorous physical activity in ≥ 10-minute bouts.

129

5.3.2. Secondary Outcomes

Mean ± SD scores for QoL (3.2 ± 0.77 vs 3.1 ± 0.78), SAPF (9.4 ± 3.4 vs 11.3 ± 3.53),

SRH (8.2 ± 1.38 vs 8.3 ± 1.22), and SEE (28.8 ± 10 vs 33.6 ± 9.56) did not differ

significantly between women and men, respectively. Significant gender, age category

and SES-group (F(4, 188) =3.440, p =0.01) differences were observed with QoL, and

gender and age category (F(2, 188) =3.899, p =0.022) differences with SRH. No

significant post-hoc pairwise comparisons were present for these results. SAPF score

was 17.6% higher for males compared to females (F(1, 201) =7.893, p =0.005) and

26.5% and 11% higher for the middle-SES group when compared to the low- and

high-SES groups, respectively (F(2, 201) =5.449, p =0.005). Post-hoc analysis revealed

significant interactions between low- and middle-SES groups (p =0.009) and low- and

high-SES groups (p =0.047). SEE scores were 16.4% higher for males compared to

females (F(1, 205) =8.137, p =0.005).

5.3.3. Regression Analysis

After adjusting for gender, age category, SES-group, and season, there was a

significant negative association between sitting (min‧w-1) and SAPF (β =-97.187, p

<0.001) and SEE (β =-17.819, p <0.029). Significant positive associations between

MPA10min (min‧w-1) and SEE (β =7.340, p =0.002), and MVPA10min and SRH (β =13.176, p

=0.005) and SAPF (β =5.762, p =0.002) were also observed. No significant

associations were obtained from self-reported sitting, MPA10min and MVPA10min.

130

5.4. Discussion

This study investigated gender, age and SES differences in older adults’ SB, PA and

self-reported physical and psychosocial health outcomes, and explored associations

between SB and PA with self-reported physical and psychosocial health outcomes.

No significant gender, age category and SES-group differences were observed

between self-reported and accelerometer-derived SB and PA. These findings are

counter to those in previous studies which have noted that MVPA10min is lower

among those who are female and older (Amagasa et al., 2017; Lohne-Seiler, Hansen,

Kolle & Anderssen, 2014; Ramires et al., 2017; Shiroma et al., 2018) due to

difficulties in mobility, general health status, and lower levels of self-efficacy

(Ramires et al., 2017). Men are more physically active than women in almost every

country throughout the adult and older adult age range when evaluated based on

current PA guidelines (Hallal et al., 2012; Sallis et al., 2016; Sun, Norman & While,

2013). Participants were recruited whilst attending the GHGA PA sessions and

therefore, both men and women across the age range were likely more inclined to

be active. Moreover, gender bias in the sample could have further affected any

potential gender and age category differences between SB and PA. Previous studies

have reported significantly lower levels of MVPA10min in older adults with low-SES

compared to those with middle and high-SES (Mendoza-Vasconez et al., 2016). A

variety of barriers including cost, transport, lower levels of education (e.g,, lack of

knowledge about PA benefits) and neighbourhood safety have been suggested as

reasons for these disparities (Buckman et al., 2014; Greaney et al., 2016; Lindgren,

Borjesson, Ekblo, Bergstrom, Lappas & Rosengren, 2016; Mendoza-Vasconez et al.,

131

2016; Xiao, Keadle, Berrigan, & Matthews, 2018). Participants in the current study

were recruited whilst attending PA programme sessions which were free to attend,

with free transport (local taxis) provided to and from each PA session and hence,

barriers associated with SES-group were not apparent as the PA sessions largely

eliminated cost, transport and neighbourhood safety barriers noted in previous

research (Baert et al., 2016).

In support of findings outlined in Chapter 3 (Study 1), significant gender, age

category and SES-group differences were observed with QoL, gender, and age

category differences with SRH, gender and SES-group differences with SAPF, and

gender differences with SEE. Older adults who are male, younger-old (60 to 69

years), and of higher SES are more likely to report favourable ratings of self-reported

physical and psychosocial health (Bamia et al., 2017; Kuosmanen et al., 2016).

Negative associations between SB and SAPF and SEE, and positive associations

between MVPA10min and SRH and SAPF were also observed. A number of self-

reported physical conditions including number of falls, balance, pain interference,

and lower-extremity function have been shown to be associated with time spent in

SB and MVPA (Haywood et al., 2018; de Rezende et al., 2014). Previous studies have

also noted negative associations of SB, and positive associations of MVPA on QoL,

wellbeing, depression, and self-efficacy (Ku et al., 2016; Withall et al., 2014).

No significant effects of Walk10min on secondary outcomes was found. Conversely,

previous studies have demonstrated significant positive effects of time spent in LPA

behaviours such as walking on SRH and SAPF (Buman et al., 2010; Kuosmanen et al.,

132

2016; Loprinzi et al., 2013), and psychosocial outcomes including increased QoL,

social well-being, socialization, and reduced stress (Ku et al., 2016; Sun et al., 2013).

Among older adults, the physical health benefits (e.g., number of falls, balance and

strength) of LPA are as beneficial as those for total accumulated MVPA (Ku et al.,

2016), and greater than all forms of other activity, including MVPA10min, in terms of

benefits to psychosocial well-being (Buman et al., 2010). It has been suggested that

two sessions per week of ‘light-to-moderate’ intensity PA each of a minimum of 45

minutes duration are optimal for improving self-reported physical and psychosocial

outcomes in older adults (Windle et al., 2010). Although less than the recommended

PA guidelines, fewer sessions of a lower intensity are more realistic for encouraging

long-term adherence to PA in older adults regardless of gender, age and SES-group

status (Kuosmanen et al., 2016). Further longitudinal studies in older adults are

warranted examining the effects of replacing sedentary behaviours not just with

MVPA, but also LPA (Chastin et al., 2018; Jefferis et al., 2016; McMahon et al., 2017;

Phoenix & Tulle, 2017).

Self-report and accelerometer assessed total time spent in SB (411.9 vs 772 min‧d-1)

and MVPA10min (62.7 vs 7.7 min‧d-1) is comparable to recent self-report and

accelerometer assessed SB and PA studies in older adults by López-Rodríguez et al.

(2017) and Amagasa et al. (2017), respectively. Within these two studies the total

number of participants was 80 (mean age =72, SD =5.5) and 450 (mean age =74.3, SD

=2.9) respectively, and total time spent in SB was 407.4 and 548.3 min‧d-1, and

MVPA10min was 27.4 and 17.9 min‧d-1, respectively. Compared to GA wrist-based

accelerometer-derived MVPA10min (cutpoint of ≥104.mg), self-reported MVPA10min

133

levels were higher by 58 min‧d-1 and 40 min‧d-1 for women and men, respectively. A

recent study by Menai et al. (2017) reported similar results when comparing

accelerometer-assessed and self-reported MVPA10min. Self-reported levels of

MVPA10min were 19.3 min‧d-1 higher than GA accelerometer-assessed MVPA10min after

applying an MVPA cutpoint of ≥100 mg. These findings further confirm the inherent

limitation of recall bias within self-report measures (Barnett et al., 2016). The

ubiquitous presence of total accumulated and sporadic PA in older adults makes it

difficult to recall in questionnaire surveys (Washburn, 2000), though such behaviours

may be of particular importance, especially for older adults who tend to perform

shorter duration exercises (Amagasa et al., 2017; Jefferis et al., 2016; Sparling et al.,

2015). Consequently, this population tend to misreport time spent in such activities

when compared with objective measures such as accelerometry (Ku et al., 2016).

The use of raw accelerometry presents many advances, such as transparency in the

analytical process and enhanced comparability between data collected from

different devices; however, there are still only limited triaxial wrist-based

acceleration data to compare current results to, owing to this attachment site only

becoming more commonly used in very recent studies (Menai et al., 2017; Ramires

et al., 2017; Troiano et al., 2014; Wijndaele et al., 2015). Total accumulated 1 second

bouts of MVPA were 74.7 and 72.8 min‧d-1, MVPA10min were 7.8 and 8.4 min‧d-1, and

mean acceleration were 26.3 and 25.4 mg‧d-1 for women and men, respectively.

These results are comparative to a recent study in older adults which reported total

accumulated 1 second bouts of MVPA to be 56.7 and 64.6 min‧d-1, MVPA10min to be

4.5 and 9.5 min‧d-1, and mean acceleration to be 21.5 and 22 mg‧d-1 for women and

men, respectively (Ramires et al., 2017). A study by Sun et al. (2013) reported that

134

under half (47.6%) of older adults accumulated at least one 10 minute bout of MVPA

per day. This figure was even lower in the current study (32.5%). PA guidelines (CDC,

2015; Department of Health, 2011a), were achieved by 88.1% and 12.9% of

accelerometer wearing participants when total accumulated 1 second bouts and 10

minute bout criteria were applied, respectively. The large differences between total

accumulated 1 second bouts of MVPA and MVPA10min outlines the considerable effect

that the use of different bout criteria has on the final estimate of MVPA in older

adults (Menai et al., 2017; Ramires et al., 2017). These differences might be even

more pronounced compared to other age groups since older adults are less likely to

sustain MVPA for longer periods (Amagas et al., 2017). Recent research notes that

the same physical and psychosocial health benefits associated with MVPA10min can be

achieved through total accumulated 1 second bouts of MVPA in older adults (Jefferis

et al., 2016; Sparling et al., 2015). Consequently, revised PA guidelines soon to be

published in the US (Office of Disease Prevention and Health Promotion, 2018) now

recognise that any amount of time spent in MVPA counts toward meeting PA

recommendations.


Strengths of the study included the use of both questionnaire- and accelerometer-

assessed SB and PA. Furthermore, high wearer compliance was observed among the

accelerometer-wearing sample. Previous research has also demonstrated superior

wearer compliance in older adults when compared to younger age groups (Doherty

et al., 2017). Some limitations of this study need to be considered when interpreting

the results. The cross-sectional design is a limitation, precluding causal inferences

135

and any associations (or lack of associations) between SB and PA with self-reported

health outcomes. Participant recruitment strategies increased the risk of sampling

bias and affected external validity as it is possible that those who were already

inclined to be active were more likely to participate. Furthermore, men tend to

decrease participation in leisure-time PA as they get older; whereas this dose-

response is not seen among women (Amagasa et al., 2017). Consequently, there is

the possibility of gender bias given the higher number of female participants

recruited across both samples, especially given the relatively even gender split of

older adults across Sefton Borough (43.5% men) (ONS, 2017). Due to the limited

number of available GA devices during the period of contact with the older adults’

population, uneven sample sizes occurred. Although accelerometers provide

objective measures, they cannot accurately detect postural information (e.g.,

standing vs. sitting) and capture some types of PA (e.g., bicycling), which may have

influenced estimations of SB and PA and caused some misclassification of time spent

in SB and LPA (Shephard & Tudor-Locke, 2016). Furthermore, only self-reported

physical and psychosocial health outcomes were adopted and no objective health

indicators were assessed. Consequently, recall bias is a probability (Barnett et al.,

2016). Lastly, although IMD is commonly adopted as a measure of SES (Ramsay et al.,

2015), it refers to street level deprivation rather than individual level SES. Hence, the

results may not be truly representative of the entire socioeconomic spectrum of

older adults across the three areas of differing SES participants were recruited from.

5.6. Conclusions

136

No significant gender, age category or SES-group differences were observed between

self-reported and accelerometer-derived SB and PA outcomes. There were

significant gender, age category and SES-group differences between QoL, SRH, SAPF,

and SEE. Results also provided evidence of a negative association of self-report SB,

and positive association of self-reported MPA10min and MVPA10min with physical and

psychosocial health outcomes. Large differences in participants achieving PA

guidelines were noted both between self-report and accelerometer-assessed

samples, and between accelerometer-assessed total accumulated 1 second bouts of

MVPA and MVPA10min. Future longitudinal studies are warranted to further confirm

these baseline findings and to further explore the most reliable methods to

successfully decrease SB and increase PA among older adults.

137

Thesis Study Map





Objectives





the community.




initiative.

Key Findings:










adults.




adults.

Objectives








138


Key Findings







respectively.




respectively.










Objectives



behaviour, physical activity and self-reported

health indicators.



health outcomes.

Key Findings

No significant gender, age category or

socioeconomic status differences were observed

between self-reported and accelerometer-derived

sedentary behaviour and physical activity

outcomes.

Significant gender, age category and

socioeconomic status differences between self-

reported quality of life, self-rated health, self-

assessment of physical fitness, and self-efficacy

for exercise were observed.

139

Study 5. Implementation fidelity of the Get Healthy Get Active physical activity programme for community-dwelling older adults

A negative association of self-reported sedentary

behaviour, and positive association of self-

reported moderate and moderate-to-vigorous

physical activity with health indicators was also

evident.

Study 4. A pragmatic evaluation of the Get

Healthy Get Active physical activity

programme for community-dwelling older

adults.

Objectives

To evaluate the effectiveness of Sport England’s

Get Healthy Get Active physical activity

intervention on older adults physical activity,

sedentary behaviour and self-reported health

indicators.

140

Chapter 6

Study 4: A pragmatic evaluation of the Get

Healthy Get Active physical activity programme

for community-dwelling older adults.

141

6.1. Introduction

Chapter 5 (Study 3) established that objectively assessed PA levels of older adults

were low, thus confirming that interventions targeting increased PA among this

population were warranted. These findings are also supportive of previous research

which has highlighted the need for PA promotion strategies among this population

(Plotnikoff et al., 2014; ten Brinke et al., 2015). The current study progresses the

work of Chapter 5 (Study 3) by evaluating the effectiveness of Sport England’s GHGA

PA intervention on older adults PA, SB and self-reported health indicators. Despite

the benefits associated PA (Devereux-Fitzgerald et al., 2016; Greaney et al., 2016;

Zhu et al., 2017), less than 12% of older adults globally perform PA on a daily basis

(Centers for Disease Control and Prevention, 2016). Large scale repeated measures

studies have shown that PA further declines with increasing age, among females,

those of lower SES, and among individuals with lower levels of self-reported health

status and SEE (Lehne & Bolte, 2017; Murtagh et al., 2015; Smith et al., 2015).

There is also growing public health interest in the amount of time spent in SB

(Tremblay et al., 2017). Older adults are the most sedentary segment of society

(Chastin et al., 2017). Observational studies show that more than 60% of an adults

non-sleeping hours are spent in SB, corresponding to around ten hours per day

(Dunstan, Howard, Healey, & Owen, 2012). Epidemiological evidence indicates that

time spent in SB is associated with physical (e.g., premature mortality, chronic

142

diseases and all-cause dementia risk) and psychosocial (e.g., self-reported QoL,

wellbeing and SEE) risk factors, even among individuals who meet current PA

guidelines (Meneguci et al., 2015; Thorp, Owen, Neuhaus, & Dunstan, 2011). Time

spent in SB further increases with increasing age, decreased MVPA levels, lower SES,

and on days with lower temperature, less sunshine, inclement weather, and fewer

daylight hours (Diaz et al., 2016; Eisinga, Franses & Vergeer, 2011).

The high levels of SB and low levels of PA among older adults are concerning given

their negative associations with self-reported health (Beyer et al., 2015), fitness

(Kuosmanen et al., 2016) and psychosocial outcomes including QoL and self-efficacy

(French et al., 2014; Greaney et al., 2016; Kim et al., 2016; Olson et al. 2016). SB

represents a unique and clinically important aspect of an individual’s overall activity

profile and is no longer considered simply to be the extreme low end of the PA

continuum (Dunstan et al., 2012). Consequently, safe, effective, inclusive, and

sustainable interventions are needed to address not only the single effect of either

SB or PA level, but a balance of both (ten Brinke et al., 2015).

A recent systematic review noted that the successful reduction of SB and promotion

of PA among community-dwelling older adults requires a whole system-oriented

approach tailored to meet the needs of older adults and aligned with social,

individual and environmental factors (Zubula et al., 2017). Interventions also need to

be adaptable and offer choice in order to be suitable to all, including those with

mobility restrictions, disabilities, or other limiting health conditions, and those in

different socio-demographic groups (Zubula et al., 2017). Such multi-modal and

143

multi-component interventions have had positive effects on reducing SB and

increasing levels of PA, SEE, and QoL in older adults (Gardner, Smith, Lorencatto,

Hamer, & Biddle, 2016; Zubula et al., 2017). However, effects on maintenance

beyond post-intervention remain unclear due to a lack of high quality longitudinal

studies (Olanrewaju et al., 2016; Richards et al., 2013).

Older adults' engagement in PA may also benefit from key facilitators of participation

in PA such as enjoyment, access, timing, and social support (Olanrewaju et al., 2016).

Social support is particularly relevant to ageing populations (Franco et al., 2015;

Sanders, Roe, Knowles, Kaehne, & Fairclough, 2018; Warner, Wolff, Ziegelmann,

Schwarzer, & Wurm, 2016), and is associated with PA adherence and maintenance in

older adults (Brown et al., 2015). Overall, participant centred, personalised

interventions starting with professional and tailored guidance and providing ongoing

support throughout and beyond the intervention have been found to lead to the

greatest improvements in PA levels in community-dwelling older adults (Zubula et

al., 2017). The characteristics of the GHGA PA intervention were in line with the

Consolidated Standards of Reporting Trials (CONSORT) checklist for pragmatic

evaluations (Zwarenstein et al., 2008) and thus, a pragmatic approach was adopted.

Pragmatic evaluations aim to inform a policy decision by providing evidence for

adoption of the intervention into ‘real-world’ practice (Schwartz & Lellouch, 1967).

Features include the recruitment of investigators and participants, the intervention

and its delivery, follow-up, and the determination and analysis of outcomes (Ford &

Norrie, 2016). In line with objective 7 of the thesis objectives, the aim of the current

study was to evaluate the effectiveness of Sport England’s GHGA PA intervention on

144

older adults PA, SB and self-reported health indicators. If effective, intentions were

for the GHGA programme to be scaled up and lead to bigger, sustainable, national

level research projects whose results would have policy ramifications and inform the

thought and practice of professionals in PA, social work and care settings.

6.2. Methods


This study protocol was prepared according to the CONSORT checklist of items for

reporting pragmatic trials (Zwarenstein et al., 2008). Between January 2016 and

December 2017, a quasi-experimental study with repeated follow-ups was

undertaken. All 207 participants who provided informed consent and met the

eligibility criterion for study three were invited to take part in the current study. Data

were collected from participants at 3-months (n =193; attrition =6.8%), 6-months (n

=168; attrition =18.8%), and 12-months (n =118; attrition =43%) post-baseline.

Participants invited to participate in the programme received a covering letter,

participant information sheet, and consent form, and provided written informed

consent prior to participation at baseline. Before the study commenced, institutional

ethical approval was received (#SPA-REC-2015-329). Participation was voluntary with

no incentives provided. Details of the flow of participants through the study from

baseline to follow-up are displayed in Figure 6.1.

Figure 6.1. Flow of participants through the study.

145Baseline

Total number of participants approached = 380

Total number of participants providing written informed consent and meeting inclusion

criteria = 207

6.2.2. Intervention

GHGA was a three-year project aimed at engaging inactive older adults in PA at least

once a week for 30 minutes, via a 12 week PA intervention. The project was funded

by Sport England and delivered by SMBC employees trained in delivering PA sessions

among this population. These trained employees are referred to as deliverers herein.

The intervention was implemented throughout Sefton Borough within differing

locations (e.g., leisure centres, a church hall, a theatre, a retirement homes, and a

library) with each 12 week PA intervention implemented at the same venue.

Deliverers were required to incorporate exercises outlined in table 6.1 which

targeted five core aspects of fitness (balance, endurance, flexibility, resistance, and

strength exercises), whilst also incorporating a warm-up and a cool-down.

Exercise component Example exercises

Warm-up

Balance

Seated stretching of all major muscle groups

Standing walking on the spot

Standing knee raises

Single limb stance

Single limb stance with arm

Side leg raise

Back leg raise

146

Follow-up 3 (12-months post baseline)Participants lost from baseline = 89 (74 missed due to data collection ending, 15 out of reach)Total number of participants assessed and analysed = 118

Table 6.1. Exercises typical of a GHGA PA session.

Endurance

Flexibility

Resistance (with exercise band)

Strength

Cool-down

Walking heel to toe

Seated jumping jacks

Arm circles

Bicycle kicks

Spine twist

Shoulder rolls

Ankle circles

Head circles

Overhead stretch

Seated bicep curl

Seated chest press

Seated shoulder press

Seated leg press

Standing side deltoid raise

Seated dumbbell curls

Seated weighted ankle circles

Standing weighted ankle circles

Standing squats

Standing calf raises

Seated stretching of all major muscle groups

In line with the Evidence Integration Triangle (Glasgow et al., 2012), the exploration

of the three main evidence-based components of intervention programme/policy,

implementation processes, and measures of progress were achieved via a 12 week

pilot GHGA programme delivered by SMBC throughout Sefton Borough from

September to December 2015. A total of 281 older adults took part in individual

interviews. Data regarding current trends of inactivity, barriers to participation, and

147

preferred sports and venues were collected via informal focus groups and individual

interviews. Results and analysis from this pilot, along with relevant intervention

literature and findings from Chapter 3 (Study 1) of the current thesis (Sanders et al.,

2018), were fed back to Sport England as the funder, as well as SMBC in order to

assess, evaluate and promptly inform adapted future iterations of the GHGA PA

intervention. Through adopting key elements of the ecological model of behaviour

change (Stokols, 1992), the GHGA PA intervention targeted individual (e.g., session

components tailored to individual needs), interpersonal (e.g., social networking),

organisational (e.g., advertising through Older Peoples' Forums throughout Sefton

Borough), and community (e.g., advertising through general practitioner surgeries

and leisure centres throughout Sefton Borough) level factors. PA sessions consisted

of chair-based PA, including a wide range of progressive exercise modalities, at

differing intensities. Sessions were designed to have no financial cost to the

participants and hence, were free to attend with transport to and from each session

provided if necessary. Sessions lasted for one hour and consisted of a ten minute

warm-up, forty minutes of varying aerobic, endurance, strength, flexibility, and

balance exercises, and a ten minute cool-down undertaken by each individual within

a group setting.

6.2.3. Primary Outcome Measures

6.2.3.1. International Physical Activity Questionnaire for the Elderly

To assess SB and PA levels, the IPAQ-E (Hurtig-Wennlöf et al., 2010) was adopted as

required by funder. IPAQ-E is based on the short version of the IPAQ

(www.ipaq.ki.se) and assesses time spent sitting, walking in bouts of 10 minutes or

148

more, MPA in bouts of 10 minutes or more, and VPA in bouts of 10 minutes or more

during the previous 7 days. The categorical outcome from IPAQ-E assigns the

participants into one of three PA categories (e.g., low, moderate, or high-PA). The

IPAQ-E provides favourable levels of both direct, and indirect levels of criterion

validity for sitting (Spearman r =0.28, p <0.05), walking (Spearman r =0.35, p <0.01),

MPA (Spearman r =0.40, p <0.01), and VPA (Spearman r =0.37, p <0.01) (Hurtig-

Wennlöf et al., 2010). However, varying levels of test-retest reliability (intraclass

correlation ranging from 0.30 to 0.82) have also been reported (Tomioka, Iwamoto,

Saeki, & Okamoto, 2011).

6.2.4. Secondary Outcome Measures

6.2.4.1. Self-Assessment of Physical Fitness

To assess self-reported levels of fitness, the SAPF (Weening-Dijksterhuis et al., 2012)

was adopted. The questionnaire asks three questions including: How do you rate

your strength?; How do you rate your aerobic endurance?; and How do you rate your

balance? The SAPF uses a rating method from 0 (indicating the lowest rating) to 10

(indicating the highest rating) for each of the three items. Sound psychometric

properties have been demonstrated for the SAPF scale among frail older adults with

acceptable internal consistency (Cronbach alpha =0.71) (Weening-Dijksterhuis et al.,

2012), one week test-retest validity (0.70), and moderate concurrent validity against

the Groningen Fitness Test for the Elderly (Lemmink, 1996).

6.2.4.2. Kemp Quality of Life Scale

149

To assess QoL, the single item Kemp Quality of Life Scale (Kemp & Ettelson, 2001)

was adopted. Following the question: Thinking about the good and bad things that

make up your quality of life, how would you rate the quality of your life as a whole? ,

respondents are asked to tick the box next to the answer that best describes their

QoL on a 7-item scale anchored by “so good, it could not be better” (score of 1) to

“so bad, it could not be worse” (score of 7). Although there is no published

information about the distribution or error of data obtained using this scale, the

scale is interval level, can provide data for parametric and non-parametric analysis,

and has been adopted previously in older adult populations (Siebens et al., 2015;

Roe et al., 2011).

6.2.4.3. Self-Rated Health Questionnaire

The SRH (Sargent-Cox et al., 2010) is a 3-item questionnaire assessing global, age-

comparative, and self-comparative HS. Global HS is measured via the question: How

would you rate your overall health at the present time?. Participants answer on a five

point scale (1= excellent; 5= poor). Age-comparative HS is measured on a three-point

scale via the question: Would you say your health is: (1) better, (2) about the same,

or (3) worse than most people your age?. Self-comparative HS is also measured on a

three point scale via the question: Is your health now… (1) better, (2) about the

same, or (3) not as good as it was 12 months ago?. Previous research has shown

good prognostic validity for mortality, and high internal reliability (α =.75) (Ferraro &

Wilkinson, 2013).

6.2.4.4. Self-Efficacy for Exercise Scale

150

The SEE (Resnick & Jenkins, 2000) is an 11-item scale designed to assess confidence

to continue exercising in the face of perceived barriers. The SEE consists of 11

situations that might affect participation in exercise. Items are rated from 0 (not

confident) to 10 (very confident). The mean score of numerical ratings from each

response indicates the strength of efficacy expectations. The SEE scale has

demonstrated excellent internal consistency (Cronbach α =0.92) and significant

predictive validity of exercise activity in older adults when controlled for age and

gender (F =78.8; p <0.05) (Resnick & Jenkins, 2000; Resnick et al., 2004).

All measures were administered by the first author to each participant whilst they

attended the GHGA sessions. Measures took no longer than 30 minutes to complete

per participant. Regular breaks were incorporated if needed, to reduce fatigue.

Additional visits were also arranged if needed, to aid completion of a full dataset due

to participant time constraints. All participants recorded their age, gender, and home

postal code. Ages were categorised as 65 to 69 years, 70 to 79 years, and ≥80 years.

Postal codes were used to estimate SES by generating IMD (Department for

Communities and Local Government, 2015) using an online conversion tool

(http://imd-by-postcode.opendatacommunities.org/). The IMD is a UK government

metric used to rank area-level deprivation within and between different

communities. The IMD scores rank each super output area in England from 1 (most

deprived area) to 32,844 (least deprived area). Super output areas were then split

into 3 equal groups to represent low (1 to 10,947), middle (10,948 to 21,895), and

high-SES (21,896 to 32,844). Questionnaire data were input manually into Microsoft

Excel 2013 and checked for missing data. Confidentiality and data storage

151

procedures were adhered to as is set out in Edge Hill University’s research data

management (Edge Hill University, 2017) and code of practice for the conduct of

research guidelines (Edge Hill University, 2017). All participant data were

anonymised and coded to prevent identification, and securely stored using

password-protected files on the Edge Hill University computing network. The server

hosting the files was backed up every four hours and nightly to tape, to ensure data

attrition was minimised. Only research team members had access to the anonymised

data. This was shared between the team strictly for the purposes of research.

6.2.5. Statistical Analysis

Descriptive statistics were calculated for the outcomes of all participants at baseline

and follow-up. Multilevel modelling was performed using MLwiN Version 3.00

(Centre for Multilevel Modelling, University of Bristol, UK) (Rabesh, Charlton,

Browne, Healy, & Cameron, 2009) to determine the effects of the intervention.

Multilevel modelling was appropriate for use in this study design where data

collection time points (e.g., observations) are clustered within older adults (Twisk,

2006). Therefore, a 2-level data structure was used with data collection time point

defined as the first level of analysis, and participant ID as the second level of analysis.

An issue of longitudinal studies is missing data. Previous literature has suggested

imputation of missing data to obtain a ‘complete dataset’ (Little & Rubin, 2014).

However, imputation may result in over- or underestimation of the importance of

covariates, underestimation of random coefficient and effect variance corresponding

to time-varying covariates with missing values, and conflation of within-person and

152

between-person effects (Enders, Mistler & Keller, 2016; Grund, Lüdtke & Robitzsch,

2016; Lüdtke, Robitzsch & Grund, 2017; van Buuren, 2011). Applying multilevel

analysis to an incomplete longitudinal dataset is even better than applying

imputation methods (Twisk, 2013) due to its ability to analyse the variance of

random intercept and regression coefficients regardless of individual missing values

(Twisk, 2006).

Continuous outcome variables were sitting, MVPA, QoL, SRH, SAPF and SEE.

Additionally, dichotomous outcome variables of achieving/not achieving MVPA

guidelines as well as PA level (low and moderate/high active (according to the IPAQ-

E MET min‧w-1 scoring protocol; IPAQ Research Committee, 2005) were studied.

Initially, ‘crude’ analyses were conducted with only the outcome variables at three,

six and 12-months included in the model (Twisk, 2006). Potential confounding

covariates were then added to construct the ‘adjusted’ models. These potential

confounding covariates were selected based on previous research which has

deemed them to be influential the dependent outcomes. For sitting these covariates

included gender (Greaney et al., 2016), age (Diaz et al., 2016; Heo et al., 2017), SES

(Diaz et al., 2016; Gray et al., 2015), season (Diaz et al., 2016), QoL (Kim et al., 2016),

SRH (Beyer et al., 2015; Kuosmanen et al., 2016) and MVPA (Beyer et al., 2015). For

the PA level categorical outcome the covariates included gender (Amagasa et al.,

2017), age (Amagasa et al., 2017), SES (Ku, Steptoe, Liao, Sun, & Chen, 2018), season

(Parsons et al., 2016), QoL (Buman et al., 2010), SAPF (Ku et al., 2016), SRH (Ku et al.,

2016), and MVPA (Barone Gibbs et al., 2017). For MVPA and achieving/not achieving

153

MVPA guidelines, confounding covariates of gender (Amagasa et al., 2017; Shiroma

et al., 2018), age (Doherty et al., 2017; Ramires et al., 2017), SES (Mendoza-Vasconez

et al., 2016), season (Prins & van Lenthe, 2015), QoL (Lok, Lok & Canbaz, 2017), SRH

(Kuosmanen et al., 2016; Ramires et al., 2017) and PA self-efficacy (Dionigi, 2007;

French et al., 2014) were included. Confounding covariates for the health indicators

of QoL, SRH, SAPF, and SEE included gender (Bamia et al., 2017; Kirchengast &

Haslinger, 2008; Moreno et al., 2017; Overdorf, Coker & Kollia, 2016), age (Bamia et

al., 2017; Hong, 2015; Langan & Marotta, 2000; Overdorf et al., 2016), sitting (Ku et

al., 2016; Kuosmanen et al., 2016), and MVPA (Beyer et al., 2015; Ku et al., 2016).

Regression coefficients from the models were assessed for significance using the

Wald statistic and the following equation, (regression coefficient/standard error)2.

Statistical significance was set at p <0.05. The evaluation of potential effect

modification was also carried out on the gender, age category, SES-group, and

season to determine whether the intervention effects were different for these

subgroups. Interaction terms were added to the models, consisting of a

multiplication of the main determinant (intervention) and the potential effect

modifier (Twisk, 2006). Due to the reduced power which interaction terms have,

statistical significance for these analyses was set at p <0.1 (Twisk, 2006).

6.3. Results

A total of 318 participants were recruited for participation in this study. Two

hundred and seven participants (mean age =77.8, SD =7.7; 65.1% compliance rate)

met the inclusion criterion, provided written informed consent, and were

154

subsequently included in the analysis. Table 6.2 highlights baseline characteristics of

the participants and Table 6.3 highlights unadjusted self-reported PA and

psychosocial outcome measures at baseline and follow-up time points.

Table 6.2. Descriptive baseline characteristics of the participants.

SD, Standard Deviation; SES, Socioeconomic status.

Table 6.3. Unadjusted self-reported physical activity and psychosocial outcome measures.

Baseline(n =207)

3-months (n =193)

6-months (n =168)

12-months (n =118)

155

Baseline(n =207)

Age (years)

Gender n (%)

Female

Male

Age Category n (%)

65-69 yrs

70-79 yrs

80+ yrs


Autumn/Winter

Spring/Summer

SES Group n (%)

Low-SES

Mid-SES

High-SES

77.8 (7.7)

165 (79.7)

42 (20.3)

32 (15.5)

92 (44.7)

83 (39.8)

86 (41.7)

121 (58.3)

74 (35.9)

51 (24.3)

82 (39.8)

Sitting (min‧w-1) 2883.7 (1137.2)

3348 (916.4)

3435 (886.9)

3735.5 (895.1)

MVPA (min‧d-1)

Meeting PA Guidelines n (%)

YES

NO

62.7 (92.2)

114 (55.3)

92 (44.7)

62.5 (63.2)

49 (25.4)

144 (74.6)

57.6 (49)

38 (22.6)

130 (77.4)

61.7 (44.1)

29 (24.6)

89 (75.4)

IPAQ-E Activity Category n (%)

Low

Moderate

High

118 (57)

41 (19.8)

48 (23.2)

67 (34.7)

64 (33.7)

60 (31.6)

60 (35.7)

64 (38.1)

44 (26.2)

46 (39)

36 (30.5)

36 (30.5)

Quality of Life 3.2 (.77) 3.4 (.77) 2.8 (.7) 2.8 (.8)


9.8 (3.48) 11.7 (4) 12.3 (3.4) 12 (3.3)

Self-Rated Health 8.1 (1.5) 6.9 (2) 6.6 (1.9) 6.7 (2)

Self-Efficacy for Exercise 29.8 (10.1) 37.9 (12.3) 39.9 (12.4) 40.1 (13 )

6.3.1. Intervention Effects

Tables 6.4 and 6.5 show the results of the crude and adjusted multilevel analyses,

respectively. In the adjusted models, significant increases in sitting time of 526, 650

and 974 minutes per week were observed at three, six and 12-months compared to

baseline, respectively. Results also indicated decreases in MVPA of 12.95, 21.85 and

18.99 minutes per day at three, six and 12-months, respectively, although three (p

=0.16) and 12-months (p =0.1) did not reach significance. Significant odds of meeting

MVPA guidelines were 9.09, 11.1 and 9.09 times lower across the follow-up time

points compared to baseline. Finally, significant odds at three and six-month follow-

156

up time points indicated participants were 5.88 times more likely to be classed as

low active compared to baseline.

Significant changes in health indicator scores were also observed across all follow-up

time points. At 12-months, favourable changes in QoL (β =-0.09 (95% CI =-0.36, 0.18),

p< .001), SRH (β =-1.41 (95% CI =-2.02, -0.8), p <.001), SAPF (β =2.97 (95% CI =1.71,

4.23), p <.001), and SEE (β =11.57 (95% CI =7.7, 15.44), p <.05) scores were observed

compared to baseline.

157

Table 6.4. Crude multilevel model analyses of the outcome measures at three, six and 12-months follow-up.

3-months 6-months 12-months

β or OR 95% CI p β or OR 95% CI p β or OR 95% CI p

Sitting (min‧w-1) 402.68 c 244.49 to 560.87 <0.001 599.4 c 434.04 to 764.76 <0.001 919.02 c 732.41 to 1105.63 <0.001

MVPA (min‧d-1)

Meeting PA Guidelines (No)

IPAQ-E Activity Category (Low)

Quality of Life


-0.58 c

3.7* d

1.85* d

-0.36 c

2.04 c

-10.99 to 9.83

3.27 to 4.13

1.55 to 2.15

-0.49 to -0.23

1.49 to 2.59

0.91

<0.001

<0.001

<0.001

<0.001

-7.18 c

4.35* d

1.85* d

-0.37 c

2.28 c

−18.1 to 3.71

3.9 to 4.8

1.53 to 2.17

-0.5 to -0.24

1.7 to 2.86

0.2

<0.001

<0.001

<0.001

<0.001

0.62 c

3.85* d

1.61* d

-0.44 c

2.3 c

−11.69 to 12.93

3.35 to 4.35

1.24 to 1.98

-0.59 to -0.29

1.65 to 2.95

0.92

<0.001

0.01

<0.001

<0.001

Self-Rated Health -1.1 c -1.4 to -0.8 <0.001 -1.48 c −1.79 to -1.17 <0.001 -1.4 c −1.75 to -1.05 <0.001

Self-Efficacy for Exercise 9.01 c 7.22 to 10.8 <0.001 10.07 c 8.2 to 11.95 <0.001 10.82 c 8.7 to 12.94 <0.001

Values reflect the intervention effects (e.g., within group differences) between baseline and follow-up time points. Use of bold denotes beta significant intervention effects (p < 0.05). * Significant odds ratios <1 have been inverted (Osborne, 2006). c β value. d OR. OR, odds ratio; CI, confidence interval; PA, physical activity; MVPA, Moderate-to-vigorous physical activity. IPAQ-E; International Physical Activity Questionnaire for the Elderly; Quality of Life and Self-Rated Health are negatively scored (e.g., lower score = better outcome). Note. Lower score = more favourable result for Quality of Life and Self-Rated Health.

158

3-months 6-months 12-months

β or OR 95% CI p β or OR 95% CI p β or OR 95% CI p

Sitting (min‧w-1) 526.91 c 242.69 to 811.13 <0.001 650.09 c 339.93 to 960.25 <0.001 974.41 c 616.47 to 1332.36 <0.001

MVPA (min‧d-1)

Meeting PA Guidelines (No)

IPAQ-E Activity Category (Low)

Quality of Life


-12.95 c

9.09* d

5.88* d

-0.37 c

2.59 c

−30.92 to 5.02

8.16 to 10.02

2.12 to 9.64

-0.58 to -0.17

1.62 to 3.56

0.16

<0.001

0.01

<0.001

<0.001

-21.85 c

11.1* c

5.88* d

-0.40 c

2.11 c

-41.08 to -2.62

10.07 to 12.13

1.78 to 9.98

-0.17 to -0.62

1.04 to 3.18

0.03

<0.001

0.02

<0.001

<0.001

-18.99 c

9.09* d

1.63 d

-0.09 c

2.97 c

-41.56 to 3.58

8.01 to 10.17

-1.04 to 4.3

-0.36 to 0.18

1.71 to 4.23

0.1

<0.001

0.72

<0.001

<0.001

Self-Rated Health -0.89 c -1.37 to -0.41 <0.001 -1.19 c -1.71 to -0.67 <0.001 -1.41 c -2.02 to -0.8 <0.001

Self-Efficacy for Exercise

9.89 c 6.93 to 12.85 <0.001 9.92 c 6.63 to 13.21 <0.001 11.57 c 7.7 to 15.44 <0.001

Table 6.5. Adjusted multilevel model analyses of the outcome measures at three, six and 12-months follow-up.

159

Values reflect the intervention effects (e.g., within group differences) adjusted for confounding covariates at baseline, between baseline and follow-up time points. Values in bold denote beta (95% CI) and significance values of outcomes with significant intervention effects (p < 0.05). * Significant odds ratios <1 have been inverted (Osborne, 2006). c β value. d OR. OR, odds ratio; CI, confidence interval; PA, physical activity; MVPA, Moderate-to-vigorous physical activity; IPAQ-E; International Physical Activity Questionnaire for the Elderly. Note. Lower score = more favourable result for Quality of Life and Self-Rated Health.

160

6.3.2. Interaction Analyses

Subgroup interaction analyses (see Table 6.6) revealed that improvement in SAPF

score at 12-months was significant among males (β =3.38 (95% CI =2.01, 4.75), p

<.001), and non-significant among females (β =1.51 (95% CI =-0.89, 3.91), p =.22). No

other significant subgroup interaction effects were evident.

Table 6.6. Significant intervention subgroup interactions.

Values in bold denote beta (95% CI) and significance values of crude and adjusted intervention outcomes with significant subgroup effect modifiers (p < 0.1). CI, confidence interval.

6.4. Discussion

This study aimed to evaluate the impact of Sport England’s GHGA PA intervention on

time spent in SB and MVPA, as well as health indicators. After accounting for

confounding variables, significant increases in sitting time of 526, 650 and 974

minutes per week at three, six and 12-months, respectively were observed. Given

the decreasing levels of MVPA noted, increased sitting time was to be expected as

lower amounts of time spent in MVPA has been associated with greater total

sedentary time (Diaz et al., 2016). This result further confirms the need for

interventions to go beyond simply increasing MVPA levels, and to actively seek

methods of decreasing SB. It is suggested that interventions should concentrate not

161

Interactions SAPF Score

β 95% CI p

Adjusted model x gender

Male x 12-months

Female x 12-months

3.38

1.51

2.01, 4.75

-0.89 to 3.91

<0.001

0.22

only on the single effect of either SB or PA level, but a balance of both (ten Brinke et

al., 2015), as older adults have been found to compensate for increased MVPA levels

by decreasing LPA and increasing SB during the remainder of the day (Barone Gibbs

et al., 2017). Concurrently, results of the current study showed significant increased

odds of participants being classed as low active at three and six-month follow-up

time points compared to baseline.

Across all follow-up time points participants spent on average 8.3 hours of waking

time per day sitting. This is comparative to previous research suggesting that older

adults spend up to 8.5 hours of waking time per day sedentary regardless of PA level

(Shaw et al., 2017). Individuals aged 65 to 74 and 75 years or older have a 2-fold and

4-fold greater likelihood of exhibiting prolonged SB compared to individuals aged 45

to 54 years (Diaz et al., 2016). Hence, targeting less SB might be a successful

approach to increase whole day PA in older adults who may have limited mobility,

motivation, or self-efficacy for more intense or prolonged bouts of exercise

(Gardiner, Eakin, Healy, & Owen, 2011; Manns, Dunstan, Owen, & Healy, 2012). To

date however, interventions aimed at both promoting PA and reducing SB have

reported mixed results (Martin et al., 2015; Prince et al., 2014). Results from a recent

meta-analysis by Prince et al. (2014) identified that on average, interventions

targeting both PA and SB resulted in significant, but modest reductions in time spent

engaged in SB (standardised mean difference (SMD) =−0.37 [95% CI =−0.69, −0.05])

equating to a mean difference of approximately 35 min‧d-1 less sedentary time in the

intervention groups compared with the controls. However, a more recent systematic

review by Martin et al. (2015) concluded that there was no evidence that combined

162

PA and SB interventions reduced SB. Interventions solely targeting SB seem

promising given meta-analysis results indicating significant and large reductions of

up to 91 min‧d-1 of sedentary time in intervention groups compared with the

controls (Prince et al., 2014). To further build evidence-based approaches for

reducing SB, there is a need to understand the factors that influence patterns of

prolonged SB (Diaz et al., 2016). Studies exploring the barriers and facilitators of

reducing SB are warranted in order to inform future interventions.

Results indicated a significant decrease in MVPA of 22 min‧d-1 at six months follow-

up, and non-significant decreases in MVPA of 13 and 19 min‧d-1 at three and 12-

months, respectively. These results are in contrast to a recent meta-analysis of 53

exercise intervention studies in community-dwelling older adults which reported a

pooled effect equivalent to a 73 min‧w-1 increase in MVPA when comparing

intervention with control groups (Chase, 2015). Similarly, Barone Gibbs et al. (2017)

observed increases of both self-reported (67 min‧w-1) and objectively assessed MVPA

(75 min‧w-1) compared to baseline following a 12-week PA intervention in older

adults. The GHGA PA sessions aimed to provide sustainable PA sessions which would

lead to PA maintenance among inactive older adults. Adult and older adult

guidelines advise ≥150 minutes of MVPA weekly, or 75 minutes of vigorous PA, or a

combination, in ≥10 minute bouts (Centers for Disease Control and Prevention, 2015;

Department of Health, 2011a), but any increase in PA for inactive people is valuable

(Sparling et al., 2015). Fewer PA sessions of a lighter intensity are more realistic for

encouraging long-term PA adherence in older adults regardless of gender, age and

SES-group status (Kuosmanen et al., 2016; McMahon et al. 2017). Furthermore, LPA

163

is reported to be optimal for improving self-reported physical and psychosocial

outcomes in older adults (Windle et al., 2010). Given such results, Chapter 7 (Study

5) will evaluate whether or not the GHGA multi-component intervention was

implemented as intended. Process evaluation of intervention delivery is deemed to

be particularly important in multi-component interventions delivered across varying

locations where the same intervention may be implemented and received in

different ways (Koorts et al., 2018).

Significant increases in QoL, SAPF, SRH and SEE scores were observed throughout the

three, six and 12-month follow-up time points, providing further evidence for the

benefits of PA interventions in this population. Although type of PA does not seem to

influence effectiveness, there is an indication that light-to-moderate intensity PA

interventions may be preferred in older adults (Zubula et al., 2017). Among older

adults it is reported that the physical health benefits of LPA are as beneficial as those

for MVPA (Ku et al., 2016), and greater than all forms of other activity in terms of

benefits to psychosocial well-being (Buman et al., 2010). Significant favourable

changes in all health indicators were observed throughout the follow-up time points.

A recent meta-analysis also reported significant increases in self-reported QoL at six

and 12-months (Chase, 2015), suggesting that PA interventions among community-

dwelling older adults can be effective in prompting long-term increases in

psychosocial health indicators.

Research has also explored the relationship between SB and health indicators in

older adults, subsequently highlighting the detrimental effects that increased time

164

spent in SB can have on self-reported physical and psychosocial outcomes including

SRH (Beyer et al., 2015; Kuosmanen et al., 2016) and QoL (Kim et al., 2016).

Moreover, social support has been recognised as an important social determinant of

health and studies have demonstrated a relationship between social support and

QoL (Siedlecki, Salthouse, Oishi, & Jeswani, 2014), SRH (Dai, Zhang, Zhang, Li, Jiang,

& Huang, 2016), and SEE (Warner, Ziegelmann, Schüz, Wurm, & Schwarzer, 2011).

Resultantly, the thirty minutes set aside at the end of each weekly GHGA PA session

to allow participants to socialise could have further emphasised the long-term health

indicator increases observed. Given the decrease in MVPA, increase in sitting time,

and significant increases in health indicator scores observed, further research

exploring the individual effects of PA, SB and social support on health indicators, and

the interactions between them are warranted.

Subgroup analysis of this study highlighted that the GHGA intervention had a

significant positive effect on males SAPF score at 12-months, but this effect was not

significant for females. Previous studies also highlight a strong positive relationship

between gender and self-reported physical health indicators (Bamia et al., 2017;

Overdorf et al., 2016). A recent meta-analysis by Bamia et al. (2017) examining

confounding covariates of SRH scores among 424,791 European and US older adults

(aged ≥60 years) noted gender (male) to be a long-term confounding covariate

favourably associated with increased SRH score. Additional covariates included age

(younger-old), education (high), marital status (married/cohabiting), PA (active),

body mass index (non-obese), alcohol consumption (low to moderate), and previous

morbidity (absence). Similarly, Overdorf et al. (2016) reported gender (male) and age

165

(younger-old) to be confounding covariates favourably associated with physical self-

perception (PSPP; Fox & Corbin, 1989). A contributing factor to the gender

differences noted previously is that self-esteem is highly correlated with masculinity

and self-efficacy but not with femininity (Delignières, Marcellini, Brisswalter, &

Legros, 1994). Furthermore, it has been suggested that the masculine-role

endorsement could have a major influence on physical self-worth and hence, males

do consider themselves to be of higher levels of fitness and health (Delignières et al.,

1994).

Results of the current study also showed that regardless of gender, age and SES, the

GHGA intervention had significant effects on SB, MVPA and health indicators. These

findings are counter to those in previous studies which have noted that time spent in

SB is higher among those who are male, older and of low-SES compared to those

with middle and high-SES (Bellettiere et al., 2015; Shaw et al., 2017). MVPA is noted

as being lower among those who are female, older and of low-SES compared to

those with middle and high-SES (Lehne & Bolte, 2017; Mendoza-Vasconez et al.,

2016; Murtagh et al., 2015; Smith et al., 2015). Men are more physically active than

women in almost every country throughout the adult and older adult age range

when evaluated based on current PA guidelines (Hallal et al., 2012; Sallis et al., 2016;

Sun et al., 2013). Among self-reported health indicators, older adults who are male,

younger-old (60-69 years), and of higher SES are more likely to report favourable

ratings of self-reported physical and psychosocial health (Bamia et al., 2017;

Kuosmanen et al., 2016). Despite repeated requests by previous systematic reviews

(Richards et al., 2013; Vijay, Wilson, Suhrcke, Hardeman, & Sutton 2016) and

166

guidelines (National Institute of Health and Care Excellence, 2014) for interventions

to be conducted with longer follow-up periods and objective PA measures, there

remains a lack of data from interventions assessed using objectively measured PA

levels beyond 12-months (Harris et al., 2018). In order to progress this field of

research, future interventions targeting PA in older adults should be designed and

implemented based upon the requirements of recognised intervention guidelines

such as the SEF for PA interventions (National Obesity Observatory, 2012), and

intervention effect should be assessed via objective measures.


The GHGA PA intervention had several strengths. The repeated measures design at

three follow-up time points post-intervention is a strength, as well as the statistical

analysis which took into account these differing follow-up time points. In line with

the SEF for PA interventions (National Obesity Observatory, 2012), the design,

delivery and recruitment strategies were developed through prior formative

research (Sanders et al., 2018) and were theoretically underpinned by conceptual

behaviour change models (McLeroy et al., 1988; Stokols, 1992). Additionally, GHGA

provided a rare opportunity for participants to participate in 12 weeks of PA sessions

free of charge. Due to high engagement throughout the initial 12 week PA sessions,

and high demand from participants for the continuation of the GHGA PA sessions,

paid maintenance sessions were set up by SMBC. This demand regardless of cost,

suggests that the GHGA PA intervention can be self-sustained by SMBC beyond the

initial funding by Sport England. However, the decreases in MVPA and increases in

sitting time across all follow-up time points suggests that the potential for long-term

167

implementation throughout Sefton Borough and scaling up of the intervention to

reach broader populations across multiple settings is not warranted.

This study also acknowledges several limitations. The major limitation of the current

study was the pre-post design. This design results in lowered levels of causal validity

due to the uncontrollable effects of regression to the mean (RTM), maturation,

history and test effects (Marsden & Torgerson, 2012). The absence of a control group

is also a limitation. A methodological review of studies of psychological, educational

and behavioural treatments (Lipsey & Wilson, 1993) showed that pre-post designs

consistently overestimate effectiveness by an average of 61% compared with studies

with a control group. However, the feasibility of a clustered controlled research

design was rejected by SMBC due to both low GHGA deliverer and participant

numbers. Coverage and sampling errors are also limitations due to the modest

sample size and non-randomisation of participants. Consequently, selection bias is

an issue (Marsden & Torgerson, 2012). However, participant non-randomisation was

justified given the pragmatic approach adopted (Thomas et al., 2006) and the low

number of participants recruited from intact groupings (Sefton’s Older People Forum

and care homes) throughout Sefton Borough. Furthermore, minimal initial

participant characteristic data is also a concern further inhibiting causal validity.

Future studies should obtain additional participant characteristic data including

height and weight, current sedentary time and PA levels, history of PA, family history

of PA, ethnicity, employment status, and educational achievements as such are

reported as confounding covariates of older adults SB and PA levels (Greaney et al.,

2016; Keadle et al., 2016).

168

An additional limitation is the lack of objective measures of sitting time and/or PA

level. Recall bias is an inherent limitation of self-report measures, particularly for

routine and sedentary activities that are not encoded in memory as discrete events

(Altschuler et al., 2009). Given that recall is an ability that can decline with ageing,

recall bias is a possibility (Barnett et al., 2016). Given the complexity of PA

constructs, and the variety of applications available for their measurement in

surveillance, epidemiology, clinical, and intervention research, it is recommended

that future repeated measures research among this population adopt objective

measures such as accelerometry in order to provide the most accurate evidence for

intervention effect on PA levels and SB (Martin et al., 2015). Objective measurement

in the form of GA accelerometry was adopted as a measure of PA level and SB to

examine baseline participant characteristics prior to participation in the GHGA PA

intervention. Due to time constraints and a limited number of available GA

accelerometer devices during the period of contact with the older adults’

population, this was not possible in the current study.

Systematic attrition is a limitation of repeated measure research designs and causes

biases in all results that are influenced by these variables (Asendorpf, Van De Schoot,

Denissen, & Hutteman, 2014). Consequently, nonresponse bias is a concern given

attrition levels of 6.8, 18.8 and 43% compared to baseline at three, six and 12-month

follow-up time points, respectively. However, comparable attrition rates of 15.7,

20.2 and 47.2% at three, six and 12-month follow-up time points were reported in a

recent study examining the impact of a PA intervention among older adults

169

(Eggenberger, Theill, Holenstein, Schumacher, & de Bruin, 2015). Furthermore,

multilevel analysis is flexible to missing data and has been shown to be more

effective at analysing incomplete datasets than applying imputation methods (Twisk,

2013). However, examining correlates of attrition and attempting to understand

reasons for attrition is a vital step in the research process that needs further

exploring (McDonald, Haardoerfer, Windle, Goodman, & Berg, 2017). Finally,

implementation bias is a limitation as the GHGA PA sessions were delivered by

differing SMBC staff, and session content varied according to participant feedback

and group capabilities. Gaining an accurate and objective record of session content is

important in determining intervention suitability and overall effectiveness. However,

providing older adults with both choice and a wide range of PA components is a

facilitator of both initial adoption and maintenance of PA engagement and so

content variation was justified (Petrescu-Prahova et al., 2015; Sanders et al., 2018).

6.6. Conclusions

The GHGA PA intervention resulted in favourable changes in QoL, SAPF, SRH and SEE

scores throughout all follow-up time points. These results add further support for

the effectiveness of PA interventions to impact upon self-reported physical and

psychosocial health indicators in older adults. However, significant increases in

sitting time across all follow-up time points were observed, as well as a significant

decrease in time spent in MVPA at the six-month follow-up time point. The odds of

meeting MVPA guidelines also decreased significantly across the three follow-up

time points. Results from this pragmatic evaluation indicate that the potential for

long-term implementation of the GHGA programme throughout Sefton Borough, and

170

scaling up of the intervention to inform the thought of policy and practice of

professionals in PA, social work and care settings is not warranted in its current

capacity. After taking into account the barriers and facilitators of PA participation

outlined in Chapter 3 (Study 1), future research should continue to explore the

feasibility of interventions targeting PA among inactive older adults. It is

recommended that future research studies explore the potential of PA promotion

interventions to effect sustained improvements. Large scale longitudinal projects

with follow-up beyond two years are needed to identify the interventions capable of

achieving long-term results and establishing maintained PA engagement post-

intervention.

171

Thesis Study Map





Objectives





the community.




initiative.

Key Findings:










adults.




adults.

Objectives









Key Findings


172






respectively.




respectively.










Objectives




health indicators.



health outcomes.

Key Findings





outcomes.










173

evident.




adults.

Objectives

To evaluate the effectiveness of Sport England’s

Get Healthy Get Active physical activity

intervention on older adults physical activity,

sedentary behaviour and self-reported health

indicators.

Key Findings:

The Get Healthy Get Active physical activity

intervention was effective in increasing quality of

life, self-rated health, self-assessment of physical

fitness, and self-efficacy for exercise scores over

time after adjustment for covariates.

There was no significant intervention effect on

time spent in moderate-to-vigorous physical

activity.

The intervention also led to a significant increase

in sitting time across all three follow-up time

points.

Study 5. Implementation fidelity of the Get



adults

Objectives:

To evaluate whether or not the GHGA multi-

component intervention was implemented as

intended.

To evaluate sustainability of the GHGA multi-

component intervention in terms of its feasibility

and acceptability of long-term implementation

across multiple settings.

174

Chapter 7

Study 5: Implementation fidelity of the Get

Healthy Get Active physical activity programme

for community-dwelling older adults.

175

7.1. Introduction

Results from Chapter 6 (Study 4) revealed that Sport England’s GHGA PA

intervention was ineffective at increasing MVPA levels across all three follow-up time

points. Significant increases in sitting time were also observed across the follow-up

time points. The current study provides quantitative and qualitative data to assess

the implementation fidelity of the GHGA PA intervention. It is important to

understand how the GHGA intervention was implemented in practice so that

changes in the behaviours of intervention participants can be attributed to the novel

remedy that the intervention represents rather than to variations in the delivery and

receipt of the intervention. Efficacious interventions can then be scaled up in order

to inform the thought of policy and practice of professionals in PA, social work and

care settings.

Intervention research in the field of PA in older adults has primarily focused on pre-

and post-intervention measurements and less on longer term follow-up

measurements after intervention completion (McMahon et al., 2017). Follow-up

measures post-intervention are critical for understanding implementation

sustainability and maintenance patterns (McMahon et al., 2017). Home-based,

group-based, community-based, and educational whole-system oriented multi-

component PA interventions can result in both short- and long-term increases in PA

(McMahon et al., 2017). Findings from a recent Pedometer Accelerometer

Consultation Evaluation (PACE)-Lift Cluster RCT (Harris et al., 2015; Harris et al.,

176

2018) showed at 3-months that both average daily step-counts and weekly MVPA in

≥10 minute bouts were significantly higher in the intervention than control group: by

1,037 (95% CI 513–1,560) steps/day and 63 (95% CI 40–87) minutes/week,

respectively. At 12-months corresponding differences were 609 (95% CI 104–1,115)

steps/day and 40 (95% CI 17–63) minutes/week (Harris et al., 2015), and at 4-years

post-baseline versus control results revealed sustained intervention effects resulting

in: +407 (95% CI: −177±992), p = 0.17 steps/day, and +32 (95% CI: 5±60), p = 0.02

minutes/week MVPA in ≥10 minute bouts in the intervention compared to the

control group, respectively. A systematic review of reviews by Zubala et al. (2017)

found that PA interventions among community-dwelling older adults often resulted

in sustained improvements in PA over the study period, typically at 12 months.

However, effects on maintenance beyond 12 months remains unclear, due to a lack

of high quality longitudinal studies (Olanrewaju et al., 2016; Richards et al., 2013;

Zubala et al., 2017). Consequently, only a minority of interventions move from

research into practice, and those that do provide limited information on

sustainability or institutionalisation within routine practice (Reis et al., 2016). Across

three decades of PA intervention research, the majority of publications have been

efficacy/effectiveness trials and only 3% comprised dissemination studies

(Gottfredson et al., 2015). This continued lack of evidence for the successful

institutionalisation of PA interventions in real-world settings, combined with

unacceptably high levels of PA inactivity worldwide (Hallal et al., 2012; Kohl et al.,

2012) makes addressing the research-to-practice gap a significant public health

priority (Koorts et al., 2018). It is recommended therefore, that process evaluations

177

of intervention implementation and fidelity become an integral part of the conduct

and evaluation of all health behaviour intervention research (Castillo et al., 2017).

Intervention fidelity is the degree to which an intervention is implemented as

intended by its developers and ensures that the intervention maintains its intended

effects (Carroll et al., 2007). When research is inattentive to fidelity, changes in the

behaviours of intervention participants can be attributed to variations in the delivery

and receipt of the intervention just as plausibly as they can be credited to the novel

remedy that the intervention represents (Bellg et al., 2004). It is now widely

acknowledged that it is important that interventions are studied in terms of their

implementation and fidelity, as this process evaluation research can improve

understanding of how interventions have been implemented in practice, so that they

can be further integrated into ‘real world’ community settings (Bellg et al., 2004;

Oakley et al., 2006; Craig et al., 2008). Process evaluation of intervention delivery is

deemed to be particularly important in multi-component interventions delivered

across varying locations where the same intervention may be implemented and

received in different ways (Koorts et al., 2018).

Assessing intervention fidelity has been identified to be a key challenge for health

behaviour interventions (Koorts et al., 2018). Public health impact is dependent on

the extent to which efficacious PA interventions are disseminated with fidelity into

real world settings, then maintained, and institutionalised (Lewis et al., 2017). If an

178

intervention is not implemented as directed and no effect is found, then one cannot

be sure whether this is due to lack of efficacy of the intervention or simply that it has

not been implemented correctly (Hasson, 2010). When analysing fidelity of a large-

scale community-based PA intervention study, Hardeman et al. (2008) found that

facilitators delivered only around 44% of the specified intervention techniques

across four key sessions. It is recommended, therefore, that evaluations of

intervention fidelity become an integral part of the conduct and evaluation of all

health behaviour intervention research (Castillo et al., 2017). The process evaluation

of interventions is advocated by the SEF, which deems it to be an essential part of

designing and testing multi-component interventions (National Obesity Observatory,

2012). Assessment of fidelity requires a mixed methods approach, using quantitative

and qualitative methods to understand processes which influence implementation,

and their variation across contexts (Moore et al., 2015).

Assessing the degree of fidelity for intervention design and implementation is a

critical feature in translating research-based studies with positive outcomes into

successful programmes (Frank et al., 2008). Despite this recommendation, there has

been considerable heterogeneity and variability in the conceptualisation and

measurement of intervention fidelity in the quality of measurement of delivery

fidelity in interventions promoting PA (Lambert et al., 2017; Quested et al., 2017).

Fidelity assessment includes appraisal of the intervention itself, by addressing core

elements of treatment integrity and treatment differentiation (Calsyn, 2000;

Moncher & Prinz, 1991). The characteristics and actions of the deliverer(s) also need

179

to be scrutinised. These include adherence (accuracy in delivering the components of

an intervention) and competence (the ability of the deliverer to engage the

participants effectively) (Santacroce, Maccarelli & Grey, 2004). Increasingly

sophisticated conceptual models of fidelity measurement have been developed and

tested (Pérez et al., 2015; Moore et al., 2015). A comprehensive treatment fidelity

framework specifically developed to provide guidance for the assessment,

enhancement and monitoring of fidelity for tailored health behaviour interventions

is the NIH BCC framework (Bellg et al., 2004). The BCC framework conceptualises

fidelity across five core domains: Study Design, Provider Training, Intervention

Delivery, Intervention Receipt and Enactment. Assessing all these elements enables

more accurate inferences to be made about programme effectiveness, and (if

appropriate) any implications for wider roll out/implementation (Dane & Schneider,

1998). The model has been previously adopted among health behaviour

interventions in older adults (Chiang et al., 2006; Quijano et al., 2007) and provides a

set of guidelines for translating research into practice and improving the successful

implementation of interventions into real world settings (Demiris et al., 2014). For

these reasons the NIH BCC framework (Bellg et al., 2004) was adopted in the current

study to assess programme fidelity of the Get Healthy Get Active (GHGA) multi-

component PA intervention. In line with thesis objectives 8 and 9, this study aimed;

to evaluate whether the GHGA multi-component intervention was implemented as

intended, and evaluate sustainability of the GHGA multi-component intervention in

terms of its feasibility and acceptability of being implemented and incorporated in

the long-term across multiple settings.

180

7.2. Methods

The GHGA PA intervention was a pragmatic, quasi-experimental study with repeated

follow-ups, delivered from January 2016 and July 2018 and conducted in various

locations throughout Sefton Borough in the North West of England. The intervention

aimed to engage inactive older adults in PA at least once a week for 30 minutes, via a

12 week PA intervention. The project was funded by Sport England and delivered by

SMBC. The characteristics of the GHGA PA intervention were in line with the

CONSORT checklist for pragmatic evaluations (Zwarenstein et al., 2008). A detailed

description of the intervention can be found in Chapter 6 (Study 4), but a brief

summary is provided here. Funded by Sport England and delivered by SMBC, the

GHGA PA intervention was a three-year project aimed at engaging inactive older

adults in PA at least once a week for 30 minutes, via a 12 week PA intervention. The

intervention was implemented throughout Sefton Borough within differing locations

(e.g., leisure centres, a church hall, a theatre, a retirement homes, and a library) with

each 12 week PA intervention implemented at the same venue. However, deliverer

retention throughout the entirety of each 12 week PA intervention was not possible

due to competing time demands, holidays and sickness of deliverers. The potential,

therefore, for the fidelity of the intervention to vary considerably both across

separate locations, and within the same location was high.

181

Given the varying levels of functional ability and physical and psychosocial health

among this population (Van Cauwenberg et al., 2016), the exercises within each

session were designed to be flexible (e.g., variations of the same exercise each with a

differing difficulty) in order to meet the needs of all participants and the varying

needs of deliverers themselves (e.g., if a deliverer was injured and could not perform

certain activities). Consequently, the precise standardisation of the delivery of the

intervention was neither desirable nor feasible. Thus, it was predicted that there

would be variation in both the delivery of the intervention (by deliverers) and the

response to the intervention (by participants), and therefore the fidelity of the

intervention. This study draws upon various quantitative and qualitative session

observation and qualitative interview data sources to provide a comprehensive

exploration of GHGA PA intervention fidelity. A total of 43 mixed-gender GHGA

session observations, and 21 GHGA deliverer semi-structured interviews were

completed by the author.

7.2.1. Design

A mixed-methods research design was adopted as adopting both quantitative and

qualitative methods in combination can enable a deeper understanding of

programme implementation and maintenance to be gained (Moore et al., 2015).

7.2.2. Measures

7.2.2.1. Session Observations

182

Due to the nature of the intervention and logistical and time constraints of the

research staff, randomisation of sessions was not possible and consequently, a

convenience sample of 43 mixed-gender GHGA session observations took place

between June 2016 and December 2017. Given the demographically and

geographically local nature of the intervention sessions, this type of sampling is

acceptable in qualitative research (Robinson, 2014). Observations adhered to

Merriam’s framework for session observations (Merriam, 1998) and were completed

in a contrived (non-natural setting), non-disguised, human, direct, structured

manner. Merriman’s framework conceptualises session observations across four

core domains. Firstly, frequency counts of the physical environment (e.g., number of

participants, size of the space, number and type of equipment available, and number

and type of activities performed) are noted. Participants’ and deliverers’ perceptions

of delivery, organisation, venue, facilities, timings and engagement are also recorded

by the first author who was knowledgeable and experienced in the requirements of

delivering a successful GHGA PA session. Activities and interactions between the

participants and the deliverer(s) including the intervention itself (e.g., delivery,

content and structure), the deliverer (e.g., competence, adherence to

objectives/exercise content, consistency and enthusiasm), and perceived participant

enjoyment are then taken into account. Finally, frequency and duration of any other

subtle factors (e.g., unplanned activities, symbolic meanings, nonverbal

communication, physical clues, and what should happen that has not happened) are

noted. A total of ~60 hours of sessions observations were aggregated and analysed

in Microsoft Word (version 2013, Microsoft Corporation, Redmond, WA, US)

resulting in 215 pages of typeset data with Arial font, size 12, double-spaced. All

183

participant and deliverer data were anonymised and coded throughout the

transcripts to ensure confidentiality. Session observations served to triangulate

interview findings and thus, aid the latter stages of the analysis by exploring further

intervention fidelity.

7.2.2.2. Deliverer Interviews

Due to time constraints of both the GHGA deliverers and the first author, a

convenience sample of seven GHGA intervention deliverers participated across 21

interviews, with all seven deliverers participating in at least one interview.

Specifically, one deliverer took part in six, one deliverer in four, one deliverer in

three, and finally four deliverers in two interviews, respectively. Convenience

sampling has been adopted in older adult interview research previously (Van

Cauwenberg et al., 2018) and thus, the current study extends the applicability of this

method within this population. Interviews took place between June 2016 and

November 2017. Deliverer interviews were conducted using a semi-structured

interview guide including open- and closed-ended items. Interviews included 12

questions with probes and follow-up questions used as needed. Guide development

was informed by Merriam’s framework for session observations (Merriam, 1998) and

consequently, elicited information in line with the four core domains of, physical

environment, participants and deliverer perceptions, activities and interactions

between both the participants and the deliverer(s), and the frequency and duration

of any other subtle factors. An example question was: ‘Are the facilities appropriate

for what you need in order to fully complete the session? Please explain.’ To

184

maximise the interaction between the deliverer and the first author, interview

questions were reviewed by the project team for appropriateness of question

ordering and flow prior to the deliverer interviews. All interviews were led by the

first author. An example question from a section exploring interactions between

participants and deliverers was: “How would you rate participant engagement with

this session out of 10? Please explain.” Questions therefore demonstrated aspects of

face validity as they were transparent and relevant to both the topic and target

population (French et al., 2015). Interviews took place within multiple venues,

multiple times including two leisure centres (x5 interviews), a church hall (x8

interviews), a theatre (x3 interviews), a retirement home (x2 interviews), and a

library (x3 interviews). All locations were free from background noise, and deliverers

could be overlooked but not overheard. Interviews averaged 14 minutes in duration

(ranging from 11 to 23 minutes), were digitally recorded, and transcribed verbatim.

Transcribed data was aggregated and analysed in Microsoft Word (version 2013,

Microsoft Corporation, Redmond, WA, US) resulting in 70 pages of typeset data with

Arial font, size 12, double-spaced. The text for each interview was sequentially

labelled with numbers to identify the sentences that belonged to the participant or

interviewer (Silverman, 1994). All data were anonymised and coded throughout the

transcripts to ensure confidentiality.

Verbatim transcripts were read and re-read to allow familiarisation of the data.

Participants of the GHGA PA intervention received a covering letter, participant

information sheet, and consent form. Prior to the commencement of the study,

185

institutional ethical approval was received (#SPA-REC-2015-329) and written

informed consent was obtained for all participants prior to participation.

7.2.3. Data Coding and Analysis

Previous research within this population has adopted analytical procedures including

thematic analysis (Van Dyck et al., 2017), content analysis (Middelweerd et al., 2014)

and used specialist qualitative data analysis packages, such as NVivo (Warmoth et al.,

2016). Deductive content analysis (Braun & Clarke, 2006) was initially adopted to

categorise session observation and interview data into a priori themes from

Merriam’s session observation framework (Merriam, 1998). Inductive analysis then

allowed for emerging themes to be created beyond the pre-defined categories.

As well as being used as a tool to shape treatment fidelity plans in intervention

development (Sineat et al., 2017), the NIH BCC framework (Bellg et al., 2004) has

also been adopted to assess and evaluate treatment fidelity of PA interventions in

both older adults (Frank et al., 2008) and adults (Lambert et al., 2017). Consequently,

to exemplify operationalisation of the NIH BCC framework, relevant a priori and

emergent themes obtained from both interview and session observation data were

retrospectively applied into the five core fidelity domains of: Study Design, Provider

Training, Intervention Delivery, Intervention Receipt and Enactment. Study design is

concerned with whether a study adequately tests its hypotheses in relation to its

underlying theoretical and clinical processes. Quality criteria includes delivery and

186

adherence to the study protocol, and ensuring an environment where the protocol

can be fully operationalised (Lambert et al., 2017). Provider training involves

standardising training between providers and ensuring they are trained to clear

criteria and monitored over time (Bellg et al., 2004). Quality criteria includes both

adherence to treatment components and adherence to process (e.g., interactional

style) (Lambert et al., 2017). Intervention delivery involves assessing and monitoring

deliverer differentiation (differences between the intervention and any comparison

treatments), competency (skills set of deliverer), and adherence (delivery of

intended components) to treatment components and competence to deliver the

treatment in the manner specified (Frank et al., 2008). Assessment of non-specific

treatment effects (e.g., infrastructure) should also be noted (Lambert et al., 2017).

Intervention Receipt refers to whether the intervention was understood and

‘received’ by participants (Bellg et al., 2004). Assessment of participant receipt can

involve pre-post tests and verbal/non-verbal cues (Lambert et al., 2017). Fidelity to

treatment enactment refers to intervention sustainability and in particular, whether

participants used intervention related skills in day to day settings (Bellg et al., 2004;

Borrelli, 2011; Borrelli et al., 2005). Assessment includes objective observations and

psychometric properties (Lambert et al., 2017). Verbatim quotations were

subsequently used to provide context and verify participant responses. Quotations

relating to the deliverer interviews were labelled by interview number (In), and

subsequent deliverer number (Dn) within that interview. Characterising traits of this

protocol include details of frequency counts of the physical environment, and

extracts of verbatim quotes to provide context to the a priori and emergent themes.

The pen profile approach adopted in Chapter 3 (Study 1) was not adopted for the

187

current study as the minimum threshold for theme inclusion criteria required for this

approach would have severely reduced the already limited data available.

Methodological rigour was demonstrated through a process of interrater reliability

(McHugh, 2012) whereby coding checks were undertaken for a 10% random sample

of all data collected from observations (n =5) and interview transcripts (n =2)

independently by a member of the project team. This involved cross-checking

placement of data into the four main themes, within 19 sub-themes obtained from

Merriam’s framework for session observations (Merriam, 1998), and then into the

five core themes associated with the NIH BCC framework (Bellg et al., 2004). A total

of nine emergent themes were identified which were placed within the appropriate

main theme within Merriam’s framework for session observations (Merriam, 1998),

and then if relevant, within the appropriate domain within the NIH BCC framework

(Bellg et al., 2004). Any omissions and discrepancies with coding analysis were

identified and discussed until subsequent agreement on data themes in relation to

verbatim extracts was reached. Agreement for coding of themes from the data

ranged from 84% to 95% across the seven complete data sets. 80% and above

agreement is considered acceptable (McHugh, 2012). This process ensured

transparency, credibility and trustworthiness of the results (Smith & Caddick, 2012).

7.3. Results

188

In total, seven GHGA intervention deliverers and 388 older adults (54 male)

participated across the 21 GHGA deliverer semi-structured interviews (mean number

of participants per session =1.1, SD= 0.3) and 43 session observations (mean number

of participants per session= 9.02, SD =4.08), respectively.

7.3.1. Fidelity to study design

7.3.1.1. Number and type of exercises included

Although deliverers often did not have a specific session plan outlining specific

exercises for each individual session, deliverers noted that they had to include

exercises which targeted five core aspects of fitness including balance, endurance,

flexibility, resistance, and strength exercises, whilst also incorporating a warm-up

and a cool-down.

“So there is a warm-up, endurance, resistance, balance, strength, flexibility, and then a cool-down.” (In12, Dn3, Lines 69-70).

Table 7.1 shows that 93.1% of GHGA sessions delivered the required five core

components, and the associated warm-ups and cool-downs. Deliverers also noted

that they were given flexibility to introduce essential exercise components of varying

difficulty, thus allowing all participants regardless of ability, to engage with the

entirety of the session protocol.

“Yeh it’s nice to see people moving onto more difficult exercises or even just getting out of the chair if they couldn’t do it before coming.” (In6, Dn7, Lines 54-55)

189

“…there are different progressions people can do like increasing or decreasing

resistance in the exercise bands or holding onto the chair or not during the balance

exercises and again they (GHGA participants) know to go at their own pace and stop

if anything starts to hurt or anything.” (In9, Dn5, Lines, 51-54)

7.3.1.2. Environment

Deliverers emphasised the usefulness of the pilot-phase of the project in aiding with

session set-up, timings, adherence to essential protocol elements, and overall

session environment.

“Yes we had practice sessions before we fully started which has made the running of the actual sessions much smoother.” (In1, Dn2, Lines 4-5).

7.3.2. Fidelity of provider training

7.3.2.1. The intervention itself

Deliverers noted that they were trained by SMBC employees who were experienced

in delivering chair-based PA sessions.

“I only started on the programme around 6 months ago and didn’t have any training on how to deliver chair based exercises but in only 8 weeks I was trained and now I am delivering classes around Sefton on my own.” (In11, Dn7, Lines 3-5).

“That’s the good thing about having set exercises is that in only a month or so I went from not teaching anything to being fully trained and running sessions on my own. It is scary but enjoyable now at the same time.” (In8, Dn7, Lines 103-105).

190

Deliverers also said how much participants were progressing which meant that

deliverers were soon expected to devise sessions of greater difficulty in order to

keep participants engaged and subsequently benefitting from the sessions.

“At the beginning (of the GHGA PA intervention delivery) we were told we would just be doing chair based exercises in care homes but it is very different to that and we now do walking and balance and more advanced things which I’m not really trained for.” (In10, Dn5, Lines 83-85).

7.3.2.2. Deliverer experience

A limited number of deliverers were trained and subsequently conducted the

sessions throughout the programme’s entirety. Consequently, all deliverers were

experienced in all aspects of session delivery.

“I have session plans for all of them but because I have done the sessions for a long time now, I know them off the back of my hand, but there is a set thing we have to do.” (In13, Dn3, Lines 68-69).

However, it was noted that the limited number of deliverers available had an impact

on the number of participants that could attend each session.

191

“I can only have a limited number of participants because I wouldn’t be able to observe them all carefully enough if we had too many… there is only me delivering we need more deliverers.” (In13, Dn3, Lines 95-97).

7.3.2.3. Rapport

Deliverers noted the importance of building rapport with the participants and in

particular how much they enjoyed delivering the sessions.

“Yes, it is a good group! It is an engaged group there is a lot of feedback from the group and we have a laugh and a joke. Good group to engage with, one of the best I would say.” (In14, Dn1, Lines 33-35).

“…after the first 12 weeks we actually did satisfaction forms with them and each one of them pretty much rated it excellent which is a good sign. They were all anonymous so they could of wrote anything down and we wouldn’t know who said what but they all came back said it was positive, and you can tell by their body language as when they come they all love to see each other and love having a little gathering seeing what has happened over the week. So I do generally think they are positive and because they are coming back as some of them have been there for 6 months now and they come back every week that has to be a good sign.” (In15, Dn3, Lines 130-137).

7.3.2.4. Motivation

Deliverers talked about the freedom they were given to introduce a variety of

exercises of varying levels of difficulty. This kept deliverers themselves engaged and

motivated.

192

“I try to do different things (exercises) each week so it’s not the same it keeps me excited as I’m not doing the same thing over and over and it keeps people (GHGA participants) working different areas as for some this will be the only exercise they do all week!” (In5, Dn1, Lines 44-46).

“Everybody is really pushing themselves in this class and although there are lots of different levels of ability and balance and flexibility in the class… they are moving more freely and their balance has improved a lot as I keep making it harder for them!” (In8, Dn7, Lines 19-23).

7.3.3. Fidelity of treatment delivery

7.3.3.1. Delivery

Many deliverers talked about the different ways in which they adapted the exercises

to fit the characteristics and needs of the participants whilst still adhering to the

GHGA PA intervention protocol.

“In any group you will have people who perhaps can’t do certain exercises and that is mainly due to us having such varying levels of ability of participants turning up so we try and cater for all as there are always lesser and harder variations of exercises we can do to keep all involved.” (In4, Dn4, Lines 19-22).

Some also noted the importance of knowing the health history of class members to

make sure exercises are appropriate and safe.

“We have to do health checks with them all to see if there is anything restricting them in exercise, certain medication can also affect the exercises they can do but we

193

will always give adaptions (of exercises) for them so regardless of what is going on we can give adaptions”. (In15, Dn3, Lines 169-172).

7.3.3.2. Venue

Deliverers noted how the on the whole the venues used were suitable in both size

and location for the target population and allowed for the sessions to be completed

fully.

“Yes, I like it, it’s good! The space is accessible and its quite central in the area, I don’t know the area too much but everyone seems to know it and how to get here.” (In14, Dn1, Lines 22-23).

“Yes I think it’s perfect really it’s a nice cheap room it’s plenty big enough for what we need it for.” (In1, Dn6, Lines 30-31).

Table 7.1 shows that there was sufficient room space to fully complete required

GHGA session components 93.1% of the time. Some venues however were deemed

to be too small and inaccessible, which impacted upon session delivery, especially

the elements which required more space such as the walking balance exercises.

“We were in a small room for the last couple of weeks… so it wasn’t good enough, wasn’t big enough, couldn’t actually operate the session very well.” (In6, Dn7, Lines 7-9).

194

“…it’s (the venue) a little dark and it’s upstairs.” (In9, Dn5, Line 56).

7.3.3.3. Facilities and Equipment

Deliverers spoke highly of the facilities and equipment available for each session.

“…there are plenty of chairs in the room and also the weight box of dumbbells which is good and also the good speakers in the corner of the room.” (In10, Dn5, Lines 59-60).

Table 7.1 shows that sufficient equipment was available for 95% of the sessions.

Kitchen facilities were also noted as being important given the strong focus of

socialisation portrayed by GHGA and were available throughout 88% of the sessions.

“…the kitchen is brilliant for making the teas and coffees at the end (of the session).” (In8, Dn7, Lines 50-51).

“I think that everything is there to be honest because they have got the kitchen as well, the kitchen is a good idea for when we are doing tea and coffee and it’s obviously got the kettle facilities and stuff.” (In15, Dn3, Lines 96-98).

7.3.4. Fidelity of receipt of treatment

7.3.4.1. Confidence

195

Participant confidence in performing everyday tasks was observed via the timed up

and go test (Podsiadlo & Richardson, 1991) at baseline, and at 12-weeks post-

baseline assessment.

“We do an assessment at the beginning (first session of the 12-week GHGA PA intervention) and after 12-weeks and each participant has progressed massively in both confidence and the physical element of it… I think the majority of them improved by about 7 seconds on the walking (based on the timed up and go test; Podsiadlo & Richardson, 1991) and their ability to actually get up out the chair improved massively, they’re all quite confident now to get up out of the chair.” (In15, Dn3, Lines 86-92).

7.3.4.2. Perceived participant engagement and enjoyment

As well as being physically tested, participants noted their progress to deliverers

verbally, allowing deliverers to further assess their understanding and progress.

Specifically, deliverers talked about how the flexibility of the session protocol and

their delivery styles themselves allowed participants of all abilities to perform,

understand and enjoy the entirety of each session.

“Nobody is left behind or left with nothing to do as almost every exercise can be adapted so all can be involved… We describe the exercise and show it before the participants do it so all know exactly what to do and can get involved.” (In10, Dn5, Lines 127-129).

“When people actually come to the session more often than not they keep coming as they enjoy it so much!” (In7, Dn6, Lines 88-89).

196

Table 7.1 shows participant fidelity to receipt of treatment on a scale from 1-10, with

10 representing full participant engagement and 1 representing no engagement.

Deliverers reported average participant engagement throughout the sessions to be

8.1. Furthermore, participants themselves rated overall session enjoyment (based

upon the four main themes outlined in Merriam’s session observation framework)

on a scale from 1-10, with 10 representing excellent enjoyment and 1 representing

no enjoyment. Data collected from 388 older adults revealed an overall mean score

of 8.6 (SD 2.21).

7.3.4.3. Retention

Deliverers also talked about the high rates of participant retention throughout the

12-week GHGA PA sessions.

“Yes the same participants keep coming back which is good and we are slowly getting more participants coming. The sessions are really motivating and you can see the benefits to the participants even after just a few short weeks.” (In10, Dn5, Lines 98-99).

“I can really push participants and they thrive off the energy. Participants seem hooked once they have been to one session as the same faces keep coming back… people just don’t believe how much it actually helps them.” (In1, Dn2, Lines 52-56).

7.3.5. Fidelity to treatment enactment

7.3.5.1. Benefits to health

197

Deliverers noted that GHGA measured falls efficacy, falls management, and falls

control using self-report scales at baseline and 12-weeks post-baseline.

“From the falls questionnaires we can see participants are increasing mobility reducing falls and improving their flexibility and you know their strength as well which is fantastic as it means they will be able to stay independent for longer and that is kind of the aim of these sessions.” (In7, Dn6, Lines 18-20).

Deliverers also talked about how participants themselves recognised the benefits

that the sessions were having and how this was helping participants retain their

independence outside of the sessions and consequently, reducing burden on

healthcare services.

“…one participant said they had a problem with their ankle but that is now getting better after coming to these sessions and keeping people healthy puts less pressure on the NHS.” (In14, Dn1, Lines 46-47).

7.3.5.2. Socialising

The social aspect of the GHGA PA sessions was noted by deliverers as being a key

aspect by both deliverers and participants.

“Yes definitely I’d like to think they (GHGA participants) enjoy the sessions and the social aspect which is the tea and coffee after which everyone stays for and brings biscuits and things. Participants have made new friends which they keep in touch with outside of these sessions which is fantastic.” (In9, Dn5, Lines 40-43).

198

Table 7.1. Frequency counts and descriptives.

7.3.5.3. Sustainability

Deliverers talked about wanting to set up paid-maintenance sessions after the initial

free 12-week block of sessions as despite being given leaflets and DVDs outlining the

GHGA session components, participants were not utilising these outside of the

sessions themselves.

“I mean a lot of them (GHGA participants) don’t use the bands or do the exercises at home even though we give out DVDs and booklets explaining and showing the exercises we do so it’s important to keep these sessions going if we can do especially as these participants are willing to pay for the sessions.” (In7, Dn6, Lines 29-32).

Session Observation

Number

Number of

participants (male)

Sufficient room space

to fully complete required

GHGA session

components

Sufficient equipment available

for a complete session

(e.g., chairs, exercise bands)

Kitchen Facilities

Core components completed

(%)

Perceived participant

engagement by GHGA deliverers

(/10)

Perceived participant

session fidelity (/10)

1

2

3

4

5

6

7

8

9

10

11

23 (3)

20 (0)

14 (1)

12 (1)

16 (0)

11 (3)

8 (0)

10 (2)

3 (0)

3 (0)

5 (1)

No

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

No

No

Yes

Yes

Yes

Yes

No

No

No

Yes

Yes

No

No

Yes

80%

100%

100%

100%

100%

100%

100%

100%

100%

100%

100%

7

7

8

7

8

7

7

7

8

9

9

8

8

7

9

10

8

7

8

8

10

8

199

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

7 (2)

7 (3)

8 (1)

4 (0)

4 (0)

9 (0)

11 (3)

7 (2)

10 (1)

10 (1)

7 (2)

11 (2)

9 (1)

9 (1)

7 (0)

7 (2)

11 (2)

8 (1)

9 (1)

5 (1)

10 (2)

3 (0)

9 (1)

7 (2)

9 (1)

8 (1)

7 (2)

7 (2)

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

No

No

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

100%

100%

100%

100%

100%

100%

100%

100%

100%

100%

100%

100%

100%

100%

100%

100%

80%

80%

100%

100%

100%

100%

100%

100%

100%

100%

100%

100%

10

10

10

7

8

7

7

9

9

9

8

9

8

8

9

7

7

7

8

7

8

9

7

7

8

7

10

8

9

9

9

8

10

10

9

8

8

9

7

9

9

10

10

10

7

9

9

9

9

8

9

10

10

8

8

7

200

40

41

42

43

7 (2)

12 (2)

9 (0)

15 (2)

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

100%

100%

100%

100%

10

10

8

8

9

7

8

9

7.4. Discussion

Through the adoption of a comprehensive treatment fidelity framework developed

by the BCC for tailored health behaviour interventions (Bellg et al., 2004), this study

aimed to evaluate whether the GHGA multi-component PA intervention was

implemented as intended with a view of better understanding the results outlined in

Chapter 6 (Study 4).

Study design fidelity ensures procedures are put in place to ensure equivalent

content both within and across conditions, as well as creating plans to deal with

possible setbacks during implementation (Frank et al., 2008). Results revealed that

although 93.1% of GHGA sessions adhered to and included a set structure of

exercises targeting the five core aspects of fitness (endurance, resistance, balance,

strength, and flexibility), variation of session content and session exercises were also

built into the sessions. This resulted in varying numbers, types and timings of

exercise components in every session, even between sessions delivered by the same

GHGA deliverers. However, given the varying levels of functional ability and physical

and psychosocial health associated with community-dwelling older adults (Van

Cauwenberg et al., 2016), these results support the idea that a one size fits all

201

approach to multi-component interventions is not appropriate (Hawe, Shiell & Riley,

2004). A strict protocol consisting of the same exercises for all participants regardless

of ability may have resulted in decreased participant engagement, motivation and

subsequent retention (Rose, 2018). Consequently, the core components of each

session were included but with a degree of variation whereby the same exercise

could be delivered at differing difficulties to enable all participants to successfully

take part regardless of ability. Previous research has also advocated for certain levels

of flexibility and progressions in session content based upon participant requests and

levels of ability given that such serves to allow better tailoring of the intervention to

the local context (Lawton et al., 2014). However, results from Chapter 6 (Study 4)

showed that despite the varied and flexible sessions, the GHGA PA intervention was

ineffective at increasing MVPA levels. Hence, delivering a block of sessions with

consistent content and delivery techniques may better allow researchers to

understand the approaches that are both effective and ineffective in eliciting PA

behaviour change among this population (Chase, 2014). Incorporating both

quantitative and qualitative measures of intervention fidelity through

comprehensive frameworks such as Merriam’s framework for session observations

(Merriam, 1998) and NIH BCC’s framework (Bellg et al., 2004) can allow future

researchers to accurately measure both session content and delivery consistency

and consequently, whether the intervention is efficacious to PA behaviour change.

Fidelity to provider training ensures that deliverers are satisfactorily trained to

deliver the intervention to participants (Frank et al., 2008). Training practitioners to

202

faithfully deliver multi-component interventions is a major challenge and thus,

assessment and ongoing evaluation of those who implement the programme is a key

element of fidelity as this ensures that deliverers have been satisfactorily trained to

deliver the programme as intended to participants (Bellg et al., 2004). GHGA

deliverers noted that they were trained by SMBC employees who were experienced

in delivering chair-based exercise sessions. This ensured that deliverers had baseline

knowledge of exercise science and safety, which they then built upon further whilst

gaining experience delivering the GHGA sessions themselves. Session observation

findings provided further agreement for fidelity to provider training given the sound

knowledge and descriptions of each exercise provided by GHGA deliverers during the

sessions. Effective deliverers as noted by both intervention participants and the first

author were those who; provided clear, concise instructions both before and during

each exercise, demonstrated each exercise both verbally and visually, performed the

exercises with the participants and therefore provided a reference for required

speed and intensity, and set out a target for participants during each set (i.e. number

of reps, time). GHGA deliverers also received a GHGA instructor manual, detailing

the exercises to be included within the sessions. A DVD outlining the exercises was

also available for the deliverers. A previous evidence-based group exercise

intervention for older adults (EnhanceFitness) also noted that providing deliverers

with detailed scripts, descriptions, and guidelines for each intervention component

could increase fidelity to provider training (Quijano et al., 2007).

203

The importance of deliverer engagement and motivation has also been identified as

a key determinant affecting fidelity to provider training (Schoenwald et al., 2011).

Session observations revealed that all deliverers were fully engaged and motivated

to deliver GHGA sessions due to their strong beliefs in the potential benefits of the

GHGA intervention to participant’s physical and psychosocial health. Consequently,

fidelity to provider training was further ensured as those who believe in the value of

the intervention are more likely to fully engage with the training (Castillo et al.,

2017).

Fidelity to treatment delivery is considered the ‘heart of fidelity assessment in

behavioural interventions’ (Gearing et al., 2011, p.82) but has historically been

insufficiently considered (Miller & Rollnick, 2014). Treatment delivery is crucial to

ensure intervention results are truly attributable to the programme (internal validity)

and that the results are generalizable to other study populations (external validity)

(Frank et al., 2008). In line with previous interventions in older adults (Quijano et al.,

2007; Tennstedt et al., 1998), GHGA monitored treatment delivery through onsite

observations of new deliverers throughout the 12 weeks by senior SMBC staff

experienced in the design and structure of the GHGA sessions. Staff involved in

GHGA from the pilot stage were not observed. However, participant satisfaction with

delivery was assessed for all deliverers after 12 weeks through a written programme

evaluation, which asked about satisfaction with the programme and the deliverer. It

is to be expected that deliverers potentially became more proficient in delivery with

increased experience throughout the 12 week intervention and consequently, future

204

process evaluations should assess participant satisfaction across time (e.g., baseline,

three-, six- and 12-weeks) (Thompson, Lambert, Greaves, & Taylor, 2018).

Infrastructure is also a key aspect of treatment fidelity and includes venue location

and size, availability of equipment and materials, and session timing (Petrescu-

Prahova et al., 2015). GHGA sessions were implemented throughout Sefton Borough

within several differing locations (e.g., leisure centres, a church hall, a theatre, a

retirement homes, and a library). Deliverers noted that on the whole the venues

adopted were in locations that were suitable for participants, and 93.1% were of a

size which allowed for the full completion of the required GHGA session

components. Session observations provided further agreement for the size of

venues, however participants noted that locations were often not suitable to be

reached by public transport. This could have affected session numbers as frequency

and reliability of affordable public transport are all associated with decreased PA

participation (Newitt et al., 2016). Treatment fidelity was further ensured through

the sufficient availability of equipment at each venue which was either provided by

the venue itself (e.g., chairs and music systems) or by SMBC (e.g., fitness bands,

ankle weights and dumbbells).

Fidelity related to receipt of treatment concerns both documenting participant

exposure to the treatment and the ability of participants to understand and perform

treatment-related activities and strategies during treatment delivery. In GHGA,

205

participant confidence in performing everyday tasks was observed via the timed up

and go test (Podsiadlo & Richardson, 1991) at baseline, and at 12-weeks post-

baseline. This measure, along with participants informally noting their progress to

deliverers verbally at the end of each session, ensured participants were able to

comprehend and perform the exercises as instructed (Frank et al., 2008). Session

observations provided further evidence of participant receipt and engagement. As is

recommended in the NIH BCC guidelines (Bellg et al., 2004; Lambert et al., 2017),

GHGA deliverers demonstrated each exercise both verbally and visually throughout

the sessions in order to ensure participant comprehension of each exercise.

Participant confidence in performing the behaviours and success in meeting goals

was also assessed during classes through videotapes of participants at baseline and

12 weeks post-baseline (Lambert et al., 2017). Concurrent with recent PA

intervention research in older adults, (Lyons et al., 2017), trained GHGA deliverers

monitored “dose” by tracking older adults’ participation in programme activities

(e.g., attendance) and those who discontinued (e.g., dropped out) to assess level of

receipt. The subsequent high rates of participant retention throughout the 12 week

GHGA PA sessions further solidified the efficacy of receipt of treatment.

Fidelity to treatment enactment concerns the participants’ ability to implement the

learned skills and activities in real world settings (Frank et al., 2008). The repeated

measures design of the research element of the GHGA PA intervention allowed for

fidelity strategies to track participants’ implementation of the learned behaviours,

skills, and cognitive strategies presented in the GHGA PA sessions in relevant real

206

world settings at three-, six- and 12-months post-baseline via self-reported physical

and psychosocial outcomes (see Chapter 6). Additionally, falls efficacy, falls

management, and falls control were measured by GHGA deliverers using self-report

scales at baseline and 12 weeks post-baseline. Session observations also revealed

that participants themselves recognised the benefits that the sessions were having

on their physical health and how this was helping them retain their independence

outside of the sessions. Participants also noted that they had made friends for life

which they continued seeing outside of the GHGA PA sessions. A recent systematic

review of qualitative studies of PA in older age highlighted social influences in its

thematic synthesis of findings, identifying social interactions as important facilitators

for PA sustainability, and hence fidelity to treatment enactment, in this age group

(Franco et al., 2015).

Based upon NIH BCC’s treatment fidelity framework assessment criteria (Bellg et al.,

2004), this process evaluation has provided a comprehensive review of fidelity

relating to the GHGA PA intervention for community-dwelling older adults. Results

derived from this post-hoc fidelity analysis can be used to conceptualise best

practices as a process for planning future interventions that will be appropriate

within this setting and population (Green, 2001).


207

A strength of the current study was the comprehensive assessment of intervention

fidelity using multiple sources of data based upon NIH BCC’s treatment fidelity

framework assessment criteria (Bellg et al., 2004). The triangulation of data, utilising

multiple methods of qualitative data alongside quantitative data is a further strength

which enhanced understanding of intervention implementation and subsequently,

overall intervention fidelity (Farquhar, Ewing & Booth, 2011). Adherence to

intervention components and content delivery were both self-assessed by deliverers

and by an independent observer thus, increasing the validity of the data. Study

limitations are also noted. The subjective nature of the data is a limitation, as is the

presence of self-selection bias which resulted from the convenience sampling

methods adopted. Furthermore, as was noted in Chapter 3 (Study 1), men tend to

decrease participation in leisure-time PA as they get older; whereas this trend is not

seen among women (Amagasa et al., 2017). Consequently, gender bias is a possibility

given the large discrepancy between male (n =54) and female (n =334) attendees

noted throughout the session observations. With regard to the deliverer interviews,

not all GHGA deliverers participated in the study and thus, results may not be

representative of the larger group of GHGA deliverers. Interviews did however

reflect a broad range of experiences and opinions encompassing a spectrum of

implementation factors. One of the key benefits of assessing treatment fidelity is to

allow for the early detection of errors to prevent protocol deviations from becoming

widespread and long lasting before their implementation into real world settings

(Borrelli, 2011). Consequently, the post-hoc analysis design is a limitation (Lawton et

al., 2014). However, when a multi-component intervention is being tested within

‘real world’ settings there is a much greater blurring of the boundaries between

208

evaluations of efficacy and effectiveness (Glasgow et al., 2003). Consequently, it is

entirely appropriate to measure fidelity and to use this information to explain

variations in effectiveness as the intervention is being delivered in ‘real world’

settings (Craig et al., 2008). This allows for more informed decision making about the

commissioning and roll out of the intervention in any subsequent settings (Craig et

al., 2008). Furthermore, post-hoc fidelity analysis has been adopted previously in

multi-component PA interventions in older adults (McMahon et al., 2017; Vidovich et

al., 2015) and thus, was suitable for adoption in the current study.

7.6. Conclusions

Through the adoption of a comprehensive fidelity framework, GHGA PA intervention

fidelity was assessed under ‘real world’ settings. Subsequent analysis of the factors

influencing fidelity to delivery, provide valuable evidence to aid interpretation of

overall programme findings and effectiveness. Results from both deliverer interviews

and session observations revealed that a high degree of intervention fidelity was

maintained throughout the GHGA PA sessions, across all venues and deliverers.

Although the GHGA PA sessions were implemented as intended, findings reported in

Chapter 6 revealed that the programme was ineffective in increasing PA, and

decreasing time spent in SB. Consequently, the potential for long-term

implementation of the GHGA programme throughout Sefton Borough, and scaling up

209

of the intervention to inform the thought of policy and practice of professionals in

PA, social work and care settings is not warranted in its current capacity.

Results can be used to further strengthen the design, delivery and recruitment

strategies of future community-based PA interventions in older adults. However,

understanding of the optimal formatting and content of multi-component

interventions, of relevance for implementation in older adults remains

underdeveloped (Mc Sharry, Olander & French, 2014; Nigg & Long, 2012), and

further process evaluations of multi-component community-based PA interventions

in older adults are warranted to guarantee future success.

Thesis Study Map





Objectives





the community.




initiative.

Key Findings:




210







adults.




adults.

Objectives









Key Findings







respectively.




respectively.








psychosocial wellbeing among UK community-


Objectives




211

health indicators.



health outcomes.

Key Findings





outcomes.










evident.




adults.

Objectives

To evaluate the impact of Sport England’s Get

Healthy Get Active physical activity intervention

on older adults physical activity, sedentary

behaviour and self-reported health indicators.

Key Findings:

The Get Healthy Get Active physical activity

intervention was effective in increasing quality of

life, self-rated health, self-assessment of physical

fitness, and self-efficacy for exercise scores over

time after adjustment for covariates.

There was no significant intervention effect on

time spent in moderate-to-vigorous physical

activity.

The intervention also led to a significant increase

in sitting time throughout the follow-up time

points.

Study 5. Implementation fidelity of the Get Objectives:

212



adults

To evaluate whether or not the GHGA multi-

component intervention was implemented as

intended.

To evaluate sustainability of the GHGA multi-

component intervention in terms of its feasibility

and acceptability of long-term implementation

across multiple settings.

Key Findings:

A high degree of intervention fidelity was

maintained throughout the GHGA PA sessions

within the five core domains of: Study Design,

Provider Training, Intervention Delivery,

Intervention Receipt and Enactment.

Chapter 8. Synthesis of Findings,

Recommendations and Conclusions

213

8.1. Synthesis of findings

Overwhelming evidence shows that regular PA is among the most important

modifiable determinants for maintenance of physical (Zhu et al., 2017) and

psychosocial (Devereux-Fitzgerald et al., 2016; Franco et al., 2015; Greaney et al.,

2016) health at older ages (Lehne & Bolte, 2017). Government recommendations

state that older adults (≥65 years) should engage in at least 150 minutes of MPA (or

75 minutes of VPA) per week in bouts of at least 10 minutes, with muscle-

strengthening and balance activities included on at least two of those days

(Department of Health, 2011; CDC, 2015). Objectively collected data indicates that

only 15 per cent of males and ten percent of females within the UK, and 9.5% of

males and 7% of females within the US are meeting the recommended PA guidelines

(Tucker et al., 2011; Jefferis et al., 2014). To improve population health, efficacious

214

PA interventions in controlled research settings must be scaled up to reach broader

populations across multiple settings (Milat et al., 2016).

The overarching aim of the research programme was to assess the effectiveness and

implementation of Sport England’s GHGA intervention with community-dwelling

older adults. This aim was assessed through the following Research Questions:

1. What are the current knowledge and attitudes towards PA among older adults

living in the community, as well as perceived barriers, facilitators and opportunities

for PA participation?

2. What are the most appropriate wrist- and hip-worn raw acceleration cutpoints for

SB and MVPA activity in the GHGA sample of older adults?

3. Are there any gender, age, and socio-economic status differences in older adults’

SB, PA and self-reported health indicators?

4. What are the associations between SB and PA with self-reported health

indicators?

5. Is Sport England’s GHGA PA intervention effective in increasing community-

dwelling older adults PA levels?

6. Was the GHGA PA intervention implemented as intended?

215

The programme of work firstly aimed to explore older adults’ current knowledge and

attitudes towards PA, as well as perceived barriers, facilitators and opportunities for

PA participation among older adults living in the community. Secondly, a laboratory-

based protocol to generate behaviourally valid, population specific wrist-based GA

and hip-based AG raw acceleration cutpoints for SB and MVPA in older adults was

conducted. Thirdly, based upon the GA cutpoints generated, gender, age, and SES

differences in older adults’ SB, PA, and self-reported health outcomes were explored.

Associations between SB and PA with self-reported health outcomes were also

investigated. Fourthly, a mixed-methods evaluation was adopted in order to test the

feasibility, acceptability and effectiveness of Sport England’s GHGA PA intervention.

Finally, implementation fidelity of the GHGA PA intervention was assessed. The

studies conducted as part of this thesis were theoretically underpinned by two

conceptual models.

The programme of work was grounded in the socio-ecological model (McLeroy et al.,

1988) and the PRECEDE-PROCEED model of health programme design,

implementation, and evaluation (Green & Kreuter, 2005). Use of these frameworks

ensured that the studies considered a range of multidimensional interacting

influences on older adults’ PA, including intrapersonal, interpersonal, organisational,

community, environmental, and policy levels. This then enabled the community-

based PA strategies which were appropriate for use. Few other community-based PA

interventions targeting older adults’ PA have been underpinned by appropriate

theory (Chase, 2015). Interventions which are guided by theoretical frameworks,

216

consider implementation at scale across levels of the socioecological model, and are

designed, implemented and delivered in close partnership with stakeholders are

warranted among this population (Harris et al., 2015; Harris et al., 2018; Sink et al.,

2015).

This final chapter of the thesis will summarise the findings of each study and

synthesise them in relation to each other and the existing literature base. The overall

strengths and limitations of the thesis will then be discussed. Finally,

recommendations are outlined to inform future practice and research and overall

thesis conclusions are presented.

Several social (e.g., social awkwardness and peer/family support), behavioural (e.g.,

ageing stereotypes and lack of time), physical (e.g., improved balance and flexibility),

and environmental (e.g., transport and neighbourhood safety) correlates of PA

among older adults have been noted in previous formative (van Schijndel-Speet et

al., 2014; Banerjee et al., 2015) and qualitative research (Franco et al., 2015;

Devereux-Fitzgerald et al., 2016; Phoenix & Tulle, 2017). Such findings are a first step

in enabling policymakers and healthcare professionals to implement effective PA

interventions and promote active ageing (Franco et al., 2015).

Chapter 3 was the first to adopt a pen profiling protocol in order to analyse the

barriers and facilitators to PA among both assisted living and community dwelling

older adults. Results from this chapter revealed a variety of predisposing, enabling

217

and reinforcing correlates of PA participation. Consistent with previous research

(Gray et al., 2015; Kosteli et al., 2016), motivation was the most highly cited

predisposing correlate of PA participation and was perceived to be both a facilitator

and barrier to PA participation. The importance of pre-intervention intrinsic

motivation (e.g., participating for enjoyment) among older adults is key for both

initial adoption and maintenance of PA participation (Gray et al., 2015). A key

enabling correlate of PA behaviour was financial cost which was viewed as being a

barrier to PA participation (Franco et al., 2015; Borodulin et al., 2016). Community-

dwelling participants were either unable, or unwilling to pay the perceived high costs

associated with both attending and travelling to PA intervention sessions.

Consequently, future research is warranted to source ways that can sustain low-cost,

and easy reachable PA opportunities (Petrescu-Prahova et al., 2015; Borodulin et al.,

2016). An additional correlate of PA adherence in community-dwelling older adults

noted in previous literature is peer support (Brown et al., 2015). This reinforcing

correlate of PA participation was also identified to be a key facilitator (n=13) to PA

participation in Chapter 3. Hence, sustainable exit routes in order to retain the

provision of group activities which continue to facilitate, build and retain social

bonds post-intervention should be considered by PA programmers and policymakers

when designing PA interventions in this population (Wu et al., 2015).

In answering Research Question 1, Chapter 3 revealed that older adults

acknowledged the benefits of PA, not only for health but also those relating to

socialising, enjoyment, relaxation, and physical and psychological wellbeing. This

218

confirmed the need for efficacious PA interventions among this population. These

data were used to strengthen the design, delivery and recruitment strategies of the

GHGA PA intervention which was implemented and evaluated in Chapters 6 and 7.

Chapter 4 was the first to test a laboratory-based protocol to generate behaviourally

valid, population specific wrist-based GENEActiv (GA) and hip-based Actigraph GT3X+

(AG) raw acceleration cutpoints for SB and MVPA in older adults. ROC curve analyses

revealed that both wrist-based GA and hip-based AG accelerometer raw acceleration

cutpoints provide good and excellent discriminations of SB and MVPA, respectively.

This study was also the first to test the effect of both Youden and Se/Sp-based

cutpoints among older adults in order to make an informed decision as to which may

be most suitable once the data were analysed. In answering Research Question 2, GA

cutpoints of 57 mg and 104 mg were generated for SBSe and MVPASp, respectively.

For AG the cutpoints were 15 mg and 69 mg for SB Se and MVPASp thresholds,

respectively. These results are comparable to values reported previously for SB and

MVPA (Hildebrand et al., 2014; Hildebrand et al., 2016; Menai et al., 2017). Cross-

validation analysis revealed moderate agreement for GA and AG SB cutpoints, and

fair to substantial agreement for GA and AG MVPA cutpoints, respectively. Therefore

future interventions should seek to further demonstrate the utility of these cutpoints

in this population.

Based on the novel GA cut-points obtained in Chapter 4, Chapter 5 investigated

objectively measured time spent in MVPA and SB, as well as self-reported SB and PA

219

of community-dwelling older adults. Health comprises not only physical, but also

psychological and social components, and it is therefore important to take self-rated

measurements into consideration when evaluating health status (Kuosmanen et al.,

2016). Consequently, self-reported health indicators including QoL, SRH, SAPF, and

SEE, and their associations with SB and PA were also explored. Among older adults,

health indicators have been found to be influenced by sociodemographic attributes

including gender, age, and SES (Bamia et al., 2017; Meyer et al., 2014), as well as

being positively associated with PA level (Beyer et al., 2015; Haywood et al., 2018),

and negatively associated with SB (Haywood et al., 2018).

Comparable to recent self-report and accelerometer assessed SB and PA studies in

older adults (Amagasa et al., 2017; López-Rodríguez et al., 2017), results from

Chapter 5 revealed self-report and accelerometer assessed total time spent in SB

and MVPA in at least 10 minute bouts to be 411.9 vs 772 min‧d-1 and 62.7 vs 7.7 min‧

d-1, respectively. The lower levels of self-reported SB, and higher levels of self-

reported MVPA in at least 10 minute bouts reported when compared to objective GA

accelerometer-assessed SB and MVPA further confirms the inherent limitation of

recall bias within self-report measures (Barnett et al., 2016). The ubiquitous

presence of total accumulated and sporadic PA in older adults makes it difficult to

recall in questionnaire surveys (Washburn, 2000), though such behaviours may be of

particular importance, especially for older adults who tend to perform shorter

duration exercises (Amagasa et al., 2017; Jefferis et al., 2016; Sparling et al., 2015).

Consequently, this population tend to misreport time spent in such activities when

220

compared with objective measures such as accelerometry (Ku et al., 2016). In

answering Research Question 3, results revealed no significant gender, age or SES

differences between self-reported and accelerometer-derived SB and PA. These

findings are counter to those in previous studies which have noted that MVPA is

lower among those who are female and older (Amagasa et al., 2017; Lohne-Seiler et

al., 2014; Ramires et al., 2017; Shiroma et al., 2018) due to difficulties in mobility,

general health status, and lower levels of self-efficacy (Ramires et al., 2017).

Participants were recruited whilst attending the GHGA PA sessions and therefore,

both men and women across the age range were likely more inclined to be active.

Moreover, gender bias in the sample could have further affected any potential

gender and age category differences between SB and PA.

Results from answering Research Question 4 revealed significant gender, age

category and SES-group differences between QoL, SRH, SAPF, and SEE. Concurrently,

previous research has noted that older adults who are male, younger-old (60-69

years), and of higher SES are more likely to report favourable ratings of self-reported

physical and psychosocial health (Bamia et al., 2017; Kuosmanen et al., 2016).

Further results from Chapter 5 also provided evidence of a negative association of

self-reported SB, and positive association of self-reported MVPA in at least 10

minute bouts with health indicators. A number of self-reported physical conditions

including number of falls, balance, pain interference, and lower-extremity function

are associated with time spent in SB and MVPA (Haywood et al., 2018; Rezende et

al., 2014). Previous studies have also noted negative associations of SB, and positive

221

associations of MVPA on QoL, wellbeing, depression, and self-efficacy (Ku et al.,

2016; Withall et al., 2014).

The findings of Chapter 5 provided baseline information of participants due to

participate in the GHGA PA intervention. Chapter 6 evaluated the overall

effectiveness of Sport England’s GHGA PA intervention on older adults’ PA, SB and

self-reported health indicators. A major strength of this study was the 2-level data

structured multilevel modelling statistical analysis which took into account the three

follow-up time points and multiple covariates associated with the primary and

secondary outcome measures. Results indicated a significant decrease in self-

reported MVPA at six months follow-up. The odds of meeting MVPA guidelines

decreased significantly across all three follow-up time points. These results contrast

with a recent meta-analysis of 53 exercise intervention studies in community-

dwelling older adults which reported a positive pooled effect equivalent to a 73

minute per week increase in MVPA when comparing intervention with control

groups (Chase, 2015). Results also revealed significant increases in self-reported

sitting time across all three follow-up time points. However, given the decreasing

levels of MVPA noted, increased sitting time was to be expected as lower amounts of

time spent in MVPA has been associated with greater total sedentary time (Diaz et

al., 2016). In line with recent research (Zubula et al., 2017), significant increases in

QoL, SAPF, SRH and SEE health indicator scores were observed throughout the three,

six and 12-month follow-up time points. Previous research has demonstrated

positive associations between social support and QoL (Siedlecki et al., 2014), SRH

222

(Dai et al., 2016), and SEE (Warner et al., 2011). Resultantly, the thirty minutes set

aside at the end of each weekly GHGA PA session to allow participants to socialise

could explain the increases in health indicator scores despite the lower levels of

MVPA, and increased levels of sitting observed. In answering Research Question 5,

these results revealed that Sport England’s GHGA PA intervention was ineffective in

increasing community-dwelling older adults’ PA levels.

Gaining an accurate and objective record of session content is important in

determining intervention suitability and overall fidelity (Moore et al., 2015).

Consequently, Research Question 6 was answered in Chapter 7 through the

assessment of implementation fidelity of the GHGA PA intervention. The prominent

strength of Chapter 7 was the use of qualitative and quantitative data sources. This

approach is advocated when conducting implementation fidelity research as it allows

for a comprehensive understanding of the processes which influence

implementation, and their variation across contexts to be gained (Moore et al.,

2015). Through the adoption of a comprehensive treatment fidelity framework

developed by the NIH BCC for tailored health behaviour interventions (Bellg et al.,

2004), results from both deliverer interviews and session observations revealed that

a high degree of session content fidelity was maintained throughout the delivery of

the GHGA PA intervention within the five BCC core domains of: Study Design,

Provider Training, Intervention Delivery, Intervention Receipt and Enactment.

Despite such results, the GHGA PA intervention was ineffective at increasing MVPA

levels. Consequently, further exploration of the most suitable advertising, session

223

content and delivery strategies to sustain increases in PA among older adults are

warranted and hence, should remain a nationwide public health priority.

The results associated with this thesis underline the importance of conducting

process evaluations prior to intervention roll out in ‘real world’ settings. The GHGA

PA sessions were predominantly chair-based sessions aiming to engage inactive

older adults in PA at least once a week for 30 minutes. A major limitation of the

current thesis was the IPAQ-E (Hurtig-Wennlöf et al., 2010) which was adopted as

the measure of PA as required by the funder. The IPAQ-E lacks a measure of LPA, a

PA activity which is most associated with older adults (McMahon et al., 2017) and

one which was most associated with the GHGA PA sessions. Consequently, any

changes in LPA as a consequence of the programme were not captured by the self-

reported outcome measure of PA. A more specific aim (e.g., a certain PA intensity

such as MVPA, or an aim anchored to current PA guidelines) would have provided

more specificity over the required outcome measures required. Specifically, national

governing bodies, stakeholders and key partners should work together prior to

intervention implementation in ‘real world’ settings’. Clear intervention aims and

subsequent intervention measures can then be decided upon in order to avoid null

intervention effects.


224

Given that study-specific strengths and limitations have already been discussed in

each chapter, this section will discuss in more detail the main strengths and

limitations that were consistent across the whole programme of research.

8.2.1. Physical activity measurement

Both self-report and objective measures of PA were adopted throughout the

programme. PA levels have traditionally been measured via subjective self-report

questionnaires in older adults (Kowalski et al., 2012) as they are relatively cheap to

conduct and have the potential to reach a large number of participants (Aguilar-

Farías et al., 2015; Celis-Morales et al., 2012; Chastin et al., 2014; Healy et al., 2011).

Specifically, the IPAQ-E (Hurtig-Wennlöf et al., 2010) was adopted as required by the

funder. The IPAQ-E is tailored to and validated for older adults and hence was

appropriate for use throughout programme. However, the ubiquitous presence of

total accumulated and sporadic PA in older adults makes it difficult to recall in

questionnaire surveys (Washburn, 2000), though such behaviours may be of

particular importance, especially for older adults who tend to perform shorter

duration exercises (Amagasa et al., 2017; Jefferis et al., 2016; Sparling et al., 2015).

Consequently, recall bias is a probability (Barnett et al., 2016) given evidence that

such methods of data collection can lead to underestimations of SB (Aguilar-Farías et

al., 2015; Chastin & Granat, 2010; Harvey et al., 2014) and overestimations of time

spent engaged in PA of all intensities (Tucker et al., 2011). A further limitation of the

IPAQ-E is that there is no measure for LPA, a PA activity which is most associated

with older adults (McMahon et al., 2017). Objective assessment such as

225

accelerometers can record more detailed and accurate patterns of personal daily

activity (Jefferis et al., 2016; Shephard & Tudor-Locke, 2016). Accelerometers can

precisely obtain this type of information. Therefore, the adoption of both

questionnaire- and accelerometer-assessed SB and PA in Chapter 5 further

strengthened the overall study outcome. Most commonly used outputs from

accelerometers are counts which are dependent on internal proprietary algorithms

(Welk et al., 2012). Uncertainties of pre-processed count data include the possibility

that signal filtering methods can alter study results (Freedson, Bowles, Troiano, &

Haskell, 2012; Peach, Van Hoomissen, & Callender, 2014). Consequently, the

adoption of raw accelerations in Chapter 4 provided greater methodological

transparency in post-data collection analytical processes (Hildebrand et al., 2014).

However, the use of raw data is still in its infancy and the increased control which

researchers can have over this form of data means that there is a lack of consensus

over the procedures. A major limitation of Chapter 6 was the lack of objective

measures of sitting time and/or PA level, especially given the recall bias associated

with older adults’ assessment of PA and SB when compared to objective

accelerometry which resulted in Chapter 5. Due to time constraints and a limited

number of available GA accelerometer devices during the period of contact with the

older adult population, this was not possible in the current study. The GHGA PA

sessions were predominantly chair-based sessions aiming to engage inactive older

adults in PA at least once a week for 30 minutes. A more specific aim (e.g., a certain

PA intensity such as MVPA, or an aim anchored to current PA guidelines) would have

provided more specificity over the required outcome measures required.

Considering the large variation in participant abilities, sessions were relatively low

226

intensity and consequently, a major limitation of Chapter 6 was that any changes in

LPA as a consequence of the programme, were not captured by the self-reported

outcome measure of PA which was adopted as a requirement of the funder in

Chapters 5 and 6.

The repeated measures design at three follow-up time points post-intervention is a

major strength, in light of recommendations for longer-term post-intervention

follow-ups among PA interventions (McMahon et al., 2017). The multi-level analysis

which took into account the clustered nature of the design, with measures clustered

in the differing follow-up time points, was also a strength. The major limitations of

the study included the pre-post design and absence of a control group. A clustered

RCT research design was initially advocated but was rejected by SMBC due to both

low GHGA deliverer and participant numbers.

8.2.2. Methodological approach

The mixed methods approach adopted across the five chapters further strengthens

the overall results of the thesis. The GHGA PA intervention employed a

methodological approach that enabled researchers to capture PA data from

participants and contextual data from all key stakeholders (e.g., GHGA session

deliverers) in order to comprehensively assess both effectiveness and

implementation fidelity. The triangulation of data sources in Chapter 1 and interrater

reliability (McHugh, 2012) procedures in Chapter 7 allowed for the comparison and

confirmation of data, which provided the study with a high degree of credibility and

227

methodological rigour. Quantitative methodologies are most useful for testing

effectiveness but cannot identify mistakes, limitations or unintended consequences

of PA strategies which can influence effectiveness outcomes (Beltran-Carrillo, Ferriz,

Brown, & Gonzalez-Cutre, 2017). Learning from the target population (older adults)

and incorporating evaluation techniques based on the qualitative approaches

undertaken in Chapters 1 and 7 can attend to this gap in evaluative knowledge. The

insight provided from qualitative data is an important complimentary contribution to

the research base particularly when combined with quantitative methodologies

(Beltran-Carrillo et al., 2017). This approach allows for a review of not just overall

intervention effectiveness, but also implementation fidelity. Therefore, programme

suitability, feasibility and long-term sustainability in ‘real world’ settings can be

reviewed. The results of the current programme of research advocates the further

use of mixed methodologies in community-based PA intervention research to better

understand strategies which are both effective and also feasible.

8.2.3. Sampling

The formative research strategies adopted to recruit participants throughout the

chapters is a limitation and introduces a level of sampling and selection biases into

the results. A major limitation of Chapter 6 is the pre-post design. This design results

in lowered levels of causal validity due to the uncontrollable effects of regression to

the mean, maturation, history, and test effects (Marsden & Torgerson, 2012). The

absence of a control group is also a limitation. Coverage and sampling errors are also

limitations due to the modest sample size and non-randomisation of participants.

228

Consequently, selection bias is an issue (Marsden & Torgerson, 2012). However,

participant non-randomisation was justified given the pragmatic approach adopted

(Thomas et al., 2006) and the low number of participants recruited from intact

groupings (Sefton’s Older People Forum and care homes) throughout Sefton

Borough. However, the project was limited by the resources available, for example

access to accelerometers, research staff to collect data and also time available.

Subsequently, the modest sample sizes of Chapter 5 and 6 are recognised as

limitations of this research. Given that statistical power increases as the number of

participants increases and overall sample size is extremely influential on power,

some of the analyses may have lacked sufficient power to detect significant changes

in outcomes (Thomas, Silverman, & Nelson, 2015). Gender bias is also a concern

because of the disproportionate male to female participation ratio throughout the

five chapters. This is of particular concern in Chapter 5 and could have affected any

potential gender and age category differences between SB and PA. There was a lack

of characteristic data collected throughout the programme due to the formative

research strategies adopted to recruit the participants. Such have been shown to

affect the perceived barriers and facilitators to PA participation among older adults

(Greaney et al., 2016; Keadle et al., 2016) and thus, future research should obtain

data including, participants’ current sedentary time and PA levels, history of PA,

family history of PA, ethnicity, employment status, and educational achievements.

8.3. Recommendations

229

As a result of the findings presented from this programme of research, various

recommendations for future work are proposed. These are separated into

recommendations for research and practice.

8.2.1: Recommendations for practice

Community-based PA interventions should explore the effects of delivering

blocks of sessions with consistent content and delivery techniques in order to

allow for a better understanding of the approaches that are both effective

and ineffective in eliciting both short- and long-term PA behaviour change

among this population.

Community-based PA interventions should take into account location and

timing of sessions. Locations should be easy to access by both public

transport and by car, and sessions should avoid taking place during either

morning or afternoon rush hour periods.

Health-promotion strategies should advocate for the provision of low-cost,

and easy reachable PA opportunities.

If adopting self-report questionnaires, studies should include assessments of

LPA as well as MPA and VPA. This is especially applicable to older adults

whose PA participation is sporadic and often of a lower intensity.

The adoption of local/national mass media messages may be a cost effective

educational solution at a time when there is a growing ageing population.

Post-hoc fidelity analysis can be adopted to conceptualize best practices as a

process for planning future interventions that will be appropriate within

specific settings and populations.

230

8.2.2: Recommendations for future research

Future researchers should work in partnership with funders and key

stakeholders prior to intervention delivery to establish outcome measures

which are fit for purpose for each specific project.

Future research should seek to identify barriers and facilitators among

diverse samples (e.g., community/assisted living and young-old, middle-old

and old-old participants) that are more representative of the older adult

population.

Future interventions targeting PA in older adults should be designed and

implemented based upon the requirements of the SEF for PA interventions

(National Obesity Observatory, 2012) to enhance implementation consistency

across settings.

PA research among this population should use accelerometers to provide

more accurate estimates of PA levels and SB. Wrist-worn, 24-hour protocols

are also recommended to optimise compliance.

The raw acceleration wrist-worn GA and hip-worn AG SB and MVPA cutpoints

obtained in Chapter 4 should be further cross-validated with independent

samples, ideally from other settings and within free-living environments.

It is recommended that local funders and commissioners of research obtain

participant characteristic data even when it is not central to the intervention

outcomes as these data are confounding covariates of older adults SB and PA

levels.

231

Examining correlates of attrition and initial non-participation in older adults is

a vital step in the research process that needs further exploring.

Further research with more socio-demographically diverse older populations

is warranted to improve understanding of the relationship between gender,

age, and SES and, self-reported physical and psychosocial outcomes, as well

as the independent factors affecting them such as SB and PA levels.

Future research should make efforts to collect qualitative data from older

adults across differing locations and SES-strata in order to understand

perceived feasibility and acceptability of PA strategies. The triangulation of

data and utilising qualitative data alongside quantitative data can enhance

understanding.

Process evaluations of intervention implementation and fidelity based upon

comprehensive frameworks such as the NIH BCC (Bellg et al., 2004) should

become an integral part of the conduct and evaluation of all health behaviour

intervention research. Such will improve understanding of how interventions

have been implemented in practice, so that they can be further integrated

into ‘real world’ community settings and contribute to overall public health.

8.4. Conclusions

The overall aim of this thesis was to assess the effectiveness and implementation of

Sport England’s GHGA intervention on inactive community-dwelling older adults’ PA

levels. Although a high degree of intervention fidelity was maintained throughout

232

the GHGA PA sessions, across all venues and deliverers, the intervention was

ineffective in reducing time spent in SB and increasing time spent in MVPA across

three, six and 12-month follow-up time points. However, the GHGA intervention was

effective in increasing QoL, SRH, SAPF, and SEE scores at 12-months post-baseline

measurement after adjustment for covariates. These results indicate that the

potential for long-term implementation and scaling up of the GHGA programme

throughout Sefton Borough is not warranted in its current format. Major facilitators

and barriers to PA participation were first identified in order to inform the design,

delivery and recruitment strategies of the intervention. Facilitators for PA included

motivation, enjoyment, health benefits and social support. Barriers for PA

participation included age, isolation, opportunities and awareness for physical

activity participation, cost, transport, location, season/weather, and time of day.

References

233

Active People Survey. (2014). Active People Survey 8: Active People Interactive. Retrieved from http://activepeople.sportengland.org/.

Age UK. (2018). Later life in the United Kingdom. Retrieved from https://www.ageuk.org.uk/globalassets/age-uk/documents/reports-and-publications/later_life_uk_factsheet.pdf.

Aggio, D., Fairclough, S., Knowles, Z. and Graves, L. (2016). Validity and Reliability of a Modified English Version of the Physical Activity Questionnaire for Adolescents. Archives of Public Health, 74(1), 3-11.

234

http://activepeople.sportengland.org/

Aguilar-Farías, N., Brown, W. J., Olds, T. S., & Peeters, G. M. E. E. (2015). Validity of self-report methods for measuring sedentary behaviour in older adults. Journal of Science and Medicine in Sport, 18(6), 662–666.

Altschuler, A., Picchi, T., Nelson, M., Rogers, J. D., Hart, J., & Sternfeld, B. (2009). Physical Activity Questionnaire comprehension-lessons from cognitive interviews. Medicine and science in sports and exercise, 41(2), 336-343.

Amagasa, S., Fukushima, N., Kikuchi, H., Takamiya, T., Oka, K., & Inoue, S. (2017). Light and sporadic physical activity overlooked by current guidelines makes older women more active than older men. International Journal of Behavioral Nutrition and Physical Activity, 14(1), 59-66.

Amagasa, S., Machida, M., Fukushima, N., Kikuchi, H., Takamiya, T., Odagiri, Y., & Inoue, S. (2018). Is objectively measured light-intensity physical activity associated with health outcomes after adjustment for moderate-to-vigorous physical activity in adults? A systematic review. International Journal of Behavioral Nutrition and Physical Activity, 15(1), 65-77.

Antikainen, I., & Ellis, R. (2011). A RE-AIM evaluation of theory-based physical activity interventions. Journal of sport and exercise psychology, 33(2), 198-214.

Armstrong, T., & Bull, F. (2006). Development of the world health organization global physical activity questionnaire (GPAQ). Journal of Public Health, 14(2), 66-70.

Arnardottir, N.Y., Koster, A., Van Domelen, D.R., Brychta, R.J., Caserotti, P., Eiriksdottir, G., Sverrisdottir, J.E., Launer, L.J., Gudnason, V., Johannsson, E., & Harris, T.B. (2012). Objective measurements of daily physical activity patterns and sedentary behaviour in older adults: Age, Gene/Environment Susceptibility-Reykjavik Study. Age and ageing, 42(2), 222-229.

Asendorpf, J. B., Van De Schoot, R., Denissen, J. J., & Hutteman, R. (2014). Reducing bias due to systematic attrition in longitudinal studies: The benefits of multiple imputation. International Journal of Behavioral Development, 38(5), 453-460.

ATS Committee on Proficiency Standards for Clinical Pulmonary Function Laboratories. (2002). Statement AT: guidelines for the six-minute walking-test. American Journal of Respiratory and Critical Care Medicine, 166 , 111-117.

Atun, R., de Jongh, T., Secci, F., Ohiri, K., & Adeyi, O. (2010). Integration of targeted health interventions into health systems: a conceptual framework for analysis. Health policy and planning, 25(2), 104-111.

Australian Government Department of Health and Ageing 2013. Recommendations on Physical Activity for Health for Older Australians. Retrieved from http://www.health.gov.au/internet/main/publishing.nsf/content/130D93778A64136DCA257BF0001DACF2/$File/pa-guidelines.pdf.

235

Autenrieth, C. S., Baumert, J., Baumeister, S. E., Fischer, B., Peters, A., Döring, A., & Thorand, B. (2011). Association between domains of physical activity and all-cause, cardiovascular and cancer mortality. European journal of epidemiology, 26(2), 91-99.

Baert, V., Gorus, E., Calleeuw, K., De Backer, W. and Bautmans, I. (2016). An Administrator's Perspective on the Organization of Physical Activity for Older Adults in Long-Term Care Facilities. Journal of the American Medical Directors Association, 17(1), 75-84.

Baert, V., Gorus, E., Mets, T. and Bautmans, I. (2015). Motivators and Barriers for Physical Activity in Older Adults with Osteoporosis. Journal of Geriatric Physical Therapy, 38(3), 105-114.

Balboa-Castillo, T., León-Muñoz, L. M., Graciani, A., Rodríguez-Artalejo, F., & Guallar-Castillón, P. (2011). Longitudinal association of physical activity and sedentary behavior during leisure time with health-related quality of life in community-dwelling older adults. Health and quality of life outcomes, 9(1), 47-56.

Bamia, C., Orfanos, P., Juerges, H., Schöttker, B., Brenner, H., Lorbeer, R., ... & Lagiou, P. (2017). Self-rated health and all-cause and cause-specific mortality of older adults: Individual data meta-analysis of prospective cohort studies in the CHANCES Consortium. Maturitas, 103, 37-44.

Bampton, E. A., Johnson, S. T., & Vallance, J. K. (2015). Profiles of resistance training behavior and sedentary time among older adults: Associations with health-related quality of life and psychosocial health. Preventive Medicine Reports, 2, 773–776.

Bandura, A. (1986). Social foundation of thought and action: A social-cognitive view. Englewood Cliffs.

Banerjee, A. T., Kin, R., Strachan, P. H., Boyle, M. H., Anand, S. S. and Oremus, M. (2015). Factors Facilitating the Implementation of Church-Based Heart Health Promotion Programs for Older Adults: A Qualitative Study Guided by the Precede-Proceed Model. American Journal of Health Promotion, 29(6), 365-373.

Bankoski, A., Harris, T. B., McClain, J. J., Brychta, R. J., Caserotti, P., Chen, K. Y., ... & Koster, A. (2011). Sedentary activity associated with metabolic syndrome independent of physical activity. Diabetes care, 34(2), 497-503.

Bann, D., Hire, D., Manini, T., Cooper, R., Botoseneanu, A., McDermott, M. M., ... & Church, T. (2015). Light intensity physical activity and sedentary behavior in relation to body mass index and grip strength in older adults: cross-sectional findings from the lifestyle interventions and independence for elders (LIFE) study. PloS one, 10(2), e0116058.

236

Barber, S.E., Forster, A. & Birch, K.M. (2015). Levels and patterns of daily physical activity and sedentary behavior measured objectively in older care home residents in the United Kingdom. Journal of aging and physical activity, 23(1), 133-143.

Barnett, A., van den Hoek, D., Barnett, D. & Cerin, E. (2016). Measuring moderate-intensity walking in older adults using the ActiGraph accelerometer. BMC geriatrics, 16(1), 211-219.

Barnett, I., van Sluijs, E., Ogilvie, D. & Wareham, N.J. (2014). Changes in household, transport and recreational physical activity and television viewing time across the transition to retirement: longitudinal evidence from the EPIC-Norfolk cohort. J Epidemiol Community Health, 68(8), 747–753.

Barone Gibbs, B., Brach, J.S., Byard, T., Creasy, S., Davis, K.K., McCoy, S., Peluso, A., Rogers, R.J., Rupp, K. and Jakicic, J.M. (2017). Reducing sedentary behavior versus increasing moderate-to-vigorous intensity physical activity in older adults: a 12-week randomized, clinical trial. Journal of Aging and Health, 29(2), 247-267.

Bauman, A. E., Reis, R. S., Sallis, J. F., Wells, J. C., Loos, R. J., Martin, B. W., & Lancet Physical Activity Series Working Group. (2012). Correlates of physical activity: why are some people physically active and others not? The lancet, 380(9838), 258-271.

Bauman, A. E., Sallis, J. F., Dzewaltowski, D. A., & Owen, N. (2002). Toward a better understanding of the influences on physical activity: the role of determinants, correlates, causal variables, mediators, moderators, and confounders. American journal of preventive medicine, 23(2), 5-14.

Bauman, A., Merom, D., Bull, F.C., Buchner, D.M. & Singh, M.A.F. (2016). Updating the evidence for physical activity: summative reviews of the epidemiological evidence, prevalence, and interventions to promote “Active Aging”. The Gerontologist, 56(2), S268-S80.

Bellettiere, J., Carlson, J.A., Rosenberg, D., Singhania, A., Natarajan, L., Berardi, V., LaCroix, A.Z., Sears, D.D., Moran, K., Crist, K. & Kerr, J. (2015). Gender and age differences in hourly and daily patterns of sedentary time in older adults living in retirement communities. PloS one, 10(8), p.e0136161.

Bellg, A. J., Borrelli, B., Resnick, B., Hecht, J., Minicucci, D. S., Ory, M., ... & Czajkowski, S. (2004). Enhancing treatment fidelity in health behavior change studies: best practices and recommendations from the NIH Behavior Change Consortium. Health Psychology, 23(5), 443-451.

Beltran-Carrillo, V. J., Ferriz, R., Brown, D. H. K., & Gonzalez-Cutre, D. (2017). Qualitative evaluation of a school intervention for the promotion of physical activity: Learning from the perspective of the target population. European Journal of Human Movement, 38, 69-92.

237

Berger-Schmitt, R. (2000). Social cohesion as an aspect of the quality of societies: Concept and measurement. ZUMA.

Berkemeyer, K., Wijndaele, K., White, T., Cooper, A.J.M., Luben, R., Westgate, K., Griffin, S.J., Khaw, K.T., Wareham, N.J. & Brage, S. (2016). The descriptive epidemiology of accelerometer-measured physical activity in older adults. International Journal of Behavioral Nutrition and Physical Activity, 13(1), 2-11.

Beyer, A. K., Wolff, J. K., Warner, L. M., Schüz, B., & Wurm, S. (2015). The role of physical activity in the relationship between self-perceptions of ageing and self-rated health in older adults. Psychology & Health, 30(6), 671-685.

Biddle, S. (1995). Exercise and psychosocial health. Research quarterly for exercise and sport, 66(4), 292-297.

Bird, M.L., Clark, B., Millar, J., Whetton, S. and Smith, S. 2015. Exposure to “Exergames” Increases Older Adults’ Perception of the Usefulness of Technology for Improving Health and Physical Activity: A Pilot Study. JMIR Serious Games, 3(2), 1-8.

Birkel, R., Dessem, E., Eldridge, S., Kulinski, K., Lachenmayr, S., Spafford, M., Suwal, B., Terrillion, A., Walsh, M. & Zenker, W. (2015). Improving lives through evidence-based health promotion programmes: a national priority. Evidence-Based Programmeming for Older Adults, 2, 255-256.

Biswas, A., Oh, P.I., Faulkner, G.E., Bajaj, R.R., Silver, M.A., Mitchell, M.S. & Alter, D.A. (2015). Sedentary time and its association with risk for disease incidence, mortality, and hospitalization in adults a systematic review and meta-analysis sedentary time and disease incidence, mortality, and hospitalization. Annals of internal medicine, 162(2), 123-132.

Blodgett, J., Theou, O., Kirkland, S., Andreou, P. & Rockwood, K. (2015). The association between sedentary behaviour, moderate–vigorous physical activity and frailty in NHANES cohorts. Maturitas, 80(2), 187-191.

Blondell, S. J., Hammersley-Mather, R., & Veerman, J. L. (2014). Does PA prevent cognitive decline and dementia? A systematic review and meta-analysis of longitudinal studies. BMC public health, 14(1), 510-537.

Boddy, L. M., Knowles, Z. R., Davies, I. G., Warburton, G. L., Mackintosh, K. A., Houghton, L. & Fairclough, S. J. (2012). Using Formative Research to Develop the Healthy Eating Component of the CHANGE! School-Based Curriculum Intervention. BMC Public Health, 12(1), 710-720.

Booth, F. W., & Hawley, J. A. (2015). The Erosion of Physical Activity in Western Societies: an Economic Death March. Diabetologia, 58(8), 1730-1734.

Borodulin, K., Sipilä, N., Rahkonen, O., Leino-Arjas, P., Kestilä, L., Jousilahti, P., & Prättälä, R. (2016). Socio-Demographic and Behavioral Variation in Barriers to

238

Leisure-Time Physical Activity. Scandinavian Journal of Public Health, 44(1), 62-69.

Borrelli, B. (2011). The assessment, monitoring, and enhancement of treatment fidelity in public health clinical trials. Journal of Public Health Dentistry, 71(Suppl 1), S52-63.

Bouma, A. J., van Wilgen, P., & Dijkstra, A. (2015). The Barrier-Belief Approach in the Counseling of Physical Activity. Patient Education and Counseling, 98(2), 129-136.

Bowling, A. (2005). Ageing well: Quality of life in old age. McGraw-Hill Education (UK).

Braun, V. & Clarke, V. (2006). Using Thematic Analysis in Psychology. Qualitative Research in Psychology, 3(2), 77-101.

Breckon, J. D., Johnston, L. H., & Hutchison, A. (2008). Physical activity counseling content and competency: a systematic review. Journal of Physical Activity and Health, 5(3), 398-417.

Brown, D., Spanjers, K., Atherton, N., Lowe, J., Stonehewer, L., Bridle, C., Sheehan, B. & Lamb, S.E. (2015). Development of an Exercise Intervention to Improve Cognition in People with Mild to Moderate Dementia: Dementia and Physical Activity (DAPA) Trial, Registration ISRCTN32612072. Physiotherapy, 101(2), 126-134.

Bryan, A., Hutchison, K. E., Seals, D. R., & Allen, D. L. (2007). A transdisciplinary model integrating genetic, physiological, and psychological correlates of voluntary exercise. Health Psychology, 26(1), 30-39.

Buchman, A.S., Boyle, P.A., Yu, L., Shah, R.C., Wilson, R.S., & Bennett, D.A. (2012). Total daily physical activity and the risk of AD and cognitive decline in older adults. Neurology, 78(17), 1323-1329.

Buman, M.P., Hekler, E.B., Haskell, W.L., Pruitt, L., Conway, T.L., Cain, K.L., Sallis, J.F., Saelens, B.E., Frank, L.D., & King, A.C. (2010). Objective light-intensity physical activity associations with rated health in older adults. American Journal of Epidemiology, 172(10), 1155-1165.

Butte, N.F., Ekelund, U. & Westerterp, K.R. (2012). Assessing physical activity using wearable monitors: measures of physical activity. Medicine & Science in Sports & Exercise, 44(1 Suppl 1), S5-S12.

Byrne, N.M., Hills, A.P., Hunter, G.R., Weinsier, R.L. & Schutz, Y. (2005). Metabolic equivalent: one size does not fit all. Journal of Applied physiology, 99(3), 1112-1119.

239

Cadmus-Bertram, L. A., Marcus, B. H., Patterson, R. E., Parker, B. A., & Morey, B. L. (2015). Randomized trial of a Fitbit-based physical activity intervention for women. American journal of preventive medicine, 49(3), 414-418.

Calsyn, R. (2000). A checklist for critiquing treatment fidelity. Mental Health Services Research, 2(2), 107-113.

Camhi, S.M., Sisson, S.B., Johnson, W.D., Katzmarzyk, P.T., & Tudor-Locke, C. (2011). Accelerometer-determined moderate intensity lifestyle activity and cardiometabolic health. Preventive medicine, 52(5), 358-360.

Canadian Society for Exercise Physiology 2016. Canadian Physical Activity Guidelines for Older Adults – 65 Years and Older. Retrieved from https://www.participaction.com/sites/default/files/downloads/Participaction-Canadian-physical-activity-guidelines-older-adult.pdf.

Cardinal, B.J., & Cardinal, M.K. (2000). Preparticipation physical activity screening within a racially diverse, older adult sample: comparison of the original and Revised Physical Activity Readiness Questionnaires. Research quarterly for exercise and sport, 71(3), 302-307.

Cardinal, B.J., Esters, J., & Cardinal, M.K. (1996). Evaluation of the revised physical activity readiness questionnaire in older adults. Medicine and science in sports and exercise, 28(4), 468-472.

Carlson, J.A., Sallis, J.F., Conway, T.L., Saelens, B.E., Frank, L.D., Kerr, J., Cain, K.L., & King, A.C. (2012). Interactions between psychosocial and built environment factors in explaining older adults' physical activity. Preventive medicine, 54(1), 68-73.

Carroll, C., Patterson, M., Wood, S., Booth, A., Rick, J., & Balain, S. (2007). A conceptual framework for implementation fidelity. Implementation science, 2(1), 40-48.

Carstensen, L. L., Isaacowitz, D. M., & Charles, S. T. (1999). Taking Time Seriously: A Theory of Socioemotional Selectivity. American Psychologist, 54(3), 165-181.

Caspersen, C. J., Powell, K. E., & Christenson, G. M. (1985). Physical activity, exercise, and physical fitness: definitions and distinctions for health-related research. Public health reports, 100(2), 126-131.

Castillo, J. M., Wang, J. H., Daye, J. G., Shum, K. Z., & March, A. L. (2017). A Longitudinal Analysis of the Relations Among Professional Development, Educators’ Beliefs and Perceived Skills, and Response-to-Intervention Implementation. Journal of Educational and Psychological Consultation, 1-32. Doi: 10.1080/10474412.2017.1394864

Celis-Morales, C. A., Perez-Bravo, F., Ibañez, L., Salas, C., Bailey, M. E. S., & Gill, J. M. R. (2012). Objective vs. self-reported physical activity and sedentary time:

240

effects of measurement method on relationships with risk biomarkers. PloS One, 7(5), e36345.

Centers for Disease Control and Prevention. (2008). Physical Activity Guidelines for Americans: Fact Sheet for Health Professionals on Physical Activity Guidelines for Older Adults. Retrieved from http://www.cdc.gov/physicalactivity/downloads/pa_fact_sheet_olderadults.pdf.

Centers for Disease Control and Prevention. (2015). How Much Physical Activity do Older Adults Need?. Retrieved from https://www.cdc.gov/physicalactivity/basics/older_adults/.

Centers for Disease Control and Prevention. (2016). National Center for Health Statistics. Health Indicators Warehouse. Retrieved from http://www.healthindicators.gov.

Chad, K.E., Reeder, B.A., Harrison, E.L., Ashworth, N.L., Sheppard, S.M., Schultz, S.L., Bruner, B.G., Fisher, K.L., & Lawson, J.A. (2005). Profile of physical activity levels in community-dwelling older adults. Medicine & Science in Sports & Exercise, 37(10), 1774-1784.

Chase, J. D. (2015). Interventions to increase physical activity among older adults: A meta-analysis. The Gerontologist, 55(4), 706-718.

Chastin, S. F. M., & Granat, M. H. (2010). Methods for objective measure, quantification and analysis of sedentary behaviour and inactivity. Gait and Posture, 31(1), 82–86.

Chastin, S. F., De Craemer, M., De Cocker, K., Powell, L., Van Cauwenberg, J., Dall, P., ... & Stamatakis, E. (2018). How does light-intensity physical activity associate with adult cardiometabolic health and mortality? Systematic review with meta-analysis of experimental and observational studies. British Journal of Sports Medicine, 1-8.

Chastin, S. F., Fitzpatrick, N., Andrews, M., & DiCroce, N. (2014). Determinants of sedentary behavior, motivation, barriers and strategies to reduce sitting time in older women: a qualitative investigation. International journal of environmental research and public health, 11(1), 773-791.

Chastin, S., Gardiner, P.A., Ashe, M.C., Harvey, J.A., Leask, C.F., Balogun, S., Helbostad, J.L., & Skelton, D.A. (2017). Interventions for reducing sedentary behaviour in community dwelling older adults. ‐ The Cochrane Library, 1-13.

Chastin, S.F., Buck, C., Freiberger, E., Murphy, M., Brug, J., Cardon, G., O’Donoghue, G., Pigeot, I., & Oppert, J.M. (2015). Systematic literature review of determinants of sedentary behaviour in older adults: a DEDIPAC study. International Journal of Behavioral Nutrition and Physical Activity, 12(1), 127-138.

241

Chastin, S.F.M., Culhane, B., & Dall, P.M. (2014). Comparison of self-reported measure of sitting time (IPAQ) with objective measurement (activPAL). Physiological measurement, 35(11), 2319-2327.

Chen, Y. M., Li, Y. P., & Yen, M. L. (2015). Gender differences in the predictors of physical activity among assisted living residents. Journal of Nursing Scholarship, 47(3), 211-218.

Chiang, K., Seman, L., Belza, B., & Tsai, J. (2006). “It’s our exercise family”: Ethnic older adults’ experiences in a group-based exercise program. Preventing Chronic Disease, 5(1). Retrieved from http://www.cdc.gov/pcd/issues/2008/jan/06_0170.

Chudyk, A.M., McKay, H.A., Winters, M., Sims-Gould, J., & Ashe, M.C. (2017). Neighborhood walkability, physical activity, and walking for transportation: A cross-sectional study of older adults living on low income. BMC geriatrics, 17(1), 82-95.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed., 413). Hillsdale (NJ): Lawrence Erlbaum Associates.

Cole, R.E., & Horacek, T. (2009). Applying PRECEDE-PROCEED to develop an intuitive eating nondieting approach to weight management pilot program. Journal of nutrition education and behavior, 41(2), 120-126.

Conn, V. S., Hafdahl, A. R., & Mehr, D. R. (2011). Interventions to increase physical activity among healthy adults: meta-analysis of outcomes. American journal of public health, 101(4), 751-758.

Cooper, A. J., Simmons, R. K., Kuh, D., Brage, S., & Cooper, R. (2015). Physical activity, sedentary time and physical capability in early old age: British birth cohort study. PLoS One, 10(5), e0126465.

Copeland, J. L., Ashe, M. C., Biddle, S. J., Brown, W. J., Buman, M. P., Chastin, S., ... & Owen, N. (2017). Sedentary time in older adults: a critical review of measurement, associations with health, and interventions. Br J Sports Med, 51(21), 1539-1539.

Copeland, J. L., Clarke, J., & Dogra, S. (2015). Objectively measured and self-reported sedentary time in older Canadians. Preventive medicine reports, 2, 90-95.

Copeland, J.L., & Esliger, D.W. (2009). Accelerometer assessment of physical activity in active, healthy older adults. Journal of aging and physical activity, 17(1), 17-30.

Costello, E., Kafchinski, M., Vrazel, J., & Sullivan, P. (2011). Motivators, barriers, and beliefs regarding physical activity in an older adult population. Journal of geriatric physical therapy, 34(3), 138-147.

242

Craig, P., Dieppe, P., Macintyre, S., Michie, S., Nazareth, I., & Petticrew, M. (2008). Developing and evaluating complex interventions: the new Medical Research Council guidance. Bmj, 337, a1655.

Craig, P., Dieppe, P., Macintyre, S., Michie, S., Nazareth, I., & Petticrew, M. (2008). Developing and evaluating complex interventions: the new Medical Research Council guidance. Bmj, 337, a1655.

Craske, M. G., & Stein, M. B. (2016). Anxiety. The Lancet. doi:10.1016/S0140-6736(16)30381-6.

Cyarto, E. V., Lautenschlager, N. T., Desmond, P. M., Ames, D., Szoeke, C., Salvado, O., ... & Cox, K. L. (2012). Protocol for a randomized controlled trial evaluating the effect of PA on delaying the progression of white matter changes on MRI in older adults with memory complaints and mild cognitive impairment: The AIBL Active trial. BMC psychiatry, 12(1), 167-178.

Dai, Y., Zhang, C.Y., Zhang, B.Q., Li, Z., Jiang, C., & Huang, H.L. (2016). Social support and the self-rated health of older people: A comparative study in Tainan Taiwan and Fuzhou Fujian province. Medicine, 95(24), S268-S280.

Dane, A. V., & Schneider, B. H. (1998). Program integrity in primary and early secondary prevention: are implementation effects out of control?. Clinical psychology review, 18(1), 23-45.

Davis, M. G., Fox, K. R., Stathi, A., Trayers, T., Thompson, J. L., & Cooper, A. R. (2014). Objectively measured sedentary time and its association with physical function in older adults. Journal of Aging and Physical Activity, 22(4), 474–481.

de Almeida Mendes, M., da Silva, I.C.M., Ramires, V., Reichert, F., Martins, R. & Tomasi, E. (2017). Calibration of raw accelerometer data to measure physical activity: a systematic review. Gait & posture, 61, 98-110.

de Kam, D., Smulders, E., Weerdesteyn, V., & Smits-Engelsman, B.C.M. (2009). Exercise interventions to reduce fall-related fractures and their risk factors in individuals with low bone density: a systematic review of randomized controlled trials. Osteoporosis International, 20(12), 2111-2125.

de Rezende, L. F. M., Rey-López, J. P., Matsudo, V. K. R., & do Carmo Luiz, O. (2014). Sedentary behavior and health outcomes among older adults: a systematic review. BMC public health, 14(1), 333-341.

Delignières, D., Marcellini, A., Brisswalter, J., & Legros, P. (1994). Self-perception of fitness and personality traits. Perceptual and motor Skills, 78(3), 843-851.

Demiris, G., Parker Oliver, D., Capurro, D., & Wittenberg-Lyles, E. (2014). Implementation science: Implications for intervention research in hospice and palliative care. The Gerontologist, 54, 163–171.

243

Department for Communities and Local Government. (2015). The English Index of Multiple Deprivation (IMD) 2015 – Guidance. Retrieved from https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/464430/English_Index_of_Multiple_Deprivation_2015_-_Guidance.pdf.

Department of Health. (2011a). Physical Activity Guidelines in the UK: Review and Recommendations..Retrieved.from.https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/213743/dh_128255.pdf.

Department of Health. (2011b). Start Active, Stay Active: A report on physical activity for health from the four home countries’ Chief Medical Officers. Retrieved from https://www.sportengland.org/media/2928/dh_128210.pdf.

Devereux-Fitzgerald, A., Powell, R., Dewhurst, A., & French, D.P. (2016). The Acceptability of PA Interventions to Older Adults: A Systematic Review and Meta-Synthesis. Social Science & Medicine, 158, 14-23.

Diaz, K. M., Howard, V. J., Hutto, B., Colabianchi, N., Vena, J. E., Blair, S. N., & Hooker, S. P. (2016). Patterns of sedentary behavior in US middle-age and older adults: the REGARDS study. Medicine and science in sports and exercise, 48(3), 430-438.

Dionigi, R. (2007). Resistance training and older adults’ beliefs about psychological benefits: the importance of self-efficacy and social interaction. Journal of Sport and Exercise Psychology, 29(6), 723-746.

Dishman, R. K., & Buckworth, J. (1996). Increasing physical activity: A quantitative synthesis. Medicine and Science in Sports and Exercise, 28(6), 706-719.

Dogra, S., Ashe, M. C., Biddle, S. J., Brown, W. J., Buman, M. P., Chastin, S., ... & Owen, N. (2017). Sedentary time in older men and women: an international consensus statement and research priorities. Br J Sports Med, bjsports-2016.

Dogra, S., & Stathokostas, L. (2012). Sedentary behavior and physical activity are independent predictors of successful aging in middle-aged and older adults. Journal of aging research, 2012, 1-9.

Doherty, A., Jackson, D., Hammerla, N., Plötz, T., Olivier, P., Granat, M.H., White, T., van Hees, V.T., Trenell, M.I., Owen, C.G., & Preece, S.J. (2017). Large scale population assessment of physical activity using wrist worn accelerometers: The UK Biobank Study. PloS one, 12(2), p.e0169649.

Donaldson, L. J. (2004). At least five a week: Evidence on the impact of PA and its relationship to health. Department of Health.

Duncan, T. E., & McAuley, E. (1993). Social support and efficacy cognitions in exercise adherence: A latent growth curve analysis. Journal of behavioral medicine, 16(2), 199-218.

244

Dunstan, D. W., Howard, B., Healy, G. N., & Owen, N. (2012). Too much sitting–a health hazard. Diabetes research and clinical practice, 97(3), 368-376.

Durlak, J. A., & DuPre, E. P. (2008). Implementation matters: A review of research on the influence of implementation on program outcomes and the factors affecting implementation. American journal of community psychology, 41(3-4), 327-350.

Edge Hill University. (2017). Research data management. Retrieved from https://www.edgehill.ac.uk/research/files/2012/05/Strategy-Output-Research-Data-Management-Guidelines-RO-GOV-04.pdf.

Edge Hill University. (2017). Code of Practice for the conduct of research. Retrieved from https://www.edgehill.ac.uk/documents/files/code-of-practice-for-research.pdf.

Edwards, M.K., & Loprinzi, P.D. (2016). The Association between Sedentary Behavior and Cognitive Function among Older Adults May Be Attenuated with Adequate Physical Activity. Journal of Physical Activity and Health, 14(1), 52-58.

Eggenberger, P., Theill, N., Holenstein, S., Schumacher, V., & de Bruin, E. D. (2015). Multicomponent physical exercise with simultaneous cognitive training to enhance dual-task walking of older adults: a secondary analysis of a 6-month randomized controlled trial with 1-year follow-up. Clinical interventions in aging, 10, 1711-1732.

Eisinga, R., Franses, P. H., & Vergeer, M. (2011). Weather conditions and daily television use in the Netherlands, 1996–2005. International journal of biometeorology, 55(4), 555-564.

Ekelund, U., Brage, S., Besson, H., Sharp, S., & Wareham, N.J. (2008). Time spent being sedentary and weight gain in healthy adults: reverse or bidirectional causality?–. The American journal of clinical nutrition, 88(3), 612-617.

Elwood, P., Galante, J., Pickering, J., Palmer, S., Bayer, A., Ben-Shlomo, Y., ... & Gallacher, J. (2013). Healthy lifestyles reduce the incidence of chronic diseases and dementia: evidence from the Caerphilly Cohort Study. PLOS One, 8(12), 1-7.

Emdadi, S., Hazavehie, S.M.M., Soltanian, A., Bashirian, S., & Heidari Moghadam, R. (2015). Predictive Factors of Regular Physical Activity among Middle-Aged Women in the West of Iran, Hamadan: Application of PRECEDE Model. Journal of research in health sciences, 15(4), 244-249.

Enders, C.K., Mistler, S.A. and Keller, B.T., 2016. Multilevel multiple imputation: A review and evaluation of joint modeling and chained equations imputation. Psychological methods, 21(2), 222-240.

245

Erickson, K. I., Weinstein, A. M., & Lopez, O. L. (2012). Physical activity, brain plasticity, and Alzheimer's disease. Archives of medical research, 43(8), 615-621.

Esliger, D. W., Rowlands, A. V., Hurst, T. L., Catt, M., Murray, P., & Eston, R. G. (2011). Validation of the GENEA Accelerometer, 43(6), 1085-1093.

Evenson, K. R., Catellier, D. J., Gill, K., Ondrak, K. S., & McMurray, R. G. (2008). Calibration of two objective measures of physical activity for children. Journal of Sports Sciences, 26, 1557–1565.

Evenson, K.R., Wen, F., Herring, A.H., Di, C., LaMonte, M.J., Tinker, L.F., Lee, I.M., Rillamas-Sun, E., LaCroix, A.Z. & Buchner, D.M. (2015). Calibrating physical activity intensity for hip-worn accelerometry in women age 60 to 91 years: The Women's Health Initiative OPACH Calibration Study. Preventive medicine reports, 2, 750-756.

Ewald, B., McEvoy, M., & Attia, J. (2010). Pedometer counts superior to physical activity scale for identifying health markers in older adults. British Journal of Sports Medicine, 44(10), 756-761.

Fairclough, S., Noonan, R., Rowlands, A., Van Hees, V., Knowles, Z., & Boddy, L. (2016). Wear compliance and activity in children wearing wrist and hip mounted accelerometers. Medicine & Science in Sports & Exercise, 48(2), 245-253.

Falck, R.S., Davis, J.C., & Liu-Ambrose, T. (2016). What is the association between sedentary behaviour and cognitive function? A systematic review. British Journal of Sports Medicine, 51(10), 800-811.

Farquhar, M. C., Ewing, G., & Booth, S. (2011). Using mixed methods to develop and evaluate complex interventions in palliative care research. Palliative Medicine, 25(8), 748-757.

Fernandez-Alonso, L., Muñoz-García, D., & Touche, R. La. (2016). The level of physical activity affects the health of older adults despite being active. Journal of Exercise Rehabilitation, 12(3), 194–201.

Ferraro, K. F., & Wilkinson, L. R. (2013). Alternative measures of self-rated health for predicting mortality among older people: Is past or future orientation more important?. The Gerontologist, 55(5), 836-844.

Field, A.P., & Wilcox, R.R. (2017). Robust statistical methods: a primer for clinical psychology and experimental psychopathology researchers. Behaviour research and therapy, 98, 19-38.

Fisher, K.L., Harrison, E.L., Bruner, B.G., Lawson, J.A., Reeder, B.A., Ashworth, N.L., Sheppard, M.S., & Chad, K.E. (2018). Predictors of Physical Activity Levels in Community-dwelling Older Adults: A Multivariate Approach Based on a Socio-Ecological Framework. Journal of Aging and Physical Activity, 26(1), 114-120.

246

Fleig, L., McAllister, M. M., Brasher, P., Cook, W. L., Guy, P., Puyat, J. H., ... & Ashe, M. C. (2016). Sedentary behavior and physical activity patterns in older adults after hip fracture: a call to action. Journal of aging and physical activity, 24(1), 79-84.

Ford, I., & Norrie, J. (2016). Pragmatic trials. New England journal of medicine, 375(5), 454-463.

Fox, K. R., & Corbin, C. D. (1989). The physical self-perception profile: Development and preliminary validation. Journal of Sport and Exercise Psychology, 11, 408–430.

Fox, K.R., Ku, P.W., Hillsdon, M., Davis, M.G., Simmonds, B.A., Thompson, J.L., Stathi, A., Gray, S.F., Sharp, D.J., & Coulson, J.C. (2014). Objectively assessed physical activity and lower limb function and prospective associations with mortality and newly diagnosed disease in UK older adults: an OPAL four-year follow-up study. Age and ageing, 44(2), 261-268.

Forbes, S. C., Forbes, D., Forbes, S., Blake, C. M., Chong, L. Y., Thiessen, E. J., ... & Rutjes, A. W. (2015). Exercise interventions for preventing dementia or delaying cognitive decline in people with mild cognitive impairment. The Cochrane Library, 4, CD011706.

Franco, M.R., Tong, A., Howard, K., Sherrington, C., Ferreira, P.H., Pinto, R.Z., & Ferreira, M.L. (2015). Older People's Perspectives on Participation in Physical Activity: A Systematic Review and Thematic Synthesis of Qualitative Literature. British Journal of Sports Medicine, 49(19), 1268-1276.

Frank, J. C., Coviak, C. P., Healy, T. C., Belza, B., & Casado, B. L. (2008). Addressing fidelity in evidence-based health promotion programs for older adults. Journal of Applied Gerontology, 27(1), 4-33.

Freedson, P., Bowles, H. R., Troiano, R., & Haskell, W. (2012). Assessment of physical activity using wearable monitors: recommendations for monitor calibration and use in the field. Med Sci Sports Exerc, 44(1 Suppl 1), S1-4.

French, D.P., Olander, E.K., Chisholm, A., & McSharry, J. (2015). Which Behaviour Change Techniques are Most Effective at Increasing Older Adults’ Self-efficacy and Physical Activity Behaviour? A Systematic Review. Annals of Behavioural Medicine, 48(2), 225-234.

Furtado, G. E., Uba-Chupel, M., Carvalho, H. M., Souza, N. R., Ferreira, J. P., & Teixeira, A. M. (2016). Effects of a chair-yoga exercises on stress hormone levels, daily life activities, falls and physical fitness in institutionalized older adults. Complementary therapies in clinical practice, 24, 123-129.

Gagliardi, A. R., Faulkner, G., Ciliska, D., & Hicks, A. (2015). Factors Contributing to the Effectiveness of Physical Activity Counselling in Primary Care: A Realist Systematic Review. Patient Education and Counseling, 98(4), 412-419.

247

Garcia, J.A., Schoene, D., Lord, S.R., Delbaere, K., Valenzuela, T., & Navarro, K.F. (2016). A Bespoke Kinect Stepping Exergame for Improving Physical and Cognitive Function in Older People: A Pilot Study. Games for Health Journal, 5(6), 382-388.

Gardiner, C., Geldenhuys, G., & Gott, M. (2016). Interventions to Reduce Social Isolation and Loneliness Among Older People: An Integrative Review. Health & Social Care in the Community, 26(2), 1-17.

Gardner, B., Smith, L., Lorencatto, F., Hamer, M., & Biddle, S. J. (2016). How to reduce sitting time? A review of behaviour change strategies used in sedentary behaviour reduction interventions among adults. Health psychology review, 10(1), 89-112.

Gardiner, P. A., Eakin, E. G., Healy, G. N., & Owen, N. (2011). Feasibility of reducing older adults’ sedentary time. American Journal of Preventive Medicine, 41, 174-177.

Gearing, R. E., El-Bassel, N., Ghesquiere, A., Baldwin, S., Gillies, J., & Ngeow, E. (2011). Major ingredients of fidelity: A review and scientific guide to improving quality of intervention research implementation. Clinical Psychology Review, 31(1), 79–88.

Gennuso, K. P., Thraen-Borowski, K. M., Gangnon, R. E., & Colbert, L. H. (2016). Patterns of sedentary behavior and physical function in older adults. Aging Clinical and Experimental Research, 28(5), 943–950.

Gillespie, L.D., Robertson, M.C., Gillespie, W.J., Sherrington, C., Gates, S., Clemson, L.M., & Lamb, S.E. (2012). Interventions for Preventing Falls in Older People Living in the Community. Cochrane Handbook for Systematic Reviews of Interventions, 9, 1-420.

Glasgow, R.E., Green, L.W., Taylor, M.V., & Stange, K.C. (2012). An evidence integration triangle for aligning science with policy and practice. American journal of preventive medicine, 42(6), 646-654.

Glasgow, R. E., Vogt, T. M., & Boles, S. M. (1999). Evaluating the public health impact of health promotion interventions: the RE-AIM framework. American journal of public health, 89(9), 1322-1327.

Godfrey, A., Lord, S., Galna, B., Mathers, J. C., Burn, D. J., & Rochester, L. (2013). The association between retirement and age on physical activity in older adults. Age and ageing, 43(3), 386-393.

Gomersall, S. R., Rowlands, A. V., English, C., Maher, C., & Olds, T. S. (2013). The ActivityStat hypothesis: The concept, the evidence and the methodologies. Sports Medicine, 43, 135-149.

Gorman, E., Hanson, H.M., Yang, P.H., Khan, K.M., Liu-Ambrose, T., & Ashe, M.C. (2014). Accelerometry analysis of physical activity and sedentary behavior in

248

older adults: a systematic review and data analysis. European Review of Aging and Physical Activity, 11(1), 35-49.

Gottfredson, D. C., Cook, T. D., Gardner, F. E., Gorman-Smith, D., Howe, G. W., Sandler, I. N., & Zafft, K. M. (2015). Standards of evidence for efficacy, effectiveness, and scale-up research in prevention science: Next generation. Prevention Science, 16(7), 893-926.

Gray, P.M., Murphy, M.H., Gallagher, A.M., & Simpson, E.E. (2015). Motives and Barriers to Physical Activity among Older Adults of Different Socio-Economic Status. Journal of Aging & Physical Activity, 24(3), 419-429.

Greaney, M.L., Lees, F.D., Blissmer, B.J., Riebe, D., & Clark, P.G. (2016). Psychosocial Factors Associated With Physical Activity in Older Adults. Annual Review of Gerontology and Geriatrics, 36(1), 273-91.

Green, L. W. (2001). From research to “best practices” in other settings and populations. American journal of health behavior, 25(3), 165-178.

Green, L. W., & Kreuter, M. W. (2005). Health Program Planning: An Educational and Ecological Approach (4th ed.). New York: McGraw-Hill.

Green, L. W., & Kreuter, M. W. 1999. Health Promotion Planning: An Educational and Ecological Approach. New York: McGraw-Hill.

Green, L. W., Kreuter, M. W., Deeds, S. G., & Partidge, K. B. (1980): Health education planning: A diagnostic approach, Mayfield, California.

Grund, S., Lüdtke, O., & Robitzsch, A. (2016). Multiple imputation of missing covariate values in multilevel models with random slopes: A cautionary note. Behavior Research Methods, 48(2), 640-649.

Guralnik, J. M., Simonsick, E. M., Ferrucci, L., Glynn, R. J., Berkman, L. F., Blazer, D. G., ... & Wallace, R. B. (1994). A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. Journal of gerontology, 49(2), M85-M94.

Haggis, C., Sims-Gould, J., Winters, M., Gutteridge, K., & McKay, H.A. (2013). Sustained impact of community-based physical activity interventions: key elements for success. BMC Public Health, 13(1), p.892-899.

Hagströmer, M., Troiano, R. P., Sjöström, M., & Berrigan, D. (2010). Levels and patterns of objectively assessed physical activity—a comparison between Sweden and the United States. American journal of epidemiology, 171(10), 1055-1064.

Hall, K.S., Howe, C.A., Rana, S.R., Martin, C.L., & Morey, M.C. (2013). METs and accelerometry of walking in older adults: standard versus measured energy cost. Medicine and science in sports and exercise, 45(3), 574-582.

249

Hall, K. S., & McAuley, E. (2010). Individual, social environmental and physical environmental barriers to achieving 10 000 steps per day among older women. Health education research, 25(3), 478-488.

Hallal, P.C., Andersen, L.B., Bull, F.C., Guthold, R., Haskell, W., Ekelund, U., & Lancet Physical Activity Series Working Group. (2012). Global physical activity levels: surveillance progress, pitfalls, and prospects. The lancet, 380(9838), 247-257.

Hamer, M., Kivimaki, M., & Steptoe, A. (2012). Longitudinal patterns in physical activity and sedentary behaviour from mid-life to early old age: a substudy of the Whitehall II cohort. J Epidemiol Community Health, 66(12), 1110-1115.

Hamer, M., Lavoie, K.L., & Bacon, S.L. (2014). Taking up physical activity in later life and healthy ageing: the English longitudinal study of ageing. British Journal of Sports Medicine, 48(3), 239-243.

Hardeman, W., Michie, S., Fanshawe, T., Prevost, A. T., Mcloughlin, K., & Kinmonth, A. L. (2008). Fidelity of delivery of a physical activity intervention: predictors and consequences. Psychology and Health, 23(1), 11-24.

Hardeman, W., Michie, S., Fanshawe, T., Prevost, A. T., Mcloughlin, K., & Kinmonth, A. L. (2008). Fidelity of delivery of a physical activity intervention: predictors and consequences. Psychology and Health, 23(1), 11-24.

Hargens, T. A., Deyarmin, K. N., Snyder, K. M., Mihalik, A. G., & Sharpe, L. E. (2017). Comparison of wrist-worn and hip-worn activity monitors under free living conditions. Journal of medical engineering & technology, 41(3), 200-207.

Harris, T., Kerry, S.M., Limb, E.S., Furness, C., Wahlich, C., Victor, C.R., Iliffe, S., Whincup, P.H., Ussher, M., Ekelund, U., & Fox-Rushby, J. (2018). Physical activity levels in adults and older adults 3–4 years after pedometer-based walking interventions: Long-term follow-up of participants from two randomised controlled trials in UK primary care. PLoS medicine, 15(3), e1002526.

Harris, T., Kerry, S.M., Victor, C.R., Ekelund, U., Woodcock, A., Iliffe, S., Whincup, P.H., Beighton, C., Ussher, M., Limb, E.S., & David, L. (2015). A primary care nurse-delivered walking intervention in older adults: PACE (pedometer accelerometer consultation evaluation)-Lift cluster randomised controlled trial. PLoS medicine, 12(2), e1001783.

Hart, P. D. (2016). Meeting Recommended Levels of Physical Activity and Health-Related Quality of Life in Rural Adults. Journal of Lifestyle Medicine, 6(1), 1–6.

Harvey, J. a, Chastin S, F. M., & Skelton, D. A. (2014). How Sedentary are Older People? A Systematic Review of the Amount of Sedentary Behavior. Journal of Aging and Physical Activity, 23(3), 471–87.

Haselwandter, E. M., Corcoran, M. P., Folta, S. C., Hyatt, R., Fenton, M., & Nelson, M. E. (2015). The Built Environment, Physical Activity, and Aging in the United

250

States: A State of the Science Review. Journal of Aging and Physical Activity, 23(2), 323-329.

Haskell, W.L., Lee, I.M., Pate, R.R., Powell, K.E., Blair, S.N., Franklin, B.A., Macera, C.A., Heath, G.W., Thompson, P.D., & Bauman, A. (2007). Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation, 116(9), 1081-1093.

Hasson, H. (2010). Systematic evaluation of implementation fidelity of complex interventions in health and social care. Implementation Science, 5(1), 67-75.

Hawe, P., Shiell, A., & Riley, T. (2004). Complex interventions: how “out of control” can a randomised controlled trial be?. BMJ: British Medical Journal, 328(7455), 1561-1563.

Hawley-Hague, H., Horne, M., Skelton, D. A., & Todd, C. (2016). Older Adults’ Uptake and Adherence to Exercise Classes: Instructors’ Perspectives. Journal of Aging and Physical Activity, 24(1), 119-128.

Haywood, N., Foley, K., Pugh, A., Warden, H., Donelle, J., Telega, L., ... & Talarico, R. (2018). Physical activity, sedentary time, sleep duration, and self-rated health in older adults: A compositional analysis, (Doctoral dissertation).

Healthy People 2020. (2014). Healthy People 2020 Framework. The Vision, Mission, and Goals of Healthy People 2020. Overarching Goals. Retrieved from http://www.healthypeople.gov/sites/default/files/ HP2020Framework.pdf.

Healy, G. N., Clark, B. K., Winkler, E. A. H., Gardiner, P. A., Brown, W. J., & Matthews, C. E. (2011). Measurement of Adults’ Sedentary Time in Population-Based Studies. American Journal of Preventive Medicine, 41(2), 216–227.

Healy, G.N., Winkler, E.A., Owen, N., Anuradha, S., & Dunstan, D.W. (2015). Replacing sitting time with standing or stepping: associations with cardio-metabolic risk biomarkers. European Heart Journal, 36(39), 2643-2649.

Hensley, L. D., Ainsworth, B. E., & Ansorge, C. J. (1993). Assessment of Physical Activity—Professional Accountability in Promoting Active Lifestyles. Journal of Physical Education, Recreation & Dance, 64(1), 56-64.

Heo, J., Chun, S., Kim, B., Ryu, J., & Lee, Y. (2017). Leisure activities, optimism, and personal growth among the young-old, old-old, and oldest-old. Educational Gerontology, 43(6), 289-299.

Hildebrand, M., & Neufeld, P. (2009). Recruiting Older Adults into a Physical Activity Promotion Program: Active Living Every Day Offered in a Naturally Occurring Retirement Community. The Gerontologist, 49(5), 702-710.

251

http://www.healthypeople.gov/sites/default/files/

Hildebrand, M., Hansen, B. H., van Hees, V. T., & Ekelund, U. (2016). Evaluation of raw acceleration sedentary thresholds in children and adults. Scandinavian Journal of Medicine and Science in Sports, 1-10.

Hildebrand, M., Van Hees, V. T., Hansen, B. H., & Ekelund, U. (2014). Age group comparability of raw accelerometer output from wrist- and hip-worn monitors. Medicine and Science in Sports and Exercise, 46(9), 1816-1824.

Hoare, E., Stavreski, B., Jennings, G.L., & Kingwell, B.A. (2017). Exploring motivation and barriers to physical activity among active and inactive Australian adults. Sports, 5(3), 47-54.

Hobbs, N., Godfrey, A., Lara, J., Errington, L., Meyer, T. D., Rochester, L., ... & Sniehotta, F. F. (2013). Are behavioral interventions effective in increasing physical activity at 12 to 36 months in adults aged 55 to 70 years? A systematic review and meta-analysis. BMC medicine, 11(1), 75-86.

Hong, E. (2015). Age differences in health-related quality of life among South Korean elderly. Journal of Nursing and Health Sciences, 1(4), 34-39.

Hoppmann, C.A., Lee, J.C.M., Ziegelmann, J.P., Graf, P., Khan, K.M., & Ashe, M.C. (2015). Precipitation and Physical Activity in Older Adults: The Moderating Role of Functional Mobility and Physical Activity Intentions. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 72(5), 792–800.

Huerta, J. M., Chirlaque, M. D., Tormo, M. J., Buckland, G., Ardanaz, E., Arriola, L., ... & Molina-Montes, E. (2016). Work, household, and leisure-time physical activity and risk of mortality in the EPIC-Spain cohort. Preventive medicine, 85, 106-112.

Hughes, S. L., Leith, K. H., Marquez, D. X., Moni, G., Nguyen, H. Q., Desai, P., & Jones, D. L. (2011). Physical activity and older adults: expert consensus for a new research agenda. The Gerontologist, 51(6), 822-832.

Hurtig-Wennlöf, A., Hagströmer, M., & Olsson, L. A. (2010). The International PA Questionnaire modified for the elderly: aspects of validity and feasibility. Public health nutrition, 13(11), 1847-1854.

Husted, H. M., & Llewellyn, T. L. (2017). The accuracy of pedometers in measuring walking steps on a treadmill in college students. International journal of exercise science, 10(1), 146-153.

Ikezoe, T., Asakawa, Y., Shima, H., Kishibuchi, K., & Ichihashi, N. (2013). Daytime physical activity patterns and physical fitness in institutionalized elderly women: An exploratory study. Archives of Gerontology and Geriatrics, 57(2), 221–225.

Innes, E., & Straker, L. (1999). Validity of work-related assessments. Work, 13(2), 125-152.

252

IPAQ Research Committee. 2005. Guidelines for Data Processing and Analysis of the International Physical Activity Questionnaire (IPAQ) – Short and Long Forms, 17. 1-15.

Jago, R., Zakeri, I., Baranowski, T., & Watson, K. (2007). Decision boundaries and receiver operating characteristic curves: new methods for determining accelerometer cutpoints. Journal of sports sciences, 25(8), 937-944.

Jancey, J. M., Clarke, A., Howat, P. A., Lee, A. H., Shilton, T., & Fisher, J. (2008). A physical activity program to mobilize older people: A practical and sustainable approach. The Gerontologist, 48(2), 251-257.

Jansen, F.M., Prins, R.G., Etman, A., van der Ploeg, H.P., De Vries, S.I., van Lenthe, F.J., & Pierik, F.H. (2015). Physical activity in non-frail and frail older adults. PloS one, 10(4), e0123168.

Jefferis, B.J., Parsons, T.J., Sartini, C., Ash, S., Lennon, L.T., Wannamethee, S.G., Lee, I.M., & Whincup, P.H. (2016). Does duration of physical activity bouts matter for adiposity and metabolic syndrome? A cross-sectional study of older British men. International Journal of Behavioral Nutrition and Physical Activity, 13(1), 36-46.

Jefferis, B.J., Sartini, C., Lee, I.M., Choi, M., Amuzu, A., Gutierrez, C., Casas, J.P., Ash, S., Lennnon, L.T., Wannamethee, S.G., & Whincup, P.H. (2014). Adherence to Physical Activity Guidelines in Older Adults, Using Objectively Measured Physical Activity in a Population-Based Study. BMC Public Health, 14(1), 1-9.

John, D., Sasaki, J., Staudenmayer, J., Mavilia, M., & Freedson, P. S. (2013). Comparison of raw acceleration from the GENEA and ActiGraph™ GT3X+ activity monitors. Sensors, 13(11), 14754-14763.

Johnson, L. G., Butson, M. L., Polman, R. C., Raj, I. S., Borkoles, E., Scott, D., ... & Jones, G. (2016). Light physical activity is positively associated with cognitive performance in older community-dwelling adults. Journal of science and medicine in sport, 19(11), 877-882.

Kato, Y., Islam, M. M., Koizumi, D., Rogers, M. E., & Takeshima, N. (2018). Effects of a 12-week marching in place and chair rise daily exercise intervention on ADL and functional mobility in frail older adults. Journal of physical therapy science, 30(4), 549-554.

Keadle, S. K., McKinnon, R., Graubard, B. I., & Troiano, R. P. (2016). Prevalence and Trends in Physical Activity among Older Adults in the United States: A Comparison across Three National Surveys. Preventive Medicine, 89, 37-43.

Kemp, B., & Ettelson, D. (2001). Quality of life while living and aging with a spinal cord injury and other impairments. Topics in Spinal Cord Injury Rehabilitation, 6(3), 116-127.

253

Kerr, J., Rosenberg, D., Millstein, R. A., Bolling, K., Crist, K., Takemoto, M., ... & Buchner, D. (2018). Cluster randomized controlled trial of a multilevel physical activity intervention for older adults. International Journal of Behavioral Nutrition and Physical Activity, 15(1), 32-40.

Kerr, J., Rosenberg, D.E., Nathan, A., Millstein, R.A., Carlson, J.A., Crist, K., Wasilenko, K., Bolling, K., Castro, C.M., Buchner, D.M., & Marshall, S.J. (2012). Applying the ecological model of behavior change to a physical activity trial in retirement communities: description of the study protocol. Contemporary clinical trials, 33(6), 1180-1188.

Kim, H. and Lee, T.H. (2018). Strategic CSR Communication: A Moderating Role of Transparency in Trust Building. International Journal of Strategic Communication, 12(2), 107-124.

Kim, H., & Lee, T.H. (2018). Strategic CSR Communication: A Moderating Role of Transparency in Trust Building. International Journal of Strategic Communication, 12(2), 107-124.

Kim, J., Im, J.S., & Choi, Y.H. (2016). Objectively measured sedentary behavior and moderate-to-vigorous physical activity on the health-related quality of life in US adults: The National Health and Nutrition Examination Survey 2003–2006. Quality of Life Research, 26(5), 1315-1326.

Kim, Y., Barry, V. W., & Kang, M. (2015). Validation of the ActiGraph GT3X and activPAL Accelerometers for the Assessment of Sedentary Behavior. Measurement in Physical Education and Exercise Science, 19(3), 125-137.

Kirchengast, S., & Haslinger, B. (2008). Gender differences in health-related quality of life among healthy aged and old-aged Austrians: cross-sectional analysis. Gender Medicine, 5(3), 270-278.

Knowles, Z. R., Parnell, D., Stratton, G., & Ridgers, N. D. (2013). Learning from the Experts: Exploring Playground Experience and Activities using a Write and Draw Technique. Journal of Physical Activity & Health, 10(3), 406-415.

Ko, S.U., Jerome, G.J., Simonsick, E.M., Studenski, S., & Ferrucci, L. (2018). Differential Gait Patterns by Falls History and knee pain status in Healthy Older Adults: Results From the Baltimore Longitudinal Study of Aging. Journal of aging and physical activity, 2018, 1-18.

Koeneman, M. A., Verheijden, M. W., Chinapaw, M. J., & Hopman-Rock, M. (2011). Determinants of physical activity and exercise in healthy older adults: a systematic review. International Journal of Behavioral Nutrition and Physical Activity, 8(1), 142-156.

Kohl 3rd, H.W., Craig, C.L., Lambert, E.V., Inoue, S., Alkandari, J.R., Leetongin, G., Kahlmeier, S., & Lancet Physical Activity Series Working Group. (2012). The

254

pandemic of physical inactivity: global action for public health. The Lancet, 380(9838), 294-305.

Koorts, H., Eakin, E., Estabrooks, P., Timperio, A., Salmon, J., & Bauman, A. (2018). Implementation and scale up of population physical activity interventions for clinical and community settings: the PRACTIS guide. International Journal of Behavioral Nutrition and Physical Activity, 15(1), 51-61.

Kosteli, M.C., Williams, S.E., & Cumming, J. (2016). Investigating the psychosocial determinants of physical activity in older adults: a qualitative approach. Psychology & health, 31(6), 730-749.

Koster, A., Caserotti, P., Patel, K. V., Matthews, C. E., Berrigan, D., Van Domelen, D. R., ... & Harris, T. B. (2012). Association of sedentary time with mortality independent of moderate to vigorous physical activity. PloS one, 7(6), e37696.

Kowalski, K., Rhodes, R., Naylor, P. J., Tuokko, H., & MacDonald, S. (2012). Direct and indirect measurement of physical activity in older adults: a systematic review of the literature. International Journal of Behavioral Nutrition and Physical Activity, 9(1), 148-168.

Ku, P.W., Fox, K.R., Liao, Y., Sun, W.J., & Chen, L.J. (2016). Prospective associations of objectively assessed physical activity at different intensities with subjective well-being in older adults. Quality of Life Research, 25(11), 2909-2919.

Ku, P. W., Steptoe, A., Liao, Y., Sun, W. J., & Chen, L. J. (2018). Prospective relationship between objectively measured light physical activity and depressive symptoms in later life. International journal of geriatric psychiatry, 33(1), 58-65.

Kuosmanen, K., Rovio, S., Kivipelto, M., Tuomilehto, J., Nissinen, A., & Kulmala, J. (2016). Determinants of self-rated health and self-rated physical fitness in middle and old age. European Journal of Mental Health, 11, 128-143.

Kwan, M., Woo, J., & Kwok, T. (2004). The standard oxygen consumption value equivalent to one metabolic equivalent (3.5 ml/min/kg) is not appropriate for elderly people. International journal of food sciences and nutrition, 55(3), 179-182.

Lambert, J. D., Greaves, C. J., Farrand, P., Cross, R., Haase, A. M., & Taylor, A. H. (2017). Assessment of fidelity in individual level behaviour change interventions promoting physical activity among adults: a systematic review. BMC public health, 17(1), 765-777.

Landi, F., Abbatecola, A.M., Provinciali, M., Corsonello, A., Bustacchini, S., Manigrasso, L., Cherubini, A., Bernabei, R., & Lattanzio, F. (2010). Moving against frailty: does physical activity matter?. Biogerontology, 11(5), 537-545.

255

Landry, G.J., Falck, R.S., Beets, M.W., & Liu-Ambrose, T. (2015). Measuring physical activity in older adults: calibrating cutpoints for the MotionWatch 8©. Frontiers in aging neuroscience, 7, 166-173.

Langan, M. E., & Marotta, S. A. (2000). Physical Activity and Perceived Self Efficacy in‐ Older Adults. Adultspan Journal, 2(1), 29-43.

Lautenschlager, N. T., Cox, K., & Kurz, A. F. (2010). Physical activity and mild cognitive impairment and Alzheimer’s disease. Current neurology and neuroscience reports, 10(5), 352-358.

Lawton, R., Mceachan, R., Jackson, C., West, R., & Conner, M. (2014). Intervention fidelity and effectiveness of a UK worksite physical activity intervention funded by the Bupa Foundation, UK. Health promotion international, 30(1), 38-49.

Lean, M., Lara, J., & Hill, J. O. (2007): Strategies for preventing obesity, 20-21. In S. N andL. M (Eds): ABC of Obesity Blackwell Publishing, Oxford.

Lee, I.M., & Shiroma, E.J. (2014). Using accelerometers to measure physical activity in large-scale epidemiological studies: issues and challenges. British Journal of Sports Medicine, 48, 197–201.

Lehne, G., & Bolte, G. (2017). Impact of universal interventions on social inequalities in physical activity among older adults: an equity-focused systematic review. International Journal of Behavioral Nutrition and Physical Activity, 14(1), 20-34.

Leitzmann, M.F., Jochem, C., & Schmid, D. (2017). Sedentary Behaviour Epidemiology. Cham, Switzerland: Springer International Publishing.

Lemmink, K.A.P.M. (1996). De Groninger Fitheidstest voor Ouderen: Ontwikketing van een meetinstrument [The Groningen Fitness Test for the Elderly: Development of a measuring instrument. Thesis in Dutch language, English summary]. Groningen, The Netherlands: Department of Human Movement Sciences, University of Groningen.

Lewis, B.A., Napolitano, M.A., Buman, M.P., Williams, D.M., & Nigg, C.R. (2017). Future directions in physical activity intervention research: expanding our focus to sedentary behaviors, technology, and dissemination. Journal of behavioral medicine, 40(1), 112-126.

Lewis, M., Peiris, C. L., & Shields, N. (2017). Long-term home and community-based exercise programs improve function in community-dwelling older people with cognitive impairment: a systematic review. Journal of physiotherapy, 63(1), 23-29.

Li, F., Fisher, K. J., Bauman, A., Ory, M. G., Chodzko-Zajko, W., Harmer, P., ... & Cleveland, M. (2005). Neighborhood influences on physical activity in middle-

256

aged and older adults: a multilevel perspective. Journal of aging and physical activity, 13(1), 87-114.

Li, J., & Siegrist, J. (2012). Physical activity and risk of cardiovascular disease—a meta-analysis of prospective cohort studies. International journal of environmental research and public health, 9(2), 391-407.

Lillo, N., Palomo-Vélez, G., Fuentes, E., & Palomo, I. (2015). Role of physical activity in cardiovascular disease prevention in older adults. Sport Sciences for Health, 11(3), 227–233.

Lim, J., Kim, H.K., Hwang, C.L., Yoo, J.K., Hwang, M.H., Handberg, E.M., Nichols, W.W., & Christou, D.D. (2017). Peripheral Vascular Adaptations to All-Extremity Aerobic Exercise Training in Healthy Older Adults. The FASEB Journal, 31(1_supplement), lb669-lb669.

Lindgren, M., Börjesson, M., Ekblom, Ö., Bergström, G., Lappas, G., & Rosengren, A. (2016). Physical activity pattern, cardiorespiratory fitness, and socioeconomic status in the SCAPIS pilot trial—a cross-sectional study. Preventive medicine reports, 4, 44-49.

Linnan, L.A., Sterba, K.R., Lee, A.M., Bontempi, J.B., Yang, J., & Crump, C. (2005). Planning and the professional preparation of health educators: implications for teaching, research, and practice. Health promotion practice, 6(3), 308-319.

Lipsey, M. W., & Wilson, D. B. (1993). The efficacy of psychological, educational, and behavioral treatment: Confirmation from meta-analysis. American psychologist, 48(12), 1181-1209.

Little, R. J., & Rubin, D. B. (2014). Statistical analysis with missing data (Vol. 333). John Wiley & Sons.

Livingston, G., Johnston, K., Katona, C., Paton, J., Lyketsos, C. G., & Old Age Task Force of the World Federation of Biological Psychiatry. (2014). Systematic review of psychological approaches to the management of neuropsychiatric symptoms of dementia. American Journal of Psychiatry, 162(11), 1-37.

Lohne-Seiler, H., Hansen, B.H., Kolle, E., & Anderssen, S.A. (2014). Accelerometer-determined physical activity and self-reported health in a population of older adults (65–85 years): a cross-sectional study. BMC Public Health, 14(1), 284-293.

Lok, N., Lok, S., & Canbaz, M. (2017). The effect of physical activity on depressive symptoms and quality of life among elderly nursing home residents: randomized controlled trial. Archives of gerontology and geriatrics, 70, 92-98.

López-Rodríguez, C., Laguna, M., Gómez-Cabello, A., Gusi, N., Espino, L., Villa, G., ... & Aznar, S. (2017). Validation of the self-report EXERNET questionnaire for

257

measuring physical activity and sedentary behavior in elderly. Archives of gerontology and geriatrics, 69, 156-161.

Loprinzi, P. D. (2013). Objectively measured light and moderate-to-vigorous physical activity is associated with lower depression levels among older US adults. Aging & Mental Health, 17(7), 801–5.

Loprinzi, P.D., Lee, H., & Cardinal, B.J. (2013). Dose response association between physical activity and biological, demographic, and perceptions of health variables. Obesity facts, 6(4), 380-392.

Lugade, V., Fortune, E., Morrow, M. & Kaufman, K. (2014). Validity of using tri-axial accelerometers to measure human movement—Part I: Posture and movement detection. Medical Engineering and Physics, 36(2), 169-176.

Lynch, B. M., Dunstan, D. W., Winkler, E., Healy, G. N., Eakin, E., & Owen, N. (2011). Objectively assessed physical activity, sedentary time and waist circumference among prostate cancer survivors: findings from the National Health and Nutrition Examination Survey (2003–2006). European journal of cancer care, 20(4), 514-519.

Lyons, E. J., Swartz, M. C., Lewis, Z. H., Martinez, E., & Jennings, K. (2017). Feasibility and acceptability of a wearable technology physical activity intervention with telephone counseling for mid-aged and older adults: a randomized controlled pilot trial. JMIR mHealth and uHealth, 5(3), e28-e63.

Mackintosh, K. A., Knowles, Z. R., Ridgers, N. D., & Fairclough, S. J. (2011). Using Formative Research to Develop Change!: A Curriculum-Based Physical Activity Promoting Intervention. BMC Public Health, 11(1), 831-843.

Mackintosh, K.A., Fairclough, S.J., Stratton, G., & Ridgers, N.D. (2012). A calibration protocol for population-specific accelerometer cutpoints in children. PLoS One, 7(5), e36919.

Mañas, A., del Pozo-Cruz, B., García-García, F.J., Guadalupe-Grau, A., & Ara, I. (2017). Role of objectively measured sedentary behaviour in physical performance, frailty and mortality among older adults: A short systematic review. European journal of sport science, 17(7), 940-953.

Mañas, A., del Pozo-Cruz, B., Guadalupe-Grau, A., Marín-Puyalto, J., Alfaro-Acha, A., Rodríguez-Mañas, L., García-García, F.J., & Ara, I. (2018). Reallocating Accelerometer-Assessed Sedentary Time to Light or Moderate-to Vigorous-Intensity Physical Activity Reduces Frailty Levels in Older Adults: An Isotemporal Substitution Approach in the TSHA Study. Journal of the American Medical Directors Association, 19(2), 185-190.

Manns, P. J., Dunstan, D. W., Owen, N., & Healy, G. N. (2012). Addressing the nonexercise part of the activity continuum: a more realistic and achievable

258

approach to activity programming for adults with mobility disability?. Physical therapy, 92(4), 614-625.

Marsden, E., & Torgerson, C. J. (2012). Single group, pre-and post-test research designs: Some methodological concerns. Oxford Review of Education, 38(5), 583-616.

Marshall, S. J., Levy, S. S., Tudor-Locke, C. E., Kolkhorst, F. W., Wooten, K. M., Ji, M., ... & Ainsworth, B. E. (2009). Translating physical activity recommendations into a pedometer-based step goal: 3000 steps in 30 minutes. American journal of preventive medicine, 36(5), 410-415.

Martin, A., Fitzsimons, C., Jepson, R., Saunders, D.H., van der Ploeg, H.P., Teixeira, P.J., Gray, C.M., & Mutrie, N. (2015). Interventions with potential to reduce sedentary time in adults: systematic review and meta-analysis. British Journal of Sports Medicine, 49(16), 1056-63.

Martinez Gomez, D., Bandinelli, S., Del Panta, V., Patel, K. V., Guralnik, J. M., &‐ ‐ Ferrucci, L. (2017). Three Year Changes in Physical Activity and Decline in‐ Physical Performance Over 9 Years of Follow Up in Older Adults: The‐ Invecchiare in Chianti Study. Journal of the American Geriatrics Society, 65(6), 1176-1182.

Matchar, D,B., & Orlando, L,A. (2007). The Relationship Between test and Outcome. In: Price CP, editor. Evidence-Based Laboratory Medicine; Principles, Practice and Outcomes, 2nd Edition. Washington DC, USA: AACC Press, 53-66.

Mathie, M.J., Coster, A.C., Lovell, N.H., & Celler, B.G. (2004). Accelerometry: providing an integrated, practical method for long-term, ambulatory monitoring of human movement. Physiological measurement, 25(2), R1-R20.

Matthew, C.E. (2005). Calibration of accelerometer output for adults. Medicine & Science in Sports & Exercise, 37(11 Suppl), S512-522.

Mc Sharry, J., Olander, E. K., & French, D. P. (2015). Do single and multiple behavior change interventions contain different behavior change techniques? A comparison of interventions targeting physical activity in obese populations. Health Psychology, 34(9), 960-977.

McCluskey, A., & Lovarini, M. (2005). Providing education on evidence-based practice improved knowledge but did not change behaviour: a before and after study. BMC medical education, 5(1), 40-51.

McDonald, B., Haardoerfer, R., Windle, M., Goodman, M., & Berg, C. (2017). Implications of Attrition in a Longitudinal Web-Based Survey: An Examination of College Students Participating in a Tobacco Use Study. JMIR public health and surveillance, 3(4), e73-e99.

McHugh, M. L. (2012). Interrater reliability: the kappa statistic. Biochemia medica: Biochemia medica, 22(3), 276-282.

259

McKay, H.A, Sims-Gould, J., Nettlefold, L., Hoy, C.L., & Bauman, A.E. (2017). Implementing and Evaluating an Older Adult Physical Activity Model at Scale: Framework for Action. Translational Journal of the American College of Sports Medicine, 2(2), 10-19.

McLeroy, K.R., Bibeau, D., Steckler, A., & Glanz, K. (1988). An ecological perspective on health promotion programs. Health education quarterly, 15(4), 351-377.

McMahon, S.K., Lewis, B., Oakes, J.M., Wyman, J.F., Guan, W., & Rothman, A.J. (2017). Assessing the effects of interpersonal and intrapersonal behavior change strategies on physical activity in older adults: a factorial experiment. Annals of Behavioral Medicine, 51(3), 376-390.

McNeill, L., Wyrwich, K. W., Brownson, R. C., Clark, E. M., & Kreuter, M. W. (2006). Individual, social environmental, and physical environmental influences on physical activity among black and white adults: a structural equation analysis. Annals of Behavioral Medicine, 31(1), 36-44.

McPhee, J.S., French, D.P., Jackson, D., Nazroo, J., Pendleton, N., & Degens, H. (2016). Physical activity in older age: perspectives for healthy ageing and frailty. Biogerontology, 17(3), 567-580.

Menai, M., Van Hees, V.T., Elbaz, A., Kivimaki, M., Singh-Manoux, A., & Sabia, S. (2017). Accelerometer assessed moderate-to-vigorous physical activity and successful ageing: results from the Whitehall II study. Scientific reports, 7, 45772-45780.

Mendoza-Vasconez, A.S., Linke, S., Muñoz, M., Pekmezi, D., Ainsworth, C., Cano, M., Williams, V., Marcus, B.H., & Larsen, B.A. (2016). Promoting physical activity among underserved populations. Current sports medicine reports, 15(4), 290-305.

Meneguci, J., Sasaki, J.E., Santos, A., Scatena, L.M., & Damião, R. (2015). Sitting Time and Quality of Life in Older Adults: A Population-Based Study. Journal of Physical Activity and Health, 12(11), 1513-1519.

Meng, Q., Xie, Z., & Zhang, T. (2014). A single-item self-rated health measure correlates with objective health status in the elderly: a survey in suburban Beijing. Frontiers in public health, 2, 27-35.

Merriam, S. B. (1998). Qualitative Research and Case Study Applications in Education. Revised and Expanded from" Case Study Research in Education.". Jossey-Bass Publishers, 350 Sansome St, San Francisco, CA 94104.

Metz, C.E. (1978) Basic principles of ROC analysis. In Seminars in nuclear medicine (Vol. 8, No. 4, 283-298). Elsevier.

Meyer, O. L., Castro-Schilo, L., & Aguilar-Gaxiola, S. (2014). Determinants of mental health and self-rated health: a model of socioeconomic status, neighborhood

260

safety, and physical activity. American journal of public health, 104(9), 1734-1741.

Michie, S., Pilling, S., Garety, P., Whitty, P., Eccles, M. P., Johnston, M., & Simmons, J. (2007). Difficulties implementing a mental health guideline: an exploratory investigation using psychological theory. Implementation Science, 2(1), 8-15.

Middelweerd, A., Mollee, J.S., van der Wal, C.N., Brug, J., & te Velde, S.J. (2014). Apps to promote physical activity among adults: a review and content analysis. International journal of behavioral nutrition and physical activity, 11(1), 97-106.

Middleton, L. E., & Yaffe, K. (2009). Promising strategies for the prevention of dementia. Archives of neurology, 66(10), 1210-1215.

Milat, A.J., Newson, R., King, L., Rissel, C., Wolfenden, L., Bauman, A., Redman, S., & Giffin, M. (2016). A guide to scaling up population health interventions. Public Health & Research Practice, 26(1), e2611604.

Miller, N.E., Strath, S.J., Swartz, A.M., & Cashin, S.E. (2010). Estimating absolute and relative physical activity intensity across age via accelerometry in adults. Journal of aging and physical activity, 18(2), 158-170.

Miller, W. R., & Rollnick, S. (2014). The effectiveness and ineffectiveness of complex behavioral interventions: impact of treatment fidelity. Contemporary Clinical Trials, 37(2), 234–41.

Milligan, C., Payne, S., Bingley, A., & Cockshott, Z. (2015). Place and Wellbeing: Shedding Light on Activity Interventions for Older Men. Ageing and Society, 35(01), 124-149.

Milton, K., Bull, F. C., & Bauman, A. (2010). Reliability and validity testing of a single-item PA measure. British Journal of Sports Medicine. 1-7.

Ministry of Health. (2013). Guidelines on Physical Activity for Older People (Aged 65 Years and Over). Retrieved from https://www.health.govt.nz/system/files/documents/publications/guidelines-on-physical-activity-older-people-jan13-v3.pdf.

Molinuevo, J. L., Valls-Pedret, C., & Rami, L. (2010). From mild cognitive impairment to prodromal Alzheimer disease: A nosological evolution. European Geriatric Medicine, 1(3), 146-154.

Monahan, T., & Fisher, J. A. (2010). Benefits of ‘observer effects’: lessons from the field. Qualitative research, 10(3), 357-376.

Moncher, F. J., & Prinz, R. J. (1991). Treatment fidelity in outcome studies. Clinical Psychology Review, 11, 247-266.

261

https://www.health.govt.nz/system/files/documents/publications/guidelines-on-physical-activity-older-people-jan13-v3.pdf

https://www.health.govt.nz/system/files/documents/publications/guidelines-on-physical-activity-older-people-jan13-v3.pdf

Moore, G.F., Audrey, S., Barker, M., Bond, L., Bonell, C., Hardeman, W., Moore, L., O’Cathain, A., Tinati, T., Wight, D., & Baird, J. (2015). Process evaluation of complex interventions: Medical Research Council guidance. bmj, 350, h1258.

Moran, F., MacMillan, F., Smith-Merry, J., Kilbreath, S., & Merom, D. (2015). Perceived Barriers, Facilitators and Patterns of Physical Activity of Older-old Adults Living in Assisted Retirement Accommodation. Journal of Gerontology & Geriatric Research, 4(6), 1-6.

Moreno, X., Albala, C., Lera, L., Sánchez, H., Fuentes-García, A., & Dangour, A. D. (2017). The role of gender in the association between self-rated health and mortality among older adults in Santiago, Chile: A cohort study. PloS one, 12(7), e0181317.

Mudrak, J., Stochl, J., Slepicka, P., & Elavsky, S. (2016). Physical activity, self-efficacy, and quality of life in older Czech adults. European Journal of Ageing, 13(1), 5–14.

Murphy, S.L. (2009). Review of physical activity measurement using accelerometers in older adults: considerations for research design and conduct. Preventive medicine, 48(2), 108-114.

Murtagh, E. M., Murphy, M. H., Murphy, N. M., Woods, C., Nevill, A. M., & Lane, A. (2015). Prevalence and correlates of physical inactivity in community-dwelling older adults in Ireland. PloS one, 10(2), e0118293.

National Institute for Health and Clinical Excellence. (2007): Behaviour Change. Quick reference guide. Retrieved from http://www.nice.org.uklnicemedia/pdf/PH006 quickrefguide.pdf.

National Institute of Health and Care Excellence. (2014). Behaviour change: individual approaches. Retrieved from https://www.nice.org.uk/guidance/ph49.

National Obesity Inventory. (2012). Standard Evaluation Framework for physical activity interventions. Retrieved from http://webarchive.nationalarchives.gov.uk/20170110173326/http://www.noo.org.uk/uploadsdoc/vid_16722_SEF_PA.pdf.

Neidrick, T. J., Fick, D. M., & Loeb, S. J. (2012). Physical activity promotion in primary care targeting the older adult. Journal of the American Academy of Nurse Practitioners, 24(7), 405-416.

Nero, H., Wallén, M.B., Franzén, E., Ståhle, A., & Hagströmer, M. (2015). Accelerometer Cutpoints for Physical Activity Assessment of Older Adults with Parkinson’s Disease. PloS one, 10(9), 1-11.

Nes, A. A. G., van Dulmen, S., Brembo, E. A., & Eide, H. (2018). An mHealth Intervention for Persons with Diabetes Type 2 Based on Acceptance and

262

https://www.nice.org.uk/guidance/ph49

Commitment Therapy Principles: Examining Treatment Fidelity. JMIR mHealth and uHealth, 6(7), e151.

Newitt, R., Barnett, F., & Crowe, M. (2016). Understanding factors that influence participation in physical activity among people with a neuromusculoskeletal condition: A review of qualitative studies. Disability and rehabilitation, 38(1), 1-10.

Ng, J. Y., Ntoumanis, N., & Thøgersen Ntoumani, C. (2014). Autonomy support and‐ control in weight management: What important others do and say matters. British Journal of Health Psychology, 19(3), 540-552.

NHS. (2015). Physical activity guidelines for older adults. Retrieved from http://www.nhs.uk/Livewell/fitness/Pages/physical-activity-guidelines-for-older-adults.aspx.

Nicholson, N.R. (2012). A review of social isolation: an important but underassessed condition in older adults. The journal of primary prevention, 33(2-3), 137-152.

Nigg, C. R., & Long, C. R. (2012). A systematic review of single health behavior change interventions vs. multiple health behavior change interventions among older adults. Translational behavioral medicine, 2(2), 163-179.

Noonan, R. J., Boddy, L. M., Fairclough, S. J., & Knowles, Z. R. (2016a). Parental perceptions on childrens out-of-school physical activity and family-based physical activity. Early Child Development and Care, 187(12), 1-16.

Noonan, R.J., Boddy, L.M., Fairclough, S.J., & Knowles, Z.R. (2016b). Write, draw, show, and tell: a child-centred dual methodology to explore perceptions of out-of-school physical activity. BMC public health, 16(1), 326-344.

Notthoff, N., Klomp, P., Doerwald, F., & Scheibe, S. (2016). Positive Messages Enhance Older Adults’ Motivation and Recognition Memory for Physical Activity Programmes. European Journal of Ageing, 13(3), 1-7.

Oakley, A., Strange, V., Bonell, C., Allen, E., & Stephenson, J. (2006). Process evaluation in randomised controlled trials of complex interventions. Bmj, 332(7538), 413-416.

Oestergaard, A. S., Mathiesen, M. H., Karlsen, A., Turtumoeygaard, I. F., Vahlgren, J., Kjaer, M., & Beyer, N. (2018). In acutely admitted geriatric patients, offering increased physical activity during hospitalization decreases length of stay and can improve mobility. Translational Sports Medicine, 1(1), 46-53.

Office for National Statistics. (2017). Overview of the UK Population: July 2017: An overview of the UK population, how it’s changed, what has caused it to change and how it is projected to change in the future. The UK population is also compared with other European countries. Retrieved.from.https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/articles/

263

http://www.nhs.uk/Livewell/fitness/Pages/physical-activity-guidelines-for-older-adults.aspx

http://www.nhs.uk/Livewell/fitness/Pages/physical-activity-guidelines-for-older-adults.aspx

overviewoftheukpopulation/july2017#the-uks-population-is-getting-older-but-its-not-the-same-in-all-areas-of-the-uk.

Office for National Statistics. (2017). Population Estimates for UK, England and Wales, Scotland and Northern Ireland: time-series. Retrieved from https://beta.ons.gov.uk/filter-outputs/936835f1-2514-4d87-a981-02d8fc804383.

Office of Disease Prevention and Health Promotion. (2018). PART A. EXECUTIVE SUMMARY. Retrieved.from.https://health.gov/paguidelines/secondedition/report/pdf/02_A_Executive_Summary.pdf.

Oguma, Y., Osawa, Y., Takayama, M., Abe, Y., Tanaka, S., Lee, I.M., & Arai, Y. (2017). Validation of questionnaire-assessed physical activity in comparison with objective measures using accelerometers and physical performance measures among community-dwelling adults aged≥ 85 years in Tokyo, Japan. Journal of Physical Activity and Health, 14(4), 245-252.

Olanrewaju, O., Kelly, S., Cowan, A., Brayne, C. and Lafortune, L., 2016. Physical activity in community-dwelling older people: a systematic review of reviews of interventions and context. PLoS One, 11(12), p.e0168614.

Olson, E.A., Mullen, S.P., Raine, L.B., Kramer, A.F., Hillman, C.H., & McAuley, E. (2016). Integrated Social-and Neurocognitive Model of Physical Activity Behavior in Older Adults with Metabolic Disease. Annals of Behavioral Medicine, 85, 78-83.

Orr, N., & Phoenix, C. (2015). Photographing physical activity: using visual methods to ‘grasp at’the sensual experiences of the ageing body. Qualitative Research, 15(4), 454-472.

Overdorf, V., Coker, C., & Kollia, B. (2016). Perceived Competence and Physical Activity in Older Adults. Activities, Adaptation & Aging, 40(4), 285-295.

Owen, N., Healy, G.N., Matthews, C.E., Dunstan, D.W. (2010). Too much sitting: the population-health science of sedentary behavior. Exercise and sport sciences reviews, 38(3), 105-113.

Parsons, T.J., Sartini, C., Ash, S., Lennon, L.T., Wannamethee, S.G., Lee, I.M., Whincup, P.H., & Jefferis, B.J. (2017). Objectively measured physical activity and kidney function in older men; a cross-sectional population-based study. Age and Ageing, 1-5.

Pate, R.R., O'neill, J.R., & Lobelo, F. (2008). The evolving definition of" sedentary". Exercise and sport sciences reviews, 36(4), 173-178.

Peach, D., Van Hoomissen, J., & Callender, H. L. (2014). Exploring the ActiLife((R)) filtration algorithm: converting raw acceleration data to counts. Physiol Meas, 35(12), 2359-2367.

264

Pérez, D., Van der Stuyft, P., del Carmen Zabala, M., Castro, M., & Lefèvre, P. (2015). A modified theoretical framework to assess implementation fidelity of adaptive public health interventions. Implementation science, 11(1), 91-101.

Perkins, N. J., & Schisterman, E. F. (2006). The inconsistency of "optimal" cutpoints obtained two criteria based on the receiver operating characteristic curve. American Journal of Epidemiology, 163(7), 670-675.

Peterson, M.J., Giuliani, C., Morey, M.C., Pieper, C.F., Evenson, K.R., Mercer, V., Cohen, H.J., Visser, M., Brach, J.S., Kritchevsky, S.B., & Goodpaster, B.H. (2009). Physical activity as a preventative factor for frailty: the health, aging, and body composition study. The Journals of Gerontology: Series A, 64(1), 61-68.

Petrescu-Prahova, M., Belza, B., Kohn, M., & Miyawaki, C. (2015). Implementation and Maintenance of a Community-Based Older Adult Physical Activity Program. The Gerontologist, 56(4), 1-10.

Phillips, S.M., Awick, E.A., Conroy, D.E., Pellegrini, C.A., Mailey, E.L., & McAuley, E. (2015). Objectively measured physical activity and sedentary behavior and quality of life indicators in survivors of breast cancer. Cancer, 121(22), 4044-4052.

Phoenix, C., & Tulle, E. (2017). Physical activity and ageing. IN: J. Piggin, L. Mansfield, and M. Weed, eds. The Routledge handbook of physical activity policy and practice. London: Routledge.

Plassman, B. L., Williams, J. W., Burke, J. R., Holsinger, T., & Benjamin, S. (2010). Systematic review: factors associated with risk for and possible prevention of cognitive decline in later life. Annals of internal medicine, 153(3), 182-193.

Plotnikoff, R. C., Lubans, D. R., Penfold, C. M., & Courneya, K. S. (2014). Testing the utility of three social cognitive models for predicting objective and self report‐ ‐ physical activity in adults with type 2 diabetes. British journal of health psychology, 19(2), 329-346.

Podsiadlo, D., & Richardson, S. (1991). The timed “Up & Go”: a test of basic functional mobility for frail elderly persons. Journal of the American geriatrics Society, 39(2), 142-148.

Potter, R., Ellard, D., Rees, K., & Thorogood, M. (2011). A systematic review of the effects of PA on physical functioning, quality of life and depression in older people with dementia. International journal of geriatric psychiatry, 26, 1000-1011.

Prince, S.A., Adamo, K.B., Hamel, M.E., Hardt, J., Gorber, S.C., & Tremblay, M. (2008). A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. International Journal of Behavioral Nutrition and Physical Activity, 5(1), 56-79.

265

Prince, S. A., Saunders, T. J., Gresty, K., & Reid, R. D. (2014). A comparison of the effectiveness of physical activity and sedentary behaviour interventions in reducing sedentary time in adults: A systematic review and meta-analysis of controlled trials. Obesity Reviews, 15(11), 905-919.

Prins, R.G., & van Lenthe, F.J. (2015). The Hour-to-Hour Influence of Weather Conditions on Walking and Cycling among Dutch Older Adults. Age & Ageing, 44(5), 1-5.

Public Health England. (2017). Health Report 2017: Sefton. Retrieved from http://fingertipsreports.phe.org.uk/health-profiles/2017/e08000014.pdf.

Pulsford, R. M., Stamatakis, E., Britton, A. R., Brunner, E. J., & Hillsdon, M. (2015). Associations of sitting behaviours with allcause mortality over a 16-year follow-up: The Whitehall II study. International Journal of Epidemiology, 44(6), 1909-1916. doi:10.1093/ije/dyv191.

QSR International Pty Ltd. (2017). NVIVO: Version 11. Reference guide. Doncaster Victoria, Australia: Author.

Quested, E., Ntoumanis, N., Thøgersen-Ntoumani, C., Hagger, M. S., & Hancox, J. E. (2017). Evaluating quality of implementation in physical activity interventions based on theories of motivation: current challenges and future directions. International Review of Sport and Exercise Psychology, 10(1), 252-269.

Quijano, L. M., Stanley, M. A., Petersen, N. J., Casado, B. L., Steinberg, E. H., Cully, J. A., & Wilson, N. L. (2007). Healthy IDEAS: A depression intervention delivered by community-based case managers serving older adults. Journal of Applied Gerontology, 26(2), 139-156.

Rabesh, J., Charlton, C., Browne, W.J., Healy, M., & Cameron, B. (2009). MLwiN Version 2.10.; Centre for Multilevel Modelling, University of Bristol: Bristol, UK.

Ramires, V.V., Wehrmeister, F.C., Böhm, A.W., Galliano, L., Ekelund, U., Brage, S., & da Silva, I.C.M. (2017). Physical activity levels objectively measured among older adults: a population-based study in a Southern city of Brazil. International Journal of Behavioral Nutrition and Physical Activity, 14(1), 13-21.

Ramsay, S. E., Morris, R. W., Whincup, P. H., Subramanian, S. V., Papacosta, A. O., Lennon, L. T., & Wannamethee, S. G. (2015). The influence of neighbourhood-level socioeconomic deprivation on cardiovascular disease mortality in older age: longitudinal multilevel analyses from a cohort of older British men. Journal of epidemiology and community health, 69(12), 1224-1231.

Reid, H., & Foster, C. (2016). Infographic. Physical Activity Benefits for Adults and Older Adults. British Journal of Sports Medicine, 51(19), 1441-1442.

266

Reis, R. S., Salvo, D., Ogilvie, D., Lambert, E. V., Goenka, S., Brownson, R. C., & Lancet Physical Activity Series 2 Executive Committee. (2016). Scaling up physical activity interventions worldwide: stepping up to larger and smarter approaches to get people moving. The Lancet, 388(10051), 1337-1348.

Rejeski, W.J., Axtell, R., Fielding, R., Katula, J., King, A.C., Manini, T.M., Marsh, A.P., Pahor, M., Rego, A., Tudor-Locke, C., & Newman, M. (2013). Promoting Physical Activity for Elders with Compromised Function: The Lifestyle Interventions and Independence for Elders (LIFE) Study Physical Activity Intervention. Clinical Interventions in Aging 8, 1119-1131.

Resnick, B., & Jenkins, L. S. (2000). Testing the reliability and validity of the self-efficacy for exercise scale. Nursing research, 49(3), 154-159.

Resnick, B., Luisi, D., Vogel, A., & Junaleepa, P. (2004). Reliability and validity of the self-efficacy for exercise and outcome expectations for exercise scales with minority older adults. Journal of Nursing Measurement, 12(3), 235-248.

Rhodes, R. E., Mark, R. S., & Temmel, C. P. (2012). Adult sedentary behavior: a systematic review. American journal of preventive medicine, 42(3), e3-e28.

Richards, J., Thorogood, M., Hillsdon, M., & Foster, C. (2013). Face-to-face versus remote and web 2.0 interventions for promoting physical activity. Cochrane Database of Systematic Reviews, 9(9), 1-11.

Roe, B., Beech, R., Harris, M., Beech, B., Russell, W., Gent, D., ... & Dickinson, A. (2011). Improving quality of life for older people in the community: findings from a local Partnerships for Older People Project innovation and evaluation. Primary health care research & development, 12(03), 200-213.

Rose, D. J. (2018). Physical Activity Instruction of Older Adults, 2E. Human Kinetics.

Rosenberg, D. E., Bellettiere, J., Gardiner, P. A., Villarreal, V. N., Crist, K., & Kerr, J. (2015). Independent associations between sedentary behaviors and mental, cognitive, physical, and functional health among older adults in retirement communities. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences, 71(1), 78-83.

Rosenberger, M.E., Haskell, W.L., Albinali, F., Mota, S., Nawyn, J., & Intille, S. (2013). Estimating activity and sedentary behavior from an accelerometer on the hip or wrist. Medicine and science in sports and exercise, 45(5), 964-975.

Rowlands, A.V., Fraysse, F., … & Catt, M. (2015). Comparison of measured acceleration output from accelerometry-based activity monitors. Medicine and science in sports and exercise, 47, 201-210.

Rowlands, A.V., Mirkes, E.M., Yates, T., Clemes, S., Davies, M., Khunti, K., & Edwardson, C.L. (2017). Accelerometer-assessed Physical Activity in Epidemiology: Are Monitors Equivalent? Medicine and science in sports and exercise. doi:10.1249/MSS.0000000000001435.

267

Rowlands, A. V., Yates, T., Davies, M., Khunti, K., & Edwardson, C. L. (2016). Raw Accelerometer Data Analysis with GGIR R-package: Does Accelerometer Brand Matter? Medicine and science in sports and exercise 48, 1935–1941.

Sallis, J.F., Bull, F., Guthold, R., Heath, G.W., Inoue, S., Kelly, P., Oyeyemi, A.L., Perez, L.G., Richards, J., Hallal, P.C., & Lancet Physical Activity Series 2 Executive Committee. (2016). Progress in physical activity over the Olympic quadrennium. The Lancet, 388(10051), 1325-1336.

Sallis, J.F., Cervero, R.B., Ascher, W., Henderson, K.A., Kraft, M.K., & Kerr, J. (2006). An ecological approach to creating active living communities. Annu. Rev. Public Health, 27, 297-322.

Sallis, J. F., Owen, N., & Fisher, E. B. (2008). Ecological models of health behaviour. Health behaviour and health education: Theory, research, and practice, 4, 465-486.

Sallis, J. F., Owen, N., & Fotheringham, M. J. (2000). Behavioral epidemiology: a systematic framework to classify phases of research on health promotion and disease prevention. Annals of Behavioral Medicine, 22(4), 294-298.

Sanders, G. J., Roe, B., Knowles, Z. R., Kaehne, A., & Fairclough, S. (2018). Using formative research with older adults to inform a community physical activity programme: Get Healthy, Get Active. Journal of Primary Health Care Research & Development. 1-10. Doi: 10.1017/S1463423618000373

Santacroce, S. J., Maccarelli, L. M., & Grey, M. (2004). Intervention fidelity. Nursing Research, 53, 63-66.

Santos-Lozano, A., Torres-Luque, G., Marín, P.J., Ruiz, J.R., Lucia, A., & Garatachea, N. (2012). Intermonitor variability of GT3X accelerometer. International journal of sports medicine, 33(12), 994-999.

Sargent-Cox, K. A., Anstey, K. J., & Luszcz, M. A. (2010). The choice of self-rated health measures matter when predicting mortality: evidence from 10 years follow-up of the Australian longitudinal study of ageing. BMC geriatrics, 10(1), 18-29.

Sartini, C., Wannamethee, S. G., Iliffe, S., Morris, R. W., Ash, S., Lennon, L., ... & Jefferis, B. J. (2015). Diurnal patterns of objectively measured physical activity and sedentary behaviour in older men. BMC Public Health, 15(1), 609-621.

Schaefer, C. A., Nigg, C. R., Hill, J. O., Brink, L. A., & Browning, R. C. (2014). Establishing and evaluating wrist cutpoints for the GENEActiv accelerometer in youth. Medicine and Science in Sports and Exercise, 46(4), 826-833.

Schneider, J.L., Goddard, K.A., Davis, J., Wilfond, B., Kauffman, T.L., Reiss, J.A., Gilmore, M., Himes, P., Lynch, F.L., Leo, M.C., & McMullen, C. (2016). “Is It Worth Knowing?” Focus Group Participants’ Perceived Utility of Genomic

268

Preconception Carrier Screening. Journal of Genetic Counseling, 25(1), 135-145.

Schoenwald, S. K., Garland, A. F., Chapman, J. E., Frazier, S. L., Sheidow, A. J., & Southam-Gerow, M. A. (2011). Toward the effective and efficient measurement of implementation fidelity. Administration and Policy in Mental Health and Mental Health Services Research, 38(1), 32-43.

Schuler, P. B., & Marzilli, T. S. (2003). Use of self-reports of physical fitness as substitutes for performance-based measures of physical fitness in older adults. Perceptual and motor skills, 96(2), 414-420.

Schwartz, D., & Lellouch, J. (1967). Explanatory and pragmatic attitudes in therapeutical trials. Journal of Clinical Epidemiology, 20(8), 637-648.

Scully, T. (2012). Demography: To the limit. Nature, 492(7427), S2–3.

Sergi, G., Coin, A., Sarti, S., Perissinotto, E., Peloso, M., Mulone, S., Trolese, M., Inelmen, E.M., Enzi, G., & Manzato, E. (2010). Resting VO2, maximal VO2 and metabolic equivalents in free-living healthy elderly women. Clinical nutrition, 29(1), 84-88.

Shaltout, H.A., Eggebeen, J., Marsh, A.P., Brubaker, P.H., Laurienti, P.J., Burdette, J.H., Basu, S., Morgan, A., Dos Santos, P.C., Norris, J.L., & Morgan, T.M. (2017). Effects of supervised exercise and dietary nitrate in older adults with controlled hypertension and/or heart failure with preserved ejection fraction. Nitric Oxide, 69, 78-90.

Shaw, R.J., Čukić, I., Deary, I.J., Gale, C.R., Chastin, S.F., Dall, P.M., Skelton, D.A., & Der, G. (2017). Relationships between socioeconomic position and objectively measured sedentary behaviour in older adults in three prospective cohorts. BMJ Open, 7(6), 1-11.

Shephard, R. (2011). Compendium of Physical Activities: A Second Update of Codes and MET Values. Yearbook Of Sports Medicine, 2012, 126-127.

Shephard, R.J., & Tudor-Locke, C. (2016). The objective monitoring of physical activity: contributions of Accelerometry to epidemiology, exercise science and rehabilitation. Springer International Publishing. http://www.springer.com/us/book/9783319295756.

Shiroma, E.J., Schrack, A., Harris, T.B. (2018). Accelerating Accelerometer Research in Aging. The Journals of Gerontology, 73(5), 619–621.

Siebens, H.C., Tsukerman, D., Adkins, R.H., Kahan, J., & Kemp, B. (2015). Correlates of a Single-Item Quality-of-Life Measure in People Aging with Disabilities. American journal of physical medicine & rehabilitation, 94(12), 1065-1074.

269

Siedlecki, K.L., Salthouse, T.A., Oishi, S., & Jeswani, S. (2014). The relationship between social support and subjective well-being across age. Social indicators research, 117(2), 561-576.

Siervo, M., Bertoli, S., Battezzati, A., Wells, J.C., Lara, J., Ferraris, C. & Tagliabue, A. (2014). Accuracy of predictive equations for the measurement of resting energy expenditure in older subjects. Clinical nutrition, 33(4), 613-619.

Simmonds, B., Fox, K., Davis, M., Ku, P.W., Gray, S., Hillsdon, M., Sharp, D., Stathi, A., Thompson, J., Coulson, J., & Trayers, T. (2014). Objectively assessed physical activity and subsequent health service use of UK adults aged 70 and over: a four to five year follow up study. PloS one, 9(5), e97676.

Simplican, S.C., Leader, G., Kosciulek, J., & Leahy, M. (2015). Defining social inclusion of people with intellectual and developmental disabilities: An ecological model of social networks and community participation. Research in developmental disabilities, 38, 18-29.

Sineath, A., Lambert, L., Verga, C., Wagstaff, M., & Wingo, B. C. (2017). Monitoring intervention fidelity of a lifestyle behavioral intervention delivered through telehealth. Mhealth, 3, 35-46.

Sink, K.M., Espeland, M.A., Castro, C.M., Church, T., Cohen, R., Dodson, J.A., Guralnik, J., Hendrie, H.C., Jennings, J., Katula, J., & Lopez, O.L. (2015). Effect of a 24-month physical activity intervention vs health education on cognitive outcomes in sedentary older adults: the LIFE randomized trial. Jama, 314(8), 781-790.

Skender, S., Ose, J., Chang-Claude, J., Paskow, M., Brühmann, B., Siegel, E.M., Steindorf, K., & Ulrich, C.M. (2016). Accelerometry and physical activity questionnaires-a systematic review. BMC public health, 16(1), 515-524.

Skjæret, N., Nawaz, A., Morat, T., Schoene, D., Helbostad, J. L., & Vereijken, B. (2016). Exercise and Rehabilitation Delivered through Exergames in Older Adults: An Integrative Review of Technologies, Safety and Efficacy. International Journal of Medical Informatics, 85(1), 1-16.

Smith, G.L., Banting, L., Eime, R., O’Sullivan, G., & van Uffelen, J.G. (2017). The association between social support and physical activity in older adults: a systematic review. International Journal of Behavioral Nutrition and Physical Activity, 14(1), 56-76.

Smith, B., & Caddick, N. (2012). Qualitative Methods in Sport: A Concise Overview for Guiding Social Scientific Sport Research. Asia Pacific Journal of Sport and Social Science, 1(1), 60-73.

Smith, L., Gardner, B., Fisher, A., & Hamer, M. (2015). Patterns and correlates of physical activity behaviour over 10 years in older adults: prospective analyses from the English Longitudinal Study of Ageing. BMJ open, 5(4), e007423.

270

Sofi, F., Valecchi, D., Bacci, D., Abbate, R., Gensini, G. F., Casini, A., & Macchi, C. (2011). Physical activity and risk of cognitive decline: a meta analysis of‐ prospective studies. Journal of internal medicine, 269(1), 107-117.

Song, J., Lindquist, L.A., Chang, R.W., Semanik, P.A., Ehrlich-Jones, L.S., Lee, J., Sohn, M.W., & Dunlop, D.D. (2015). Sedentary behavior as a risk factor for physical frailty independent of moderate activity: results from the osteoarthritis initiative. American journal of public health, 105(7), 1439-1445.

Sparling, P.B., Howard, B.J., Dunstan, D.W., & Owen, N. (2015). Recommendations for physical activity in older adults. BMJ: British Medical Journal (Online), 350, 1-5.

Sport England. (2012). GET HEALTHY, GET ACTIVE: Learn more about our initial investments into tackling inactivity from 2012-2016. Retrieved from https://www.sportengland.org/our-work/health-and-inactivity/get-healthy-get-active/.

Sport England. (2014). Active People Survey 8: Active People Interactive. Retrieved from http://activepeople.sportengland.org/.

Sport England. (2015). Local Profile Tool. Retrieved from http://www.sportengland.org/our-work/local-work/local-government/local-sport-profile/.

Stewart, D. W., & Shamdasani, P. N. (2014). Focus Groups: Theory and Practice (Vol. 20). Newbury Park: Sage publications.

Stiles, V. H., Griew, P. J., & Rowlands, A. V. (2013). Use of accelerometry to classify activity beneficial to bone in premenopausal women. Medicine and science in sports and exercise, 45(12), 2353-2361.

Stokols, D. (1992). Establishing and maintaining healthy environments: toward a social ecology of health promotion. American Psychologist, 47(1), 6-22.

Strain, T., Fitzsimons, C., Foster, C., Mutrie, N., Townsend, N., & Kelly, P. (2016). Age-related comparisons by sex in the domains of aerobic physical activity for adults in Scotland. Preventive medicine reports, 3, 90-97.

Sun, F., Norman, I.J., & While, A.E. (2013). Physical activity in older people: a systematic review. BMC public health, 13(1), 449-465.

Susan, J., Mallan, K., Callaway, L., Daniels, L.A., & Nicholson, J.M. (2017). A cross sectional comparison of predisposing, reinforcing and enabling factors for lifestyle health behaviours and weight gain in healthy and overweight pregnant women. Maternal and child health journal, 21(3), 626-635.

Szklo, M., & Nieto, F.J. (2007). Stratification and adjustment: multivariate analysis in epidemiology. Epidemiology Beyond the basics, 227-295.

271

Taylor, L.M., Klenk, J., Maney, A.J., Kerse, N., MacDonald, B.M., & Maddison, R. (2014). Validation of a body-worn accelerometer to measure activity patterns in octogenarians. Archives of physical medicine and rehabilitation, 95(5), 930-934.

Teixeira, P. J., Carraça, E. V., Markland, D., Silva, M. N., & Ryan, R. M. (2012). Exercise, physical activity, and self-determination theory: a systematic review. International journal of behavioral nutrition and physical activity, 9(1), 78-107.

ten Brinke, L.F., Bolandzadeh, N., Nagamatsu, L.S., Hsu, C.L., Davis, J.C., Miran-Khan, K., & Liu-Ambrose, T. (2015). Aerobic exercise increases hippocampal volume in older women with probable mild cognitive impairment: a 6-month randomised controlled trial. British Journal of Sports Medicine, 49(4), 248-54.

Tennstedt, S., Howland, J., Lachman, M., Peterson, E., Kasten, L., & Jette, A. (1998). A randomized, controlled trial of a group intervention to reduce fear of falling and associated activity restriction in older adults. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 53(6), P384-P392.

Teychenne, M., Abbott, G., Ball, K., & Salmon, J. (2014). Prospective associations between sedentary behaviour and risk of depression in socio-economically disadvantaged women. Preventive Medicine, 65, 82–86.

Teychenne, M., Costigan, S. A., & Parker, K. (2015). The association between sedentary behaviour and risk of anxiety: a systematic review. BMC Public Health, 15(1), 513-520.

Thøgersen-Ntoumani, C., Cumming, J., Ntoumanis, N., & Nikitaras, N. (2012). Exercise imagery and its correlates in older adults. Psychology of Sport and Exercise, 13(1), 19-25.

Thomas, D. R. (2006). A general inductive approach for analyzing qualitative evaluation data. American journal of evaluation, 27(2), 237-246.

Thomas, J. R., Silverman, S., & Nelson, J. (2015). Research Methods in Physical Activity - 7th Edition. Champaign, IL: Human Kinetics.

Thompson, T.P., Lambert, J.D., Greaves, C.J., & Taylor, A.H. (2018). Intervention delivery fidelity assessment of a counseling-based intervention for promoting smoking reduction and increasing physical activity. Health Psychology, 37(7), 627-668.

Thornton, C.M., Kerr, J., Conway, T.L., Saelens, B.E., Sallis, J.F., Ahn, D.K., Frank, L.D., Cain, K.L., & King, A.C. (2017). Physical activity in older adults: An ecological approach. Annals of Behavioral Medicine, 51(2), 159-169.

Thorp, A.A., Owen, N., Neuhaus, M., & Dunstan, D.W. (2011). Sedentary behaviors and subsequent health outcomes in adults a systematic review of longitudinal studies, 1996-2011. American Journal of Preventive Medicine, 41(2), 207–215.

272

Thraen-Borowski, K. M., Trentham-Dietz, A., Edwards, D. F., Koltyn, K. F., & Colbert, L. H. (2013). Dose–response relationships between physical activity, social participation, and health-related quality of life in colorectal cancer survivors. Journal of Cancer Survivorship, 7(3), 369–378.

Tomioka, K., Iwamoto, J., Saeki, K., & Okamoto, N. (2011). Reliability and validity of the International Physical Activity Questionnaire (IPAQ) in elderly adults: the Fujiwara-kyo Study. Journal of Epidemiology, 21(6), 459-465.

Träff, A.M., Cedersund, E., & Nord, C. (2017). Perceptions of physical activity among elderly residents and professionals in assisted living facilities. European Review of Aging and Physical Activity, 14(1), 2-11.

Tremblay, M. S., Aubert, S., Barnes, J. D., Saunders, T. J., Carson, V., Latimer-Cheung, A. E., … & Chinapaw, M. J. M. (2017). Sedentary Behavior Research Network (SBRN) - Terminology Consensus Project process and outcome. International Journal of Behavioral Nutrition and Physical Activity, 14(1), 75-92.

Treuth, M.S., Schmitz, K., Catellier, D.J., McMurray, R.G., Murray, D.M., Almeida, M.J., Going, S., Norman, J.E. & Pate, R. (2004). Defining accelerometer thresholds for activity intensities in adolescent girls. Medicine and science in sports and exercise, 36(7), 1259-1266.

Troiano, R. P., Berrigan, D., Dodd, K. W., Mâsse, L. C., Tilert, T., & McDowell, M. (2008). Physical activity in the United States measured by accelerometer. Medicine and Science in Sports and Exercise, 40(1), 181–8.

Troiano, R. P., McClain, J. J., Brychta, R. J., & Chen, K. Y. (2014). Evolution of accelerometer methods for physical activity research. British Journal of Sports Medicine, 48(13), 1019-1023.

Troiano, R.P., McClain, J.J., Brychta, R.J. and Chen, K.Y. (2014). Evolution of accelerometer methods for physical activity research. British Journal of Sports Medicine, 48(13), 1019–23.

Trost, S. G., Owen, N., Bauman, A. E., Sallis, J. F., & Brown, W. (2002). Correlates of adults' participation in physical activity: review and update. Medicine & Science in Sports & Exercise, 34, 1996-2001.

Tucker, J. M., Welk, G. J., & Beyler, N. K. (2011). Physical Activity in US Adults: Compliance with the Physical Activity Guidelines for Americans. American Journal of Preventive Medicine, 40(4), 454-461.

Turner-McGrievy, G. M., Beets, M. W., Moore, J. B., Kaczynski, A. T., Barr-Anderson, D. J., & Tate, D. F. (2013). Comparison of traditional versus mobile app self-monitoring of physical activity and dietary intake among overweight adults participating in an mHealth weight loss program. Journal of the American Medical Informatics Association, 20(3), 513-518.

273

Twisk, J.W.R. (2006). Applied Multilevel Analysis. Cambridge University Press: Cambridge, UK.

Twisk, J. W.R. (2013). Applied longitudinal data analysis for epidemiology: a practical guide. Cambridge University Press.

UK Office for National Statistics. (2017). Mid-2016 Population Estimates. Retrieved from https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/bulletins/annualmidyearpopulationestimates/latest.

United Nations. (2015). World Population Ageing Report. Retrieved from http://www.un.org/en/development/desa/population/publications/pdf/ageing/WPA2015_Report.pdf.

Vahlberg, B., Cederholm, T., Lindmark, B., Zetterberg, L., & Hellström, K. (2017). Short-term and long-term effects of a progressive resistance and balance exercise program in individuals with chronic stroke: a randomized controlled trial. Disability and rehabilitation, 39(16), 1615-1622.

Valenti, G., Bonomi, A.G., & Westerterp, K.R. (2017). Diurnal patterns of physical activity in relation to activity induced energy expenditure in older adults. Physical activity in older adults, IN PRESS, 83-101.

van Baal, P.H., Hoogendoorn, M., & Fischer, A. (2016). Preventing dementia by promoting physical activity and the long-term impact on health and social care expenditures. Preventive medicine, 85, 78-83.

Van Cauwenberg, J., Clarys, P., De Bourdeaudhuij, I., Ghekiere, A., de Geus, B., Owen, N., & Deforche, B. (2018). Environmental influences on older adults’ transportation cycling experiences: A study using bike-along interviews. Landscape and Urban Planning, 169, 37-46.

Van Cauwenberg, J., De Bourdeaudhuij, I., Clarys, P., Nasar, J., Salmon, J., Goubert, L., & Deforche, B. (2016). Street characteristics preferred for transportation walking among older adults: a choice-based conjoint analysis with manipulated photographs. International journal of behavioral nutrition and physical activity, 13(1), 6-22.

Van Cauwenberg, J., De Donder, L., Clarys, P., De Bourdeaudhuij, I., Owen, N., Dury, S., De Witte, N., Buffel, T., Verté, D., & Deforche, B. (2014). Relationships of individual, social, and physical environmental factors with older adults’ television viewing time. Journal of Aging and Physical Activity, 22(4), 508-517.

van Der Berg, J.D., Bosma, H., Caserotti, P., Eiriksdottir, G., Arnardottir, N.Y., Martin, K.R., Brychta, R.J., Chen, K.Y., Sveinsson, T., Johannsson, E., & Launer, L.J. (2014). Midlife determinants associated with sedentary behavior in old age. Medicine and science in sports and exercise, 46(7), p.1359-1365.

274

Van Dyck, D., Mertens, L., Cardon, G., De Cocker, K., & De Bourdeaudhuij, I. (2017). Opinions toward Physical Activity, Sedentary Behavior, and Interventions to Stimulate Active Living During Early Retirement: A Qualitative Study in Recently Retired Adults. Journal of aging and physical activity, 25(2), 277-286.

van Hees, V. T., Fang, Z., Langford, J., Assah, F., Mohammad, A., da Silva, I. C., . . . & Brage, S. (2014). Autocalibration of accelerometer data for free-living physical activity assessment using local gravity and temperature: an evaluation on four continents. Journal of Applied Physiology, 117(7), 738-744.

van Hees, V. T., Golubic, R., Ekelund, U., & Brage, S. (2013a). Impact of study design on development and evaluation of an activity-type classifier. Journal of Applied Physiology, 114(8), 1042-1051.

van Hees, V. T., Gorzelniak, L., Dean Leon, E. C., Eder, M., Pias, M., Taherian, S., . . . & Brage, S. (2013b). Separating movement and gravity components in an acceleration signal and implications for the assessment of human daily physical activity. PLoS One, 8(4), e61691.

van Hees, V. T., Renstrom, F., Wright, A., Gradmark, A., Catt, M., Chen, K. Y., . . . & Franks, P. W. (2011). Estimation of daily energy expenditure in pregnant and non-pregnant women using a wrist-worn tri-axial accelerometer. Plos One, 6, e22922.

van Schijndel-Speet, M., Evenhuis, H.M., van Wijck, R., van Empelen, P., & Echteld, M.A. (2014). Facilitators and barriers to physical activity as perceived by older adults with intellectual disability. Mental Retardation, 52(3), 175-186.

van Uffelen, J. G. Z., Heesch, K. C., Hill, R. L., & Brown, W. J. (2011). A qualitative study of older adults’ responses to sitting-time questions: do we get the information we want? BMC Public Health, 11, 458-468.

Van Uffelen, J. G., Heesch, K. C., van Gellecum, Y. R., Burton, N. W., & Brown, W. J. (2012). Which older women could benefit from interventions to decrease sitting time and increase physical activity?. Journal of the American Geriatrics Society, 60(2), 393-396.

Varela Mato, V., Yates, T., Stensel, D., Biddle, S., & Clemes, S.A. (2017). Concurrent Validity of Actigraph-Determined Sedentary Time Against the Activpal Under Free-Living Conditions in a Sample of Bus Drivers. Measurement in Physical Education and Exercise Science, 21(4), 212-222.

Vidovich, M.R., Lautenschlager, N.T., Flicker, L., Clare, L., McCaul, K., & Almeida, O.P. (2015). The PACE study: a randomized clinical trial of cognitive activity strategy training for older people with mild cognitive impairment. The American Journal of Geriatric Psychiatry, 23(4), 360-372.

275

Vijay, G. C., Wilson, E. C., Suhrcke, M., Hardeman, W., & Sutton, S. (2016). Are brief interventions to increase physical activity cost-effective? A systematic review. British Journal of Sports Medicine, 50(7), 408-417.

Walston, J., Hadley, E.C., Ferrucci, L., Guralnik, J.M., Newman, A.B., Studenski, S.A., Ershler, W.B., Harris, T., & Fried, L.P. (2006). Research agenda for frailty in older adults: toward a better understanding of physiology and etiology: summary from the American Geriatrics Society/National Institute on Aging Research Conference on Frailty in Older Adults. Journal of the American Geriatrics Society, 54(6), 991-1001.

Warburton, D. E. R., & Bredin, S. S. D. (2016). Reflections on Physical Activity and Health: What Should We Recommend? Canadian Journal of Cardiology. Pulsus Group Inc.

Warmoth, K., Lang, I.A., Phoenix, C., Abraham, C., Andrew, M.K., Hubbard, R.E., & Tarrant, M. (2016). ‘Thinking you're old and frail’: a qualitative study of frailty in older adults. Ageing & Society, 36(7), 1483-1500.

Warner, L.M., Wolff, J.K., Ziegelmann, J.P., Schwarzer, R. and Wurm, S., 2016. Revisiting self-regulatory techniques to promote physical activity in older adults: null-findings from a randomised controlled trial. Psychology & health, 31(10), pp.1145-1165.

Warner, L.M., Ziegelmann, J.P., Schüz, B., Wurm, S. and Schwarzer, R., 2011. Synergistic effect of social support and self-efficacy on physical exercise in older adults. Journal of Aging and Physical Activity, 19(3), pp.249-261.

Washburn, R. A., Smith, K. W., Jette, A. M., & Janney, C. A. (1993). The physical activity scale for the elderly (PASE): Development and evaluation. Journal of Clinical Epidemiology, 46(2), 153–162.

Washburn, R.A. (2000). Assessment of physical activity in older adults. Research Quarterly for Exercise and Sport, 71(sup2), 79-87.

Weening-Dijksterhuis, E., de Greef, M. H., Krijnen, W., & van der Schans, C. P. (2012). Self-reported physical fitness in frail older persons: reliability and validity of the self-assessment of physical fitness (SAPF) 1. Perceptual & Motor Skills, 115(3), 797-810.

Welk, G. (2002). Physical activity assessments for health-related research. Human Kinetics.

Welk, G. J. (2005). Principles of design and analyses for the calibration of accelerometry-based activity monitors. Medicine Science Sports Exercise, 37(11), S501-S511.

Welk, G. J., Going, S. B., Morrow, J. R., & Meredith, M. D. (2011). Development of new criterion-referenced fitness standards in the FITNESSGRAM® program:

276

rationale and conceptual overview. American journal of preventive medicine, 41(4), S63-S67.

Welk, G. J., Laurson, K. R., Eisenmann, J. C., & Cureton, K. J. (2011). Development of Youth Aerobic-Capacity Standards Using Receiver Operating Characteristic Curves. American Journal of Preventive Medicine, 41(4, Supplement 2), S111-S116.

Welk, G.J., McClain, J., & Ainsworth, B.E. (2012). Protocols for evaluating equivalency of accelerometry-based activity monitors. Medicine and science in sports and exercise, 44(1 Suppl 1), S39-49.

White, M.N., King, A.C., Sallis, J.F., Frank, L.D., Saelens, B.E., Conway, T.L., Cain, K.L.,& Kerr, J. (2016). Caregiving, Transport-Related, and Demographic Correlates of Sedentary Behavior in Older Adults: The Senior Neighborhood Quality of Life Study. Journal of aging and health, 28(5), 812-33.

White, D.A., Rothenberger, S.D., Hunt, L.A., & Goss, F.L. (2016). Comparison of Affect and Cardiorespiratory Training Responses Between Structured Gym Activities and Traditional Aerobic Exercise in Children. International journal of exercise science, 9(1), 16-24.

Wijndaele, K., Westgate, K., Stephens, S.K., Blair, S.N., Bull, F.C., Chastin, S.F., Dunstan, D.W., Ekelund, U., Esliger, D.W., Freedson, P.S., & Granat, M.H. (2015). Utilization and harmonization of adult accelerometry data: review and expert consensus. Medicine and science in sports and exercise, 47(10), 2129-2140.

Wilcox, R. R. (2010). Fundamentals of modern statistical methods: Substantially improving power and accuracy (2nd ed.). New York: Springer.

Wilmot, E.G., Edwardson, C.L., Achana, F.A., Davies, M.J., Gorely, T., Gray, L.J., Khunti, K., Yates, T., & Biddle, S.J. (2012). Sedentary time in adults and the association with diabetes, cardiovascular disease and death: Systematic review and meta-analysis. Diabetologia, 55, 2895-2905.

Windle, G., Hughes, D., Linck, P., Russell, I., & Woods, B. (2010). Is exercise effective in promoting mental well-being in older age? A systematic review. Aging and Mental Health, 14(6), 652–669.

Withall, J., Stathi, A., Davis, M., Coulson, J., Thompson, J. L., & Fox, K. R. (2014). Objective indicators of physical activity and sedentary time and associations with subjective well-being in adults aged 70 and over. International Journal of Environmental Research and Public Health, 11(1), 643–656.

Wolitzky Taylor, K., Castriotta, N., Lenze, E. J., Stanley, M. A., & Craske, M. G. (2010).‐ Anxiety disorders in older adults: a comprehensive review. Depression and Anxiety, 27(2), 190–211.

277

World Health Organisation. (2017). Global Strategy on Diet, Physical Activity and Health: Physical Activity and Older Adults. Retrieved from http://www.who.int/dietphysicalactivity/factsheet_olderadults/en/.

World Health Organisation. (2006). Global Physical Activity Questionnaire (GPAQ). Geneva: World Health Organisation.

World Health Organization. (2010). Global recommendations on physical activity for health. Geneva: WHO World Health Organization.

World Health Organization. (2012). DEPRESSION A Global Public Health Concern. Retrieved from.http://www.who.int/mental_health/management/depression/wfmh_paper_depression_wmhd_2012.pdf?ua=1.

Wu, E., Barnes, D. E., Ackerman, S. L., Lee, J., Chesney, M., & Mehling, W. E. (2015). Preventing Loss of Independence through Exercise (PLIÉ): Qualitative Analysis of a Clinical Trial in Older Adults with Dementia. Aging & Mental Health, 19(4), 353-362.

Wu, S., Wang, R., Zhao, Y., Ma, X., Wu, M., Yan, X., & He, J. (2013). The relationship between self-rated health and objective health status: a population-based study. BMC public health, 13(1), 320-328.

Wullems, J.A., Verschueren, S.M., Degens, H., Morse, C.I., & Onambélé, G.L. (2016). A review of the assessment and prevalence of sedentarism in older adults, its physiology/health impact and non-exercise mobility counter-measures. Biogerontology, 17(3), 547-565.

Wullems, J.A., Verschueren, S.M., Degens, H., Morse, C.I., & Onambélé, G.L. (2017). Performance of thigh-mounted triaxial accelerometer algorithms in objective quantification of sedentary behaviour and physical activity in older adults. PloS one, 12(11), e0188215.

Xiao, Q., Keadle, S. K., Berrigan, D., & Matthews, C. E. (2018). A prospective investigation of neighborhood socioeconomic deprivation and physical activity and sedentary behavior in older adults. Preventive medicine, 111, 14-20.

Yeo, M., Berzins, S., & Addington, D. (2007). Development of an early psychosis public education program using the PRECEDE–PROCEED model. Health education research, 22(5), 639-647.

Yuan, C.T., Nembhard, I.M., Stern, A.F., Brush, J.E. Jr, Krumholz, H.M., & Bradley, E.H. (2010). Blueprint for the dissemination of evidence-based practices in health care. Issue Brief, 86, 1-16.

Zhu, W., Wadley, V.G., Howard, V.J., Hutto, B., Blair, S.N., & Hooker, S.P. (2017). Objectively Measured Physical Activity and Cognitive Function in Older Adults. Medicine and science in sports and exercise, 49(1), 47-53.

278

Zwarenstein, M., Treweek, S., Gagnier, J.J., Altman, D.G., Tunis, S., Haynes, B., Oxman, A.D., & Moher, D. (2008). Improving the reporting of pragmatic trials: an extension of the CONSORT statement. Bmj, 337, a2390.

Appendices

279

Appendix 1. Ethical Approval

280

Appendix 1.1. Cover Letter

George Sanders Department of Sport and Physical Activity Edge Hill University St Helens Road Ormskirk Lancs, UK L39 4QP

E: [email protected] T: 01695 657 344

22nd January 2016

Dear Professor McNaughton

Please find enclosed an application for ethical approval by the Department of Sport and Physical Activity REC for a study entitled Get Healthy Get Active. The project has confirmed funding by, and will be conducted in collaboration with Sefton Metropolitan Borough Council.

This project aims to increase physical activity levels in adults with intellectual disabilities and older adults (over 65s) with or at risk of dementia within Sefton Borough. By providing the target groups with the opportunity to access bespoke sport and physical activity programmes, the project aims to increase physical activity levels and in doing so reduce the health inequalities currently experienced by these populations. A major strength of the project is its inclusivity and great potential for generalisation. Almost all people with varying abilities will be able to participate in the programme and potentially obtain physical and mental health benefits from it.

The project supervisory team involves individuals with specific knowledge and research experience of both the proposed target groups and methodologies and includes:

Professor Stuart Fairclough - Dept. Sport and Physical ActivityProfessor Brenda Roe - Faculty of Health and Social CareDr Axel Kaehne - Faculty of Health and Social Care

281

This project has already been presented to; the REC Chair for the Faculty of Arts and Sciences, the REC Chair for the Faculty of Health and Social Care, and a selection of professors and Edge Hill University academic staff with specific knowledge of the both the proposed target groups and methodologies. This presentation was positively received and the project subsequently obtained professional indemnity and insurance cover by Edge Hill University. This was approved by the Director of the Research Office, Dr Nikki Craske.

This application for ethical approval is upon recommendation from the Social Care Research Ethics Committee (SCREC), and follows an unsuccessful application for ethical approval by SCREC. Feedback from SCREC highlighted that:

There was a lack of clarity between the activity of the ‘Get Healthy Get Active’ programme and the PhD research project.The Chief Investigator (Mr George Sanders) has insufficient experience of undertaking research with the proposed target groups.There were no breaching confidentiality statements on any of the Participant Information Sheets. The Committee recommended the phrase: ‘Everything you say/report is confidential unless you tell us something that indicates you or someone else is at risk of harm. We would discuss this with you before telling anyone else.’The language used within the consultee consent forms was not appropriate as the forms implied that the consultee was giving proxy consent on behalf of a person lacking capacity to consent. The use of the Mini-Mental State Examination and Short Portable Mental Status Questionnaire as screening tools were inappropriate due to their clinical connotations and American wording.

These points have now been addressed, and following the recommendation of SCREC to apply for ethical approval through the Department of Sport and Physical Activity at Edge Hill University, please find enclosed all relevant details of the project and additional materials. My academic supervisors and I look forward to hearing the outcome of the submission.

Yours faithfully,

George Sanders

282

Appendix 1.2. Participant Invitation Letter

Dept. Sport and Physical ActivityEdge Hill UniversitySt. Helens RoadOrmskirkL39 4QP

January 2016

Dear Sir/Madam,

In partnership with Sefton Metropolitan Borough Council we are conducting a

research project called Get Healthy Get Active. The aim of the project is to increase

physical activity levels in older adults aged 65 years and above. The information

gathered from the project will help us to know whether a programme such as this

can be of benefit to individual’s physical activity levels, fitness, and health.

I am writing to enquire whether you would like to take part in this project. To take

part in the project you need to complete and return the participant consent form to

a project team member.

Your participation in this project is really important to us and as a way of saying

thank you, the project team will be giving all participants a free Sefton Metropolitan

Borough Council ‘Choices’ card, which provides discounted access for participants to

all Active Sefton Leisure facilities. In addition, discounted gym/swim memberships at

just £21 per month will be available. Should you wish to take part please complete

and return the project participant consent form.

Kind regards,

283

George Sanders

PhD Researcher

Sport and Physical Activity Department

Edge Hill University

Email: [email protected]

Tel: 01695 657 344

284

mailto:[email protected]

Appendix 1.3. Participant Information Sheet


E: [email protected]

T: 01695 657 344

Date:

Dear Sir/Madam,

You are being invited to take part in a research project to evaluate Sefton’s ‘Get Healthy Get Active’ programme. Before you decide whether or not to take part, it is vital that you understand why the research is being conducted and what will be required of you should you choose to participate. Please read the following information carefully and ask a researcher involved with the study for assistance if you have any further questions or queries.

Before taking part

This participant information sheet is intended to help you make an informed decision about whether or not to take part in the project.

If you decide to take part, it would be useful if you could inform us before the start of the project of a person that you might want to be contacted should you need support at any point. Whilst the research has been designed to avoid upsetting you, this is always a wise precaution in case the project unintentionally leads to any kind of physical or psychological harm.

What is the purpose of the study?

285

The purpose of the study is to increase physical activity levels in older adults over 65 years of age.

Who is organizing and funding the research and why?

This study is being funded and supported by Sport England, Sefton Metropolitan Borough Council, and Edge Hill University in order to promote physical activity levels and well-being.

The main researcher is George Sanders, who is completing this project with support and supervision from;

- Professor Stuart Fairclough (Edge Hill University) - Dr Axel Kaehne (Edge Hill University) - Professor Brenda Roe (Edge Hill University)

Are there any exclusion criteria?

To be included within the project we ask that you:

Are over 65 years of ageBe without physical disabilities which prevent participation in physical activities

Do I have to take part?

No. There is no obligation on you to take part. This information sheet is designed to help you make an informed decision about whether or not to do so.

What will happen if I agree to take part?

Should you choose to participate, you will be enrolled upon a six/twelve week sport and physical activity programme. Both the programme content and length are dependent upon your current health status and capabilities and will be decided by the research team should you wish to participate. The full programme is as follows:

Participation in a six/twelve week sport and physical activity programme. Participation in three short follow-up sessions 3, 6, and 12 months after the start of the sport and physical activity programme.

At the start of the sport and physical activity programme, and at the three short follow-up sessions the participant will be asked to complete various questionnaires that assess; (i) physical activity levels (ii) falls risk (iii) general health (iv) quality of life (v) self-confidence during sport and physical activity, and (vi) thoughts about the programme. The participant will also be asked to wear an activity monitor on their wrist for 7-days at the start of the programme, and at each of the three follow-up sessions.

286

There will also be opportunities to take part in:

A focus group study discussing your current preferences and attitudes towards physical activity, as well as the provision of services across Sefton Borough.A laboratory-based study assessing physical activity and sedentary time in older adults. During this study you will be asked to wear an activity monitor on your wrist and also to complete a questionnaire assessing your physical activity levels.

Once I take part, can I change my mind?

Yes. Should you wish to withdraw from the study, you are able to do so at any time, for any reason, and you will not be asked to explain your reasons for withdrawing.

However, once the results of the study have been submitted (expected to be by November 2018), it will not be possible to withdraw your individual data from the research.

How long will it take to complete the project?

Including the follow-up period, each participant will be in the study for 12 to 14 months.

What personal information will be required from me?

Information regarding: (i) age (ii) gender (iii) height (iv) weight and (v) current physical activity levels will be required.

Are there any risks in participating?

As with any sport or physical exercise, there is a possibility of both psychological (e.g. anxiety, stress) and physical (e.g. exertion, injury) distress. These risks should be minimal but could occur before, during or after the project. If you do decide to take part in the project, the most important thing is that you feel safe. Therefore, the project team are all specifically trained in delivering sessions that are individually tailored to each participant’s confidence, ability, and skill levels.

The project team has also been given guidance on how to work sensitively and supportively with anyone who might experience upset or distress before, during or after the project.

What are the benefits of taking part?

Taking part offers you the opportunity to access a bespoke sport and physical activity programme, which aims to increase sport and physical activity levels, and might improve your physical and psychological health and well-being, self-confidence for exercise, and general quality of life.

287

By participating in the project you could also help us influence national guidelines regarding physical activity in older adults.

Expenses and payments

As a way of saying thank you, the team will be giving all participants a free Sefton Metropolitan Borough Council ‘Choices’ card, which provides discounted access to all Active Sefton Leisure facilities.

In addition, discounted gym/swim memberships at just £21 per month will be available.

Will the participant’s identity be kept anonymous and confidential throughout the study?

Yes. Any information provided will remain strictly confidential and if you do decide to take part in the project, no names of participants will appear in any work we might publish. All participant and consultee information will be kept safely on a University computer to which only the researchers have access.

Everything you say/report is confidential unless you tell us something that indicates you or someone else is at risk of harm. We would discuss this with you before telling anyone else.

It is entirely your choice about whether to take part. We don’t want to put pressure on anyone.

What will happen to the information?

The information will be collected in order to assist in the completion of:

Sport England’s ‘Get Healthy, Get Active’ initiative, details of which can be found at the link below: http://www.sportengland.org/media/388302/Get-Healthy-Get-Active-Prospectus-FINAL.pdfA PhD research project at Edge Hill University evaluating the results of the ‘Get Healthy Get Active’ programme.

Any information collected will be anonymised and coded to prevent identification, and securely stored using password-protected files on the Edge Hill University computing network. The information will only be used for the purposes of academic research and only research team members will have access to such information.

Upon completion of the study a summary copy of the finalized research study will be made available to you upon request.

288

What happens when the project stops?

Upon completion of the programme, you will be asked to participate in three short follow-up sessions 3, 6, and 12 months after the start of the programme. Should you regret disclosing anything, or want to withdraw your contribution in part or whole throughout this time period, please contact the lead researcher, George Sanders. His contact details can be found on this information sheet. You will need to let the lead researcher know by November 2018 at the latest and he will ensure that any data you have contributed that you want to withdraw does not appear in any academic journals or media reports.

If you later feel distressed about something you have contributed, you might find it helpful to discuss it with a person you trust, and you and/or that person could get in touch with a member of the research team and/or project staff and we will try to do what you advise or negotiate with you to resolve any upset. (See also below in case this might not satisfy you).

What will happen to the results of the project?

We will provide all those helping with the project with a summary of the key findings upon request. The project team also hopes to publish the results in the media, and in professional and academic journals.

We would also hope to keep the anonymous data from the project - securely archived - for up to ten years so that we can compare it with any new findings from other research.

I have some more questions; who should I contact?

Should you have any further questions regarding any part of the research process, please contact the lead researcher, George Sanders, via the contact details provided below:


Telephone number: 01695 657 344

What if I am not happy with how the research was conducted?

If you have a complaint or concern about any aspect of the project, you should contact a member of the research team and/or project staff who will do their best to answer your questions. If they are unable to resolve any concern or you wish to make a complaint regarding the project, please contact Ms Joanne Morris, Edge Hill Universities Research Ethics Secretary at:

289


Ms J Morris Research Office The Lodge Edge Hill University St Helens Road Ormskirk Lancs, UK L39 4QP

Tel: 01695 650925.


The University also has a policy relating to Research Misconduct which is available online at: https://www.edgehill.ac.uk/research/files/2012/05/Strategy-Output-Code-of-Practice-for-the-Investigation-of-Research-Misconduct-RO-GOV-02.pdf

Who has approved the project?

This project has been approved by Edge Hill Universities Department of Sport and Physical Activity Research Ethics Committee.

Further information is available from:


01695 657 344

[email protected]

Thank you for taking the time to read this participant information sheet.

290


https://www.edgehill.ac.uk/research/files/2012/05/Strategy-Output-Code-of-Practice-for-the-Investigation-of-Research-Misconduct-RO-GOV-02.pdf

https://www.edgehill.ac.uk/research/files/2012/05/Strategy-Output-Code-of-Practice-for-the-Investigation-of-Research-Misconduct-RO-GOV-02.pdf

Appendix 1.4. Participant Consent Form

Project Title: Get Healthy Get Active

Lead Researcher Name: Mr George Sanders, Edge Hill University

PARTICIPANT CONSENT FORM

Thank you for considering taking part in this project aiming to increase physical activity levels in Older Adults (over 65 years of age). If you have any questions at all, please ask a member of the project team and/or the lead researcher before you decide whether or not to take part. You will be given a copy of this consent form to keep, which you can refer to at any time.

Please tick one of the relevant ‘Yes’ or ‘No’ boxes below each question.

The purpose and details of this project have been explained to me. I understand that this project is designed to further scientific knowledge and that all procedures have been approved by Edge Hill Universities Department of Sport and Physical Activity Research Ethics Committee.

Yes No

I confirm that I have read and understood the participant information sheet for the project titled ‘Get Healthy Get Active’ dated ............................, and have had the opportunity to consider the information, ask questions and have had these answered satisfactorily.

Yes No

I understand that my participation in this project is voluntary and will have no effect on the care that I receive.

Yes No

I understand that I have the right to withdraw from this project at any stage for any reason without my care or legal rights being affected, and that I will not be required to explain my reasons for withdrawing.

I understand that if I withdraw from the project before November 2018, any data contributed that I want to withdraw will not appear in

Yes

Yes

No

No

291

any academic journals or media reports upon request.

I understand that all the information I provide will be treated in strict confidence and will be kept anonymous and confidential to the research team at Edge Hill University unless (under the statutory obligations of the agencies which the researchers are working with), it is judged that confidentiality will have to be breached for the safety of the participant or others.

Yes No

I understand that the research team may use my telephone/ mobile number to call/ text message regarding the Get Healthy Get Active project and for no other reason.

Yes No

I agree to participate in this project. Yes No

Name of participant ________________________________(please print)

Participant signature ________________________________

Date ________________________________

Name of researcher _________________________________(please print)

Signed __________________________________

Date __________________________________

292

Appendix 1.5. Ethical Approval Letter

293

Appendix 2. Accelerometer Instructions

294

295

Appendix 2.1. GENEActiv Accelerometer Instructions

HOW TO WEAR THE ACTIVITY MONITOR

As part of the Get Healthy Get Active project, we will be investigating how active you are during one week (7 days). To measure your activity you will wear a GENEactiv accelerometer. It is a wrist-worn lightweight monitor that detects activity by sensing movement.

How do I wear it?Wear the monitor around your non-dominant wrist, just like a watch.

Adjust the GENEactiv wrist strap so that it is tight enough so that the monitor does not move when you are being active

When do I wear it?

Please wear the monitor for 24 hours

The monitor is waterproof, so there is no need to remove it before showering or swimming

You should wear the monitor for at least 10 hours each day

When and how do I give the monitor back?George Sanders will arrange collection of the monitor after you have worn it for 7 days.The research team may use your mobile number to call/ send text message reminders to wear the physical activity monitor and for no other reason.

If you have any questions you can contact George by telephone: 01695 657 344 or by email: [email protected]

295

296

ACTIVITY MONITOR DIARY Monitor Number: ___________

In the table below write down the times that you put the monitor on and take it off during each day. The first row is an example for you to see

how to fill it out.

If you take the monitor off for more than 5 minutes, please record when you take it off and put it back on.

My name is: ____________________________________________________________

Day: Time periods when activity monitor was taken off

Reason activity monitor was taken off

Signature

Example:Wednesday

8:00am-8:15am Getting changed Mrs Smith

296

https://doi.org/10.1017/S1463423618000373

297

Appendix 3. Associated Publications

https://doi.org/10.1017/S1463423618000373

298

Primary Health Care Research & Development

cambridge.org/phc

ResearchCite this article: Sanders GJ, Roe B, Knowles ZR, Kaehne A, Fairclough SJ. (2018) Using formative research with older adults to inform a community physical activity programme: Get Healthy, Get Active. Primary Health Care Research & Developmentpage 1 of 10. doi: 10.1017/S1463423618000373

Received: 2 November 2017Revised: 18 April 2018Accepted: 1 May 2018

Key words:ageing; community groups; formative; physical activity; primary care

Author for correspondence:George J. Sanders, Department of Sport and Physical Activity, Edge Hill University,St Helens Road, Ormskirk L39 4QP, UK.E-mail: [email protected]

Using formative research with older adults to inform a community physical activity programme: Get Healthy, Get Active

George J. Sanders1, Brenda Roe2,3, Zoe R. Knowles4, Axel Kaehne2 and Stuart J. Fairclough1,5

1Physical Activity and Health Research Group, Department of Sport and Physical Activity, Edge Hill University, Ormskirk, UK, 2Faculty of Health & Social Care, Edge Hill University, Ormskirk, UK, 3Personal Social Services Research Unit, University of Manchester, Manchester, UK, 4The Research Institute for Sport and Exercise Sciences, Liverpool John Moores University, Liverpool, UK and 5Department of Physical Education and Sports Sciences, University of Limerick, Ireland

Introduction

In the United Kingdom there are over 11 million older adults aged 65 years and over who make up 18% of the population (UK Office for National Statistics, 2017). Aligning with the United States and other developed countries (United Nations, 2015) this proportion is projected to increase to at least 24% by 2039 (UK Office for National Statistics, 2017). Although prolongation of life remains an important public health goal, of even greater significance is that extended life should involve preservation of the capacity to live independently, function well and quality of life (Rejeski et al., 2013). The purpose of this formative descriptive study was to explore current knowledge and attitudes towards physical activity (PA), as well as perceived barriers, facilitators and opportunities for PA participation among older adults living in the community. The findings were used to inform the design, delivery and recruitment strategies of an ongoing three-year community PA intervention project, Get Healthy, Get Active (GHGA), which forms part of Sport England’s national GHGA programme (Sport England, 2012).

© Cambridge University Press 2018.

Abstract

Aim: The purpose of this formative study was to explore current knowledge and attitudes towards physical activity, as well as perceived barriers, facilitators and opportunities for physical activity participation among older adults living in the community. The findings have subsequently informed the design, delivery and recruitment strategies of a local community physical activity intervention programme which forms part of Sport England’s national Get Healthy, Get Active initiative. Background: There is a growing public health concern regarding the amount of time spent in sedentary and physical activity behaviours within the older adult population. Methods: Between March and June 2016, 34 participants took part in one of six focus groups as part of a descriptive formative study. A homogenous purposive sample of 28

community-dwelling white, British older adults (six male), aged 65–90 years (M = 78, SD = 7

years) participated in one of five focus group sessions. An additional convenience pragmatic sub-sample of six participants (three male), aged 65–90 years (M = 75, SD = 4 years), recruited from an assisted living retirement home participated in a sixth focus group. Questions for focus groups were structured around the PRECEDE stage of the PRECEDE– PROCEDE model of health programme design, implementation and evaluation. Questions addressed knowledge, attitudes and beliefs towards physical activity, as well as views on

barriers and opportunities for physical activity participation. All data were transcribed verbatim. Thematic analysis was then conducted with outcomes represented as pen

https://doi.org/10.1017/S1463423618000373

299

BackgroundGuidelines issued by the UK Chief Medical Officers and the US Surgeon Generals recommend that older adults (⩾65 years) engage in at least 150 min of moderate (or 75 min of vigorous) PA per week in bouts of at least 10 min, with muscle-strengthening and balance activities

Downloaded from https://www.cambridge.org/core. Edge Hill University, on 24 Jul 2018 at 12:39:18, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.

2 George J. Sanders et al.

Downloaded from https://www.cambridge.org/core. Edge Hill University, on 24 Jul 2018 at 12:39:18, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1463423618000373

included on at least two of those days (Department of Health, 2011; Centers for Disease Control and Prevention (CDC), 2015). Despite the recognised evidence base for the benefits of regular PA (CDC, 2015; Reid and Foster, 2016; World Health Organi- zation (WHO), 2017), objective summaries of PA levels among older adults show that only 15% of males and 10% of females within the United Kingdom, and 9.5% of males and 7% of females within the United States meet the recommended PA guidelines (Tucker et al., 2011; Jefferis et al., 2014). Given that current PA guidelines remain the same for both adults (18–64 years) and older adults (⩾65 years), such high levels of inactivity suggests that PA guidelines appear too demanding for the latter population (Booth and Hawley, 2015).Accumulating evidence suggests that prolonged and continuous bouts of sedentary behaviours [SB; defined as waking behaviours in a sitting, reclining or lying posture with energy expenditure ⩽1.5 metabolic equivalents (Tremblay et al., 2017)] have similar physical (eg, premature mortality, chronic diseases and all-cause dementia risk) and psychosocial (eg, self-perceived quality of life, well-being and self-efficacy) risk factors to that of physical inactivity (Wilmot et al., 2012; Edwards and Loprinzi, 2016; Falck et al., 2016; Kim et al., 2016). In fact, SB is now an identifiable risk factor independent of other PA behaviours (Tremblay et al., 2017). Spending on average 80% of their time in a seated posture, and with 67% being sedentary for more than8.5 h/day (Shaw et al., 2017), older adults are the most sedentary segment of society and seldom engage in moderate-to-vigorous PA (Chastin et al., 2017).Several social (eg, social awkwardness and peer/family support), behavioural (eg, ageing stereotypes and lack of time), physical (eg, improved balance and flexibility) and environmental (eg, transport and neighbourhood safety) correlates of PA among older adults have been noted in recent formative (van Schijndel-Speet et al., 2014; Banerjee et al., 2015) and qualitative research (Franco et al., 2015; Devereux-Fitzgerald et al., 2016; Phoenix and Tulle, 2017). Such findings are a first step in enabling policymakers and health care professionals to implement effective PA interventions and promote active ageing (Franco et al., 2015). Given the potential benefits associated with PA outlined, such interventions have the potential to reduce, age-related morbidity and declines in activities of daily living, maintain muscle strength and mass, improve quality of life, and thus reduce the primary and total health care costs associated with SB and physical inactivity among this population (Bauman et al., 2016).Prior research notes that interventions aimed at promoting PA participation should adopt an appropriate conceptual health promotion model to prioritise the key assets of the target group (Plotnikoff et al., 2014). The PRECEDE–PROCEED model of health programme design, implementation and evaluation (Green and Kreuter, 2005) provides the target population with a comprehensive and structured assessment of their own needs and barriers to a healthy lifestyle. The PRECEDE component of the model comprises of, predisposing, enabling and reinforcing factors has previously been used as a formative framework to guide PA intervention content and design (Mackintosh et al., 2011; Banerjee et al., 2015). This model has also been adopted as a method for the identification of perceived PA barriers and facilitators among older adults (Banerjee et al., 2015; Gagliardi et al., 2015) and other populations (Mackintosh et al., 2011; Emdadi et al., 2015; Susan et al., 2017).The purpose of this formative study was to (i) explore current knowledge and attitudes towards PA, as well as the perceived

barriers, facilitators and opportunities for PA participation among older adults living in the community who had agreed to take part in an ongoing PA programme; and (ii) use this data to inform the design, delivery and recruitment strategies of an ongoing community PA intervention programme, as well as international PA interventions among this population. Given the purpose and aims outlined, the Evidence Integration Triangle (Glasgow et al., 2012) was adopted as the overarching theoretical framework. Through the prompt identification of success and failures across individual-focussed and patient–provider interventions, as well as health systems and policy-level change initiatives, the framework allows for the exploration of the three main evidence-based components of intervention program/policy, implementation processes and measures of progress. Hence, this framework enabled a steep learning cycle through an initial 12-week pilot GHGA programme delivered by the Metropolitan Borough Council within the chosen local authority. Results and analysis from this pilot were fed back to Sport England as the funder, as well as deliverers and participants in order to assess, evaluate and promptly inform adapted future iterations of the GHGA programme.

Methods

Participants and proceduresA descriptive formative study was undertaken from March to June 2016. Participants were recruited from one local authority in North West England recognised as having the highest percentage of inactive older adults (80%) compared to the UK national average, and the highest national health costs associated with physical inactivity (Active People Survey, 2014; Sport England’s Local Profile Tool, 2015). The first author facilitated six, mixed- gender focus groups. Representative of the uptake of participants within the target GHGA initiative, a homogenous purposive sample of 28 community-dwelling white, British older adults (five male) participated in five of the focus groups, with an additional convenience pragmatic sub-sample of six participants (three male) recruited from an assisted living retirement home, participating in the sixth focus group. In total, 34 older adults (eightmale), aged 65–90 years (M = 78, SD = 7 years), participated across the six sessions. Four focus groups involved a group size of six to ten participants, and two involved three participants (mean focus group size of 6 ± 5 participants). Previous focus groups inPA studies have been conducted effectively with as many as 12 (Moran et al., 2015), and as few as four (Schneider et al., 2016) participants. Focus groups took place in two church halls, an assisted living retirement home lounge, and a theatre. All locations were free from background noise, and participants could be overlooked but not overheard. The inclusion criterion set out by Sport England as funders of the GHGA programme were that participants must be 65 years of age or over, reside within one local authority in North West England, could provide written informed consent to participate.GHGA is an ongoing three-year project which seeks to increase the number of inactive older adults participating in PA at least once a week for 30 min, via a 12-week PA intervention delivered by the Metropolitan Borough Council within the assigned local authority. Participants due to participate in GHGA received a covering letter, participant information sheet, and consent form. Prior to the commencement of the study, institutional ethical approval was received (#SPA-REC-2015-329) and


3


written informed consent was obtained for all participants prior to participation. All focus groups utilised the PRECEDE stage of the PRECEDE–PROCEDE model (Green and Kreuter, 2005) within their design allowing for the exploration of predisposing, enabling and reinforcing correlates of PA participation. To maximise the interaction between participants, focus group questions were reviewed by the project team for appropriateness of question ordering and flow. Subsequent minor additions were made to questions on social isolation and PA advertisement. The semi-structured discussion guide included open ended questions structured to prompt discussion with equal chance for participants to contribute (Stewart and Shamdasani, 2014). Focus groups were led by a trained facilitator and with an observer/ note taker also present. Questions addressed knowledge, attitudes and beliefs towards PA as well as views on barriers and opportunities for PA participation. An example question from a section exploring barriers to PA was: ‘Can you tell me about what stops you from participating in physical activity?’ Questions therefore demonstrated aspects of face validity as they were transparent and relevant to both the topic and target population (French et al., 2015).

Data coding and analysisFocus groups lasted between 20 and 45 min (M = 29, SD = 12), were audio recorded, and later transcribed verbatim, resulting in 66 pages of raw transcription data with Arial font, size 12 and double-spaced. Verbatim transcripts were read and re-read to allow familiarisation of the data and then imported into the QSR NVivo 11 software package (QSR International Pty Ltd., Doncaster, Victoria, Australia, 2017).Previous research within this population has adopted analytical procedures including thematic analysis (Van Dyck et al., 2017), content analysis (Middelweerd et al., 2014) and used specialist qualitative data analysis packages, such as NVivo (Warmoth et al., 2016). In supporting new methodologies and data representation within qualitative research (Orr and Phoenix, 2015), the current study followed the pen profiling protocol. The pen profile approach has been used in recent child PA research (Mackintosh et al., 2011; Boddy et al., 2012; Knowles et al., 2013; Noonan et al., 2016b) and presents findings from content analysis via a diagram of composite key emerging themes. In summary, data were initially analysed deductively via content analysis (Braun and Clarke, 2006), using the PRECEDE component of the PRECEDE–PROCEED model (Green and Kreuter, 2005) as a thematic framework which reflects the underlying study purpose. Inductive analysis then allowed for emerging themes to be created beyond the pre-defined categories. Data were then organised schematically to assist with interpretation of the themes (Aggio et al., 2016). As akin to more traditional qualitative research, verbatim quotations were subsequently used to expand the pen profiles, provide context and verify participant responses. Previous studies have demonstrated this method’s applicability in representing analysis outcomes within PA research (Mackintosh et al., 2011; Boddy et al., 2012; Knowles et al., 2013; Noonan et al., 2016a) making it accessible to researchers who have an affinity with both quantitative and qualitative backgrounds (Knowles et al., 2013; Noonan et al., 2016a). Recent findings suggest that the discrepancy between objective isolation and felt loneliness may be associated with undesirable health outcomes such as cognitive dysfunction.Three pen profiles were developed to display themes within the data aligned to the PRECEDE component of the

PRECEDE–PROCEED model (Green and Kreuter, 2005). Quo- tations were labelled by focus group number (Fn) and subsequent participant number (Pn) within that focus group. Characterising traits of this protocol include details of frequency counts and extracts of verbatim quotes to provide context to the themes. A minimum threshold for theme inclusion was based upon comparable participant numbers within previous research adopting a pen profiling approach (Boddy et al., 2012; Noonan et al., 2016a)and hence, was set as ⩾n = 6, with n representing individual mentions per participant. However, multiple ‘mentions’ by thesame participant were only counted once. Methodological rigour was demonstrated through a process of triangular consensus (Hawley- Hague et al., 2016) between the authors. This offered transparency, credibility and trustworthiness of the results, as the data were critically reviewed using a reverse tracking process from pen profiles to verbatim transcripts, providing alternative interpretations of the data (Smith and Caddick, 2012). The process was repeated through cross-verification and discussion until subsequent agreement on data themes in relation to verbatim extracts was reached (Aggio et al., 2016).

Findings and discussion

Predisposing correlatesFigure 1 displays the predisposing correlates of PA participation. In agreement with previous research (Gray et al., 2015; Kosteli et al., 2016), the most highly cited theme of motivation (n = 29) was perceived to be both a facilitator (n = 15) and barrier (n = 14) to PA participation throughout. Some participants were proactivein seeking out opportunities for PA.I’m a lung cancer survivor and I just ran a mile last month and I raised£550.

(Focus group (F) 1: Participant (P) 2)

Contrastingly, others expressed disinterest in PA altogether believing that they would not derive any health benefit.I’ve pushed these [PA] classes to lots and lots of friends and they still ignore it, they will not come to anything like this.(F1: P3)

Participants also reported laziness or apathy to prevent participation.It’s [lack of PA] apathy, just apathy, people can’t be bothered.(F4: P3)

The importance of pre-intervention intrinsic motivation (eg, participating for enjoyment) among older adults is key for both initial adoption and maintenance of PA participation (Gray et al., 2015). Hence, future interventions could promote intrinsic motivation for PA through the adoption of socio-emotional selectivity theory (Carstensen et al., 1999). Recent findings support this theory’s notion that motivation for PA is more effectively promoted when paired with positive messages about the benefits of PA rather than with negative messages about the risks of inactivity (Notthoff et al., 2016).The theme of age (n = 20) was identified as a key barrier (n = 13) to PA participation throughout.They [older adults] get to a certain age and just give up.

(F1: P7)Social norms and cultural misconceptions often influence not only the type of PA in which older adults engage, but whether they participate at all (Greaney et al., 2016). Moreover,



Figure 1. Predisposing correlates of physical activity participation among older adults. n = Individual mentions per person (multiple mentions not included); Fn = focus group number; Pn = participant number.

participants noted that lifestyle (n = 20) often affects individual views regarding ageing stereotypes, and therefore PA participation. Some participants felt that physically active older adults were more likely to be habituated to PA engagement over many years.Well if you’ve kept healthy, kept fit all your life, you can keep doing it.(F1: P4)

Conversely, it was felt that inactive older adults were reluctant to start exercising.You see the ones who haven’t been doing it [PA] are not going to be able to start and do it now.(F2: P1)

Previous research has also reported prior PA behaviours (eg, being sedentary or active) to be key correlates affecting older adults’ current PA participation levels (Franco et al., 2015). Additionally, ageing is associated with a decrease in the size of social networks and hence, older adults are at increased risks of isolation (Devereux-Fitzgerald et al., 2016; Greaney et al., 2016).Corroborating with prior research (Greaney et al., 2016), participants throughout perceived isolation (n = 15) to be a key barrier (n = 14) to PA participation.

It’s so easy to get trapped inside and not go out. People sit in front of the television from the moment they wake up to when they go to bed.(F6: P5)

Isolation is associated with decreased social and psychological well-being (Owen et al., 2010; Milligan et al., 2015) and increased SB among older adults (Nicholson, 2012). Certain targeted intervention strategies can reduce isolation by providing an opportunity for older adults from differing socio-economic areas to take part in PA within local community spaces (eg, parks, leisure centres and churches), that promote social networking by encouraging camaraderie, adaptability and productive engagement, without the pressure to perform (Milligan et al., 2015; Gardiner et al., 2016). Given that SB is an independent and

modifiable behavioural target for interventions (Lewis et al., 2017), opportunities to replace SB with health-enhancing behaviours such as moderate-to-vigorous PA (Prince et al., 2014), light PA (McMahon et al., 2017; Phoenix and Tulle, 2017) and standing (Healy et al., 2015) should be promoted. However, none of the participants in the current study noted negative health effects of prolonged sitting, or the importance of breaks in sedentary time. Previous research has noted that older adults are not yet familiar with the concept of SB and hence, are not motivated to reduce such behaviours (Van Dyck et al., 2017). Hence, it is first crucial to increase knowledge about the negative health consequences of SB independent from PA among both older adults and other populations (Van Dyck et al., 2017).Participants also emphasised the importance of having a wide range of choice and opportunities for PA (n = 22), and in general their perceptions of community provision were positive (n = 16).Yes it’s quite a good place [the local authority where the study took place]. There are a lot of different physical activity sessions to try.(F2: P1)

However, in line with recent research (Baert et al., 2016; Träff et al., 2017), key barriers noted by the participants within the assisted living group included a lack of advertisement regarding PA opportunities, and few opportunities to take part in PA within the assisted living facility itself.It’s hard to know what is on if you don ’t read the noticeboards and to be honest most of us have even stopped looking at that [noticeboard] because there is never anything on it.(F3: P3)

Further research into the most effective advertisement strategies to engage older adults in assisted living facilities is warranted (Hildebrand and Neufeld, 2009). Regardless of living status, participants noted a strong preference not to engage with online and/or social media channels for advertising and awareness- raising.


5


A lot of people our age don’t like that technology stuff at all. I would not know where to start.(F5: P2)

These results suggest educational strategies outlining the potential benefits of technology in aiding PA participation are needed (Bird et al., 2015). This is especially salient given that recent research has shown technology-based interventions to have good adherence and provide a sustainable means of reducing SB and promoting PA participation among older adults (Garcia et al., 2016; Skjæret et al., 2016).

Enabling correlatesFigure 2 displays the enabling correlates of PA participation. Consistent with previous research findings (Franco et al., 2015; Borodulin et al., 2016), cost (n = 21) was perceived to be a key barrier (n = 12) to PA participation exclusively among the community dwelling participants who were either unable, or unwillingto pay the perceived high costs associated with both attending and travelling to such programmes.Money is the big bug bear [barrier to PA participation] isn’t it.(F2: P5)

Examples of competing programmes were also noted, with free and lower cost programmes taking precedence over the more expensive.We like it [a local chair-based PA programme] because it’s free.(F4: P3)

Thus, to effectively increase PA participation within this population, health-promotion strategies should go further than merely educating and raising awareness about potential health benefits, and should also advocate for the provision of low-cost, and easy reachable PA opportunities regardless of financial status (Petrescu-Prahova et al., 2015; Borodulin et al., 2016). It is worth

noting that for the participants recruited from the assisted living retirement home, any PA sessions delivered were included within the cost of the overall living fee, and hence lack of financial resources was rejected as a potential barrier for PA participation (Baert et al., 2016).Participants’ views on the theme of location (n = 11) centred on neighbourhood safety. Declining health and physical impairmentsassociated with ageing increase the time spent in ones’ neighbourhood and thus, neighbourhood environmental factors such as, PA provision, proximity, traffic volume and overall neighbourhood safety are considered to be important correlates affecting older adults’ PA participation (Greaney et al., 2016). Perceived neigh-bourhood safety was identified as a barrier (n = 7) to PA participation exclusively among the community-dwelling older adults.You wouldn’t go out on your own at night around here.(F1: P5)

Participants from the assisted living retirement home did not view neighbourhood safety to be either a barrier to or facilitator of PA. This neighbourhood environment was perhaps viewed as the norm and therefore they did not associate safety concerns so acutely (Moran et al., 2015). This association could have also affected results obtained for the theme time/day of the week as such participants did not recognise this to be a barrier to PA participation either.Time of day wouldn’t make much difference [to PA participation]. To be fair you aren’t doing much at the weekend so day of the week isn’t going to make much difference [to PA participation] either.(F3: P1)

Conversely, community-dwelling participants reported time/ day of the week to be a barrier (n = 15), with early morning or early evening sessions identified as reducing PA participation, especially during the winter months when daylight hours are more limited. These findings could have been further

Figure 2. Enabling correlates of physical activity participation among older adults. n = Individual mentions per person (multiple mentions not included); Fn = focus group number; Pn = participant number.



Figure 3. Reinforcing correlates of physical activity participation among older adults. n = Individual mentions per person (multiple mentions not included); Fn = focus group number; Pn = participant number.

amplified by the neighbourhood safety concerns also identified by this group (Hoppmann et al., 2015; Prins and van Lenthe, 2015).The theme of transportation (n = 14) has been extensively reported to be both a barrier and facilitator to PA participation among older adults (Bouma et al., 2015; Haselwandter et al., 2015; Kosteli et al., 2016; Van Dyck et al., 2017). Within the currentstudy transportation was identified as a barrier (n = 10) restricting access to PA sessions regardless of living status.I would like to go to the baths [swimming pool] but it’s difficult to get there and back so I just don’t bother.(F4: P5)

Transport is especially important for those lacking the ability to be more independently mobile as it allows individuals to bridge larger distances than they could by walking alone (Van Cauwenberg et al., 2016). Thus, lack of access to a car and inadequate availability, frequency and reliability of affordable public transport are all associated with decreased PA participation (Newitt et al., 2016). Additionally, being dependent upon others (eg, family, friends and peers) for transportation has been identified as a barrier to PA participation within this population (Baert et al., 2015). This was also noted in the current study.I think the worst thing is having to rely on somebody else to take you [to a PA session] as anything can happen in your own life let alone somebody else’s. (F5: P2)

Prior research suggests the promotion of walking for transportation to PA sessions among physically independent older adults (Chudyk et al., 2017). However, given the neighbourhood safety concerns noted by participants, and the varying levels of functional ability among this population, further research examining access to PA sessions including walking facilities (eg, path and crossing quality), traffic safety and safety from crime is warranted (Van Cauwenberg et al., 2016).

Reinforcing correlatesFigure 3 displays the reinforcing correlates of PA participation. Peer support is associated with PA adherence in older adults (Brown et al., 2015), and was identified as a key theme (n = 18) and subsequent facilitator (n = 13) to PA participation in the current study.I’ve got to know everybody now and I’m used to you all. I feel more comfortable and I don’t feel anxious or anything.(F3: P6)

Unsurprisingly, in light of the above several participants reported peers to be a barrier to PA participation (n = 5) because of an unwillingness to attend other PA sessions due to anxieties about meeting new people.I wouldn’t like to go somewhere else as I wouldn’t like to walk in on a crowd of new people.(F3: P6)

Although group-based activities offer older adults the chance to gain a sense of belonging, enjoyment and establish friendships, designing sustainable exit routes in order to retain the provision of group activities which continue to facilitate, build and retain social bonds post-intervention should be considered by PA programmers and policymakers (Wu et al., 2015).In line with recent research (Devereux-Fitzgerald et al., 2016; Smith et al., 2017), family members were identified as being both barriers (n = 2) and facilitators (n = 4) to PA participation. Specifically, a barrier often reported is overprotectiveness, in which family members may not allow older adults to participatein PA out of concern for their safety or health (Greaney et al., 2016). Participants among the community-dwelling groups also noted this.My sons in for a shock that we’re coming to this as he’s like, ‘no long walks, no boat rides’, he goes ‘you’re past it’.(F6: P2)


7


Such results suggest a need to educate family members on the importance and benefits of PA among older adults. Educational resources such as the older adults PA guidelines infographics for the, United Kingdom (Reid and Foster, 2016), Canada (Canadian Society for Exercise Physiology, 2016), Australia (Australian Government Department of Health and Ageing, 2013), New Zealand (Ministry of Health, 2013) and the United States (CDC, 2008) are appropriate tools advocating for older adults to be active safely, and can be understood by family members plus health care providers. Furthermore, the adoption of local/national mass media messages may be a cost effective educational solution at a time when there is a growing ageing population (United Nations, 2015; UK Office for National Statistics, 2017). However, given the resistance to technology-based PA noted in the current study, further educational strategies promoting enjoyable, easy-to-use technology within a family environment are needed for community-dwelling older adults (Bird et al., 2015). Participants within the assisted living group did not perceive family members to be either barriers or facilitators to PA participation and thus, further research is needed to identify approaches to involve family members as additional facilitators of PA participation within this group.Participants viewed the theme of perceived health benefits (n = 23) to be both a facilitator (n = 14) and barrier (n = 9) to PA participation regardless of living status. Participants were knowledgeable regarding the potential benefits of PA for their physical health.

It [PA] loosens all your limbs up.(F2: P2)

Participants also noted the potential benefits of PA for their psychological health.The wellbeing [from PA participation] makes you feel better.(F1: P3)

Despite the irrefutable evidence demonstrating the benefits of PA among older adults (CDC, 2015; Reid and Foster, 2016; WHO, 2017), participants also noted health to be a potential barrier (n = 14) to PA participation due to doubts about their capabilities, or fear of causing themselves harm, particularly if they were unfamiliar with it.People have to be sure they can come to PA sessions because my sister had a heart attack … and she can’t do a lot of these exercises.(F1: P5)

To overcome such perceptions, educational strategies at a population level should focus on communicating the role of PA in gaining health benefits for all as well as how well-designed PA programmes can aid in the management of common comorbidities specific to this age group (Gillespie et al., 2012; Hamer et al., 2013).Taken together with the findings of recent qualitative studies examining correlates of PA participation among older adults living in both assisted living (Baert et al., 2016; Träff et al., 2017) and community-dwelling older adults (Fisher et al., 2017; Phoenix and Tulle, 2017), results from this formative research study have been used to inform the design, delivery and recruitment strategies of an ongoing community PA intervention project. Specifically, changes implemented to programme design have included the introduction of, increased intervention duration from 6 to 12-weeks, maintenance sessions post-initial 12-week intervention, tea and coffee after each session to promote social interaction, and a reduction of early morning and late afternoon sessions. Changes to programme delivery have

included the introduction of, participant choice in session activities, videoing participants at week 1 and week 12 to show participants their progression, and signposting participants to other local PA programmes. Finally, changes implemented to recruitment strategies have included, improved relationships with general practitioners to enable them to refer participants onto the programme, leafleting in church halls and charity shops, and deliverers attending and subsequently advertising the programme at several Older Peoples’ Forums. Such methods could also be adopted throughout similar community PA programmes elsewhere in order to increase programme fidelity, representativeness and effectiveness.

Strengths and limitations

Methodological strengths include the exploration of consensus and associated discussion through the focus groups and subsequent analysis process which allowed insight into the predisposing, enabling and reinforcing correlates of PA participation among older adults. Consistency of themes, data credibility, transferability, and dependability were achieved through the triangulation consensus of data between authors and methods. While this study reiterates important insights into the perceived barriers, facilitators and opportunities for PA participation among both community-dwelling and assisted living older adults, value outside of this to the wider research community may be limited due to programme funding which only allowed for formative research strategies to recruit participants who had agreed to take part in an ongoing PA programme. Consequently, sampling bias is a potential issue as it could be assumed that a high proportion of the participants were already inclined to be and/or currently physically active given the positive predisposing comments with regard to motivation towards PA and current lifestyle choices (Costello et al., 2011). This is especially important given that motivators and barriers towards regular PA vary among currently active and inactive adults across the age range (Costello et al., 2011; Hoare et al., 2017). Considering that less than 10% of older adults ( ⩾ 65 years of age) meet the recommended PA guidelines (Jefferis et al., 2014), future research should seek to identify barriers and facilitators among larger sample sizes of currently inactive older adults living within both the community and assisted living facilities.Additionally, a small convenience pragmatic sub-sample of participants from one assisted living facility were recruited and hence results cannot be considered representative. Furthermore, men tend to decrease participation in leisure-time PA as they get older; whereas this dose-response is not seen among women (Amagasa et al., 2017). Consequently, there is the possibility of gender bias given the higher number of female participants recruited. However, the sample size, participants’ ages and gender distribution are comparable to those reported in two recent studies examining barriers and facilitators to PA participation among older adults (Baert et al., 2015; Moran et al., 2015). Within these two studies the total number of participants was 15 (five male) and 40 (13 male), and the mean age of the respondents was 74 years, and 84 years, respectively. This compares to a total number of 34 participants (eight male) with a mean age of 78 years in the current study. Nevertheless, as well as exploring correlates of PA participation in relation to gender, functional status and age differences between the young–old (60–69 years), old–old (70–79 years) and oldest–old (80 + years) (Heo et al., 2017), future research should obtain additional participant



characteristic data prior to the intervention including, participants’ current sedentary time and PA levels, history of PA, family history of PA, ethnicity, employment status and educational achievements as such have been shown to potentially affect the perceived barriers and facilitators to PA participation among older adults (Greaney et al., 2016; Keadle et al., 2016).

Conclusions

Older adults acknowledged the benefits of PA, not only for health but also those relating to socialising, enjoyment, relaxation, and physical and psychological well-being. The themes of opportunities and awareness for PA participation, cost, transport, location and season/weather varied dependent upon living status. These findings suggest current living status to be a separate correlate of PA participation among older adults. This data can be used to further strengthen the design, delivery and recruitment strategies of both the target GHGA PA intervention programme and international PA intervention programmes among older adults. Future interventions should consider educational strategies to communicate the role of PA in gaining health benefits for all, reducing SB, and countering the negative implicit attitudes that may undermine PA within this population. Given the small sample of participants in the current study, further comparativeresearch exploring the barriers and facilitators between assisted living and community-dwelling, and active and inactive older adults on both national and international levels is warranted.

Acknowledgements. The authors express their deepest gratitude to all the participants involved. Edge Hill University institutional ethical approval number: # SPA-REC-2015-329.

Conflicts of Interest. No potential conflict of interest was reported by the authors.

References

Active People Survey (2014) Active People Survey 8: Active People Interactive. Retrieved 9 October 2017 from http://activepeople. sportengland.org/Aggio D, Fairclough S, Knowles Z and Graves L (2016) Validity and reliability of a modified English version of the physical activity questionnaire for adolescents. Archives of Public Health 74, 3–11.Amagasa S, Fukushima N, Kikuchi H, Takamiya T, Oka K and Inoue S (2017) Light and sporadic physical activity overlooked by current guidelines makes older women more active than older men. International Journal of Behavioral Nutrition and Physical Activity 14, 59–65.Australian Government Department of Health and Ageing (2013) Recommendations on physical activity for health for older Australians. Retrieved 9 October 2017 from http://www.health.gov.au/internet/main/ publishing.nsf/content/130D93778A64136DCA257BF0001DACF2/$File/paguidelines.pdfBaert V, Gorus E, Calleeuw K, De Backer W and Bautmans I (2016) An administrator’s perspective on the organization of physical activity for older adults in long-term care facilities. Journal of the American Medical Directors Association 17, 75–84.Baert V, Gorus E, Mets T and Bautmans I (2015) Motivators and barriers for physical activity in older adults with osteoporosis. Journal of Geriatric Physical Therapy 38, 105–114.Banerjee AT, Kin R, Strachan PH, Boyle MH, Anand SS and Oremus M (2015) Factors facilitating the implementation of church-based heart health promotion programs for older adults: a qualitative study guided by the precede-proceed model. American Journal of Health Promotion 29, 365–373.

Bauman A, Merom D, Bull FC, Buchner DM and Fiatarone Singh MA (2016) Updating the evidence for physical activity: summative reviews of the epidemiological evidence, prevalence, and interventions to promote “Active Aging”. The Gerontologist 56, (Suppl 2), S268–S280.Bird ML, Clark B, Millar J, Whetton S and Smith S (2015) Exposure to ‘exergames’ increases older adults’ perception of the usefulness of technology for improving health and physical activity: a pilot study. JMIR Serious Games 3, 1–8.Boddy LM, Knowles ZR, Davies IG, Warburton GL, Mackintosh KA, Houghton L and Fairclough SJ (2012) Using formative research to develop the healthy eating component of the CHANGE! School-based curriculum intervention. BMC Public Health 12, 710–720.Booth FW and Hawley JA (2015) The erosion of physical activity in western societies: an economic death March. Diabetologia 58, 1730–1734.Borodulin K, Sipilä N, Rahkonen O, Leino-Arjas P, Kestilä L, Jousilahti P and Prättälä R (2016) Socio-demographic and behavioral variation in barriers to leisure-time physical activity. Scandinavian Journal of Public Health 44, 62–69.Bouma AJ, van Wilgen P and Dijkstra A (2015) The barrier-belief approach in the counseling of physical activity. Patient Education and Counseling 98, 129–136.Braun V and Clarke V (2006) Using thematic analysis in psychology.Qualitative Research in Psychology 3, 77–101.Brown D, Spanjers K, Atherton N, Lowe J, Stonehewer L, Bridle C, Sheehan B and Lamb SE (2015) Development of an exercise intervention to improve cognition in people with mild to moderate dementia: Dementia and Physical Activity (DAPA) trial, registration ISRCTN32612072. Physiotherapy 101, 126–134.Canadian Society for Exercise Physiology (2016) Canadian physical activity guidelines for older adults – 65 years and older. Retrieved

9 October 2017 from https://www.participaction.com/sites/default/files/ downloads/Participaction-Canadian-physical-activity-guidelines-older-adult.pdf

Carstensen LL, Isaacowitz DM and Charles ST (1999) Taking time seriously: a theory of socioemotional selectivity. American Psychologist

54, 165–181.Centers for Disease Control and Prevention (CHD) (2008) Physical activity guidelines for Americans: fact sheet for health professionals on physical activity guidelines for older adults. Retrieved 9 October 2017 from http:// www.cdc.gov/physicalactivity/downloads/pa_fact_sheet_olderadults.pdfCenters for Disease Control and Prevention (CHD) (2015) How much physical activity do older adults need?. Retrieved 12 October 2017 from https://www.cdc.gov/physicalactivity/basics/older_adults/Chastin S, Gardiner PA, Ashe MC, Harvey JA, Leask CF, Balogun S, Helbostad JL and Skelton DA (2017) Interventions for reducing sedentary behaviour in community‐dwelling older adults. The Cochrane Library 9, 1–13.Chudyk AM, McKay HA, Winters M, Sims-Gould J and Ashe MC (2017) Neighborhood walkability, physical activity, and walking for transportation: a cross-sectional study of older adults living on low income. BMC Geriatrics 17, 82–95.Costello E, Kafchinski M, Vrazel J and Sullivan P (2011) Motivators, barriers, and beliefs regarding physical activity in an older adult population. Journal of Geriatric Physical Therapy 34, 138–147.Department of Health (2011) Physical activity guidelines in the UK: review and recommendations. Retrieved 9 October 2017 from https://www.gov.uk/ government/uploads/system/uploads/attachment_data/file/213743/dh_128255. pdfDevereux-Fitzgerald A, Powell R, Dewhurst A and French DP (2016) The acceptability of PA interventions to older adults: a systematic review and meta-synthesis. Social Science & Medicine 158, 14–23.Edwards MK and Loprinzi PD (2016) The association between sedentary behavior and cognitive function among older adults may be attenuated with adequate physical activity. Journal of Physical Activity and Health 14, 52–58.Emdadi S, Hazavehie SMM, Soltanian A, Bashirian S and Heidari Moghadam R (2015) Predictive factors of regular physical activity among middle-aged women in the West of Iran, Hamadan: application of PRECEDE model. Journal of Research in Health Sciences 15, 244–249.


9


Falck RS, Davis JC and Liu-Ambrose T (2016) What is the association between sedentary behaviour and cognitive function? A systematic review. British Journal of Sports Medicine 51, 800–811.Fisher KL, Harrison EL, Bruner BG, Lawson JA, Reeder BA, Ashworth NL, Sheppard MS and Chad KE (2017) Predictors of physical activity levels in community dwelling older adults: a multivariate approach based on a socio-ecological framework. Journal of Aging and Physical Activity 26, 1–23.Franco MR, Tong A, Howard K, Sherrington C, Ferreira PH, Pinto RZ and Ferreira ML (2015) Older people’s perspectives on participation in physical activity: a systematic review and thematic synthesis of qualitative literature. British Journal of Sports Medicine 49, 1268–1276.French DP, Olander EK, Chisholm A and McSharry J (2015) Which behaviour change techniques are most effective at increasing older adults’ self-efficacy and physical activity behaviour? A systematic review. Annals of Behavioural Medicine 48, 225–234.Gagliardi AR, Faulkner G, Ciliska D and Hicks A (2015) Factors contributing to the effectiveness of physical activity counselling in primary care: a realist systematic review. Patient Education and Counseling 98, 412–419.Garcia JA, Schoene D, Lord SR, Delbaere K, Valenzuela T and Navarro KF (2016) A bespoke Kinect stepping exergame for improving physical and cognitive function in older people: a pilot study. Games for Health Journal 5, 382–388.Gardiner C, Geldenhuys G and Gott M (2016) Interventions to reduce social isolation and loneliness among older people: an integrative review. Health & Social Care in the Community 26, 1–17.Gillespie LD, Robertson MC, Gillespie WJ, Sherrington C, Gates S, Clemson LM and Lamb SE (2012) Interventions for preventing falls in older people living in the community. Cochrane Handbook for Systematic Reviews of Interventions 9, 1–420.Glasgow RE, Green LW, Taylor MV and Stange KC (2012) An evidence integration triangle for aligning science with policy and practice. American Journal of Preventive Medicine 42, 646–654.Gray PM, Murphy MH, Gallagher AM and Simpson EE (2015) Motives and barriers to physical activity among older adults of different socioeconomic status. Journal of Aging & Physical Activity 24, 419–429.Greaney ML, Lees FD, Blissmer BJ, Riebe D and Clark PG (2016) Psychosocial factors associated with physical activity in older adults. Annual Review of Gerontology and Geriatrics 36, 273–291.Green LW and Kreuter MW (2005) Health Program Planning: An Educational and

Ecological Approach, 4th edn. New York: McGraw-Hill.Hamer M, Lavoie KL and Bacon SL (2013) Taking up physical activity in later life and healthy ageing: the English longitudinal study of ageing. British Journal of Sports Medicine 48, 239–243.Haselwandter EM, Corcoran MP, Folta SC, Hyatt R, Fenton M and Nelson ME (2015) The built environment, physical activity, and aging in the United States: a state of the science review. Journal of Aging and Physical Activity 23, 323–329.Hawley-Hague H, Horne M, Skelton DA and Todd C (2016) Older adults’ uptake and adherence to exercise classes: instructors’ perspectives. Journal of Aging and Physical Activity 24, 119–128.Healy GN, Winkler EA, Owen N, Anuradha S and Dunstan DW (2015) Replacing sitting time with standing or stepping: associations with cardiometabolic risk biomarkers. European Heart Journal 36, 2643–2649.Heo J, Chun S, Kim B, Ryu J and Lee Y (2017) Leisure activities, optimism, and personal growth among the young-old, old-old, and oldest-old. Educational Gerontology 43, 289–299.Hildebrand M and Neufeld P (2009) Recruiting older adults into a physical activity promotion program: active living every day offered in a naturally occurring retirement community. The Gerontologist 49, 702–710.Hoare E, Stavreski B, Jennings GL and Kingwell BA (2017) Exploring motivation and barriers to physical activity among active and inactive Australian adults. Sports 5, 47–54.Hoppmann CA, Lee JCM, Ziegelmann JP, Graf P, Khan KM and Ashe MC (2015) Precipitation and physical activity in older adults: the moderating role of functional mobility and physical activity intentions. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 72, 792–800.

Jefferis BJ, Sartini C, Lee IM, Choi M, Amuzu A, Gutierrez C, Casas JP, Ash S, Lennnon LT, Wannamethee SG and Whincup PH (2014) Adherence to physical activity guidelines in older adults, using objectively measured physical activity in a population-based study. BMC Public Health 14, 1–9.

Keadle SK, McKinnon R, Graubard BI and Troiano RP (2016) Prevalence and trends in physical activity among older adults in the United States: a comparison across three national surveys. Preventive Medicine 89, 37–43. Kim J, Im JS and

Choi YH (2016) Objectively measured sedentary behavior and moderate-to-vigorous physical activity on the health-related quality of life in US adults: the

National Health and Nutrition Examination Survey2003–2006. Quality of Life Research 26, 1315–1326.Knowles ZR, Parnell D, Stratton G and Ridgers ND (2013) Learning from the experts: exploring playground experience and activities using a write and draw technique. Journal of Physical Activity & Health 10, 406–415.Kosteli MC, Williams SE and Cumming J (2016) Investigating the psychosocial determinants of physical activity in older adults: a qualitative approach. Psychology & Health 31, 730–749.Lewis BA, Napolitano MA, Buman MP, Williams DM and Nigg CR (2017) Future directions in physical activity intervention research: expanding our focus to sedentary behaviors, technology, and dissemination. Journal of Behavioral Medicine 40, 112–126.Mackintosh KA, Knowles ZR, Ridgers ND and Fairclough SJ (2011) Using formative research to develop change!: a curriculum-based physical activity promoting intervention. BMC Public Health 11, 831–843.McMahon SK, Lewis B, Oakes JM, Wyman JF, Guan W and Rothman AJ (2017) Assessing the effects of interpersonal and intrapersonal behavior change strategies on physical activity in older adults: a factorial experiment. Annals of Behavioral Medicine 51, 376–390.

Middelweerd A, Mollee JS, van der Wal CN, Brug J and te Velde SJ (2014) Apps to promote physical activity among adults: a review and content analysis.

International Journal of Behavioral Nutrition and Physical Activity 11, 97–106. Milligan C, Payne S, Bingley A and Cockshott Z (2015) Place and wellbeing:

shedding light on activity interventions for older men. Ageing and Society35, 124–149.Ministry of Health (2013) Guidelines on physical activity for older people (aged 65 years and over). Retrieved 9 October 2017 from https://www. health.govt.nz/system/files/documents/publications/guidelines-on-physical- activity-older-people-jan13-v3.pdfMoran F, MacMillan F, Smith-Merry J, Kilbreath S and Merom D (2015) Perceived barriers, facilitators and patterns of physical activity of older-old adults living in assisted retirement accommodation. Journal of Gerontology & Geriatric Research 4, 1–6.Newitt R, Barnett F and Crowe M (2016) Understanding factors that influence participation in physical activity among people with a neuromusculoskeletal condition: a review of qualitative studies. Disability and Rehabilitation 38, 1–10.Nicholson NR (2012) A review of social isolation: an important but underassessed condition in older adults. The Journal of Primary Prevention 33, 137–152.Noonan RJ, Boddy LM, Fairclough SJ and Knowles ZR (2016a) Parental perceptions on childrens out-of-school physical activity and family-based physical activity. Early Child Development and Care 187, 1–16. https://doi. org/10.1080/03004430.2016.1194409Noonan RJ, Boddy LM, Fairclough SJ and Knowles ZR (2016b) Write, draw, show, and tell: a child-centred dual methodology to explore perceptions of out-of-school physical activity. BMC Public Health 16, 326–344.Notthoff N, Klomp P, Doerwald F and Scheibe S (2016) Positive messages enhance older adults’ motivation and recognition memory for physical activity programmes. European Journal of Ageing 13, 1–7.Orr N and Phoenix C (2015) Photographing physical activity: using visual methods to ‘grasp at’ the sensual experiences of the ageing body. Qualitative Research 15, 454–472.Owen N, Healy GN, Matthews CE and Dunstan DW (2010) Too much sitting: the population-health science of sedentary behavior. Exercise and Sport Sciences Reviews 38, 105–113.Petrescu-Prahova M, Belza B, Kohn M and Miyawaki C (2015) Implementation and maintenance of a community-based older adult physical activity program. The Gerontologist 56, 1–1


Phoenix C and Tulle E (2017) Physical activity and ageing. . In Piggin J, Mansfield L and Weed M, editors. The Routledge handbook of physical activity policy and practice. London: Routledge, 1–18.Plotnikoff RC, Lubans DR, Penfold CM and Courneya KS (2014) Testing the utility of three social‐cognitive models for predicting objective and self‐ report physical activity in adults with type 2 diabetes. British Journal of Health Psychology 19, 329–346.Prince SA, Saunders TJ, Gresty K and Reid RD (2014) A comparison of the effectiveness of physical activity and sedentary behaviour interventions in reducing sedentary time in adults: a systematic review and meta-analysis of controlled trials. Obesity Reviews 15, 905–919.Prins RG and van Lenthe FJ (2015) The hour-to-hour influence of weather conditions on walking and cycling among Dutch older adults. Age & Ageing 44, 1–5.QSR International Pty Ltd (2017) NVIVO: Version 11. Reference guide. Doncaster Victoria, Australia: Author.Reid H and Foster C (2016) Infographic. Physical activity benefits for adults and older adults. British Journal of Sports Medicine 51, 1441–1442.Rejeski WJ, Axtell R, Fielding R, Katula J, King AC, Manini TM, Marsh AP, Pahor M, Rego A, Tudor-Locke C and Newman M (2013) Promoting physical activity for elders with compromised function: the lifestyle interventions and independence for elders (LIFE) study physical activity intervention. Clinical Interventions in Aging 8, 1119–1131.Schneider JL, Goddard KA, Davis J, Wilfond B, Kauffman TL, Reiss JA, Gilmore M, Himes P, Lynch FL, Leo MC and McMullen C (2016) Is it worth knowing? ‘Focus group participants’ perceived utility of genomic preconception carrier screening. Journal of Genetic Counseling 25, 135–145. Shaw RJ, Čukić I, Deary IJ, Gale CR, Chastin SF, Dall PM, Skelton DA and Der G (2017) Relationships between socioeconomic position and objectively measured sedentary behaviour in older adults in threeprospective cohorts. BMJ Open 7, 1–11.Skjæret N, Nawaz A, Morat T, Schoene D, Helbostad JL and Vereijken B (2016) Exercise and rehabilitation delivered through exergames in older adults: an integrative review of technologies, safety and efficacy. Interna- tional Journal of Medical Informatics 85, 1–16.Smith B and Caddick N (2012) Qualitative methods in sport: a concise overview for guiding social scientific sport research. Asia Pacific Journal of Sport and Social Science 1, 60–73.Smith GL, Banting L, Eime R, O’Sullivan G and van Uffelen JG (2017) The association between social support and physical activity in older adults: a systematic review. International Journal of Behavioral Nutrition and Physical Activity 14, 56–77.Sport England (2012) GET HEALTHY, GET ACTIVE: learn more about our initial investments into tackling inactivity from 2012-2016. Retrieved 9 October 2017 from https://www.sportengland.org/our-work/health-and- inactivity/get-healthy-get-active/Sport England (2015) Local Profile Tool. Retrieved 9 October 2017 from http://www.sportengland.org/our-work/local-work/local-government/local- sport-profile/Stewart DW and Shamdasani PN (2014) Focus Groups: Theory and Practice. Volume 20, Newbury Park: Sage Publications.

Susan J, Mallan K, Callaway L, Daniels LA and Nicholson JM (2017) A cross sectional comparison of predisposing, reinforcing and enabling factors for lifestyle health behaviours and weight gain in healthy and overweight pregnant women. Maternal and Child Health Journal 21, 626–635.Träff AM, Cedersund E and Nord C (2017) Perceptions of physical activity among elderly residents and professionals in assisted living facilities. European Review of Aging and Physical Activity 14, 2–11.Tremblay MS, Aubert S, Barnes JD, Saunders TJ, Carson V, Latimer- Cheung AE, Chastin SF, Altenburg TM and Chinapaw MJM (2017) Sedentary behavior research network (SBRN) – terminology consensus project process and outcome. International Journal of Behavioral Nutrition and Physical Activity 14, 75–92.Tucker JM, Welk GJ and Beyler NK (2011) Physical activity in US adults: compliance with the physical activity guidelines for Americans. American Journal of Preventive Medicine 40, 454–461.UK Office for National Statistics (2017) Mid-2016 population estimates. Retrieved 9 October 2017 from https://www.ons.gov.uk/peoplepopulation andcommunity/populationandmigration/populationestimates/bulletins/ annualmidyearpopulationestimates/latestUnited Nations (2015) World population ageing report. Retrieved 9 October 2017 from http://www.un.org/en/development/desa/population/publications/pdf/ageing/WPA2015_Report.pdfVan Cauwenberg J, De Bourdeaudhuij I, Clarys P, Nasar J, Salmon J, Goubert L and Deforche B (2016) Street characteristics preferred for transportation walking among older adults: a choice-based conjoint analysis with manipulated photographs. International Journal of Behavioral Nutrition and Physical Activity 13, 6–22.Van Dyck D, Mertens L, Cardon G, De Cocker K and De Bourdeaudhuij I (2017) Opinions toward physical activity, sedentary behavior, and interventions to stimulate active living during early retirement: a qualitative study in recently retired adults. Journal of Aging and Physical Activity 25, 277–286.van Schijndel-Speet M, Evenhuis HM, van Wijck R, van Empelen P and Echteld MA (2014) Facilitators and barriers to physical activity as perceived by older adults with intellectual disability. Mental Retardation 52, 175–186.Warmoth K, Lang IA, Phoenix C, Abraham C, Andrew MK, Hubbard RE and Tarrant M (2016) ‘Thinking you’re old and frail’: a qualitative study of frailty in older adults. Ageing & Society 36, 1483–1500.Wilmot EG, Edwardson CL, Achana FA, Davies MJ, Gorely T, Gray LJ, Khunti K, Yates T and Biddle SJ (2012) Sedentary time in adults and the association with diabetes, cardiovascular disease and death: systematic review and meta-analysis. Diabetologia 55, 2895–2905.World Health Organization (WHO) (2017) Global strategy on diet, physical activity and health: physical activity and older adults. Retrieved 12 October 2017 from http://www.who.int/dietphysicalactivity/factsheet_olderadults/en/Wu E, Barnes DE, Ackerman SL, Lee J, Chesney M and Mehling WE (2015) Preventing loss of independence through exercise (PLIÉ): qualitative analysis of a clinical trial in older adults with dementia. Aging & Mental Health 19, 353–362.

repository.edgehill.ac.uk Sanders 23103400... · Web viewPrimary Health Care Research &...

Documents

Transcript of repository.edgehill.ac.uk Sanders 23103400... · Web viewPrimary Health Care Research &...