PREVENTION OF LINEAR GROWTH FALTERING AND REVERSAL OF STUNTING AMONG LOW BIRTH WEIGHT INFANTS IN RURAL BANGLADESH:

A 2X2 FACTORIAL CLUSTER RANDOMIZED TRIAL

By Sohana Shafique

Advisor: Stanley H. Zlotkin, CM, MD, PhD, FRCPC

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Graduate Department of Nutritional Sciences University of Toronto

© Copyright by Sohana Shafique 2013

ii  

PREVENTION OF LINEAR GROWTH FALTERING AND REVERSAL OF STUNTING AMONG LOW BIRTH WEIGHT INFANTS IN RURAL BANGLADESH:

A 2X2 FACTORIAL CLUSTER RANDOMIZED TRIAL  

PhD, 2013 Sohana Shafique

Graduate Department of Nutritional Sciences University of Toronto

Abstract

Background: Low birth weight (LBW) infants in low income countries are vulnerable to

frequent infections and suboptimal breastfeeding and complementary feeding practices, which

together may result in postnatal linear growth faltering and poor neurodevelopment. Objectives:

To measure the relative effect of directed use of benzalkonium chloride-containing, water-based

hand sanitizers (HS) and broad-range multiple micronutrient powder (MNP) along with nutrition,

health and hygiene education (NHHE) to prevent infections and linear growth faltering among

LBW infants. Methods: A prospective 2X2 factorial, cluster-randomized controlled trial was

conducted among 467 full-term LBW infants from 0-12 months, using 48 clusters randomly

assigned as follows: A. From 0 to 6 months, i) NHHE alone or ii) NHHE plus HS; B. From 6-12

months i) NHHE alone; ii) NHHE plus HS; iii) NHHE plus MNP (to be provided with

complementary foods); and iv) NHHE plus both HS and MNP. Results: Combined intervention

of ‘directed hand-sanitizer use’ and ‘micronutrient powder’ along with ‘NHHE’ significantly

improved linear growth of low birth weight infants compared to ‘NHHE’ alone. At the end of

12-month intervention, the ‘NHHE+HS+MNP’ and ‘NHHE+MNP’ groups had significantly

lower prevalence of moderate and severe stunting (length-for-age <−2 Z-score) compared to the

‘NHHE only’ group, (P<0.01, for both). A marked increase in linear growth was found only

during the second half of infancy (from 6-12 months) through provision of MNP with

iii  

complementary foods. Mean duration of diarrhea episodes during infancy (0-6 mo) was

significantly lower in the ‘HS plus NHHE’ group compared to the ‘NHHE alone’ group (3.1±1.2

vs. 5.8±1.6 days, P<0.01). Upper respiratory tract infections (URTI) and cough were also

significantly lower in the ‘HS plus NHHE’ group compared to the ‘NHHE alone’ group (22.5%

vs. 27.1% and 8.1% vs. 11.2%, P<0.05, for both). Although HS independently reduced

infections, no impact of HS on linear growth was found in the first six months.

Conclusions: The use of hand sanitizer in combination with MNP reduced infections and

impacted on rates of stunting among term LBW infants in the first year of life.

Implications: Further evidence is needed to confirm the beneficial effects of hand sanitizers and

MNPs among other populations of infants.

Trial registration: http://www.clinicaltrials.gov (identifier NCT01455636)

Key Words: Low birth weight, linear growth, stunting, infection, hand hygiene

iv  

STUDENT’S CONTRIBUTION TO THE THESIS

In collaboration with my supervisor Dr. Stanley Zlotkin, I prepared the research protocol,

submitted applications to the research ethics committees in Toronto, Canada (SickKids) and

Bangladesh, did a first draft of a grant application to the Alive and thrive Program of the Gates

Foundation (successful), and in addition, played a significant management role in Bangladesh

with a MSc project by one of Dr. Zlotkin’s former students, Waqas Khan. I am a co-author of the

manuscript that was recently accepted (May 28, 2013) for publication in Public Health Nutrition

(Home-fortification with calcium reduces hemoglobin response to iron among anemic

Bangladeshi children consuming a new multi-micronutrient powder (MNP) formulation). I also

managed the study in Bangladesh, which included the hiring of staff, preparation of SOPs,

training of intervention and assessment staff, and supervised data collection and data

management activities. I completed the statistical analysis of the data from the study and of

course wrote my thesis.

v  

ACKNOWLEDGEMENTS

This project would not have been possible without the personal and professional support and

encouragement I received from countless friends, family, and colleagues. First and foremost, I

would like to thank my dissertation supervisor Dr. Stanley Zlotkin for his intellectual and

emotional support through this journey. I’m extremely grateful for his attentive, thoughtful and

expert guidance and constant trust and confidence in my abilities and for allowing me to work

independently. I thank him for providing a role model for scientific dedication to the expansion

of knowledge benefitting children globally. I want to express that it has been an invaluable

experience to have Stan as a guide, who I not only admire, but trust as a colleague and a friend.

I would like to give special thanks to my excellent co-supervisor Dr. Daniel W. Sellen for his

keen methodological advice, commitment and continuous encouragement. I am very grateful to

the other members of my advisory committee, Dr. Debbie O’Connor, Dr. Carol Greenwood and

Dr. Wendy Lou for their valuable suggestions and encouragement. I offer my deep appreciation

to each one of them. I would also like to thank the examiners that attended my defenses, Dr.

Christopher Tomlinson, who acted as the internal examiner at my departmental exam and Dr.

Stephanie Atkinson, who was the external examiner at my SGS exam. I wish to express gratitude

for their excellent suggestions and insights into my research, providing critical feedback as well

as challenging my thinking that made me a confident person.

I would like to give special thanks to our study collaborators at Research Evaluation Division

(RED) at BRAC, Bangladesh. I extend my sincere thanks to Dr. Ziauddin Hyder, Dr. Chowdhury

S.B. Jalal and my colleagues Hasina Shikder and Saira Parvin Jolly for their tremendous support

to this study. I would like to thank our community health workers, field interviewers, field

managers, and data entry personnel for their hard work and commitment to the project. Thank

you also to Eshetu Atenafu for his help with statistical analysis. Special thanks to the infants,

their mothers and the families for their participation in the research.

I would like to thank members of Alive and Thrive Technical Advisory Group Dr. Kathryn

Dewey and Dr. Rukhsana Haider, who provided insightful comments to the study design; and

Dr. Zulfiqar Bhutta, who visited the study sites in Bangladesh and gave valuable feedback and

vi  

encouragement. I would like to extend my thanks to Dr. Frances Aboud and Dr. Charles Larson

for their intellectual support and most importantly, for being excellent mentors.

I am appreciative of colleagues at the Hospital for Sick Children: especially Brenda Hartman, Jo-

Anna Baxter, Waqas Khan and members of The Global Child Health group (especially Ashley

Aimone, Dr. Daniel Roth and Dr. Diego Bassani); and the Sellen laboratory group.

I also wish to acknowledge the generous funding support I received in pursuing my studies and

research: Connaught Scholarship, International Development Research Centre (IDRC) Doctoral

Research Award, School of Graduate Studies Doctoral Completion Award and bursary from

Child Health and Evaluative Sciences, Hospital for Sick Children. Special thanks to Alive and

Thrive Small Grants Program, an initiative of Bill and Melinda Gates Foundation for providing

financial support, interest in and exposure to my work. A major financial assistance for fieldwork

also came from the Heinz Foundation.

Thank you to my family and friends around the world, who encouraged and inspired me to hang

in there during this critical period of my life. I would especially like to mention my dear friends

Walid Hossain, Sadika Akhter, Kerrie Proulx, Bright Drah and Tibra Ali for their love and

support along the way. Thank you to Imon, who always believed in me and was always there

when I needed him most. Finally, I owe my most profound gratitude to my lovely parents, who

raised me to love reading and learning about this world, and gave unconditional love and support

to become the person I am today.

vii  

TABLE OF CONTENTS

CHAPTER 1 INTRODUCTION…………………..………………………....………..1 CHAPTER 2 LITERATURE REVIEW…………………………………………..….4

2.1. Global epidemiology of stunting………………………………………………………...4 2.1.1. Definition of stunting and linear growth faltering 2.1.2. Trends of stunting 2.1.3. Social and economic cost of stunting 2.1.4. Timing of linear growth faltering 2.1.5. Early interventions to prevent stunting

2.2. Epidemiology of low birth weight (LBW)………………………………………………7

2.2.1. Definition and prevalence of low birth weight 2.2.2. Intrauterine growth restriction and risk of mortality 2.2.3. LBW and risk of stunting 2.2.4. LBW and risk of chronic diseases in adulthood

2.3. Risk factors associated with stunting……………………………………………………9

2.3.1. Sub-optimum infant and young child feeding practices 2.3.2. Lack of breastfeeding 2.3.3. Poor complementary feeding and micronutrient deficiencies 2.3.4. Infectious diseases

2.4. Current evidence for interventions to prevent and reduce stunting…………………….12

2.4.1. Nutrition and health education

2.5. Micronutrients to prevent stunting……………………………………………………...14 2.5.1. Multiple micronutrients (MMN) and linear growth 2.5.2. Home fortification of complementary foods 2.5.3. The potential of improved micronutrient powders to enhance growth

2.6. Current evidence for hand hygiene to prevent infections……………………………... 19

2.6.1. Handwashing to prevent infectious diseases in developing countries 2.6.2. Limitations of handwashing with soap and water 2.6.3. Efficacy of alcohol-based hand sanitizers and inherent disadvantages 2.6.4. Alcohol-free hand sanitizers

2.7. Infant growth monitoring………………………………………………………………22

2.7.1. The development of WHO growth standard 2.7.2. Multicentre growth reference study

2.8. Assessing nutritional status…………………………………………………………… 24

2.8.1. Anthropometry 2.8.2. Anthropometric indices

viii  

2.8.2.1. Length-for-age 2.8.2.2. Weight-for-age 2.8.2.3. Weight-for-length

2.8.3. Evaluation of anthropometric indices using Z-scores 2.8.4. Composite measures from anthropometric indices

2.8.4.1. Stunting 2.8.4.2. Underweight 2.8.4.3. Wasting

2.9. Child undernutrition in the context of Bangladesh…………………………………… 29

2.9.1. Prevalence of low birth weight and stunting in Bangladesh 2.9.2. Infant and young child feeding practices in Bangladesh

2.10. Conclusions

CHAPTER 3 RESEARCH RATIONALE AND OBJECTIVES ……………..……40 3.1 Rationale of the proposed research 3.2 Goals and Research Question 3.3 Objectives

3.3.1 Primary objective: 3.3.2 Secondary objectives

CHAPTER 4 METHODS……………………………..………………………..……..44 4.1 Overview of study design 4.2 Study participants 4.3 Sample size 4.4 Study setting 4.5 Ethical oversight 4.6 Sampling and randomization 4.7 Intervention training 4.8 Preparation for safe delivery, birth notification and community mobilization 4.9 Screening and enrolment 4.10 Intervention components

4.10.1 Directed hand-hygiene with hand sanitizers (HS) 4.10.2 Improved micronutrient powders (MNP) 4.10.3 Nutrition, health and hygiene education (NHHE)

4.11 Intervention management and monitoring 4.12 Outcome measures

4.12.1 Anthropometry 4.12.2 Morbidity

ix  

4.12.3 Infant and young child feeding

4.13 Compliance and acceptability of intervention 4.14 Quality control and data management 4.15 Data analysis plan and statistical approach 4.16 Benefits to the control communities CHAPTER 5 PAPER 1: ……………………….…………..…………………..……. 63

The effect of a water-based hand sanitizer to reduce neonatal infections and improve linear growth among low birth weight infants in the first month of life: a community-based cluster randomized trial in rural Bangladesh (Running head: Effect of HS on neonatal infections)

Abstract 5.1 Introduction 5.2 Methods 5.3 Results 5.4 Discussion CHAPTER 6 PAPER 2: ……………………….…………..…………….…………. 87

The effect of a water-based hand sanitizer to reduce infections and improve linear growth in full-term low birth weight infants in the first six months of life: a community-based cluster randomized trial in rural Bangladesh (Running head: Effect of HS on growth in 0-6mo)

Abstract 6.1 Introduction 6.2 Methods 6.3 Results 6.4 Discussion CHAPTER 7 PAPER 3: ……………………….…………..……………………….115

Relative effect of a water-based hand sanitizer and micronutrient powder to improve linear growth among low birth weight infants in the first year of life: a community-based cluster randomized trial in rural Bangladesh

(Running head: Effect of HS and MNP on growth in 0-12mo)

Abstract 7.1 Introduction 7.2 Methods

x  

7.3 Results 7.4 Discussion CHAPTER 8 SUMMARY AND CONCLUSIONS………………………………155 8.1 Key Findings and Potential Significance 8.2 Strengths of Current study 8.3 Limitations of the current study 8.4 Implications for Policy 8.5 Future Research Directions 8.6 Conclusions LITERATURE CITED.……………………….…..………………………………………..165 APPENDIX ……………………….…………..……………………………………..…….188

xi  

LIST OF TABLES Table 2.1. Summary of studies investigating effect of multiple micronutrients on linear

growth. Table 4.1. Critical points for hand washing and using hand sanitizers Table 4.2. Composition of improved Micronutrient Powder (MNP) Table 5.1. Maternal, demographic and household socioeconomic characteristics by study

group at enrolment

Table 5.2. Comparison of newborn care practices among low birth weight infants in the neonatal period (0-28 days) by study group

Table 5.3. Comparison of complications among low birth weight infants in the neonatal

period (0-28 days) by study group Table 5.4. Anthropometry of low birth weight infants at birth by study group

Table 6.1. Comparison of selected infant, maternal, socioeconomic characteristics by

intervention groups at enrolment

Table 6.2. Anthropometric indices of low birth weight infants at birth by study group at birth, 3 and 6 months of age

Table 6.3. Nutritional status of low birth weight infants by study group at the end of 6

months

Table 7.1. Comparison of selected infant, maternal, socioeconomic characteristics by

intervention groups at enrolment Table 7.2. Anthropometric values of low birth weight infants at 0, 6 and 12 months of age by

4 intervention groups

Table 7.3. Morbidity of low birth weight infants at 6, 9 and 12 months by intervention groups

xii  

LIST OF FIGURES Figure 2.1. Worldwide timing for growth faltering.

Figure 2.2. Malnutrition and infection cycle.

Figure 2.3. Conceptual framework for the relationship of specific child care behaviors and infant health outcomes.

Figure 2.4. Etiological relationships between various biological outcomes of child development.

Figure 4.1. Study design: 2X2 factorial cluster randomized controlled trial

Figure 4.2. Study sites: two adjacent sub-districts, Kaliganj in Gazipur; and Palash in Narsingdi, in Dhaka division, Bangladesh

Figure 4.3. Chemical structure of benzalkonium chloride

Figure 5.1. Trial profile of the study (0-28 days) Figure 5.2. Point prevalence of reported morbidity in the neonatal period (0-28 days) by

week by study group Figure 5.3. Household expenditure on health care seeking (Mean, SD) for the low birth

weight inafnts in the neonatal period (0-28 days) by study group Figure 6.1. Trial profile of the study (0-6mo)

Figure 6.2. Length-for-age Z-score, weight-for-age Z-score and weight-for-length Z-score of low birth weight infants from 0-6 months by intervention groups

Figure 6.3.1 Length-for-age Z-score of low birth weight infants from 0-6 months by gender

Figure 6.3.2 Weight-for-age Z-score of low birth weight infants from 0-6 months by gender

Figure 6.4. Proportion of low birth weight infants who had any reported illness in the previous week during first six months

Figure 6.5. Proportion of low birth weight infants with upper respiratory tract infections (URTI) and cough in the previous week during first six months.

Figure 6.6. Mean (SD) duration of diarrhea (in days) among low birth weight infants in Bangladesh during the first 6 months

Figure 7.1. Trial profile of the study (0-12mo)

xiii  

Figure 7.2. Unadjusted mean length-for-age Z-score (LAZ), weight-for-age Z-score (WAZ), and weight-for-length Z-score (WLZ) of all study participants from 0-12 months1,2

Figure 7.3.1 Longitudinal changes in length-for-age Z-score (LAZ) of low birth weight infants

from 0-12 months by study group

Figure 7.3.2 Longitudinal changes in weight-for-age Z-score (WAZ) of low birth weight infants from 0-12 months by study group

Figure 7.4. Proportion of low birth weight infants with severe (length-for-age Z-score <−3) and moderate stunting (length-for-age Z-score ≥ −3 and <−2) at the end of 12 months by study group

Figure 7.5.1. Cumulative incidence of recovery from moderate stunting (length-for-age Z-score

≥ −3 and <−2) from 0-12 months

Figure 7.5.2 Cumulative incidence of recovery from severe stunting (length-for-age Z-score <−3) from 0-12 months

xiv  

LIST OF APPENDICES Appendix 1 The World Health Organization (WHO) growth charts for length-for-age

for boys. Appendix 2 The World Health Organization (WHO) growth charts for length-for-age

for girls. Appendix 3 The World Health Organization (WHO) growth charts for weight-for-age

for boys. Appendix 4 The World Health Organization (WHO) growth charts for weight-for-age

for girls.

Appendix 5 Consent form of the study. Appendix 6 Ethics approval for the study from Hospital for Sick Children (Canada).

Appendix 7 Ethics approval for the study from JPG School of Public Health BRAC University (Bangladesh).

Appendix 8 Danger signs associated with neonatal complications (Courtesy: Maternal, Neonatal and Child Health (MNCH) Programme, BRAC)

Appendix 9 Study Questionnaires

xv  

LIST OF ABBREVIATIONS ARI Acute respiratory tract infections

EB Exclusive breastfeeding

EDD Expected-date of delivery

ENC Essential newborn care

FE Field enumerators

Hb Hemoglobin

HNA Health and Nutrition Promoters

HNP Health and Nutrition Assistants

HS Hand sanitizers

ICC Intra-cluster correlation

KMC Kangaroo mother care

LBW Low birth weight

LMP Last menstrual period

MNPs Micronutrient powders

MGRS Multicentre Growth Reference Study

NHHE Nutrition, health and hygiene education

ORS Oral rehydration solution

SES Socio-economic status

SGA Small for gestational age

URTI Upper respiratory tract infections

1

CHAPTER 1

INTRODUCTION

Investing in nutrition has high potential for child survival and ensuring better growth and

development for those who survive. It is evident that the first two years of life is the time of

greatest vulnerability and risk of irreversible long-term physical and developmental damage.

Providing nutrition interventions during infancy is critical, when nutrient needs are highest and

the ability of the child to respond to a nutrition intervention is greatest. It is believed that the

major contributing factors for post-natal linear growth faltering are sub-optimal breastfeeding,

inadequate and inappropriate complementary feeding as well as micronutrient deficiencies that

result in suppressed immune system among children. Infections are associated with diminished

food intake, impaired nutrient absorption, cause direct nutrient loss, increase metabolic

requirements or catabolic losses of nutrients and possibly impair the transport of nutrients to

target tissues. Episodes of infectious diseases, such as diarrhea and respiratory illness likely play

a pivotal role in length growth faltering.

The global prevalence of low birth weight is 15.5% and 96.5% of them are born in developing

countries. South Asian countries have very high incidence of low birth weight infants with the

highest rate in Bangladesh (36%). These infants are more vulnerable to infectious morbidity and

mortality. Although LBW infants are shorter at birth compared to normal birth weight babies,

unfortunately they share similar patterns of linear growth faltering at around 3 months of age.

This suggests that growth faltering in the postnatal period is primarily caused by environmental

factors.

2

While impact of optimal infant and young child feeding (IYCF) practices on improving child

nutrition and survival are well documented, questions on how infections and sub-optimal feeding

alter patterns of early post-partum linear growth of low birth weight infants remain to be

answered. Exclusive breastfeeding is appropriately recommended for the first six months of life,

thus early food based interventions to influence linear growth faltering are not possible. During

this same period, rates of infection are very high, thus, there is an urgent need for non food-based

interventions in the first 6 months of life to break the malnutrition-infection cycle.

An innovative and central feature of the research described in this thesis is the use of multiple

preventive interventions to deal with growth faltering and stunting. We hypothesized that linear

growth faltering among LBW infants could be averted through the early provision of nutrition

education combined with directed hand hygiene to caregivers, and later fortification of

complementary food with an improved micronutrient (mineral and vitamin) powder (MNP)

formulation.

A 2x2 factorial community-based cluster randomized control trial design was used to investigate

the relative effects of using water-based hand sanitizer and improved micronutrient powder to

prevent linear growth faltering among 467 term-LBW infants in rural Bangladesh. The literature

review and rationale of the study is presented in Chapter 2 and Chapter 3, respectively. Detailed

description of the study design and methods are presented in Chapter 4.

Results from this study are presented in three distinct but interrelated papers. The first paper

(Chapter 5) describes impact of using hand sanitizer to reduce infections during the neonatal

3

period (0-28 days), an extremely critical period for low birth weight infants. The second paper

(Chapter 6) examines the impact of hand sanitizer in preventing linear growth faltering during

the critical 0-6 month. Finally, the third paper (Chapter 7) relative effect of micronutrient

powders and hand sanitizers on linear growth of low birth weight infants in the first year of life

compared with the international growth standards.

Interpretation and discussion of the results of these studies provides useful information on the

biological basis for altered patterns of linear growth and also provide greater insights into

strategies to optimize growth and survival in this high-risk population.

4

CHAPTER 2

LITERATURE REVIEW

2.1 Global epidemiology of stunting

2.1.1 Definition of stunting and linear growth faltering

Child undernutrition remains as one of the major public health challenges of the 21st century,

particularly in low- and middle-income countries (Black et al., 2008; de Onis et al., 2012;

Stevens et al., 2012; Bhutta and Salam, 2012). Undernutrition is expressed in various forms,

including stunting, wasting, underweight and deficiencies of essential vitamins and minerals

(Waterlow, 1973). Stunting, wasting, and intrauterine growth restriction are responsible for 2.2

million deaths and 21% of disability-adjusted life years (DALYs) lost among children younger

than 5 years (Morris et al., 2008). Stunting is defined as a chronic restriction of growth in

length/height indicated by short stature among children (Black et al., 2008). In low and middle

income countries, stunting remains as the most invariable characteristic of children less than five

years (Bhutta and Salam, 2012). Linear growth faltering is the process by which length-for-age

(LAZ) of a child starts to deviate from the international growth standards and if LAZ reaches

below -2 Z-scores, it is called stunting (de Onis, 2006). Therefore, preventing linear growth

faltering is the crucial first step to reduce the rates of stunting.

 

2.1.2 Trends of stunting

Stunting prevalence is a critical indicator of progress in child survival, reflecting long-term

exposure to poor health and nutrition, especially in the first two years of life (Victora et al.,

2010). All children around the world have the same growth potential, and prevalence of stunting

above the 3% threshold in a country indicates the need for immediate actions (Bhutta et al.,

5

2010). Globally, an estimated 178 million under-five children are stunted with majority of

children in Sub-Saharan Africa and South-Central Asia (Black et al., 2008; de Onis, 2013). An

analysis of 61 countries that have >3% stunting showed that stunting is particularly high among

the poorest populations (Bhutta et al., 2010). In the majority of these countries, more than 1 in 3

children are stunted. A recent analysis shows that in developing countries, prevalence of

moderate-and-severe stunting declined from 47·2% in 1985 to 29·9% in 2011 (Stevens et al.,

2012). Even after this dramatic reduction, 314 million children younger than 5 years were

mildly, moderately, or severely stunted in 2011. This paper also showed that developing

countries as a whole have less than a 5% chance of meeting the millennium development goal

(MDG) of improving undernutrition.

2.1.3 Social and economic cost of stunting

Stunting in early childhood is associated with poor cognitive and language development, delay in

the acquisition of fine motor skills and increased morbidity and mortality (Pelletier et al.,

2003;Victora et al., 2008). Stunting also has long-term consequences, which include short adult

height, reduced intellectual development and economic productivity and increased risk of

chronic diseases in adulthood (Victora et al., 2008). Notably, stunted girls grow as women with

short stature and are more likely to give birth to low birth weight (LBW) babies perpetuating the

intergenerational cycle of malnutrition (Kramer, 1987). Child stunting alone is responsible for

7% of total global disease burden, which accounts for the loss of 91 million DALYs (Black et al.,

2008).

6

2.1.4 Timing of linear growth faltering

The first two years of life is the time of greatest vulnerability and risk of irreversible long-term

physical and developmental damage (Emond et al., 2006; Grantham-McGregor et al., 2007;

Black et al., 2008). In a landmark study, Shrimpton et al. (2001) highlighted the worldwide

timing for growth faltering in early post-natal period. A recent study was carried out to describe

the growth faltering patterns by using the new World Health Organization (WHO) standards

comprising data from national anthropometric surveys from 54 countries (Victora et al., 2010).

This analysis showed that the decline in length-for-age Z-score (LAZ) starts close to the standard

and falters dramatically until 24 months (Figure 2.1). Growth faltering in early childhood was

even more pronounced than suggested by previous analyses based on the National Center for

Health Statistics (NCHS) reference (Shrimpton et al., 2001). The study emphasized the need to

scale up interventions during the window of opportunity defined by pregnancy and the first 2

years of life.

2.1.5 Early interventions to prevent stunting

Providing nutrition interventions during infancy is crucial, when nutrient needs relative to energy

intake are highest and the ability of the child to respond to a nutrition intervention is greatest

(Dewey and Adu-Afarwuah, 2008). Environmental factors have more effect on growth faltering

in the very early stage of life because growth velocity is highest during this period. Globally,

most nutrition programs are designed to target undernourished children identified through

growth monitoring activities and a large proportion of these programs are targeted to severely

underweight children (Shekhar and Lee 2006; Mason et al., 2005). As these programs invested in

treating undernutrition when substantial damage had already taken place, not surprisingly, they

7

showed very little success on reducing the overall burden of undernutrition. Ruel et al. (2008)

demonstrated that a preventive approach with efficient nutrition interventions is more effective

than this recuperation approach to reduce child undernutrition.

2.2 Epidemiology of low birth weight

2.2.1 Definition and global prevalence of low birth weight

Globally, 20 million newborns, representing 16% of all births in developing countries, are born

with low birth weight (LBW) (UNICEF/WHO 2004; Imdad and Bhutta 2013). More than 95%,

of these births are in developing countries. The greatest incidence of LBW occurs in South-

Central Asia followed by Africa (Imdad and Bhutta, 2013). Low birth weight is often used as an

indicator of fetuses that have not reached their expected standard of growth. The standard

definition of LBW is a newborn that weighs less than 2500g regardless of gestational age. A

newborn that weighs less than 1500g is defined as having Very Low Birth Weight (VLBW); a

newborn weighing less than 1000g is considered of Extremely Low Birth Weight (ELBW)

(UNICEF/WHO 2004).

2.2.2 Intra-uterine growth restriction and risk for mortality

The two main reasons for LBW are intrauterine growth restriction (IUGR) and preterm birth

(<37 weeks). More than two thirds (68%) of all LBW infants are born with intrauterine growth

retardation (IUGR) (Black et al., 2008). Such IUGR infants mostly include those born at term.

They may also include preterm infants or those born with a combination of prematurity and

IUGR. Term IUGR infants have much higher rates of morbidity and neonatal complications

including a higher risk of mortality (Black et al., 2008). It has been estimated that for term

8

infants weighing 2000–2499 g at birth, the risk of neonatal death is four times higher than for

infants weighing 2500–2999 g, and 10 times higher than for infants weighing 3000–3499 g

(Ashworth, 1998). One recent study showed that infants born at term weighing 1500-1999g

were 8.1 (95% CI 3.3-19.3) times more likely to die and those weighing 2000-2499g, were 2.3

(95% CI 1.3-4.1) times more likely to die from all cause during the neonatal period than the

infants born with normal birth weight (Black et al., 2008).

2.2.3 Consequences of low birth weight

The likelihood of stunting is inversely associated with birth weight (Adair and Guilkey, 1997).

LBW babies have poor immune function and higher risk of morbidity that makes them more

vulnerable to linear growth faltering and stunting in the post-natal period (Raqib et al., 2007).

LBW infants who survive are unlikely to catch up significantly on this lost growth and are more

likely to experience developmental deficits (Martorell et al., 1998). In addition, epidemiological

data have shown an important association between LBW and an increased prevalence of obesity,

hypertension, diabetes, and cardiovascular disease in the adult (Barker, 1998). This association

has been known as the Barker hypothesis from the name of its proponent or the Developmental

Origin of Health and Disease (DOHaD) hypothesis. This hypothesis holds that stressful in utero

environment alters the developmental programming of the embryo and fetus. This will result in

LBW and long-term consequences in the adult. The prenatal period is indeed a sensitive period

and organisms are more vulnerable to stressors during this phase of development (Barker, 1998).

Recently, strong epidemiological evidence linked IUGR or LBW to adult-onset insulin

resistant/type 2 diabetes mellitus, obesity, hypertension, and coronary artery disease (Varvarigou,

2010;). There is new evidence that epigenetic regulation is likely to be associated with this

9

phenomenon of fetal origins of adult disease (Morrison et al., 2010). Animal models for IUGR

induced by maternal nutrition restriction or uterine artery ligation have confirmed the association

between epigenetics and IUGR (Xu and Du, 2010).

2.3 Risk factors associated with linear growth faltering and stunting

Stunting is associated with the complex interaction between nutrition specific and nutrition

sensitive indicators, such as, food insecurity, suboptimal feeding practices, frequent infections,

poor sanitary conditions, inappropriate hygiene practices and poor access to health care services

(Black et al., 2008). Some of the immediate factors associated with linear growth faltering are:

lack of exclusive breastfeeding; inappropriate and inadequate complementary feeding,

micronutrient deficiencies and importantly, high rates of infectious diseases. In this review we

will focus in these immediate factors affecting linear growth.

2.3.1 Suboptimum breastfeed practices

The most important first nutritional intervention for a newborn is breastfeeding (Bhutta et al.,

2008). Breast milk is the natural and ideal first food for the neonate and provides numerous

immunologic, psychological and social, economic, and environmental benefits (Ip et al., 2009).

Breast milk has a significant positive impact on child survival, growth, and development and

decreases the risk for many acute and chronic diseases, including infections, such as, diarrhea

and respiratory tract infections during infancy (Bhutta et al., 2008). Adair and Guilkey (1997)

showed that breastfeeding has a protective effect on linear growth faltering in the first 6 months.

Exclusive breastfeeding remains uncommon in most countries both developed and developing.

Exclusive breastfeeding rates in infants less than 6 months of age varied from as low as 20% in

10

central and eastern European countries to 44% in south Asia (UNICEF, 2011). The reasons for

the low prevalence of exclusive breastfeeding could be a lack of support for breastfeeding by

social workers and healthcare providers, emotional stress in mothers and their perception of not

having enough breast milk, and pressure from close relatives to introduce other liquid and solid

foods, unsupportive hospital practices that delay early initiation of breast-feeding, maternal

employment etc.

2.3.2 Poor complementary feeding and micronutrient deficiencies

Complementary feeding for infants refers to the timely introduction of solid/semi solid foods in

addition to breast-feeding, that is, safe and nutrient-dense additional foods introduced at about 6

months of infant age (WHO, 2002). The timing of growth faltering often coincides with the

period of initiating complementary feeding i.e. during the dietary transition from the liquid diet

to eventually eating solid foods from the family pot. These transition diets are more commonly

qualitatively deficient than quantitatively deficient (Allen, 2003). Often these are inadequate and

inappropriate in terms of hygiene, energy and nutrient density, safety and feeding style (Dewey

and Adu-Afarwuah, 2008). The complementary foods in most developing countries are generally

cereal-based and commonly deficient in a number of micronutrients especially in iron, zinc, and

vitamins A (Dewey and Brown, 2003). The low micronutrient content of complementary foods

has been associated with growth faltering (Kariger et al., 2005). Poor micronutrient status of a

child leads to increased risk of infections thus influence growth faltering among children (Black,

2003).

11

2.3.3 Infectious diseases

The direct interaction between infection and feeding has been well described (Scrimshaw et al.,

1959; Bhutta 2006) (Figure 2.2). An infant or young child with an active infection is likely to be

lethargic, anorexic and often disinterested in food and the process of eating, even breastfeeding.

Acute infections are accompanied by changes in acute phase reactants, including inflammatory

cytokines. It is believed that inflammatory cytokines interfere with the metabolism and

utilization of nutrients (Stephensen et al., 1999). Indeed, nutrients are either channeled to ‘fight

infection’ (e.g. increased zinc availability during infection likely enhances the immune response)

or are sequestered so as not to enhance bacterial growth (e.g. iron is sequestered in ferritin).

A recent analysis showed that the odds of stunting at 24 months increased multiplicatively with

each episode of diarrhea (Bhutta, 2006). The adjusted odds of stunting at 24 months increased by

a factor of 1.05 (odds ratio 1.05, 95% CI 1.03-1.07) with each episode of diarrhea in the first 24

months. During each of the first two years of life, if such illness recurs 6-8 times or more, the net

effect of these acute self-limiting infections on feeding, nutrient utilization and ultimately growth

will be significant. A community-based longitudinal study in rural Bangladesh showed that

overall diarrheal incidence rate among children less than five years old was 4.6 episodes per

child per year and the incidence of persistent diarrhea was 34/100 child-years (Baqui et al.,

1992). Thus any intervention that can decrease the number of infections in an individual infant is

likely to impact on their appetite, ingestion of food and ultimately on growth. If a mother is

unaware of infant cues for more frequent feeding, if the child is on a feeding regimen that does

not allow for additional food or more frequent feeding episodes or if there is not additional food

available to provide extra food for the young child, nutrient needs may not be met for the days

during the acute illness and often for a few days thereafter (Piwoz et al., 1994).

12

In summary, the major contributing factors for post-natal linear growth faltering are

inappropriate breastfeeding, complementary feeding as well as micronutrient deficiencies that

result in poor growth and suppressed immune system among children (Figure 2.3 and 2.4).

Consequently, episodes of infectious diseases, such as diarrhea and respiratory illness play a

pivotal role in length growth faltering. In addition to the above-mentioned individual factors,

there are certain underlying factors that need to be addressed to control stunting in a sustained

manner, such as, low maternal education, poverty, inequity, and most importantly women's status

in the society.

2.4 Current evidence for interventions to prevent and reduce stunting

In a systematic review, Dewey and Adu-Afarwuah (2008) argued that there is no single universal

best package of components to reduce stunting. Therefore, a combination of intervention needs

to be considered to combat the multi-faceted issue of linear growth faltering and stunting. In this

review, we will discuss three effective community-based preventive interventions to tackle linear

growth faltering: i) nutrition and health education; ii) multiple micronutrient supplementation or

fortification; and iii) hand hygiene to control infections.

2.4.1 Nutrition and health education to improve Infant and Young Child Feeding

Appropriate infant feeding practices and better dietary quality were found to be associated with

better linear growth in Bangladeshi children (Saha et al., 2008). Community-based culturally

appropriate nutrition education to caregivers can improve infant-feeding practices, increase

energy and nutrient intake, and thus reduce stunting (Bhandari et al., 2003, Roy et al., 2007). It is

13

crucial to ensure exclusive breastfeeding and timely introduction of complementary foods to

prevent growth faltering. Over the past decades, evidence for the health advantages of breast-

feeding and recommendations for practice have continued to increase. Breast-feeding reduces

child mortality and has health benefits that extend into adulthood (Bhutta et al., 2008). WHO

recommends exclusive breast-feeding for the first 6 months of life followed by continued breast-

feeding with appropriate complementary foods for up to 2 years or beyond (WHO, 2009).

Complementary foods should be added to the diet of the child at 6 months of age, as breast milk

is no longer enough to meet the nutritional needs of the growing infant (WHO, 2003). According

to WHO, the complementary feeding should be timely, meaning that all infants should start

receiving foods in addition to breast milk from 6 months onwards; adequate, meaning that the

nutritional value of complementary foods should fulfill the needs of the rapidly growing child;

and appropriate, meaning that foods should be diverse, of appropriate texture, and given in

sufficient quantity (WHO, 2003). Several strategies have been employed to improve

complementary feeding practices. These include nutritional counseling to mothers designed to

promote healthy feeding practices in food secure populations, provision of complementary foods

in food insecure populations, and supplementation with foods either fortified with MMNs or with

increased energy content. A review by Dewey and Adu-Afarwuah (2008) showed that

educational interventions for complementary feeding had a modest effect on linear growth

[weighed mean difference (WMD)=0.20; SD; range 0.04–0.64]. Another review and model

simulation published in the Lancet Nutrition Series demonstrated that provision of

complementary food (with/without nutritional counseling) was associated with a significant

increase in linear growth in food insecure populations (Bhutta et al., 2008)

14

Psychosocial stimulation and responsive feeding is another prerequisite for optimal growth of

young children (Engle et al., 2000; Aboud et al., 2009). Education on proper illness management,

e.g. recognizing the symptoms of child illnesses, taking appropriate care and feeding during

illness, recognizing when to seek outside care and seeking appropriate care is also critical to

prevent infections. One study in rural Bangladesh showed that nutrition education on infant

feeding, disease control and stimulation prevented weight growth faltering among children aged

6-9 months, however the study did not show prevention of linear growth faltering in those infants

(Roy et al., 2007). It implies that nutrition education alone was not sufficient to prevent linear

growth faltering in this population.

In a recent review Briscoe & Aboud (2012) highlighted that the most successful health behavior

change interventions in developing countries used 3-4 behavior change techniques. Authors

suggest that engaging participants at behavioral, social, sensory and cognitive levels is crucial to

sustain change. Authors also argued that behavior change techniques were rarely linked to

theories of behavior change. Therefore, inclusion of behavior change theories to address

determinants of stunting at multiple levels (e.g. individual, household, community, etc) should be

considered to bring about change (Aboud and Singla, 2012).

2.5 Micronutrients to prevent stunting

Recently a few reviews on multiple micronutrients interventions argued that prevention of

multiple micronutrient (MMN) malnutrition is key to improve child health and survival (Semba,

2012; Christian and Tielsch, 2012). As deficiencies of important micronutrients such as iron,

zinc, vitamin A, among others, are prevalent in children in developing countries, efforts have

15

been made to supplement infants and children with MMNs (Ramakrishnan et al., 2008; Allen et

al., 2009; Ramakrishnan et al., 2009). There has been increasing recognition that micronutrient

deficiencies do not occur in isolation and that providing several micronutrients at the same time

would be more beneficial and cost effective than single nutrient interventions (Ramakrishnan et

al., 2008).

2.5.1 Multiple micronutrients (MMN) and linear growth

One recent meta-analysis showed that no single micronutrient has positive impact on linear

growth rather a combination of three has some modest impact (Ramakrishnan et al., 2011). Allen

et al. (2009) compared the effects of MMN interventions containing 3 micronutrients to a

placebo or only 1 or 2 micronutrients. They estimated that the overall effect size for length is

0.25 for supplements and 0.18 for fortified food [overall effect size for length: 0.13 (95% CI:

0.06, 0.21)]. For weight, these values were 0.26 for supplements and 0.15 for fortified food

[overall effect size for weight: 0.14 (95% CI: 0.03, 0.25)]. These effects were significant. There

was no evidence of a greater effect in younger children or of interactions with baseline

nutritional status. They also found significant effects for linear growth, however, the overall

effect size was much smaller [ES = 0.09 (95% CI = 0.006, 0.17)]. There were no differences by

type of supplement or baseline nutritional status of children. The composition of supplements

varied between studies; therefore, it is not easy to discern one or two specific formulations for

future use from this evaluation.

A previous meta-analysis has demonstrated that iron alone or vitamin A alone does not

significantly affect either linear or ponderal growth (Ramakrishnan et al., 2004). In contrast, five

16

trials of multiple micronutrient supplementation in which the intervention was more uniform

containing vitamin A, iron, zinc, B vitamins, and folic acid had a somewhat higher effect size for

height and weight [0.28 (95%CI: 0.16, 0.41)], but this was not significant [0.28 (95%CI: −0.07,

0.63)] (Shrimpton et al., 2009). In four collaborative parallel trials conducted in Indonesia, Peru,

South Africa, and Vietnam using the same multiple micronutrient formulation, a pooled analysis

found higher (P < 0.05) weight gain/month (0.21 ± 0.09 g) compared to the control (0.21 ± 0.09

g) but not for height (Smuts et al., 2005). This trial excluded infants who were born with low

birth weight.

Recently, Chaagan et al. (2010) reported that stunted children who received MMN supplements

had a 0.7 length-for-age Z-score improvement compared to a 0.3 Z-score decline among those

who received only vitamin A and zinc and a 0.2 Z-score increase among those who received only

vitamin A. After 12 mo of follow-up among HIV-infected children who received twice the RDA

of 14 micronutrients compared to those who received a single RDA of 6 vitamins (standard of

care) for 6 mo, Ndeezi et al. (2010) found no significant differences in mean HAZ scores

measured.

In summary, MMN interventions have demonstrated to improve linear growth among children;

however, results vary depending on the study setting, MMN intervention, comparison group, and

the study population.

2.5.2 Home fortification of complementary foods

17

In order to supplement infants with adequate micronutrients, the concept of ‘home fortification’

of complementary foods has been evaluated. Currently, three approaches are being used for

home fortification: Micronutrient powders (e.g., Sprinkles), crushable tablets (e.g., foodlet), and

lipid-based [e.g., Nutributter (Nutriset, Malaunay, France)] or soy-based products. Micronutrient

Powder (or sprinkles) are the most extensively studied product among these and consists of

sachets containing micronutrients in a powdered form, to be sprinkled onto a portion of the

child's food just before it is consumed (de Pee et al., 2008). Crushable or chewable tablets are

MMN tablets that can be easily dissolved in a small amount of liquid, or crushed and added to

complementary foods (Smuts et al., 2005). Lipid-based nutrient supplements (LNS) are products

that provide some energy, predominantly from fat, in addition to MMNs (Chaparro et al., 2010).

Dewey et al. (2009) have reviewed home fortification strategies of complementary foods. Home

fortification was highly effective at reducing iron deficiency and decreases the prevalence of

anemia by half. There was no effect on growth with micronutrients only; however, there was

significant effect when micronutrients combined with a small amount of energy (including fat

and protein). These results were based on the pooled data from two efficacy trials in Africa

(Kuusipalo et al., 2006; Adu-Afarwuah et al., 2007). Home fortification also had beneficial

impact on morbidity in high-risk populations in some studies; however, there was no overall

significant impact. Acceptability of home fortification by caregivers and young children was

high, and side-effects were rare (Dewey et al., 2009).

2.5.3 The potential of improved Micronutrient Powders to enhance growth

Home fortification of complementary foods with micronutrient powders (MNP), is so far the

most feasible and cost-effective strategy to combat micronutrient deficiencies (Sharieff et al.,

18

2006; Menon et al., 2007; Sherieff 2008; Semba 2012). Micronutrient powder (or Sprinkles) was

originally developed by Stanley Zlotkin et al. in the late 1990s as a strategy to combat nutritional

anemia (Zlotkin et al., 2001, Zlotkin et al., 2003). Known as the “point of use” fortification, a

small sachet of the powder is opened and sprinkled on an individual serving of food prior to

consumption to increase its nutrient density. The MNP can be sprinkled on the most basic bowl

or plate of food. So, one of the advantages of home fortification is that it can reach the poorest

families, where lack of dietary diversity is a major concern. Currently, MNPs are being used in

large-scale programs, mainly in development settings, but also in a few emergency and refugee

settings (de Pee et al., 2007). Micronutrient powders were also distributed in refugee camps in

Kenya, Nepal, and Bangladesh, and in disaster relief in Bangladesh and the Philippines, as

reviewed by Rah et al. (2012).

In the past, MNPs, particularly Sprinkles have been designed to include only those

micronutrients needed to prevent or treat various types of anemia, or alternatively to provide a

variety of minerals and vitamins to mimic the RDA for young children (Adu-Afarwuah et al.,

2008). The MNP formulations, however, have not contained high enough amounts of minerals,

such as calcium and magnesium, to enhance bone growth nor additional zinc to enhance appetite.

The effects of zinc deficiency on the sense of taste and on appetite have been well described

(Jing et al., 2007). Individuals with zinc deficiency (e.g. acrodermatitis enteropathica) present

with anorexia, while environmentally induced zinc deficiency is associated with altered taste and

reduced appetite. With zinc treatment, these lost senses are improved and appetite is increased

(Gibson, 2000). In a child with ‘normal’ zinc status, additional zinc above the RDA is unlikely to

either enhance growth or the child’s sense of hunger. However, if an infant or child has limited

19

zinc intake and increased losses (through, for example, frequent bouts of diarrhea), additional

zinc may improve appetite and total food intake. The addition of zinc as well as calcium,

magnesium, manganese and other necessary minerals in MNPs may further enhance linear

growth among infants.

2.6 Current evidence for hand hygiene to prevent infections

2.6.1 Handwashing to prevent infectious diseases in developing countries

Water supply, sanitation, and hygiene are important for the prevention of malnutrition because of

their direct impact on infectious disease, especially diarrhea (Fewtrell et al., 2005; Bartram and

Cairncross, 2010; Dangour et al., 2013). In a recent meta-analysis of data from cluster-

randomized controlled trials carried out in low and middle income countries, it was demonstrated

that there was a positive effect of water sanitation and hygiene (WASH) interventions on linear

growth of children under five years of age (Dangour et al., 2013). Recent research initiatives in

developing countries have included education on a number of home-based practices such as

hand-washing, safe disposal of stool, improving water quality and storage, and reducing food-

borne contamination. These interventions have shown to be effective in controlling disease

transmission at home and in the community (Curtis and Cairncross, 2003; Luby et al., 2007;

Luby and Curtis, 2008; Ejemot et al., 2008). In particular, hand-washing with soap and water can

significantly reduce diarrheal and respiratory diseases among children (Luby et al., 2005; Luby

et al., 2008). A review by Cairncross et al., (2010) showed a reduced risk of diarrhea by 48, 17,

and 36%, associated respectively, with handwashing with soap, improved water quality, and

excreta disposal as the estimates of effect for the Live Saved Tool (LiST) model. In a recent

review Blencowe et al., 2011 showed that improved handwashing practices among birth

20

attendants reduced all-cause neonatal mortality (19%), cord infection (30%), and neonatal

tetanus (49%). Postnatal maternal handwashing was associated with 44% reductions in all-cause

mortality and 24% reduction in cord infections. Interestingly, one study showed reduction in

diarrhea among children of mothers who wash hands with water alone compared to mothers who

do not wash hands at all (Luby et al., 2011).

2.6.2 Limitations of handwashing with soap and water

Adherence to handwashing practice, however, requires convenient access to water and soap, and

sufficient time to perform the procedure, all of which are not readily available in resource-poor

settings. In Bangladesh, for example, having water and soap available at the place to wash hands

after toileting were significantly associated with washing both hands with soap after fecal contact

(Luby et al., 2009). In Ghana Scott et al. surveyed 449 mothers from 5 major socio-cultural

zones, and found that those with household water connections (only 12% of households) were

over twice as likely to wash their hands after defecation as those without this type of access

(Scott et al., 2008). It has been well documented in hospital- or institutional-based settings that

hand sanitizers may be effective alternatives to hand washing with soap and water, since they do

not require rinsing, and can kill most bacteria, protozoa and several types of viruses (Kampf and

Kramer, 2004). It has not been established whether the community (household) use of hand

sanitizers would be as successful.

2.6.3 Efficacy of alcohol-based hand sanitizers and inherent disadvantages

Currently, the most widely used hand sanitizers in health care facilities are alcohol-based, due to

their proven in-vitro and in-vivo germicidal activity against bacteria, several viruses, and fungi

21

(Simon et al., 2004; Sandora et al., 2008). At the community level, significant reductions in the

transmission of gastrointestinal and respiratory infections in the homes of American families

with young children have been reported (Sandora et al., 2005; Lee et al., 2005). Besides this

handful of studies, however, the efficacy of hand sanitizers in the community has largely not

been investigated; and virtually no studies have been conducted in developing countries.

Although the current WHO-recommended hand sanitizer formulations for health-care facilities

are alcohol-based (Pittet et al., 2009) their acceptability is often reduced by their tendency to

cause excessive dryness and skin irritation (Larson et al., 2009). Disruption of the protective

lipid layer and interfere with the skin’s natural defense against bacterial infections (Houben et

al., 2006). Other disadvantages of using alcohol-based hand sanitizers in resource poor settings

include safety and culture-related concerns. Due to their high flammability, alcohol-based

sanitizers may pose a fire hazard if they are exposed to an open flame. Since cooking in many

resource-poor settings is via open flame, this potential hazard is of significant concern.

Accidental ingestion of alcohol-based products may lead to poisoning, especially in young

children (Miller et al., 2009). Also, the use of alcohol-based agents may not be acceptable in

certain religious communities where alcohol abstinence is practiced. Finally, the unpleasant taste

of product residue left on the fingers may deter those who eat primarily with their hands from

using a sanitizer before preparing and eating meals.

2.6.4 Alcohol-free hand sanitizers

Due to the inherent disadvantages of alcohol-based hand sanitizers, the efficacy of alcohol-free

antimicrobial agents have been investigated in various settings (White et al., 2001). One such

group of alcohol-free preparations is quaternary ammonium compounds (QAC’s), which include

22

the most widely used antiseptic agent, benzalkonium chloride. Benzalkonium chloride is

primarily bacteriostatic and fungistatic and generally well tolerated and have low allergenic

potential (Thorsteinsson et al., 2003). One recent study showed use of waterless hand sanitizer

reduced hand contamination but did not improve the frequency of handwashing compared with

soap (Luby et al., 2010). Pickering et al., (2010) showed that hand sanitizer was significantly

better than handwashing at reducing fecal streptococci on hands. No significant difference in

efficacy between hands that were clean versus dirty or oily (Pickering et al., 2011).

2.7 Infant growth monitoring

The objective of infant growth monitoring is to promote a growth rate that will optimise

immediate and long-term health and wellbeing. Regular examinations of infants and

measurement of growth are recommended in most countries. In developing countries the use of

growth charts in the regular monitoring of infants and children was popularised by Morley and

Woodland (1980). The regular weighing and plotting against a growth reference is an important

monitoring tool as parents (or health professionals) do not recognise the rate of growth (Jeffery

et al., 2005). Consequently the growth chart has become the most widely used paediatric

intervention, with routine vaccination the only intervention in wider use, in countries and

communities at all stages of economic development.

2.7.1 The development of the World Health Organization (WHO) growth standard

Until recently the growth reference in most widespread use around the world was the NCHS

2000 reference, based on longitudinal data, which was also used by the World Health

Organisation. For a number of years it was recognised that this reference includes a high

23

proportion of formula-fed infants and may not accurately reflect the growth of breastfed infants.

It was apparent that exclusively breastfed babies grow at a slightly lower rate than the NCHS

reference.

Concern was expressed that the existing growth reference was too high and might lead mothers

to introduce complementary foods unnecessarily (de Onis et al., 1997). The author expressed

concerns about the existing growth reference: “the NCHS curves are inappropriate for healthy,

breastfed infants. Recent research shows that infants fed according to recommendations by the

WHO and who live under conditions that favour the achievement of genetic growth potentials

grow less rapidly than, and deviate significantly from, the NCHS reference. The negative

deviations are large enough to lead health workers to make faulty decisions regarding the

adequate growth of breastfed infants, and thus to mistakenly advise mothers to supplement

unnecessarily or to stop breastfeeding altogether. Given the health and nutritional benefits of

breastfeeding, this potential misinterpretation of the growth pattern of healthy breastfed infants

has great public health significance. The premature introduction of complementary foods can

have life-threatening consequences for young infants in many settings, especially where

breastfeeding’s role in preventing severe infectious morbidity is crucial to child survival”. These

concerns led to the development of a new growth reference by the WHO (de Onis et al., 2006).

2.7.2 Multi centre growth reference study

The new WHO standards based on the Multicentre Growth Reference Study (MGRS) adopt a

fundamentally prescriptive approach designed to describe how all children should grow rather

than the more limited goal of describing how children grew at a specified time and place (de

Onis et al., 2006). Since the publication of the new WHO growth standards, statements have

24

often been made stating that they are a major advance in child health such as “WHO growth

standards are an international gold standard describing how children should grow when

measured by weight and height” (Kerac et al., 2009). A recent study showed that, the WHO

standards have been widely scrutinized and implemented. Countries have adopted and

harmonized this best practices in child growth assessment and established the breast-fed infant as

the norm against which to assess compliance with children’s right to achieve their full genetic

growth potential (de Onis et al., 2012).

2.8 Assessing nutritional status

2.8.1 Anthropometry

Changes in body dimensions reflect the overall health and welfare of individuals and

populations. Anthropometry is used to assess and predict performance, health and survival of

individuals and reflect the economic and social well-being of populations (Gibson 2005).

Anthropometry is a widely used, inexpensive and non-invasive measure of the general nutritional

status of an individual or a population group. Growth monitoring is described by anthropometric

measurement including weight, length and head circumference. These measures are relatively

simple inexpensive safe and non-invasive, however accuracy is crucial to overall growth

assessment (Gibson, 2005) Standardized measurement techniques, trained personnel and the use

of calibrated equipment are imperative in order to obtain precise measurements (de Onis et al.,

2006).). Also, to reduce errors and ensure accuracy, measurement should be repeated at least

twice and it is preferred if the same person obtain measurements over time.

2.8.2 Anthropometric indices

25

When anthropometric variables (length, weight, and age) are used together, they can provide

important information about a person’s nutritional status and is called an index. Three indices are

commonly used in assessing the nutritional status of children: length-for-age, weight-for-age;

and weight-for-length. The advantages and disadvantages of the three indices and the

information they can provide is summarized below:

2.8.2.1 Length-for-age

Low length-for-age index identifies past undernutrition or chronic malnutrition. It cannot

measure short term changes in malnutrition (Gibson 2000). For children below 2 years of age,

the term is length-for-age; above 2 years of age, the index is referred to as length or height-for-

age. Deficit in length-for-age is referred to as stunting. However, length-for-age is not a sensitive

index, in that it does not provide information regarding the proportionality of body mass relative

to length and should not be used in isolation (Gibson 2000).

2.8.2.2 Weight-for-age

Low weight-for-age index identifies the condition of being underweight, for a specific age

(Gibson 2000). The advantage of this index is that it reflects both past (chronic) and/or present

(acute) undernutrition (although it is unable to distinguish between the two). However, the major

limitation with weight-for-age curves is that length or height is not taken into consideration.

2.8.2.3 Weight-for-length

Low weight-for- length helps to identify children suffering from current or acute undernutrition

or wasting and is useful when exact ages are difficult to determine. Weight-for-length (in

26

children under 2 years of age) or weight-for-height (in children over 2 years of age) is

appropriate for examining short-term effects such as seasonal changes in food supply or short-

term nutritional stress brought about by illness (Gibson 2000).

2.8.3 Evaluation of anthropometric indices using Z-scores

Z-scores are more commonly used by the international nutrition community (Gibson 2000; de

Onis 2004, de Onis et al., 2006) because they offer two major advantages. First, using Z-scores

allows us to identify a fixed point in the distributions of different indices and across different

ages. For all indices for all ages, 2.28% of the reference population lie below a cut-off of -2 Z-

scores. The percent of the median does not have this characteristic. For example, because weight

and length have different distributions (variances), -2 Z-scores on the weight-for-age distribution

is about 80% of the median, and -2 Z-scores on the length-for-age distribution is about 90% of

the median. Further, the proportion of the population identified by a particular percentage of the

median varies at different ages on the same index.

The second major advantage of using Z-scores is that useful summary statistics can be calculated

from them. The approach allows the mean and standard deviation to be calculated for the Z-

scores for a group of children. The Z-score application is considered the simplest way of

describing the reference population and making comparisons to it. It is the statistic recommended

for use when reporting results of nutritional assessments.

The Z-score or standard deviation unit (SD) is defined as the difference between the value for an

individual and the median value of the reference population for the same age or height, divided

27

by the standard deviation of the reference population. This can be written in equation form as

follows:

For normal distribution

Z-score or SD= (Observed value)-(mean reference value)

Standard deviation of reference population

For non-normal distribution

Z-score or SD= (Observed value) ÷(M)L-1

L*S

(Where, M=reference median; L=the power needed to normalize the data; S=coefficient of variation)

The distribution of Z-scores follows a normal (bell-shaped or Gaussian) distribution. The

commonly used cut-offs of -3, -2, and -1 Z- scores are, respectively, the 0.13th, 2.28th, and 15.8th

percentiles. The most commonly-used cut-off with Z-scores is -2 standard deviations,

irrespective of the indicator used. This means children with a Z-score for underweight, stunting

or wasting, below -2 SD are considered moderately or severely malnourished. For example, a

child with a Z-score for length-for-age of -2.56 is considered stunted, whereas a child with a Z-

score of -1.78 is not classified as stunted. Growth Charts using Z-scores are presented separately

for boys and girls (Appendix 1-4).

2.8.4 Composite measures from anthropometric indices

28

The three indices are used to identify three nutritional conditions: underweight, stunting and

wasting (WHO 2006).

2.8.4.1 Stunting

Low length-for-age, stemming from a slowing in the growth of the fetus and the child and

resulting in a failure to achieve expected length as compared to a healthy, well-nourished child of

the same age, is a sign of stunting. Stunting is an indicator of past growth failure. It is associated

with a number of long-term factors including chronic insufficient protein and energy intake,

frequent infection, sustained inappropriate feeding practices and poverty.

2.8.4.2 Underweight

Underweight, based on weight-for-age, is a composite measure of stunting and wasting and is

recommended as the indicator to assess changes in the magnitude of malnutrition over time.

2.8.4.3 Wasting

Wasting is the result of a weight falling significantly below the weight expected of a child of the

same length. Wasting indicates current or acute malnutrition resulting from failure to gain weight

or actual weight loss. Causes include inadequate food intake, incorrect feeding, disease, and

infection or, more frequently, a combination of these factors. Wasting in individual children and

population groups can change rapidly and shows marked seasonal patterns associated with

changes in food availability or disease prevalence to which it is very sensitive and therefore used

in emergency situations.

29

2.9 Child undernutrition in the context of Bangladesh

2.9.1 Prevalence of Low birth weight and stunting in Bangladesh

Bangladesh is a tropical country located in South Asia with very high rates of child mortality and

undernutrition (NIPORT, 2011) (Figure 2.5 and Figure 2.6). It is estimated that as many as

35% of infants are born with LBW in South Asia, although this number may be an underestimate

because of the high proportion of home births (UNICEF, 2012). The rates of LBW among

Bangladeshi infants, though reduced from 40%, are still among the highest in the world, ranging

from 20% to 22% (WHO 2011, Ahmed et al., 2005, UNICEF 2010). The national LBW survey

suggested that approximately 25% are born prematurely, while the remainder of LBW infants

were IUGR-LBW (BBS/UNICEF, 2005). However, a study in rural Bangladesh, using the Parkin

method, to determine gestational age, reported that IUGR was the major contributor to LBW

(96.4%) while only 3.6% of babies were born preterm (Nahar et al., 2004).

Conversely, one recent study, produced by active birth surveillance, demonstrated that low birth

weight is far more common than suggested by cross-sectional survey estimates (Klemm et al.,

2013). In this study, over one-half (55.3%) of infants were born low birth weight and 46.6%,

37.0% and 33.6% had a weight, length and head circumference below -2 Z-scores of the WHO

growth standard at birth. Infants in this typical rural Bangladesh setting were commonly born

small, reflecting a high burden of fetal growth restriction and preterm birth. Despite the high

prevalence of LBW infants in Bangladesh, to date, few studies have explored interventions to

improve linear growth in this vulnerable population.

30

In Bangladesh, 43% of under-five children are stunted (length or height-for-age Z-score <-2),

and wasting is present in 17% of children, with 3.4% being severely wasted (NIPORT, 2011).

This accounts for more than 0.5 million children with severe acute malnutrition (SAM) in the

country. These children who are suffering from various forms and degrees of undernutrition are

at a high risk of death or severe impairment of growth and development. Linear growth of

surviving infants in Bangladesh starts to falter from three months of age and progressively

deteriorates until they reach 24 months resulting in very high rates of stunting (Saha et al., 2009)

(Figure 2.7).

2.9.2 Infant and young child feeding practices in Bangladesh

Breastfeeding is common in Bangladesh; 99% of infants aged less than 12 months are breastfed.

However, the prevalence of EBF is 43% in infants aged less than six months, and this has not

changed over the last decade, with serious implications for the nutritional well-being and

mortality of children (NIPORT, 2011). Forty-three percent of neonates are breastfed within one

hour of birth. The median duration of breastfeeding is 32.8 months while that of EBF is only 1.8

months.

Inappropriate feeding practices, particularly after the age of six months when breastmilk alone is

no longer sufficient to meet the increasing nutrient requirements for growth, result in high rates

of childhood undernutrition in developing countries. In Bangladesh, complementary feeding

starts too early or too late, and foods that are offered are often inappropriate, leading to high rates

of childhood undernutrition. Even among infants aged less than two months, 17% drink milk

31

other than breastmilk (fresh or powdered cow's milk, infant formula, etc.); 6% are given other

liquids; and another 6% receive solid or semi-solid foods (NIPORT, 2011).

On the other hand, 25% of infants aged 6-9 months are not fed any solid or semi-solid foods in

addition to breastmilk. Of the older children who have crossed exclusive breastfeeding age, only

42% are fed according to the recommended infant and young child-feeding practices.

Furthermore, results of a study in Bangladesh showed that the amount of energy from

complementary foods offered to infants was about 74% of the recommended amount (Kimmons

et al., 2005). The mean intakes of vitamin A from breastmilk and complementary food together

were 44% and 48% of the required nutrient intakes for children aged 6-8 months and 9-12

months respectively. The intakes of vitamin D and zinc were 12-13% of the recommended

nutrient intake (RNI) and 40-45% respectively. The intake of iron was very low, accounting for

only 8-9% of the RNI. The findings of the study imply that it is difficult to improve the

micronutrient intakes of children by simply increasing the amount of complementary food

currently consumed in Bangladesh. Other interventions, including home fortification of food or a

complementary food for children that has added micronutrients and other nutrients require

special consideration.

2.10 Conclusion

Currently, there is an emerging interest towards micronutrient supplementation and prevention of

stunting in early childhood. However limited data is available on growth of low birth weight

infants from developing countries, who are at the highest risk of stunting. In addition to

improving infant and young child feeding, there is a growing recognition that it is necessary to

32

investigate in non-nutrition strategies, such as infection control. High rates of LBW and stunting

mean that many children in developing countries will not reach their full developmental

potential. Improving nutritional status in early life is crucial for long term health. Strong data to

support positive and long-term effects of multiple micronutrients on infant growth are still

lacking. It has only been recently that studies have started to examine impact of MMN on growth

and development on developing countries. These studies have been conducted in South Asian

countries where LBW rates are high. These handful of studies have reported promising results of

the impact of MMN supplementation on linear growth. To date no study has explored the effect

of hand sanitizers in developing country to prevent growth faltering and interventions for

improved infant growth outcomes among LBW babies have yet to be determined.

33

Figure 2.1. Worldwide timing for growth faltering (Victora et al., 2010)

34

Figure 2.2. Malnutrition and infection cycle (Modified from Tomkins and Watson, 1989).

Pathogen  in    

environment  

Pathogen  invade  host  

Pathogen  causes  illness  

 

Prolonged  recovery  Appetite  loss  Nutrient  loss  Malabsorption  Altered  metabolism    

COMPLETE  RECOVERY  

AND  SURVIVAL  

STUNTING  

DEATH    

OR  DISABILITY  

Inadequate  

dietary  intake  

Weight  loss  Growth  faltering  Malnutrition  

Weak  host  defense  system  

Appropriate  nutrition  and  feeding  during  illness,  Effective  illness  management  

Lack  of  care  

Lowered  Immunity  

Mucosal  damage  

Malnutrition-­‐Infection    

Cycle  

35

Figure. 2.3. Conceptual framework for the relationship of specific child care behaviors and

infant health outcomes (Modified from Engle et al., 1996).

Safe  and  Hygienic  

Environments  

Appropriate  Infant  Feeding  practices  

Appropriate  Illness  prevention  and  treatment  

Illness   Nutritional  Status  

Linear  growth,  Development  and  

Survival  

Child  Health  

Outcomes  

Precipitators  of  Child  Health  Outcomes  

Maternal  Care  

behaviours  

36  

Figure 2.4. Etiological relationships between various biological outcomes of child development (adapted from Sellen, 1995).

ENVIRONMENT:  Ecology  Subsistence    Cultural  practices  Parental  behaviors  

FOOD  SECURITY,                

FOOD  PREFERENCES  

RESPONSE  TO  

INFECTION  

Dietary    Quality    

LINEAR  GROWTH  

MORTALITY  

MORBIDITY  

SOCIAL  INTERACTIONS  

Neuroendocrinological completence; brain

tissue

PHYSICAL  CONDITIONS  

SUSCEPTIBILITY  TO  

INFECTIONS    

CHILD  FEEDING  PRACTICES  

 

EXPOSURE  TO  

INFECTION  

 

Dietary  

Intake  

 

Nutrient  

Supply  

 

Nutrient          Use  

 

MAINTENANCE    

 

IMMUNE  FUNCTION    

 

COGNITIVE  

DEVELOPMENT  

Food preparation Meals offered

Anorexia Lesions Metabolism

Gut size Gut maturation

Injury

Infections

T4 cells

Epithelial integrity

NEW  TISSUE  

 

37  

Figure 2.5. Trends in child mortality in Bangladesh, 1989-2011. (NIPORT, 2011)

   

38  

Figure 2.6. Trends in child nutrition in Bangladesh, 2004-2011 (NIPORT, 2011)

   

39  

Figure 2.7. Linear growth patterns (Length-for age) and stunting among Bangladeshi infants

across all socioeconomic groups indicated by household food security (Saha et al., 2009).

   

40  

CHAPTER 3

RESEARCH RATIONALE AND OBJECTIVES

3.1 Rationale of the proposed research

Improving nutrition has a high potential for increasing child survival and quality of life.

Globally, more than half of the 9 million deaths among children less than 5 years are associated

with undernutrition (Pelletier et al., 1995; Black et al., 2008a; Bhutta et al., 2008). The first two

years of life present a period of greatest vulnerability and risk of irreversible long-term physical

and developmental damage from malnutrition (Rivera and Ruel, 1997; Shrimpton et al., 2001;

Emond et al., 2006; Grantham-McGregor et al., 2007; Black et al., 2008b; Victora et al., 2008).

At population level, prevention with targeted and effective nutrition interventions is

recommended as more ethically desirable and cost-effective than therapeutic approaches to

reduce child undernutrition (Ruel et al., 2008; Dewey and Adu-Afarwuah, 2008).

The direct interaction between infections in infancy and feeding is well described (Scrimshaw et

al., 1959; Mata et al., 1977; Bhutta, 2006). Acute infections are accompanied by changes in acute

phase response metabolites, including inflammatory cytokines that likely interfere with the

metabolism and utilization of nutrients (Piwoz et al., 1994). Infections are also associated with

impaired nutrient absorption and utilization, cause direct nutrient loss, increase metabolic

requirements or catabolic losses of nutrients and possibly impair the transport of nutrients to

target tissues (Bhutta, 2006). With new episode of infection, nutrients are either channeled to

‘fight infection’ (e.g. increased zinc availability during infection likely enhances the immune

response) or are sequestered so as not to enhance bacterial growth (e.g. iron sequestered in

ferritin) (Stephensen, 1999). If such illness recurs 6-8 times or more during each of the first two

41  

years of life, the net effect of these acute self-limiting infections on feeding, nutrient absorption

and utilization, and ultimately growth can be significant (Bhutta, 2006). In low income countries

like Bangladesh, recurrent episodes of diarrhea and respiratory infections in the first months of

life undoubtedly affect the appetite as well as the utilization of nutrients for many infants and

likely play a pivotal role in early malnutrition (Baqui et al., 1992; Shankar and Prasad, 1998).

Low birth weight (LBW) infants are particularly vulnerable to such infections in increased

frequency.

Exclusive breastfeeding (EBF) is recommended for the first 6 months of life of infants because it

can support well the nutrient needs of most healthy infants. Ingestion of adequate volume of

breast milk through EBF, therefore, is expected to result in ‘normal’ growth throughout the first

half of infancy (Bhutta et al., 2008). During this same period, however, infants are highly

susceptible to clinically apparent as well as sub-clinical infections, particularly in resource-poor

settings. Thus, there is urgent need to ensure exclusive breastfeeding and prevent infection to

break the infection-malnutrition cycle. In the second 6 months of life, and thereafter, the poor

quality of locally available complementary foods may lead to a deficiency of micronutrients that

are needed to support of the infant’s immune system and adequate linear growth. During this

period, infections may continue to interfere with the feeding process and prevent the

amelioration of the malnutrition that started in the first 6 months of life (Bhutta, 2006). The

resultant sub-optimal breastfeeding, limited ingestion and utilization of complementary foods,

and ultimately micronutrient deficiencies contribute significantly to post-natal linear growth

faltering and stunting.

42  

Any intervention that can decrease the number of infectious episodes in infants is likely to

improve their appetite, ingestion of food, and ultimately growth. It has been repeatedly

demonstrated that targeted use of hand hygiene (with soap and hand sanitizers) in day-care

centers and elementary schools in developed countries and in hospitals globally, decreases

iatrogenic infection rates (Curtis and Cairncross; 2003; Luby et al., 2004; Simon, 2004; Fewtrell

et al., 2005; Lee et al., 2005; Sandora et al., 2005; Luby and Curtis, 2008; Sandora et al., 2008).

However, no study yet provided similar evidence in non-institutional community settings in

developing countries. While soap is widely available in such resource-poor settings, clean water

for hand washing is often in short supply due to which frequent ‘soap and water’ hand washing is

often not possible.

3.2 Goals and Research Question

The goal of the trial was to gain a better understanding of the patterns of linear growth of low

birth weight infants and provide greater insights into ways to optimize linear growth of LBW

infants in low and middle income countries. The study was designed to answer the following

question: Can linear growth faltering among LBW infants be prevented or reversed through the

early provision of nutrition education and promotion of directed hand hygiene to caregivers, in

combination with later home fortification of complementary food using an improved formulation

of micronutrient powders (MNPs)?

43  

3.3 Objectives

3.3.1 Primary objective

To investigate the independent and combined effects of ‘water-based hand sanitizers (HS)’ and

‘MNPs’ in combination with ‘Nutrition, health and hygiene education (NHHE)’ to reduce rates

of stunting among LBW infants by prevention or reversal of linear growth faltering (Ha:

Combination of all three interventions will reduce rates of stunting by preventing or reversing

linear growth faltering).

3.3.2 Secondary objectives

(i) Investigate the independent and combined effects of intervention packages on reducing

rates of underweight and wasting

(ii) Investigate the independent effect of HS on neonatal infections

(iii) Investigate the independent effect of HS on reducing rates of gastrointestinal infections

(e.g. diarrhea) and acute respiratory tract infection (ARI) in the first six months

(iv) Investigate the independent and combined effect of HS and MNP on reducing rates of

diarrhea, ARI and other infections in the first year of life

44  

CHAPTER 4

STUDY DESIGN AND METHODS

4.1 Overview of the study design

This study followed a prospective 2X2 factorial, community-based cluster-randomized design to

ensure better representativeness and prevent contamination between intervention groups

(Donner, 2000). The design also supported the evaluation of the combined and independent

effects on infant growth and morbidity of the two main intervention components of the study: a)

Water-based hand sanitizer (HS) and b) micronutrient powders (MNPs) to be provided with

complementary foods. In the first phase of the study (0-6 months), 48 clusters were allocated to

either ‘HS’ or ‘no HS’ groups. The second phase included following up of the infants from age 7

to 12 months of both ‘HS’ and ‘no HS’ groups further allocating them to either ‘MNPs’, or ‘no

MNPs’ groups. All four groups were provided a third intervention component, ‘Nutrition, health

and hygiene education (NHHE)’, during the entire period of study (0-12 months). Thus, 48

clusters were randomly assigned as follows: from 0 to 6 months, i) NHHE only or ii) NHHE plus

HS (number of clusters were 24, for both); from 7-12 months i) NHHE plus both HS and MNP;

ii) NHHE plus HS; iii) NHHE plus MNP; and iv) NHHE only (number of clusters were 12, for

all 4 groups) (Figure 4.1).

4.2 Study participants

Dyads of LBW infants and their biological mothers were recruited for the study. Mothers were

provided nutrition and hygiene education and also served as the study respondents.

45  

4.3 Sample size

Based on data from a previous study in rural Bangladesh (Saha et al., 2009), estimates of

required sample size assumed a 20% reduction in the proportion of stunting (length-for-age <-2

Z-score) between Group 1 and Group 4 by the end of the intervention. Intra-cluster correlation

coefficient (ICC) was set low at 0.03 and the cluster sizes (number of subjects in a cluster, m)

were expected to be at least 6 (Aboud et al., 2008). The sample size was multiplied by a design

effect of 1.15, calculated using DE=1+ICC(m–1), to accommodate the clustering effect. The

sample was further adjusted for a potential 20% loss to follow-up over one year, thus requiring a

sample of 120 per group. We estimated that a total of 48 clusters inhabited by 480 low birth

weight infants would have 80% power (1-β) to detect the 20% reduction in the proportion in

stunting at 5% level of significance (α). Thus, the study aimed at a total sample of 480 LBW

infants with complete data retained for analysis of the primary outcome.

4.4 Study setting

The study was conducted in Bangladesh, a tropical country in South Asia (Figure 4.2 and Figure

4.3) with very high rates of infant mortality and stunting (NIPORT, 2011). Between 2010 and

2011 the national infant mortality rate was 43 deaths per 1,000 live births, the risk of dying in the

first month of life (32 deaths per 1,000 live births) was three times greater than in the subsequent

11 months of infancy (10 deaths per 1,000 live births), and deaths in the neonatal period

accounted for 60 percent of all under-five deaths (BBS/UNICEF, 2005). Linear growth of

surviving infants starts to falter from three months of age and progressively deteriorates until

they reach 24 months resulting in very high rates of stunting (Saha et al., 2009). According to the

only nationally representative community-based survey in Bangladesh, 36% of all infants were

46  

born with LBW, which was more than twice the 15% threshold that indicates a public health

problem (UNICEF, 2012). Estimates are that at least 77% of these LBW infants are growth

retarded at birth, confirming that intrauterine growth restriction was the major cause of LBW in

Bangladesh. The prevalence of LBW is higher in rural areas and the highest prevalence is found

in Dhaka division (province). The present study was conducted in two adjacent rural upazila

(sub-districts) in Dhaka division that are representative of many rural communities in

Bangladesh: Palash under Narsingdi district and Kaliganj under Gazipur district (Figure 4.4).

4.5 Ethical oversight

Verbal consent was taken from all pregnant mothers and their families to collect weight and

length of the newborn within 24 hours of birth. Written informed consent was obtained from the

respondents before enrollment if the infants met eligibility criteria (Appendix 5). The Research

Ethics Committees of Hospital for Sick Children, Canada and the James P. Grant School of

Public Health, BRAC University, Bangladesh provided ethical approval to the study (Appendix

6 and Appendix 7). The trial is registered with NCT01455636.

4.6 Sampling and randomization

The study sampling frame was constructed from a list of all pregnant married women identified

through a household survey in Kaliganj and Palash using a structured data collection form. We

assumed that at least 20% of pregnant women identified in the study area would give birth to

LBW babies, which is conservative compared to available data (BBSéUNICEF, 2005). A total of

2198 pregnant women were identified from 43,832 households and interviewed to ascertain a

date of last menstrual period (LMP) using a ‘calendar of local events’ prepared by study

47  

investigators to help women recall their LMPs. To reduce recall bias in estimating gestational

age women reporting LMP more than four months prior to the survey were not included in the

study. The information on LMP was also used to calculate the ‘Expected date of delivery (EDD)’

using a pre-plotted LMP-EDD calendar designed by Maternal Neonatal and Child Health

Program of BRAC, to identify infants born as full-term low birth weight (small for gestational

age, SGA) babies, to screen for exclusion infants born prematurely, and to organize the

interventions and schedule data collection home visits. The entire study area was divided into 48

clusters, i.e., the unit of randomization. A cluster represented the area covered by one Health and

Nutrition Assistant (HNA) who provided service to about 45-50 pregnant women. Each cluster

was randomly assigned to one of the four study groups. A computer generated random digit table

was used to ensure equal probability of allocation. The likely bias in field implementation was

eliminated by concealing allocation throughout the study.

4.7 Intervention Training

The intervention team Health and Nutrition Assistants (HNAs) and health and Nutrition

promoters (HNPs) received extensive basic training for seven days and monthly refresher

trainings for the entire study period. Trainings were organized and conducted separately for

different study groups.

4.8 Preparation for safe delivery, birth notification and community mobilization

The study team and local officials from the BRAC Development Program organized community

meetings and informed the local leaders about the importance of the study and distributed a four-

color pictorial information leaflets. Each pregnant woman was visited at 8th month of pregnancy

48  

and a group meeting was conducted including the respondent (the mother), a senior female

family member of the household, and if available, the potential birth attendant identified by the

family. Safe delivery kits (locally produced by the BRAC Health Program) were distributed free

of charge to all mothers. In these pre-delivery meetings, the importance of breastfeeding and

hand-washing in preventing newborn infections was discussed. Education on essential newborn

care (ENC), use of the safe delivery kit, hand washing and early initiation of breastfeeding, and

exclusive breastfeeding was provided along with demonstrations. In addition, leaflets with clear

hand-washing and breastfeeding messages were provided to the male and literate members of the

families. Particular emphasis was placed on birth notification by ensuring that husbands or other

male member of the families send the ‘birth notification card’ to the HNAs or call them

immediately after the birth of an infant.

4.9 Screening and enrolment

Within 24 hours of a birth, each infant-caregiver dyad was visited by an ‘Assessment Team’

(consisted of two trained Field Enumerators, FEs) that was kept separated from the ‘Intervention

team’ throughout the study period. The FEs performed anthropometric measurement of the

mother and the newborn and completed the recruitment form. Infants were eligible to be enrolled

in this study provided the following eligibility conditions were met and informed consent was

received from the respondents. Inclusion criteria consisted of: i) born as singletons and full term

(≥37 weeks of gestation); ii) birth weight: 1800 to <2500g; iii) families planning to reside in the

study area for at least a year; while exclusion criteria consisted of: i) severe illness; ii) birth

asphyxia; (iii) congenital abnormalities or severe malformations; and iv) mother’s health

49  

significantly compromised during birth. A total of 467 infants met the eligibility conditions and

thus were enrolled and followed for 12 months.

4.10 Intervention components

4.10.1 Directed hand-hygiene with hand sanitizers (HS)

The active ingredient of the hand sanitizer (HS) used in this study was Benzalkonium Chloride

(alkyl dimethyl benzyl ammonium chloride) 0.15% (w/w) (Figure 4.5). Additional ingredients

were aqua, polysorbate 20, aloe barbadensis and parfume. This is a nitrogenous cationic surface-

acting agent belonging to the quaternary ammonium group. Benzalkonium chloride solutions are

rapidly acting biocidal agents with a moderately long duration of action. It is one of the safest

synthetic biocides known and is safe for human use as disinfectants. The water-based hand

sanitizer was produced in India by Hexagon Inc and packaged into 250ml foam producing plastic

dispensers for the study. Dispensing in the form of foam was preferred over the standard gel

form because of the natural tendency to rub foam into one’s hands until it disappears, usually in

about 30 seconds. This extended rubbing time is crucial for disinfection.

In mid 2009, well before the start of the implementation, the investigators completed a formative

research to understand the common hygiene practice in the intervention area to prepare messages

for information, education and communication (IEC) materials and training modules. Given hand

sanitizers were not meant to replace the practice of hand washing with soap, it was necessary to

identify and separate prior to the intervention the ‘critical points’ for using hand sanitizer and

washing with soap (Table 4.1). The formative research also contributed to the formulation of the

hand sanitizer by providing information on the socio-cultural and religious aspects of its likely

use. At the beginning of the intervention each family received a dispenser full of water-based

50  

hand sanitizers, which was replaced as required throughout the intervention period. All

households received a poster including pictorial messages describing how to wash hands and

critical points for hand-hygiene.

4.10.2 Improved micronutrient powders (MNPs)

From the beginning of the infants’ 7th months (181 days), infants in the MNP groups (Groups 1

and 3) were assigned to receive one sachet of MNP per day for six months. Caregivers were

instructed to sprinkle the MNP contents of the sachet on the complementary foods before

feeding. A modified and improved formulation of MNPs with 22 micronutrients was used (Table

4.2) in this study. MNP was procured from a local pharmaceutical company in Dhaka,

Bangladesh (Renata Pharmaceuticals Ltd.). Caregivers were educated on the dosage, proper use,

and possible side-effects of MNP through use of a 4-color flipchart and practical demonstration

followed by an on-site practice to ensure appropriate usage.

4.10.3 Nutrition, health and hygiene education (NHHE)

Throughout the intervention period, caregivers of all study groups received simple, standardized,

and age- and culturally-appropriate nutrition and hygiene education that aimed to improve infant

feeding and prevent infections. A 4-colour pictorial flipchart was used to highlight the key

messages which included: i) Kangaroo Mother Care (KMC), a technique used for thermal

protection of the newborn through skin to skin contact between mother and newborn (Conde-

agudelo et al., 2011) ; ii) early initiation of breastfeeding; iii) exclusive breastfeeding; iv) timely,

adequate, age-appropriate and safe complementary feeding; vii) responsive feeding and

psychosocial stimulation; v) hand-washing with soap; vi) recognition of common diseases and

51  

care during illness; vii) danger signs (Appendix 8); and viii) father’s involvement and family

support. A behavior change communication strategy was designed using cognitive learning

theory, which emphasized coaching, demonstration, and practice for each message.

Mothers were counseled to interact and feed their infants in a responsive manner, including a

better recognition of hunger cues and increased verbal communication. In the first phase (0-6

months), each HNA visited the household every other day from 0-7 days postpartum; weekly

from 2nd week to 3 months postpartum and bi-weekly from 3-6 months. In the second phase (7-

12 mo), visits were scheduled for every other day in the first week of the 7th month, weekly until

8mo postpartum, and bi-weekly from 9-12months. The health and hygiene education emphasized

early recognition and home-based care of infections.

4.11 Intervention management and monitoring

Mothers were visited in a preplanned scheduled meeting by a Health and Nutrition Assistant

(HNA), who was recruited from the same community and was responsible for one cluster. Four

HNAs were supervised by one Health and Nutrition Promoter (HNPs). While the 48 HNAs were

responsible for the nutrition and hygiene education and distribution of MNPs and Hand

Sanitizers in respective clusters, the 12 HNPs were responsible for counseling on breastfeeding

and complementary feeding and problem-solving. HNPs would meet each HNAs separately in

their clusters on a weekly basis and would discuss the progress and problems faced during

implementation of the intervention and documented the meetings in ‘weekly cluster-level

meeting forms’. Every two weeks the HNPs reported to the 2 Field Supervisors (FS) (each

responsible for 24 clusters in each sub-district). The issues identified in this meeting were

documented in the ‘bi-weekly sub-division-level meeting forms’. At the end of every month FSs

52  

submitted all forms to the Project Coordinator and Functional PI. All critical issues identified

were documented in ‘monthly meeting forms’ and were discussed in the refresher training of

HNAs and HNPs during the following month.

4.12 Outcome measures

All infants were followed from 0-12 months in varying schedules to obtain data on the primary

and secondary outcomes. Information on demography, socio-economic status (SES), and other

covariates were collected during baseline (at birth). All study questionnaires were pretested in

the field (Appendix 9). Training modules for the ‘Assessment Team’ included: i) Data collection

methods and ii) Detailed description of tools to be used and explanation of questionnaires. The

outcome measures include the following:

4.12.1 Anthropometry

Recumbent length was measured to the nearest 1mm on a locally constructed high quality

wooden stadiometer. Weight was measured using a digital weighing scale with 15 g precision

(Tanita Model HD-314, Japan). Mid-upper-arm and head circumferences were measured with

non-stretchable plastic tapes. All measurements were performed twice. Monthly standardization

was also performed among the assessment group to monitor inter- and intra-observer variations.

All measuring equipments were calibrated monthly. Based on the methods described in the

MGRS, inter-observer variation was categorized into 3 categories: small, medium and large.

Among the quality assurance measurements done by the assessors for height and weight, nearly

all fell into the small category. For the few that were medium, the training was continued or

repeated until all fell into the 'small' category.

53  

4.12.2 Morbidity

Diarrhea was defined as >3 loose stools in a 24-h period or >1 loose stool containing blood.

Mothers were expected to use ORS and zinc tablets for all diarrhea episodes (provided free to

study subjects). ARI was diagnosed if the child had reported symptoms of cough or difficulty in

breathing, and rapid breathing with or without chest in-drawing. Episodes with chest in-drawing

were categorized as severe ARI. Information on the occurrence, type and severity of diarrhea and

ARI and other infections were collected weekly from 0-12 months.

4.12.3 Infant and young child feeding

Standardized infant and young child feeding questionnaires were used to obtain information on

age-specific feeding practices weekly from 0-6 months and bi-weekly from 7-12 months.

Estimate of dietary intakes was also recorded using 2-repeated multiple-pass 24-hour dietary

recall per month from 7-12 months.

4.13 Compliance and acceptability of intervention

Compliance was assessed by counting total number of bottles of hand sanitizer and MNP sachets

used by families at the end of the month using structured data collection forms. At the end of the

intervention, acceptability of and barrier to using hand sanitizers and MNPs and assessment of

the overall intervention was carried out qualitatively by using focus group discussions and semi-

structured interviews among randomly selected mothers.

4.14 Quality control and data management

Throughout the study the Assessment Team’ was kept independent of the ‘Intervention team.

Selection bias was minimized by strictly following the eligibility criteria. Measurement bias was

54  

minimized by following standard procedure in training the field workers by the same set of

trainers, using structured questionnaires, and standard methods for anthropometry and

hemoglobin analyses. All questionnaires and data forms were reviewed by researchers for

accuracy, consistency and completeness. To ensure data quality, the study supervisors and

investigators made random spot checks. In addition, a 10% sample of study children were re-

interviewed and re-measured within 48 hours of original interview by two data collectors for

quality control to provide continuous feedback on quality of data. Data were entered twice into a

custom designed Microsoft Access 2010 database program (Microsoft Corp, Redmond, WA).

Data were electronically checked for extreme values and were corrected after range-check and

mismatch check and considering biological plausibility.

4.15 Data analysis plan and statistical approach

Anthropometric indexes length-for-age (LAZ), weight-for-age (WAZ) and, weight-for-length

(WLZ) were calculated using ANTHRO 2011 (de Onis et al., 2006). The analyses was performed

using SAS for WINDOWS release 9.3 (SAS Institute, Cary, NC) on an intention-to-treat basis.

The primary outcome of the study were rates of stunting (LAZ <−2 Z-score). A changing trend

in Z-score over time may represent growth faltering or catch-up growth depending on the child's

preceding Z-score. Secondary outcomes included mean changes in LAZ, WAZ and WLZ;

prevalence of moderate and severe stunting; underweight; and infections (e.g. diarrhea, acute

respiratory tract infections, fever etc).

Baseline characteristics of the different intervention groups were examined for group

comparability to determine if randomization was successful. All outcome variables were checked

55  

for normality. Continuous and categorical outcomes in growth and morbidity for the four

intervention groups were compared by using analysis of variance (ANOVA) (or independent

sample t-tests as appropriate) and chi-square tests respectively. We used mixed effects regression

models using PROC MIXED for repeated measures of length-for-age, weight-for-age and

weight-for length Z scores. PROC GLIMMIX was used to compare proportions of infants with

infections over time. Analysis was performed taking into consideration the design effect of

clustering and controlling for confounding factors such as gender, anthropometric status at

enrolment, mother’s age, mother education level and household socioeconomic status. The main

predictor included in the analysis was treatment group. Kaplan-Meyer survival analysis method

was used to determine cumulative probability of recovery from severe or moderate stunting

among different groups.

4.16 Benefits to the control communities

Nutrition, health, and hygiene education intended to improve breastfeeding, complementary

feeding and other nutritional and health practices among all subjects. Irrespective of their

eligibility to be included in the intervention group, the participant mothers received safe delivery

kits at birth while their children received zinc tablet and oral rehydration solutions (ORS) for all

episodes of diarrhea throughout the study. Promotion of hand washing with soap likely benefited

the families by reduction in the incidence of infections.

56  

 

Table 4.1. Critical points for hand washing and using hand sanitizers When to wash hands with soap When to use hand sanitizers

Before touching the newborn Before preparing food Before eating Before feeding a child After defecation After cleaning child’s anus After cleaning house After handing waste

Before touching the newborn Before eating snacks After sneezing or coughing or picking nose After returning home from field or outside After handling animals After attending a sick person After treating a wound

57  

 

Table 4.2. Composition of Improved Micronutrient Powder (MNP) 1

Micronutrients Amount/ sachets

Detailed technical specifications

Vitamin A 300 µg 999 IU (as Palmitate, USP-FCC)2

Vitamin D 5 µg 200 IU (as Cholecalciferol, USP-FCC)3

Vitamin E 6 mg α-TE 9 IU (as dl-alpha-Tocopheryl Acetate, USP-FCC)4

Vitamin B1 0.5 mg 0.5 mg (as Thiamin HCl, USP-FCC) Vitamin B2 0.5 mg 0.5 mg (as Riboflavin, USP-FCC) Vitamin B6 0.5 mg 0.5 mg (as Pyridoxine HCl, USP-FCC) Vitamin B12 0.5 µg 0.5 mcg (as Cyanocobalamin, USP) Niacin 6.0 mg 6 mg (as Niacinamide, USP-FCC) Folic Acid 160 µg 0.16 mg (USP) Vitamin C 30 mg 30 mg (as Ascorbic Acid, USP-FCC) Pantothenic Acid 1.8 mg 1.8 mg (as Calcium d-Pantothenate, USP-FCC) Biotin 6.0 µg 6 mcg (USP) Vitamin K 20 µg 20 mcg (as Phytonadione, USP) Calcium 100 mg 100 mg (as Calcium Carbonate, USP) Copper 0.5 mg 0.5 mg (as Copper Gluconate, USP) Iodine 90 µg 90 mcg (as Potassium Iodide, USP-FCC) Iron 10 mg 10 mg (as Ferrous Fumarate) Magnesium 20 mg 20 mg (as Magnesium Oxide, USP) Phosphorus 100 mg 100 mg (as Dicalcium Phosphate, FCC) Zinc 10 mg 10 mg (as Zinc Gluconate, USP) Selenium 20 µg 20 mcg (as Sodium Selenite) Manganese 0.6 mg 0.6 mg (as Manganese Sulfate (1H2O), USP-FCC) 1 Dose: one sachet per day, from 6-12months 2 1RE vit A=3.3 IU 3 1 µg vit D=40 IU 4 1mg TE=1.5 IU.

 

58  

Figure 4.1. Study design: 2X2 factorial cluster randomized controlled trial

Hand Sanitizer (HS)

(24 clusters)

 

No Hand Sanitizers

(24 clusters)

 

MNP

(12 clusters)

No MNP

(12 clusters)

Randomization of 48 clusters (at 0 mo)

Enrolment: at birth

Randomization (at 6 mo) Randomization (at 6 mo)

Identification of mothers (at 8th months of pregnancy)

MNP

(12 clusters)

No MNP

(12 clusters)

Phase I

(0-6mo)

Phase II

(6-12mo)

Nutrition, Health and Hygiene

Education (NHHE)

(48  clusters)  

 

GROUP  1    

(NHHE  +  HS  +  MNP)  

 

Preparatory Phase

(Before birth)  

 

 

GROUP  2    

(NHHE  +  HS)  

 

 

GROUP  3    

(NHHE  +  MNP)  

 

 

GROUP  4  

(NHHE  Only)    

 

59  

Figure 4.2. Map of South Asia.

60  

Figure 4.3. Map of Bangladesh and location of study sites.

Study  Site  1  

Study  Site  2  

61  

   

Figure 4.4. Study sites: two adjacent sub-districts, Kaliganj in Gazipur; and Palash in Narsingdi, in Dhaka division, Bangladesh 1

1The gray dots represent ‘NHHE+HS’ clusters and white dots represent ‘NHHE only’ clusters.

 

62  

 

Figure 4.5. Chemical structure of Benzalkonium Chloride.  

63  

CHAPTER 5

PAPER 1

Benzalkonium chlroride containing hand sanitizer reduces minor infections but did not

impact linear growth among full-term low birth weight infants in the neonatal period: a

2X2 factorial cluster randomized trial in rural Bangladesh

Abstract

Background: Of the 4 million neonatal deaths that occur every year in low- and middle-income

countries infections account for about half. Low birth weight (LBW) infants are at highest risk

for both infections and post-natal growth faltering.

Objective: To investigate the use of water-based hand sanitizer combined with nutrition and

hygiene education to prevent neonatal infection and improve linear growth among full-term

LBW infants during the first month of life.

Methods: A community-based, cluster randomized, controlled trial was conducted in rural

Bangladesh using 48 clusters randomly assigned to two groups: i) Nutrition, health and hygiene

education (NHHE) plus directed use of benzalkonium chloride-based hand sanitizer (HS) by

mothers and other family members; ii) NHHE only. We followed 467 LBW infants with weekly

morbidity reports and base-line and end-point anthropometry during the first four weeks of life.

64  

Results: At 4 weeks postpartum there was no difference in neonatal sepsis although infants in

the ‘NHHE+HS’ group had reduced rates of fever (P<0.01); skin rashes (P<0.001) and apparent

jaundice (P<0.05) compared to the ‘NHHE only’ group. There were two neonatal deaths in the

‘NHHE only’ group and one in the ‘NHHE+HS’ group. Reported household expenditure on care

seeking due to neonatal illness was significantly less in the ‘NHHE+HS’ group compared to the

‘NHHE only’ group (P<0.01). Mean ± SD anthropometric measures were comparable between

groups both at baseline and end-point (weight: 2.2±0.2 vs. 2.3±0.2 kg; length: 44.1±1.8 vs.

43.9±1.8 cm, at baseline; and weight: 2.9±0.2 vs. 3.0±0.1 kg; length: 47.8±1.3 vs. 48.1±1.4 cm,

at end-point).

Conclusions: Although the use of hand sanitizer reduced minor morbidities, its use did not

impact linear growth during the neonatal period.

Trial registration: http://www.clinicaltrials.gov (identifier NCT01455636)

Key words

Low birth weight, neonatal infections, linear growth, weight, length, hand-washing, developing

countries

(We intent to publish the article in a peer reviewed journal.)

65  

Introduction

The fourth Millennium Development Goal (MDG-4) targets the reduction of childhood mortality

less than five years of age by two-thirds by the year 2015. Thirty-eight per cent of this mortality

is attributable to deaths in the neonatal period (0-28 days) (Lawn et al., 2005). The Lancet

neonatal survival series highlighted the gaps existing internationally, regionally and at country

level in terms of policy, advocacy and implementation (Lawn et al., 2005). Globally, 4 million

neonatal deaths occur every year in low- and middle-income countries and infections account for

an estimated 1·44 million deaths (Lawn et al., 2005; Black et al., 2010). In regions with high

neonatal mortality, almost half of the deaths are attributable to neonatal infections (Lawn et al.,

2005, Baqui et al., 2006). It has been suggested that up to three-quarters of the perinatal and

newborn deaths could be prevented at an additional cost of less than U$1 per infant with the use

of low cost technology and a set of evidence-based practices (Darmstadt et al., 2005).

In Bangladesh and similar resource-poor settings, deliveries usually occur at home in unhygienic

conditions assisted by unskilled birth attendants (Black et al., 2010). A review of umbilical cord

infections in Oman noted that the incidence rate in home-delivered neonates was 5·9 times

higher than in hospital births (Sawardekar et al., 2004). The presence of umbilical cord infection

and symptoms of neonatal sepsis have been directly associated with increased mortality risk

(Mullany et al., 2011). In South Asia, community-based case-control studies have focused on

risk factors for life-threatening illnesses like neonatal tetanus or sepsis, however, the high burden

of upper respiratory tract illnesses, fever and hypothermia suggests that ‘minor’ infections are

very common in the neonatal period (Darmstadt et al., 2005; Baqui et al., 2008).

66  

Infections are associated with impaired nutrient absorption and utilization and nutrient loss and

the effect of these infections on growth can be significant, even in the first months of life

(Scrimshaw et al.,1959; Mata et al., 1977). Recurrent episodes of infection in the first weeks of

life may play a pivotal role in influencing rates of growth among neonates (Bhutta et al., 2006;

Stephensen et al., 1999; Shankar et al., 1998). Low birth weight (LBW) infants are particularly

vulnerable to such infections due to their compromised immune system (Singh et al., 1978;

Raqib et al., 2007). In developing countries, there are few data available on the impact of

infections on growth in the neonatal period, especially among the LBW infants born at home. In

this study, we hypothesize that interventions to improve exclusive breastfeeding and hand

hygiene to control infections in the neonatal period could permanently place these children on a

more normal growth trajectory, thereby reducing the prevalence of stunting.

Hand-washing with soap and water can significantly reduce infectious diseases among children

in developing countries (Luby et al., 2004; Luby et al., 2005; Luby et al., 2008). In a recent

meta-analysis of data from cluster-randomized controlled trials carried out in low and middle

income countries, it was demonstrated that there was a positive effect of water sanitation and

hygiene (WASH) interventions on linear growth of children under five years of age (Dangour et

al., 2013). However, in these resource-poor settings adherence to frequent hand-washing requires

access to water and soap, which are not always readily available (Scott et al., 2008; Luby et al.,

2009). With limited access to water, frequent hand washing with ‘soap and water’ is often not

possible. Hand sanitizers are regarded as effective alternatives to hand washing with soap and

water in institutional-based settings since they do not require rinsing, and can disinfect most

bacteria, protozoa and several types of viruses (Kampf et al., 2004).

67  

Although the current WHO-recommended hand sanitizer formulations for health-care facilities

are alcohol-based (Pittet et al., 2009) their acceptability is often reduced by their tendency to

cause excessive dryness and skin irritation (Larson et al., 2006). Moreover, due to their high

flammability, alcohol-based hand sanitizers may pose a fire hazard if they are exposed to an open

household cooking flame. In addition, accidental ingestion of alcohol-based products may lead to

poisoning, especially in young children (Miller et al., 2009). Due to these inherent disadvantages

of alcohol-based hand sanitizers, the efficacy of alcohol-free, water-based hand sanitizers have

been investigated (White et al., 2001), however, their efficacy to prevent infections in a

community setting has not been widely investigated and no study in developing countries has

targeted LBW neonates.

The objective of the present study was to investigate the effect of a water-based hand sanitizer to

reduce neonatal infections and improve rates of linear growth of LBW infants in the first month

of life in a setting with limited access to water for hand washing.

Methods

Overview of study design

The complete design of this prospective randomized trial to explore efficacy of a water-based

hand sanitizer (HS) on linear growth in the first year of life among the low birth weight infants is

presented in Chapter 4. The present chapter presents data on 1-month post-natal growth, neonatal

infections and weekly morbidity in the neonatal period (0-28 days).

68  

Study participants

Full-term LBW infants from two rural communities in Bangladesh (Figure 1) were recruited as

study subjects along with their biological mothers. Inclusion criteria consisted of: i) born as

singletons and full-term (≥37 weeks of gestation); ii) birth weight: 1800 to <2500g; iii) families

planning to reside in the study area for the study duration. Exclusion criteria consisted of: i)

severe illness; ii) birth asphyxia; (iii) congenital abnormalities or severe malformations; and iv)

mother’s health significantly compromised during birth.

Sampling and randomization

A total of 2198 pregnant women were identified and interviewed to ascertain the date of their last

menstrual period (LMP) using a ‘calendar of local events’ to help women recall their LMPs. To

reduce recall bias in estimating gestational age, women reporting their LMP more than four

months prior to the survey were not included in the study. The information on LMP was also

used to calculate the expected date of delivery (EDD) using a pre-plotted LMP-EDD calendar

designed by Maternal Neonatal and Child Health Program of BRAC. The entire study area was

divided into 48 clusters, i.e., units of randomization. A cluster represented the area covered by

one Health and Nutrition Assistant (HNA) who provided service to about 45-50 pregnant

women. Each cluster was randomly assigned to one of the two study groups. A computer

generated random digit table was used to ensure equal probability of allocation. The likely bias

in field implementation was eliminated by concealing allocation throughout the study. Education

on essential newborn care and community mobilization was started at beginning of the 8th month

of pregnancy, which was same for both the study groups.

69  

Screening and enrolment

Within 24 hours of birth, each newborn-mother dyad was visited by an ‘Assessment Team’

consisting of two trained Field Enumerators (FEs). The FEs performed anthropometric

measurement of the mother and the newborn and completed the recruitment form. A total of 467

infants met the eligibility conditions and thus were enrolled and followed for 4 weeks.

Ethical oversight

Verbal consent was obtained from all pregnant mothers and their families within 24 hours of

birth prior to weight and length measurements of their newborns. Written informed consent was

obtained from the respondents before enrolment if the infants met eligibility criteria. The

Research Ethics Committees of Hospital for Sick Children, Canada and the James P. Grant

School of Public Health, BRAC University, Bangladesh provided ethical approval for the study.

The trial is registered with NCT01455636.

Intervention components

Directed hand-hygiene with hand sanitizers (HS)

The study investigated the effect of hand sanitizers to reduce neonatal illness. The active

ingredient of the hand sanitizer (HS) was benzalkonium chloride (0.15% alkyl dimethyl benzyl

ammonium chloride). This is a nitrogenous cationic surface-acting agent belonging to the

quaternary ammonium group. Benzalkonium chloride solutions are rapidly acting biocidal

agents, with a moderately long duration of action. It is water-based and is one of the safest

synthetic biocides known and is safe for human use as disinfectants. The hand sanitizers for this

trial were produced in India by Hexagon Inc. and were packaged into 250ml foam producing

70  

plastic dispensers. Dispensing in the form of foam was preferred over the standard gel form

because of the natural tendency to rub foam into one’s hands until it disappears, usually in about

30 seconds. This extended rubbing time is crucial for disinfection.

Given that hand sanitizers were not meant to replace the practice of hand washing with soap, it

was necessary to identify the critical points for using hand sanitizer and washing with soap and

water. This was accomplished through a formative research project to understand common

hygiene practices in the intervention area and to prepare information, education and

communication (IEC) materials and training modules.

At the beginning of the intervention, each family received a dispenser containing water-based

hand sanitizer, which was replaced as required throughout the intervention period. All

households received a poster including pictorial messages describing how to wash their hands

and critical points for hand-hygiene. Figure 2 presents the critical points recommended for

‘washing hands with soap and water’ and ‘using hand sanitizer’. A demonstration session

involving all members of the household on hand-washing and the use of hand sanitizer was also

conducted.

Nutrition, health and hygiene education (NHHE)

Throughout the neonatal period, caregivers of both study groups received simple, standardized,

and age- and culturally-appropriate nutrition and hygiene education that aimed to improve infant

feeding and prevent infections. A four-colour pictorial flipchart was used to highlight the key

messages that included: i) Kangaroo Mother Care (KMC), a technique used for thermal

protection of the newborn through skin to skin contact between mother and newborn; ii) early

71  

initiation of breastfeeding; iii) exclusive breastfeeding; v) hand-washing with soap and water; vi)

recognition of common diseases and care during illness; and vii) danger signs associated with

neonatal complications. A behavior change communication strategy was designed using

cognitive learning theory that emphasized coaching, demonstration and practice for each

message. The intervention were delivered by 48 Health and Nutrition Assistants (HNAs) and 12

Health and Nutrition Promoters (HNPs) who consisted of the ‘Intervention team’.

Outcome measures

Information on maternal, household socio-economic status (SES), and pregnancy and delivery

care were collected during enrolment. Linear growth was the primary outcome measure of this

study. Recumbent length was measured to the nearest 1mm on a locally constructed high quality

wooden stadiometer. Other anthropometric data were also collected: weight was measured using

a digital scale with 15g precision (Tanita Model HD-314, Japan) and head circumferences were

measured with a non-stretchable plastic tape. All measurements were performed twice and all

measuring equipment was calibrated monthly. Data on neonatal complications were collected at

the end of neonatal period (28th day) and morbidity surveillance data were collected weekly for 4

weeks using structured questionnaires. Information on care seeking due to neonatal illness was

also collected at the end of neonatal period.

Quality control and data management

Throughout the study the ‘Assessment Team’ was kept independent of the ‘Intervention team’.

Selection bias was minimized by strictly following the eligibility criteria. Measurement bias was

minimized by following standard procedure in training the field workers by the same set of

72  

trainers, using structured questionnaires, and standard methods for anthropometry and

hemoglobin analyses. Monthly standardization was also performed among the assessment group

to monitor inter- and intra-observer variations. "Based on the methods described in the MGRS,

inter-observer variation was categorized into 3 categories: small, medium and large. Among the

quality assurance measurements done by the assessors for height and weight, nearly all fell into

the small category. For the few that were medium, the training was continued or repeated until

all fell into the 'small' category." All questionnaires and data forms were reviewed by researchers

for accuracy, consistency and completeness. To ensure data quality, the study supervisors and

investigators made random spot checks. In addition, a 10% sample of study children were re-

interviewed and re-measured within 48 hours of original interview by two data collectors for

quality control to provide continuous feedback on quality of data. Data were entered twice into a

custom designed Microsoft Access 2010 database program (Microsoft Corp, Redmond, WA).

Data were electronically checked for extreme values and were corrected after range-check and

mismatch check and considering biological plausibility.

Data analysis and statistical approach

Characteristics of newborn baby, mother and household socio-demographic status of the

intervention groups were examined for group comparability to determine if randomization was

successful. The primary outcome of the study was length-for-age Z-scores (LAZ). Secondary

outcome included mean changes in length, weight, and head circumference; indexes of weight-

for-age Z-scores (WAZ) and weight-for-length Z-scores (WLZ). Anthropometric indexes (LAZ,

WAZ, WLZ) were calculated using ANTHRO 2011.

73  

In addition, data on morbidity were collected weekly as secondary outcome. Presence of any

neonatal complications (danger signs associated with neonatal sepsis) and health care seeking

behaviour were collected at the end of neonatal period. Every week, body temperature readings

were recorded on the Fahrenheit scale. Respiratory rate was measured if the infant was reported

to have rapid or difficulty breathing. All continuous outcome variables were checked for

normality. For continuous and categorical outcomes in growth and morbidity for the two

intervention groups were compared by using independent sample t-test and chi-square test

respectively. PROC GLIMMIX was used to compare proportions of infants with infections over

time. The analyses were performed using SAS for WINDOWS release 9.3 (SAS Institute, Cary,

NC) on an intention-to-treat basis.

Results

Figure 5.1 shows the trial profile and outcomes for the subjects from the 48 clusters. Between

October 14, 2010 and April 30, 2011, a total of 1295 infants were born in the study area and

were assessed for enrolment in the trial. Of these, 814 infants did not meet the eligibility criteria

and 14 families declined to participate. A total of 467 newborn babies were enrolled in the

‘NHHE only’ (n=236) and ‘NHHE+HS’ (n=231) groups with approximately 10 infants per

cluster.

At enrolment, maternal demographic and household socioeconomic characteristics were well

balanced across the two study groups (Table 5.1). Maternal care seeking during the antenatal

period was low and almost half of the mothers did not receive iron tablets or TT injections

during pregnancy. About 80% mothers delivered at home and the safe delivery kits, provided to

74  

the families free of cost, were used almost universally. In almost all home deliveries, a new and

sterile blade was used to cut the umbilical cord.

Analysis of essential newborn care (ENC) practices data reveal that compliance to the ENC

practices were comparable between groups (Table 5.2). More than two-thirds of the newborn

(70%) were breastfed within one hour of birth and almost all infants were breastfed on demand

or at least every two hours. Feeding expressed breast milk (EBM) was recommended only if the

low birth weight infants had sucking problems. Use of EBM was low (15%) in both groups.

More than 85% mothers provided skin-to-skin contact for thermal care and infants were kept

warm almost universally by being covering head to toe with a soft blanket. Although delayed

bathing for 7 days and delayed shaving of the head until 28 days after birth were recommended

as essential newborn care for LBW babies, fewer than a third mothers followed these specific

recommendations. As a component of the essential newborn care recommendations, about two-

thirds reported not applying mustard oil, boric powder or other traditional products to the

umbilical cord, a common practice in Bangladesh.

Table 5.3 presents the data collected on complications associated with neonatal sepsis at end-

line. Throughout the neonatal period (0-28 days), there were few neonatal complications in both

groups with no significant differences between groups.

In addition to the neonatal complications collected at end-line, neonatal morbidity surveillance

data were collected weekly (Figure 5.2). Analysis of the morbidity surveillance data identified

that a lower proportion of neonates in the ‘NHHE+HS’ group were reported to have had fever

75  

(P<0.01); skin rashes (P<0.001) and jaundice (P<0.05) compared to the ‘NHHE only’ group.

However, there were no differences in the prevalence of upper or lower respiratory tract or eye

infections.

Data on actual body temperature collected by the field enumerators also showed that average

body temperature was significantly higher in the ‘NHHE only’ group compared to the

‘NHHE+HS’ group (98.4±1.06 °F vs. 97.2±0.96 °F P<0.001). Prevalence of fever (defied as

temperature >100°F) was significantly higher in the ‘NHHE only’ group compared to the

‘NHHE+HS’ group (7.6% vs. 3.4%, P<0.05). Respiratory rates (RR) were objectively counted

by the data collectors only from infants who had reported to have breathing difficulties. No

differences were found between groups (data not shown).

Household expenditure for treatment seeking for neonatal illness is presented in Figure 5.3.

Total amount spent for any neonatal illness was significantly less in the ‘NHHE+HS’ group

compared to that of the ‘NHHE only’ group (Tk.2143 ± 305 vs. Tk.3057± 443, P<0.01).

Similarly, total distance travelled (in kilometer) in seeking care for neonatal infections was also

significantly less in the ‘NHHE+HS’ group compared to the ‘NHHE only’ group (2.87 ± 3.9 km

vs. 4.23 ± 3.9 km, P<0.05, data not shown).

Weight, height, and head circumference at 1month post-partum were no different between

groups (Figure 5.4). Similarly, no differences were found in one-month postnatal length-for-age

Z-score (LAZ), weight-for-age-Z-score (WAZ) and weight-for length Z-score (WLZ) between

groups after adjusting for gender, maternal age, maternal education and clustering effects.

76  

Discussion

The combined intervention of directed hand-sanitized use and ‘nutrition, health and hygiene

education’ did not result in improved linear growth during the first four weeks of life compared

to ‘nutrition, health and hygiene education’ alone, although our data suggests fewer non-severe

infections in the hand-sanitizer group. To our knowledge, this was the first study to explore the

efficacy of water-based hand sanitizers to reduce infections among low birth weight infants in a

resource-poor setting. It should be noted that there was no true ‘no-intervention’ control group

since requesting one group to not wash their hands would have been unethical.

The major strength of the study is its robust design to investigate the impact of hand sanitizers in

combination with an ‘essential nutrition, health and hygiene education package’. The cluster

randomized design was used to ensure better representativeness and prevent contamination

between intervention groups (Donner et al., 2000). An important aspect of our study was that our

study intervention was formulated based on formative research on hand sanitizers during the

design phase. All intervention and assessment tools were prepared taking into consideration the

findings of the formative research. For example, we used separate messages for ‘hand-washing

with soap and water’ and ‘using hand sanitizer’ so that hand sanitizers did not replace or compete

with the regular hand-washing practices of the households. Data quality was ensured through

rigorous routine field supervision and ongoing training in data collection methods.

It is important to note that the primary outcome of the study was linear growth and the study was

not powered to detect differences in neonatal sepsis or other complications. Our observation of

77  

few neonatal complications in this high risk population should be interpreted in light of the fact

that both groups received an intensive package of essential newborn care and feeding

recommendations as well as training on hand-washing with soap and water. Moreover, all

mothers were provided with a safe delivery kit free of cost including a sterile blade to cut the

umbilical cord. Although there has been a significant reduction in the neonatal mortality rate

(NMR) in Bangladesh over the last decade (from 41 per 1000 live births, in 2000 to 27 per 1000

live births, in 2010 (NIPORT, 2011), LBW remains a major risk factor for neonatal mortality. A

study conducted in a periurban setting in Bangladesh in 2001 showed that the NMR in LBW

infants were approximately double those of full-term normal weight infants (133 per 1000 live

births) (Yasmin et al., 2001). Although this study did not distinguish between those born LBW

full-term and LBW preterm, the authors concluded that preterm birth was the most important

contributing factor to NMR. In our study, we had few neonatal deaths most likely because at

enrollment we excluded those born prematurely and did not include infants whose birth weight

were less than 1.8 kg. Another reason for the low mortality may have been that these infants

were in an intensive research project that emphasized proper essential newborn care for both

groups, equally. A study in Ethiopia to determine the etiology of neonatal septicemia found that

low birth weight was strongly associated with blood culture proven neonatal sepsis (Shitaye et

al., 2010). Another study conducted in Nigeria found that birth weight was a significant predictor

of mortality (OR = 0.04; 95% CI: 0.005-0.310, P= 0.002) (Onwuanaku et al., 2011).

 

One possible limitation of our study was that the documentation of neonatal complications was

based on self-reported data by the mothers rather than direct observation by our field staff.

However, to assist mothers to recognize neonatal complications and to avoid under- or over-

reporting, a four-color pictorial poster was used with pictures of neonatal danger signs

78  

(Appendix 8). The weekly morbidity data (symptoms that occurred in the previous seven days)

may also have been limited by recall bias. One study in Brazil concluded that parental recall was

an unreliable method to estimate incidence of diarrhea among children; the longer the recall

period, the greater was the error in under-reporting episodes of diarrhea (Melo et al., 2007). In

the current study, only respiratory rates and body temperature were measured directly by data

collectors every week for the entire first month.

Findings from the present study suggest that use of hand sanitizers in combination with nutrition,

health and hygiene education may reduce minor infections among the low birth weight neonates.

The study also found that household expenditure on care seeking related to neonatal illness was

lower in the group that used hand sanitizers. However, despite apparent differences in fever and

minor infections, there was no impact on length or weight growth at the end of neonatal period.

There is independent evidence that hand washing with soap and water is an effective intervention

to prevent infections and other evidence to demonstrate that the use of alcohol-based hand

sanitizers is also effective in preventing infections. Few of the studies that examined the value of

hand washing and hand sanitizers took place in a community setting, nor were they compared to

routine hand-washing. In our study it would have been inappropriate (and unethical) to

recommend that subjects stop the practice of washing their hands with soap and water. Thus our

design assumed that hand washing and the use of the hand sanitizer would be better than hand

washing alone. Additionally, both groups received the most up-to-date recommendations on the

use of an intensive package of essential newborn care and feeding recommendations for LBW

babies, which also would impact on the risk of infection. So, again, our design assumed that

preventive care and hand sanitizer use would be better than preventive care alone. Given the

79  

results of our study, we can only conclude that in this setting and under the circumstances of the

study, water-based hand sanitizer was not an effective intervention.

In summary, caregiver use of hand sanitizer did not result in differential growth of term, low

birth weight, rural Bangladeshi infants in the neonatal period. We did document a small decline

in subjective symptoms associated with infections and the use of hand sanitizer was associated

with reduced cost of neonatal care seeking suggesting that it may have impacted on the severity

of non life-threatening infections. Identifying a research design in a community setting to test the

efficacy of water-based hand sanitizers in high-risk groups is a significant challenge.

80  

1 Data are number (percentage), unless otherwise mentioned; groups were compared with chi-square test for proportions; 2 Data are mean (SD); groups were compared with independent sample t-test; 3 Safe delivery kit was provided to all mothers free of cost.

Table 5.1. Maternal, demographic and household socioeconomic characteristics by study group at enrolment. Socio-demographic characteristics NHHE only

(n=231) NHHE+HS (n=236)

P value

Socio-economic status Religion of household: Islam 215 (93.1)1 222 (94.1) 0.579 Ownership of house 217 (93.9) 223 (94.4) 0.798 Electricity in home 182 (78.8) 182 (77.1) 0.664 Access to cell phone 180 (77.9) 182 (77.1) 0.835 Ownership of farming land 2 25.80 (57.3) 20.97 (54.7) 0.352 Expenditure on food in the last month 2 4764.9

(2599.7) 4756.8 (2402.2)

0.972

Water sanitation Access to tap water in household 4 (1.7) 2 (0.8) 0.292 Access to tube-well water in the household 131 (56.7) 142 (60.1) 0.384 Access to sanitary latrine 43 (18.61) 46 (19.5) 0.341 Maternal Characteristics Mother’s age (years)2 23.64 (5.4) 23.61 (5.4) 0.938 Mothers education (Illiterate) 188 (80.8) 186 (79.1) 0.634 Mothers education level (years of schooling)2 6.28 (6.9) 6.13 (3.1) 0.777 Antenatal care Number of antenatal care (ANC) Visits 2 2.74 (1.8) 2.61 (1.7) 0.497 Received iron supplementation 128 (55.4) 120 (55.4) 0.354 Received tetanus toxoid (TT) injection 121 (52.4) 127 (53.8) 0.952 Delivery care Delivered at home 177 (77.3) 191 (82.3) 0.516 Used safe delivery kit during childbirth3 145 (94.4) 156 (98.6) 0.859 Used sterile blade for cutting umbilical cord3 145 (96.9) 153 (97.9) 0.352

81  

1 Data are number (%); groups were compared with chi-square test; 2 Expressing breast milk was recommended if newborn was unable to suckle milk.

Table 5.2. Comparison of newborn care practices among full term low birth weight infants in the neonatal period (0-28 days) by study group. Essential newborn care NHHE only

(n=231) NHHE+HS (n=232)

P value

Feeding colostrum to the newborn 225 (96.1)1 223 (94.7) 0.378 Breastfed newborn within one hour of birth 162 (70.1) 164 (70.7) 0.919 Breastfed the newborn exclusively 223 (95.8) 226 (95.5) 0.770 Breastfed the newborn on demand or at least every two hours

221 (95.7) 222 (94.5) 0.670

Fed expressed milk with cup or spoon 2 35 (15.2) 35 (15.0) 1.000 Covered newborn from head to toe with soft cloth 223 (96.9) 228 (96.9) 1.000 Skin-to-skin care (kangaroo mother care) 205 (88.7) 201 (85.5) 0.334 Withhold bathing until seven days after birth 78 (33.8) 86 (36.6) 0.561 Withhold shaving hair until 28 days after birth 51 (22.1) 55 (23.4) 0.742 Avoiding applying anything to the umbilical stump 155 (67.1) 169 (72.2) 0.267

82  

Table 5.3. Comparison of complications among full term low birth weight infants in the neonatal period (0-28 days) by study group. Neonatal infections3 NHHE only

(n=231) NHHE+HS (n=232)

P value

Stopped breastfeeding/ unable to suck 3.1 (4)1,2 0.7 (1) 0.191 Umbilical infection or discharge 1.4 (2) 0.8 (1) 1.000 Lethargic 0.7 (1) 0.0 (0) 0.100 Redness of eye 2.8 (4) 0.0 (0) 0.125 Shivering or hypothermia 0.7 (1) 0.0 (0) 1.000 Difficult or fast breathing 4.9 (7) 3.1(4) 0.548 Chest in-drawing 0.7 (1) 0.8 (1) 1.000 Convulsion 0.0 (0) 0.8 (1) 0.471 Distended abdomen 1.4 (2) 2.3 (3) 0.669 Severe vomiting4 2.1 (3) 1.6 (2) 1.000 1 Data are number (%); groups were compared with chi-square test; 2 Groups were compared with Fisher exact test; 3 All complications were self-reported by the infant’s mother on a weekly basis; 4 Severe vomiting was defined as throwing up everything the newborn was fed.

83  

1 Values are Mean ± SD. Z-scores are calculated using World Health Organization Growth Standard (Anthro 2011);

2 Groups were compared with independent sample t-test at each time point; 3 All values at 28 days were significantly different (P<0.01) compared to values at birth using independent sample t-test.

Table 5.4. Anthropometry of full term low birth weight infants at birth (baseline) and end of 28 days (end-point) by study group. Anthropometry NHHE only

(n=231) NHHE+HS (n=236)

P value

At birth Length (cm) 43.9 ± 1.81 44.1 ± 1.8 0.212 Weight (kg) 2.2 ± 0.2 2.3 ± 0.2 0.44 Head circumference (cm) 32.4 ± 1.0 32.4 ± 1.1 0.86 Length-for-age Z-score -3.0 ± 0.9 -2.9 ± 0.9 0.50 Weight-for-age Z-score -2.5 ± 0.5 -2.4 ± 0.5 0.42 Weight-for-length Z-score -1.3 ± 0.7 -1.4 ± 0.9 0.66 At the end of 28 days Length (cm) 47.8 ± 1.33 48.1 ± 1.4 0.20 Weight (kg) 3.2 ± 0.1 3.2 ± 0.1 0.16 Head circumference (cm) 35.3 ± 1.3 35.4 ± 1.2 0.45 Length-for-age Z-score -2.1 ± 1.0 -2.2 ± 1.0 0.24 Weight-for-age Z-score -1.9 ± 0.8 -2.0 ± 0.8 0.22 Weight-for-length Z-score -0.3 ±1.5 -0.3 ± 1.3 0.80

84  

Figure 5.1. Trial profile of the study (0-28 days). 1 NHHE, nutrition health and hygiene education (received by both groups); 2 HS, hand sanitizers.

 

24 clusters allocated to ‘No HS’ (n=231)

 

467 newborns enrolled in 48 clusters and received NHHE1

24 clusters allocated to ‘HS’2 (n=236)

Study area divided into 48 clusters (with 45-50 pregnant women per cluster)  

828 excluded [814 did not meet inclusion criteria (Birth weight not between 1.8 -<2.5 kg=805; severe illness or birth asphyxia=3; born <37 weeks of gestation=2; twin=4); and 14 refused to participate]

1295 newborns assessed for eligibility at birth  

229 completed trial

GROUP 2 (NHHE only)

2 died 0 migrated out 0 declined  

 

 

1 died 0 migrated out 0 declined  

 

 

48 clusters randomized at birth

235 completed trial

GROUP 1 (NHHE +HS)

 

85  

Figure 5.2. Point prevalence of reported morbidity among full term low birth weight infants in the neonatal period (0-28 days) by week by study group.1, 2

1NHHE, nutrition health and hygiene education (received by both groups); HS, hand sanitizers. 2The study groups were compared using PROC GLIMMIX.

P = NS P<0.05

P<0.001

P = NS

P<0.01

P<0.01

86  

Figure 5.3. Household expenditure on health care seeking (Mean, SD) for the full term low birth weight infants in the neonatal period (0-28 days) by study group. 1, 2 1 1 US$=BDT 83. 2 Groups were compared using independent sample t-test.

87  

CHAPTER 6

PAPER 2

Benzalkonium chlroride containing water-based hand sanitizer reduced infections but did

not influence linear growth among low birth weight infants in the first six month of life: a

2X2 factorial cluster randomized trial in rural Bangladesh

Abstract

Background: Low birth weight (LBW) infants in low-income countries are vulnerable to

frequent infections resulting in postnatal linear growth faltering and poor neurodevelopment.

Innovations to reduce infection can protect young lives.

Objectives: To investigate the efficacy of water-based hand sanitizers combined with nutrition

and hygiene education to prevent infections and linear growth faltering among full term LBW

infants in rural Bangladeshi communities in the first six months of life.

Methods: A community-based, cluster randomized, controlled trial was conducted in rural

Bangladesh using 48 clusters randomly assigned to two groups: i) Nutrition, health and hygiene

education (NHHE) plus directed use of benzalkonium chloride-based hand sanitizer (HS) by

mothers and other family members; ii) NHHE only. We followed 467 LBW infants with weekly

morbidity reports and monthly anthropometry during the first six months of life.

88  

Results: At the end of six months, no significant difference was found in length-for-age z score

or prevalence of stunting between intervention groups. Descriptive analysis indicates important

patterns of linear growth of low birth of infants from resource poor settings. A lower proportion

of LBW infants in the ‘NHHE+HS’ group had caregiver reported symptoms of infections

compared to the ‘NHHE only’ group (P<0.01). Prevalence of upper respiratory tract infections

(URTI) and cough was also significantly lower in the ‘NHHE+HS’ group (29.8% vs. 33.0%; and

10.6% vs. 12.2%, respectively; P<0.05, for both).

Conclusion: Use of hand sanitizers and nutrition and hygiene education may reduce infections

among LBW infants but did not influence linear growth.

Trial registration: http://www.clinicaltrials.gov (identifier NCT01455636)

Key words

Low birth weight, linear growth, weight-for-age, length-for-age, hand-washing, developing

countries

(We intent to publish the article in a peer reviewed journal.)

89  

Introduction

The first two years of life is the time of greatest vulnerability and risk of irreversible long-term

physical and developmental damage (Rivera et al., 1997; Shrimpton et al., 2001; Grantham-

McGregor et al., 2007; Black et al., 2008). Infections are associated with impaired nutrient

absorption and utilization and their effect on growth can be significant (Scrimshaw et al., 1959;

Mata et al., 1977). Recurrent episodes of infections during the first two years are associated with

linear growth faltering and increased rates of stunting (Stephensen et al., 1999; Shankar et al.,

1998; Bhutta et al., 2006). Low birth weight (LBW) infants are particularly vulnerable to such

infections due to their compromised immune system (Singh et al., 1978; Raqib et al., 2007).

South Asian countries have a very high incidence of low birth weight (LBW) infants with the

highest rate in Bangladesh (36%). These infants in the low-income countries are more vulnerable

to infectious morbidity and mortality. In Bangladesh and similar developing countries, there are

few data available on the impact of infections on growth, especially among the LBW infants.

During infancy the nutrient needs relative to energy intake are highest and the ability of the child

to respond to nutrition interventions is greatest (Dewey et al., 2008). Thus any intervention that

can decrease the number of infections in an individual infant is likely to impact on their appetite,

ingestion of food and ultimately on growth. In a recent systematic review it was found that

provision of appropriate counselling on infant and young child feeding lead to significant

increase in weight and length gain among infants (Imdad et al., 2011). Exclusive breastfeeding is

appropriately recommended for the first six months of life, thus early food-based interventions to

influence linear growth faltering are not possible. Because the rates of infection are very high

during this same period, to break the malnutrition-infection cycle, there is an urgent need for non

food-based interventions in the first 6 months of life. Interventions to improve hand hygiene to

90  

control infections along with exclusive breastfeeding in the first six months could prevent linear

growth faltering and thereby reduce the prevalence of stunting.

In a recent meta-analysis of data from cluster-randomized controlled trials carried out in low and

middle income countries, it was demonstrated that there was a positive effect of water sanitation

and hygiene (WASH) interventions on linear growth of children under five years of age

(Dangour et al., 2013). In developing countries, hand-washing with soap and water has been

demonstrated as an evidence-based strategy to reduce rates of infection among children (Luby et

al., 2004; Luby et al., 2005; Luby et al., 2008). However, with limited access to running water at

the household, frequent hand washing with ‘soap and water’ is often not possible (Scott et al.,

2008; Luby et al., 2009). Hand sanitizers are regarded as effective alternatives to hand washing

with soap and water in institutional-based settings since they do not require rinsing, and can kill

most pathogens (Kampf et al., 2004). The current hand sanitizer formulations for health-care

facilities are alcohol-based (Pittet et al., 2009) but their acceptability is often reduced by their

tendency to cause excessive dryness and skin irritation (Larson et al., 2006). Besides, alcohol-

based hand sanitizers are toxic if ingested and are highly flammable and may pose a fire hazard

if they are exposed to an open household cooking flame. As a result of these inherent

disadvantages of alcohol-based hand sanitizers, the efficacy of alcohol-free, water-based hand

sanitizers have been investigated (White et al., 2001), however, their efficacy to prevent

infections in a community setting has not been widely investigated in developing countries.

The objective of this study was to investigate the efficacy of a water-based hand sanitizer in

combination with nutrition, health and hygiene education to reduce infections and improve rates

91  

of stunting in LBW Bangladeshi infants. Secondary analyses focused on the growth of the LBW

infants in our cohort since there is a marked lack of growth data on full term LBW infants.

Methods

Overview of study design

The complete design of this prospective cluster randomized trial to explore efficacy of a water-

based hand sanitizer (HS) to reduce linear growth faltering in the first year of life is presented in

Chapter 4. The present chapter presents data on monthly growth and weekly morbidity in the

first six months of full term LBW infants.

Study area

The study was conducted in Bangladesh, a tropical country in South Asia with very high rates of

infant mortality and stunting (NIPORT 2011; UNICEF 2012). According to the only nationally

representative community-based survey in Bangladesh, 36% of all infants were born with LBW,

the prevalence of LBW is higher in rural areas and the highest prevalence is found in Dhaka

division (province). The present study was conducted in two adjacent rural upazila (sub-districts)

in Dhaka division that are representative of many rural communities in Bangladesh: Palash

located in Narsingdi district and Kaliganj in Gazipur district (Figure 4.1).

Study participants

Full-term LBW infants were recruited as study subjects. Inclusion criteria consisted of: i) born as

singletons and full term (≥37 weeks of gestation); ii) birth weight: 1800 to <2500g; and iii)

families planning to reside in the study area for the study duration. Exclusion criteria consisted

92  

of: i) severe illness at birth; ii) birth asphyxia; (iii) congenital abnormalities or severe

malformations; and iv) mother’s health significantly compromised during birth.

Sampling and randomization

A total of 2198 pregnant women were identified through a household survey. The entire study

area was divided into 48 clusters, where one cluster represented the area able to be covered by

one Health and Nutrition Assistant (HNA). Each cluster provided service to about 45-50

pregnant women. Each cluster was randomly assigned to one of the two study groups. A

computer generated random digit table was used to ensure equal probability of allocation. The

potential bias in field implementation was eliminated by concealing allocation throughout the

study.

Screening and enrolment

Within 24 hours of birth, each infant-mother dyad was visited by an ‘Assessment Team’

consisting of two trained Field Enumerators (FEs). The FEs performed anthropometric

measurement of the mother and the newborn and completed the recruitment form. Infants were

eligible to be enrolled in this study provided that all eligibility conditions were met and informed

consent was received. A total of 467 infants met the eligibility conditions and thus were enrolled

and followed for 6 months for monthly anthropometry and weekly morbidity.

Ethical oversight

Verbal consent was obtained from all pregnant mothers and their families within 24 hours of

birth, prior to weight and length measurements of their newborns. Written informed consent was

93  

obtained from the respondents before enrolment if the infants met eligibility criteria. The

Research Ethics Committees of Hospital for Sick Children, Canada and the James P. Grant

School of Public Health, BRAC University, Bangladesh provided ethical approval for the study.

The trial is registered with NCT01455636.

Intervention components

Directed hand-hygiene with hand sanitizers (HS)

The active ingredient of the hand sanitizer (HS) used in this study was benzalkonium chloride

(alkyl dimethyl benzyl ammonium chloride) 0.15% (w/w). This is a nitrogenous cationic surface-

acting agent belonging to the quaternary ammonium group. Benzalkonium chloride solutions are

rapidly acting biocidal agents, with a moderately long duration of action. It is one of the safest

synthetic biocides known and is safe for human use as disinfectants. The water-based hand

sanitizers were packaged into 250ml foam producing plastic dispensers specifically for the study.

Dispensing in the form of foam was preferred over the standard gel form because of the natural

tendency to rub foam into one’s hands until it disappears, usually in about 30 seconds. This

extended rubbing time is crucial for disinfection. Given that hand sanitizers were not meant to

replace the practice of hand washing with soap, it was necessary to identify the critical points for

using hand sanitizer and washing with soap and water. This was accomplished through a

formative research project to understand common hygiene practices in the intervention area and

to prepare information, education and communication (IEC) materials and training modules. At

the beginning of the intervention, each family received a dispenser containing water-based hand

sanitizer, which was replaced as required throughout the intervention period. All households

received a poster including pictorial messages describing how to wash their hands and critical

94  

points for hand-hygiene. Table 4.1 presents the critical points recommended for ‘washing hands

with soap and water’ and ‘using hand sanitizer’. A demonstration session involving all members

of the household on hand-washing and the use of hand sanitizer was also conducted.

Nutrition, health and hygiene education (NHHE)

Throughout the first six months, caregivers of both study groups received simple, standardized,

and age- and culturally-appropriate nutrition and hygiene education that aimed to improve infant

feeding and prevent infections. A four-colour pictorial flipchart was used to highlight the key

messages that included: i) Kangaroo Mother Care (KMC), a technique used for thermal

protection of the newborn through skin to skin contact between mother and newborn; ii) early

initiation of breastfeeding; iii) exclusive breastfeeding; v) hand-washing with soap and water; vi)

recognition of risk factors for common diseases and basic care during illnesses; and vii) danger

signs associated with neonatal complications. A behavior change communication strategy was

designed using cognitive learning theory that emphasized coaching, demonstration and practice

for each message. The intervention were delivered by 48 Health and Nutrition Assistants (HNAs)

and 12 Health and Nutrition Promoters (HNPs) who consisted of the ‘Intervention team’.

Measurement of outcome variables

The primary outcome was the prevalence of stunting (LAZ <−2) at the end of the 6-month

follow-up period. Secondary outcomes included the prevalence of severe stunting (LAZ <−3),

underweight (WAZ <−3 or <−2) or wasting (WLZ <−3 or <−2), at 3 or 6 months of age; change

in the anthropometric indices of WAZ, LAZ, and WLZ; length and weight gain; change in head

or mid-upper arm circumference; and longitudinal prevalence of symptoms for common

95  

childhood illnesses. Anthropometric data were collected by observers working in pairs. All study

sites used identical measuring equipment that were be highly accurate and precise, yet sturdy and

portable for home visits. Recumbent length was measured to the nearest 1mm on a locally

constructed high-quality length-board with digit counter readings precise to 1 mm). Unclothed

infants were weighed using an electronic infant weighing scale (Tanita HD-314, Tanita Corp.,

Tokyo, Japan), and weights were recorded to the nearest 10 g. MUAC and head circumference

were measured using non-stretchable plastic tape measures (Lasso-o Tape, Harlow Printing Ltd.,

South Shields, Tyne & Wear, England). Each observer measured and recorded a complete set of

measurements independently, after which the two compared their readings. If any pair of

readings exceeded the maximum allowable difference for a given variable (e.g. weight, 100 g;

length/height, 7 mm), both observers once again independently measured and recorded a second

and, if necessary, a third set of readings for the variable(s) in question (de Onis et al., 2004c). All

measuring equipment was calibrated monthly. Anthropometric indices (WAZ, LAZ, and WLZ)

were calculated using WHO Child Growth Standards (2011 SAS igrowup package) (WHO,

2006).

Morbidity surveillance data were collected weekly for 6 months using structured questionnaires.

With the aid of a pictorial morbidity calendar, mothers recorded whether a child had diarrhea,

cough, fever, or any other symptom of common childhood illnesses on a daily basis. Information

on the calendar was verified by fieldworkers through a recall interview done every week.

Diarrhea was defined as >3 loose stools in a 24-h period. Mothers were expected to use ORS and

zinc tablets for all diarrhea episodes (provided free to study subjects). Acute respiratory infection

(ARI) was diagnosed if the child had reported symptoms of cough or difficulty in breathing, and

96  

rapid breathing with or without chest in-drawing. Episodes with rapid breathing and chest in-

drawing were categorized as severe ARI. Respiratory rate was measured if the infant was

reported to have rapid or difficulty breathing. In addition, every week, body temperature data

were recorded by data collectors using an axillary thermometer. Information on the occurrence,

type and severity of diarrhea and ARI and other infections were collected weekly from 0-6

months. Information on maternal and household socio-economic status (SES) were collected at

enrolment.

Quality control and data management

Throughout the study the ‘Assessment Team’ was kept independent of the ‘Intervention team’.

Selection bias was minimized by strictly following the eligibility criteria. Measurement bias was

minimized by following standard procedures including the training the field workers by the same

set of trainers, using structured questionnaires, and standard methods for anthropometry and

hemoglobin analyses. Monthly standardization was also performed among the assessment group

to monitor inter- and intra-observer variations. Based on the methods described in the MGRS,

inter-observer variation was categorized into 3 categories: small, medium and large. Among the

quality assurance measurements done by the assessors for height and weight, nearly all fell into

the small category. For the few that were medium, the training was continued or repeated until

all fell into the 'small' category. All questionnaires and data forms were reviewed by researchers

for accuracy, consistency and completeness. To ensure data quality, the study supervisors and

investigators made random spot checks. In addition, a 10% sample of study children were re-

interviewed and re-measured within 48 hours of original interview by two data collectors for

quality control to provide continuous feedback on quality of data. Data were entered twice into a

97  

custom designed Microsoft Access 2010 database program (Microsoft Corp, Redmond, WA).

Data were electronically checked for extreme values and were corrected after range-check and

mismatch check and considering biological plausibility.

Data analysis plan and statistical approach

Anthropometric indexes length-for-age (LAZ), Weight-for-age (WAZ) and, Weight-for-length

(WLZ) were calculated using ANTHRO 201145. The analyses were performed using SAS for

WINDOWS release 9.3 (SAS Institute, Cary, NC) on an intention-to-treat basis. The primary

outcome of the study was rates of stunting (LAZ <−2 Z-score). A changing trend in Z-scores

over time may represent growth faltering or catch-up growth given the child's preceding Z-score.

Secondary outcome included mean changes in LAZ, WAZ and WLZ; prevalence of moderate

and severe stunting; underweight; and infections (e.g. diarrhea, acute respiratory tract infections,

fever etc).

Baseline characteristics of the two intervention groups were examined for group comparability to

determine if randomization was successful. All outcome variables were checked for normality.

For continuous and categorical outcomes in growth and morbidity for the four intervention

groups were compared by using independent sample t-test and chi-square test respectively. We

used mixed effects regression models using PROC MIXED for repeated measures of length-for-

age Z scores. PROC GLIMMIX method was used to compare proportion of illness over time by

study group. Analysis was performed considering the design effect of clustering.

Results

98  

Of the 1,295 infants who were assessed for enrollment in the study area (between October 14,

2010 and April 30, 2011), 814 did not meet the eligibility criteria: 805 infants had birth weight

less than 1.8 kg or greater than 2.5 kg; 3 had severe illness or birth asphyxia at birth; 2 were born

before 37 weeks of gestation; 4 were twins; and 14 declined to participate. The remaining 467

were included in 48 clusters (Figure 6.1). These clusters were randomized into 2 intervention

groups, the ‘NHHE only’ (n=236) and ‘NHHE+HS’ (n=231) group, with 8-12 infants per cluster.

The infant, maternal and household socioeconomic characteristics between two study groups

were comparable at baseline (Table 6.1).

Table 6.2 shows the mean length-for-age Z-score (LAZ), weight-for-age Z-score (WAZ), and

weight-for-length Z-score (WLZ) values in the study infants at birth, 3 and 6 months of age.

There were no differences between groups in terms of LAZ, WAZ and WLZ values at any of

these given time points. However, within group analysis showed that for all indices, values at 6

months was significantly higher compared to their respective values at birth and 3 months.

Because no significant differences were found between study groups when the anthropometric

indices were compared by intervention group, we have presented LAZ, WAZ and WLZ of the

entire sample irrespective of the study groups (Figure 6.2). These figures show useful

information on the patterns of growth among low birth weight infants in the first six months. As

shown, LBW newborns already had severely compromised length growth at birth. Although

LAZ showed accelerated (catch up) growth in the first month (from LAZ −3 to −2.2) it showed

no further catch up and remained stable below −2 until 6 months of age. On the other hand, in

both study groups WAZ showed accelerated growth in the first month that continued to increase

99  

until the end of 6 months (from WAZ −2.5 to −1.5). Mean WLZ accelerated in the first two

months (from WLZ−1.3 to +0.2), and remained above zero for six months. The pattern of growth

suggests that weight and growth faltering of LBW infants was prevented by the combination of

interventions in the first six months. Although the pattern of growth was similar between boys

and girls, the mean Z-scores for LAZ and WAZ were significantly lower (P<0.05) for boys

compared to the girls (Figure 6.3.1 and 6.3.2).

Table 6.2 shows the prevalence of stunting, underweight and wasting by severity (severe: <-3 Z-

score; moderate: >-3 and <-2 Z-score; and moderate and severe: <-2 Z-score) at six months

between ‘NHHE+HS’ and ‘NHHE only’ groups. There was no difference between groups.

The overall longitudinal prevalence of any symptom for common childhood illnesses (i.e.,

cough, fever, diarrhea, other symptoms) in the previous week (0-6mo) was significantly lower in

the ‘NHHE+HS’ group compared to the ‘NHHE only’ group (P<0.01) (Figure 6.4). The

proportion of infants who had URTI and cough in the previous week in the first six months was

significantly lower in the ‘NHHE+HS’ compared to that of the ‘NHHE only’ group (29.8% vs.

33.0% and10.6% vs. 12.2%, P<0.05, for both) (Figure 6.5).

Average duration of diarrhea among low birth weight infants in the first 6 months was

significantly lower in the ‘HS plus Education’ group (3.4 days) compared to that of the ‘NHHE

only’ group (3.4 vs. 6.0 days, P<0.05) (Figure 6.6).

Discussion

100  

The present trial was conducted to investigate the efficacy of a 6-month-long intervention of

water-based hand sanitizer use in combination with nutrition, health and hygiene education to

reduce infections and linear growth faltering and prevent development of stunting. The main

finding of the study was that hand sanitizers had no effect on reducing stunting in the first six

months among rural Bangladeshi full-term LBW infants. However, infants who received hand

sanitizer were reported to have fewer infections and shorter bouts of diarrhea.

Patterns of growth of the entire population irrespective of study group designation provide

important insights into the longitudinal growth patterns of Bangladeshi low birth weight infants

born in the community. Global patterns of linear growth from mixed population (low birth

weight; term and preterm children) shows faltering at 3-4 months of age and progressive

deterioration until the age of two years (Victora et al., 2010). Similar patterns of growth were

also shown in Bangladesh (Arifeen et al., 2001; Saha et al., 2008). Unlike the above studies that

generally demonstrated linear growth faltering after three months of age, results from the current

study showed that after a sharp increase in the first month, linear growth remains stable between

1 and 6 months. More positive results were found in terms of weight-for age, where the weight

of the infants continued to increase over the first six months showing no sign of faltering at all.

These results suggest that with extensive counselling, including exclusive breastfeeding and

promotion of handwashing with soap and water, both linear and ponderal growth faltering can be

prevented in the critical first six months of life.

Several recent studies on infant growth from Bangladesh are worth exploring. One study showed

that nutrition education with a focus on complementary foods improved weight gain among

101  

children in Bangladesh but no effect on linear growth of infants was found (Roy et al., 2007).

Another study showed that improved infant and young child (IYCF) feeding practices were

significantly associated with increased linear growth (Saha et al., 2008). The gain in length

during the first year of life, however, was not biologically meaningful.

Although linear growth faltering in the first six months was prevented in the current study, no

reversal of stunting was seen. Moreover, since the LBW infants already had extremely

compromised length growth at birth, the prevalence of stunting remained very high at the end of

six months. Again, there was no significant difference in rates of stunting between study groups.

This result is consistent with the previous studies that show a strong relationship between low

weight at birth and risk of stunting in Bangladesh and other low-income countries (Arifeen et al.

2001). When the growth patterns were explored by gender, it was found that boys had slower

linear and ponderal growth compared to girls. This is in line with the findings of the Bangladesh

Demographic and Health survey that showed higher rates of stunting and underweight among

boys during infancy (NIPORT 2011). However, the NIPORT results are from a cross-sectional

study and not derived from longitudinal growth data.

Interestingly, the current study found that hand sanitizer use significantly reduced respiratory

tract infections, diarrhea and a few minor infections in the first six months. Although very few

studies from developing countries have examined the effect of hand sanitizers on morbidity, it is

evident that hand-washing with soap and water can significantly reduce diarrheal and respiratory

diseases among children (Luby et al., 2005; Luby et al., 2008). A review by Cairncross et al.,

(2010) shown a reduced risk of diarrhea by 48, 17, and 36%, associated with handwashing with

soap, improved water quality, and excreta disposal, respectively. Interestingly, one study in

102  

Bangladesh showed reduction in diarrhea among children of mothers who wash hands with water

alone compared to mothers who do not wash hands at all (Luby et al., 2011).

There have been very few studies conducted in low-income country on the efficacy of hand

sanitizers (Pickering et al., 2010; Pickering et al., 2011). The first study was conducted in Dar es

Salaam, Tanzania, where the use of alcohol-based hand sanitizer resulted in significant reduction

of Escherichia coli (0.66 log units) and fecal streptococci (0.71 log units) on hands compared to

hand washing with soap and water alone (P<0.01). It was carried out in two community settings:

schools and local health clinics. In the local health setting, mothers of infants aged <10 months

were included. The study however was not randomized and the results were not adjusted for

baseline values. Examples from developed countries repeatedly show that at the community level

in USA, there was a significant reduction in the transmission of gastrointestinal and respiratory

infections among young children (Sandora et al., 2005; Lee et al., 2005). Another American

study explored the efficacy of alcohol based hand sanitizers on artificially contaminated hands

with Escherichia coli. There was no significant difference in efficacy between hand sanitizers

and handwashing with soap and water in reducing pathogens on hands. These results support the

use of alcohol-based hand sanitizers for moderately soiled hands in setting with limited access to

water (Pickering et al., 2011). The study was conducted by volunteers and the sample size was

very low (n=15).

It is worth pointing out that the hand sanitizer formulations in all of the above mentioned studies

were alcohol-based. To date, only one study examined the effect of water-based, alcohol-free

hand sanitizer in low income setting (Luby et al., 2010). The study, completed in Bangladesh,

showed that hand sanitizer use reduced presence of microorganisms in hand. None of the above

103  

mentioned studies were, however, targeted to low birth weight infants to prevent growth

faltering.

To our knowledge, ours was the first study to explore the efficacy of water-based hand sanitizers

to influence growth among low birth weight infants in a resource-poor setting. The major

strength of the study is its robust design to investigate the efficacy of hand sanitizers in

combination with an essential ‘NHHE’ package. The cluster-randomized design was used to

ensure better representativeness and prevent contamination among intervention groups. An

important aspect of our study was a formative pilot study on hand sanitizers during the design

phase. All intervention and assessment tools were prepared taking into consideration the findings

of the formative research. For example, we used separate messages for ‘hand-washing with soap

and water’ and ‘using hand sanitizer’ so that hand sanitizers did not replace or compete with the

regular hand-washing practices of the households.

In this trial the probability of bias was low because of broad inclusion criteria, random group

allocation, similarity of the intervention groups at enrollment, separation of the outcome

assessors from intervention staff and comprehensive follow-up. The inherent consistency of the

findings, their biological plausibility and coherence with earlier studies, and their logical

chronology lead us to believe that the sample findings were reliable and representative of the

target population from which the sample was drawn. When considering the generalizability of

the results it should be considered that our study excluded pre-term infants (born less than 37

weeks of gestation), infants who were born with very low birth weight (<1800g) and normal

weight full-term infants. Furthermore, twins were excluded because they might differ markedly

104  

from the rest of the population. Thus the growth curves and other results from this study only

apply to term singleton LBW infants.

We conclude that in rural Bangladesh, a six month intervention of a combination of nutrition,

health and hygiene education with hand sanitizer may reduce the symptoms of common

childhood infections among low birth weight infants but did not support the hypothesis that use

of hand sanitizer in combination with nutrition and hygiene education would have a marked

impact on reducing stunting in the first six months. Although a true control group was not

included, we have documented that the expected linear growth faltering was prevented in this

sample but was not reversed. Further evidence is needed to confirm the beneficial effects of hand

sanitizers on reducing morbidity on stunting among term low birth weight infants.

105  

1NHHE, nutrition health and hygiene education; HS, hand sanitizers. There was no significant difference between the groups. 2 Data are number (percentage), unless otherwise mentioned; groups were compared with chi-square test for proportions; 3 Data are mean (SD); groups were compared with independent sample t-test; 4Safe delivery kit was provided to all mothers free of cost.

Table 6.1. Comparison of selected infant, maternal, socioeconomic characteristics by intervention group at enrolment1

Characteristics NHHE only

(n=231) NHHE+HS (n=236)

P value

Infant characteristics Gender, male (%) 51.22 49.8 0.93 Birth weight (g) 2.2 ± 0.23 2.2 ± 0.2 0.76 Birth length (cm) 44.3 ±1.9 43.9 ± 1.7 0.06 Head circumference at birth (cm) 32.6 ± 1.2 32.3 ± 1.0 0.13 Socio-economic status

Religion of household: Islam (%) 215 (93.1)1 222 (94.1) 0.579 Ownership of house (%) 217 (93.9) 223 (94.4) 0.798 Electricity in home (%) 182 (78.8) 182 (77.1) 0.664 Access to cell phone (%) 180 (77.9) 182 (77.1) 0.835 Ownership of farming land (%) 25.80 (57.3) 20.97 (54.7) 0.352 Expenditure on food in the last month 4764.9 (2599.7) 4756.8

(2402.2) 0.972

Water sanitation Access to tap water in household (%) 4 (1.7) 2 (0.8) 0.292 Access to tube-well water in the household (%) 131 (56.7) 142 (60.1) 0.384 Access to sanitary latrine (%) 43 (18.61) 46 (19.5) 0.341 Maternal Characteristics Mother’s age (years) 23.64 (5.4) 23.61 (5.4) 0.938 Mothers education (Illiterate) (%) 188 (80.8) 186 (79.1) 0.634 Mothers education level (years of schooling) 6.28 (6.9) 6.13 (3.1) 0.777 Antenatal care Number of antenatal care (ANC) Visits (%) 2.74 (1.8) 2.61 (1.7) 0.497 Received iron supplementation (%) 128 (55.4) 120 (55.4) 0.354 Received tetanus toxoid (TT) injection (%) 121 (52.4) 127 (53.8) 0.952 Delivery care

Delivered at home (%) 177 (77.3) 191 (82.3) 0.516 Used safe delivery kit during childbirth (%)4 145 (94.4) 156 (98.6) 0.859 Used sterile blade for cutting umbilical cord (%)4 145 (96.9) 153 (97.9) 0.352

106  

1 NHHE, nutrition health and hygiene education; HS, hand sanitizers. 2 Mean ± SD (all such values). 3 Between groups comparison at 0, 3 and 6 months were carried out using independent sample t-test. 4 Within group comparisons were carried out by PROC-MIXED. LAZ, WAZ and WLZ for both groups were significantly different at 6 months (P< 0.05) compared to values at birth and 3 months. Results were adjusted for clusters.

Table 6.2. Anthropometric indices of full term low birth weight infants at birth by study group at birth, 3 and 6 months of age. 1

Intervention groups Anthropometry

NHHE only (n=231)

NHHE+HS (n=236)

P value3

Length-for-age z score At birth −3.0 ± 0.92 −2.9 ± 0.9 0.50 3 mo −2.4 ± 0.9 −2.4 ± 0.9 0.86 6 mo −2.2 ± 1.04 −2.3 ± 0.94 0.82 Weight-for-age z score At birth −2.5 ± 0.5 −2.4 ± 0.5 0.42 3 mo −1.9 ± 0.9 −1.9 ± 0.9 0.88 6 mo −1.5 ± 0.94 −1.6 ± 0.94 0.52 Weight-for-length z score At birth −1.3 ± 0.8 −1.4 ± 0.9 0.66 3 mo 0.2 ± 1.3 0.2 ± 1.4 0.99 6 mo 0.1 ± 1.14 0.0 ± 1.24 0.69

107  

1 NHHE, nutrition health and hygiene education; HS, hand sanitizers. 2 Groups were compared by chi-square test.

Table 6.3. Nutritional status of full term low birth weight infants by study group at the end of 6 months.1, 2

Intervention groups Anthropometry

NHHE only (n=231)

NHHE+HS (n=236)

P value

Stunting (%) Moderate (<−2 to >−3 Z-score) 39.8 38.1 0.733 Severe (<−3 Z-score) 18.5 21.5 0.756 Moderate and severe (<−2 Z-score) 58.3 59.6 0.846 Underweight (%) Moderate (<−2 to >−3 Z-score) 22.2 29.0 0.261 Severe (<−3 Z-score) 5.6 5.4 0.262 Moderate and severe (<−2 Z-score) 27.8 34.4 0.150 Wasting (%) Moderate (<−2 to >−3 Z-score) 4.2 3.1 0.599 Severe (<−3 Z-score) 0.5 1.3 0.588 Moderate and severe (<−2 Z-score) 4.6 4.5 0.670

108  

Figure 6.1. Trial profile of the study (0-6mo). 1NHHE, nutrition health and hygiene education (received by both groups); 2 HS, hand sanitizers.

 

 

 

24 clusters allocated to ‘No HS’ (n=231)

 

467 newborns enrolled in 48 clusters and received NHHE1

24 clusters allocated to ‘HS’2 (n=236)

Study area divided into 48 clusters (with 45-50 pregnant women per cluster)  

828 excluded [814 did not meet inclusion criteria (Birth weight not between 1.8 -<2.5 kg=805; severe illness or birth asphyxia=3; born <37 weeks of gestation=2; twin=4); and 14 refused to participate]

1295 newborns assessed for eligibility at birth  

216 completed trial

GROUP 2 (NHHE only)

3 died 12 migrated out 0 declined  

 

 

1 died 12 migrated out 0 declined  

 

 

48 clusters randomized at birth

223completed trial

GROUP 1 (NHHE +HS)

109  

Figure 6.2. Length-for-age Z-score, weight-for-age Z-score and weight-for-length Z-score of full term low birth weight infants from 0-6 months.1, 2 1Z-scores were calculated using World Health Organization Growth Standard (Anthro 2011). 2 Repeated measures for each index was carried out by PROC-MIXED (P<0.05) (n=439).

110  

Figure 6.3.1. Length-for-age Z-score of full term low birth weight infants from 0-6 months by gender 1, 2. 1Z-scores were calculated using World Health Organization Growth Standard (Anthro 2011). 2 LAZ of two groups were compared by PROC-MIXED (P<0.05) (n=439).

111  

Figure 6.3.2. Weight-for-age Z-score of full term low birth weight infants from 0-6 months by gender 1, 2. 1Z-scores were calculated using World Health Organization Growth Standard (Anthro 2011). 2 WAZ of two groups were compared by PROC-MIXED (P<0.05) (n=439).

112  

Figure 6.4. Proportion of full term low birth weight infants who had any reported illness in the previous week during first six months. 1, 2

1 NHHE, nutrition health and hygiene education (n=236); HS, hand sanitizers (n=231). 2 P<0.01. Groups were compared by PROC-GLIMMIX.

113  

Figure 6.5. Proportion of full term low birth weight infants with upper respiratory tract infections (URTI) and cough in the previous week during first six months.1, 2

1 NHHE, nutrition health and hygiene education (n=236); HS, hand sanitizers (n=231). 2P<0.05. Groups were compared by Chi-square test.

114  

Figure 6.6. Mean (SD) duration of diarrhea (in days) among full term low birth weight infants in Bangladesh during the first 6 months.1, 2

1 NHHE, nutrition health and hygiene education (n=236); HS, hand sanitizers (n=231). 2 P<0.05. Groups were compared by independent sample t-test.

   

115  

CHAPTER 7

PAPER 3

Use of water-based hand sanitizer and home fortification of complementary foods with

micronutrient powder combined with nutrition, health and hygience education prevent

linear growth faltering and reduce stunting among low birth weight infants in the first year

of life: a 2X2 factorial cluster randomized trial in rural Bangladesh.

Abstract

Background: Low birth weight (LBW) infants in low income countries are vulnerable to

frequent infections resulting in postnatal linear growth faltering and poor neurodevelopment.

Objectives: To measure the relative effect of directed use of benzalkonium chloride containing,

water-based hand sanitizers (HS) and broad-range multiple micronutrient powder (MNP) along

with nutrition, health and hygiene education (NHHE) to prevent infections and linear growth

faltering among LBW infants.

Methods: A prospective 2X2 factorial, cluster-randomized controlled trial was conducted among

467 full-term LBW infants from 0-12 months, using 48 clusters randomly assigned as follows:

A. From 0 to 6 months, i) NHHE alone or ii) NHHE plus HS. B. From 6-12 months i) NHHE

alone; ii) NHHE plus HS; iii) NHHE plus MNP (to be provided with complementary foods); and

iv) NHHE plus both HS and MNP.

Results: Combined intervention of ‘directed hand-sanitizer use’ and ‘micronutrient powder’

along with ‘NHHE’ significantly improved linear growth of low birth weight infants compared

116  

to ‘NHHE’ alone. At the end of 12-month intervention, the ‘NHHE+HS+MNP’ (moderate:

23.7%; severe: 14.9%), ‘NHHE+MNP’(moderate: 22.1%; severe: 11.5%) and ‘NHHE+HS’

(moderate: 30.5%; severe: 17.1%) groups had significantly lower prevalence of moderate and

severe stunting (LAZ scores <−2 Z-score) compared to the ‘NHHE only’ group (moderate:

32.4.1%; severe: 23.5%) (P<0.01, for all). A marked increase in linear growth was found only

during the second half of infancy (from 6-12 months) through provision of MNP with

complementary foods.

Conclusions: A novel approach of using hand sanitizer in combination with MNP may reduce

infections and prevent linear growth faltering among term LBW infants in the first year of life.

Further evidence is needed to confirm the beneficial effects of hand sanitizers and MNPs on

reducing morbidity on linear stunting among term low birth weight infants.

Trial registration: http://www.clinicaltrials.gov (identifier NCT01455636)

Key Words: Low birth weight, linear growth, stunting, infection, hand hygiene

(We intent to publish the article in a peer reviewed journal.)

117  

INTRODUCTION

Globally, more than half of the 9 million deaths among children less than 5 years are associated

with undernutrition (Pelletier et al., 1995; Black et al., 2008; Bhutta et al., 2008). The first two

years of life present a vulnerable period of risk for irreversible long-term physical and

developmental damage associated with malnutrition (Rivera and Ruel, 1997; Shrimpton et al.,

2001; Emond et al., 2006; Grantham-McGregor et al., 2007; Black et al., 2008b; Victora et al.,

2008). Prevention with targeted and effective nutrition interventions is recommended as cost-

effective approaches to reduce child undernutrition (Ruel et al., 2008; Dewey and Adu-

Afarwuah, 2008).

The direct interactions among infections, nutrition and feeding in infancy are well described

(Scrimshaw et al., 1959; Mata et al., 1977; Bhutta, 2006). An infant or young child with an

active infection is likely to be lethargic, anorexic and often disinterested in food and the process

of eating, even breastfeeding. Acute infections are accompanied by changes in acute phase

response metabolites, including inflammatory cytokines that likely interfere with the metabolism

and utilization of nutrients (Piwoz et al., 1994). Infections are also associated with impaired

nutrient absorption, cause direct nutrient loss, increase metabolic requirements or catabolic

losses of nutrients and possibly impair the transport of nutrients to target tissues (Bhutta, 2006).

With new episode of infection, nutrients are either channeled to ‘fight infection’ (e.g. increased

zinc availability during infection likely enhances the immune response) or are sequestered so as

not to enhance bacterial growth (e.g. iron sequestered in ferritin) (Stephensen, 1999).   If such

illness recurs 6-8 times or more during each of the first two years of life, the net effect of these

acute self-limiting infections on feeding, nutrient absorption and utilization, and ultimately

118  

growth can be significant (Bhutta, 2006). In low income countries like Bangladesh, recurrent

episodes of diarrhea and respiratory infections in the first months of life undoubtedly affect

appetite as well as the utilization of nutrients for many infants and likely play a pivotal role in

early malnutrition (Baqui et al., 1992; Shankar and Prasad, 1998). Low birth weight (LBW)

infants are particularly vulnerable to such infections.

Exclusive breastfeeding (EBF) is recommended for the first 6 months of life because it can

optimally support the nutrient needs of most healthy infants. Ingestion of an adequate volume of

breast milk, therefore, is expected to result in ‘normal’ growth throughout the first half of

infancy (Bhutta et al., 2008). During this same period, however, infants are highly susceptible to

clinically apparent as well as sub-clinical infections, particularly in resource-poor settings. Thus,

there is urgent need to ensure the exclusivity of breastfeeding to prevent infection to break the

infection-malnutrition cycle. In the second 6 months of life, and thereafter, the poor quality of

locally available complementary foods may lead to a deficiency of micronutrients that are

needed to support of the infant’s immune system and adequate linear growth. During this period,

infections may continue to interfere with the feeding process and prevent the amelioration of the

malnutrition that started in the first 6 months of life (Bhutta, 2006). The resultant sub-optimal

breastfeeding, limited ingestion and utilization of nutrient-rich complementary foods, and

ultimately micronutrient deficiencies contribute significantly to post-natal linear growth faltering

and stunting.

Bangladesh has very high rates of low birth weight infants, infant mortality and stunting

(NIPORT, 2011). Linear growth of surviving infants often starts to falter by three months of age

119  

and progressively deteriorates until 24 months resulting in very high rates of stunting (Arifeen et

al., 2001; Saha et al., 2009). According to the only nationally representative community-based

survey in Bangladesh, 36% of all infants were born with LBW, which is more than twice the

15% threshold that defines a public health problem (BBS/UNICEF, 2005). It is suggested that at

least 77% of these LBW infants are growth retarded at birth, confirming that intrauterine growth

restriction is the major cause of LBW in Bangladesh.

Any intervention that can decrease the number of infection episodes in infants is likely to

improve their appetite, ingestion of food, and ultimately growth. It has been repeatedly

demonstrated that targeted use of hand hygiene (with soap and hand sanitizers) in day-care

centers and elementary schools in developed countries and in hospitals globally, decreases

iatrogenic infection rates (Curtis and Cairncross; 2003; Luby et al., 2004; Simon, 2004; Fewtrell

et al., 2005; Lee et al., 2005; Sandora et al., 2005; Luby and Curtis, 2008; Sandora et al., 2008).

However, no study has provided similar evidence in non-institutional community settings in

developing countries. While soap is widely available in such resource-poor settings, clean water

for hand washing is often in short supply thus frequent hand washing with ‘soap and water’ is

often not possible. The current hand sanitizer formulations for health-care facilities are alcohol-

based (Pittet et al., 2009) but their acceptability is often reduced by their tendency to cause

excessive dryness and skin irritation (Larson et al., 2006). Besides, alcohol-based hand sanitizers

are toxic if ingested and are highly flammable and may pose a fire hazard if they are exposed to

an open household cooking flame. As a result of these inherent disadvantages of alcohol-based

hand sanitizers, the efficacy of alcohol-free, water-based hand sanitizers have been investigated

120  

(White et al., 2001). Their efficacy, however, to prevent infections in a community setting has

not been widely investigated in developing countries.

Home fortification of complementary foods with micronutrient powders (MNP) is a feasible and

cost-effective strategy to prevent and treat micronutrient deficiencies (Zlotkin et al., 2003;

Sharieff et al., 2004; Sharieff et al., 2006; Menon et al., 2007). In the past, MNPs, particularly

Sprinkles™ have been designed to include only those micronutrients needed to prevent or treat

various anemia, or alternatively to provide a variety of minerals and vitamins to achieve the

RDA for infants and young children (Adu-Afarwuah et al., 2008). The formulations, however,

have not contained high enough amounts of minerals, such as calcium, phosphorus and

magnesium, to enhance bone growth nor additional zinc to potentially enhance appetite. The

effects of zinc deficiency on the sense of taste and on appetite have been well described (Shay,

2000; Jing et al., 2007). Individuals with zinc deficiency (e.g. acrodermatitis enteropathica)

present with anorexia, while environmentally induced zinc deficiency is associated with altered

taste and reduced appetite. With zinc treatment, these lost senses are improved and appetite is

increased (Gibson, 2000). In a child with ‘normal’ zinc status, additional zinc above the RDA is

unlikely to either enhance growth or the child’s sense of hunger. However, if an infant or child

has limited zinc intake and increased losses (through, for example, frequent bouts of diarrhea), as

might be the case in rural Bangladesh, additional zinc in MNPs may improve appetite and total

food intake. The additional calcium, phosphorus and magnesium may enhance linear growth.

We hypothesized that linear growth faltering among LBW infants could be averted through the

early provision of nutrition education and promotion of directed hand hygiene to caregivers, and

121  

later fortification of complementary food with an expanded formulation of micronutrients. The

aim of the trial was to optimize the linear growth of LBW infants. The study was designed to

answer the following question: Can linear growth faltering among LBW infants be prevented or

reversed through the early provision of nutrition education and promotion of directed hand

hygiene to caregivers, in combination with later home fortification of complementary food using

an improved formulation of micronutrient powders (MNPs)?

METHODS

Overview of study design

This study followed a prospective 2X2 factorial, community-based cluster-randomized design to

ensure better representativeness and prevent contamination between intervention groups

(Donner, 2000). The design also supported the evaluation of the combined and independent

effects on infant growth and morbidity of the two main intervention components of the study: a)

water-based hand sanitizer (HS) and b) micronutrient powders (MNPs) to be provided with

complementary foods. In the first phase of the study (0-6 months), infants from 48 clusters were

allocated to either ‘HS’ or ‘no HS’ groups. In the second phase (7-12 months) infants in each of

the original two groups were randomly re-allocated to either ‘MNPs’, or ‘no MNPs’ groups.

Thus, 48 clusters were randomly assigned as follows: from 0 to 6 months, i) NHHE only or ii)

NHHE plus HS (number of clusters were 24, for both); from 7-12 months i) NHHE only; ii)

NHHE plus HS; iii) NHHE plus MNP; and iv) NHHE plus both HS and MNP (number of

clusters were 12, for all 4 groups) (Figure 4.1). All four groups were provided with, ‘Nutrition,

health and hygiene education (NHHE)’, during the entire period of study (0-12 months).

122  

Study participants

Dyads of LBW infants and their biological mothers were recruited for the study. While LBW

infants were the study subjects, their mothers were provided with nutrition and hygiene

education and also served as the respondents of the study.

Sample size

Based on data from a previous study in rural Bangladesh (Saha et al., 2009) estimates of required

sample size assumed a 20% reduction in the proportion of stunting (length-for-age <-2 Z-score)

between Group 1 (NHHE+HS+MNP) and Group 4 (NHHE only) by the end of the intervention.

Intra-cluster correlation coefficient (ICC) was set low at 0.03 and the cluster sizes (number of

subjects in a cluster, m) were expected to be at least 6 (Aboud et al., 2008). The sample size was

multiplied by a design effect (DE) of 1.15, calculated using DE=1+ICC (m–1), to accommodate

the clustering effect. The sample was further adjusted for a potential 20% loss to follow-up over

one year, thus requiring a sample of 120 infants per group. We estimated that a total of 48

clusters inhabited by 480 low birth weight infants would have 80% power (1-β) to detect the

20% reduction in the proportion in stunting at 5% level of significance (α). Thus, the study

aimed for a total sample of 48 clusters comprising 480 LBW infants for analysis of the primary

outcome.

Study setting

The present study was conducted in rural Bangladesh. The prevalence of LBW is higher in rural

areas and the highest prevalence is found in Dhaka division (province). The study was conducted

in two adjacent rural upazila (sub-districts) in Dhaka division that are representative of many

123  

rural communities in Bangladesh: Palash was in Narsingdi district and Kaliganj in Gazipur

district.

Study procedures

Ethical oversight

For recruitment, verbal consent was obtained from all pregnant mothers and the male members

of the family (husbands or father-in-laws) to collect the weight and length of their newborns

within 24 hours of birth. If the infants met eligibility criteria, written informed consent was

obtained from the respondents before enrollment. The Research Ethics Committees of Hospital

for Sick Children, Canada and the James P. Grant School of Public Health, BRAC University,

Bangladesh provided ethical approval to the study. This study was registered as a clinical trial on

the clinicaltrials.gov protocol registration website (NCT01455636).

Sampling and randomization

The study sampling frame was constructed from a list of all pregnant married women identified

through a household survey in Kaliganj and Palash using a structured data collection form. We

assumed that at least 20% of pregnant women identified in the study area would give birth to

LBW babies, which is conservative compared to available data (BBS/UNICEF, 2005). A total of

2198 pregnant women were identified and interviewed from 43,832 households to ascertain the

date of their last menstrual period (LMP) using a ‘calendar of local events’ prepared by study

investigators to help women recall their LMPs. To reduce recall bias in estimating gestational

age women reporting LMP more than four months prior to the survey were not included in the

study. The information on LMP was also used to calculate the ‘Expected date of delivery (EDD)’

124  

using a pre-plotted LMP-EDD calendar designed by Maternal Neonatal and Child Health

Program of BRAC, our partner organization. The LMP-EDD calendar was used to identify

infants born as full term LBW babies, to screen out infants born prematurely, and to organize the

interventions and schedule data-collection home visits.

The study area was divided into 48 clusters, with clusters serving as the units of randomization.

A cluster represented the geographical area covered by one Health and Nutrition Assistant

(HNA) who provided service to about 45-50 pregnant women. Each cluster was randomly

assigned to one of the two study groups at birth and then further allocated to one of the four

study groups at 6 months, as mentioned in the study design section. A computer generated

random digit table was used to ensure equal probability of allocation. The likely bias in field

implementation was eliminated by concealing allocation throughout the study.

Intervention Training

The intervention team ‘Health and Nutrition Assistants (HNAs)’ and ‘Health and Nutrition

Promoters (HNPs)’ received basic training for seven full days and monthly refresher training for

the entire study period. Trainings were conducted separately for each of the study groups.

Preparation for safe delivery, birth notification and community mobilization

The study team and local officials from BRAC organized community meetings and informed the

local leaders about the importance of the study and distributed color pictorial information

leaflets. Each pregnant woman was visited at the 8th month of pregnancy and a family meeting

was conducted including the respondent (the mother), a senior female family member of the

125  

household, and if available, the potential birth attendant identified by the family. Safe delivery

kits (locally produced by the BRAC Health Program) were distributed free of charge to all

mothers. In these pre-delivery meetings, education on essential newborn care (ENC), use of the

safe delivery kit, hand washing and early initiation of breastfeeding, and exclusive breastfeeding

was provided along with demonstrations. In addition, leaflets with clear messages on hand-

washing and breastfeeding were provided to the male and literate members of the families.

Particular emphasis was placed on birth notification by ensuring that husbands or other male

member of the families send the ‘birth notification card’ to the HNAs or called them

immediately after the birth of an infant.

Screening and enrolment

Within 24 hours of a birth, each infant-mother dyad was visited by an ‘Assessment Team’ that

consisted of two trained Field Enumerators (FEs). The FEs were separated from the ‘Intervention

team’ throughout the study period by arranging separate administrative meetings and training

sessions. The FEs performed anthropometric measurement of the mother and the newborn and

completed the recruitment form. Infants were eligible to be enrolled in this study provided the

following eligibility conditions were met and informed consent was received from the

respondents. Inclusion criteria consisted of: i) born as singletons and full term (≥37 weeks of

gestation); ii) birth weight: 1800 to <2500g; iii) families planning to reside in the study area for

at least a year; while exclusion criteria consisted of: i) severe illness; ii) birth asphyxia; (iii)

congenital abnormalities or severe malformations; and iv) mother’s health significantly

compromised during birth. During the months allocated for recruitment, a total of 467 infants

who met the eligibility conditions were enrolled in the study and followed for 12 months.

126  

Intervention components

Directed hand-hygiene with hand sanitizers (HS)

The active ingredient of the hand sanitizer (HS) used in this study was benzalkonium chloride

(alkyl dimethyl benzyl ammonium chloride) 0.15% (w/w) (Figure 4.2). This is a nitrogenous

cationic surface-acting agent belonging to the quaternary ammonium group. Benzalkonium

chloride solutions are rapid acting biocidal agents with a moderately long duration of action. It is

one of the safest synthetic biocides known and is safe for human use as a disinfectant. The water-

based hand sanitizer was produced in India by Hexagon Inc. and packaged into 250ml foam

producing plastic dispensers for the study. Dispensing in the form of foam was preferred over the

standard gel form because of the natural tendency to rub foam into one’s hands until it

disappears, usually in about 30 seconds. This extended rubbing time is crucial for disinfection.

In 2009, well before of the start of the study, the investigators completed a formative research

project to understand common hygiene practice in the intervention area to prepare messages for

information, education and communication (IEC) materials and training modules. Given that

hand sanitizers were not meant to replace the practice of hand washing with soap, it was

necessary to identify and separate the ‘critical points’ for using hand sanitizer and washing with

soap (Table 4.1). The formative research also contributed to the formulation of the hand sanitizer

by providing information on the socio-cultural aspects of its likely use.

At the beginning of the main study each family received a dispenser full of water-based hand

sanitizers, which was replaced as required throughout the intervention period. The families

received one bottle at a time unless they had planned to be away from the study area for long

127  

period. All households received a poster including pictorial messages describing how to wash

hands and critical points for hand-hygiene.

Improved micronutrient powders (MNP)

From the beginning of the 7th month (181 days) of postnatal age, infants in the MNP groups

(Groups 1 and 3) were assigned to receive one sachet of MNP per day for six months. Caregivers

were instructed to sprinkle the entire contents of the sachet on the complementary foods before

feeding. A modified and improved formulation of MNPs with 22 micronutrients was used (Table

4.2) in this study. MNP was procured from a local pharmaceutical company in Dhaka,

Bangladesh (Renata Pharmaceuticals Ltd.). Caregivers were educated on the dosage, proper use,

and possible side-effects of MNP through a 4-color flipchart and practical demonstration

followed by an on-site practice session with the mothers to ensure appropriate use.

Nutrition, health and hygiene education (NHHE)

Throughout the intervention period, caregivers of all study groups received simple, standardized,

and age- and culturally-appropriate nutrition and hygiene education that aimed to improve infant

feeding and prevent infections. A 4-colour pictorial flipchart was used to highlight the key

messages which included: i) Kangaroo Mother Care (KMC), a technique used for thermal

protection of the newborn through skin to skin contact between mother and newborn; ii) early

initiation of breastfeeding; iii) exclusive breastfeeding; iv) timely, adequate, age-appropriate and

safe complementary feeding; vii) responsive feeding and psychosocial stimulation; v) hand-

washing with soap; vi) recognition of danger signs of common diseases and care during illness;

vii) father’s involvement and family support. A behavior change communication strategy was

128  

designed using cognitive learning theory, which emphasized coaching, demonstration, and

practice for each message. Mothers were counseled to interact and feed their infants in a

responsive manner, including a better recognition of hunger cues and increased verbal

communication. In the first phase (0-6 months), each HNA visited the household every other day

from 0-7 days postpartum; weekly from 2nd week to 3 months postpartum and bi-weekly from 3-

6 months. In the second phase (7-12 mo), visits were scheduled for every other day in the first

week of the 7th month, weekly for another month, and bi-weekly from 9-12months. The health

and hygiene education emphasized early recognition and home-based care of infections.

Intervention management and monitoring

Mothers met with a Health and Nutrition Assistant (HNA), who was recruited from the same

community and was responsible for one cluster. Four HNAs were supervised by one Health and

Nutrition Promoter (HNPs). While the 48 HNAs were responsible for the nutrition and hygiene

education and distribution of MNPs and Hand Sanitizers in respective clusters, the 12 HNPs

were responsible for counseling on breastfeeding and complementary feeding and problem-

solving. HNPs would meet each HNA separately in their clusters on a weekly basis and would

discuss the progress and problems faced during implementation of the intervention and document

the meetings in ‘weekly cluster-level meeting forms’. Every two weeks the HNPs reported to the

2 Field Supervisors (FS) (each responsible for 24 clusters in each sub-district). The issues

identified in this meeting were documented in the ‘bi-weekly sub-division-level meeting forms’.

At the end of every month FSs submitted all forms to the Project Coordinator and Functional PI.

All critical issues identified were documented in ‘monthly meeting forms’ and were discussed in

the refresher training of HNAs and HNPs during the following month.

129  

Outcome measures

All infants were followed from 0-12 months to obtain data on the primary and secondary

outcomes. Information on demography, socio-economic status (SES), and other covariates were

collected during baseline visits (at birth). All study questionnaires were pretested in the field.

Training modules for the ‘Assessment Team’ included i) Data collection methods and ii)

Detailed description of tools to be used and explanation of questionnaires. The outcome

measures include the following:

Anthropometry

Anthropometric data were collected by observers working in pairs. Identical portable measuring

equipment was used that was routinely checked for accuracy and precision.

Recumbent length was measured to the nearest 1mm on a locally constructed high-quality

length-board (with digit counter readings precise to 1 mm). Unclothed infants were weighed

using an electronic infant weighing scale (Tanita HD-314, Tanita Corp., Tokyo, Japan), and

weights were recorded to the nearest 10 g. MUAC and head circumference were measured using

non-stretchable plastic tape measures (Lasso-o Tape, Harlow Printing Ltd., South Shields, Tyne

& Wear, England). Each observer measured and recorded a complete set of measurements

independently, after which the two compared their readings. If any pair of readings exceeded the

maximum allowable difference for a given variable (e.g. weight, 100 g; length, 7 mm), both

observers once again independently measured and recorded a second and, if necessary, a third set

of readings for the variable(s) in question. All measuring equipment was calibrated monthly.

130  

Anthropometric indices (WAZ, LAZ, and WLZ) were calculated using WHO Child Growth

Standards (2011 SAS igrowup package (de Onis et al., 2006).

Morbidity

Morbidity surveillance data were collected weekly for 6 months using structured questionnaires.

With the aid of a pictorial morbidity calendar, mothers recorded daily whether a child had

diarrhea, cough, fever, or any other symptom of common childhood illnesses. Information on the

calendar was verified by fieldworkers through a recall interview done every week. Diarrhea was

defined as >3 loose stools in a 24-h period. Mothers were expected to use ORS and zinc tablets

for all diarrhea episodes (provided free to study subjects). Acute respiratory infection (ARI) was

diagnosed if the child had reported symptoms of cough or difficulty in breathing, and rapid

breathing with or without chest in-drawing. Episodes with rapid breathing and chest in-drawing

were categorized as severe ARI. Respiratory rate was measured if the infant was reported to have

rapid or difficulty breathing. In addition, every week, body temperature data were recorded by

data collectors using an axillary thermometer. Information on the occurrence, type and severity

of diarrhea and ARI and other infections were collected weekly from 0-6 months.

Compliance and acceptability of intervention

Compliance was assessed by counting the total number of bottles of hand sanitizer and MNP

sachets used by families at the end of the month using structured data collection forms. At the

end of the intervention, acceptability of and barriers to using hand sanitizer and MNPs and

assessment of the overall intervention was carried out qualitatively by using focus group

discussions and semi-structured interviews among randomly selected mothers.

131  

Quality control and data management

Throughout the study the Assessment Team was kept independent of the Intervention team.

Selection bias was minimized by strictly following the eligibility criteria. Measurement bias was

minimized by following standard procedure in training the field workers by the same set of

trainers, using structured questionnaires, and standard methods for anthropometry and

hemoglobin analyses. All questionnaires and data forms were reviewed by researchers for

accuracy, consistency and completeness. To ensure data quality, the study supervisors and

investigators made random spot checks. In addition, a 10% sample of study children were re-

interviewed and re-measured within 48 hours of original interview by two data collectors for

quality control to provide continuous feedback on quality of data. Data were entered twice into a

custom designed Microsoft Access 2010 database program (Microsoft Corp, Redmond, WA).

Data were electronically checked for extreme values and were corrected after range-check and

mismatch check and considering biological plausibility.

Data analysis plan and statistical approach

Anthropometric indexes length-for-age (LAZ), weight-for-age (WAZ) and, weight-for-length

(WLZ) were calculated using ANTHRO 2011 (de Onis et al., 2006). The analyses were

performed using SAS for WINDOWS release 9.3 (SAS Institute, Cary, NC) on an intention-to-

treat basis. The primary outcome of the study was rates of stunting (LAZ <−2 Z-score). A

changing trend in Z-score over time may represent growth faltering or catch-up growth

depending on the child's preceding Z-score. Secondary outcomes included mean changes in

LAZ, WAZ and WLZ; prevalence of moderate and severe stunting; underweight; and infections

(e.g. diarrhea, acute respiratory tract infections, fever etc).

132  

Baseline characteristics of the different intervention groups were examined for group

comparability to determine if randomization was successful. All outcome variables were checked

for normality. Continuous and categorical outcomes in growth and morbidity for the four

intervention groups were compared by using analysis of variance (ANOVA) (or independent

sample t-tests as appropriate) and chi-square tests respectively. We used mixed effects regression

models using PROC MIXED for repeated measures of length-for-age, weight-for-age and

weight-for-length Z scores. PROC GLIMMIX was used to compare proportions of infants with

infections over time. Analysis was performed taking into consideration the design effect of

clustering and controlling for confounding factors such as gender, anthropometric status at

enrolment, mother’s age, mother education level and household socioeconomic status. The main

predictor included in the analysis was treatment group. Kaplan-Meyer survival analysis method

was used to determine cumulative probability of recovery from severe or moderate stunting

among different groups.

RESULTS

Between October 14, 2010 and April 30, 2011, 1,295 infants were born and assessed for

enrollment in the study area (48 clusters). Of these newborns, 814 did not meet the eligibility

criteria (805 infants had birth weight less than 1.8 kg or greater than 2.5 kg; 3 had severe illness

or birth asphyxia at birth; 2 were born before 37 weeks of gestation; 4 were twins); and 14

declined to participate. The remaining 467 newborns in 48 clusters were randomized into two

intervention groups at birth (in 24 clusters) and further randomized into 4 intervention groups at

six months (in 12 clusters): the NHHE+HS+MNP (n=123); NHHE+HS (n=113); NHHE+MNP

133  

(n=117) and NHHE only (n=114); with approximately 10 infants per cluster (Figure 7.1). There

were no differences in loss to follow-up by treatment group.

Comparison of intervention groups for several infant, maternal and household socioeconomic

characteristics at enrolment showed no significant differences across the groups (Table 7.1). The

mean anthropometric indices were also comparable among the intervention groups. Average

birth weight was 2.2 kg and birth length was 44.1 cm; and ~80% infants were born at home.

Maternal demographic and household socioeconomic characteristics were also well balanced

across groups. The infants who dropped out did not differ in their enrolment characteristics from

those who completed the trial at 12 months of age (data not shown).

Growth 0-12 months

Figure 7.2 presents the unadjusted mean length-for-age Z-scores (LAZ), weight-for-age Z-scores

(WAZ) and weight-for-length Z-scores (WLZ) of the entire sample of low birth weight infants

from 0-12 months irrespective of the intervention groups. The LBW newborns of the study

already had severely compromised length growth at birth. In the first month, mean LAZ scores

increased from −2.9 to −2.2 but showed no further catch up and remained stable below −2 until

12 months of age. On the other hand, mean WAZ scores continued to increase until the end of 12

months (from WAZ scores −2.5 to −1.2). Mean WLZ scores showed a marked increase in the

first two months (from WLZ scores −1.3 to +0.2) then declined throughout the first year. Mean

LAZ, WAZ and WLZ scores by gender revealed that boys had significantly slower length and

weight growth compared to the girls throughout the first 12 months (data not shown).

134  

Mean LAZ, WAZ, and WLZ score values at birth, 6 months and 12 months by intervention

group are presented in Table 7.2. Although at birth and 6 months no differences were found in

anthropometric indices among groups, at 12 months, the ‘NHHE+HS+MNP’, ‘NHHE+HS’ and

‘NHHE+MNP’ groups were significantly different from the ‘NHHE only’ group (P<0.05, for all)

(Table 4). Within group comparisons of LAZ and WAZ scores showed that measures were

significantly different at 12 months (P < 0.05) compared to their respective values at birth and 6

months, except for the ‘NHHE only’ group.

Longitudinal changes in LAZ and WAZ scores of infants from 0-12 months by study group are

presented in Figure 7.3.1 and Figure 7.3.2, respectively. While in the first phase (from 0 to 6

mo), there was no difference in longitudinal trajectories of LAZ scores among study groups,

different growth trajectories in LAZ scores were found across groups in the second phase (from

6 to 12 mo). After adjustment for initial LAZ scores, gender, maternal height, maternal education

and household asset score, mean LAZ scores in all three intervention groups

(‘NHHE+HS+MNP’, ‘NHHE+HS’ and ‘NHHE+MNP’) were significantly different from the

‘NHHE only’ group (P<0.05). There was no detectable effect of interventions on longitudinal

trends of WAZ scores among infants.

Degree of stunting

Figure 7.4 represents the proportion of infants with severe (LAZ scores <−3 Z-score) and

moderate stunting (LAZ scores ≥−3 and <−2 Z-score) by study group at the end of intervention

period. At 12 months, the ‘NHHE+HS+MNP’ and ‘NHHE+MNP’ groups had significantly

lower prevalence of moderate and severe stunting (LAZ scores <−2 Z-score) compared to the

135  

‘NHHE only’ group, (P<0.01, for both). Similar results were found for moderate and severe

stunting separately.

The overall cumulative incidence of recovery from moderate stunting (LAZ scores ≥ −3 and

<−2) and severe stunting (LAZ scores <−3) by study group during the 12-month period were

calculated by Kaplan-Meier survival analysis (Figure 7.5.1 and Figure 7.5.2, respectively).

Figures show that, severe stunting developed less often and at later age in the three intervention

groups (‘NHHE+HS+MNP’, ‘NHHE+HS’ and ‘NHHE+MNP’) compared to the ‘NHHE only’

groups (P<0.05). No statistically significant differences among groups were found for

cumulative incidence of moderate stunting.

Morbidity in second phase (6-12 mo)

Comparison of morbidity at 6, 9 and 12 months by four intervention groups are presented in

Table 7.3. There were no differences across study groups in terms of prevalence of URTI, cough

and diarrhea. Only at 12 months, infants in the ‘NHHE+HS+MNP’ and ‘NHHE+MNP’ groups

had significantly lower incidence of fever compared to ‘NHHE+HS’ and ‘NHHE only’ groups

(P<0.05).

Mortality

Although our study was not powered to detect differences in mortality, we found that the ‘NHHE

only’ group had the highest incidence of child deaths (n=5), while the number of deaths in

‘NHHE+HS’, ‘NHHE+MNP’ and ‘NHHE+HS+MNP’ groups were 3, 1 and 1, respectively (data

not shown). Results from verbal autopsy suggested that the major causes of death were

136  

pneumonia and severe respiratory illnesses followed by gastrointestinal complications (severe

diarrhea and vomiting). The 3 infants who died in the neonatal period were due to delivery

complications and jaundice.

DISCUSSION

The present trial was carried out to investigate the independent and combined effects of ‘water-

based hand sanitizers’ and ‘MNPs’ in combination with ‘nutrition, health and hygiene education’

to reduce rates of stunting among a unique and vulnerable population of LBW infants in rural

Bangladesh. The results suggest that the combined intervention of ‘directed hand-sanitized use’

and ‘micronutrient powder’ along with ‘NHHE’ significantly improved linear growth of low

birth weight infants compared to ‘NHHE’ alone. At the end of 12-month intervention, among the

study participants, those receiving both MNP and HS along with NHHE had significantly less

moderate and severe stunting compared to the NHHE only group. Although HS independently

reduced infections, no impact of on linear growth was found in the first six months; a marked

increase in linear growth was found only during the second half of infancy (from 6-12 months)

through provision of MNP with complementary foods.

Results from the present study are consistent with earlier trials that used a broad range of

multiple micronutrients (MMN) to improve linear growth. A recent meta-analysis restricted to

studies conducted in children less than 5 years of age found significant effects of MMN on linear

growth (Ramakrishnan et al., 2011). In another meta-analysis, effects of MMN interventions

containing 3 micronutrients were compared with a placebo or only 1 or 2 micronutrients. The

overall effect size for length was 0.25 for supplements and 0.18 for fortified food; however the

137  

results were not statistically significant (Allen et al., 2009). Similar to our study, both of these

meta-analyses did not find any impact of MMN on weight growth. However, it must be

highlighted that none of these systematic studies was focused on LBW full-term infants. Indeed

very few studies have been completed on this population.

A study in Malawi showed that infants of a mixed population, who received 17-micronutrient

rich complementary foods from 6-18 months showed a noticeable difference in incidence in

severe stunting (Phuka et al., 2009). In another trial in Malawi, infants who were provided with

MMN-rich foods lost about 0.3 Z-score units of mean LAZ between 6-12 months (Lin et al.,

2008). In a randomized controlled trial in a non-malaria endemic area in South Africa,

stunted children aged 6-24 months who received MMN had significantly higher rates of linear

growth (P<0.05) (Chhagan et al., 2010). This study included a mixed population of infants who

were either HIV-infected; HIV-uninfected born to HIV-infected mothers; or HIV-uninfected

children born to HIV-infected mothers. On average, supplements had a 0.7 increase in LAZ over

18 months compared to a 0.3 Z-score decline among those who received only vitamin A and zinc

and 0.2 Z-score in vitamin A only group. The study further showed that MMN ameliorated the

effects of diarrhea on growth.

Another trial in Chinese children found that MMN with zinc was better than zinc alone in

improving growth velocity (Sandstead et al., 1998). In an attempt to improve linear growth

velocity a recent cluster randomized efficacy trial was conducted in 4 countries: Democratic

Republic of Congo, Zambia, Guatemala, and Pakistan (Krebs et al., 2012). The study compared

daily intakes of 30 to 45 g meat from 6 to 18 mo with an equicaloric multimicronutrient-fortified

138  

cereal. Neither the meat nor the micronutrient-fortified cereal reversed the progression of

stunting. None of these trials were conducted among low birth weight infants.

In a systematic review, Dewey et al. (2008) showed inconsistent effects of the provision of

micronutrient fortified complementary foods (CF) on linear growth, with improved linear growth

in Ghana and Malawi (malnourished children), but no effect in South Africa, Indonesia, or

Brazil. Of the 6 studies providing micronutrient-fortified CF, only one study, which provided

fortified milk, had a significant positive effect on linear growth.

Although previous studies examining the impact of multiple micronutrient supplementations on

stunting found no significant or only minor improvement (Adu-Afarwuah et al., 2007, Dewey et

al., 2008), recent evidence from various settings showed a positive effect of micronutrients on

reducing stunting. One recent study in Vietnam demonstrated that the general pattern of growth

faltering during late infancy could be partly prevented by interventions including the provision of

complementary foods with a greater energy and micronutrient density than traditional gruels

(Hop and Berger, 2005). A large scale evaluation of micronutrient powder distribution program

in refugee children (ages 6 to 24 months) in Bhutan showed that the prevalence of stunting

decreased significantly from 39.2% at baseline to 23.4% at endpoint, a relative decrease of 40%

(Bilukha et al., 2011). In the Cyclone Sidr response program in Bangladesh, stunting prevalence

was reported to be lower among children who had consumed 75% of the distributed MNP

sachets compared to those with a compliance rate of <75% (Rah et al., 2011). Moreover,

according to the cross-sectional nutrition surveys conducted at the Damak refugee camp in

Nepal, there was a clear downward trend in stunting, with a decline in prevalence from 39% in

139  

2007 to 23% in 2010 (Bilukha et al., 2011). Apparently, the extended duration of MNP

supplementation and enrolment of children at a young age (6 months) may have contributed to

the positive findings in these programs.

In the present study, providing a wider range on minerals and vitamins in the micronutrient

powder may have impacted on the observed improvement in linear growth and reduced rates of

stunting. The greater micronutrient intake could have resulted in better immuno-competence,

thereby reducing morbidity and indirectly leading to improved growth or may have contributed

greater substrate for linear growth. In one recent policy review to address stunting in India, the

authors suggested that effective coverage of children below two years of age with a package of

interventions (e.g. breastfeeding; immunization; appropriate complementary feeding; treatment

of infections; safe water supply; and sanitation) merits urgent investigation for its impact on

reducing stunting (Sachdev, 2012). In the current study, our results suggest that an integrated

intervention that included a state-of-the-art NHHE component, expanded hand hygiene as well as

an expanded range of micronutrients, may explain the cumulative positive impact on linear

growth.

In the current study, water-based hand sanitizer was not an effective intervention for preventing

linear growth faltering in the first six months. However, infants in the group that received hand

sanitizer had significantly fewer respiratory tract infections, diarrhea and a few minor infections

in the neonatal period and in the first six months. There is independent evidence that hand

washing with soap and water is an effective intervention to prevent infections and other evidence

to demonstrate that the use of alcohol hand sanitizers is also effective in preventing infections

140  

(Sandora et al., 2005; Luby and Curtis, 2008). Few of the studies that examined the value of

hand sanitizers took place in a community setting, nor were they compared to routine hand-

washing. In our study it would have been inappropriate (and unethical) to recommend that

subjects stop the practice of washing their hands with soap and water. Thus our design assumed

that results from hand washing and the use of the hand sanitizer would be better than hand

washing alone.

Our observation on linear growth and morbidity in this high risk population should be interpreted

in light of the fact that all groups received an intensive package of essential newborn care and

feeding recommendations as well as training on hand-washing with soap and water. Our results

strongly suggest that the nutrition, health and hygiene education promotion successfully

prevented both length and weight growth faltering, as the patterns of linear growth in all groups

did not demonstrate the typical growth faltering trends in early infancy which have been reported

globally (Victora et al., 2010) and across all socioeconomic groups in Bangladesh (Saha et al.,

2009). It should be noted that the global growth patterns were from combined cohorts of both

premature and full-term LBW and normal birth weight infants. In a study in Vietnam of low

birth weight ethnic minority infants, linear growth was compared with international references.

The authors concluded that the LBW infants in their study did not achieve comparable length

growth or catch-up growth at 12 months (Nguyen et al., 2007).

We would like to highlight that our study excluded pre-term infants (born less than 37 weeks of

gestation) and infants who were born with very low birth weight (<1800g). Furthermore, twins

were excluded because their growth patterns are known to differ from the rest of the population.

141  

Thus the results from the current study only apply to term singleton LBW infants. We had few

neonatal and infant deaths in the study most likely because high-risk infants were excluded at

enrollment. Another reason for the low mortality may have been that these infants were in an

intensive research project that emphasized nutrition and health care for all groups, equally.

To our knowledge, this was the first study to explore the relative efficacy of water-based hand

sanitizers and a modified formulation of micronutrient powder to influence growth among low

birth weight infants in a resource-poor setting. The major strength of the study is its robust

design to investigate the relative efficacy of hand sanitizers and MNP in combination with an

essential ‘NHHE’ package. The cluster-randomized design was used to ensure better

representativeness and prevent contamination among intervention groups (Donner, 2000). An

important aspect of our study was a formative pilot study on hand sanitizers during the design

phase. All intervention and assessment tools were prepared taking into consideration the findings

of the formative research. For example, we used separate messages for ‘hand-washing with soap

and water’ and ‘using hand sanitizer’ so that hand sanitizers did not replace or compete with the

regular hand-washing practices of the households. We believe that the probability of bias in this

trial was low because of broad inclusion criteria, random group allocation, similarity of the

intervention groups at enrollment, separation of the outcome assessors and comprehensive

follow-up. Data quality of our study was ensured through rigorous routine field supervision and

ongoing training in data collection methods. The inherent consistency of the findings, their

biological plausibility and coherence with earlier studies, and their logical chronology lead us to

believe that the sample findings were reliable and representative of the target population from

which the sample was drawn.

142  

One possible limitation of our study was that the documentation of weekly morbidity was based

on self-reported data by the mothers rather than direct observation by our field staff. The data

(symptoms that occurred in the previous seven days) may have been limited by recall bias. One

study in Brazil concluded that parental recall was an unreliable method to estimate incidence of

diarrhea among children; the longer the recall period, the greater was the error in under-reporting

episodes of diarrhea (Melo et al., 2007).

Conclusions

We conclude that in rural Bangladesh, daily home fortification of complimentary foods with

MNP along with nutrition and hygiene education may significantly reduce stunting among full

term low birth weight infants. Thus, MNP interventions targeting LBW infants may accelerate

rates of stunting reduction during the second half of infancy. Use of hand sanitizers may prevent

infections during the neonatal period and first six months of life in infants with high risk of

infection. Further evidence is needed to confirm the beneficial effects of hand sanitizers on

reducing morbidity and use of MNP on linear growth among term low birth weight infants.

143  

(Cont.)

Table 7.1. Comparison of selected infant, maternal, socioeconomic characteristics by intervention group at enrolment.1

Intervention groups NHHE

+ HS + MNP (n = 123)

NHHE + HS (n = 113)

NHHE + MNP (n = 117)

NHHE Only (n = 114)

P value

Infant characteristics Gender, male (%) 51.22 37.8 41.9 50.0 0.122 Birth weight (g) 2.2 ± 0.23 2.2 ± 0.2 2.2 ± 0.2 2.2 ± 0.2 0.508 Birth length (cm) 44.3 ±1.9 43.9 ± 1.7 44.0 ± 1.8 43.8 ± 1.9 0.098 Head circumference at birth (cm)

32.6 ± 1.2 32.3 ± 1.0 32.5 ±1.1 32.4 ± 0.1 0.134

Maternal Characteristics Mother’s age (%) <20 y 24.4 25.7 24.8 21.1 0.524 20-24 y 40.7 35.4 33.3 46.5 25-29 y 24.4 22.1 24.8 16.7 ≥30 y 10.6 16.8 17.1 15.8 Maternal height (m) 148.2 ± 1.5 152.1 ± 0.8 150.2 ± 1.1 147.6 ± 1.4 0.953 Maternal BMI (kg/m2) 4

22.2 ± 4.5 21.2 ± 3.2 22.7 ± 4.5 23.2 ± 3.6 0.189

Mother’s education level (%) Illiterate 14.9 15.0 12.1 8.9 0.125 Primary (grade 5) 38.6 33.6 25.0 27.6 Secondary 43.0 47.8 61.2 61.8 Higher secondary or more

3.5 3.5 1.7 1.6

No. of ANC visits 5 2.5 ± 1.6 2.6 ± 1.8 2.7 ± 0.9 2.7 ± 1.8 0.911 Received iron supplementation (%)

56.1 45.1 53.8 57.0 0.256

Place of delivery, home (%)

82.1 82.3 75.2 78.9 0.452

Socio-economic status Ownership of house (%) 96.7 92.0 96.6 91.2 0.138 Use electricity (%) 79.7 74.3 78.6 78.9 0.760 Area of main house (sqft)

151.0 ± 88.0 155.7 ± 96.1 138.5 ± 66.7 160.8 ± 93.3 0.367

Ownership of land (decimal)6

79.5 ± 125.9 43.9 ± 61.8 68.7 ± 177.1 75.9 ± 99.8 0.127

Expenditure on food in last month (Tk)

4591 ± 2417 4937 ± 2384 4944 ± 2891 4621 ± 2228 0.554

Household asset score7 37.2 ± 3.3 37.4 ± 2.9 37.1 ± 3.2 37.2 ± 2.9 0.955

144  

 

1 NHHE, nutrition health and hygiene education; HS, hand sanitizers; MNP, micronutrient powder. There was no significant difference among the groups. 2 Data are number (%), unless otherwise indicated; groups were compared by using chi-square test (for proportions). 3 Mean±SD (all such values); groups were compared by using ANOVA (for continuous variables). 4 BMI, body mass index. 5ANC, Antenatal Care; 61Decimal equal to 1/100 acre (40.4m²).    7This index was derived based on the quality of housing, occupation and household possessions.

Water sanitation Access to tap water in household (%)

0.0 1.8 0.9 2.6 0.876

Access to tube-well water (%)

70.7 48.7 56.4 57.0 0.245

Access to sanitary latrine (%)

14.6 24.8 16.2 21.1 0.248

145  

 

1 NHHE, nutrition health and hygiene education; HS, hand sanitizers; MNP, micronutrient powder. 2 Mean ± SD (all such values); 3 Between groups comparison at 0, 6 and 12 months were carried out using ANOVA. Values in the same row with different superscript letters are significantly different, P < 0.05.

Table 7.2. Anthropometric values of full term low birth weight infants at 0, 6 and 12 months of age by intervention group 1

Intervention groups

NHHE + HS + MNP (n = 123)

NHHE + HS (n = 113)

NHHE + MNP (n = 117)

NHHE only (n = 114)

P value

Length-for-age z score

At birth −2.8 ± 1.12 −3.0 ± 0.9 −2.9 ± 1.0 −3.0 ± 0.8 0.2853 6 mo −2.3 ± 0.9 −2.3 ± 1.0 −2.2 ± 1.1 −2.3 ± 1.0 0.870 12 mo −2.0 ± 1.1 a −1.9 ± 1.0 b −1.8 ±0.9 b −2.3 ± 1.0 c 0.005 Weight-for-age z score

At birth −2.4 ± 0.45 −2.4 ± 0.45 −2.4 ± 0.43 −2.5 ± 0.49 0.345 6 mo −1.6 ± 0.9 −1.5 ± 1.0 −1.4 ± 0.9 −1.6 ± 1.0 0.715 12 mo −1.0 ± 1.1 b −1.4 ± 1.1 a −1.1 ± 1.0 b −1.1 ± 1.0 b 0.043 Weight-for-length z score

At birth −1.5 ± 0.8 −1.2 ± 0.9 −1.3 ± 0.7 −1.4 ± 0.9 0.258 6 mo 0.0 ± 1.1 0.1 ± 1.2 0.1 ± 1.1 0.0 ± 1.1 0.825 12 mo 0.0 ± 1.1 c −0.5 ± 1.1 a −0.3 ± 1.1 b 0.0 ± 1.2c 0.002

146  

1 URTI, upper respiratory tract infection; 2 n (percentage) in parenthesis (all such values); 3 P values derived by using chi-square test; 4P values derived by using fisher-exact test ; 5P=NS, for within-group comparison at 12 mo compared to 6 mo and 9 mo, for each group; Results were adjusted for clusters. 6 Values in the same row with different superscript letters are significantly different, P < 0.05.Results were adjusted for clusters.

Table 7.3. Morbidity of full term low birth weight infants at 6, 9 and 12 months by intervention group.

Intervention groups

NHHE + HS + MNP

NHHE + HS

NHHE + MNP

NHHE only

P value

(n = 116) (n = 108) (n = 113) (n = 104)

URTI 1 6 mo 31 (26.7) 2 32 (29.6) 39 (34.5) 31 (29.8) 0.6403 9 mo 33 (28.9) 23 (21.7) 29 (27.9) 36 (35.6) 0.174 12 mo 5 37 (32.5) 28 (26.2) 34 (32.7) 32 (31.7) 0.697 Cough

6 mo 16 (13.8) 8 (7.4) 16 (14.2) 13 (12.5) 0.387 9 mo 10 (8.8) 6 (5.7) 10 (9.6) 13 (12.9) 0.351 12 mo 5 9 (7.9) 8 (7.5) 16 (15.4) 12 (11.9) 0.197 Fever

6 mo 7 (6.0) 7 (6.5) 9 (8.0) 6 (5.9) 0.261 9 mo 11(9.6) 9 (8.5) 8 (7.7) 10 (9.9) 0.938 12 mo 5 8 (7.0)a 13 (12.5)b 9 (8.4) a 12 (11.9) c 0.0416 Diarrhea4

6 mo 4 (4.0) 4 (4.5) 8 (7.0) 10 (9.9) 0.283 9 mo 4 (3.6) 4 (3.5) 4 (3.7) 6 (5.9) 0.842 12 mo 5 2 (2.0) 3 (3.4) 1 (1.4) 9 (8.9) 0.690

147  

 

Figure 7.1. Trial profile of the study (0-12mo).1 1NHHE, nutrition health and hygiene education (received by all groups from 0-12 mo); HS, hand sanitizers; MNP, micronutrient powder.

 

 

 

 

 

24 clusters allocated to ‘No HS’ (n=231)

 

5 died 6 migrated out 1 declined  

 

 

1 died 4 migrated out 0 declined  

 

 

467 newborns enrolled in 48 clusters and received NHHE1

24 clusters allocated to ‘HS’ (n=236)

Study area divided into 48 clusters (with 45-50 pregnant women per cluster)  

828 excluded [814 did not meet inclusion criteria (Birth weight not between 1.8 -<2.5 kg=805; severe illness or birth asphyxia=3; born <37 weeks of gestation=2; twin=4); and 14 refused to participate]

1295 newborns assessed for eligibility at birth  

12 clusters allocated to ‘MNP’

(n=123)

12 clusters allocated to ‘No MNP’

(n=113)

12 clusters allocated to ‘MNP’

(n=117)

105 completed trial

GROUP 2 (NHHE +HS)

113 completed trial

GROUP 3 (NHHE+MNP)

102 completed trial

GROUP 4 (NHHE Only)

12 clusters allocated to ‘No MNP’

(n=114)

3 died 5 migrated out 0 declined  

 

 

1 died 8 migrated out 0 declined  

 

 

48 clusters randomized at birth

24 clusters randomized at 6 mo 24 clusters randomized at 6 mo

114completed trial

GROUP 1 (NHHE +HS+MNP)

148  

Figure 7.2. Unadjusted mean length-for-age Z-score (LAZ), weight-for-age Z-score (WAZ), and weight-for-length Z-score (WLZ) of full term low birth weight infants from 0-12 months.1,2

1 Z-scores were calculated by using World Health Organization Growth Standard (Anthro 2011) 2 Error bars represent confidence intervals of mean.

149  

Figure 7.3.1. Longitudinal changes in length-for-age Z-score (LAZ) of full term low birth weight infants from 0-12 months by study group. 1,2

1 Z-scores were calculated using World Health Organization Growth Standard (Anthro 2011); 2 Error bars represent confidence intervals of mean; 3 Cohorts with different superscript letters are significantly different from each other (P < 0.05). Results are adjusted for initial LAZ, gender, mother’s age, mother’s height, mother’s education and household asset score.

a      b    b      c    

150  

Figure 7.3.2. Longitudinal changes in weight-for-age Z-score (WAZ) of full term low birth weight infants from 0-12 months by study group. 1-3

1 Z-scores were calculated using World Health Organization Growth Standard (Anthro 2011) 2 Error bars represent confidence intervals of mean. 3 P = NS. Results are adjusted for initial LAZ, gender, mother’s age, mother’s height, mother’s education and household asset score.

151  

Figure 7.4. Unadjusted proportion of full term low birth weight infants with severe (length-for-age Z-score <−3) and moderate stunting (length-for-age Z-score ≥−3 and <−2) at the end of 12 months by study group.1, 2 1 NHHE, nutrition health and hygiene education; HS, hand sanitizers; MNP, micronutrient powder; 2 Groups with different superscript letters imply significant difference of moderate to severe (length-for-age Z-score <−2) (P < 0.01), obtained by chi-square test.

152  

SEVERE STUNTING

MODERATE STUNTING

Figure 7.4. Unadjusted proportion of full term low birth weight infants with severe (length-for-age Z-score <−3) and moderate stunting (length-for-age Z-score ≥−3 and <−2) at the end of 12 months by study group.1, 2 1 NHHE, nutrition health and hygiene education; HS, hand sanitizers; MNP, micronutrient powder; 2 Groups with different superscript letters imply significant difference of moderate to severe (length-for-age Z-score <−2) (P < 0.01), obtained by chi-square test.

P=0.43,  log  rank  test  

1P=NS, log rank test using Kaplan-Meier survival analysis.

153  

Figure 7.5.2. Cumulative incidence of recovery from severe stunting (length-for-age Z-score <−3) among full term low birth weight infants from 0-12 months.1

P=0.02,  log  rank  test  

1P=0.02, log rank test using Kaplan-Meier survival analysis.

154  

CHAPTER 8

SUMMARY AND CONCLUSION

8.1. Key Findings and potential significance

This piece of work provides novel information about the effect of directed use of hand sanitizer

and home fortification of multiple micronutrient powders combined with nutrition health and

hygiene education (NHHE) to mothers on improving linear growth of term low birth weight

infants in a rural community in Bangladesh.

At the end of the 12 months, both interventions, combined with NHHE, markedly reduced

both moderate and severe stunting among full term low birth weight infants in rural

Bangladesh compared to NHHE alone. The combination of MNP plus NHHE had the most

robust impact. Thus, we believe, the intervention package investigated in this trial has potential

public health implications.

This study also demonstrated that the use of water-based hand sanitizer reduced the symptoms of

common childhood infections among term low birth weight infants during first six months of

life. Infants who received hand sanitizer, reported to have lower rates of infections, respiratory

infections and diarrhea compared to those who received education only. However, it did not

support the hypothesis of the study that use of hand sanitizer in combination with nutrition and

hygiene education would have a marked impact on reducing stunting in the first half of infancy.

Findings from the present study suggest that caregiver use of benzalkonium chloride containing

hand-sanitizer combined with ‘nutrition, health and hygiene education’ did not result in

155  

differential growth among term, low birth weight, rural Bangladeshi infants during the first four

weeks of life. However, a small decline in subjective symptoms associated with infections

suggests that hand sanitizers may have impacted on the severity of non life-threatening

infections. The study also found that household expenditure on care seeking related to neonatal

illness was lower in the group that used hand sanitizers compared to the group that received

education only. Given the results of our study, we can only conclude that in this setting and

under the circumstances of the study, water-based hand sanitizer was not an effective

intervention for improving linear growth in the neonatal period.

8.2. Strength of the study

To our knowledge, this was the first study to explore the relative efficacy of water-based hand

sanitizers and a modified formulation of micronutrient powder to influence growth among low

birth weight infants in a resource-poor setting. The major strength of the study is its robust

design to investigate the relative efficacy of hand sanitizers and MNP in combination with an

essential ‘NHHE’ package.

An important aspect of our study was a formative pilot study during the design phase. All

intervention and assessment tools were prepared taking into consideration the findings of the

formative research and preliminary explorations.

The probability of bias in this trial was low because of broad inclusion criteria, random group

allocation, similarity of the intervention groups at enrollment, separation of the outcome

assessors from intervention staff and comprehensive follow-up. Data quality of our study was

156  

ensured through rigorous routine field supervision and ongoing training in data collection

methods. The inherent consistency of the findings, their biological plausibility and coherence

with earlier studies, and their logical chronology lead us to believe that the sample findings were

reliable and representative of the target population from which the sample was drawn.

In this trial we had an excellent response and attrition rate. Few neonatal and infant deaths were

observed in this trial most likely because LBW premature infants who were at the high risk of

mortality were excluded at enrollment. Another reason for the low mortality may have been that

infants were in an intensive research project that emphasized nutrition and health care for all

groups, equally.

8.3. Limitations of the study

Results of the study are based on a carefully conducted cluster randomized trial. However, there

are several limitations that are important to note.

Our observation on linear growth and morbidity in this high risk population should be interpreted

in light of the fact that all groups received an intensive package of essential newborn care, infant

feeding recommendations as well as training on hand-washing with soap and water. In our study

it would have been inappropriate (and unethical) to recommend that subjects stop the practice of

washing their hands with soap and water. Therefore the design assumed that hand washing plus

the use of the hand sanitizer would be more effective than hand washing alone.

It should also be noted that our study excluded pre-term infants (born less than 37 weeks of

gestation) and infants who were born with very low birth weight (<1800g). Furthermore, twins

157  

were excluded because their growth patterns are known to differ from the rest of the population.

Thus the direct results from the current study only apply to term singleton LBW infants, although

the principles of improved hand sanitation, the emphasis on early and continuous breastfeeding

and infants care and the provision of nutrition-enhanced complementary foods may be more

widely applicable.

Another possible limitation of our study was that the documentation of neonatal complications

and weekly morbidity was based on self-reported data by the mothers rather than direct

observation by our field staff. The morbidity data (symptoms that occurred in the previous seven

days) may have been limited by recall bias.

8.4. Implications for policy

The results of the studies presented in this thesis provide evidence-based research on preventing

linear growth faltering and reducing stunting among full-term LBW infants in low-income

settings, who are at the high risk of infection and growth faltering.

The results from this thesis provide evidence:

1) To suggest that caregiver use of water-based hand sanitizers along with nutrition

health and hygiene education from 0-6 months of age is efficacious in reducing

infections among LBW infants in rural Bangladesh.

2) To suggest that home fortification of complementary foods with broad-range

MNP to infants from 6-12 months and continued use of hand sanitizers combined

158  

with age-appropriate nutrition, health and hygiene education is also efficacious in

reducing both moderate and severe stunting among LBW infants in rural

Bangladesh.

The present study was a small-scale research trial that reflects the impact of HS and MNPs under

conditions that optimized the distribution and consumption of MNPs. It has yet to be determined

whether HS or MNP is a cost-effective strategy for preventing and controlling stunting and

whether it can be effectively implemented at the operational level where hand sanitizers and

supplements would likely be supplied through the primary health care system over vast rural

areas.

This study was a research trial, in which both HS and MNPs were provided free-of-charge under

supervised conditions. It is uncertain whether use of HS and MNP are feasible and effective

under real world conditions, which would require a sufficient supply of HS and MNP (dependent

on cost and availability), an efficient health care service delivery for distribution, and adequate

adherence under non-supervised conditions.

Use of HS and MNP supplementation programs would face these challenges and additionally

would have to find creative and efficient ways to identify LBW infants at birth. Although

currently there is a birth registry system in Bangladesh, very few infants from rural areas are

registered immediately after birth because a majority are born at home. Identifying newborn in

rural areas would therefore be challenging and expensive. However, targeting pregnant women

through antenatal care services through community health workers may be a possible place to

159  

start. Therefore, before large-scale HS and MNP programs are implemented in Bangladesh,

effectiveness studies need to be conducted. It would be important for future studies to measure

functional outcome of growth and morbidity. Trials that include large sample sizes for detecting

differences in outcomes such as mortality between intervention groups may also be investigated.

These types of studies might take long time to complete. A cost-effectiveness study of HS and

MNP supplementation that could provide valuable information for policy decision-making is

greatly needed.

In this trial mothers and their infants benefited by receiving nutrition and health education. This

new knowledge may be used for future children in her family or in her immediate community.

The results of this trial will be used to improve understanding of the factors associated with

growth faltering and help elucidate the relationship between micronutrient status, infectious

diseases and linear growth of the low birth weight infants and inform decision making at national

regional and global level. To the best of our knowledge, neither liquid HS nor the combination of

nutrition education and an MNP product have been previously evaluated as a means to prevent

infection, improve feeding and prevent malnutrition in a non-institutional community setting in a

developing country. Both HS and MNP products can be procured locally and are inexpensive to

produce. Bangladesh has a high disease burden from low birth weight babies. Families with low

birth weight babies were the intended direct beneficiaries of this research. As we partnered with

BRAC the largest NGO in Bangladesh and Global South, there is a high likelihood that the two

interventions could be quickly implemented at scale.

160  

8.5. Future Research Directions

Prevention of linear growth faltering among LBW infants with a combined intervention

package of improved nutrition and hand hygiene is a new area of research, and therefore

very few studies to date have been conducted to evaluate the efficacy, acceptability and

cost-effectiveness of such package during infancy. For ethical reason identifying a

research design in a community setting to test the efficacy of water-based hand sanitizers

in high-risk groups is a significant challenge. Areas of future research include the

following:

A RCT that compares efficacy of HS and MNP that includes a control group of

normal birth weight infants and include infants who are born with prematurity or

extreme low birth weight. The study should include follow-up measures beyond

intervention period of MNP and HS to explore sustainability.

An effectiveness trial that examines the relative effect of intervention packages

under real world conditions and includes measures for a cost-effectiveness

analysis.

We  have  additional  data  to  analyze,  including  exclusive  breastfeeding  rates,  actual  dietary  

intake  of  infants  collected  with  24  hour  recall  from  6  to  12  months,  hemoglobin  status  at  6  

and  12  months  and  neurocognitive  developmental  outcomes  (at  age  18  months  in  two  sub-­‐

groups).

161  

8.5. Conclusion

The results of this thesis provide some preliminary evidence that in rural Bangladesh,

daily home fortification of complimentary foods with MNP and use of water-based hand

sanitizers along with nutrition and hygiene education may significantly reduce stunting

among full term low birth weight infants. Additionally, use of hand sanitizers may

prevent or limited the extent of infections during the neonatal period and first six months

of life in infants at high risk of infection and have a small impact on stunting when used

during the second six months of life. Further evidence is needed to confirm the beneficial

effects of hand sanitizers and MNP on reducing morbidity and improve linear growth

among different populations of infants.

162  

LITERATURE CITED  

Aboud FE, Moore AC, Akhter S. Effectiveness of a community-based responsive feeding

programme in rural Bangladesh: a cluster randomized field trial. Matern Child Nutr 2008;4:275-

86.

Aboud FE, Shafique S, Akhter S. A responsive feeding intervention increases children's self-

feeding and maternal responsiveness but not weight gain. J Nutr 2009;139:1738-43.

Aboud, FE and Singla DR. Challenges to changing health behaviours in developing countries: A

critical overview. Soc Sci Med. 2012;75(4):589-94.

Adair LS, Guilkey DK. Age-specific determinants of stunting in Filipino children. J Nutr

1997;127:314-20.

Adu-Afarwuah S, Lartey A, Brown KH, et al. Randomized comparison of 3 types of

micronutrient supplements for home fortification of complementary foods in Ghana: effects on

growth and motor development. Am J Clin Nutr 2007; 86:412-420.

Adu-Afarwuah S, Lartey A, Brown KH, Zlotkin S, Briend A, Dewey KG. Home fortification of

complementary foods with micronutrient supplements is well accepted and has positive effects

on infant iron status in Ghana. Am J Clin Nutr 2008;87:929-38.  

Ahmed T, Roy SK, Alam N, Ahmed AS, Ara G, Bhuiya AU, et al. Baseline survey 2004:

National Nutrition Programme. Dhaka: International Centre for Diarrheal Disease Research,

Bangladesh; 2005. p. 319.

163  

Alam, MA., Ali, NA, Sultana, N, et al.,. Newborn umbilical cord and skin care in Sylhet District,

Bangladesh: implications for the promotion of umbilical cord cleansing with topical

chlorhexidine. J Perinatol 2008;28:61-68.

Allen LH, Peerson JM, Olney DK. Provision of multiple rather than two or fewer micronutrients

more effectively improves growth and other outcomes in micronutrient-deficient children and

adults. J Nutr 2009; 139:1022-1030.

Allen LH. Interventions for micronutrient deficiency control in developing countries: past,

present and future. J Nutr 2003;133: S3875- S3878.

Arifeen SE, Black RE, Caulfield LE, Antelman G, Baqui AH. Determinants of infant growth in

the slums of Dhaka: size and maturity at birth, breastfeeding and morbidity. Eur J Clin Nutr

2001;55:167-78.

Ashworth A. Effects of intrauterine growth retardation on mortality and morbidity in infants and

young children. Eur J Clin Nutr 1998;52:S34-S41.

Bang AT, Bang RA, Baitule SB, Reddy HM, Deshmukh, MD. Manage- ment of birth asphyxia

in home deliveries in rural Gadchiroli: the effect of two types of birth attendants and of

resuscitating with mouth-to-mouth, tube- mask or bag-mask. J Perinatol 2005;25:82-91.

Baqui AH, Black RE, Sack BR, Yunus MD, Siddique AK, Chowdhury HR. Epidemiological

and clinical characteristics of acute and persistent diarrhea in rural Bangladeshi children. Acta

Paediatr 1992;381:15-21.

164  

Baqui AH, Darmstadt GL, Williams EK, et al. Rates, timing, and causes of neonatal deaths in

rural India: implications for neonatal interventions. Bull World Health Organ 2006; 84: 706-13.

Baqui AH, EL-Arifeen S, Darmstadt GL, AL E. Effect of community-based newborn-care

intervention package implemented through two service-delivery strategies in Sylhet district,

Bangladesh: a cluster-randomised controlled trial. The Lancet 2008;371:1936-1944.

Barker DJ. In utero programming of chronic disease. Clin Sci (Lond) 1998;95:115-28.

Bartram J, Cairncross S. Hygiene, sanitation, and water: forgotten foundations of health. PLoS

Med 2010;7:e1000367.

BBS/UNICEF. National Low Birth Weight Survey of Bangladesh 2003-2004. Bangladesh

Bureau of Statistics, 2005.

Bhandari N, Bahl R, Mazumdar S, Martines J, Black RE, Bhan MK. Effect of community-based

promotion of exclusive breastfeeding on diarrheal illness and growth: a cluster randomised

controlled trial. Lancet 2003;361:1418-23.

Bhutta ZA. Effect of infections and environmental factors on growth and nutritional status in

developing countries. J Peddiatr Gastroenterol Nutr 2006;43:S13-S21.

Bhutta ZA, Ahmed T, Black RE, et al. What works? Interventions for maternal and child

undernutrition and survival. Lancet 2008;371:417-40.

Bhutta ZA, Chopra M, Axelson H, et al: Countdown to 2015 decade report (2000–10): taking

stock of maternal, newborn, and child survival. Lancet 2010;375:2032–2044.

165  

Bhutta ZA, Salam RA. Global nutrition epidemiology and trends. Ann Nutr Metab 2012;61:S19-

S27.

Bilukha O, Howard C, Wilkinson C, Bamrah S, Husain F. Effects of multimicronutrient home

fortification on anemia and growth in Bhutanese refugee children. Food Nutr Bull 2011;32:264-

76.

Black MM, Walker SP, Wachs TD, et al. Policies to reduce undernutrition include child

development. Lancet 2008;371:454-5.

Black RE. Zinc deficiency, infectious disease and mortality in the developing world. J Nutr

2003;133: S1485-S9.

Black RE, Allen LH, Bhutta ZA, et al. Maternal and child undernutrition: global and regional

exposures and health consequences.Lancet. 2008;371(9608):243–260.

Black RE, Cousens S, Johnson HL, et al, and the Child Health Epidemiology Reference Group

of WHO and UNICEF. Global, regional, and national causes of child mortality in 2008: a

systematic analysis. Lancet 2010; 375:1969-87.

Briscoe C and Aboud F. Behaviour change communication targeting four health behaviours in

developing countries: A review of change techniques. Soc Sci Med 2012;75:612-21.

Bryce J, Coitinho D, Darnton-Hill I, Pelletier D, Pinstrup-Anderson P; Maternal and Child

Undernutrition Study Group. Maternal and child undernutrition: effective action at national level.

Lancet 2008;371:510-526.

166  

Cairncross S, Hunt C, Boisson S, et al. Water, sanitation and hygiene for the prevention of

diarrhea. Int J Epidemiol 2010;39:193-205.

Chaparro CM, Dewey KG. Use of lipid-based nutrient supplements (LNS) to improve the

nutrient adequacy of general food distribution rations for vulnerable sub-groups in emergency

settings. Matern Child Nutr 2010;6:1-69.

Chhagan MK, Van den Broeck J, Luabeya K-K, Mpontshane N, Tomkins A, Bennish M. Effect

on longitudinal growth and anemia of zinc or multiple micronutrients added to vitamin A: a

randomized controlled trial in children aged 6–24 months. BMC Public Health 2010;10:145.

Christian P, Tielsch J. Evidence for multiple micronutrient effects based on randomized

controlled trials and meta-analyses in developing countries. J Nutr 2012;142: S173-S77.

Conde-agudelo A, Belizan JM, Diaz-Rossello J. Kangaroo mother care to reduce morbidity and

mortality in low birthweight infants.Cochrane Database Syst Rev 2011;16:cd0002771.

Curtis V, Cairncross S. Effect of washing hands with soap on diarrhea risk in the community: a

systematic review. Lancet Infect Dis 2003;3:275-81.

Curtis V and Luby SP. Hygiene: new hopes, new horizons. Lancet Infect Dis 2011;11:312-21.

Dangour AD, Watson L, Cumming O, Boisson S, Che Y, Velleman Y, Cavill S, Allen E, Uauy

R. Interventions to improve water quality and supply, sanitation and hygiene practices, and their

effects on nutritional status of children. Chchrane Database Syst Rev 2013 Aug 1;8:CD009382.

[Epub ahead of print]

167  

Darmstadt GL, Saha SK, Ahmed AS, Chowdhury MA, Law PA, Ahmed S, Alam MA, Black

RE, Santosham M. Effect of topical treatment with skin barrier-enhancing emollients on

nonsocomial infections in preterm infants in Bangladesh: a randomised control trial. Lancet

2005;365:1039-45.

de Onis M, Blössner M, Villar J. Levels and patterns of intrauterine growth retardation in

developing countries. Eur J Clin Nutr 1998;52:S5- S15.

de Onis M, Onyango AW, Borghi E, Garza C, Yang H. Comparison of the World Health

Organization (WHO) Child Growth Standards and the National Center for Health

Statistics/WHO international growth reference: implications for child health programmes. Public

Health Nutr 2006;9:942-7.

de Onis M, Onyango AW, Van den Broeck J, Chumlea WC, Martorell R, for the WHO

Multicentre Growth Reference Study Group. Measurement and standardization protocols for

anthropometry used in the construction of a new international growth reference. Food and

Nutrition Bulletin 2004;25:S27-S36.

de Onis M, Onyango AW, Borghi E, Garza C, Yang H; WHO Multicentre Growth Reference

Study Group.Comparison of World health organization 9WHO Child Growth standrads and the

National Centre for Health Statistics/WHO international growth reference:implications for child

health performance. Public Health Nutr 2006;9:942-7.

de Onis M, Garza C, Habicht JP. Time for a new growth reference. Pediatrics. 1997;100:E8.

168  

de Onis M. Measuring nutritional status in relation to mortality. Bull World Health

Organ.2000;78:1271-4.

de Onis, Blossner M, Borghi E. Prevalence and trends of stunting among pre-school children,

1990-2020. Public Health Nutr 2012;15:142-8.

de Onis. Update on the implementation of the WHO child growth standards. World Review Nutr

Diet 2013;106:75-82.

de Pee S, Kraemer K, van den Briel T, Boy E, Grasset C, Moench- Pfanner R, Zlotkin S, Bloem

MW. Quality criteria for micronutrient powder products: report of a meeting organized by the

World Food Programme and Sprinkles Global Health Initiative. Food Nutr Bull 2008;29:232-41.

de Pee S, Moench-Pfanner R, Martini E, Zlotkin SH, Darnton-Hill I, Bloem MW. Home

fortification in emergency response and transition programming: experiences in Aceh and Nias,

Indonesia. Food Nutr Bull 2007;28:189-97.

Dewey KG, Adu-Afarwuah S. Systematic review of the efficacy and effectiveness of

complementary feeding interventions in developing countries. Matern Child Nutr 2008;4:24-85.

Dewey KG, Brown KH. Update on technical issues concerning complementary feeding of young

children in developing countries and implications for intervention programs. Food Nutr Bull

2003;24:5-28.

Dewey KG, Chaparro CM. Session 4: mineral metabolism and body composition iron status of

breast-fed infants.Proc Nutr Soc 2007;66:412-22.

169  

Dewey KG, Yang Z, Boy E. Systematic review and meta-analysis of home fortification of

complementary foods. Matern Child Nutr 2009;5:283-321.

Dimond HJ, Ashworth A. Infant feeding practices in Kenya, Mexico and Malaysia. The rarity of

the exclusively breast-fed infant. Hum Nutr Appl Nutr 1987;41:51-64.

Donner A, Klar N. Design and analysis of Cluster randomization Trials in Health Research.

London, Arnold, 2000.

Ejemot RI, Ehjri JE, Meremikwu MM, Critchlwy JA. Hand washing for preventing diarrhea.

Cochrane Database Sys Rev 2008;23:cd004265.

Emond AM, Lira PI, Lima MC, Grantham-McGregor SM, Ashworth A. Development and

behaviour of low-birthweight term infants at 8 years in northeast Brazil: a longitudinal study.

Acta Paediatr 2006;95:1249-57.

Engle PL, Bentley M, Pelto G. The role of care in nutrition programmes: current research and a

research agenda. Proc Nutr Soc 2000;59:25-35.

Fewtrell L, Kaufmann RB, Kay D, Enanoria W, Haller L, Colford JM, Jr. Water, sanitation, and

hygiene interventions to reduce diarrhea in less developed countries: a systematic review and

meta-analysis. Lancet Infect Dis 2005;5:42-52.

Gibson RS. Zinc supplementation for infants. Lancet 2000;355:2008-9.

Grantham-McGregor S, Cheung YB, Cueto S, Glewwe P, Richter L, Strupp B. Developmental

potential in the first 5 years for children in developing countries. Lancet 2007;369:60-70.

170  

Gulani A, Bhatnagar S, Sachdev HP. Neonatal zinc supplementation for prevention of mortality

and morbidity in breastfed low birth weight infants: systematic review of randomized controlled

trials. Indian Pediatr 2011;48:111-7.

Hajipour L. Hand washing rituals in trauma theatres: clean or dirty. Ann R Coll Surg Engl 2006;  

Hop le T, Berger J.Multiple micronutrientsupllementation improves anemia, micronutrient

status, and growth of Vietnamese infants:double blind, randomized, placebo-controlled trial. J

Nutr 2005;135:S660-S665.

Houben E, De Paepe K, Rogiers V. Skin condition associated with intensive use of alcoholic gels

for hand disinfection: a combination of biophysical and sensorial data. Contact Dermatitis 2006;

54:261-267.

Huda, T.M., et al., Interim evaluation of a large scale sanitation, hygiene and water improvement

programme on childhood diarrhea and respiratory disease in rural Bangladesh. Soc Sci Med

2012;75:604-11.

Imdad A, Yakoob MY, Bhutta ZA. Impact of maternal education about complementary feeding

and provision of complementary foods on child growth in developing countries. BMC Public

Health 2011;11:S25-S29.

Imdad A, Bhutta ZA. Nutritional management of the low birth weight/preterm infant in

community settings: a perspective from the developing world. J Pediatr 2013;162:S107-14.

Ip S, Chung M, Raman G, et al. A summary of the Agency for Healthcare Research and Quality's

evidence report on breastfeeding in developed countries. Breastfeed Med 2009;4:S17-S30.

171  

Jeffery AN, Voss LD, Metcalf BS, Alba S, Wilkin TJ. Parents’ awareness of overweight in

themselves and their children: cross-sectional study within a cohort. Br Med J 2005;330:23-4

Jing MY, Sun JY, Weng XY. Insights on zinc regulation of food intake and macronutrient

selection. Biol Trace Elem Res 2007;115:187-94.

Kampf G, Kramer A. Epidemiological background of hand hygiene and evaluation of the most

important agents for scrubs and rubs. Clin Microbiol Rev 2004;17:863-893.

Kariger PK, Stoltzfus RJ, Olney D, et al. Iron deficiency and physical growth predict attainment

of walking but not crawling in poorly nourished Zanzibari infants. J Nutr 2005;135:814-9.

Kimmons JE, Dewey KG, Haque E, Chakraborty J, Osendarp SJ, Brown KH. Low nutrient

intakes among infants in rural Bangladesh are attributable to low intake and micronutrient

density of complementary foods. J Nutr 2005;135:444-51.

Kramer MS. Determinants of low birth weight: methodological assessment and meta-analysis.

Bull World Health Organ 1987;65:663-737.

Kumar D, Goel NK, Mittal PC, Misra P. Influence of infant-feeding practices on nutritional

status of under-five children. Indian J Pediatr 2006;73:417-21.

Kuusipalo H, Maleta K, Briend A, et al. Growth and change in blood haemoglobin concentration

among underweight Malawian infants receiving fortified spreads for 12 weeks: a preliminary

trial. J Pediatr Gastroenterol Nutr 2006;43:525-532.

172  

Larson E, Girard R, Pessoa-Silva CL, Boyce J, Donaldson L, Pittet D. Skin reactions related to

hand hygiene and selection of hand hygiene products. Am J Infect Control 2006; 34:627–635.

Lawn JE, Cousens S, Zupan J, and the Lancet Neonatal Survival Steering Team. 4 million

neonatal deaths: when? Where? Why? Lancet 2005;365:891-900.

Lee GM, Salomon JA, Friedman JF, et al. Illness transmission in the home: a possible role for

alcohol-based hand gels. Pediatrics 2005;115:852-60.

Lin CA, Manary MJ, Maleta K, Ashorn P. An energy dense complementary food is associated

with a modest increase in weight gain when compared with fortified porridge in Malawian

children aged 6-18 months. J Nutr 2008;138;593-598.

Luby SP, Agboatwalla M, Billhimer W, Hoekstra RM. Field trial of a low cost method to

evaluate hand cleanliness. Trop Med Int Health 2007;12:765-71.

Luby SP, Agboatwalla M, Feikin DR, Painter J, Billhimer W, Altaf A, Hoekstra RM. Effect of

handwashing on child health: a randomized control trial. Lancet. 2005;366:225-33.

Luby SP, Agboatwalla M, Painter J, Altaf A, Billhimer WL, Hoekstra RM. Effect of intensive

handwashing promotion on childhood diarrhea in high-risk communities in Pakistan: a

randomized control trial. JAMA 2004;291:2547-54.

Luby SP, Curtis V. Hand washing for preventing diarrhea. Int J Epidemiol 2008;37:470-3.

173  

Luby SP, Halder AK, Tronchet C, Akhter S, Bhuiya A, Johnston RB. Household characteristics

associated with handwashing with soap in rural Bangladesh. Am J Trop Med Hyg 2009;81:882-

7.  

Luby SP, Halder AK. Associations among handwashing indicators, wealth and symptoms of

childhood respiratory illness in urban Bangladesh. Trop Med Int Health 2008;13:835-44.

Luby SP et al. A community-randomised controlled trial promoting waterless hand sanitizer and

handwashing with soap, Dhaka, Bangladesh. Trop Med Int Health, 2010;15:1508-16.

Luby SP et al., The effect of handwashing at recommended times with water alone and with soap

on child diarrhea in rural Bangladesh: an observational study. PLoS Med 2011;8:e1001052.

Luby SP et al., Using child health outcomes to identify effective measures of handwashing. Am J

Trop Med Hyg 2011;85:882-92.

Martorell R, Ramakrishnan U, Schroeder DG, et al. Intrauterine growth retardation, body size,

body composition and physical performance in adolescence. Eur J Clin Nutr 1998;52:S43-S52.

Mason J, Bailes A, Beda-Andourou M, et al. Recent trends in malnutrition in developing regions:

vitamin A deficiency, anemia, iodine deficiency, and child underweight. Food Nutr Bull

2005;26:59-108.

Mata LJ, Kromal RA, Urrutia JJ, Garcia B. Effect of infection on food intake and the nutritional

state: perspectives as viewed from the village. Am J Clin Nutr 1977;30:1215-27.

174  

McCann JC, Ames BN. An overview of evidence for a causal relation between iron de fi ciency

during development and de fi cits in cognitive or behavioral function. Am J Clin Nutr.

2007;85:931-45.

Melo MC, Taddei JA, Diniz-santos DR, may DS, Carneiro NB, Silva LR. Incidence of diarrhea:

poor parental recall ability. Braz J Infect Dis. 2007;11:571-9

Menon P, Ruel MT, Loechl CU, et al. Micronutrient Sprinkles reduce anemia among 9- to 24-

mo-old children when delivered through an integrated health and nutrition program in rural Haiti.

J Nutr 2007;137:1023-30.

Miller M, Borys D, Morgan D. Alcohol-based hand sanitizers and unintended pediatric

exposures:a retrospective review. Clin Pediatr 2009;48:429-31.  

Morley D, Woodland M. See how they grow: monitoring child growth for appropriate health

care in developing countries. Oxford, NY: Oxford University Press; 1980.

Morris SS, Cogill B, Uauy R. Effective international action against undernutrition: why has it

proven so difficult and what can be done to accelerate progress? Lancet. 2008;371:608-621

Morrison JL, Duffield JA, Muhlhausler BS, et al. Fetal growth restriction, catch-up growth and

the early origins of insulin resistance and visceral obesity. Pediatr Nephrol 2010; 25:669-677.

Mukhtar-Yola M, Ilivasu Z. A review of neonatal morbidity and mortality in Aminu Kano

Teaching Hospital, northern Nigeria. Trop Doct 2007;37:130-2.

Mullany LC, Darmstadt GL, Khatry SK, et al. Topical applications of chlorhexidine to the

175  

umbilical cord for prevention of omphalitis and neonatal mortality in southern Nepal: a

community-based, cluster-randomised trial. Lancet 2006;367:910-18.

Mullany LC, Darmstadt GL, Katz J, et al. Risk of mortality subsequent to umbilical cord

infection among newborns of southern Nepal: cord infection and mortality. Pediatr Infect Dis J

2009;28:17-20.

Mullany LC, Faillace S, Tielsch JM, et al. Incidence and risk factors for newborn umbilical cord

infections on Pemba Island, Zanzibar, Tanzania. Pediatr Infect Dis J 2009;28:503-09.

Nahar S. Impact of food supplementation on pregnancy weight gain and birth weight in rural

Bangladesh. Cambridge: University of Cambridge; 2004. pp. 50-70. PhD dissertation.

National Institute of Population Research and Training (NIPORT). Bangladesh demographic and

health survey 2010. Dhaka: National Institute of Population Research and Training; 2011. p.346.

Ndeezi G, Tylleskar T, Ndugwa C, Tumwine J. Effect of multiple micronutrient supplementation

on survival of HIV-infected children in Uganda: a randomized, controlled trial. J Int AIDS Soc.

2010;13:18.

Nguyen T and Ushijima H. Nutritional status of low birth weight ethnic minority infants in

Backan province, Vietnam. Pediatrics International 2007;49:266-272.  

NIPORT/Mitra &Associates/Measure DHS. Bangladesh Demographic and Health Survey

Preliminary Report, 2011.

176  

Onwuanaku CA, Okolo SN, Ige KO, Okpe SE, Toma BO. The effects of birth weight and gender

on neonatal mortality in north central Nigeria. BMC Res Notes 2011;24:562.

Pelletier DL, Frongillo EA, Jr., Schroeder DG, Habicht JP. The effects of malnutrition on child

mortality in developing countries. Bull World Health Organ 1995;73:443-8.

Pelletier DL, Frongillo EA. Changes in child survival are strongly associated with changes in

malnutrition in developing countries. J Nutr 2003;133:107-19.

Phuka JC, Maleta K, Thakwalakwa C, Cheung YB, Briend A, Manary Mj, Ashorn P.

Postintervention growth of Malawian children who received 12-mo dietary complementation

with a lipid-based nutrient supplement or maize-soy flour. Am J Clin Nutr 2009;89:382-90.

Pickering AJ et al., Bacterial hand contamination among Tanzanian mothers varies temporally

and following household activities. Trop Med Int Health, 2011;16:233-9.

Pickering, AJ et al., Efficacy of waterless hand hygiene compared with handwashing with soap:

a field study in Dar es Salaam, Tanzania. Am J Trop Med Hyg 2010;82:270-8.

Pittet D, Allegranzi B, Boyce J; WHO World Alliance for Patient Safety. The World Health

Organization Guidelines on Hand Hygiene in Health Care and their consensus recommendations.

Infect Control Hosp Epidemiol 2009;30:611-22.  

Piwoz EG, Black RE, Lopez de Romaña G, Creed de Kanashiro H, Brown KH. The relationship

between infants preceding appetite, illness and growth performance and mothers subsequent

feeding practice decisions. Soc Sci Med 1994;39:851-60.

177  

Rah JH, de Pee S, Kraemer K, Steiger G, Bloem MW, Spiegel P, Wilkinson C, Bilukha O.

Program experience with micronutrient powders and current evidence. J Nutr 2012;142: S191-

S196.

Ramakrishnan U, Huffman S. Multiple mircronutrient malnutrition:what can be done? In: Semba

RD, Bloem M, editors. Nutrition and health in developing countries. 2nd ed. Totowa (NJ):

Humana Press; 2008. p. 531-76.

Ramakrishnan U, Nguyen P, Martorell R. Effects of micronutrients on growth of children under

5 y of age: meta-analyses of single and multiple nutrient interventions. Am J Clin Nutr

2009;89:191-203.

Ramakrishnan U, Goldenberg T, Allen LH. Do multiple micronutrient intervention improve

child growth and development? J Nutr 2011;141:2066-75.

Ramakrishnan U, Aburto N, McCabe G, Martorell R.Multimicronutrient interventions but not

vitamin A or iron interventions alone improve child growth: results of 3 meta-analysis. J Nutr.

2004;134:2592-602.

Raqib R, Alam DS, Sarker P, et al. Low birth weight is associated with altered immune function

in rural Bangladeshi children: a birth cohort study. Am J Clin Nutr 2007;85:845-52.

Rivera J, Ruel MT. Growth retardation starts in the first three months of life among rural

Guatemalan children. Eur J Clin Nutr 1997;51:92-6.

178  

Klemm RD, Merrill RD, Wu L, Shamim AA, Ali H, Labrique A, Christian P, West KP Jr. Low-

birthweight rates higher among Bangladeshi neonates measured during active birth surveillance

compared to national survey data. Matern Child Nutr 2013. (doi:10.1111/mcn.12041)

Roy SK, Jolly SP, Shafique S, et al. Prevention of malnutrition among young children in rural

Bangladesh by a food-health-care educational intervention: a randomized, controlled trial. Food

Nutr Bull 2007;28:375-83.

Ruel MT, Menon P, Habicht JP, et al. Age-based preventive targeting of food assistance and

behaviour change and communication for reduction of childhood undernutrition in Haiti: a

cluster randomised trial. Lancet 2008;371:588-95.

Sachdev HP. Overcoming challenges to accelerating linear growth in Indian children. Indian

Pediatr 2012;49:271-5.

Saha KK, Frongillo EA, Alam DS, Arifeen SE, Persson LA, Rasmussen KM. Appropriate infant

feeding practices result in better growth of infants and young children in rural Bangladesh. Am J

Clin Nutr 2008;87:1852-9.

Saha KK, Frongillo EA, Alam DS, Arifeen SE, Persson LA, Rasmussen KM. Household food

security is associated with growth of infants and young children in rural Bangladesh. Public

Health Nutr 2009;12:1556-62.

Sandora TJ, Shih MC, Goldmann DA. Reducing absenteeism from gastrointestinal and

respiratory illness in elementary school students: a randomized, controlled trial of an infection-

control intervention. Pediatrics 2008;121:1555-62.

179  

Sandora TJ, Taveras EM, Shih MC, et al. A randomized, controlled trial of a multifaceted

intervention including alcohol-based hand sanitizer and hand-hygiene education to reduce illness

transmission in the home. Pediatrics 2005;116:587-94.

Sandstead HH, Penland JG, Alcock NW, Dayal HH, Chen XC, Li JS, Zhao F, Yang JJ: Effects

of repletion with zinc and other micronutrients on neuropsychologic performance and growth of

Chinese children Am J Clin Nutr 1998;68: S470- S475.

Sawardekar KP. Changing spectrum of neonatal omphalitis. Pediatr Infect Dis J 2004;23:22-26.

Sazawal S, Black RE, Ramsan M, Chwaya HM, Stoltzfus RJ, Dutta A, et al. Effects of routine

prophylactic supplementation with iron and folic acid on admission to hospital and mortality in

preschool children in a high malaria transmission setting: community-based, randomised,

placebo-controlled trial. Lancet. 2006;367:133-43.

Sazawal S, Dhingra U, Dhingra P, et al. Micronutrient fortified milk improves iron status,

anemia and growth among children 1–4 years: a double masked, randomized, controlled trial.

PLoS One 2010;5:e12167.

Scott BE, Schmidt WP, Aunger R, Garbrah-Aidoo N, Animashaun R. Marketing hygiene

behaviours: the impact of different communication channels on reported handwashing behaviour

of women in Ghana.Health Educ Res 2008;23:392-401.

Scrimshaw NS, Taylor CE, Gordon JE. Interactions of nutrition and infection. Am J Med Sci

1959;237:367-403.

180  

Sellen DW. The socio-ecology of young child growth among the Datoga Pastoralists of Northern

Tanzania. UC Davis, USA. 1995. PhD dissertation.

Semba RD. The historical evolution of thought regarding multiple micronutrient nutrition. J Nutr

2012;142: S143- S156.

Shankar AH, Prasad AS. Zinc and immune function: the biological basis of altered resistance to

infection. Am J Clin Nutr 1998;68: S447- S63.

Sharieff W, Bhutta Z, Schauer C, Tomlinson G, Zlotkin S. Micronutrients (including zinc)

reduce diarrhea in children: the Pakistan Sprinkles Diarrhea Study. Arch Dis Child 2006;91:573-

9.

Sharieff W, Horton SE, Zlotkin S. Economic gains of a home fortification program: evaluation

of "Sprinkles" from the provider's perspective. Can J Public Health 2006;97:20-3.

Sharieff W, Zlotkin SH, Ungar WJ, Feldman B, Krahn MD, Tomlinson G. Economics of

preventing premature mortality and impaired cognitive development in children through home-

fortification: a health policy perspective. Int J Technol Assess Health Care 2008;24:303-11.

Shay NF. Neurobiology of zinc-influenced eating behavior. J Nutr 2000;130: S1493- S9.

Shekhar M HR, Lee Y-K. . World Bank. Repositioning nutrition as central to development: A

strategy for large scale action. Washington, DC: World Bank, 2006.

Shitaye D, Asrat D, Woldeamanuel Y, Worku B. Risk factors and etiology of neonatal sepsis in

Tikur Anbessa University Hospital, Ethiopia. Ethiop Med J 2010;48:11-21

181  

Shrimpton R, Huffman SL, Zehner ER, Darnton-Hill I, Dalmiya N. Multiple micronutrient

supplementation during pregnancy in developing-country settings: policy and program

implications of the results of a meta-analysis. Food Nutr Bull 2009;30:S556-S73.

Shrimpton R, Victora CG, de Onis M, Lima RC, Blossner M, Clugston G. Worldwide timing of

growth faltering: implications for nutritional interventions. Pediatrics 2001;107:e75.

Simon AC. Hand hygiene, the crusade of the infection control specialist. Alcohol-based handrub:

the solution! Acta Clin Belg 2004;59:189-93.

Singh M, Manerikar S, Malaviya AN, Premawathi, Gopalan R, Kumar R. Immune status of low

birth weight babies. Indian Pediatr 1978;15:563-7.

Smuts CM, Lombard CJ, Benade AJ, et al. Efficacy of a foodlet-based multiple micronutrient

supplement for preventing growth faltering, anemia, and micronutrient deficiency of infants: the

four country IRIS trial pooled data analysis. J Nutr 2005;135:S631- S638.

Stephensen CB. Burden of infection on growth failure. J Nutr 1999;129: S534- S538.

Stevens GA, Finucane MM, Paciorek CJ, et al: Trends in mild, moderate, and severe stunting

and underweight, and progress towards MDG 1 in 141 developing countries: a systematic

analysis of population representative data. Lancet 2012;380:824-834.

Taneja S, Bhandari N, Rongsen-Chandola T, Mahalanabis D, Fontaine O, Bhan MK. Effect of

zinc supplementation on morbidity and growth in hospital born, low birth weight infants. Am J

Clin Nutr 2009;90:385-91.  

182  

Thorsteinsson T, Loftsson T, Masson M. Soft antibacterial agents. Curr Med Chem

2003;10:1129-36.

Tomkins A and Watson F. Malnutrition and Infection: A review. Nutrition policy discussion

paper No. 5. UNSSCN, 1989, 144 p.

United Nations Children’s Fund and (UNICEF). Infant and young child feeding (2000–2007).

http://www.childinfo.org/breastfeeding_countrydata.php. [Accessed 15 January 2013]

United Nations Children’s Fund and (UNICEF). The state of the world’s children 2012: children

in an urban world.

United Nations Children’s Fund (UNICEF) and World Health Organization (WHO). Low

Birthweight: Country, regional and global estimates. New York:UNICEF;2004.

United Nations Children's Fund and World Health Organization (WHO). Low birth weight:

country, regional and global estimates. New York, NY: United Nations Children's Fund; 2004. p.

27.

United Nations Children's Fund (UNICEF) Bangladesh statistics. New York, NY: United

Nations Children's Fund.

(http://www.unicef.org/infobycountry/bangladesh_bangladesh_statistics.html, accessed on 31

July 2010).

Varvarigou AA. Intrauterine growth restriction as a potential risk factor for disease onset in

adulthood. J Pediatr Endocrinol Metab 2010; 23:215-224.

183  

Victora CG, Adair L, Fall C, et al. Maternal and child undernutrition: consequences for adult

health and human capital. Lancet 2008;371:340-57.

Victora CG, de Onis M, Hallal PC, Blössner M, Shrimpton R: Worldwide timing of growth

faltering: revisiting implications for interventions. Pediatrics 2010; 125:e473-e480.

Waterlow JC. Note on the assessment and classification of protein-energy malnutrition in

children. Lancet. 1973;2(7820):87-89.

White CG, Shinder FS, Shinder AL, Dyer DL.Reduction of illness absenteeism in elementary

schools using an alcohol free instant hand sanitizer. J Sch Nurs 2001;17:258-65.  

World Health Organization (WHO). Guiding principles for complementary feeding of the

breastfed child. 2003.

World Health Organization (WHO). Infant and young child feeding: model chapter for textbooks

for medical students and allied health professionals.

2009.http://www.who.int/child_adolescent_health/documents/9789241597494/en/index.html.

World Health Organization (WHO). Report of informal meeting to review and develop

indicators for complementary feeding. Washington, DC; 3–5 December 2002.

http://www.who.int/child_adolescent_health/documents/a85622/en/index.html.

Wieringa FT, Berger J, Dijkhuizen MA, Hidayat A, Ninh NX, Utomo B, et al. Sex differences in

prevalence of anaemia and iron deficiency in infancy in a large multi-country trial in South-East

Asia. Br J Nutr 2007;98:1070–6.

184  

Wieringa FT, Berger J, Dijkhuizen MA, Hidayat A, Ninh NX, Utomo B, et al. Combined iron

and zinc supplementation in infants improved iron and zinc status, but interactions reduced

efficacy in a multicountry trial in southeast Asia. J Nutr 2007;137:466-71.

World Food Programme Bangladesh household food security and nutrition assessment report

2009. Dhaka: World Food Programme 2009. p.176.

Xu XF, Du LZ. Epigenetics in neonatal diseases. Chin Med J (Engl) 2010; 123:2948–2954.

Yanik CS, Deshmukh US. Maternal nutrition, intrauterine programming and consequential risks

in the offspring. Rev Endocr Metab Disord 2008;9:203-11.

Yasmin S, Osrin D, Paul E, Costello A.Neonatal mortality of low-birth-weight infants in

Bangladesh. Bull World Health Organ 2001;79:608-14.

Zlotkin S, Arthur P, Antwi KY, Yeung G. Treatment of anemia with microencapsulated ferrous

fumarate plus ascorbic acid supplied as sprinkles to complementary (weaning) foods. Am J Clin

Nutr 2001;74:791-5.

Zlotkin S, Arthur P, Schauer C, Antwi KY, Yeung G, Piekarz A. Home-fortification with iron

and zinc sprinkles or iron sprinkles alone successfully treats anemia in infants and young

children. J Nutr 2003;133:1075-80.

185  

APPENDICES Appendix 1. The World Health Organization (WHO) growth charts for length-for-age for boys.

 

186  

Appendix 2. The World Health Organization (WHO) growth charts for length-for-age for girls.

 

187  

Appendix 3. The World Health Organization (WHO) growth charts for weight-for-age for boys.

 

188  

Appendix 4. The World Health Organization (WHO) growth charts for weight-for-age for girls.

 

189  

Appendix 5. Consent form of the study.

Research Consent Form

Title of Research Project:

Preventing linear growth faltering and reversal of stunting among low birth weight infants in Bangladesh: A community-based cluster randomized controlled trial

Investigator(s):

Dr. Stanley Zlotkin, MD, SickKids Staff (Principal Investigator) (416) 813-6171

Sohana Shafique, PhD Candidate (Co-Investigator) (416) 939-7222

Purpose of the Research:

Improving nutrition has a high potential to improve child survival and ensure better growth and development for those who survive. It is evident that the first two years of life are the time of greatest vulnerability and risk of irreversible long-term physical and mental damage. Providing nutrition interventions during infancy is critical, when nutrient needs are highest and the ability of the child to respond to a nutrition intervention is greatest. The major contributing factors for post-natal linear growth faltering are sub-optimal breastfeeding, inadequate and inappropriate complementary feeding as well as micronutrient deficiencies that result in suppressed immune system among children. Infections are associated with diminished food intake and impaired nutrient absorption, cause direct nutrient loss, increase metabolic requirements or catabolic losses of nutrients and possibly impair the transport of nutrients to target tissues. Episodes of infectious diseases, such as diarrhea and respiratory illness may play a pivotal role in length growth faltering.

As many as 33% of infants are born with low birth weight in South Asia. These infants are more vulnerable to infectious morbidity and mortality. Although LBW infants are shorter at birth compared to normal birth weight babies, unfortunately they share similar patterns of linear growth faltering at around 3 months of age. This suggests that growth faltering in the postnatal period is primarily caused by environmental factors. Despite the high prevalence of LBW, to date, few studies have examined linear growth faltering among LBW infants. Studying factors affecting growth faltering may provide greater insights into interventions for optimizing growth and survival of this high-risk population.

 

190  

Exclusive breastfeeding is appropriately recommended for the first six months of life, thus early food based interventions to influence linear growth faltering are not possible. During this same period, rates of infection are very high, thus, we believe that there is an urgent need for non food-based interventions in the first 6 months of life to break the malnutrition infection cycle and prevent growth faltering.

We hypothesize that linear growth faltering among LBW infants can be averted through the early provision of nutrition education and promotion of directed hand hygiene to caregivers, and later fortification of complementary food with a high Zn micronutrient powder combined with a broad set of micronutrients.

Description of the Research:

A community-based cluster randomized controlled trial will be carried out among LBW infants in rural areas of Bangladesh. A 2x2 factorial design will be used with two main components: Water-based Hand Sanitizer (HS) and Micronutrient Powder (MNP). Infants will be recruited at birth and allocated to either (i) HS or (ii) No HS from 0 to 6 months. From 6 to 12 months, children in each group will receive either (a) MNP or (b) Placebo MNP (to be provided with complementary foods). All four groups will be provided with education on nutrition and hygiene for the entire period (0-12 months). Recumbent length is the primary outcome measure while morbidity and dietary intake will be assessed as secondary outcomes. In a subset of population, hemoglobin concentration will be assessed by field staff using a portable HemoCue® Photometer system. One drop of capillary blood will be collected from each subject at 6th and 12th months. Data on the household socio-economic, demographic conditions and food security will be collected only at baseline by field staff, whereas morbidity and feeding practices will be assessed at regular intervals. Anthropometric measurements will be carried out monthly and will be expressed as an increase in the mean weight-for age, weight-for-length, and length-for-age z-scores. The WHO standards will be used to calculate these indices from weight, height, and age data. Compliance will be assessed on the basis of the mothers’ reporting and by counting the total number of unused sachets during weekly visits by field staff. An intention to treat analysis will be performed, accounting for the design effect of clustering and controlling for confounders. Structural Equation Modeling will be used to measure direct and indirect factors associated with linear growth faltering.

Potential Harms:

A sub-group of participants will undergo blood sampling, which carries a small but inherent risk of infection. There may be a small amount of bleeding when blood is taken that might cause a slight discomfort, bruising, or redness that will usually disappear in a few days. In the case of any negative outcome observed with the interventional treatment of Sprinkles, the children from this group will be immediately removed from the study and receive proper medical attention.

Potential Discomforts or Inconvenience:

Responding to the questionnaires will last approximately 45-60 minutes and may inconvenience mothers. Infants may experience a potential discomfort when receiving the HemoCue® finger prick.

Potential Benefits:

191  

To individual subjects:

The benefits for participants will be improved iron and other micronutrient status as well as improved breastfeeding, complementary feeding and other nutritional and health practices. All the study subjects would receive nutrition education and free medical advice during the whole course of the study. Hand sanitizer and hygiene education can reduce incidence of infections.

To society:

Bangladesh has a high disease burden from low birth weight babies - a high risk group which is prone to growth faltering and infant mortality. Families with low birth weight babies will be the direct beneficiaries of this research. The society thus could be saved from many immediate and long term adverse health and economic outcomes. Mothers will also benefit by receiving nutrition and health education, which is very sustainable. Her knowledge may be used for future children in her family or in the immediate neighborhood. The results of this trial will be used to improve our understanding of the factors associated with growth faltering and help elucidate the relationship between micronutrient status, infectious diseases and linear growth of the low birth weight infants. The results of this trial will be used to inform decision-making.

Confidentiality:

We will respect you and your family’s privacy. No information about who you are/your child is will be given to anyone or be published without your permission, unless the law requires us to do this. For example, the law requires us to give information about you/your child if a child has been abused, if you/your child has an illness that could spread to others, if you or someone else talks about suicide (killing themselves), or if the court orders us to give them the study papers.

Sick Kids Clinical Research Monitors, employees of the funder or sponsor of the study [Renata Inc], or the regulator of the study may see your/your child’s health record to check on the study.

By signing this consent form, you agree to let these people look at your/ your child’s records. We will put a copy of this research consent form in your/your child’s patient health records. We will give you a copy for your/your child’s files.

The data produced from this study will be stored in a secure, locked location. Only members of the research team (and maybe those individuals described above) will have access to the data. This could include external research team members. Following completion of the research study, the data will be kept as long as required and then destroyed as required by Sick Kids policy. Published study results will not reveal your/your child’s identity.

Reimbursement:

At the end of the study, we will provide your child with a month supply of Sprinkles in recognition of your time and effort.

Participation:

192  

If you choose to let your child take part in this study you can take your child out of the study at any time. The care your child gets will not be affected in any way by whether your child takes part in this study. New information that we get while we are doing this study may affect your decision to take part in this study. If this happens, we will tell you about this new information. And we will ask you again if you still want to be in the study.

During this study we may create new tests, new medicines, or other things that may be worth some money. Although we may make money from these findings, we cannot give you/your child any of this money now or in the future because your child took part in this study.

In some situations, the study doctor or the company paying for the study may decide to stop the study. This could happen even if the medicine (or treatment) given in the study is helping your child. If this happens, the study doctor will talk to you about what will happen next.

If your child become ill or are harmed because of study participation, we will treat your child for free. Your signing this consent form does not interfere with your legal rights in any way. The study staff, any people who gave money for the study, or the hospital are still responsible, legally and professionally, for what they do.

Sponsorship:

The sponsor of this research is The Hospital for Sick Children Foundation.

Conflict of Interest:

Dr. Zlotkin and the other research team members have no conflict of interest to declare.

Consent:

By signing this form, I agree that:

1) You have explained this study to me. You have answered all my questions.

2) You have explained the possible harms and benefits (if any) of this study.

3) I know what I could do instead of having my child take part in this study. I understand that I have the right to refuse to let my child take part in the study. I also have the right to take my child out of the study at any time. My decision about my child taking part in the study will not affect my child’s health care at Sick Kids.

4) I am free now, and in the future, to ask questions about the study.

5) I have been told that my child’s medical records will be kept private except as described to me.

6) I understand that no information about my child will be given to anyone or be published without first asking my permission. ”

7)I have read and understood pages 1 to 5 of this consent form. I agree, or consent, that my child___________________ may take part in this study.

193  

_______ _________________________________ ___________________________________

Printed Name of Parent/Legal Guardian Parent/Legal Guardian’s signature & date

_________ _________________________________

Printed Name of person who explained consent Signature & date

_________________________________________________________________________

Printed Witness’ name (if the parent/legal guardianWitness’ signature & date

does not read English)

If you have any questions about this study, please call at BRAC office.

If you have questions about your rights as a subject in a study or injuries during a study, please call

the Research Ethics Manager at Research and Evaluation Division BRAC.

 

194  

Appendix 6. Ethics approval for the study from Hospital for Sick Children (Canada).

 

195  

 

Appendix 7. Ethics approval for the study from JPG School of Public Health BRAC University

(Bangladesh).

 

 

196  

Appendix 8. Danger signs for neonatal infections

 

 

197  

 

Appendix 9: Study Questionnaires

198  

199  

 

200  

201  

 

202  

203  

204  

205  

206  

     

207  

208  

209  

210  

211  

 

212  

213  

214  

 

215  

 

216  

217  

 

218    

219  

 

220  

221  

 

222  

 

223  

224  

 

225  

226  

 

227  

228