University of Groningen On the use and determinants of prenatal healthcare services de ... ·...

University of Groningen

On the use and determinants of prenatal healthcare servicesde Jong, Esthelle Idberga

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2015

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):de Jong, E. I. (2015). On the use and determinants of prenatal healthcare services. University ofGroningen.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 14-11-2020

https://www.rug.nl/research/portal/en/publications/on-the-use-and-determinants-of-prenatal-healthcare-services(5eef33b9-31fb-466d-8422-67c05fa36d50).html

https://www.rug.nl/research/portal/en/persons/esther-feijende-jong(58c2874e-649e-4e6c-9745-6c6b1ce14f5a).html

https://www.rug.nl/research/portal/en/publications/on-the-use-and-determinants-of-prenatal-healthcare-services(5eef33b9-31fb-466d-8422-67c05fa36d50).html

On the use and determinants of prenatal

healthcare services

Esther Feijen – de Jong

COLOFON

The studies in this thesis were conducted within the Research Institute SHARE of the Graduate School of Medical Sciences, University Medical Center Groningen/University of Groningen, the department of Midwifery Science of the EMGO Institute for Health and Care Research at the VU Medical Center, and the AVAG Midwifery Academy Amsterdam Groningen.

The author gratefully acknowledges financial support for printing this thesis by the Royal Dutch Organisation of Midwives, Research Institute SHARE, and the University of Groningen.

Cover: Bert Smidt & Ard BodewesLay-out: Gildeprint - EnschedePrinted by: Gildeprint - Enschede

ISBN: 978-90-367-7714-8 (book)ISBN: 978-90-367-7715-5 (digital version)© Esther Feijen-de Jong, 2015

On the use and determinants of prenatal

healthcare services

Proefschrift

ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen

Op gezag van de rector magnificus prof. dr. E. Sterken

en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

woensdag 27 mei 2015 om 16.15 uur

door

Esthelle Idberga de Jong

geboren op 31 mei 1972te Kampen

PromotoresProf. dr. S.A. ReijneveldProf. dr. F. Schellevis

CopromotoresDr. D.E.M.C. JansenDr. F. Baarveld

BeoordelingscommissieProf. dr. M.Y. BergerProf. dr. J.C. van der VeldenProf. dr. R.G. de Vries

Paranimfen:Esther Nieuwschepen-EnsingPaul de Cock

Voor Casper, Carlijn & Anne LotteTer nagedachtenis aan mijn vader

CONTENTS

Chapter 1 General introduction 9

Chapter 2 Determinants of late and/or inadequate use of prenatal healthcare in 21 high-income countries: a systematic review

Chapter 3 Determinants of prenatal healthcare utilization by low-risk women 41 in primary midwifery-led care in the Netherlands: a prospective cohort study

Chapter 4 Prenatal care use in Belgium and the Netherlands: predisposing, 61 enabling and pregnancy-related determinants

Chapter 5 Do pregnant women contact their general practitioner? 79 A register-based comparison of healthcare utilization of pregnant and non-pregnant women in general practice

Chapter 6 Determinants of use of care provided by complementary and 93 alternative healthcare practitioners to pregnant women in primary care: a prospective cohort study

Chapter 7 General discussion and implications 115

Chapter 8 Summary 129

Chapter 9 Samenvatting 137

Appendices Curriculum vitae 147 Dankwoord 149 SHARE publications 153

CHAPTER 1

General introduction

R1R2R3R4R5R6R7R8R9

R10R11R12R13R14R15R16R17R18R19R20R21R22R23R24R25R26R27R28R29R30R31R32R33R34R35R36R37R38R39

10 |

Chapter 1

R1R2R3R4R5R6R7R8R9R10R11R12R13R14R15R16R17R18R19R20R21R22R23R24R25R26R27R28R29R30R31R32R33R34R35R36R37R38R39

| 11


1‘Julia is a primigravida, aged 25, living in the Netherlands. Her pregnancy was not planned, but it was certainly wanted. In the first trimester of her pregnancy Julia and her partner moved from a big city to a small town. Therefore, her first appointment in a primary care midwifery practice was only possible at the fourteenth week of gestation. In the second trimester of her pregnancy, Julia suffered from lower back pain. She consulted her General Practitioner (GP) about this problem. The GP referred her to a physiotherapist. In the third trimester the midwife palpated a breech position. Julia decided to contact an acupuncturist to receive Moxa therapy to turn the baby in the womb. She did not tell the midwife about her visit to the acupuncturist because she did not think this was necessary. In one of the following consultations, the midwife tried to turn the baby in the womb. Fortunately, the midwife succeeded. Except for this lower back pain and the foetus’s temporary breech position, Julia’s pregnancy was uncomplicated. At 41 weeks gestation Julia gave birth to a healthy daughter at home. Julia visited the midwife thirteen times during this pregnancy.’

This case illustrates the use of healthcare and healthcare providers during pregnancy. Julia’s case is an example of a pregnant woman with a low-risk pregnancy. The low-risk pregnancy population consists of women who are not known to have any medical or obstetric risk factors before the onset of labour.1 Julia received care from a midwife, which is normal for the maternal healthcare system in the Netherlands. In addition to midwifery care, Julia decided that she also needed care from other healthcare providers for her specific problems (lower back pain and a breech position).

Today, many studies of heterogeneous populations (high risk and/or mixed high-risk/low-risk pregnant women) are being performed on the use of prenatal care and its determinants. However, when we started this PhD project, we found barely any data and evidence on the use of prenatal care and on the determinants of prenatal care use in women with low risk pregnancies. This lack of knowledge was the reason we started this research project: to contribute to the quality of prenatal care and to the development of midwifery science, which is still a young science in the Netherlands, despite the long history of midwifery care in the Netherlands.

This thesis aims to contribute to our understanding of the prenatal healthcare use of pregnant women receiving primary midwifery care and its determinants. Specific attention will be paid to the use of care provided by midwives, GPs and complementary/alternative medicine (CAM) providers. Insight into the healthcare use of pregnant women can be used to better meet the needs of these women and to improve the tailoring of the care provided to pregnant women. Knowledge of healthcare use can help improve care, and could contribute to change and investment in the quality of maternal care.

R1R2R3R4R5R6R7R8R9


12 |

Chapter 1

This chapter introduces the two main topics addressed in this thesis. Firstly, pregnancy care will be explained, followed by a description of healthcare use and the determinants of healthcare use. Finally, an outline of the main body of this thesis is presented.

PREGNANCY CARE

Prenatal carePrenatal care is defined as care provided to pregnant women to detect complications early, to prevent complications and to decrease the incidence of maternal and prenatal mortality.2 The term prevention can have three different connotations: primary, secondary or tertiary prevention. Primary prevention means avoiding the occurrence of a disease, for example during pregnancy this means maintaining a healthy diet, and avoiding smoking and drinking. Secondary prevention implies breaking off the disease process before the emergence of clinically recognizable disease, for example measuring blood pressure to identify the early stages of hypertension, taking blood samples to identify anaemia, etc. Tertiary prevention means the prevention of complications caused by the disease process and is therefore more or less synonymous with treatment.3 Prenatal care covers all three types of prevention. In addition to prevention, prenatal care enables providing women with the information they require to make informed decisions about their care and treatment.4

Pregnancy care systems Pregnancy care systems comprise a large range of preventive and curative health services specifically for pregnant women.5 Most high-income countries provide free or low-cost prenatal care. Usually, an obstetrician is the main pregnancy care provider6, though in the past this used to be a midwife. The involvement of GPs in specific pregnancy care is declining internationally.7 Nonetheless, GPs play an important role in pregnancy care since GPs are most familiar with a pregnant woman’s medical history and are a principal point of continuing care within the health system.8

The design of Dutch pregnancy care differs from almost all other high-income countries. In the Netherlands, roles and responsibilities did not shift from midwife to doctor in the twentieth century.8 Autonomous midwifery and home births are protected through law and regulation by the Dutch state.8 Dutch pregnancy care has been split into primary and secondary care, similar to the overall organization of the Dutch healthcare system.9 Primary pregnancy care is provided by primary care midwives and by some GPs for normal physiological pregnancies (low-risk pregnancies). Secondary pregnancy care is provided by obstetricians or residents, and secondary care midwives working under the responsibility of obstetricians (high-risk pregnancies). The costs of secondary care are reimbursed by healthcare insurance companies exclusively for medical reasons.10 An Obstetric and


| 13


1Midwifery Manual – which is acknowledged by all primary and secondary care providers – is used to optimise risk selection and referrals from primary to secondary care.11 In addition to the midwife, a small percentage (2–6%) of GPs provide primary pregnancy care.12,13 Midwives and GPs attend and supervise a large proportion of births, falling from 60% in 1910 to 30.5% in 2012.14 In 2012, 85.4% of all pregnant women started maternal care in primary care (0.4% with the GP). These women received the care of a total of 2,692 midwives.15

Prenatal care programmesA prenatal care programme is the routine care that all pregnant women can expect to receive during their pregnancies.4 Prenatal care programmes originate from models developed in Europe in the early years of the previous century.16,17 In 1929, Dr Janet Campbell (UK) stated that the health of women during pregnancy had to be supervised to prevent and avoid maternal death and/or perinatal death. Guidance to achieve this was developed by the UK Department of Health. This specified that the first prenatal care visit should take place around the sixteenth week of pregnancy, and followed by visits at 24 and 28 weeks, then fortnightly to 36 weeks and weekly thereafter.17 This was the foundation of modern prenatal care. However, prenatal care programmes today vary between high-income countries regarding the frequency of these visits. Most programmes indicate starting care before the twelfth week, with a range of a total of 7–13 visits for low-risk pregnancies.18-20 In the Netherlands, the pregnancy care programme in primary midwifery care includes an average of 13 visits.19

HEALTHCARE USE

Healthcare use can be defined as ‘the actual use of personal health services’.21 Use of prenatal healthcare provided by maternal healthcare providers (e.g. midwives, obstetricians and GPs) is an important determinant of maternal health.22,23 The risk of prematurity, stillbirth, and early and late neonatal death is linearly associated with decreasing use of prenatal care.23-26 Timely and adequate prenatal care have been shown to be effective in reducing the likelihood of adverse pregnancy outcomes.22-26 Therefore, it is widely accepted that prenatal care should begin during the first trimester.27 Early prenatal care and an adequate number of prenatal visits enable the identification of medical conditions which will require careful surveillance or treatment throughout pregnancy.18,27 Despite this knowledge, the use of prenatal healthcare varies.22,28-32

Information about the prevalence of adequate and inadequate prenatal healthcare use creates the opportunity to contribute to the prevention of adverse perinatal outcomes. From the perspective of policymakers, measuring prenatal care use enables informed decisions about efficient allocation of resources in the maternal healthcare system.33 This information

R1R2R3R4R5R6R7R8R9


14 |

Chapter 1

can help promote the efficient use of healthcare systems. With respect to healthcare providers, this thesis can contribute to collaboration and communication between healthcare providers. Moreover, it helps identify the training needs of care providers in pregnancy care. Finally, an indication of the various healthcare needs of pregnant women can be identified.

DETERMINANTS OF PRENATAL HEALTHCARE SERVICES

Measuring the use of prenatal healthcare services raises the question of what factors contribute to this use. In other words, which determinants are associated with the use of prenatal healthcare? Less information is available internationally about the determinants of the use of maternal healthcare services by low-risk pregnant women. There is research available from the Netherlands on the determinants of the use of prenatal healthcare.34,35 However, this research mainly focuses on specific subgroups of pregnant women such as ethnic minorities.

The model proposed by Andersen36 is often used to study healthcare use and its determinants. The Andersen model sets out a context in which the characteristics of the patients themselves and their context influence their decision to seek and receive professional care. Andersen’s model assists in defining and measuring multiple dimensions of access to care.37 Access means visiting a medical care provider but it also means getting to the right services at the right time.21 Originally, the Andersen model focused on the family as the unit of analysis. However, in subsequent work it shifted to the individual because of the heterogeneity between family members.36 Expanding this approach, Andersen and others revisited the model until its final version, described in 2007.21,36,38 In this version contextual and individual determinants are the focus of understanding health services use. Figure 1 presents the model of the determinants of health services use including contextual and individual characteristics.


| 15


1

HealthcareSystem

| |

ExternalEnvironment

PersonalHealthPractices | |Use ofHealthServices

PerceivedHealth Status |EvaluatedHealth Status |ConsumerSatisfaction

Predisposing Enabling Need Characteristics resources

ENVIRONMENT POPULATION CHARACTERISTICS HEALTH OUTCOMES BEHAVIOUR

Figure 1. Andersen’s behavioural model of healthcare use and its determinants (version 2007)37

Andersen stated that improving access to care is best accomplished by focusing on contextual and individual determinants.21 Individual and contextual characteristics are divided into major concepts which predispose, enable or suggest the need for individual health services use.37 Predisposing factors consist of demographic factors which represent biological imperatives, the social factors that determine the status of a person in the community, and health beliefs, which consist of people’s attitudes, values and knowledge about health and health services. Enabling variables comprise both community and personal enabling resources, which must be present for use to occur (health services costs, healthcare insurance and financial resources). It also includes travel time and waiting time for care. Need factors are divided into perceived and evaluated needs. Perceived need is how people view their own general health and functional state. It reflects the need to seek medical care (or not to do so). Evaluated need reflects professional judgment and objective measures of ill health. Health behaviour variables are the factors which influence an individual’s health status.21

Measuring the determinants of prenatal healthcare use can provide insight into women who are vulnerable to inadequate prenatal care use and into the potential routes for increasing adequate use. Based on this, tailored interventions can be developed to improve the use of prenatal care and ultimately to contribute to better perinatal outcomes.39

R1R2R3R4R5R6R7R8R9


16 |

Chapter 1

OUTLINE OF THIS THESIS

The outline of this thesis follows the conceptual model presented in Figure 2. In this thesis we measured two types of health services use, namely the use of care offered within prenatal care programmes and ancillary care use. Prenatal care programmes are based on professional guidelines and mostly concern prevention. Ancillary care is care provided alongside primary maternal healthcare provider care (in this thesis the primary maternal healthcare provider is a primary care midwife).With respect to the use of care offered within prenatal care programmes, the following aims were formulated:1. To provide a systematic review of the determinants of late and/or inadequate use

of prenatal healthcare in high-income countries.2. To examine the determinants of inadequate prenatal healthcare use by low-

risk women in primary midwifery-led care in the Netherlands, and to determine whether these differ from those who are referred to secondary prenatal care.

3. To compare prenatal healthcare use in Belgium and the Netherlands with differently designed pregnancy care systems, as measured by the Content and Timing of care in Pregnancy (CTP) tool and to identify its predisposing, enabling and pregnancy-related determinants.

With regard to the use of ancillary care, the following aims were formulated:4. To compare GP consultation rates, diagnoses and healthcare management for

pregnant women with those for non-pregnant women in the Netherlands.5. To examine the prevalence and determinants of use of complementary and

alternative medicine (CAM) by low-risk pregnant women in the Netherlands.

A systematic review was carried out to synthesize the existing knowledge of the determinants of prenatal healthcare use in high-income countries (Chapter 2). The use and determinants of care in primary midwifery care was studied using the database from the DELIVER study.40 DELIVER is an acronym of Data EersteLIjns VERloskunde – primary midwifery care data (Chapter 3). Chapter 4 presents the results of a study in which the Netherlands and Belgium are compared on the determinants and content of prenatal healthcare utilization. The healthcare use of pregnant and non-pregnant women in general practice was studied using data from the Netherlands Information Network of General Practice (LINH) (Chapter 5). Chapter 6 presents the results of a study on CAM use and its determinants in primary midwifery care in the Netherlands. Finally, Chapter 7 summarizes and discusses the results of this thesis with respect to their contribution to knowledge in the field. Their implications for practice, policy and research are also discussed.


| 17


1

Maternalhealthcareuse and its

determinants

Use and determinants ofuse of primarymidwifery care in theNetherlands(Chapter 3)1

Use and determinants ofuse of prenatal care inBelgium and theNetherlands(Chapter 4)1

Use of GP-care in theNetherlands(Chapter 5)2

Use and determinantsof CAM use in primarymidwifery care in theNetherlands(Chapter 6)2

Systematic review;Determinants ofhealthcare use(Chapter 2)1

1 Use of prenatal care programmes

2 Use of ancillary care

Figure 2. Conceptual model underlying this thesis

R1R2R3R4R5R6R7R8R9


18 |

Chapter 1

REFERENCES

(1) Birthplace in England Collaborative Group, Brocklehurst P, Hardy P, Hollowell J, Linsell L, Macfarlane A, et al. Perinatal and maternal outcomes by planned place of birth for healthy women with low risk pregnancies: the Birthplace in England national prospective cohort study. BMJ 2011 Nov 23;343:d7400.

(2) International Confederation of Midwives. Essential competencies for basic midwifery practice. 2013.

(3) Dekker G, Sibai B. Primary, secondary, and tertiary prevention of pre-eclampsia. Lancet 2001 Jan 20;357(9251):209-215.

(4) National Institute for Health and, Clinical Excellence. Routine care for the healthy pregnant woman. Antenatal care 2011.

(5) Available at: http://web.worldbank.org.(6) Hemminki E, Blondel B, Study Group on Barriers and Incentives to Prenatal Care,in Europe.

Antenatal care in Europe: varying ways of providing high-coverage services. Eur J Obstet Gynecol Reprod Biol 2001;94(1):145-148.

(7) Wiegers TA. General practitioners and their role in maternity care. Health Policy 2003 10;66(1):51-59.

(8) Rowland T, McLeod D, Froese-Burns N. Comparative study of maternity systems. 2012.(9) Wiegers TA. The quality of maternity care services as experienced by women in the Netherlands.

BMC Pregnancy Childbirth 2009 May 9;9:18.(10) Bais J, Pel M. The basis of the Dutch obstetric system: risk selection. European Clinics in Obstetrics

and Gynaecology 2006;2(4):209-212.(11) De Geus M. Midwifery in the Netherlands. 2012.(12) Hingstman L, Kenens RJ. Figures in the recordings of general practitioners. (Cijfers uit de

registratie van huisartsen, peiling 2011). Nivel, Utrecht, 2011.(13) Velden van der L, Hingstman L, Wiegers T, Kenens RJ. Obstetric care is still provided by GPs.

(Huisartsenzorg in cijfers: verloskundig actieve huisarts bestaat nog steeds). Huisarts en Wetenschap 2012;55(3):131.

(14) Stichting Perinatale Registratie Nederland. Perinatal care in the Netherlands. (Perinatale Zorg in Nederland, 2010). Utrecht, 2012.

(15) Hingstman L, Hassel van DTP, Kenens RJ. Figures in the recordings of midwives. (Cijfers uit de registratie van verloskundigen, peiling 2012). Nivel, Utrecht, 2012.

(16) Cohen AW. Scheduling the first prenatal visit: a missed opportunity. Obstet Gynecol 2010 9;203(3):192-193.

(17) Dowswell T, Carroli G, Duley L, Gates S, Gulmezoglu AM, Khan-Neelofur D, et al. Alternative versus standard packages of antenatal care for low-risk pregnancy. Cochrane Database Syst Rev 2010 Oct 6;(10).

(18) Royal College of Obstetricians & Gynaecologists. Standards of maternity care, report of a working party. 2008; Available at: http://www.rcog.org.uk/files/rcog-corp/uploaded-files/WPRMaternityStandards2008.pdf.

(19) Royal Dutch Organisation of Midwives (Koninklijke Nederlandse Organisatie van Verloskundigen). Prenatal care. Recommendations for support, interaction and counseling (Prenatale begeleiding. Aanbevelingen voor ondersteuning, interactie en voorlichting). 2008; Available at: http://www.knov.nl/docs/uploads/Kwaliteit_Richtlijnen_PrenataleVKBegeleiding.pdf. Accessed 02/14, 2012.


| 19


1

(20) Dutch Association of Obstetrics and Gynecology (Nederlandse Vereniging voor Obstetrie en Gynaecologie). Guidelines and position papers. 2002; Available at: http://www.nvog.nl/vakinformatie/Richtlijnen,+standpunten,+modelprotocollen+enz/default.aspx. Accessed 02/14.

(21) Andersen RM, Rice TH, Kominski GF. Changing the U.S. healthcare system; key issues in health services policy and management. San Francisco, CA: Jossey-Bass; 2007.

(22) LaVeist TA, Keith VM, Gutierrez ML. Black/white differences in prenatal care utilization: an assessment of predisposing and enabling factors. Health Serv Res 1995 Apr;30(1):43-58.

(23) Partridge S, Balayla J, Holcroft CA, Abenhaim HA. Inadequate prenatal care utilization and risks of infant mortality and poor birth outcome: a retrospective analysis of 28,729,765 U.S. deliveries over 8 years. Am J Perinatol 2012 Nov;29(10):787-793.

(24) Raatikainen K, Heiskanen N, Heinonen S. Under-attending free antenatal care is associated with adverse pregnancy outcomes. BMC Public Health 2007;7:268.

(25) Blondel B, Marshall B. Poor antenatal care in 20 French districts: risk factors and pregnancy outcome. J Epidemiol Community Health 1998;52:501-506.

(26) Johnson AA, Hatcher BJ, El-Khorazaty MN, Milligan RA, Bhaskar B, Rodan MF, et al. Determinants of inadequate prenatal care utilization by African American women. J Health Care Poor Underserved 2007 Aug;18(3):620-636.

(27) Wildman K, Blondel B, Nijhuis J, Defoort P, Bakoula C. European indicators of health care during pregnancy, delivery and the postpartum period. Eur J Obstet Gynecol Reprod Biol 2003 Nov 28;111 Suppl 1:S53-65.

(28) Field KS, Briggs DJ. Socio-economic and locational determinants of accessibility and utilization of primary health-care. Health Soc Care Community 2001 Sep;9(5):294-308.

(29) Ny P, Dykes AK, Molin J, Dejin-Karlsson E. Utilisation of antenatal care by country of birth in a multi-ethnic population: a four-year community-based study in Malmo, Sweden. Acta Obstet Gynecol Scand 2007;86(7):805-813.

(30) Paine LL, Lang JM, Strobino DM, Johnson TR, DeJoseph JF, Declercq ER, et al. Characteristics of nurse-midwife patients and visits, 1991. Am J Public Health 1999 Jun;89(6):906-909.

(31) Stewart SD. Economic and personal factors affecting women’s use of nurse-midwives in Michigan. Fam Plann Perspect 1998 Sep-Oct;30(5):231-235.

(32) Rowe RE, Garcia J. Social class, ethnicity and attendance for antenatal care in the United Kingdom: a systematic review. J Public Health Med 2003 Jun;25(2):113-119.

(33) Hogg W, Dyke E. Improving measurement of primary care system performance. Can Fam Physician 2011 Jul;57(7):758-60, e241-3.

(34) Chote AA, Koopmans GT, de Groot CJ, Hoefman RJ, Jaddoe VW, Hofman A, et al. Differences in timely antenatal care between first and second-generation migrants in the Netherlands. J Immigr Minor Health 2014 Aug;16(4):631-637.

(35) Chote A, de Groot C, Redekop K, Hoefman R, Koopmans G, Jaddoe V, et al. Differences in quality of antenatal care provided by midwives to low-risk pregnant dutch women in different ethnic groups. J Midwifery Womens Health 2012 Sep-Oct;57(5):461-468.

(36) Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Soc Behav 1995;36:1-10.

(37) Andersen RM. National health surveys and the behavioral model of health services use. Med Care 2008 Jul;46(7):647-653.

(38) Andersen R. Looking back at health surveys. Health Aff (Millwood) 2008 Mar-Apr;27(2):585-6; author reply 586.

(39) Feijen-de Jong EI, Jansen DE, Baarveld F, van der Schans CP, Schellevis FG, Reijneveld SA. Determinants of late and/or inadequate use of prenatal healthcare in high-income countries: a systematic review. Eur J Public Health 2012 Dec;22(6):904-913.

R1R2R3R4R5R6R7R8R9


20 |

Chapter 1

(40) Mannien J, Klomp T, Wiegers T, Pereboom M, Brug J, de Jonge A, et al. Evaluation of primary care midwifery in the Netherlands: design and rationale of a dynamic cohort study (DELIVER). BMC Health Services Research 2012;12(1):69.

CHAPTER 2

Determinants of late and/or inadequate use of prenatal

healthcare in high-income countries: a systematic review

Esther I Feijen-de Jong, Danielle EMC Jansen, Frank Baarveld, Cees van der Schans, Francois Schellevis, Sijmen A Reijneveld

European Journal of Public Health 2012 Dec;22(6):904-913

R1R2R3R4R5R6R7R8R9


22 |

Chapter 2

ABSTRACT

Background: Prenatal healthcare is likely to prevent adverse outcomes, but an adequate review of determinants of prenatal healthcare utilization is lacking.

Objective: To review systematically the evidence for the determinants of prenatal healthcare utilization in high-income countries.

Method: Search of publications in EMBASE, CINAHL, and PubMed (1992-2010). Studies that attempted to study determinants of prenatal healthcare utilization in high-income countries were included. Two reviewers independently assessed the eligibility and methodological quality of the studies. Only high-quality studies were included. Data on inadequate use (i.e. late initiation, low-use, inadequate use or non-use) were categorized as individual, contextual, and health behaviour-related determinants. Due to the heterogeneity of the studies, a quantitative meta-analysis was not possible.

Results: Ultimately eight high-quality studies were included. Low maternal age, low educational level, non-marital status, ethnic minority, planned pattern of prenatal care, hospital type, unplanned place of delivery, uninsured status, high parity, no previous premature birth, and late recognition of pregnancy were identified as individual determinants of inadequate use. Contextual determinants included living in distressed neighbourhoods, living in neighbourhoods with high rates of unemployment, single parent families, medium-average family incomes, low-educated residents, and women reporting Canadian Aboriginal status were associated with inadequate use or entering care after six months. Regarding health behaviour inadequate use was more likely among women who smoked during pregnancy.

Conclusion: Evidence for determinants of prenatal care utilization is limited. More studies are needed to ensure adequate prenatal care for pregnant women at risk.

Keywords: Prenatal healthcare utilization, Systematic review, Determinants, Maternal care


| 23

Determinants of late and/or inadequate use of prenatal healthcare in high-income countries: a systematic review

2

INTRODUCTION

Prenatal healthcare has largely contributed to the decline in perinatal and infant mortality rates in high-income countries during the last century. Prenatal care includes identification of medical conditions necessitating careful surveillance throughout pregnancy.1 Moreover, it is a way for women to integrate into the medical/obstetric care system. However, high use of prenatal healthcare services burdens the healthcare system, adds to its costs, and may medicalize pregnancy and birth.2,3

Late or inadequate use of prenatal healthcare – that is, entry after the first trimester and/or an inappropriate number of prenatal visits – may be due to individual characteristics, contextual characteristics, and health behaviour.1,4,5 These can be understood by using Andersen’s behavioural model6 on determinants for utilization of healthcare. Variations in use may help to explain differences in infant mortality and morbidity rates, and may serve as a guide for further improvements in quality of care.

The aim of this study is to provide a systematic review of the current evidence of the determinants of use of prenatal healthcare in high-income countries. A recent systematic review on this topic is lacking, the most recent ones being those by Goldenberg et al.7 in 1992 and by Rowe and Garcia4 in 2003: the latter is only comprised of UK studies. The current review includes studies focusing on all high-income countries worldwide, published since 1992.

METHODS

We followed the guidelines of the NHS Centre for Reviews and Dissemination.8

Search methodWe searched the literature that was published from January 1992 to 30 September 2010 using the following databases: PubMed, Cinahl, and Embase. Research published before 1992 was excluded as this was already included in the review by Goldenberg et al.7 Search terms were “prenatal” and “utilization”, as Mesh terms, Emtrees, and Cinahl headings, and as free text words. We made no restrictions as to language of the publication. The search was performed by a librarian and by one of the authors (EF-J), and aimed for high sensitivity, in order to ensure the inclusion of as many relevant studies as possible.

Search outcomeTwo of the authors independently scanned the resulting 880 studies as to title and abstract (when available). EF-J is an expert in prenatal healthcare, DJ in models of healthcare utilization.

R1R2R3R4R5R6R7R8R9


24 |

Chapter 2

Studies were eligible if data were presented on the determinants of prenatal healthcare utilization in high-income countries, including countries in the World Bank’s list of high-income economies (countries with a gross national income (GNI) per capita of more than USD 11 456). We removed duplicate studies (n = 66). Next, we excluded studies on the determinants of prenatal healthcare utilization by specific groups which did not contrast these groups with the general population (teenage-pregnancies, migrant-farm-workers, ethnic minorities, high-risk women, or low-income groups), studies that provided no new empirical data (reviews, editorial letters, and brief items), and studies that only provided qualitative data. The remaining 100 studies were independently read by two reviewers (EF-J and DJ or EF-J and FB). Disagreement on ambiguous citations was resolved by consensus after additional review. We also contacted the authors of studies reporting incomplete data, however, this yielded no additional information. Based on this, 59 studies were excluded for various reasons (figure 1).

Quality assessmentThe remaining 41 studies underwent quality assessment and content abstraction using the quality assessment tool developed by Gyorkos et al.9 to classify studies into three categories: weak, moderate, and strong. Strong meant that no major flaws threatened the internal validity of the study, that is, that there were minor chances of selection bias (selection of population, non-response bias), information bias (measurement of intervention and outcome variables), and confounding. The further procedure was similar to that in the previous step. Seven studies were classified as moderate, 26 studies as weak, and eight as strong. We chose to include only studies with strong methodological quality to produce reliable, unbiased, and meaningful information about our review question (Figure 1).

Data synthesisA narrative synthesis was undertaken, since a quantitative synthesis (meta-analysis) was not possible due to ample heterogeneity of research design, populations, types of interventions, and outcomes. We started with a description of the studies. Thereafter, we categorised the results using the components of Andersen’s behavioural model.6

In the Andersen model use of health services depends on individual and contextual characteristics, and on health behaviour. For the former two, the following components we measured: predisposing, enabling, or suggesting a need for use characteristics. Predisposing characteristics are existing conditions that predispose people to use (yes/no) healthcare services. Enabling/disabling characteristics facilitate or impede use. Need characteristics are conditions that patients or health providers recognize as requiring medical treatment.6


| 25


2

Records identified through database searching

(n = 880) 1992-sept 2010

Records screened on title and abstract (n=814)

Full-text articles assessed for eligibility

(n = 100)

Studies methodological assessed for eligibility

(n =41)

Studies included innarrative synthesis

(n = 8)

Duplicates removed (n=66)

Records excluded withreasons; (n=714)

Reviews, editorial letters, no quantitative study, specific

groups

Full-text studies excluded because of strength of

research method (moderate (7) or weak (26) design)

(n = 33)

Full-text studies excluded, with reasons; (n =59)

Specific groups (n=44), review (n=1), no

quantitative research (n=1), determinants not studied (n=8), other main topic of

research (n=5)

Iden

tifica

tion

Scre

enin

gEl

igib

ility

Incl

uded

Figure 1. Identification of studies

R1R2R3R4R5R6R7R8R9


26 |

Chapter 2

RESULTS

General study characteristicsTable 1 shows the general characteristics of the studies included. All eight studies were based on cross-sectional data. Some of these data were collected as part of a longitudinal (cohort) design, but none of them based findings on longitudinally collected data. The data were collected from birth certificates10-12, birth registers13, mother-baby files14, combined birth and death certificates15, and surveys16,17. Samples sizes varied from 1776516 to 593 51011 women. Studies were conducted in the United States (US)(4),10-12,15 the United Kingdom (UK) (2), 16,17, Finland, 13 and Canada14. The studies from the United States also analyzed variables on the relationship between health insurance status and prenatal healthcare utilization (enabling/disabling characteristics)10-12,15, whereas the studies from the other countries focused on only demographic variables (predisposing characteristics). Two studies12,14 assessed determinants at individual and neighbourhood (contextual) levels. The other six only examined determinants at the individual level.


| 27


2

Tabl

e 1.

Mai

n ch

arac

teri

stics

of t

he in

clud

ed s

tudi

es (N

=8):

desi

gn, s

ize,

det

erm

inan

ts, m

ain

outc

omes

, and

mai

n fin

ding

sSt

udy

Des

ign

Num

ber

of

parti

cipa

nts

Det

erm

inan

tsM

ain

outc

ome

Mai

n fin

ding

s (o

nly

sign

ifica

nt re

sult

s co

rrec

ted

for

conf

ound

ers)

Ayo

ola

et a

l.10,

USA

Dat

a an

alys

is

(200

0-20

04)

Indi

vidu

al le

vel

1363

73 (l

ive

birt

hs)

Tim

e of

pre

gnan

cy re

cogn

ition

(e

arly

; with

in 6

wee

ks o

f ge

stati

on, l

ate;

aft

er 6

wee

ks o

f ge

stati

on),

mat

erna

l age

, par

ity,

mar

ital s

tatu

s, le

vel o

f edu

catio

n,

insu

ranc

e st

atus

, soc

io-e

cono

mic

st

atus

, rac

e/et

hnic

ity, p

rior

bir

th

outc

omes

1.

Tim

e of

firs

t pre

nata

l vis

it (fi

rst

trim

este

r or

oth

er)

2.

Fre

quen

cy o

f pre

nata

l car

e vi

sits

(< 1

1 vi

sits

, 11-

15 v

isits

, > 1

5 vi

sits

)

1.

Vari

able

s pr

edic

ting

initi

ation

of c

are

befo

re 1

2 w

eeks

: Ear

ly p

regn

ancy

reco

gniti

on c

ompa

red

to la

te

reco

gniti

on, n

o pr

ior

birt

h co

mpa

red

to 1

or

mor

e pr

ior

birt

hs, m

arri

ed w

omen

com

pare

d to

unm

arri

ed

wom

en, h

igh

scho

ol a

nd a

bove

hig

h sc

hool

com

pare

d to

bel

ow h

igh

scho

ol, a

ge o

f mot

her,

non-

hisp

anic

w

hite

s co

mpa

red

to b

lack

, Asi

an a

nd H

ispa

nic

wom

en,

Med

icai

d an

d pr

ivat

e in

sura

nce

com

pare

d to

no

insu

ranc

e, p

revi

ous

prem

atur

e bi

rth

com

pare

d to

no

prev

ious

pre

mat

ure

birt

h2.

<

11 v

isits

(com

pare

d to

11-

15 v

isits

), la

te p

regn

ancy

re

cogn

ition

com

pare

d to

late

pre

gnan

cy re

cogn

ition

, 1

or m

ore

prio

r bi

rths

com

pare

d to

no

prio

r bi

rth,

un

mar

ried

wom

en c

ompa

red

to m

arri

ed w

omen

, be

low

hig

h sc

hool

com

pare

d to

hig

h sc

hool

and

abo

ve

high

sch

ool,

wom

en a

ged

16 to

40

year

s co

mpa

red

to

wom

en a

ged

11 to

15

year

s, H

ispa

nic,

Asi

an, A

mer

ican

In

dian

/Ala

ska

nativ

e, b

lack

wom

en c

ompa

red

to n

on

His

pani

c w

hite

wom

en, n

o in

sura

nce

com

pare

d to

M

edic

aid

or p

riva

te in

sura

nce.

>15

visi

ts (c

ompa

red

to 1

1-15

vis

its),

earl

y pr

egna

ncy

reco

gniti

on c

ompa

red

to la

te re

cogn

ition

, pub

lic

assi

stan

ce c

ompa

red

to n

o pu

blic

ass

ista

nce1,

2

Mar

ín15

, Pue

rto

Rico

, USA

Qua

si

expe

rim

enta

l tim

e se

ries

(199

5-20

00)

Indi

vidu

al le

vel

3706

52 (l

ive

birt

hs)

Hea

lth in

sura

nce

1.

Use

and

non

use

of p

rena

tal c

are

(any

vi

sit o

r no

vis

it)2.

Ti

mel

y of

initi

ation

of c

are

(firs

t tr

imes

ter

or o

ther

)3.

Th

e nu

mbe

r of

vis

its

for

pren

atal

car

e4.

Th

e ad

equa

cy o

f car

e A

PNCU

3

Dis

trib

ution

in tw

o ca

tego

ries

;•

Ade

quat

e (a

dequ

ate

and

adeq

uate

pl

us)

•In

adeq

uate

(int

erm

edia

te a

nd

inad

equa

te)

- m

ore

nonu

se o

f pnc

for

Med

icai

d, M

MC

and

unin

sure

d w

omen

com

pare

d to

pri

vate

insu

red

wom

en-

mor

e pn

c us

e in

the

first

trim

este

r fo

r pr

ivat

e in

sure

d w

omen

com

pare

d to

Med

icai

d, M

MC

and

unin

sure

d w

omen

- le

ss v

isits

for

Med

icai

d, M

MC

and

unin

sure

d w

omen

co

mpa

red

to p

riva

te in

sure

d w

omen

- m

ore

inad

equa

te u

se fo

r M

edic

aid,

MM

C an

d un

insu

red

wom

en c

ompa

red

to p

riva

te in

sure

d w

omen

4

R1R2R3R4R5R6R7R8R9


28 |

Chapter 2

Hea

man

et a

l.14,

Cana

daPo

pula

tion

base

d co

hort

stu

dy(1

991-

2000

)

Indi

vidu

al a

nd

neig

hbou

rhoo

d le

vel

1492

91 (l

ive

birt

hs)

Mat

erna

l var

iabl

es:

Age

, par

ity, s

ingl

e pa

rent

sta

tus,

em

ploy

men

t sta

tus,

abo

rigi

nal

stat

us, f

amily

inco

me,

sm

okin

g,

educ

ation

, rec

ent i

mm

igra

nts

Gin

dex5

dist

ribu

tion

in tw

o ca

tego

ries

;•

inad

equa

te c

are

and

no c

are

•ad

equa

te c

are

Nei

ghbo

urho

od c

hara

cter

istic

s as

soci

ated

with

inad

equa

te

pren

atal

car

e ar

e: h

igh

perc

enta

ge o

f sin

gle

pare

nt fa

mili

es

com

pare

d to

low

sin

gle

pare

nt fa

mili

es, h

igh

and

med

ium

un

empl

oym

ent r

ates

com

pare

d to

low

une

mpl

oym

ent r

ates

, hi

gh a

nd m

ediu

m p

erce

ntag

e of

peo

ple

repo

rting

Abo

rigi

nal

stat

us c

ompa

red

to lo

w p

erce

ntag

e of

peo

ple

repo

rting

A

bori

gina

l sta

tus,

hig

h an

d m

ediu

m s

mok

ing

rate

s co

mpa

red

to lo

w s

mok

ing

rate

s, h

igh

and

med

ium

per

cent

age

of w

omen

repo

rting

less

than

nin

e ye

ars

of e

duca

tion

com

pare

d to

a lo

w p

erce

ntag

e of

wom

en re

porti

ng le

ss th

an

nine

yea

rs o

f edu

catio

n an

d a

high

rate

of r

ecen

t im

mig

rant

s co

mpa

red

to n

eigh

bour

hood

s w

ith lo

w re

cent

imm

igra

nts6

Kupe

k et

al. 1

6 ,U

KCr

oss

secti

onal

st

udy

(Aug

ust 1

994-

July

19

95)

Indi

vidu

al le

vel

1776

5 (li

vebo

rn

and

stillb

orn

babi

es)

Obs

tetr

ic ri

sk fa

ctor

s:H

isto

ry o

f dia

bete

s m

ellit

us,

card

iac

dise

ase,

ess

entia

l hy

pert

ensi

on, r

enal

dis

ease

, th

rom

bosi

s, s

ubst

ance

abu

se

and

a ra

nge

of le

ss p

reva

lent

di

sord

ers7.

Dim

inuti

ve s

tatu

re

(<15

2cm

), ex

trem

ities

in w

eigt

h (<

45kg

or

>89k

g) a

nd e

xtre

miti

es

in m

ater

nal a

ge (p

rim

ipar

ous:

<18

or >

30 y

ears

, mul

tipar

ous:

<18

or

> 35

yea

rs)

For

mul

tipar

ous

wom

en; l

ower

se

gmen

t cae

sare

an s

ectio

n,

prev

ious

stil

lbir

th o

r ne

onat

al

deat

h, p

revi

ous

pret

erm

del

iver

y,

prev

ious

intr

aute

rine

gro

wth

re

tard

ation

, pre

viou

s de

liver

y of

a lo

w b

irth

wei

ght i

nfan

t and

pr

evio

us d

eliv

er o

f mor

e th

an

thre

e liv

ebor

n in

fant

s.Pr

ovid

er c

hara

cter

istic

s:Ty

pe o

f hos

pita

l, pl

anne

d pa

tter

n of

pre

nata

l car

e, p

lann

ed p

lace

of

deliv

ery

Soci

odem

ogra

phic

cha

ract

eris

tics:

Ethn

icity

, sm

okin

g st

atus

, par

ity,

mat

erna

l age

Late

initi

ation

of p

rena

tal c

are:

•la

ter

than

10

wee

ks o

f ges

tatio

n•

late

r th

an 1

8 w

eeks

of g

esta

tion

Vari

able

s as

soci

ated

with

late

initi

ation

of p

rena

tal c

are

(>10

w

eeks

): lo

w r

isk

mul

tipar

ous

com

pare

d to

all

othe

r gr

oups

, yo

ung

mat

erna

l age

, sm

oker

s co

mpa

red

to n

on-s

mok

ers,

Pa

kist

ani w

omen

com

pare

d to

whi

te B

ritis

h w

omen

, hos

pita

l ty

pe(u

rban

non

teac

hing

com

pare

d to

urb

an te

achi

ng a

nd

rura

l dis

tric

t gen

eral

), pl

anne

d pa

tter

n of

pre

nata

l car

e (G

P/m

idw

ife/T

M c

are

com

pare

d to

sha

red

care

with

out t

eam

m

idw

ifery

), pl

anne

d pl

ace

of d

eliv

ery

(isol

ated

GP

unit

com

pare

d to

hos

pita

l con

sulta

nt u

nit)

Vari

able

s as

soci

ated

with

late

initi

ation

of p

rena

tal c

are

(>18

wee

ks):

low

ris

k m

ultip

arou

s co

mpa

red

to lo

w r

isk

prim

ipar

ous,

you

ng m

ater

nal a

ge, s

mok

ers

com

pare

d to

no

n-sm

oker

s, P

akis

tani

, Ind

ian

and

all o

ther

s co

mpa

red

to w

hite

Bri

tish

wom

en, h

ospi

tal t

ype

(urb

an te

achi

ng

com

pare

d to

non

urb

an te

achi

ng),

hosp

ital t

ype

(hos

pita

l co

nsul

tant

car

e co

mpa

red

to s

hare

d ca

re w

ithou

t tea

m

mid

wife

ry),

plan

ned

plac

e of

del

iver

y (h

ospi

tal c

onsu

ltant

un

it co

mpa

red

to G

P un

it w

ithin

hos

pita

l)8


| 29


2

Hea

man

et a

l.14,

Cana

daPo

pula

tion

base

d co

hort

stu

dy(1

991-

2000

)

Indi

vidu

al a

nd

neig

hbou

rhoo

d le

vel

1492

91 (l

ive

birt

hs)

Mat

erna

l var

iabl

es:

Age

, par

ity, s

ingl

e pa

rent

sta

tus,

em

ploy

men

t sta

tus,

abo

rigi

nal

stat

us, f

amily

inco

me,

sm

okin

g,

educ

ation

, rec

ent i

mm

igra

nts

Gin

dex5

dist

ribu

tion

in tw

o ca

tego

ries

;•

inad

equa

te c

are

and

no c

are

•ad

equa

te c

are

Nei

ghbo

urho

od c

hara

cter

istic

s as

soci

ated

with

inad

equa

te

pren

atal

car

e ar

e: h

igh

perc

enta

ge o

f sin

gle

pare

nt fa

mili

es

com

pare

d to

low

sin

gle

pare

nt fa

mili

es, h

igh

and

med

ium

un

empl

oym

ent r

ates

com

pare

d to

low

une

mpl

oym

ent r

ates

, hi

gh a

nd m

ediu

m p

erce

ntag

e of

peo

ple

repo

rting

Abo

rigi

nal

stat

us c

ompa

red

to lo

w p

erce

ntag

e of

peo

ple

repo

rting

A

bori

gina

l sta

tus,

hig

h an

d m

ediu

m s

mok

ing

rate

s co

mpa

red

to lo

w s

mok

ing

rate

s, h

igh

and

med

ium

per

cent

age

of w

omen

repo

rting

less

than

nin

e ye

ars

of e

duca

tion

com

pare

d to

a lo

w p

erce

ntag

e of

wom

en re

porti

ng le

ss th

an

nine

yea

rs o

f edu

catio

n an

d a

high

rate

of r

ecen

t im

mig

rant

s co

mpa

red

to n

eigh

bour

hood

s w

ith lo

w re

cent

imm

igra

nts6

Kupe

k et

al. 1

6 ,U

KCr

oss

secti

onal

st

udy

(Aug

ust 1

994-

July

19

95)

Indi

vidu

al le

vel

1776

5 (li

vebo

rn

and

stillb

orn

babi

es)

Obs

tetr

ic ri

sk fa

ctor

s:H

isto

ry o

f dia

bete

s m

ellit

us,

card

iac

dise

ase,

ess

entia

l hy

pert

ensi

on, r

enal

dis

ease

, th

rom

bosi

s, s

ubst

ance

abu

se

and

a ra

nge

of le

ss p

reva

lent

di

sord

ers7.

Dim

inuti

ve s

tatu

re

(<15

2cm

), ex

trem

ities

in w

eigt

h (<

45kg

or

>89k

g) a

nd e

xtre

miti

es

in m

ater

nal a

ge (p

rim

ipar

ous:

<18

or >

30 y

ears

, mul

tipar

ous:

<18

or

> 35

yea

rs)

For

mul

tipar

ous

wom

en; l

ower

se

gmen

t cae

sare

an s

ectio

n,

prev

ious

stil

lbir

th o

r ne

onat

al

deat

h, p

revi

ous

pret

erm

del

iver

y,

prev

ious

intr

aute

rine

gro

wth

re

tard

ation

, pre

viou

s de

liver

y of

a lo

w b

irth

wei

ght i

nfan

t and

pr

evio

us d

eliv

er o

f mor

e th

an

thre

e liv

ebor

n in

fant

s.Pr

ovid

er c

hara

cter

istic

s:Ty

pe o

f hos

pita

l, pl

anne

d pa

tter

n of

pre

nata

l car

e, p

lann

ed p

lace

of

deliv

ery

Soci

odem

ogra

phic

cha

ract

eris

tics:

Ethn

icity

, sm

okin

g st

atus

, par

ity,

mat

erna

l age

Late

initi

ation

of p

rena

tal c

are:

•la

ter

than

10

wee

ks o

f ges

tatio

n•

late

r th

an 1

8 w

eeks

of g

esta

tion

Vari

able

s as

soci

ated

with

late

initi

ation

of p

rena

tal c

are

(>10

w

eeks

): lo

w r

isk

mul

tipar

ous

com

pare

d to

all

othe

r gr

oups

, yo

ung

mat

erna

l age

, sm

oker

s co

mpa

red

to n

on-s

mok

ers,

Pa

kist

ani w

omen

com

pare

d to

whi

te B

ritis

h w

omen

, hos

pita

l ty

pe(u

rban

non

teac

hing

com

pare

d to

urb

an te

achi

ng a

nd

rura

l dis

tric

t gen

eral

), pl

anne

d pa

tter

n of

pre

nata

l car

e (G

P/m

idw

ife/T

M c

are

com

pare

d to

sha

red

care

with

out t

eam

m

idw

ifery

), pl

anne

d pl

ace

of d

eliv

ery

(isol

ated

GP

unit

com

pare

d to

hos

pita

l con

sulta

nt u

nit)

Vari

able

s as

soci

ated

with

late

initi

ation

of p

rena

tal c

are

(>18

wee

ks):

low

ris

k m

ultip

arou

s co

mpa

red

to lo

w r

isk

prim

ipar

ous,

you

ng m

ater

nal a

ge, s

mok

ers

com

pare

d to

no

n-sm

oker

s, P

akis

tani

, Ind

ian

and

all o

ther

s co

mpa

red

to w

hite

Bri

tish

wom

en, h

ospi

tal t

ype

(urb

an te

achi

ng

com

pare

d to

non

urb

an te

achi

ng),

hosp

ital t

ype

(hos

pita

l co

nsul

tant

car

e co

mpa

red

to s

hare

d ca

re w

ithou

t tea

m

mid

wife

ry),

plan

ned

plac

e of

del

iver

y (h

ospi

tal c

onsu

ltant

un

it co

mpa

red

to G

P un

it w

ithin

hos

pita

l)8

Petr

ou e

t al.17

, U

KCr

oss

secti

onal

st

udy

(Aug

ust 1

994-

July

19

95)

Indi

vidu

al le

vel

1797

8 (li

vebo

rn

and

stillb

orn

babi

es)

All

varia

bles

sim

ilar a

s Ku

pek

et

al,a

nd a

dditi

onal

ly:

Ges

tatio

nal a

ge a

t del

iver

y,

num

ber

of h

ospi

tal a

dmis

sion

s

Tota

l num

ber

of p

rena

tal v

isit

sVa

riab

les

asso

ciat

ed w

ith d

ecre

ased

num

bers

of p

rena

tal

visi

ts:lo

w r

isk

mul

tipar

ous

com

pare

d to

hig

h ri

sk

mul

tipar

ous,

low

-ris

k pr

imip

arou

s, h

igh-

risk

pri

mip

arou

s an

d un

know

n ri

sk p

rim

ipar

ous,

hos

pita

l typ

e (u

rban

non

te

achi

ng c

ompa

red

to u

rban

teac

hing

an

rura

l dis

tric

t ge

nera

l), p

lann

ed p

atter

n of

pre

nata

l car

e (s

hare

d ca

re

with

out m

idw

ifery

com

pare

d to

GP/

mid

wife

/tea

m m

idw

ifery

ca

re a

nd h

ospi

tal c

onsu

ltant

car

e), c

hang

e in

patt

ern

of

pren

atal

car

e fo

r cl

inic

al re

ason

s co

mpa

red

to n

o ch

ange

, no

cha

nge

in p

atter

n of

pre

nata

l car

e co

mpa

red

to a

cha

nge

for

non-

clin

ical

reas

ons,

Whi

te B

ritis

h w

omen

com

pare

d to

Indi

an, P

akis

tani

and

all

othe

r w

omen

, non

-sm

oker

s co

mpa

red

to s

mok

ers,

ges

tatio

nal a

ge a

t boo

king

per

wee

k ch

ange

(dec

reas

e), g

esta

toni

al a

ge a

t del

iver

y pe

r w

eek

chan

ge (i

ncre

ase)

, mat

erna

l age

at b

ooki

ng p

er y

ear

chan

ge

(incr

ease

) , n

umbe

r of

hos

pita

l adm

issi

ons

per

adm

issi

on

chan

ge (i

ncre

ase)

and

frag

men

tatio

n of

car

e le

ads

to le

ss

visi

ts9

Perl

off a

nd

Jaff

ee, 1

99912

, U

SA

Retr

ospe

ctive

an

alys

is o

f bir

th

certi

ficat

es(1

991-

1992

)

Indi

vidu

al a

nd

neig

hbou

rhoo

d le

vel

2206

94

(live

born

an

d sti

llbor

n ba

bies

)

Pred

ispo

sing

var

iabl

es: a

ge,

educ

ation

, rac

e or

eth

nici

ty, p

arity

Enab

ling

vari

able

s:M

arita

l sta

tus,

type

of h

ealth

in

sura

nce

Nei

ghbo

urho

od le

vel f

acto

rs:

evid

ence

of m

edic

al r

isk,

eco

nom

ic

oppo

rtun

ity s

truc

ture

, hea

lthca

re

oppo

rtun

ity

Late

initi

ation

of p

rena

tal c

are:

Late

r th

an s

ix m

onth

s of

pre

gnan

cy (m

onth

s se

ven,

eig

ht a

nd n

ine

or n

ot a

t all)

Pred

ispo

sing

cha

ract

eris

tics

asso

ciat

ed w

ith la

te in

itiati

on o

f pr

enat

al c

are:

11-

19 y

ears

of a

ge, n

ot h

igh

scho

ol g

radu

ate,

H

ispa

nic

whi

te, n

on-H

ispa

nic

blac

k, H

ispa

nic

blac

k, th

ree

or

mor

e liv

e bi

rths

Enab

ling

vari

able

s as

soci

ated

with

late

initi

ation

of p

rena

tal

care

: unm

arri

ed, u

nins

ured

, Med

icai

dN

eigh

bour

hood

cha

ract

eris

tics

asso

ciat

ed w

ith la

te in

itiati

on

of p

rena

tal c

are:

livi

ng in

a s

hort

age

area

10, d

istr

esse

d zi

p co

de11

-12

Hem

min

ki a

nd

Gis

sler

13, F

inla

ndRe

tros

pecti

ve

anal

ysis

of b

irth

ce

rtific

ates

(198

7)

Indi

vidu

al le

vel

5957

9 (li

vebo

rn

and

stillb

orn

babi

es)

Age

, mar

ital s

tatu

s, e

duca

tion,

sm

okin

g st

atus

, pre

viou

s bi

rths

1.

Num

ber

of p

rena

tal v

isit

s•

Man

y vi

sits

(>1.

7)•

Aver

age

visi

ts (1

.0-1

.7)

•Fe

w v

isits

(<1.

0)13

2.

Tim

ing

of fi

rst p

rena

tal v

isit

•Ea

rly

(<8

wee

ks o

f ges

tatio

n)•

aver

age

(8-1

2 w

eeks

)•

late

(> 1

2 w

eeks

)

1.

Vari

able

s as

soci

ated

with

man

y vi

sits

:Pr

imig

ravi

da: A

ge ≥

20

year

s, b

eing

mar

ried

, edu

catio

n >

9yea

rsM

ultig

ravi

da: e

duca

tion

>9 y

ears

, pre

viou

s bi

rths

<3Va

riab

les

asso

ciat

ed w

ith fe

w v

isits

:Pr

imig

ravi

da: a

ge <

20

year

s, n

on-m

arri

ed, e

duca

tion

≤ 9y

ears

, sm

oker

Mul

tigra

vida

: age

< 2

0 ye

ars,

non

-mar

ried

, edu

catio

n ≤

9yea

rs, s

mok

er, p

revi

ous

birt

hs ≥

32.

Va

riab

les

asso

ciat

ed w

ith e

arly

att

endi

ng:

Prim

igra

vida

: bei

ng m

arri

edM

ultig

ravi

da: e

duca

tion

>9 y

ears

, pre

viou

s bi

rths

<3Va

riab

les

asso

ciat

ed w

ith la

te a

tten

ding

:Pr

imig

ravi

da: a

ge <

20

year

s, n

on-m

arri

ed, e

duca

tion

≤ 9y

ears

, sm

oker

Mul

tigra

vida

: age

< 2

0 ye

ars,

non

-mar

ried

, edu

catio

n ≤

9yea

rs, p

revi

ous

birt

hs ≥

314

R1R2R3R4R5R6R7R8R9


30 |

Chapter 2

Brav

eman

et a

l.

11, U

SARe

tros

pecti

ve

anal

ysis

of b

irth

ce

rtific

ates

(199

0)

Indi

vidu

al le

vel

5935

10 (l

ive

birt

hs)

Insu

ranc

e co

vera

geM

ater

nal c

hara

cter

istic

s: ra

cial

or

eth

nic

grou

p, b

irth

plac

e, a

ge,

pari

ty, e

duca

tion,

mar

ital s

tatu

s

1.

Tim

ing

of fi

rst v

isit

Unti

mel

y; a

fter

firs

t tri

mes

ter

2.

The

adeq

uacy

of c

are

APN

CU1

Dis

trib

ution

in tw

o ca

tego

ries

;•

Ade

quat

e (a

dequ

ate

and

adeq

uate

pl

us)

•In

adeq

uate

(int

erm

edia

te a

nd

inad

equa

te)

3.

Rece

ivin

g no

pre

nata

l car

e

1.

Fact

ors

asso

ciat

ed w

ith u

ntim

ely

initi

ation

of fi

rst v

isit:

in

sura

nce

stat

us (u

nins

ured

, med

i-Cal

, Kai

ser

Nor

th,

othe

r pr

ivat

e pr

epai

d co

mpa

red

to p

riva

te in

sura

nce)

, ra

ce/e

thni

city

(Afr

ican

Am

eric

an, A

sian

Am

eric

an,

Latin

a, N

ative

Am

eric

an c

ompa

red

to E

urop

ean

Am

eric

an),

birt

hpla

ce(f

orei

gn b

orn

com

pare

d to

US-

born

), ag

e gr

oup

(≤17

, 18-

19,2

0-34

com

pare

d w

ith ≥

35

), pr

evio

us li

ve b

irth

s (1

-3, ≥

4 co

mpa

red

to n

one)

, ed

ucati

on (0

-9, 1

0-11

, 12,

13-

15 c

ompa

red

with

≥ 1

6),

Non

-mar

ried

com

pare

d w

ith m

arri

ed

2.

Fact

ors

asso

ciat

ed w

ith in

adeq

uate

use

of p

rena

tal

care

: ins

uran

ce s

tatu

s (u

nins

ured

, med

i-Cal

, Kai

ser

Nor

th, K

aise

r So

uth,

oth

er p

riva

te p

repa

id) r

ace/

ethn

icity

(Afr

ican

Am

eric

an, A

sian

Am

eric

an, L

atina

, N

ative

Am

eric

an),

birt

hpla

ce(f

orei

gn b

orn)

, age

gro

up

(≤17

, 18-

19,2

0-34

), pr

evio

us li

ve b

irth

s (1

-3, ≥

4),

educ

ation

(0-9

, 10-

11, 1

2, 1

3-15

), N

on-m

arri

ed3.

Fa

ctor

s as

soci

ated

with

no

pren

atal

car

e us

e:

(uni

nsur

ed, K

aise

r N

orth

) rac

e/et

hnic

ity (A

fric

an

Am

eric

an, A

sian

Am

eric

an, N

ative

Am

eric

an),

birt

hpla

ce(f

orei

gn b

orn)

, age

gro

up (≤

17, 1

8-19

,20-

34),

prev

ious

live

bir

ths

(1-3

, ≥4)

, edu

catio

n (0

-9,

10-1

1, 1

2, 1

3-15

), N

on-m

arri

ed15

1 Adj

uste

d fo

r: a

ge, p

arity

, mar

ital s

tatu

s, le

vel o

f edu

catio

n, in

sura

nce

stat

us, s

ocio

econ

omic

sta

tus,

race

/eth

nici

ty, a

nd p

rior

bir

th o

utco

mes

as

prem

atur

ity o

r lo

w b

irth

wei

ght.

2 The

soc

ioec

onom

ic s

tatu

s w

as m

easu

red

acco

rdin

g to

whe

ther

a w

oman

was

rec

eivi

ng a

ny p

ublic

ass

ista

nce

prog

ram

dur

ing

preg

nanc

y or

not

, and

whe

n th

e so

urce

of

inco

me

cam

e fr

om

gove

rnm

ent a

id3 K

otel

chuc

k’s

adeq

uacy

of p

rena

tal c

are

utiliz

ation

inde

x (A

PNCU

), th

is in

dex

com

bine

s in

form

ation

on

the

time

of in

itiati

on o

f pre

nata

l car

e an

d th

e to

tal n

umbe

r of p

rena

tal v

isits

. The

Kot

elch

uck

inde

x cl

assi

fies

the

adeq

uacy

of i

nitia

tion

as fo

llow

s: p

regn

ancy

mon

ths

1 an

d 2,

mon

ths

3 an

d 4,

mon

ths

5 an

d 6

and

mon

ths

7-9,

with

the

und

erly

ing

assu

mpti

on t

hat

the

earl

ier

pren

atal

car

e be

gins

the

bett

er. T

o cl

assi

fy th

e ad

equa

cy o

f rec

eive

d se

rvic

es, t

he n

umbe

r of p

rena

tal v

isits

is c

ompa

red

to th

e ex

pect

ed n

umbe

r of v

isits

for t

he p

erio

d be

twee

n th

e fir

st p

rena

tal c

are

visi

t and

th

e de

liver

y da

te. T

he e

xpec

ted

num

ber

of v

isits

is b

ased

on

the

Am

eric

an C

olle

ge o

f Obs

tetr

icia

ns a

nd G

ynec

olog

ists

pre

nata

l car

e st

anda

rds

for

unco

mpl

icat

ed p

regn

anci

es a

nd is

adj

uste

d fo

r th

e ge

stati

onal

age

whe

n ca

re b

egan

and

for

the

gest

ation

al a

ge a

t de

liver

y. A

rati

o of

obs

erve

d to

exp

ecte

d vi

sits

is c

alcu

late

d an

d gr

oupe

d in

to fo

ur c

ateg

orie

s: In

adeq

uate

(les

s th

an 5

0% o

f ex

pect

ed v

isits

, int

erm

edia

te (5

0-79

%),

adeq

uate

(80-

109%

) and

ade

quat

e pl

us (1

10%

or

mor

e)4 A

djus

ted

for

mat

erna

l age

, yea

rs o

f for

mal

edu

catio

n, m

arita

l sta

tus,

tob

acco

use

, alc

ohol

use

, med

ical

ris

k fa

ctor

s, m

edia

n fa

mily

inco

me,

pop

ulati

on d

ensi

ty, r

ate

of p

hysi

cian

s pe

r th

ousa

nd

pers

ons,

infa

nt m

orta

lity

rate

5 G

inde

x is

cal

cula

ted

on th

e ba

sis

of th

e nu

mbe

r of

PN

C vi

sits

and

the

mon

th c

are

bega

n. It

con

sist

of t

hree

cat

egor

ies;

“no

car

e” c

ateg

ory

is a

ssig

ned

to w

omen

with

no

PNC

visi

ts, “

inad

equa

te

care

” ca

tego

ry is

ass

igne

d to

wom

en w

ho (

1) b

egan

PN

C in

the

firs

t or

sec

ond

trim

este

r an

d w

ho h

ad o

nly

one

visi

t if

deliv

erin

g at

or

befo

re 2

9 w

eeks

’ ges

tatio

n, t

wo

or fe

wer

vis

its a

t 30

-31

wee

ks, t

hree

or f

ewer

vis

its a

t 32-

33 w

eeks

, and

four

or f

ewer

vis

its a

t 34

wee

ks o

r lat

er; o

r (2)

beg

an P

NC

in th

e th

ird tr

imes

ter a

nd h

ad n

ine

or fe

wer

vis

its if

del

iver

ing

betw

een

26 a

nd 3

1 w

eeks

’ ge

stati

on, 1

0 or

few

er v

isits

at 3

2-35

wee

ks, o

r 12

or

few

er v

isits

at 3

6 w

eeks

or

late

r.13

6 A

djus

ted

for a

ge, p

arity

, sin

gle

pare

nt fa

mili

es, u

nem

ploy

men

t, a

bori

gina

l sta

tus,

ave

rage

fam

ily in

com

e, s

mok

ing,

edu

catio

n, re

cent

imm

igra

nts

and

inte

racti

ons

betw

een

age,

par

ity, a

nd w

ithin

–m

othe

r in

depe

nden

cy7 A

nore

xia,

ast

hma,

dep

ress

ion,

epi

leps

y an

d sc

hizo

phre

nia


| 31


2

8 Adj

uste

d fo

r th

e eff

ects

of r

isk

stat

us a

t boo

king

, mat

erna

l age

at b

ooki

ng, s

mok

ing

stat

us, e

thni

city

, typ

e of

hos

pita

l at b

ooki

ng, p

lann

ed p

atter

n of

pre

nata

l car

e an

pla

nned

pla

ce o

f del

iver

y9 A

djus

ted

for

the

effec

ts o

f ri

sk s

tatu

s at

boo

king

, ris

k st

atus

dur

ing

pren

atal

car

e, t

ype

of h

ospi

tal a

t bo

okin

g, p

lann

ed p

atter

n of

pre

nata

l car

e, c

hang

es in

patt

ern

of p

rena

tal c

are,

eth

nici

ty,

smok

ing

stat

us, g

esta

tiona

l age

at b

ooki

ng, g

esta

tiona

l age

at d

eliv

ery,

mat

erna

l age

at b

ooki

ng, n

umbe

r of

hos

pita

l adm

issi

ons

and

frag

men

tatio

n of

car

e10

Few

er th

an 3

2 offi

ce-b

ased

pri

mar

y ca

re p

hysi

cian

s11

60%

or

mor

e no

n w

hite

pop

ulati

on a

nd 3

0% in

com

es b

elow

the

pove

rty

line

12 A

ll th

e va

riab

les

wer

e in

clud

ed in

a fu

ll lo

gisti

c re

gres

sion

mod

el a

nd w

ere

cont

rolle

d fo

r pr

edis

posi

ng, e

nabl

ing

and

neig

hbou

rhoo

d ch

arac

teri

stics

13 Th

e ac

tual

num

ber o

f vis

its w

as d

ivid

ed b

y th

e re

com

men

ded

num

ber o

f vis

its fo

r tha

t ges

tatio

n le

ngth

(rec

omm

ende

d nu

mbe

r: u

ntil 3

2 w

eeks

eve

ry fo

ur w

eeks

one

vis

it, w

eeks

33-

36 o

ne v

isits

ev

ery

two

wee

ks, w

eeks

37-

40 e

very

wee

k on

e vi

sits

and

aft

er th

e 40

th w

eek

two

visi

ts a

wee

k.14

Adj

uste

d fo

r m

arita

l sta

tus,

age

, yea

rs o

f edu

catio

n, p

revi

ous

birt

hs, s

mok

ing

stat

us15

Adj

uste

d fo

r m

ater

nal i

nsur

ance

sta

tus,

race

/eth

nici

ty, b

irth

plac

e, a

ge, p

arity

, edu

catio

n an

d m

arita

l sta

tus

R1R2R3R4R5R6R7R8R9


32 |

Chapter 2

Main outcome measuresFour outcome measures were used by two studies each. Two studies11,15 used the same definition of initiation of care, namely, starting care after the first trimester, but without clear operationalization of ‘first trimester’. Ayoola et al.10 and Hemminki and Gissler13 defined initiation as starting before or after 12 weeks of gestation. Marin et al.15 and Braveman et al.11 used the Kotelchuck Index to measure adequacy of care. The number of prenatal visits was defined similarly by Marin et al.15,17 and Petrou et al.17

The other studies all defined the main outcomes differently (table 2). Timing of the initiation of prenatal care was an important outcome, just as the number of prenatal care visits. Adequacy of prenatal healthcare utilization was measured by using two indices: the Adequacy of Prenatal Care Utilization index (APNCU) and the Graduated INDEX of PNC utilization (GINDEX), but dichotomized into adequate (> 80% expected of visits) and inadequate care (< 80% of expected visits).11,15

Table 2. Differences and variations of the main outcomes between the included studiesInitiation of care No prenatal care Number of prenatal visits Adequacy of prenatal care- First trimester or

other (Marín et al. and Braveman et al.)

- >10 weeks (Kupek et al.)- >12 weeks (Hemminki and

Gissler and Ayoola et al.)- >18 weeks (Kupek et al.)- >6 months (Perloff and

Jaffee)- < 8 weeks (Hemminki and

Gissler)- 8 to 12 weeks (Hemminki

and Gissler)

Non-use of prenatal care and receiving no prenatal care (Marín et al. and Braveman et al.)

- Frequency (Marín et al. and Petrou et al.)

- Actual number of visits divided by the recommended number of visits (Hemminki and Gissler)

- Frequency, < 11 visits, 11-15 visits, > 15 visits (Ayoola et al.)

- Kotelchuck Index, Adequacy of Prenatal Care Utilization Index (combination of initiation of care and the received services) categorised into two groups: adequate care and non-adequate care (Marín et al. and Braveman et al.)

- Gindex (Heaman et al.)

Determinants of prenatal healthcare utilization according to Andersen’s behavioural model (Table 1)Individual predisposing characteristicsSix studies10-13,16,17 examined the association between age and prenatal healthcare utilization. All studies showed an association between young maternal age (< 20 years) and lower use of prenatal healthcare services.

Four studies10-13 showed that less education (< 9 years) was associated with low use,10,11,13 late entry of prenatal care,10-13 or receiving no care at all11.

Five studies assessed ethnicity as an independent variable, but with widely differing operationalisations. Kupek et al.16 and Petrou et al.17 categorized ethnicity as: white British, Indian, Pakistani, and others. They showed that compared to white British women, all other women were more likely to fail to initiate prenatal care by 18 weeks of gestational age16, and


| 33


2

had fewer prenatal visits in pregnancy.17 Perloff and Jaffee12 chose to categorise ethnicity into four categories to distinguish between white, black, Hispanic white, and Hispanic black women. They concluded that women of colour were more likely to enter care late or not at all. Ayoola et al.10 concluded that black, Asian, Hispanic, and American Indian women were more likely to have less than 11 prenatal visits than white women. Finally, Braveman et al.11 categorised ethnicity as African-American, Asian-American, European-American, Latina, Native-American, and other. They found that compared to European-Americans all other groups were more likely to enter prenatal care after the first trimester and to receive an inadequate number of prenatal visits. The same was found for foreign-born as compared to US-born women.

Four studies10-13 examined the effect of marital status on prenatal healthcare utilization. These studies showed that unmarried women were more likely to initiate prenatal care late11, to receive an inadequate number of prenatal visits10,11,13, and not to enter care at all11 as compared to married women.

Individual enabling/disabling characteristicsFour studies10-12,15 assessed the effect of health insurance status on the initiation of prenatal care, on non-use of prenatal care and on adequacy of care. Uninsured women,10-12,15 women with Medicaid insurance12,15 or with private prepaid insurance were more likely to enter prenatal care late as compared to private fee-for-service insurance.11 Marín et al.15 and Braveman et al.11 showed more non-use among women having Medicaid insurance,11,15 private prepaid insurance,11 or no insurance11,15 as compared to women having private insurance. Regarding adequacy of care, Marín et al.15 found more inadequate use of care among uninsured women, women with Medicaid insurance or with private prepaid insurance as compared to women with private insurance. Ayoola et al.10 showed that women with Medicaid or private insurance more often had at least 11 prenatal visits. They also showed that women participating in a public assistance program more often had at least 16 visits than the non-public assistance group.

Two studies16,17 examined the association between provider characteristics and the initiation of care. Kupek et al.16 showed that late prenatal care (after 10 weeks or after 18 weeks of gestation) was associated with type of hospital at booking, the planned pattern of prenatal care, and the planned place of delivery. Petrou et al.17 showed that women with shared care without a midwifery team had more prenatal visits as compared to women with other types of prenatal care (table 1). The same applied to women in urban non-teaching hospitals as compared to women in urban teaching hospitals and rural district general hospitals.

R1R2R3R4R5R6R7R8R9


34 |

Chapter 2

Individual need characteristicsThree studies12,16,17 assessed the association between medical/obstetric risk factors and the initiation of care. Kupek et al.16 found that women who initiated care late were more often primiparous with at least one risk factor in their medical or obstetrical history. In contrast, Perloff and Jaffee12 did not find an association between entering care after six months gestation and medical risk factors, that is, having at least one medical condition that leads to pregnancy-related medical risks. Petrou et al.17 showed that when a high-risk status arose during the prenatal care period the number of prenatal visits increased.

Five studies10-14 reported on the relationship between parity and prenatal healthcare utilization. Perloff and Jaffee12 found that women with three or more live births were more likely to enter care late – after 6 months – or not at all. Hemminki and Gissler13 concluded that multiparous (≥ 3 previous births) had fewer visits than other women. Heaman et al.14 showed that higher parity leads to inadequate use of prenatal healthcare, according to the GINDEX. Braveman et al.11 found the same, with higher risks of initiating care after the third month, of having too few visits (APNCU), and of receiving no prenatal care at all. Ayoola et al.10 showed that women with no prior births compared to other women initiated prenatal care earlier (before 12 weeks gestation) and were more likely to receive more than 11 prenatal visits than other women.

Ayoola et al.10 were the only ones that reported on the relationship between prior birth outcomes and prenatal care initiation, showing that women with a previous premature birth were more likely to initiate care before 12 weeks gestation.

Finally, Ayoola et al.10 found that early pregnancy recognition (before 6 weeks gestation) led to earlier prenatal-care initiation and to higher odds of receiving more than 15 prenatal visits.

Contextual predisposing variablesTwo studies12,14 assessed contextual predisposing variables. Perloff and Jaffee12 assessed economic opportunity structure, defined at zip-code level as distressed if 60% or more of the population was non-white and 30% or more had incomes below the poverty line. They found that residence in a distressed area increased the risk of late initiation of prenatal care (after 6 months gestation).

Heaman et al.14 defined four contextual predisposing variables. They found more inadequate prenatal care among women living in neighbourhoods with medium and high rates of unemployment, with high rates of single parent families, with medium and high rates of women reporting Canadian Aboriginal status, and with medium and high rates of low-educated residents (<9 years of education).


| 35


2

Contextual enabling/disabling variablesTwo studies12,14 reported on the relation between contextual enabling/disabling variables and prenatal healthcare utilization. Perloff and Jaffee12 showed that living in a neighbourhood with few office-based primary care physicians increased the likelihood of beginning prenatal care late.

Heaman et al.14 found that women living in areas with medium average family incomes more often had inadequate prenatal care use.

Health behaviourHealth behaviour was measured in four studies.13,14,16,17 Heaman et al.14 showed more frequent inadequate prenatal care among women living in neighbourhoods with medium and high rates of women who smoked during pregnancy. Kupek et al.16 reported that smokers were at higher risk for initiating prenatal care after 10 weeks of gestation and after 18 weeks of gestation. Petrou et al.17 showed that smokers were more likely to have fewer prenatal visits as compared to non-smokers. Finally, Hemminki and Gissler13 found that women who smoked during pregnancy had fewer prenatal visits than non-smokers.

Findings aggregated by similar outcomesAs shown in tables 2 and 3 only two studies used identically defined outcomes and determinants. Initiation of care, no prenatal care utilization, and adequacy of care were identically measured by Marín et al.15 and Braveman et al.11 Still, the only identical determinant in these two studies was health insurance status, where both studies found that being uninsured made late initiation of care, receiving no prenatal care, and receiving inadequate care more likely, as compared to having private insurance coverage.

R1R2R3R4R5R6R7R8R9


36 |

Chapter 2

Table 3. Determinants stratified according to the behavioural model of AndersenStudy Individual

predisposing variables

Individual enabling/disabling variables

Individual need variables

Contextual predisposing variables

Contextual enabling/disabling variables

Health behaviours

Ayoola et al.10 Age Marital statusEducationEthnicity

Health insurance coveragePublic assistance program

ParityPrior birth outcomesTime of pregnancy recognition

Marín et al.15 Health insurance coverage

Heaman et al.14 Age (not corrected for confounders)

Parity EmploymentFamily structurePopulation compositionEducation

Family income

Smoking

Kupek et al.16 AgeEthnicity

Hospital typePlanned pattern of prenatal carePlanned place of delivery

Obstetric risk factors

Smoking

Petrou et al.17 AgeEthnicity

Hospital typePlanned pattern of prenatal care

Obstetric risk factorsChange in pattern of prenatal care

Smoking

Perloff and Jaffee12

AgeEducationEthnicityMarital status

Health insurance coverage

Parity Medical risk factors

Economic opportunity structure

Healthcare opportunity structure

Hemminki and Gissler13

AgeMarital statusEducation

Parity Smoking

Braveman et al.11

AgeEducationMarital statusEthnicityBirth place

Health insurance coverage

Parity

DISCUSSION

This study assessed the evidence on determinants of prenatal healthcare utilization. The results show that the following variables were independently associated with late initiation or inadequate use of prenatal care: smoking, low maternal age, low educational level, non-marital status, ethnic minority, planned pattern of prenatal care, hospital type, the planned place of delivery, uninsured status, high parity, prior premature birth, obstetric risk factors, late recognition of pregnancy, and living in deprived neighbourhoods.


| 37


2

Determinants of inadequate use of prenatal healthcare mostly apply to general care, but some additional pregnancy-specific determinants were found. These were late recognition of the pregnancy and high parity. Moreover, the effects of some ‘regular’ determinants such as socioeconomic status may be altered by pregnancy related issues. Further research, quantitative and qualitative, is needed to disentangle the impact of these pregnancy-specific factors on use of prenatal care.

Our findings mostly confirm those of Goldenberg et al.7, but with more recent data of better quality. Similar to that review, we found age, parity, educational level, marital status, and ethnicity to be related to inadequate prenatal care utilization. In addition, Goldenberg et al.7 also presented findings on other variables (psychosocial variables, e.g., feelings about pregnancy, family relations) that were not assessed in the studies that we included. A likely explanation is our exclusion of lower quality studies that, for example, assessed determinants such as wantedness and timing of the pregnancy, and the mother’s belief in the necessity of prenatal care. Our findings also confirm the review of Rowe and Garcia4 on socio-demographic determinants in the UK, but now in a study on all high-income countries that also comprised other determinants.

Interestingly, all the strong evidence comes from only four countries, which encompass only some of the available prenatal healthcare arrangements, both regarding first care giver and reimbursement system. It is very likely that these characteristics modify the effects of the other determinants of prenatal healthcare utilization. To properly assess the effects of system-specific factors comparative research is needed on several countries with varying systems.

Finally, next to frequency, our attention also needs to turn to the content and quality of prenatal services and to the individual, socio-demographic, financial and other factors associated with their access and utilization.

Methodological issues of the included studiesAlthough all included studies assessed prenatal healthcare utilization, they employed 12 different definitions. Similar variations were found regarding determinants that were assessed, resulting in only two studies employing identically defined determinants and outcomes.11,15 Standardization is highly needed to be able to integrate findings.

Only eight out of 41 included studies had a strong internal validity. These eight studies employed retrospective data collection, mostly from birth certificates. This may explain why evidence is lacking on other potential determinants of prenatal care utilization, such as psychosocial variables. Moreover, only one study12 used a theoretical framework to explain the determinants of prenatal healthcare utilization. Using a theoretical framework can help to overcome deficiencies of current research about prenatal healthcare utilization. Finally, all included studies adjusted for confounders, but for a widely varying range.

R1R2R3R4R5R6R7R8R9


38 |

Chapter 2

Strengths and weaknesses of the studyA strength of our study was the use of a comprehensive search strategy by using broad search terms in order not to miss any possible relevant study. Also, we did not restricted to studies published in English. However, we did not review grey literature and did not explore bibliographies, so we may have missed relevant studies.

Conclusion and implicationsOverall, our review shows that the evidence on the determinants of prenatal care utilization is limited, but it mostly confirms the results of the earlier syntheses regarding prenatal healthcare utilization. However, comprehensive data on the determinants of prenatal healthcare utilization are lacking. A means to obtain these is the routine inclusion of possible theory-driven determinants in databases on prenatal healthcare.

We obtained findings on factors that are associated with poor use of prenatal care, but not on the mechanisms that cause these associations. Additional research is needed to disentangle these mechanisms as a basis for interventions targeting at improved use of prenatal care.

We found rather large inequities in prenatal healthcare utilization, which highlights the importance of carefully tailoring interventions, such as home visiting programs, general to the needs of deprived pregnant women. Efforts need to be expanded to ensure adequate prenatal care for those who are at risk of receiving inadequate prenatal care.


| 39


2

REFERENCES

(1) Wildman K, Blondel B, Nijhuis J, Defoort P, Bakoula C. European indicators of health care during pregnancy, delivery and the postpartum period. Eur J Obstet Gynecol Reprod Biol 2003 Nov 28;111 Suppl 1:S53-65.

(2) Gissler M, Merilainen J, Vuori E, Hemminki E. Register based monitoring shows decreasing socioeconomic differences in Finnish perinatal health. J Epidemiol Community Health 2003 Jun;57(6):433-439.

(3) Backe B. Overutilization of antenatal care in Norway. Scand J Public Health 2001 Jun;29(2):129-132.

(4) Rowe RE, Garcia J. Social class, ethnicity and attendance for antenatal care in the United Kingdom: a systematic review. J Public Health Med 2003 Jun;25(2):113-119.


(6) Andersen RM, Rice TH, Kominski GF. Changing the U.S. health care system; key issues in health services policy and management. San Francisco, CA: Jossey-Bass; 2007.

(7) Goldenberg RL, Patterson ET, Freese MP. Maternal demographic, situational and psychosocial factors and their relationship to enrollment in prenatal care: A review of the literature. Women Health 1992;19(2-3):133-151.

(8) Centre for reviews and dissemination. Systematic reviews: CRD’s guidance for undertaking reviews in health care. 2009.

(9) Gyorkos TW, Tannenbaum TN, Abrahamowicz M, Oxman AD, Scott EA, Millson ME, et al. An approach to the development of practice guidelines for community health interventions. Can J Public Health 1994 Jul-Aug;85 Suppl 1:S8-13.

(10) Ayoola AB, Nettleman MD, Stommel M, Canady RB. Time of pregnancy recognition and prenatal care use: a population-based study in the United States. Birth 2010 Mar;37(1):37-43.

(11) Braveman P, Bennett T, Lewis C, Egerter S, Showstack J. Access to prenatal care following major Medicaid eligibility expansions. JAMA 1993 03/10;269(0098-7484; 10):1285-1289.

(12) Perloff JD, Jaffee KD. Late entry into prenatal care: the neighborhood context. Soc Work 1999 Mar;44(2):116-128.

(13) Hemminki E, Gissler M. Quantity and targetting of antenatal care in Finland. Acta Obstet Gynecol Scand 1993;72(1):24-30.

(14) Heaman MI, Green CG, Newburn-Cook CV, Elliott LJ, Helewa ME. Social inequalities in use of prenatal care in Manitoba. J Obstet Gynaecol Can 2007 Oct;29(10):806-816.

(15) Marin HA, Ramirez R, Wise PH, Pena M, Sanchez Y, Torres R. The effect of medicaid managed care on prenatal care: The case of Puerto Rico. Matern Child Health J 2009 2009/;13(2):187-197.

(16) Kupek E, Petrou S, Vause S, Maresh M. Clinical, provider and sociodemographic predictors of late initiation of antenatal care in England and Wales. BJOG 2002 Mar;109(3):265-273.

(17) Petrou S, Kupek E, Vause S, Maresh M. Clinical, provider and sociodemographic determinants of the number of antenatal visits in England and Wales. Soc Sci Med 2001 Apr;52(7):1123-1134.

CHAPTER 3

Determinants of prenatal healthcare utilization by low-risk

women in primary midwifery-led care in the Netherlands:

a prospective cohort study

Esther I Feijen-de Jong, Danielle EMC Jansen, Frank Baarveld, Agatha Boerleider, Evelien Spelten, Francois Schellevis, Sijmen A Reijneveld

Women and Birth 2015, E-pub February 12

R1R2R3R4R5R6R7R8R9


42 |

Chapter 3

ABSTRACT

Background: Prenatal healthcare is pivotal in providing adequate prevention and care to pregnant women. Information about prenatal healthcare use of low-risk pregnant women is lacking. Therefore, we examined the determinants of inadequate prenatal healthcare utilization by low-risk women in primary midwifery-led care in the Netherlands.

Methods: We used longitudinal data from the population-based DELIVER study with twenty midwifery practices across the Netherlands in 2009 and 2010 as the experimental setting. The participants were 3,070 pregnant women starting pregnancy care in primary midwifery care.

Findings: We collected patient-reported data on potential determinants of prenatal care utilization derived from the Andersen model. Prenatal healthcare utilization was measured by a revised version of the Kotelchuck Index, which measures a combination of care entry and numbers of visits. A prevalence of 24.7% inadequate use was found, and 24.7% of the included women were referred to secondary care. Overall, our results showed that women of a non-Western origin (compared to native Dutch women), unemployed women, women reporting chronic illnesses or disabilities, and women who did not use folic acid periconceptionally had higher odds of using inadequate healthcare utilization. Low-risk pregnant women (not referred during pregnancy) were more likely to use prenatal care inadequately if they intended to deliver at a hospital, if they did not use folic acid periconceptionally, or if they were exposed to cigarette smoke during pregnancy. Among those who were referred to secondary care, women reporting chronic illnesses or disabilities, and women who did not use folic acid periconceptionally were more likely to make inadequate use of prenatal care.

Conclusion:Inadequate prenatal healthcare use in primary midwifery care is more likely in specific groups, and the risk groups differ when women are referred to secondary care. The findings suggest routes that can target interventions to women who are at risk of not adequately using prenatal prevention and care services.

KeywordsPregnancy, Healthcare utilization, Public health, Obstetrics, Midwifery


| 43

Determinants of prenatal healthcare utilization by low-risk women in primary midwifery-led care in the Netherlands: a prospective cohort study

3

INTRODUCTION

Adequate use of prenatal healthcare is essential for mother and baby in reducing morbidity and mortality rates.1-3 A timely start (during the first trimester) and an adequate number (according to professional guidelines) of prenatal care visits are important determinants of maternal and foetal health.2, 4, 5 Availability of prenatal care may strongly influence adequate use. However, even in the case of optimal availability, some women do not make adequate use of prenatal care.6

Several determinants of inadequate prenatal healthcare utilization have been identified. Smoking, low maternal age, low educational level, non-marital status, ethnic minority status, planned pattern of prenatal care, hospital type, planned place of delivery, uninsured status, high parity, prior premature birth, obstetric risk factors, late recognition of pregnancy, and living in deprived neighbourhoods are all associated with inadequate healthcare utilization.6 However, most studies identifying determinants of prenatal healthcare utilization include heterogeneous populations of both low- and high-risk pregnant women.6 Specific evidence on determinants of prenatal healthcare utilization by low-risk women (women who are not known to have any medical or obstetric risk factors before the onset of labour,7) is lacking. This is remarkable as the majority of pregnancies (80-90%) are considered to be low-risk.8, 9

The organization of maternity care in the Netherlands enables the study of low-risk pregnant women and also the assessment of determinants of inadequate healthcare utilization in women who may be at low-risk at the beginning of pregnancy, but become high risk later. Dutch maternity care is organized into primary, secondary and tertiary care. Low-risk women mainly attend midwives and, to a small degree, general practitioners (2-6%).10 Women are defined as low-risk and are referred to secondary care according to an Obstetrics and Midwifery Manual. This Manual aligns provider competencies with the health status of pregnant women. It is developed and revised over decades with input of midwives and gynaecologists.11 High-risk women attend obstetricians and specialized midwives in general hospitals (secondary care), while tertiary care occurs in university hospitals. There is close mutual cooperation between these different strata.10

The aim of this research is to examine determinants of inadequate prenatal healthcare use by low-risk women in primary midwifery-led care in the Netherlands, and to determine whether these differ from those who are referred to prenatal secondary care. Information and knowledge about these determinants can optimize professional guidelines in prenatal care and can support the development of tailored interventions for the groups that make inadequate use of prenatal care.

We used Andersen’s behavioural model of healthcare utilization as a guiding framework to categorize the determinants of healthcare utilization.12 This model suggests that the utilization of healthcare services depends on predisposing, enabling, need and health behaviour factors.12

R1R2R3R4R5R6R7R8R9


44 |

Chapter 3

SUBJECTS AND METHODS

Data for this analysis were obtained from the DELIVER study (Dutch acronym for ‘data primary care delivery’) conducted by the Department of Midwifery Science of VU University Medical Center, Amsterdam.13 DELIVER was a descriptive study that aimed to provide information about the organization of midwifery care, the accessibility of midwifery care, and the quality of primary midwifery care in the Netherlands.

DELIVER used a two-stage sampling procedure. Firstly, midwifery practices were recruited by using purposive sampling. Subsequently, all clients receiving care in the participating primary midwifery practices at any time in a 12-month study period in 2009-2010 were eligible to participate if they were able to understand Dutch, English, Turkish or Arabic. The participating practices (20 of the 519 midwifery practices in the Netherlands) comprised 110 midwives and a caseload of 8,200 clients per year, representing all regions of the Netherlands. The women included in our study: a. started their prenatal care in a primary care midwifery practice at the beginning of their pregnancy, b. filled in the first questionnaire in the DELIVER study, and c. the data from their questionnaire could be linked to the electronic client data and the Netherlands Perinatal Registry data (Figure 1).

Women receiving care in primary midwiferypractices (n=14,640)

Deliver participants:7,907 women in primary midwifery-led care

Women included in our sample (n=3,070)

Non-responders (n=6,733), information for 912 women regarding reasons of non-respons: miscarriage (25.9%), not interested (15.8%), language barrier (11.7%)

Women excluded with reasons (n= 4,837);Did not fill in questionnaire 1 (n=1,841)No data known from client records (due to linkingproblems) (n=1,997)Did not start care inmidwifery practice (n=536)Receiving adequate plus care (n=463)

Elig

ible

popu

latio

nD

ELIV

ERco

hort

Star

ted

in p

rim

ary

mid

wife

ry-le

d c

are

Figure 1. Eligible population, DELIVER cohort and study population


| 45


3

Clients participating in the DELIVER study completed up to three questionnaires, depending on the date of their first appointment in midwifery practice during the study period. The first questionnaire was administered before 34 weeks of gestation, the second between 34 weeks of gestation and birth, and the third six weeks postpartum. In addition, data was collected about the care provided by midwives by extracting data from electronic client records of participating clients and from the Netherlands Perinatal Registry, including obligatory reporting of a standardized set of data. Unique anonymous client identifiers and anonymous midwifery practice identifiers linked the three data sources.13 This study used data from the first questionnaire (before 34 weeks of gestation), the electronic client records, and data from the Netherlands Perinatal Registry. The Medical Ethics Committee of VU University Medical Center, Amsterdam approved the study protocol of the DELIVER study, including written informed consent.

Prenatal healthcare utilization was measured by using the Kotelchuck Index, which is widely used in the US.14 We constructed a revised index, modified according to the guidelines of the Royal Dutch Organization of Midwives, concerning the number of prenatal visits during pregnancy. In the Netherlands a relatively high number of prenatal visits is advised, starting care before the 10th week of pregnancy leading to an average of 14 visits at 40 weeks of gestation, compared to e.g. the NICE guideline15 in which 10 appointments for nulliparous women and 7 for parous women is advised for women who are healthy and whose pregnancies remain uncomplicated in the prenatal period. The adjusted Kotelchuck index combines the timing of initial prenatal healthcare and the number of prenatal healthcare visits. Prenatal visits were defined as face-to-face contact with a midwife in primary care. Late initiation of care was defined as a first visit after 12 weeks of gestation (defined by the midwife on the basis of ultrasound examination or the first day of the last menstrual period). The number of visits was calculated on the basis of the electronic client records, which were kept by midwives. More than one visit a day at the same place was counted as one visit. The gestational age at referral was determined for women who were referred to secondary care. Four categories of prenatal healthcare utilization were defined: Inadequate Care, Intermediate Care, Adequate Care and Adequate Plus (Table 1). These categories were dichotomized into Adequate (adequate) and Inadequate (inadequate and intermediate) Care. The Adequate Plus group was excluded because it was not relevant in view of the aim of this study.

Referrals to secondary care were identified using data from the Netherlands Perinatal Registry. Women with no referrals during pregnancy were classified as non-referred, while women with a referral during pregnancy were classified as referred. Referrals during labour were classified as non-referred. We assumed that these referrals were not associated with prenatal healthcare utilization.

R1R2R3R4R5R6R7R8R9


46 |

Chapter 3

Table 1. Index for assessment of the adequacy of prenatal care utilization in the Dutch primary midwifery care context (by A.W. Boerleider and E.I. Feijen-de Jong)Duration of gestation Initiation of care Number of visits Kotelchuck Index2

0 - 11+6

≤ 11+6 ≥ 3 41-2 3

20 1

12 - 26+6 ≤ 11+6 ≥ 6 4Ideally 3.75 visits1 3-5 3

2 2≤ 1 1

≥ 12+0 127 - 36+6 ≥ 10 4Ideally 7.5 visits1 6-9 3

4-5 2≤ 3 1

≥ 12+0 137+0 - 37+6Ideally 11 visits1

≤ 11+6

≥ 12+0

≥ 1310-12

6-9≤5

43211

38+0 - 38+6Ideally 12 visits1

≤ 11+6

≥ 12+0

≥ 1410-13

6-9≤5

43211


≤ 11+6

≥ 12+0

≥ 1511-14

7-10≤ 6

43211


≤ 11+6

≥ 12+0

≥ 1612-15

7-11≤ 6

43211


≤ 11+6

≥ 12+0

≥ 1712-16

8-11≤ 7

43211

1According to the guidelines of the Royal Dutch Organization of Midwives2 Kotelchuck Index:

1. Inadequate (received less than 50% of expected visits)2. Intermediate (50%-79%)3. Adequate (80%-109%)4. Adequate Plus (110% and more)


| 47


3

Possible determinants of healthcare utilization concerned predisposing, enabling, need and health behaviour variables. Data on determinants were obtained from the questionnaire. Several variables, based on Andersen’s model, were considered to be potential determinants of healthcare utilization. Operationalizations of the independent variables are shown in Figure 2. Predisposing variables encompassed socio-demographic and belief factors. Enabling variables included finance (healthcare insurance) and organization (accessibility of care) variables. Regarding health insurance, we distinguished basic and supplementary healthcare insurance. In the Netherlands every inhabitant has obligatory basic health insurance, however, reimbursement of the costs for midwifery-led hospital births requires supplementary insurance. Need variables comprised the health status (perceived and evaluated) of the client. The descriptive component of EuroQol was used to measure self-reported health status. 16 This section asks the respondent to consider and rate her health today. Health was classified on five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Health state values range from most severe impairment on all five dimensions (value = -0.109) to no problems on any dimension (value = 1.0). We converted profiles of health status into a single summary index by applying scores from a valuation set from the UK (York).16 We compared the first quartile (not in control of health status) of our study population with the other quartiles (in control of health status). The locus of control was measured by a single question about the extent of the perceived possibility of influencing lifestyle and/or health behaviour. Feelings towards pregnancy were measured by using the Pregnancy Related Anxiety Questionnaire (PRAQ).17 The scales used were ‘fear of giving birth’ (two items), ‘fear of bearing a handicapped child’ (four items) and ‘concern about one’s appearance’ (three items). Items were scored on a four-point scale (4 = very true, 3 = true, 2 = not true, 1 = certainly not true). Every scale was dichotomized based on the distribution of the median score. In addition to the variable parity, we created a variable that included the difference between the number of pregnancies and the number of deliveries. We assumed that there could be a difference in prenatal healthcare utilization between women with miscarriage(s) and/or abortion(s) in their medical history. Health behaviour variables consisted of questions related to substance use, folic acid use and Body Mass Index (BMI). We removed drug abuse because none of the pregnant women reported drug use, which concurs with our sampling of low-risk pregnancies. Folic acid use was labelled as adequate when started at least four weeks before pregnancy.18 BMI was calculated using the weight and height before pregnancy registered by the respondent. We classified BMI according to the World Health Organization classification of adult underweight, normal weight, overweight and obesity.

R1R2R3R4R5R6R7R8R9


48 |

Chapter 3

Predisposing variablesDemographic:Age: ≤20/21-35/≥36Social:Ethnicity: Native Dutch/Western Non-Dutch/Non-WesternMarital status: married or living together/living aloneOccupation: employed/unemployed/disabledEducational level: low (one to secondary education) middle (to pre-university education) high (bachelor’s degree to post-graduate degree)

Beliefs:Intended place of delivery: home/hospital birth centre with own midwife/hospital consultant-led Religion: yes/no

Need variablesPerceived healthGeneral self-rated health: excellent-very good/good/fair-bad Control of health situation(EQ-5d): no control of health situation/in control of health situation Locus of control: yes/noChronic illnesses, disabilities or disorders: yes/noFeelings towards pregnancy (PRAQ-R):fear/no fearPlanned/wanted pregnancy: planned and wanted/unplanned but wanted/unplanned and unwantedEvaluated healthParitya: nullipara/primipara, multiparaGravidity/parity differencea: difference=1/difference≥2

Enabling variablesFinancing:Health insurance: basic/supplementaryNet household income: low income (≤ €2000)/high income (> €2000)

Organization:Accessibility of care: * difficulties getting through when calling during/outside business hours (yes/no)* difficulties getting to and from the midwifery practice (yes/no)

Health behavioursFolic acid use: adequate/inadequate/noBMI: underweight/normal weight/ overweight/ obesitySmoking: yes/noPassive smoking: yes/noAlcohol use: yes/no

Healthcare utilization

a Electronic client record

Figure 2: Conceptual framework; the behavioural model of Andersen, which shows the possible determinants of healthcare utilization

Statistical analysesFirstly, we described background characteristics of the study population. Secondly, we assessed the determinants of inadequate healthcare utilization by performing binary logistic regressions for the total group. We then split the research population into non-referred and referred pregnant women. Binary logistic regression was performed for both groups. Women receiving adequate healthcare utilization were our reference group. The


| 49


3

hierarchical structure of the data (respondents clustered in midwifery practices) revealed correlations within midwifery practices. Therefore, all analyses were undertaken in a multilevel framework. The variables in the final model are presented as odds ratios (OR) with 95% confidence intervals (CI). A two-tailed p-value of 0.05 or lower was considered statistically significant. Missing data accounted for less than 2.4% for all variables, with the exception of BMI (6.8%). SPSS 21.0 (SPSS Inc., Chicago, IL) was used for all analyses.

RESULTS

Table 2 shows the characteristics of the study population, prenatal healthcare utilization and referrals. The majority of the pregnant women were between 21-35 years of age (84.5%), native Dutch (84.8%), married (96.5%), employed (79.8%) and highly educated (48.8%). Of all 3,070 pregnant women, 24.7% made inadequate use of prenatal care, and 24.7% were referred to secondary care during pregnancy. A small percentage of women (4.7%) made inadequate use of prenatal care and were also referred.

Table 3 shows the results of the analyses assessing the determinants of prenatal healthcare utilization. Overall, it revealed that women of a non-Western origin (compared to native Dutch women), unemployed women, women reporting chronic illnesses or disabilities, and women who did not use folic acid periconceptionally had higher odds of using inadequate healthcare utilization. No enabling variables showed a significant association with inadequate healthcare utilization.

Split by referral status, among non-referred women, only predisposing and health behaviour variables showed a significant association with inadequate healthcare utilization: women intending to deliver in the hospital under supervision of a midwife had higher odds of making inadequate use of prenatal care. Women who did not use folic acid periconceptionally (compared to adequate use), and women exposed to cigarette smoke during pregnancy (compared to non-exposed women) had higher odds of making inadequate use of prenatal care. Among referred women, need and health behaviour determinants showed significant results. Women reporting to have chronic illnesses or disabilities (compared to women not having chronic illnesses or disabilities) had higher odds of making inadequate use of prenatal care. This also counted for women who did not make adequate use of folic acid periconceptionally.

R1R2R3R4R5R6R7R8R9


50 |

Chapter 3

Tabl

e 2.

Hea

lthca

re u

tiliz

ation

and

refe

rral

s in

pri

mar

y m

idw

ifery

car

e (d

escr

iptiv

es, N

= 3

,070

)

Hea

lthc

are

utiliz

ation

n (%

)A

dequ

ate

Inad

equa

te2,

312

(75.

3)75

8 (2

4.7)

Refe

rral

s n

(%)

Non

-ref

erre

dRe

ferr

ed2,

313

(75.

3)75

7 (2

4.7)

Back

grou

nd c

hara

cter

isti

csn

(%)

Ade

quat

e/N

on-r

efer

red

1,69

9 (5

5.3)

Ade

quat

e/Re

ferr

ed61

3 (2

0.0)

Inad

equa

te/

Non

-ref

erre

d61

4 (2

0.0)

Inad

equa

te/

Refe

rred

144

(4.7

)

Age

(n =

3,0

65)

≤ 20

21-3

5*≥

36

40 (1

.3)

2,59

5 (8

4.5)

430

(14.

0)

22 (0

.7)

1,47

2 (4

8.0)

204

(6.7

)

4 (0

.1)

500

(16.

3)10

7 (3

.5)

10 (0

.3)

509

(16.

6)93

(3.0

)

4 (0

.1)

114

(3.7

)26

(0.8

)Et

hnic

ity (n

= 3

,065

)N

ative

Dut

ch*

Non

-Wes

tern

Wes

tern

Non

-Dut

ch

2,59

9 (8

4.8)

252

(8.2

)21

4 (7

.0)

1,44

3 (4

7.1)

128

(4.2

)12

5 (4

.1)

533

(17.

4)36

(1.2

)43

(1.4

)

507

(16.

5)68

(2.2

)39

(1.3

)

116

(3.8

)20

(0.7

)7

(0.2

) M

arita

l sta

tus

(n =

3,0

67)

Mar

ried

or li

ving

toge

ther

*Li

ving

alo

ne2,

959

(96.

5)10

8 (3

.5)

1,64

7 (5

3.7)

50 (1

.6)

588

(19.

2)25

(0.8

)59

0 (1

9.2)

23 (0

.7)

134

(4.4

)10

(0.3

)

Occ

upati

on (n

= 3

,064

)Em

ploy

ed*

Une

mpl

oyed

Dis

able

d

2,44

6 (7

9.8)

572

(18.

7)46

(1.5

)

1,37

4 (4

4.8)

304

(9.9

)19

(0.6

)

497

(16.

2)10

0 (3

.3)

13 (0

.4)

468

(15.

3)13

2 (4

.3)

13 (0

.4)

107

(3.5

)36

(1.2

)1

(0.0

)Ed

ucati

onal

leve

l (n

= 3,

063)

Low

*M

iddl

eH

igh

472

(15.

4)1,

095

(35.

7)1,

496

(48.

8)

237

(7.7

)60

3 (1

9.7)

856

(27.

9)

98 (3

.2)

238

(7.8

)27

4 (8

.9)

105

(3.4

)20

8 (6

.8)

300

(9.8

)

32 (1

.0)

46 (1

.5)

66 (2

.2)

Pari

ty (n

= 3

,062

)N

ullip

ara*

Prim

i/M

ultip

ara

1,22

7 (4

0.1)

1,83

5 (5

9.9)

706

(23.

1)98

7 (3

2.2)

237

(7.7

)37

4 (1

2.2)

222

(7.3

)39

2 (1

2.8)

62 (2

.0)

82 (2

.7)

* =

Refe

renc

e ca

tego

ries


| 51


3

Tabl

e 3.

Ass

ocia

tions

of v

ario

us c

hara

cter

istic

s w

ith in

adeq

uate

pre

nata

l hea

lthca

re u

tiliz

ation

for

all w

omen

who

initi

ated

pre

nata

l vis

its a

t pri

mar

y ca

re le

vel,

and

split

by

subs

eque

nt re

ferr

al (n

o an

d ye

s): o

dds

ratio

s (O

R) a

nd 9

5% c

onfid

ence

inte

rval

s (C

I)O

vera

llN

= 3

,070

Not

refe

rred

n =

2,31

3Re

ferr

edn

= 75

7Cr

ude

OR

(C.I.

)A

djus

ted

OR

(C.I.

)aCr

ude

OR

(C.I.

)A

djus

ted

OR

(C.I.

)aCr

ude

OR

(C.I.

)A

djus

ted

OR

(C.I.

)a

Pred

ispo

sing

var

iabl

esA

ge (y

ears

)≤

20

≥ 36

21

-35

2.48

(1.2

2-5.

06)

1.27

(0.9

8-1.

64)

1.00

(ref

.)

1.97

(0.8

6-4.

51)

1.50

(1.1

0-2.

03)

1.00

(ref

.)

6.89

(1.5

6-30

.16)

1.03

(0.6

3-1.

68)

1.00

(ref

.)Et

hnic

ity

Non

-wes

tern

Wes

tern

non

-Dut

chN

ative

Dut

ch

2.05

(1.4

8-2.

85)

0.93

(0.6

4-1.

35)

1.00

(ref

.)

1.56

(1.1

1-2.

21)

0.88

(0.6

1-1.

28)

1.00

(ref

.)

1.83

(1.2

5-2.

68)

0.95

(0.6

2-1.

46)

1.00

(ref

.)

2.43

(1.2

8-4.

60)

0.76

(0.3

3-1.

76)

1.00

(ref

.)M

arit

al s

tatu

sLi

ving

alo

ne

Mar

ried

or li

ving

toge

ther

1.77

(1.1

3-2.

78)

1.00

(ref

.)1.

89 (1

.09-

3.28

)1.

00 (r

ef.)

1.77

(0.8

1-3.

91)

1.00

(ref

.)

Occ

upati

on

Une

mpl

oyed

Dis

able

d Em

ploy

ed

1.67

(1.3

3-2.

10)

1.00

(0.4

9-2.

04)

1.00

(ref

.)

1.36

(1.0

6-1.

73)

0.77

(0.3

7-1.

59)

1.00

(ref

.)

1.56

(1.1

9-2.

03)

1.27

(0.5

5-2.

92)

1.00

(ref

.)

1.96

(1.2

4-3.

11)

0.40

(0.0

5-3.

25)

1.00

(ref

.)Ed

ucati

onal

leve

lM

iddl

eH

igh

Low

0.68

(0.5

2-0.

88)

0.67

(0.5

2-0.

87)

1.00

(ref

.)

0.74

(0.5

4-1.

02)

0.69

(0.5

1-0.

94)

1.00

(ref

.)

0.55

(0.3

3-0.

94)

0.64

(0.3

9-1.

07)

1.00

(ref

.)In

tend

ed p

lace

of d

eliv

ery

Hos

pita

l/bi

rth

cent

re

mid

wife

ry-le

dH

ospi

tal c

onsu

ltant

-led

Hom

e

1.29

(1.0

6-1.

56)

2.14

(1.1

9-3.

85)

1.00

(ref

.)

1.45

(1.1

6-1.

81)

2.05

(0.9

7-4.

30)

1.00

(ref

.)

1.43

(1.1

4-1.

79)

1.71

(0.8

1-3.

60)

1.00

(ref

.)

0.87

(0.5

9-1.

28)

2.63

(1.0

1-6.

88)

1.00

(ref

.)

Relig

ion

No

Yes

0.96

(0.7

8-1.

17)

1.00

(ref

.)0.

97 (0

.77-

1.23

)1.

00 (r

ef.)

0.91

(0.6

0-1.

36)

1.00

(ref

.)

R1R2R3R4R5R6R7R8R9


52 |

Chapter 3

Tabl

e 3.

Con

tinue

dO

vera

llN

ot re

ferr

edRe

ferr

ed

Crud

eO

R (C

.I.)

Adj

uste

dO

R (C

.I.)a

Crud

e O

R (C

.I.)

Adj

uste

dO

R (C

.I.)a

Crud

eO

R (C

.I.)

Adj

uste

d O

R (C

.I.)a

Enab

ling

vari

able

sBa

sic

and

supp

lem

enta

ry

heal

thca

re in

sura

nce

Basi

c an

d su

pple

men

tary

Basi

c0.

87 (0

.67-

1.21

)1.

00 (r

ef.)

0.87

(0.6

4-1.

17)

1.00

(ref

.)0.

88 (0

.53-

1.48

)1.

00 (r

ef.)

Net

hou

seho

ld in

com

e >

€200

0<

€200

00.

63 (0

.49-

0.81

)1.

00 (r

ef.)

0.62

(0.4

6-0.

83)

1.00

(ref

.)0.

64 (0

.39-

1.06

)1.

00 (r

ef.)

Acc

essi

bilit

y of

car

e (p

hone

)Pr

oble

ms

No

prob

lem

s1.

08 (0

.85-

1.37

)1.

00 (r

ef.)

1.10

(0.8

3-1.

44)

1.00

(ref

.)1.

06 (0

.65-

1.73

)1.

00 (r

ef.)

Acc

essi

bilit

y of

car

e (g

etting

to

and

from

the

pra

ctice

)Pr

oble

ms

No

prob

lem

s1.

26 (0

.79-

1.99

)1.

00 (r

ef.)

1.18

(0.6

9-2.

02)

1.00

(ref

.)1.

11 (0

.43-

2.88

)1.

00 (r

ef.)

Nee

d va

riab

les

Gen

eral

sel

f-ra

ted

heal

th

Exce

llent

/Ver

y go

odBa

d/Fa

irG

ood

0.92

(0.7

5-1.

11)

0.82

(0.6

1-1.

10)

1.00

(ref

.)

0.92

(0.7

4-1.

15)

0.78

(0.5

6-1.

12)

1.00

(ref

.)

0.89

(0.6

9-1.

35)

0.87

(0.4

8-1.

60)

1.00

(ref

.)Q

ualit

y of

life

(Eur

oQol

) - +

1.02

(0.8

2-1.

26)

1.00

(ref

.)1.

03 (0

.80-

1.34

)1.

00 (r

ef.)

1.12

(0.7

3-1.

71)

1.00

(ref

.)

Locu

s of

con

trol

No

Yes

1.31

(1.0

3-1.

68)

1.00

(ref

.)1.

26 (0

.94-

1.67

)1.

00 (r

ef.)

1.40

(0.8

4-2.

31)

1.00

(ref

.)


| 53


3

Ove

rall

Not

refe

rred

Refe

rred

Crud

eO

R (C

.I.)

Adj

uste

dO

R (C

.I.)a

Crud

e O

R (C

.I.)

Adj

uste

dO

R (C

.I.)a

Crud

eO

R (C

.I.)

Adj

uste

d O

R (C

.I.)a

Chro

nic

illne

sses

or d

isab

iliti

esYe

sN

o1.

39 (1

.04-

1.86

)1.

00 (r

ef.)

1.41

(1.0

5-1.

90)

1.00

(ref

.)1.

26 (0

.88-

1.82

)1.

00 (r

ef.)

2.03

(1.2

4-3.

33)

1.00

(ref

.)2.

09 (1

.27-

3.44

) 1.0

0 (r

ef.)

PRA

Q*-

Child

Fear

No

fear

0.84

(0.6

9-1.

02)

1.00

(ref

.)0.

89 (0

.71-

1.11

)1.

00 (r

ef.)

0.77

(0.5

2-1.

14)

1.00

(ref

.)

PRA

Q*-

Del

iver

yFe

arN

o fe

ar1.

00 (0

.82-

1.21

)1.

00 (r

ef.)

0.83

(0.6

6-1.

05)

1.00

(ref

.)1.

64 (1

.12-

2.40

)1.

00 (r

ef.)

PRA

Q*-

Body

Fear

No

fear

0.93

(0.7

6-1.

13)

1.00

(ref

.)0.

94 (0

.74-

1.18

)1.

00 (r

ef.)

1.02

(0.6

9-1.

51)

1.00

(ref

.)

Plan

ned

and

wan

tedn

ess

of

preg

nanc

y**

Wan

ted,

not

pla

nned

Plan

ned

and

wan

ted

1.42

(1.1

3-1.

79)

1.00

(ref

.)1.

34 (1

.02-

1.76

)1.

00 (r

ef.)

1.64

(1.0

5-2.

58)

1.00

(ref

.)Pa

rity

Pr

imi/

mul

tipar

aN

ullip

ara

1.22

(1.0

2-1.

47)

1.00

(ref

.)1.

36 (1

.10-

1.69

)1.

00 (r

ef.)

0.84

(0.5

8-1.

23)

1.00

(ref

.)

Diff

eren

ce b

etw

een

num

ber

of p

regn

anci

es a

nd n

umbe

r of

bir

ths

≥ 2

11.

17 (0

.95-

1.44

)1.

00 (r

ef.)

1.17

(0.9

2-1.

49)

1.00

(ref

.)1.

22 (0

.81-

1.85

)1.

00 (r

ef.)

R1R2R3R4R5R6R7R8R9


54 |

Chapter 3

Tabl

e 3.

Con

tinue

dO

vera

llN

ot re

ferr

edRe

ferr

edCr

ude

OR

(C.I.

)A

djus

ted

OR

(C.I.

)aCr

ude

OR

(C.I.

)A

djus

ted

OR

(C.I.

)aCr

ude

OR

(C.I.

)A

djus

ted

OR

(C.I.

)a

Hea

lth

beha

viou

r va

riab

les

Folic

aci

d uti

lizati

on

No

Yes,

inad

equa

tely

Yes,

ade

quat

ely

2.42

(1.7

8-3.

29)

1.26

(1.0

3-1.

53)

1.00

(ref

.)

1.93

(1.4

0-2.

68)

1.20

(0.9

9-1.

47)

1.00

(ref

.)

2.20

(1.5

3-3.

16)

1.20

(0.9

5-1.

50)

1.00

(ref

.)

1.89

(1.3

0-2.

75)

1.18

(0.9

4-1.

48)

1.00

(ref

.)

3.06

(1.7

0-5.

50)

1.38

(0.9

0-2.

09)

1.00

(ref

.)

3.14

(1.7

4-5.

67)

1.34

(0.8

8-2.

05)

1.00

(ref

.)BM

I ≤

18.5

25-<

30

≥ 30

18.5

-< 2

5

1.02

(0.6

0-1.

76)

0.95

(0.7

5-1.

19)

1.14

(0.7

9-1.

65)

1.00

(ref

.)

1.31

(0.7

2-2.

40)

1.06

(0.8

1-1.

39)

1.22

(0.7

7-1.

96)

1.00

(ref

.)

0.42

(0.0

9-1.

86)

0.75

(0.4

6-1.

22)

1.16

(0.6

3-2.

15)

1.00

(ref

.)

Smok

ing

Yes

No

1.11

(0.8

1-1.

52)

1.00

(ref

.)1.

19 (0

.83-

1.71

)1.

00 (r

ef.)

0.87

(0.4

2-1.

78)

1.00

(ref

.)

Pass

ive

smok

ing

Yes

No

1.53

(1.1

1-2.

09)

1.00

(ref

.)1.

70 (1

.18-

2.45

)1.

00 (r

ef.)

1.56

(1.0

7-2.

28)

1.00

(ref

.)0.

94 (0

.47-

1.87

)1.

00 (r

ef.)

Alc

ohol

use

Yes

No

1.30

(0.9

8-1.

72)

1.00

(ref

.)1.

31 (0

.94-

1.82

)1.

00 (r

ef.)

1.14

(0.6

4-2.

05)

1.00

(ref

.)

a =

corr

ecte

d fo

r al

l oth

er v

aria

bles

*PRA

Q =

Pre

gnan

cy R

elat

ed A

nxie

ty Q

uesti

onna

ire

**ca

tego

ry ‘n

ot w

ante

d, n

ot p

lann

ed’ r

emov

ed d

ue to

em

pty

cells


| 55


3

DISCUSSION

We assessed the determinants of inadequate prenatal healthcare utilization of referred and non-referred women in primary midwifery practices in the Netherlands. Low-risk pregnant women (who were not referred during pregnancy) were found to be more likely to inadequately use prenatal care if they intended to deliver at hospital under the supervision of a midwife, if they did not use folic acid periconceptionally, or if they were exposed to cigarette smoke during pregnancy. Among women who were referred to secondary care during pregnancy, those who reported chronic illnesses or disabilities, and those who did not use folic acid periconceptionally were more likely to make inadequate use of prenatal care than the remaining women.

A strength of this research is the use of a unique sample of women who were low-risk at the start of their pregnancy. Next to this, we made a distinction between pregnant women who were referred and not-referred during pregnancy in order to delineate two groups; a consistently low-risk group, and a group needing specialized care during pregnancy. This allowed us to carry out the study in a homogeneous low-risk population in primary midwifery care. In the Netherlands, 83% of women start pregnancy without any problems.19 Therefore, our study represents a large majority of all pregnant women. Finally, we used a database with many variables in a large study population. Therefore, we could include all of the dimensions of Anderson’s model, enabling us to structure the large number of variables included in our database.

In the Netherlands, a low-risk pregnancy is defined on medical and obstetric criteria, although also other criteria (e.g. social factors as domestic violence, lack of facilities for existing children) are known to influence pregnancy risks.20 Currently, pilot studies are being undertaken to test the feasibility of including non-medical/obstetric criteria in the definition of low-risk pregnancy. Unfortunately, not all known determinants of inadequate prenatal care use could be included in our analyses due to absence of information about these factors. Regarding demographic characteristics, our study population included slightly more highly educated and native Dutch women compared to all pregnant women.19 This probably led to an underestimation of inadequate users. However, we found a similar percentage of women referred during pregnancy as is reported in national statistics.19 This supports the representativeness of our study. Finally, we did not measure the content of prenatal care, which is also an important indicator of adequacy of care.21

We found that 24.7% of the women who started care in primary midwifery care used prenatal care inadequately. This percentage would have been lower using for instance the prenatal schedule of the NICE guideline.15 Comparing our figures to a guideline from another country would be incorrect as professionals may generally be expected to meet their own professionals standard. It would, however, be very useful to compare guidelines on prenatal care between countries, including their evidence base.

R1R2R3R4R5R6R7R8R9


56 |

Chapter 3

The quite high percentage of underuse of prenatal care raises the question what the underlying mechanisms are for this inadequate use. Client-related and provider related factors may contribute to this underuse of prenatal care. With regard to client-related factors Dutch research is lacking. However, internationally Phillippi20 reported barriers which may also exist in the Netherlands, such as lack of familiarity with the maternal healthcare system (for primigravidae), language problems, and not knowing being pregnant. Provider-related factors may also contribute to underuse of prenatal care. After all, midwives inform pregnant women about prenatal care programme and invite them for follow-up consultations. Maybe, our finding reflects the way midwives adjust the number of visits to the preferences and wishes of pregnant women. Also, midwives themselves may experience barriers, which lead to less prenatal visits. However, to our knowledge evidence about provider-related factors in relation to adequacy of prenatal care use is not available.

We found similar determinants of inadequate prenatal healthcare utilization as have been previously reported.6 However, regarding women who begin with primary midwifery care, who are all at low-risk, we found a limited set of determinants associated with inadequate healthcare utilization. This is probably due to the group having consistently low-risk pregnancies. In a more heterogeneous population – including both women with and without health risks during pregnancy – obstetric risk factors can confound or modify the role of determinants, which may lead to a larger set of determinants, which in fact denote subgroups of pregnant women with varying risks. Another explanation might be that because we used an adapted version of the Kotelchuck Index, the number of determinants and the strength of the associations found could have been affected. The factors that led to inadequate prenatal healthcare utilization among non-referred low-risk pregnant women concerned predisposing variables and health behaviour variables. Regarding predisposing variables, the intended place of delivery showed a significant association with healthcare utilization. Similar to Kupek et al.22, we found that women intending to deliver at a hospital under the supervision of a midwife were more likely to make inadequate use of prenatal care than women intending to deliver at home. This could be due by the characteristics of the population delivering in hospital. In a large study, De Jonge et al.23, found that Dutch women intending to deliver at a hospital were more likely to have a lower socio-economic position. The intended place of delivery may thus act as a proxy for socio-economic position. Regarding health behaviour variables, we found that folic acid utilization (periconceptional behaviour) and passive smoking (prenatal behaviour) were determinants of inadequate healthcare utilization. Both folic acid utilization and passive smoking may act as proxy indicators for a less healthy lifestyle.24, 25 Inadequate healthcare utilization can also be an expression of a less healthy lifestyle or less health consciousness.24 However, we were not able to test this hypothesis, and thus it requires further research. Low-risk pregnant women whose pregnancy became high-risk (referred women) had a change of health status during


| 57


3

pregnancy, leading to referral to secondary care. Determinants associated with inadequate prenatal healthcare utilization among referred women included need and health behaviour determinants. Regarding need factors, women reporting chronic illnesses and disabilities were more likely to make inadequate use of prenatal care. It is probable that consulting other health practitioners may substitute for primary prenatal care in this case, including consulting general practitioners26, obstetricians or other medical specialists. In addition, a significant health behaviour determinant was the failure to use folic acid periconceptionally, probably due to the same reason as that given by low-risk women. While we assessed individual characteristics associated with use of prenatal healthcare services, healthcare utilization is also determined by characteristics of health services themselves, and by the interaction of the individual with the healthcare system and/or healthcare provider. We did not include these factors but acknowledge their importance, as also noted by Andersen et al.12

Our findings have implications for both daily care and future research. We found that many pregnant women visit a midwife less frequently than they should as advised by professional guidelines, or entry care after the 12th week of gestation. Professional organisations have to be aware of this and should evaluate professional guidelines as to whether these are still adequate for Dutch primary midwifery care. The differences between guidelines on prenatal care further call for an international comparative study of these guidelines and their evidence base. Regarding care, our research can help midwives in low-risk settings to be alert on care patrons that possibly can lead to suboptimal outcomes, which deviate from standard professional guidelines. Knowing about the determinants of inadequate HCU may give midwives an indication which women are vulnerable. In addition, a redesign of prenatal care could be considered – especially for women with less healthy lifestyles – which also takes client satisfaction with the services offered into account. For example, CenteringPregnancy is a model of group-based prenatal care that increases the odds of adequate healthcare utilization27 in terms of number of visits. It includes physical assessment, education and peer support.27-29 Notwithstanding, in this model, early initiation of care is necessary and it also requires the training of maternal healthcare providers in other skills.30

Future research is needed to gain an understanding of the reasons why women make inadequate use of prenatal primary care. What are the underlying reasons for certain groups? The content of care should also be included when measuring the adequacy of care. Our study can also provide a theoretical framework for future research integrating quantitative (content) and qualitative (reasons why) approaches to prenatal healthcare utilization in primary care. Finally, next to assessing underuse of prenatal healthcare, assessment of overuse of prenatal care may be relevant as well. It may add information on the adequacy of prenatal guidelines, and on the costs associated with high use.

R1R2R3R4R5R6R7R8R9


58 |

Chapter 3

CONCLUSION

Our findings show that determinants of prenatal healthcare utilization in primary care differ between women who are and women who are not referred to secondary care. Non-referred pregnant women were more likely to make inadequate use of prenatal care if they intended to deliver in hospital under the supervision of a midwife, if they did not use folic acid adequately periconceptionally, or if they were exposed to cigarette smoke during pregnancy. Women who were referred to secondary care were more likely to make inadequate use of prenatal care if reporting a chronic illness or disability, and if not using folic acid periconceptionally. Our results can be used to target interventions to women who are at risk of inadequate prenatal healthcare utilization. At the same time, healthcare providers should be made aware of the groups we have identified because they are also at risk of making inadequate use of care.31

ACKNOWLEDGEMENTS We would like to thank Myrte Westerneng and Monique Pereboom for their assistance in the data collection process for this study and Michiel de Boer for assistance with the data analyses.


| 59


3

REFERENCES

(1) Hemminki E, Gissler M. Quantity and targetting of antenatal care in Finland. Acta Obstet Gynecol Scand. 1993;72(1):24-30.

(2) Wildman K, Blondel B, Nijhuis J, Defoort P, Bakoula C. European indicators of health care during pregnancy, delivery and the postpartum period. Eur J Obstet Gynecol Reprod Biol. 2003 Nov 28;111 Suppl 1:S53-65.

(3) Raatikainen K, Heiskanen N, Heinonen S. Under-attending free antenatal care is associated with adverse pregnancy outcomes. BMC Public Health. 2007 Sep 27;7:268.

(4) LaVeist TA, Keith VM, Gutierrez ML. Black/white differences in prenatal care utilization: An assessment of predisposing and enabling factors. Health Serv Res. 1995;30:43-58.

(5) Dowswell T, Carroli G, Duley L, Gates S, Gulmezoglu AM, Khan-Neelofur D, et al. Alternative versus standard packages of antenatal care for low-risk pregnancy. Cochrane Database Syst Rev. 2010 Oct 6;(10):CD000934.

(6) Feijen-de Jong EI, Jansen DE, Baarveld F, van der Schans CP, Schellevis FG, Reijneveld SA. Determinants of late and/or inadequate use of prenatal healthcare in high-income countries: A systematic review. Eur J Public Health. 2012 Dec;22(6):904-13.

(7) Birthplace in England Collaborative Group, Brocklehurst P, Hardy P, Hollowell J, Linsell L, Macfarlane A, et al. Perinatal and maternal outcomes by planned place of birth for healthy women with low risk pregnancies: The birthplace in England national prospective cohort study. BMJ. 2011 Nov 23;343:d7400.

(8) Netherlands perinatal registry (PRN foundation). Perinatal care in the Netherlands, 2011(9) Hingstman, L., Kenens, R.J. Figures of GPs registration (cijfers uit de registratie van huisartsen;

peiling 2011). Utrecht: NIVEL; 2011.(10) De Geus M. Midwifery in the Netherlands. Utrecht: Royal Dutch Organisation of Midwives;

2012.(11) De Vries R, Nieuwenhuijze M, Buitendijk SE, members of Midwifery Science Work Group. What

does it take to have a strong and independent profession of midwifery? Lessons from the Netherlands. Midwifery. 2013 Oct;29(10):1122-8.


(13) Mannien J, Klomp T, Wiegers T, Pereboom M, Brug J, de Jonge A, et al. Evaluation of primary care midwifery in the netherlands: Design and rationale of a dynamic cohort study (DELIVER). BMC Health Services Research. 2012;12(1):69.

(14) Kotelchuck M. The adequacy of prenatal care utilization index: Its US distribution and association with low birthweight. Am J Public Health. 1994;84(9):1486-9.

(15) National Institute for Health and, Clinical Excellence. Routine care for the healthy pregnant woman. Antenatal care. 2011.

(16) Herdman M, Gudex C, Lloyd A, Janssen M, Kind P, Parkin D, et al. Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Qual Life Res. 2011 Dec;20(10):1727-36.

(17) Huizink AC, Mulder EJ, Robles de Medina PG, Visser GH, Buitelaar JK. Is pregnancy anxiety a distinctive syndrome? Early Hum Dev. 2004 Sep;79(2):81-91.

(18) Health Council of the Netherlands. Towards an optimal use of folic acid. The Hague: Health Council of the Netherlands; 2008. Report No.: 2008/02E.

(19) Netherlands Perinatal Registry. Grote lijnen 10 jaar perinatale registratie Nederland. Utrecht: Netherlands Perinatal Registry; 2011.

R1R2R3R4R5R6R7R8R9


60 |

Chapter 3

(20) Phillippi JC. Women’s perceptions of access to prenatal care in the United States: A literature review. J Midwifery Womens Health. 2009 May-Jun;54(3):219-25.

(21) Misra DP, Guyer B. Benefits and limitations of prenatal care: From counting visits to measuring content. JAMA. 1998 May 27;279(20):1661-2.

(22) Kupek E, Petrou S, Vause S, Maresh M. Clinical, provider and sociodemographic predictors of late initiation of antenatal care in England and Wales. BJOG. 2002 Mar;109(3):265-73.

(23) de Jonge A, Mesman JA, Mannien J, Zwart JJ, van Dillen J, van Roosmalen J. Severe adverse maternal outcomes among low risk women with planned home versus hospital births in the netherlands: Nationwide cohort study. BMJ. 2013 Jun 13;346:f3263.

(24) Ribeiro ER, Guimaraes AM, Bettiol H, Lima DD, Almeida ML, de Souza L, et al. Risk factors for inadequate prenatal care use in the metropolitan area of Aracaju, northeast Brazil. BMC Pregnancy Childbirth. 2009 Jul 22;9:31,2393.

(25) Mannien J, de Jonge A, Cornel MC, Spelten E, Hutton EK. Factors associated with not using folic acid supplements preconceptionally. Public Health Nutr. 2013 Oct 10:1-7.

(26) Feijen-de Jong EI, Baarveld F, Jansen DE, Ursum J, Reijneveld SA, Schellevis FG. Do pregnant women contact their general practitioner? A register-based comparison of healthcare utilisation of pregnant and non-pregnant women in general practice. BMC Fam Pract. 2013 Jan 16;14:10,2296.

(27) Ickovics JR, Kershaw TS, Westdahl C, Magriples U, Massey Z, Reynolds H, et al. Group prenatal care and perinatal outcomes: A randomized controlled trial. Obstet Gynecol. 2007 Aug;110(2 Pt 1):330-9.

(28) Rotundo G. Centering pregnancy. Nursing for Women’s Health. 2011;15(6):508-18.(29) Trudnak TE, Arboleda E, Kirby RS, Perrin K. Outcomes of latina women in CenteringPregnancy

group prenatal care compared with individual prenatal care. J Midwifery Womens Health. 2013 Jul;58(4):396-403.

(30) Novick G, Reid AE, Lewis J, Kershaw TS, Rising SS, Ickovics JR. Group prenatal care: Model fidelity and outcomes. Obstet Gynecol. 2013 8;209(2):112.e1,112.e6.

(31) Kogan MD, Alexander GR, Jack BW, Allen MC. The association between adequacy of prenatal care utilization and subsequent pediatric care utilization in the United States. Pediatrics. 1998 Jul;102(1 Pt 1):25-30.

CHAPTER 4

Prenatal care use in Belgium and the Netherlands:

predisposing, enabling and pregnancy-related determinants

Jana Vanden Broeck, Esther I Feijen-de Jong, Trudy Klomp, Koen Putman, Katrien Beeckman

Submitted

R1R2R3R4R5R6R7R8R9


62 |

Chapter 4

ABSTRACT

Background: Examining determinants of prenatal care (PNC) trajectories is important to stimulate an equitable distribution of PNC across Europe. This study (1) compares PNC utilization in Belgium and the Netherlands and (2) identifies predisposing, enabling and pregnancy-related determinants.

Methods: Secondary data analysis was performed using pooled and matched data from Belgium, and the Netherlands. The content and timing of care during pregnancy (CTP) tool measured PNC use. Non-parametric tests and ordinal logistic regression were performed.

Results: Dutch women received appropriate PNC more often than Belgian women. Lower education, unemployment, lower continuity of care and non-attendance of antenatal classes were associated with a lower likelihood of having more appropriate PNC.

Conclusions: Women in urban Dutch regions used more appropriate PNC than women in the Brussels Metropolitan Region. However, irrespective of the region (Brussels versus urban-Netherlands) or any enabling characteristic, the content and timing of PNC was associated with predisposing and pregnancy-related variables. Lower health literacy in socially vulnerable women might explain the predisposing determinants. Regarding pregnancy-related determinants, improving continuity of care by creating new maternal healthcare models could enhance PNC use.

Keywords: Prenatal care/utilization, Health behaviour, Socio-economic factors, High-income countries, CTP or content of care


| 63

Prenatal care use in Belgium and the Netherlands: predisposing, enabling and pregnancy-related determinants

4

INTRODUCTION

An understanding of the individual determinants of prenatal care (PNC) utilization may assist the pursuit of adequate levels of care recommended for every pregnancy. A minimum level of PNC is important because it enables early and continuing risk assessment, health promotion and medical and psychosocial follow-up.1 Despite its value, some women do not make proper use of PNC.2

According to the Andersen health behavioural model, individual determinants of healthcare utilization can be divided into predisposing, enabling3 and need components.4 With respect to PNC, predisposing determinants refer to individual characteristics which exist prior to the pregnancy and affect the propensity to use care. Previous studies have concluded that low maternal age4-7, being single7, low educational level6-9, lack of a paid job9, foreign ethnic background6,9, or origin2,5,8, poor language proficiency1,7, (little) support from a social network1 and lack of knowledge of the healthcare system1 are associated with inadequate PNC utilization. Enabling determinants refer to conditions, which make PNC available to pregnant women. Uninsured status6,7, planned pattern of PNC6, hospital type at booking6, personal treatment, communication and knowledge of cultural practices of the care provider1 have been found to be associated with inadequate PNC. The need component can be extended to a more specific ‘pregnancy-related’ element of the determinants. Inadequate use of PNC is related to high parity5-7, unplanned pregnancy7, no previous premature birth6, continuity of care8, late recognition of pregnancy6 and behavioural factors such as smoking during pregnancy.6,9

The operationalisation of PNC utilization varies across studies, therefore results must be interpreted cautiously. The initiation of care1,5-7,9, the number of prenatal visits6,7 and several indices based on the timing of initiation of PNC, the total number of prenatal visits and the gestational age at birth2,6-8 have been considered. Since there is no consensus about the number of prenatal visits10, it is preferable to take into account elements of the content and timing of care during the pregnancy. One recent study measured PNC more comprehensively using the content and timing of care during pregnancy (CTP) tool.8

Previously defined determinants of PNC use should be interpreted in relation to the context of these studies. In addition to individual determinants, healthcare utilization depends on resources (e.g. number of care providers available) and the organisation of the national healthcare system, such as the nature of referrals between healthcare providers.3 Feijen-de Jong et al. identified the need for comparative research in several countries with varying prenatal healthcare arrangements as these might explain differences in the effects of individual determinants on PNC use.6 In this study, we compared PNC between cohorts in two different countries (Belgium and the Netherlands). In the Netherlands, most women with uncomplicated pregnancies receive PNC from primary care midwives who act

R1R2R3R4R5R6R7R8R9


64 |

Chapter 4

as gatekeepers to secondary obstetric care.11 They receive fixed remunerations for follow-up during the full length or part of the pregnancy.12 In Belgium, most women access an obstetrician directly for PNC as they do not need preauthorisation to gain access to specialist care.5 The majority of general practitioners, specialists and independent midwives are paid on a fee-for-service basis.13

This study aimed to 1) compare PNC utilization in Belgium and the Netherlands as measured by the CTP tool and 2) to identify its predisposing, enabling and pregnancy-related determinants.

METHODS

Data collection A secondary data analysis was performed using pooled data from two studies. For Belgium, data were obtained from a prospective observational study conducted in the Brussels Metropolitan Region (the CTP study).10 Recruitment occurred between April and July 2008 in nine out of 12 hospital centres for ultrasound to which every woman is referred. All low risk women, at the beginning of their care trajectory (attending a first or second visit or having a gestational age less than 16 weeks) were legible for inclusion. Data collection comprised a questionnaire about personal characteristics and pregnancy history at the moment of recruitment, a diary recording all prenatal visits in a standardised manner and bimonthly telephone follow-up interviews to record PNC use (n=333).10 This study was approved by all participating centers and from the Ethic Committee of the University Hospital UZ Brussel.

For the Netherlands, data were obtained from the DELIVER (Dutch acronym for ‘data primary care delivery’) study. Data were gathered in a 12-month study period in 2009-2010. The Deliver study is a descriptive study that aimed to provide information about midwifery care organization, accessibility of midwifery care, and the quality of primary midwifery care in the Netherlands.14 Data collection with regard to pregnant women recruited in primary care midwifery practices included up to two questionnaires about socio-demographic characteristics and ultrasound scans. One questionnaire was administered before 34 weeks of gestation and the other between 34 weeks of gestation and birth. In addition, information about prenatal care utilization was gathered by extracting data from electronic client records of participating clients. This study was approved by the Medical Ethics Committee of the VU University Medical Center Amsterdam. More study details can be found in the specific papers.10,14


| 65


4

Composition of the pooled data setTo have comparable inclusion criteria for the secondary data analysis, only adult women (> 18 years) residing in an urban region (2500 or more households per km²) with a low-risk onset of pregnancy were eligible for inclusion. Application of these criteria meant a reduction of the Dutch study sample to 632 women (Figure 1). Because of the unbalance in the numbers between both samples, a pooled data set was constructed by combining the entire sample from the Belgian study (n = 333) and a random matched sample from the 632 women remaining in the Dutch study.

To reduce possible pre-exisiting differences in distribution between both populations, predictors for prenatal care use were used to define a comparable dataset. Our first step was to reduce missings in the Dutch data. Multiple imputation was performed for missing values with regard to household income (97/632), using the hot deck method.15 Seen Chi-square analyses indicated that non-response concerning household income depends on a woman’s occupational status and educational level (p < 0.05). Missing values of non-respondents were replaced by observed values from a respondent similar to the non-respondent16 for these variables. Five imputed data sets were generated to calculate the mean household income for each non-respondent. There were no missings in the Belgian study.

After completing the imputation in the Dutch sample we observed that the individual characteristics of both subsamples were distributed differently. Women in the Belgian subsample were significantly (p < 0.05) younger (aged ≤ 20), were more often single, more often less educated, less likely to be active on the labour market and were more often of a foreign nationality compared to the Dutch subsample. In the Belgian sample, women had more often a less educated partner (p < 0.05) and more often a partner with a foreign nationality. Furthermore, these women more often had a low and high equivalent income (p < 0.05) and lack of health insurance and additional health insurance cover. Finally, these women were more often multiparae (p < 0.05), had more unwanted pregnancies, more unplanned pregnancies and attended fewer prenatal information classes. These observed differences might potentially influence differences in healthcare utilization, therefore exact matching without replacement17 was conducted in order to balance the distribution of individual characteristics between the subsamples of the pooled data set. The units of the Dutch subsample were ordered at random and were matched 1:1 to the units of the Belgian subsample for two variables: educational level6,7,9 and maternal age.5-7 These variables were chosen because in literature they were observed to be predisposing determinants of PNC use. For 321 women in the Belgian sample we were able to match with someone in the DELIVER study. The final pooled dataset (n=642) therefore consisted of 321 women from Belgium and 321 from the Netherlands.

R1R2R3R4R5R6R7R8R9


66 |

Chapter 4

Women receiving care in primary midwifery practices

(n=14,640)

Deliver participants:7,907 women in primary

midwifery-led care

Women included in our sample (n=632)

Non-responders (n=6,733), information for 912 women regarding reasons of non-respons: miscarriage (25.9%), not interested (15.8%),language barrier (11.7%)

Women excluded with reasons (n=7,275);Women who did not fill in questionnaire 1 & 2 (n=5,137) Women not residing in ametropolitan area (n=1,537)Women younger than 19 years of age (n=2)Women without a singleton pregnancy (n=240)Women referred to secondary care during pregnancy until 1 week before delivery (n=101)

Elig

ible

popu

latio

nD

ELIV

ERco

hort

Star

ted

in p

rim

ary

mid

wife

ry-le

dca

re


Operationalisation of PNC utilization by the CTP toolThe CTP tool considers three dimensions: the timing of initiation of care, and the number and timing of three specific interventions during pregnancy (blood screening, ultrasound and blood pressure measurement).10 Four categories of PNC use are defined by the CTP: inadequate, intermediate, sufficient or appropriate care. This classification reflects the degree to which a minimum amount of care recommended for every pregnancy was received, regardless of parity or risk status.10 As the CTP was developed based on evidence about the importance of interventions in pregnancy and the congruence of PNC guidelines, the tool is applicable in the Netherlands.18-21


| 67


4

1. Initiation of care before 14 weeks of gestation

2. Number of interventions*:

3. Timing of interventions**:

YES

YES

NO

NO = Inadequate

YES = Intermediate

NO = Inadequate

US BP BS

> 2 and > 6 and > 2

Not all interventions occurred according the time table

= SufficientAll interventions occurred according the time table

= Appropriate

Minimum number per trimester

US BP BST1 1 1 1T2 1 2 0T3 0 3 1

US BP BS

> 5 or > 12 or > 5

Minimal range Upper

US: Ultrasound BP: Blood Pressure BS: Blood Screening T: Trimester*Ranges based on the NICE (NICE, 2008) and Belgian guideline (Lodewyckx K. et al., 2014) **Ranges based on the NICE guidelines (NICE, 2008)Inadequate: initiation of care after first trimester OR the number of at least one intervention is less than the lower range and none of the interventions occurred more than the rangeIntermediate: initiation of care in the first trimester; the number of at least one intervention occurred less than the lower range and at least one intervention exceeded the range Sufficient: initiation of care in the first trimester; the number of all interventions equals at least the respective lower range but timing of at least one intervention is not as recommended Appropriate: initiation of care in the first trimester; the number of the interventions equals at least the respective lower range and timing of the actions of all basic interventions is as recommended

Figure 2. Outline of the Content and Timing of care in Pregnancy (CTP) tool (Beeckman et al., 2013)

R1R2R3R4R5R6R7R8R9


68 |

Chapter 4

Potential individual determinants of PNC utilizationThe original data collection instruments used in both studies were explored to determine the variables that had been equivalently operationalised. The common variables to form the predisposing component were age, marital status, educational level, occupational status and current nationality. In addition, educational level and current nationality of the partner were examined. A variable for region referred to the two original study samples: the Metropolitan Region of Brussels, Belgium (CTP study) and urban regions in the Netherlands (DELIVER study). The educational level of CTP and DELIVER women was classified into three categories according to the International Standard Classification of Education (ISCED).22

The variables reflecting the enabling component were equivalent income, health insurance cover and additional health insurance cover. Equivalent income was calculated by using the modified Organization for Economic Co-operation and Development (OECD) scale and classified into three categories. This scale involves adjusting monthly household income based on its size and the age of its members.23 The lowest income group was defined at < 60% of the respective median national income24, the at-risk-of-poverty threshold.24 The moderate and high-income groups were delineated at 60–120% and > 120% of the national median equivalent net income respectively.

The variables describing the pregnancy-related component were parity, wish for pregnancy, planned pregnancy, continuity of care and attendance of prenatal information classes. Continuity of care was measured by the Continuity of Carer (COC) index, based on the number of visits to each different healthcare provider and the total number of visits.25 The index, expressed in percentage, was divided into two categories, with the cut off point < 50% and ≥ 50%.

Statistical analysisFor each region, the individual characteristics of the study sample and PNC utilization were summarised. Individual characteristics and PNC utilization were compared between countries using Chi-squared tests, the association between each of the individual characteristics and PNC utilization for the whole sample was determined (Chi-squared tests). Subsequently, logistic ordinal regression analysis was used to examine the significance of each individual characteristic in terms of its likelihood of being given a higher CTP classification, while controlling for the remaining significant characteristics. Since this was an exploratory study, backward elimination was used (stay level: p < 0.05). Our model was constructed in three steps in accordance with the health behaviour model.3,4 The first step considered predisposing variables, the second step considered enabling factors, with the selected predisposing variables fixed in the model, and in the final step the pregnancy-related variables were examined while controlling for the selected predisposing and enabling variables. In order to include other variations between the subsamples, the variable region


| 69


4

was fixed in this model from the first step onwards. A Score test for the proportional odds assumption and absence of multicollinearity was undertaken for each step. In addition, the final model assessed the percentage of concordant pairs of predicted probabilities and observed responses (> 60%). Multivariate analyses were conducted in SAS 9.1, and all other analyses were performed using SPSS Statistics 20.

RESULTS

Characteristics of the women The final data set consisted of 642 women. Chi-square tests indicated significant differences between the two subsamples for marital status, occupational status, nationality, educational level of the partner, nationality of the partner, equivalent income, health insurance or additional health insurance cover, parity, desire for pregnancy and attendance of prenatal information classes (p < 0.05).

The majority of the women in the final data set were aged between 21 and 35 years (82.2%), were co-habiting or married (94.1%), employed (65.3%), did not have tertiary education (58.6%) and did not have a foreign nationality (42.1%) (Table 1). Of the women, 42.1% had a partner with tertiary education and 68.7% had a partner who did not have a foreign nationality. With regard to the enabling characteristics, 70.2% of the women had a moderate equivalent income, 97.0% had health insurance cover and 32.9% had no additional health insurance cover. The pregnancy-related characteristics revealed that 55.8% of the women were multiparae. Pregnancy was wanted for 98.0% of the women but unplanned for 20.2%. A lower continuity of care provider, represented by a COC index < 50%, was observed for 72.1% of the women, while 62.9% did not attend prenatal information classes.

R1R2R3R4R5R6R7R8R9


70 |

Chapter 4

Tabl

e 1.

Stu

dy s

ampl

e ch

arac

teri

stics

, chi

-squ

ared

test

repo

rting

sig

nific

ance

leve

l for

ass

ocia

tion

with

ant

enat

al c

are

utiliz

ation

, ord

inal

regr

essi

on

anal

ysis

repo

rting

adj

uste

d od

ds ra

tios

(OR)

for

bein

g as

sign

ed in

to a

hig

her

Cont

ent a

nd T

imin

g of

Pre

gnan

cy (C

TP) c

ateg

ory

Pren

atal

car

e uti

lizati

on c

lass

ified

by

the

CTP

tool

P va

lue

χ² te

stA

djus

ted

OR

Inad

equa

teIn

term

edia

teSu

ffici

ent

App

ropr

iate

(N =

49)

(N =

46)

(N =

214

)(N

= 3

33)

Tota

l (co

lum

n %

)fN

(row

%)f

N (r

ow %

)fN

(row

%)f

N (r

ow %

)f

Pred

ispo

sing

cha

ract

eris

tics

Age

(yea

rs)

(a)

(b)

≤ 20

8 (1

.2)

00

6 (7

5.0)

2 (2

5.0)

21–3

552

8 (8

2.2)

41 (7

.8)

40 (7

.6)

172

(32.

6)27

5 (5

2.1)

> 35

106

(16.

5)8

(7.5

)6

(5.7

)36

(34.

0)56

(52.

8)M

arita

l sta

tus

(a)

(b)

Co-h

abiti

ng o

r m

arri

ed60

4 (9

4.1)

44 (7

.3)

45 (7

.5)

197

(32.

6)31

8 (5

2.6)

Sing

le38

(5.9

)5

(13.

2)1

(2.6

)17

(44.

7)15

(39.

5)O

ccup

ation

al s

tatu

s<

0.00

10.

49 (0

.34-

0.70

)Em

ploy

ed

419

(65.

3)20

(4.8

)26

(6.2

)12

0 (2

8.6)

253

(60.

4)U

nem

ploy

ed22

3 (3

4.7)

29 (1

3.0)

20 (9

.0)

94 (4

2.2)

80 (3

5.9)

Educ

ation

al le

vel

< 0.

001

0.60

(0.4

3-0.

82)

Up

to s

econ

dary

376

(58.

6)35

(9.3

)33

(8.8

)13

9 (3

7.0)

169

(44.

9)Te

rtiar

y26

6 (4

1.4)

14 (5

.3)

13 (4

.9)

75 (2

8.2)

164

(61.

7)N

ation

ality

0.

009

(b)

Belg

ian/

Dut

ch47

5 (7

4.0)

29 (6

.1)

36 (7

.6)

149

(31.

4)26

1 (5

4.9)

All

othe

r na

tiona

lities

16

7 (2

6.0)

20 (1

2.0)

10 (6

.0)

65 (3

8.9)

72 (4

3.1)

Educ

ation

al le

vel p

artn

er

< 0.

001

(b)

No

part

ner

38 (5

.9)

5 (1

3.2)

1 (2

.6)

17 (4

4.7)

15 (3

9.5)

Up

to s

econ

dary

334

(52.

0)33

(9.9

)30

(9.0

)12

0 (3

5.9)

151

(45.

2)Te

rtiar

y27

0 (4

2.1)

11 (4

.1)

15 (5

.6)

77 (2

8.5)

167

(61.

9)N

ation

ality

of t

he p

artn

er

0.00

3(b

)N

o pa

rtne

r 38

(5.9

)5

(13.

2)1

(2.6

)17

(44.

7)15

(39.

5)Be

lgia

n/D

utch

441

(68.

7)27

(6.1

)29

(6.6

)13

3 (3

0.2)

252

(57.

1)A

ll ot

her

natio

naliti

es

163

(25.

4)17

(10.

4)16

(9.8

)64

(39.

3)66

(40.

5)Re

gion

0.00

90.

90 (0

.64-

1.26

)Br

usse

ls M

etro

polit

an R

egio

n32

1 (5

0.0)

31 (9

.7)

26 (8

.1)

118

(36.

8)14

6 (4

5.5)

Urb

an D

utch

regi

ons

321

(50.

0)18

(5.6

)20

(6.2

)96

(29.

9)18

7 (5

8.3)


| 71


4

Enab

ling

char

acte

risti

csEq

uiva

lent

inco

mec

< 0.

001

(b)

Low

11

2 (1

7.4)

17 (1

5.2)

9 (8

.0)

51 (4

5.5)

35 (3

1.3)

Mod

erat

e 45

1 (7

0.2)

29 (6

.4)

33 (7

.3)

141

(31.

3)24

8 (5

5.0)

Hig

h 79

(12.

3)3

(3.8

)4

(5.1

)22

(27.

8)50

(63.

3)H

ealth

insu

ranc

e co

ver

(a)

(b)

Yes

623

(97.

0)46

(7.4

)46

(7.4

)20

2 (3

2.4)

329

(52.

8)N

o19

(3.0

)3

(15.

8)0

(0.0

)12

(63.

2)4

(21.

1)A

dditi

onal

hea

lth in

sura

nce

cove

r<

0.00

1(b

)Ye

s 43

1 (6

7.1)

24 (5

.6)

29 (6

.7)

130

(30.

2)24

8 (5

7.5)

No

211

(32.

9)25

(11.

8)17

(8.1

)84

(39.

8)85

(40.

3)Pr

egna

ncy-

rela

ted

char

acte

risti

csPa

rity

0.04

2(b

)Pr

imip

arae

28

4 (4

4.2)

19 (6

.7)

16 (5

.6)

84 (2

9.6)

165

(58.

1)M

ultip

arae

358

(55.

8)30

(8.4

)30

(8.4

)13

0 (3

6.3)

168

(46.

9)W

ish

for

preg

nanc

yd(a

)(b

)W

ante

d pr

egna

ncy

628

(98.

0)49

(7.8

)44

(7.0

)21

0 (3

3.4)

325

(51.

8)U

nwan

ted

preg

nanc

y13

(2.0

)0

(0.0

)2

(15.

4)4

(30.

8)7

(53.

8)Pl

anne

d pr

egna

ncy

0.01

3(b

)Ye

s 51

2 (7

9.8)

35 (6

.8)

34 (6

.6)

161

(31.

4)28

2 (5

5.1)

No

130

(20.

2)14

(10.

8)12

(9.2

)53

(40.

8)51

(39.

2)CO

Ce0.

041

0.60

(0.4

2-0.

84)

< 50

%

463

(72.

1)42

(9.1

)39

(7.8

)15

8 (3

4.1)

227

(49.

0)≥

50%

179

(27.

9)7

(3.9

)10

(5.6

)56

(31.

3)10

6 (5

9.2)

Atten

ding

ant

enat

al in

form

ation

cla

sses

< 0.

001

0.67

(0.4

7-0.

94)

Yes

238

(37.

1)11

(4.6

)7

(2.9

)72

(30.

3)14

8 (6

2.2)

No

404

(62.

9)38

(9.4

)39

(9.7

)14

2 (3

5.1)

185

(45.

8)

a: T

he c

ondi

tion

for

the

chi-s

quar

ed te

st fo

r la

rger

con

tinge

ncy

tabl

es w

as n

ot m

et: v

alid

if le

ss th

an 2

0% o

f the

exp

ecte

d nu

mbe

rs a

re u

nder

5 a

nd th

e m

inim

um

expe

cted

cou

nt is

less

than

1 (K

irkw

ood

B.R.

and

Ste

rne

J.A.C

., 20

03)

b: N

ot in

clud

ed in

the

final

mod

el o

f ord

inal

logi

stic

regr

essi

on a

naly

sis

c: ∑

inco

mes

in th

e ho

useh

old/

(1 +

(x*0

.5) +

(y*0

.3))

(x: n

umbe

r of

adu

lts li

ving

in th

e sa

me

hous

ehol

d, y

: num

ber

of c

hild

ren

unde

r th

e ag

e of

18

year

s liv

ing

in th

e sa

me

hous

ehol

d [m

odifi

ed O

ECD

sca

le] (

OEC

D, 2

013)

)

d: n

= 6

41e:

Con

tinui

ty o

f Car

e in

dex:

CO

C =

( (B

ice

and

Boxe

rman

, 197

7))

f: Be

caus

e of

roun

ding

, per

cent

ages

may

not

add

up

to 1

00%

a: T

he c

ondi

tion

for t

he c

hi-s

quar

ed te

st fo

r lar

ger c

ontin

genc

y ta

bles

was

not

met

: val

id if

less

than

20%

of t

he e

xpec

ted

num

bers

are

und

er 5

and

the

min

imum

exp

ecte

d co

unt i

s les

s tha

n 1

(Kirk

woo

d B

.R. a

nd S

tern

e J.A

.C.,

2003

)

b: N

ot in

clud

ed in

the

final

mod

el o

f ord

inal

logi

stic

regr

essi

on a

naly

sis

c: ∑

inco

mes

in th

e ho

useh

old/

(1 +

(x*0

.5) +

(y*0

.3))

(x: n

umbe

r of a

dults

livi

ng in

the

sam

e ho

useh

old,

y: n

umbe

r of c

hild

ren

unde

r the

age

of 1

8 ye

ars l

ivin

g

in th

e sa

me

hous

ehol

d [m

odifi

ed O

ECD

scal

e] (O

ECD

, 201

3))

d: n

= 6

41

e: C

ontin

uity

of C

are

inde

x: C

OC

=

( (B

ice

and

Box

erm

an, 1

977)

)

f: B

ecau

se o

f rou

ndin

g, p

erce

ntag

es m

ay n

ot a

dd u

p to

100

%

Tabl

e 2.

Com

paris

on o

f pre

nata

l car

e ut

iliza

tion

betw

een

regi

ons (

N=6

42)

To

tal

Brus

sels

Met

ropo

litan

R

egio

n U

rban

Dut

ch r

egio

ns

p-va

lue

χ² te

st

(N=6

42)

(N=3

21)

(N=3

21)

N (c

olum

n %

) N (c

olum

n %

) N

(col

umn

%)

C

onte

nt a

nd T

imin

g of

Pre

gnan

cy c

are

Inad

equa

te

49 (7

.6)

31 (9

.7)

18 (5

.6)

0.00

9 In

term

edia

te

46 (7

.2)

26 (8

.1)

20 (6

.2)

Su

ffic

ient

21

4 (3

3.3)

11

8 (3

6.8)

96

(29.

9)

A

ppro

pria

te

333

(51.

9)

146

(45.

5)

187

(58.

3)

R1R2R3R4R5R6R7R8R9


72 |

Chapter 4

Table 2. Comparison of prenatal care utilization between regions (N=642)

Total Brussels Metropolitan Region

Urban Dutch regions

p-value χ² test

(N=642) (n=321) (n=321)N (column %) N (column %) N (column %)

Content and Timing of Pregnancy careInadequate 49 (7.6) 31 (9.7) 18 (5.6) 0.009Intermediate 46 (7.2) 26 (8.1) 20 (6.2)Sufficient 214 (33.3) 118 (36.8) 96 (29.9)Appropriate 333 (51.9) 146 (45.5) 187 (58.3)

Comparison of PNC utilization between both countriesPNC utilization differed significantly between countries (p = 0.009) (Tables 1-2). According to the classification by the CTP tool, 9.7% of the women from the CTP subsample had an inadequate care trajectory compared with 5.6% in the DELIVER subsample. Furthermore, only 45.5% of the women in Belgium, compared to 58.3% of Dutch women, were assigned to the appropriate PNC group (Table 2).

Individual determinants of PNC utilizationThe predisposing characteristics of occupational status (p < 0.001), educational level and nationality of the women (p < 0.001; p = 0.009 respectively) and their partners (p < 0.001; p = 0.003 respectively) were found to be significantly associated with PNC utilization (Table 1). Appropriate PNC use was higher among women with tertiary education (61.7%), who were employed (60.4%) and who were not of a foreign nationality (54.9%) than among women with secondary level education (44.9%), who were unemployed (35.9%) and had a foreign nationality (43.1%) respectively. Concerning the enabling characteristics, results showed that the higher the equivalent income, the higher the proportion of women with appropriate PNC utilization (p < 0.001). More than half of the women with moderate (55.0%) or high equivalent income (63.3%) received appropriate PNC. This proportion was 31.3% among women with low equivalent income. Women with additional health insurance cover received appropriate content and timing of pregnancy care more often than women without this cover (57.5% versus 40.3%) (p < 0.001). With respect to pregnancy-related characteristics, appropriate care use was higher among primiparae (58.1%), women with a planned pregnancy (55.1%), women who had a COC index ≥ 50% (59.2%) and women who attended prenatal information classes (62.2%) compared with multiparae (46.9%), women with an unplanned pregnancy (39.2%), women who had a COC index < 50% and women who did not attend prenatal information classes (45.8%) respectively (p < 0.05).

In the final model of the multivariate analysis, after adjustment for confounding variables (Table 1), the overall regional variable (the CTP versus the DELIVER subsamples) did not


| 73


4

remain significantly associated with PNC use. However, four variables were significantly associated with PNC utilization when controlling for the other variables. Women with no more than a secondary education (OR: 0.60; 95% CI 0.43–0.82) and unemployed women (OR: 0.49; 95% CI 0.34–0.70) had lower odds of being assigned to a higher CTP category compared with women with tertiary education and employment respectively.

In the final model no enabling characteristics remained significantly associated with the content and timing of PNC. With regard to pregnancy-related characteristics women with a COC index < 50% (OR: 0.60; 95% CI 0.42–0.84) and women who did not attend prenatal information classes (OR: 0.67; 95% CI 0.47–0.94) had lower odds of obtaining a higher CTP classification compared with women with a COC index ≥ 50% and those attending prenatal information classes respectively.

DISCUSSION

This study compared PNC utilization as classified by the CTP tool between cohorts in two different countries and identified predisposing, enabling and pregnancy-related determinants based on a pooled data set. To our knowledge this is the first international comparative study that has considered these three factors related to the content and timing of PNC. Unadjusted analysis revealed that women in urban Dutch regions received more appropriate PNC than women in the Brussels Metropolitan Region. However, multivariate analysis did not indicate that the region in itself was a determinant of PNC utilization when controlling for all individual characteristics. Irrespective of the region, the content and timing of PNC was associated with educational level, employment status, continuity of care and attendance of prenatal information classes.

Previous studies have shown that a low educational level is associated with late initiation of PNC6,7,9, a low number of prenatal visits6,10, receiving no care at all6 and a lower probability of being in a higher CTP category.8 Lack of a paid job9 and type of occupation26 have also been related to inadequate PNC use. Choté et al. suggested that education might influence PNC use due to the level of general health knowledge and health literacy.9 The knowledge and skills acquired through education may create better access to information, stimulate receptiveness to health education messages and thus enable to access and communicate with healthcare providers.27 The social network, which may be less extended in unemployed women might be a mechanism explaining the association of employment with PNC use. Information and encouragement received through a social network may stimulate women to use care.28,29

No enabling characteristics, such as income, were retained in our final model. The compulsory universal cover offered by health insurers, which includes basic PNC in both Belgium13 and the Netherlands30 may play a part. However, the provision of universal cover

R1R2R3R4R5R6R7R8R9


74 |

Chapter 4

seems to be insufficient to offset disparities in PNC utilization.26 The use of healthcare services can be measured in terms of realised access to these services.4 Inequitable access occurs when important structural aspects of society determine who receives appropriate PNC. However, a sole focus on measures designed to alter these aspects – such as educational level and employment status – for the sake of promoting equitable access, is hard due to their low mutability.4 Other measures, such as the promotion of health literacy and knowledge from an early age through the education system or the training of health professionals in communication skills to adapt to the health literacy level of the care seeker, may encourage better utilization of care.31

With regard to pregnancy-related determinants, this study demonstrated that a lower continuity of PNC provider was associated with a lower CTP category. This index was calculated without differentiating between the type of primary caregiver – in Belgium most often an obstetrician and in the Netherlands a midwife. These results indicate that the continuity of care provider is important for the appropriateness of care irrespective of the type of provider. Attending prenatal classes was related to receiving a more appropriate PNC trajectory, although the number and content of these classes were not considered. While non-attenders were not convinced that prenatal classes might benefit them, attenders considered them to be valuable.32 Similarly, non-attenders may be less convinced of the importance of and need for PNC, which may hinder appropriate PNC use. Non-attenders of prenatal education classes were found to come from more vulnerable groups, with a low level of education or being unemployed.33 Enhancing the awareness of the importance of appropriate follow-up and the advantages of prenatal classes may stimulate care use.

Cross-border data-sharing enabled the study of PNC utilization in two countries. However, the number of variables used in this study was restricted by the variables equally examined and operationalised in the original studies.8,14 For example, origin or ethnicity could not be examined in this study due to different operationalisation, although previous studies have identified these variables as important determinants of PNC use.2,5,6,8,9

The results of our study demonstrate that educational level and employment status are important factors in obtaining appropriate content and timing of PNC in both regions. One way to promote appropriate PNC would be to introduce measures encouraging women to attend prenatal classes, for example by providing classes free of charge to socially vulnerable women. Furthermore, it is important to systematically create maternal healthcare models in which the continuity of care provider is ensured. Both are modifiable factors that will contribute to more appropriate care use.

Despite the value of this study, more cross-border studies are required to examine other individual determinants, such as origin, social network and health beliefs with regard to pregnancy and care. These studies should also use a larger sample including women residing in both urban and non-urban regions. To achieve this, systematic and routine data collection


| 75


4

that provides information on elements of the CTP tool and the individual characteristics of pregnant women will be required.

ACKNOWLEDGEMENTS

The authors would like to thank M. Westerneng who enabled the data-sharing, without which this study could not have been performed.

KEY POINTS

• The CTP tool, measuring PNC utilization, was applied in two cohorts from different countries for the first time.

• Irrespective of the region women live in, the content and timing of PNC were associated with educational level, employment status, continuity of caregiver and attending prenatal classes.

• The education system, the health system and PNC providers should address health literacy and the knowledge levels of socially vulnerable women who may require a more personalised approach to improve communication.

• Stimulating attendance of prenatal classes and continuity of care in PNC models are factors that can be modified to contribute to more appropriate care use.

• Systematic, reliable and routine data collection that provides information on elements of the CTP tool and the individual characteristics of pregnant women is required as a basis that will enable joint international research on the determinants of PNC use.

R1R2R3R4R5R6R7R8R9


76 |

Chapter 4

REFERENCES

(1) Boerleider AW, Wiegers TA, Mannien J, Francke AL, Deville WL. Factors affecting the use of prenatal care by non-western women in industrialized western countries: a systematic review. BMC Pregnancy Childbirth 2013 Mar 27;13:81.

(2) Martinez-Garcia E, Olvera-Porcel MC, de Dios Luna-Del Castillo J, Jimenez-Mejias E, Amezcua-Prieto C, Bueno-Cavanillas A. Inadequate prenatal care and maternal country of birth: a retrospective study of southeast Spain. Eur J Obstet Gynecol Reprod Biol 2012 Dec;165(2):199-204.

(3) Andersen R, Newman JF. Societal and individual determinants of medical care utilization in the United States. Milbank Mem Fund Q Health Soc 1973;51:95-124.

(4) Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Soc Behav 1995;36:1-10.

(5) Baker EC, Rajasingam D. Using Trust databases to identify predictors of late booking for antenatal care within the UK. Public Health 2012 Feb;126(2):112-116.

(6) Feijen-de Jong EI, Jansen DE, Baarveld F, van der Schans CP, Schellevis FG, Reijneveld SA. Determinants of late and/or inadequate use of prenatal healthcare in high-income countries: a systematic review. Eur J Public Health 2012 Dec;22(6):904-913.

(7) Heaman M, Bayrampour H, Kingston D, Blondel B, Gissler M, Roth C, et al. Migrant women’s utilization of prenatal care: a systematic review. Matern Child Health J 2013 Jul;17(5):816-836.

(8) Beeckman K, Louckx F, Putman K. Content and timing of antenatal care: predisposing, enabling and pregnancy-related determinants of antenatal care trajectories. Eur J Public Health 2013 Feb;23(1):67-73.

(9) Chote AA, Koopmans GT, Redekop WK, de Groot CJ, Hoefman RJ, Jaddoe VW, et al. Explaining ethnic differences in late antenatal care entry by predisposing, enabling and need factors in The Netherlands. The Generation R Study. Matern Child Health J 2011 Aug;15(6):689-699.

(10) Beeckman K, Louckx F, Masuy-Stroobant G, Downe S, Putman K. The development and application of a new tool to assess the adequacy of the content and timing of antenatal care. BMC Health Services Research 2011 09/06;11:213-213.

(11) Feijen-de Jong EI, Baarveld F, Jansen DE, Ursum J, Reijneveld SA, Schellevis FG. Do pregnant women contact their general practitioner? A register-based comparison of healthcare utilisation of pregnant and non-pregnant women in general practice. BMC Fam Pract 2013 Jan 16;14:10-2296.

(12) De Geus M. Midwifery in the Netherlands. 2012.(13) Gerkens S, Merkur S. Belgium: Health system review. Health Syst Transit 2010;12(5):1-266, xxv.(14) Mannien J, Klomp T, Wiegers T, Pereboom M, Brug J, de Jonge A, et al. Evaluation of primary care

midwifery in the Netherlands: design and rationale of a dynamic cohort study (DELIVER). BMC Health Services Research 2012;12(1):69.

(15) Mander A, Clayton D. Weighted Hotdeck Imputation. Stata Technical Bulletin 1999;51:32-34.(16) Andridge RR, Little RJ. A Review of Hot Deck Imputation for Survey Non-response. Int Stat Rev

2010 Apr;78(1):40-64.(17) Stuart EA. Matching methods for causal inference: A review and a look forward. Stat Sci 2010

Feb 1;25(1):1-21.(18) Dutch Association of Obstetrics and Gynecology (Nederlandse Vereniging voor Obstetrie

en Gynaecologie). Guidelines and position papers. 2002; Available at: http://www.nvog.nl/vakinformatie/Richtlijnen,+standpunten,+modelprotocollen+enz/default.aspx. Accessed 02/14.


| 77


4

(19) Dutch Association of Obstetrics and Gynecology (Nederlandse Vereniging voor Obstetrie en Gynaecologie). Quality Standards. Prenatal screening on fetal anomalies. (Prenatale screening op foetale afwijkingen). 2005.

(20) National Institute for Public Health and the Environment (RIVM). Prenatal screening infectious diseases and and erytrocyte immunization. (Draaiboek Prenatale Screening Infectieziekten en Erytrocytenimmunisatie). 2008.

(21) Verstappen W, Jans S, Van Egmond N, Van Laere A, Schippers-van Mourik M, Labots-Vogelesang S, et al. NHG guidelines: Anaemia in pregnancy and postpartum period. (LESA Anemie tijdens zwangerschap en kraamperiode). 2007.

(22) UNESCO-UIS. International Standard Classification of Education; ISCED. 1997; Available at: http://www.uis.unesco.org/Library/Documents/isced97-en.pdf. Accessed 07/21, 2013.

(23) OECD. What are equivalence scales? Available at: http://www.oecd.org/eco/growth/OECD-Note-EquivalenceScales.pdf. Accessed 07/21, 2013.

(24) Commission E. Eurostat. national mean and median income by household type. 2009.(25) Bice TW, Boxerman SB. A quantitative measure of continuity of care. Med Care 1977;15:347-

349.(26) Simoes E, Kunz S, Munnich R, Schmahl FW. Association between maternal occupational status

and utilization of antenatal care Study based on the perinatal survey of Baden-Wuerttemberg 1998-2003. Int Arch Occup Environ Health 2006 Jan;79(1):75-81.

(27) Solar O, Irwin A. A conceptual framework for action on the social determinants of health. Social determinants of health discussion Paper 2 (Policy an Practice) WHO 2010.

(28) Deri C. Social networks and health service utilization. J Health Econ 2005 Nov;24(6):1076-1107.(29) Lin N. Building a Network Theory of Social Capital. Connections 1999;22:28-51.(30) Schafer W, Kroneman M, Boerma W, van den Berg M, Westert G, Deville W, et al. The Netherlands:

health system review. Health Syst Transit 2010;12(1):v-xxvii, 1-228.(31) Parker RM, Ratzan SC, Lurie N. Health literacy: a policy challenge for advancing high-quality

health care. Health Aff (Millwood) 2003 Jul-Aug;22(4):147-153.(32) Murphy Tighe S. An exploration of the attitudes of attenders and non-attenders towards

antenatal education. Midwifery 2010 Jun;26(3):294-303.(33) Fabian HM, Radestad IJ, Waldenstrom U. Characteristics of Swedish women who do not attend

childbirth and parenthood education classes during pregnancy. Midwifery 2004; Sep;20(3):226-235.

CHAPTER 5

Do pregnant women contact their general practitioner?

A register-based comparison of healthcare utilization of

pregnant and non-pregnant women in general practice

Esther I Feijen-de Jong, Frank Baarveld, Danielle EMC Jansen, Jennie Ursum,Sijmen A Reijneveld, Francois Schellevis

BMC Family Practice 2013; 14:10

R1R2R3R4R5R6R7R8R9


80 |

Chapter 5

ABSTRACT Background: Midwives and obstetricians are the key providers of care during pregnancy and postpartum. Information about the consultations with a general practitioner (GP) during this period is generally lacking. The aim of this study is to compare consultation rates, diagnoses and GP management of pregnant women with those of non-pregnant women.

Methods: Data were retrieved from the Netherlands Information Network of General Practice (LINH), a nationally representative register. This register holds longitudinal data on consultations, prescriptions and the referrals of all patients listed at 84 practices in the Netherlands in 2007–2009, including 15,123 pregnant women and 102,564 non-pregnant women in the same age-range (15 to 45 years). We compared consultation rates (including all contacts with the practice), diagnoses (ICPC-1 coded), medication prescriptions (coded according to the Anatomical Therapeutic Chemical classification system), and rate and type of referrals from the start of the pregnancy until six weeks postpartum (336 days).

Results: Pregnant women contacted their GP on average 3.6 times, compared to 2.2 times for non-pregnant women. The most frequently recorded diagnoses for pregnant women were ‘pregnancy’ and ‘cystitis/urinary infection’, and ‘cystitis/urinary infection’ and ‘general disease not otherwise specified’ for non-pregnant women. The mean number of prescribed medications was lower in pregnant women (2.1 against 4.4). For pregnant women, the most frequent referral indication concerned obstetric care, for non-pregnant women this concerned physiotherapy.

Conclusions: GP consultation rates in pregnancy and postpartum shows that GPs are important providers of care for pregnant women. Therefore, the involvement of GPs in collaborative care during pregnancy and postpartum should be reinforced.

Keywords: Primary healthcare, General practitioner, Pregnancy, Health services research, Prenatal care


| 81

Do pregnant women contact their general practitioner? A register-based comparison of healthcare utilization of pregnant and non-pregnant women in general practice

5

INTRODUCTION

In most industrialised countries midwives and obstetricians are the key providers of care during and shortly after pregnancy, with a recommended number of antenatal visits ranging from six to 15.1 This also applies to the Netherlands, where the role of General Practitioners (GPs) as key providers of pregnancy and obstetric care has further declined over recent decades.2-5 Currently only 2-6% of all practising GPs still provide obstetric care. For most women with uncomplicated pregnancies, primary care midwives provide routine prenatal, intrapartum and postpartum care, and they act as gatekeepers to secondary obstetric care.6,7 Furthermore, the majority of GP-trainees receive very little theoretical education regarding pregnancy. Despite this, women may still consult GPs as they remain the main providers of routine medical care during pregnancy and postpartum.

Information about the frequency of consultations, diagnoses and management by GPs during pregnancy and postpartum is very scarce. The only available study on this subject was performed by Coco.8 However, Coco compared rates of additional medical problems unrelated to pregnancy as encountered by GPs and obstetricians while providing prenatal care. He found that GPs made significantly more diagnoses unrelated to pregnancy than obstetricians. However, this study does not provide information about the healthcare utilization of pregnant women in a healthcare system where midwives and obstetricians are the main maternal healthcare providers.

Evidence on the rates and content of women’s consultations with GPs during pregnancy and postpartum, and the management of these problems by GPs could provide insight into the health and healthcare needs of pregnant women. This evidence could also provide a context for the collaboration and communication between various services and professionals, and the needs for training. Moreover, it provides insight into the role of GPs during pregnancy and the postpartum period.

Therefore, this study aims to compare consultation rates, diagnoses and GPs’ management of pregnant women with those of non-pregnant women in general practice. With this study we can contribute to the understanding of the health status and healthcare needs of pregnant women consulting a GP. We limited our study to full-term pregnancies, covering the pregnancy period and the first six weeks postpartum.

METHODS

Study design and settingWe retrieved data from a nationally representative register of GP care: the Netherlands Information Network of General Practice (LINH), including data from 15,123 pregnant women and 102,564 non-pregnant women of the same age range. LINH collects longitudinal

R1R2R3R4R5R6R7R8R9


82 |

Chapter 5

data on consultations, prescriptions and referrals9 of all patients listed in 84 practices in the Netherlands. In the Netherlands every inhabitant has to register with a GP. With respect to the LINH9 data quality rules, 74 of the 84 participating LINH practices could be used for statistical analyses. None of the participating GPs provided obstetric care. The LINH register is set up in accordance with national legislation, which does not require medical ethical approval or obtaining informed consent from individual patients. We obtained permission to access the LINH-data from the LINH steering group.

ParticipantsPregnant women were identified in the register using the birthdates of children born in 2007, 2008 and 2009. We selected their mothers and measured their healthcare utilization within a period starting at 294 days (42 weeks) before the date of birth of their child and ending 42 days (6 weeks) postpartum. Mothers of preterm and stillborn children were excluded. The comparison group of non-pregnant women comprised all non-pregnant women of reproductive age (15–45 years) listed in the same LINH practices. Non-pregnant women were defined as not having given birth to a child and without any indication related to pregnancy in the register. Their healthcare utilization was measured from July 2006 until June 2010 to create a similar study period. We randomly selected a period of 336 days during these four years for each non-pregnant woman, to get an observation period identical to that of the pregnant women.

MeasuresThe primary outcomes in this study were consultation rates, diagnoses and GP management during pregnancy and the postpartum period. Consultation rates concerned the number of recorded contacts with the GP practice in the defined 336-day period. Diagnoses included the GPs’ assessment of the presented health problem at each contact: diagnoses were coded by the GP according to the International Classification of Primary Care-1 (ICPC-1). The ICPC has been designed to classify symptoms and diagnoses in primary care.10 It discriminates between symptoms and complaints (ICPC code numbers 01–29), and diagnoses (ICPC code numbers 70–99), further denoted in this paper as ‘symptom diagnoses’ and ‘diagnoses’. Diagnoses were categorised by ICPC chapter, representing the major organ systems. Coding of health problems in the GP practice is incentivised by links in the practice software packages of ICPC codes with procedures enhancing quality of care, e.g. selection of specific patient groups, monitoring medication, or follow-up appointments. GP management concerned prescribed medication (automatically coded according to the ATC Anatomical Therapeutic Chemical classification system) and referrals to other care professionals.

We further obtained data on background characteristics, age, socioeconomic status (SES), and level of urbanisation. SES concerned the socioeconomic context of the place of


| 83


5

residence. This was measured by an existing area score based on mean income, percentage of households with a low income, percentage of inhabitants without a paid job, and percentage of households with only low education, per 4-digit postal code.11-13 This SES was categorised as high (≤ 25th percentile), middle or low (≥ 75th percentile). Level of urbanisation concerned the average number of addresses per square km within a radius of one kilometre based on the same 4-digit postal codes. It was categorised as very urban (>2500 addresses/km), intermediate urban/rural (between 500 and 2500 addresses/km) and very rural (<500 addresses/km).13

AnalysesFirstly, we described the background characteristics of our study population. Next, we compared pregnant and non-pregnant women regarding consultation rates, diagnoses, prescribed medication and referrals. If the same diagnosis, prescription or referral was recorded several times, this was counted only once per woman. Stata 11.2 was used for all analyses.

RESULTS

Characteristics of the study populationTable 1 shows the characteristics of our study population. There were only small differences between pregnant and non-pregnant women with respect to age and level of urbanisation. More pregnant women lived in both high and low SES areas.

Table 1. Characteristics of the study populationPregnant womenn = 15,123

Non-pregnant womenn = 102,564

Age (mean/SD) 30.6 (4.9) 31.1 (9.6)SES (%)a High 20.4 14.6

Middle 46.4 55.0Low 32.7 29.8

Urbanisation (%)b Very rural 49.0 48.8Intermediate urban/rural 23.3 21.8Very urban 27.2 28.9

a Missing: 0.5% (pregnant women), 0.6% (non-pregnant women).b Missing: 0.5% (pregnant women), 0.5% (non-pregnant women).

Consultation ratesTable 2 shows the consultation rates of pregnant and non-pregnant women in GPs’ practices. Pregnant women had on average 1.4 more contacts with their GP than non-pregnant women. Of the pregnant women, 35% had no contact with their GP, compared to 50% of

R1R2R3R4R5R6R7R8R9


84 |

Chapter 5

the non-pregnant women. Of the women who had contact with their GP, the number of contacts was also higher for pregnant women than for non-pregnant women.

Table 2. Number of general practitioner (GP) contacts of pregnant and non-pregnant women

Pregnant women n = 15,123

Non-pregnant women n = 102,564

During pregnancy and postpartum period (336 days)

Random period of 336 days

Number of contactsa:Median (IQRb) 2 (0–5) 0 (0–3)Mean (SDc) 3.6 (4.8) 2.2 (4.1)

Number of women not visiting the GP at all (%)

5,246 (35%) 51,494 (50%)

Number of contacts if visiting:Median (IQRb) 4 (2–7) 3 (1–6)Mean (SDc) 5.6 (4.9) 4.5 (4.8)

a Differences between the groups are all statistically significant (p < 0.001). b IQR = inter quartile range. c SD = standard deviation

Diagnoses and GP managementTable 3 shows the diagnoses recorded by GPs according to the ICPC chapters. These were mostly quite similar for pregnant and non-pregnant women, except for three ICPC chapters that showed differences of more than 10%. First, as would be expected, pregnant women had far more contacts with their GPs for diagnoses related to pregnancy, birth and family planning (ICPC chapter W). Second, pregnant women had fewer contacts with their GPs for musculoskeletal disorders (ICPC chapter L): 17 against 29 per 100 women for non-pregnant women. This difference could not be attributed to a single ICPC code in chapter L. Third, the number of diagnoses for urological problems was higher in pregnant than in non-pregnant women (23 against 9 per 100 women). This was particularly due to higher numbers of cystitis/urinary infections in pregnant women (ICPC code U71, 14 against 6 per 100 women).


| 85


5

Table 3. Number of diagnoses by ICPC chapter, per 100 women

Pregnant women n = 15,123

Non-pregnant women n = 102,564

Diff.

During pregnancy and postpartum

period (336 days)


ICPC chapterA General 34.7 28.8 5.9B: Blood, blood forming organs, lymphatic, spleen

4.9 2.8 2.1

D: Digestive 19.5 15.0 4.5F: Eye 2.5 4.5 −2.1H: Ear 4.7 6.0 −1.3K: Cardiovascular 10.9 9.2 1.7L: Musculoskeletal 17.4 29.5 −12.0N: Neurological 5.9 7.7 −1.9P: Psychosocial 7.6 14.9 −7.3R: Respiratory 23.2 26.1 −2.9S: Skin 27.3 27.4 −0.1T: Endocrine/Metabolic and Nutritional 3.2 6.8 −3.6U: Urological 23.4 8.8 14.6W: Pregnancy, Childbearing, Family Planning 121.9 10.9a 111.0X: Male Genital 19.9 17.7 2.2Z: Social Problems 2.7 3.3 −0.6Mean number of diagnoses per contact (SD)b 1.14 (0.4) 1.17 (0.5)Mean number of contacts (SD) 3.64 (4.8) 2.24 (4.1)

a Related to Family Planning codes only.b More than one diagnosis per contact is possible.

Most frequently made diagnosesThe ten most frequently recorded symptom diagnoses and diagnoses showed many similarities between pregnant and non-pregnant women; however, there were differences in their ranking. The symptom diagnoses most frequently recorded in contacts with pregnant women were ‘oral contraception’ followed by ‘other localized abdominal pain’, and ‘pregnancy-related vomiting/nausea’ (Table 4). ‘Pregnancy’, the most frequently recorded diagnosis, was recorded in 40.8% of all pregnant women.

R1R2R3R4R5R6R7R8R9


86 |

Chapter 5

Table 4. The ten most frequently recorded symptom diagnoses and diagnoses of pregnant and non-pregnant women per 100 women

Pregnant women n = 15,123 Non-pregnant women n = 102,564

During pregnancy and postpartum period (336 days)


Symptom diagnosesContraception oral (W11) 4.3 Contraception oral (W11) 3.8Abdominal pain localized other (D06) 3.2 Weakness/tiredness general (A04) 2.6Pregnancy vomiting/nausea (W05) 3.0 Cough (R05) 2.4Cough (R05) 2.7 Low back symptom/complaint (L03) 2.1Low back symptom/complaint (L03) 2.6 Abdominal pain localized other (D06) 1.9Constipation (D12) 2.5 Contraception intrauterine (W12) 1.7Pregnancy symptom complaint other (W29) 2.3 Foot/toe symptom/complaint (L17) 1.6Contraception intrauterine (W12) 2.0 Neck symptom/complaint (L01) 1.5Back symptom/complaint (L02) 1.8 Back symptom/complaint (L02) 1.5Vaginal discharge (X14) 1.8 Throat symptom/complaint (R21) 1.4DiagnosesPregnancy (W78) 40.8 Cystitis/urinary infection other (U71) 4.0Cystitis/urinary infection other (U71) 8.7 General disease NOS (A99) 3.6General disease NOS (A99) 5.9 Upper respiratory infection acute (R74) 3.4No disease (A97) 5.9 No disease (A97) 2.9Genital candidiasis female (X72) 5.3 Dermatitis contact/allergic (S88) 2.7Upper respiratory infection acute (R74) 4.8 Sinusitis acute/chronic (R75) 2.5Haemorrhoids (K96) 3.2 Allergic rhinitis (R97) 2.2Puerperal mastitis (W94) 3.2 Dermatophytosis (S74) 2.2Dermatophytosis (S74) 2.7 Genital candidiasis female (X72) 1.9Dermatitis contact/allergic (S88) 2.6 Asthma (R96) 1.8

GP managementTable 5 shows the most frequently recorded medication prescriptions and referrals. Almost half of pregnant women received a prescription, mostly for antibacterials for systemic use (ATC-code J01, 16.4%), followed by prescriptions that could be related to pregnancy symptoms or diagnoses. GPs prescribed medication at least once for 84.2% of non-pregnant women.


| 87


5

Table 5. Number of prescriptions and the three most frequently prescribed medications and referrals of pregnant and non-pregnant womenPregnant women n = 15,123 Non-pregnant women n = 102,564During pregnancy and postpartum period (336 days)


PrescriptionsNumber of prescriptions: Median (IQR) 1(0–3) 2 (1–5) Mean (SD) 2.1(3.5) 4.4(7.3)Women without a prescription 47.9% 15.8%Number of prescriptions in women who received a prescription: Median (IQR) 3(1–5) 3 (2–7) Mean (SD) 4 (4) 5.9(7.9)Most frequently prescribed medication per 100 womenAntibacterials for systemic use (J01) 16.4 Sex hormones and modulators of the

genital system (G03)20.8

Antianemic preparations (B03) 11.4 Antibacterials for systemic use (J01) 12.2Gynaecological anti-infectives and antiseptics (G01)

10.8 Anti-inflammatory and antirheumatic products (M01)

10.2

Most frequent referral to other healthcare professionals per 100 womenObstetrician/midwife 5.1 Physiotherapist 1.6Midwife 3.3 Obstetrician/midwife 1.0Physiotherapist 2.2 Dermatologist 0.9

Pregnant women were most frequently referred to the obstetrician/midwife in secondary care (7%), followed by midwives in primary care (4%). Non-pregnant women were most frequently referred to a physiotherapist (1.6%), followed by medical specialists.

DISCUSSION

Key resultsPregnant women contacted their GPs an average of 3.6 times during pregnancy and postpartum, in addition to the care provided by midwives or obstetricians. They had on average 1.4 more contacts with their GPs than non-pregnant women. The diagnoses made by GPs for pregnant women and non-pregnant women were quite similar. Most of the diagnoses recorded for pregnant women were related to pregnancy problems. However, differences appeared regarding urological problems and musculoskeletal problems. Urological problems were more often recorded with pregnant women, whereas non-pregnant women had musculoskeletal problems more often. The number of prescribed medications was much lower in pregnant women than non-pregnant women. Medication prescribed during pregnancy and postpartum mainly concerned pregnancy-related medication. Finally, pregnant women were most frequently referred to obstetrical healthcare professionals, whereas non-pregnant women were most frequently referred to physiotherapists.

R1R2R3R4R5R6R7R8R9


88 |

Chapter 5

InterpretationThis is the first study to examine the consultation rates, diagnoses and management of pregnant women in general practice. We found that, in addition to the maternal care provided by midwives and obstetricians, pregnant women have more contacts with their GP during pregnancy and the postpartum period compared to non-pregnant women. The excessive number of contacts related to pregnancy, birth and family planning could explain this difference. Pregnant women might be more worried about their health, resulting in a lower threshold for consulting their GP. Furthermore, pregnancy is a special period in which more health problems occur and extra care is required.14 However, we do not know why pregnant women choose to consult a GP instead of their obstetric care providers. This is remarkable because GPs in the Netherlands are not the key professionals providing obstetric care. Maybe, pregnant women are more familiar with their GP as compared to their midwife or obstetrician, or women do not have the knowledge to decide whether a symptom is related to pregnancy. Regarding obstetricians, difficulties in getting an appointment may also play a role. Finally, regarding midwives, some pregnant women may primarily contact their GP if they expect that they will need medication, which cannot be prescribed by their midwife.

Our finding that pregnant women consult their GPs frequently for problems unrelated to pregnancy is in agreement with Coco’s study8 that showed that GPs who provide prenatal care also address non-obstetrical problems frequently. Detailed comparisons of our results with those of Coco are not possible due to different classification systems (ICD-9 vs. ICPC-1), a different study setting (family physicians providing prenatal care vs. GPs not providing prenatal care), and different study variables (exclusion of pregnancy related diagnoses vs. inclusion of all diagnoses).

‘Pregnancy’ was recorded for 41% of all pregnant women. This could be interpreted as not every GP recording pregnancy in the electronic medical record, even though they may know about it. For instance, in the UK the percentage of women visiting a GP as the first professional seen during pregnancy is 82.5%.4 Obviously, every GP needs to know about a pregnancy and to have this recorded in the medical record: this information is indispensable when problems arise or medication has to be prescribed. On the other hand, midwives and obstetricians should inform GPs about their client’s pregnancy.

We found more diagnoses of cystitis in pregnant women, which confirms that cystitis occurs more frequently during pregnancy. The lower number of musculoskeletal problems of pregnant women presented to their GP could be explained by the commonly held belief among women that musculoskeletal symptoms are to a certain extent ‘normal’ during pregnancy and postpartum and that they need no special attention. In addition, GPs might have coded musculoskeletal problems in pregnant women under the pregnancy-related problems chapter of the ICPC (W); for instance code W29 ‘Pregnancy symptom/complaint other’ instead of musculoskeletal problems (ICPC chapter L).


| 89


5

Medication was less often prescribed to pregnant than to non-pregnant women, which reflects the justified reluctance to prescribe medication during pregnancy because of the potential teratogenic effects of medication use during pregnancy.

Finally, pregnant women were more often referred to other healthcare professionals compared to non-pregnant women. Although musculoskeletal problems were less frequently recorded in pregnant women, they were relatively frequently referred to a physiotherapist. Referring pregnant women to physiotherapists could replace a drug prescription.

Strengths and limitations of the studyA major strength of our study is the use of a very large and nationally representative dataset of contacts, prescriptions and referrals for both pregnant and non-pregnant women.

This study also has some limitations. First, the recording of data is not always complete, despite completeness being quite high. However, the amount of missing data did not differ between pregnant and non-pregnant women, making an impact on our findings unlikely. Evidently, regarding the diagnosis ‘pregnancy’, many more cases were missed (59%). We do not think that such an under-registration also holds for other diagnoses, which the GP considers to be less self-evident. Moreover, we had no data on other potentially relevant background characteristics such as ethnicity, making it impossible to assess ethnic subgroups. Third, a limitation could be that the pregnant women group had a higher proportion of women belonging to the low and high SES group than non-pregnant women. However, it is unlikely that we missed any women, as all Dutch inhabitants are listed at a general practice. This finding probably reflects that women are less likely to become pregnant in middle SES areas.

ImplicationsOur findings have implications for education, research and daily care. The apparently important role of GPs for pregnant women during their pregnancy should result in the training of GPs to recognise and manage health problems during pregnancy and obstetric emergencies.

Second, future research is needed to get more insight into the reasons pregnant women to seek GP care. A better understanding of the pregnant women’s perspective will enable all healthcare professionals involved to respond more appropriately to the needs of pregnant women.

Third, GPs have to participate in the obstetrical healthcare provider team, and may provide shared care as already occurs in some countries like Ireland.15

Collaboration and sharing of relevant information should be organised. An integrated digital environment can facilitate this communication. Software tools could be helpful for appropriately recording pregnancy in the electronic medical record, e.g. by prompting this on the occasion of a pregnancy test or a referral to a midwife or obstetrician.

R1R2R3R4R5R6R7R8R9


90 |

Chapter 5

CONCLUSIONS

Even where midwives and obstetricians are the key professionals in obstetrical healthcare, women consider their GPs as important care providers during pregnancy and postpartum. This indicates the need to involve GPs in the collaborative pregnancy and postpartum care.


| 91


5

REFERENCES

(1) Beeckman K, Louckx F, Putman K. Determinants of the number of antenatal visits in a metropolitan region. BMC Public Health 2010;10(1):527.

(2) Wiegers TA, Janssen BM. Monitor maternal health care (Monitor verloskundige zorgverlening; eindrapport). 2006.

(3) Kaczorowski J, Levitt C. Intrapartum care by general practitioners and family physicians. Provincial trends from 1984-1985 to 1994-1995. Can Fam Physician 2000 Mar;46:587-92, 595-7.

(4) Redshaw M, Rowe R, Hockley C, Brocklehurst P. Recorded delivery: a national survey of women’s experience of maternity care 2006. 2007.

(5) CHBB. GPs with special skills (College voor Huisartsen met Bijzondere Bekwaamheden). Available at: http://chbb.artsennet.nl/Home.htm. Accessed 02/06, 2012.

(6) Hingstman, L., Kenens, R.J. Figures of GPs registration (Cijfers uit de registratie van huisartsen; peiling 2011). 2011.

(7) Velden van der L, Hingstman L, Wiegers T, Kenens R. Obstetric care is still provided by GPs. (Huisartsenzorg in cijfers: verloskundig actieve huisarts bestaat nog steeds.). Huisarts en Wetenschap 2012;55(3):131.

(8) Coco A. How often do physicians address other Medical problems while providing prenatal care? Ann Fam Med 2009 /;7(2):134-138.

(9) Stirbu-Wagner I, Dorsman SA, Visscher S, Davids R, Gravestein JV, Abrahamse H, et al. Netherlands Information Network of General Practice. Facts and figures regarding GP-care in the Netherlands. (Landelijk Informatienetwerk Huisartsenzorg. Feiten en cijfers over huisartsenzorg in Nederland.). 2010; Available at: http://www.nivel.nl/oc2/page.asp?pageid=16198. Accessed 12/23, 2011.

(10) International Classification Committee of WONCA. International Classification of Primary care, second edition. Available at: http://www.globalfamilydoctor.com/wicc/edu.html. Accessed 02/06, 2012.

(11) Sociaal Cultureel Planbureau. From high to low, from low to high (Van hoog naar laag, van laag naar hoog). 2006; Available at: http://www.zorgatlas.nl/beinvloedende-factoren/sociale-omgeving/ses/sociaaleconomische-status-2006/#breadcrumb. Accessed 02/06, 2012.

(12) Reijneveld SA, Veenstra R, de Winter AF, Verhulst FC, Ormel J, de Meer G. Area Deprivation Affects Behavioral Problems of Young Adolescents in Mixed Urban and Rural Areas: The TRAILS Study. J Adolesc Health 2010 02/01;46(2):189-196.

(13) Statistics Netherlands. Urbanisation rate. 2012; Available at: http://www.cbs.nl/en-GB/menu/methoden/toelichtingen/alfabet/u/urbanisation-rate.htm. Accessed 02/06, 2012.

(14) National Institute for Health and,Clinical Excellence. Routine care for the healthy pregnant woman. Antenatal care 2011.

(15) Citizens Information Board. Citizens Information Ireland. Available at: http://www.citizensinformation.ie/en/health/women_s_health/maternity_and_infant_welfare_services.html.

CHAPTER 6

Determinants of use of care provided by complementary and

alternative healthcare practitioners to pregnant women in

primary care: a prospective cohort study

Esther I Feijen-de Jong, Danielle EMC Jansen, Frank Baarveld, Evelien Spelten, Francois Schellevis, Sijmen A Reijneveld

Submitted to BMC Pregnancy and Childbirth

R1R2R3R4R5R6R7R8R9


94 |

Chapter 6

ABSTRACT

Background: Pregnant women visit complementary/alternative healthcare practitioners in addition to regular maternal healthcare practitioners. A wide variation has been reported with regard to rates and determinants of use of complementary/alternative medicine (CAM), which may be due to heterogeneous populations. The aim of this study was to examine the prevalence and determinants of use of CAM practitioners by a homogeneous population of low-risk pregnant women in the Netherlands.

Methods: Data from the population-based DELIVER study was used, concerning 1,500 clients from twenty midwifery practices across the Netherlands in 2009 and 2010. CAM use was measured based on patient reports. Potential determinants were derived from Andersen’s behavioural model of healthcare utilization.

Results: The prevalence of CAM use by low-risk pregnant women was 9.4%. Women were more likely to use CAM if they had supplementary healthcare insurance (OR 3.11; CI 1.41-6.85), rated their health as ‘bad/fair’ (OR 2.63; CI 1.65-4.21), reported a chronic illness or handicap (OR 1.93; CI 1.14-3.27), smoked during pregnancy (OR 1.88; CI 1.06-3.33), or used alcohol during pregnancy (OR 2.30; CI 1.46-3.63).

Conclusions: CAM is relatively frequently used by low-risk pregnant women. Determinants revealed in this study diverge from other studies using heterogeneous populations. Maternal healthcare practitioners must be aware of CAM use by low-risk pregnant women and incorporate this knowledge into daily practice by actively discussing this subject with pregnant women.

Keywords: Healthcare utilization, Primary care, Maternal healthcare, Low-risk pregnancy, Complementary and Alternative Medicine (CAM)


| 95

Determinants of use of care provided by complementary and alternative healthcare practitioners to pregnant women in primary care: a prospective cohort study

6

INTRODUCTION

Pregnant women visit complementary/alternative healthcare practitioners in addition to regular maternal healthcare practitioners (e.g. obstetricians, midwives and GPs). Complementary/alternative medicine (CAM) - a group of diverse medical and healthcare systems, practices and products that are not traditionally considered part of conventional medicine1- may be used because pregnant women may be concerned about the potentially harmful effects of conventional medicine on their babies or it can be an expression of dissatisfaction with conventional medicine.2,3 A wide variation has been reported with regard to the rate of use of CAM during pregnancy, with ranges from 1% to 87% found in reviews.4,5 The available evidence on CAM use among pregnant women mostly covers women in a variety of settings, such as antenatal clinics, gynaecology wards, outreach clinics, local communities and birth clinics 4,5, which may explain this large variation in rates. Next to this, these studies mostly concern use of CAM (i.e.: herbal medicine, flower essence etc.) instead of use of CAM practitioners.

Documented determinants of CAM use by pregnant women include completion of tertiary-level education and the use of CAM prior to becoming pregnant4. Adams et al.5 also found that primiparous women, non-smoking women and women planning a natural birth were more likely to use CAM. Steel et al. reported that women who had either a vocational or university qualification were more likely to consult an acupuncturist.6 Similar to the evidence on prevalence, studies on the determinants of CAM use also concern heterogeneous populations, consisting of a combination of women with low-risk and high-risk pregnancies.

In the general population, it is known that poorer health status predicts CAM use.7 Since a low-risk pregnancy population in general consists of women who are not known to have any medical or obstetric risk factors before the onset of labour8, it can be hypothesized that such a specific low-risk population would use less CAM than high-risk or heterogeneous populations. The Dutch maternity healthcare system provides a very suitable setting to study a homogeneous low-risk pregnancy population. The system is divided into two echelons. In the first, midwives are the main care practitioners for pregnant women who have low-risk pregnancies. Of all pregnant women, 85.4% start care in primary midwifery care.9 Only when problems arise, are pregnant women referred to gynaecologists/obstetricians for secondary care. There is close mutual cooperation between these echelons.10 Regarding health insurance, basic health insurance is obligatory for all Dutch people; however, reimbursement of the costs for midwifery-led hospital births requires supplementary insurance.

Knowledge about CAM practitioner use by women with low-risk pregnancies is needed to fill a knowledge gap, to gain insight into the healthcare needs of and potential risks encountered by pregnant women, i.e. issues that maternal healthcare practitioners must

R1R2R3R4R5R6R7R8R9


96 |

Chapter 6

take seriously.3 Furthermore, CAM practitioner use can influence women’s health decision-making during pregnancy which requires focused attention of other maternal healthcare providers.6

Although it is frequently assumed that CAM use ‘will do no harm’11, evidence regarding safety and efficacy does not fully confirm this.12 For example, naturopaths, a type of CAM provider can recommend herbal medicine which, in case reports, has been linked to adverse foetal outcomes.13 Also, Steel et al. reported that consuming herbal teas is associated with a higher likelihood of medical removal of placenta/blood clots.14 On the other hand CAM can be valuable for pregnant women, for instance, Viljoen et al. concluded in a systematic review that ginger can be considered an option for women suffering pregnancy-associated nausea.15

The aim of this study was to examine the prevalence and determinants of use of CAM practitioners by low-risk pregnant women in the Netherlands. We used Andersen’s behavioural model of determinants of healthcare utilization as a guiding framework to categorize these determinants. This model suggests that the use of healthcare services depends on predisposing, enabling, need and health behaviour factors.16

METHODS

Study designData for this analysis was obtained from the DELIVER study (Dutch acronym for ‘data primary care delivery’) conducted by the Department of Midwifery Science of VU University Medical Center, Amsterdam. The DELIVER study is a descriptive study that aimed to provide information about the organization of midwifery care, the accessibility of midwifery care and the quality of primary midwifery care in the Netherlands.17

Participants, setting and procedureIn the DELIVER study, a two-stage sampling procedure was used. Firstly, midwifery practices were recruited by using purposive sampling. Three stratification criteria were used: region (north, east, south, west), level of urbanisation (urban or rural area), and practice type (dual or group practice) to ensure that different types of practices in different regions were represented. Subsequently, all clients receiving care in the participating primary midwifery practices at any moment in a 12 month study period in 2009-2010 were eligible to participate if they were able to understand Dutch, English, Turkish or Arabic. The participating practices (20 of the 519 midwifery practices in the Netherlands) comprised 110 midwives and a caseload of 8,200 clients per year, with all regions of the Netherlands being represented.17

Clients participating in the DELIVER study completed up to three questionnaires. The first questionnaire was administered before 34 weeks of gestation, the second between


| 97


6

34 weeks of gestation and birth, and the third in the postpartum period. In addition, information was collected about the care provided by midwives by extracting data from electronic client records of participating clients and from the Netherlands Perinatal Registry. The latter consists of information provided by midwives, GP’s and obstetricians. Reporting to this Registry is obligatory. The three data sources were linked using unique, anonymous client and midwifery practice identifiers.17 The Medical Ethics Committee of VU University Medical Center, Amsterdam approved the study protocol of the DELIVER study. Participants provided written informed consent.

Our study comprised those pregnant women who filled in the first and third questionnaires (up to 13 weeks postpartum) and whose questionnaire data could be linked to the electronic client record data and the Netherlands Perinatal Registry data. To maximize the homogeneity of the low-risk population, we excluded women who were referred to secondary care during pregnancy. Women who were referred during labour were classified as non-referred because the pregnancies of these women were low-risk (Figure 1). All women filled in the questionnaires at home without interference from a professional. We used data from electronic client records with regard to two independent variables; 1. Healthcare utilization in midwifery practices (Figure 2, health behaviours: ‘healthcare utilization in primary midwifery care’), 2. Parity (Figure 2, need variables: ‘parity’). These variables were shown to be invalidly measured in the client questionnaires. We assume that midwives recorded visits to their practice and parities of women more validly.

R1R2R3R4R5R6R7R8R9


98 |

Chapter 6

14,640 Women receiving care in primary

midwifery practices

DELIVER participants:7,907 women in primary

midwifery-led care

Women included in our sample (n=1,500)

Non-responders (n=6,733), information for 912 women regarding reasons of non-respons: miscarriage (25.9%), not interested (15.8%),language barrier (11.7%)

Women excluded with reasons (n=6,407);Did not fill in questionnaire 1 and 3 & questionnaire 3 ≤ 13 weeks (n=5,544)No data known from clientrecords (due to linkingproblems) (n=93)Women referred to secondary care in the pregnancy period (n=770)

Elig

ible

popu

latio

nD

ELIV

ERco

hort

Star

ted

in p

rim

ary

mid

wife

ry-le

dca

re


MeasurementsCAM practitioner use was measured by two items in the third questionnaire of the DELIVER study: ‘Please indicate whether you have seen any of the following practitioners of complementary or alternative medicine since the beginning of your pregnancy’ and ‘What other practitioner(s) of complementary or alternative medicine did you see?’ For each practitioner, women had to specify contact rates in predefined categories (0, 1-3, 4-6, 7-9, 10-12, 13-15, >15 meetings). Various types of CAM practitioners were stated explicitly in the questionnaire: acupuncturist, anthroposophical practitioner, homeopath, manual therapist (chiropractor, osteopath, manual therapist), naturopath (diet therapy, neural therapy, herbal therapy) or paranormal practitioner (psychic, faith healer, magnetic therapist), and respondents also had the option to choose other alternative practitioner. Women who reported at least one consultation with a CAM practitioner were defined as CAM users.

Potential determinants of CAM practitioner use concerned predisposing, enabling, need and health behaviour variables. Data on possible determinants were obtained from the first questionnaire and the electronic client records. Several variables, based on Andersen’s model16, were considered to be potential determinants of CAM practitioner use. In the


| 99


6

Andersen model use of health services depends on individual and contextual characteristics, and on health behaviour. The following components were measured: predisposing, enabling, need, and health behaviour characteristics. Predisposing characteristics are existing conditions that predispose people to use (yes/no) healthcare services. Enabling/disabling characteristics facilitate or impede use. Need characteristics are conditions that patients or health providers recognize as requiring medical treatment. Health behaviour characteristics are behaviours on the part of the individual that influence health status.16 Potential determinants were categorized into one of these components by using existing literature of the Andersen’s model and by discussion of the authors.

Operationalizations of the independent variables are shown in Figure 2.Predisposing variables encompassed socio-demographic and belief factors, consisting

of age, ethnicity, marital status, occupation, educational level, intended place of delivery, and religion. Enabling variables included finance (healthcare insurance) and organization (accessibility of care) variables. Regarding health insurance, we distinguished between basic and supplementary healthcare insurance.

Need variables comprised the health status of the client. The descriptive component of EuroQol (EQ) was used to measure self-reported health status.18 This component asked the respondent to consider and rate her actual health on five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Responses to questions on each of these dimensions can take one of five values, which concern five levels of severity (no problems/slight problems/moderate problems/severe problems/extreme problems). Health status values ranged from extreme problems on all five dimensions (value = -0.109) to no problems on any dimension (value = 1.0). A single health status value was calculated by applying scores from a UK valuation set.18 We then dichotomized the scores as ‘poor’ (lowest quartile) and ‘the remainder’. Feelings towards pregnancy were measured by using the Pregnancy Related Anxiety Questionnaire (PRAQ).19 The scales used were ‘fear of giving birth’ (two items), ‘fear of bearing a handicapped child’ (four items) and ‘concern about one’s appearance’ (three items). Items were scored on a four-point scale (4 = very true, 3 = true, 2 = not true, 1 = certainly not true). BMI was calculated using the weight and height before pregnancy reported by the respondent. We classified BMI according to the World Health Organization classification of adult underweight, normal weight, overweight and obesity.20 Every item score was dichotomized based on the median score. Finally, we computed a variable ‘gravidity/parity difference’ which measured the difference between the number of pregnancies and the number of deliveries. We hypothesised that there could be a difference in prenatal healthcare use between women with miscarriage(s) and/or abortion(s) in their obstetric history.

Health behaviour variables consisted of questions related to smoking, soft and hard drug use, alcohol use, adequate folic acid use, locus of control and adequacy of prenatal

R1R2R3R4R5R6R7R8R9


100 |

Chapter 6

healthcare utilization of primary midwifery care. We did not include drug use because none of the pregnant women reported drug use, which concurs with our sampling of low-risk pregnancies.17 The locus of control was measured by a single question about the extent of the perceived possibility of influencing lifestyle and/or health behaviour (‘To what extent do you feel that you can influence your health by changing your lifestyle and/or behaviour?’). Folic acid use was labelled as adequate when started at least four weeks before pregnancy.21 Adequacy of prenatal healthcare utilization in primary midwifery care was measured using the Kotelchuck Index, which is widely used in the US.22 We constructed a revised assessment index of the adequacy of prenatal care use in Dutch primary midwifery care (Table 1), modified according to the guidelines of the Royal Dutch Organization of Midwives, concerning the number of prenatal visits during pregnancy. This index combines the timing of initial prenatal healthcare and the number of prenatal healthcare visits. Prenatal care entry regarded on the gestational age at the first prenatal visit and classified into ‘timely’ (gestational age at onset < 12 weeks) and ‘late’ (gestational age at onset ≥ 12 weeks). The number of prenatal visits was derived from the electronic client record, and compared to the ‘expected’ number of visits as described by the Dutch prenatal guideline for primary midwifery care taking the gestational age at which women gave birth into account. For women who were referred to secondary care, the gestational age at the date of referral was used to calculate the ‘expected’ number of visits. Adequacy of prenatal healthcare utilization was trichotomized into ‘adequate plus’, ‘adequate’ and ‘inadequate’ (inadequate and intermediate) care.


| 101


6

Predisposing variablesDemographic:Age: ≤20/21-35/≥36Social:Ethnicity: Native Dutch/Western Non-Dutch/Non-WesternMarital status: married or living together/living aloneOccupation: employed/unemployed/disabledEducational level: low (one to secondary education) middle (to pre-university education) high (bachelor’s degree to post-graduate degree)Beliefs:Intended place of delivery: home/hospital birth centre with own midwife/hospital consultant-led Religion: yes/no

Need variablesPerceived healthGeneral self-rated health: excellent-very good/good/fair-bad Control of health situation(EQ-5d): no control of health situation/in control of health situation Chronic illnesses, disabilities or disorders: yes/noFeelings towards pregnancy (PRAQ-R):fear/no fearPlanned/wanted pregnancy: planned and wanted/unplanned but wanted/unplanned and unwantedEvaluated healthBMI: underweight/normal weight/ overweight/ obesityParitya: nullipara/primipara, multiparaGravidity/parity differencea: difference=1/difference≥2

Enabling variablesFinancing:Health insurance: basic/supplementaryNet household income: low income (≤ €2000)/high income (> €2000)

Organization:Accessibility of care: * difficulties getting through when calling during/outside business hours (yes/no)* difficulties getting to and from the midwifery practice (yes/no)

Health behavioursLocus of control: yes/noFolic acid use: adequate/inadequate/noSmoking: yes/noPassive smoking: yes/noAlcohol use: yes/noHealthcare utilization in primary midwifery carea:Inadequate/Adequate/Adequate plus

Use of Complementary and Alternative Medicine

a Electronic client record

Figure 2: Conceptual framework; the behavioural model of Andersen, which shows the possible determinants of healthcare utilization

R1R2R3R4R5R6R7R8R9


102 |

Chapter 6

Table 1. Index for assessment of the adequacy of prenatal care utilization in the Dutch primary midwifery care context (A.W. Boerleider and E.I. Feijen-de Jong, submitted)Duration of gestation (completed weeks and days of pregnancy, respectively)

Initiation of care Number of visits Kotelchuck index2

0-11+61 ≤ 11+6 ≥ 3 41-2 3

20 1

12-26+6 ≤ 11+6 ≥ 6 4Ideally 3.75 visits1 3-5 3

2 2≤ 1 1

≥ 12+0 127-36+6 ≥ 10 4Ideally 7.5 visits1 6-9 3

4-5 2≤ 3 1

≥ 12+0 137+0-37+6 ≤ 11+6 ≥ 13 4Ideally 11 visits1 10-12 3

6-9 2≤ 5 1

≥ 12+0 138+0-38+6 ≤ 11+6 ≥ 14 4Ideally 12 visits1 10-13 3

6-9 2≤ 5 1

≥ 12+0 139+0-39+6 ≤ 11+6 ≥ 15 4Ideally 13 visits1 11-14 3

7-10 2≤ 6 1

≥ 12+0 140+0-40+6 ≤ 11+6 ≥ 16 4Ideally 14 visits1 12-15 3

7-11 2≤ 6 1

≥ 12+0 141+0-41+6 ≤ 11+6 ≥ 17 4Ideally 15 visits1 12-16 3

8-11 2≤ 7 1

≥ 12+0 1

1According to the guidelines of the Royal Dutch Organization of Midwives 2 Kotelchuck Index:1. Inadequate (received less than 50% of expected visits)2. Intermediate (50%-79%)3. Adequate (80%-109%)4. Adequate Plus (110% and more)


| 103


6

Statistical analysesFirst, we described the background characteristics of the study population, and second the prevalence of CAM practitioner use. Third, we performed univariable logistic regression analyses for all determinants. Next, we performed multivariable logistic regression with a backward selection procedure, i.e. stepwise deletion of the variables that contributed least to the model that predicts use of CAM practitioners until all remaining variables contributed significantly at p<0.05 level. The results are presented as odds ratios (ORs) and 95% confidence intervals (CI). Women reporting no use of a CAM practitioner were our reference group. The structure of the data was hierarchical, i.e. respondents were clustered by midwifery practice. Characteristics of practices may affect all women who received care in that practice, which might lead to dependency of data regarding women coming from the same practice.23 To adjust for this potential clustering, multilevel analytical methods were used. A two-tailed p-value of 0.05 or lower was considered statistically significant. Missing data accounted for less than 1.5% of all variables, with the exception of 6.5% for BMI. SPSS 21.0 (SPSS Inc., Chicago, IL) was used for all analyses.

RESULTS

Our study population included 1,500 women with low-risk pregnancies. Table 2 shows the potential determinants and the rate of CAM practitioner use of these women in primary midwifery care. Regarding background variables, the majority of the pregnant women were between 21-35 years of age (85.5%), native Dutch (88.5%), married (97.8%), employed (84.3%) and higher educated (i.e. having a bachelor’s degree or higher) (55.2%). Of all the women, 9.4% reported having consulted a CAM practitioner.

Table 3 shows the distribution of CAM use by pregnant women receiving midwifery care per type of CAM practitioner. Manual therapists were visited most frequently (4.1%), followed by acupuncturists (1.9%). CAM practitioners were mostly visited 1-3 times, except for acupuncturists (4-6 times).

Table 4 shows the associations of predisposing, enabling, need and health behaviour characteristics with use of CAM. Regarding enabling characteristics, our analyses showed that women with supplementary healthcare insurance were three times more likely to visit a CAM practitioner compared to women with only basic healthcare insurance (adjusted OR = 3.11; 95% CI 1.41-6.85; see Table 4). With respect to need variables, women who rated their health as ‘bad/fair’ were 2.6 times more likely to visit a CAM practitioner compared to women who rated their health as ‘good’. Furthermore, women who reported a chronic illness or handicap were more likely to visit a CAM practitioner than women reporting no chronic illness or handicap (OR = 1.93). Regarding health behaviour variables, women who smoked (compared to non-smokers), and women using alcohol during pregnancy (compared to non-drinking women) were more likely to visit a CAM practitioner.

R1R2R3R4R5R6R7R8R9


104 |

Chapter 6

Table 2. Use of a complementary/alternative medicine practitioner (CAM) by low-risk pregnant women in primary midwifery care (N = 1500)

Consultation of a CAM practitionerYes No

Background characteristics N = 1500 (%) N = 141 (9.4) N = 1359 (90.6)Age ≤ 2021-35≥ 36Missing

14 (0.9)1283 (85.6)

202 (13.5)1

1 (0.7)118 (83.7)

22 (15.6)0

13 (1.0)1165 (85.7)

180 (13.3)1

Ethnicity Native DutchNon-WesternWestern Non-DutchMissing

1327 (88.6)65 (4.3)

106 (7.1)2

121 (85.8)6 (4.3)

14 (9.9)0

1206 (88.9)59 (4.3)92 (6.8)

2

Marital status Married or living togetherLiving aloneMissing

1467 (97.8)33 (2.2)

0

137 (97.2)4 (2.8)

0

1330 (97.9)29 (2.1)

0Occupation EmployedUnemployedDisabledMissing

1264 (84.3)220 (14.7)

16 (1.1)0

117 (83.0)22 (15.6)

2 (1.4)0

1147 (84.4)198 (14.6)

14 (1.0)0

Educational level LowMiddleHighMissing

164 (10.9)508 (33.9)828 (55.2)

0

11 (7.8)42 (29.8)88 (62.4)

0

153 (11.3)466 (34.3)740 (54.4)

0

Intended place of delivery Hospital/birth centre midwifery-ledHospital consultant-ledHomeMissing

856 (57.1)19 (1.3)

625 (41.7)0

85 (60.3)3 (2.1)

53 (37.6)0

771 (56.7)16 (1.2)

572 (42.1)0

ReligionNoYesMissing

845 (57.2)631 (42.8)

24

84 (60.0)56 (40.0)

1

761 (57.0)575 (43.0)

23Basic and supplementary healthcare insurance Basic and supplementaryBasicMissing

1307 (87.4)188 (12.6)

5

134 (95.0)7 (5.0)

0

1173 (86.6)181 (13.4)

5Net household incomea > €2000< €2000Missing

1082(72.2)170(11.3)

0

110 (78.0)11 (7.8)

0

972 (71.6)159 (11.7)

0Accessibility of care (phone)ProblemsNo problemsMissing

252 (16.8)1248 (83.2)

0

25 (17.7)116 (82.3)

0

227 (16.7)1132 (83.3)

0


| 105


6

Consultation of a CAM practitioner Yes NoAccessibility of care (getting to and from the practice)ProblemsNo problemsMissing

65 (4.3)1435 (95.7)

0

6 (4.3)135 (95.7)

0

59 (4.3)1300 (95.7)

0

General self-rated health Excellent/Very goodBad/FairGoodMissing

538 (35.9)181 (12.1)781 (52.1)

0

45 (31.9)35 (24.8)61 (43.3)

0

493 (36.3)146 (10.7)720 (53.0)

0

Quality of life (EuroQol) Poor health statusGood health statusMissing

313 (20.9)1187 (79.1)

0

44 (31.2)97 (68.8)

0

269 (19.8)1090 (80.2)

0Chronic illnesses or handicaps YesNoMissing

127 (8.5)1373 (91.5)

0

23 (16.3)118 (83.7)

0

104 (7.7)1255 (92.3)

0PRAQ-ChildFearNo fearMissing

508 (33.9)989 (66.1)

3

50 (35.5)91 (64.5)

0

458 (33.8)898 (66.2)

3PRAQ-DeliveryFearNo fearMissing

19 (1.3)1480 (98.7)

1

3 (2.1)138 (97.9)

0

16 (1.2)1342 (98.8)

1PRAQ-BodyFearNo fearMissing

418 (27.9)1079 (72.1)

3

43 (31.9)96 (68.1)

0

373 (27.5)983 (72.5)

3Planned and wantedness of pregnancyb

Wanted, not plannedPlanned and wantedMissing

231 (15.4)1268 (84.6)

1

24 (17.1)117 (83.3)

0

207 (15.2)1151 (84.4)

1BMI ≤ 18.525-< 30≥ 3018.5-< 25Missing

40 (2.7)274 (18.3)

72 (4.8)1017 (67.8)

97

6 (4.3)26 (18.4)

5 (3.5)96 (68.1)

8

34 (2.5)248 (18.2)

67 (4.9)921 (67.8)

89

Parity Primi/multiparousNulliparousMissing

653 (43.5)847 (56.5)

0

58 (41.1)83 (58.9)

0

595 (43.8)764 (56.2)

0Difference between number of pregnancies and number of births ≥ 21Missing

359 (24.2)1127 (75.8)

14

38 (27.1)102 (72.9)

1

321 (23.8)1025 (76.2)

13

R1R2R3R4R5R6R7R8R9


106 |

Chapter 6

Table 2. ContinuedConsultation of a CAM practitioner Yes NoLocus of controlNoYesMissing

174 (11.6)1325 (88.4)

1

13 (9.2)128 (90.8)

0

161 (11.9)1197 (88.1)

1Folic acid useNoYes, inadequatelyYes, adequatelyMissing

110 (7.3)683 (45.6)705 (47.1)

2

9 (6.4)70 (49.6)62 (44.0)

0

101 (7.4)613 (45.2)643 (47.4)

2

Smoking YesNoMissing

108 (7.2)1392 (92.8)

0

17 (12.1)124 (87.9)

0

91 (6.7)1268 (93.3)

0Passive smokingYesNoMissing

173 (11.5)1326 (88.5)

0

10 (7.1)131 (92.9)

0

80 (5.9)1279 (94.1)

0Alcohol useYesNoMissing

173 (11.5)1326 (88.5)

1

30 (21.4)110 (78.6)

1

143 (10.5)1216 (89.5)

0Healthcare utilization in Midwifery PracticeInadequateAdequate plusAdequateMissing

384 (25.6)95 (6.3)

1021 (68.1)0

39 (27.7)6 (4.3)

96 (68.1)0

345 (25.4)89 (6.5)

925 (68.1)0

a Missings in a third category (Prefer not to say)b Category ‘not wanted, not planned’removed due to empty cells

Table 3. Distribution of complementary/alternative medicine use by pregnant women who receive primary midwifery care (most occurring category, mode and range of frequency of use) per type of CAM provider (N = 1,500)

CAM providers Number of women (%)

Mode of frequency of consultation (if visiting)

Range (if visiting)

Acupuncturist 28 (1.9) 4-6 1-3, >15Anthroposophical practitioner 6 (0.4) 1-3 1-3, >15Homeopath 24 (1.6) 1-3 1-3, 10-12Manual therapist* 62 (4.1) 1-3 1-3, >15Naturopath 8 (0.5) 1-3 1-3, 7-9Paranormal practitioner 8 (0.5) 1-3 1-3, >15Other alternative practitioner** 29 (1.9) 1-3 1-3, >15*Osteopath, chiropractor, manual therapist**For example: shiatsu therapy, reflexology, Ayurvedic Medicine, iridology, haptonomy, kinesiology, or Analytical-Synthetical Response therapy


| 107


6

Table 4. Associations of predisposing, enabling, need and health behaviour characteristics with use of CAM (N = 1500): odds ratios (OR) and 95% confidence intervals (CI)

Crude OR (95% CI) Adjusted OR (95% CI)a

Predisposing variablesAge (years)≤ 20 ≥ 36 21-35

0.78 (0.10-6.06)1.15 (0.71-1.88)

1.00 (ref.)Ethnicity Non-westernWestern non-DutchNative Dutch

1.01 (0.42-2.40)1.43 (0.79-2.61)

1.00 (ref.)Marital statusLiving alone Married or living together

1.28 (0.44-3.72)1.00 (ref.)

Occupation UnemployedDisabled Employed

1.14 (0.70-1.86)1.46 (0.32-6.60)

1.00 (ref.)Educational levelMiddleHighLow

1.25 (0.63-2.50)1.62 (0.84-3.12)

1.00 (ref.)Intended place of delivery Hospital/birth centre midwifery-ledHospital consultant-ledHome

1.19 (0.83-1.71)1.99 (0.56-7.10)

1.00 (ref.)ReligionNoYes

1.11 (0.77-1.60)1.00 (ref.)

Enabling variablesBasic and supplementary healthcare insurance Basic and supplementaryBasic

2.92 (1.34-6.36)1.00 (ref.)

3.11 (1.41-6.85)1.00 (ref.)

Net household income > €2000< €2000

1.65 (0.86-3.14) 1.00 (ref.)

Accessibility of care (phone)ProblemsNo problems

1.05 (0.66-1.66)1.00 (ref.)

Accessibility of care (getting to and from the practice)ProblemsNo problems

1.04 (0.44-2.48)1.00 (ref.)

Need variablesGeneral self-rated health Excellent/Very goodBad/FairGood

1.07 (0.72-1.60)2.81 (1.78-4.43)

1.00 (ref.)

1.29 (0.91-1.82)2.63 (1.65-4.21)

1.00 (ref.)

R1R2R3R4R5R6R7R8R9


108 |

Chapter 6

Table 4. ContinuedCrude OR (95% CI) Adjusted OR (95% CI)a

Need variables (continued)Quality of life (EuroQol) Poor health statusGood health status

1.84 (1.26-2.70)1.00 (ref.)

Chronic illnesses or handicaps YesNo

2.36 (1.44-3.87)1.00 (ref.)

1.93 (1.14-3.27)1.00 (ref.)

PRAQ*-ChildFearNo fear

1.08 (0.75-1.56)1.00 (ref.)

PRAQ*-DeliveryFearNo fear

1.78 (0.51-6.24)1.00 (ref.)

PRAQ*-BodyFearNo fear

1.26 (0.86-1.84)1.00 (ref.)

Planned and wantedness of pregnancy**Wanted, not plannedPlanned and wanted

1.18 (0.74-1.89)1.00 (ref.)

BMI ≤ 18.525-< 30≥ 3018.5-< 25

1.67 (0.68-4.09)1.02 (0.64-1.61)0.73 (0.29-1.85)

1.00 (ref.)

Parity Primi/multiparousNulliparous

1.11 (0.78-1.58)1.00 (ref.)

Difference between number of pregnancies and number of births ≥ 21

1.15 (0.77-1.71)1.00 (ref.)

Health behaviour variablesLocus of controlNoYes

0.77 (0.42-1.40)1.00 (ref.)

Folic acid useNoYes, inadequatelyYes, adequately

0.91 (0.44-1.89)1.20 (0.83-1.72)

1.00 (ref.)


| 109


6

Table 4. ContinuedCrude OR (95% CI) Adjusted OR (95% CI)a

Health behaviour variables (continued)Smoking YesNo

1.88 (1.08-3.27)1.00 (ref.)

1.88 (1.06-3.33)1.00 (ref.)

Passive smokingYesNo

1.22 (0.62-2.42)1.00 (ref.)

Alcohol useYesNo

2.28 (1.46-3.56)1.00 (ref.)

2.30 (1.46-3.63)1.00 (ref)

Healthcare utilization in Midwifery PracticeInadequateAdequate plusAdequate

1.16 (0.76-1.76)0.66 (0.28-1.55)

1.00 (ref.)

a = Corrected for all other variables in the adjusted model, predictors were considered in the final model if p-value was < 0.05 *PRAQ = Pregnancy Related Anxiety Questionnaire **category ‘not wanted, not planned’ removed due to empty cells

DISCUSSION

We assessed the prevalence and determinants of CAM practitioner use of low-risk pregnant women in primary midwifery practices in the Netherlands, and found a prevalence of 9.4% CAM practitioner use. Low-risk pregnant women were more likely to visit a CAM practitioner if they had supplementary healthcare insurance, if they rated their health as ‘bad/fair’, if they reported a chronic illness or handicap, if they smoked during pregnancy, and if they used alcohol during pregnancy.

InterpretationWe found a rate of almost 10% of pregnant women consulting a CAM practitioner, fitting into the range reported in the literature.5,24 We expected to find a lower prevalence compared to that in the general female population because of the composition of our research population consisting of low-risk pregnant and as a consequence, healthy women. However, contrary to our expectations, we found a higher prevalence than among the general population of women in the Netherlands (9.4% vs 7.5%).25 This difference might indicate the relatively large need of pregnant women for additional care besides regular pregnancy care, which might be related to women having concomitant health problems affected by pregnancy (i.e nausea, back problems).

In accordance with the general female population in the Netherlands, osteopaths and chiropractors were the most consulted CAM practitioners in our study.25 An explanation may be that musculoskeletal problems are common in pregnancy, varying in severity from mild to severe26, leading to an increased use of these manual therapists. In addition, pregnant women may presume that manual therapy potentially provides a safe alternative to pain medication during pregnancy, for example in the case of low back and pelvic pain.27,28

Regarding need variables, we found that women rating their health as ‘bad/fair’ and women reporting a chronic illness were more likely to visit a CAM practitioner. This also holds for the general population with chronic illnesses in the Netherlands.25 It is possible that women with chronic illnesses look for comfort measures or symptom management which they cannot find in conventional medicine or midwifery care.27,29

Regarding health behaviour variables, we found an association between smoking, alcohol use and CAM practitioner use. This finding conflicts with findings from international studies on this topic.30 However, these studies mostly concerned non-pregnant women. In pregnant women CAM practitioner use may be a coping strategy reflecting the intention to stop drinking and/or smoking.14 Moreover, these health behaviour adjustments in pregnancy can cause prenatal psychological distress, which also is associated with CAM use.29,31

Surprisingly, we did not find any significant associations of predisposing variables with CAM practitioner use. In the literature, a higher educational level has been shown to be associated with CAM use.6,32 It is assumed that higher educational level may encourage the development of critical thinking, which may lead to the appraisal of healthcare options that lie outside conventional care.1 Our descriptives (Table 2) show that 62.4% of highly educated women visit a CAM practitioner as compared to 54.4% of non-users. However, when we controlled for many variables this difference was not statistically significant. We have no indication that the educational level of Dutch women differs substantially from women in other countries. An explanation may be that the variation in educational level among CAM users was not large enough to establish a statistically significant association. Bishop et al.32 found that older mothers were more likely to consult a CAM practitioner compared to younger mothers. However, the effect size of this association seems to be small. In our research there was a slight difference between CAM users and non-CAM users (15.6% vs. 13.2%) in the group of women aged 36 years and over. When we corrected for many other variables, age was not significantly related to CAM use.

Strengths and limitationsOne strength of this research is the use of a unique and large sample of women who had low-risk pregnancies. This allowed us to carry out the study in a homogeneous population in primary midwifery care. Furthermore, we used data of a large study population, which covered all components of the behavioural model of Andersen.


| 111


6

Our study population included slightly more highly educated and native-Dutch women compared to the general Dutch population.25 However, educational level and ethnic background were not significantly associated with the outcome measures. Potentially, recall bias may have occurred due to the timing of the completion of the third questionnaire for the DELIVER database. However, we attempted to reduce this risk by only including women who had filled in this questionnaire up to 13 weeks postpartum. Moreover, it seems likely that most women will remember whether they visited a CAM practitioner or not. Finally, self-report can create bias due to social desirability. However, pregnant women could fill in the questionnaire on CAM use with confidentiality, which might decrease non-disclosure because of the absence of a potentially judgmental healthcare provider.

ImplicationsThe results of this research indicate that CAM use is relatively high even in a low-risk population of pregnant women. This raises the question of how maternal care providers can become more aware of CAM use by their clients. Midwives and obstetricians must be attentive to CAM use. We know that non-disclosure can occur for different reasons.11 Therefore, we advise midwives/obstetricians to actively ask their clients whether they have contacted a CAM practitioner at every scheduled consultation. Furthermore, our findings reflect the need for informing and collaborative care approaches by all practitioners involved in the care of the same pregnant woman.

Our research shows that it is necessary for midwives to learn about CAM, which is not commonly included in midwifery education.33 This may consist of acquiring knowledge about CAM and learning how to identify safety issues regarding maternal healthcare. In addition, midwives have to encourage pregnant women to make use of professional bodies and voluntary registers if considering using CAM. For instance, in Great Britain this would be the Complementary and Natural Healthcare Council (CNHC).

Research challenges concern, specifically, understanding the reasons, attitudes and beliefs of low-risk women who consult CAM practitioners. Why do pregnant women consult CAM practitioners in addition to regular pregnancy care practitioners? Which CAM practitioners are mostly consulted? Is it used as a supplement to or a substitute for traditional care? Maternal healthcare practitioners can use this information to better meet the needs of pregnant women.29

CONCLUSION

CAM is relatively frequently used in a sample of low-risk pregnant women. The determinants of this use as revealed in this study diverge from those found in other studies using more heterogeneous populations. Maternal healthcare practitioners must become more aware of

R1R2R3R4R5R6R7R8R9


112 |

Chapter 6

CAM practitioner use and incorporate this knowledge into daily practice, actively discussing this subject with pregnant women.

ACKNOWLEDGEMENTS We would like to thank Esther Halma and Esther Blok for their valuable contribution to this study.


| 113


6

REFERENCES

(1) Frawley J, Adams J, Sibbritt D, Steel A, Broom A, Gallois C. Prevalence and determinants of complementary and alternative medicine use during pregnancy: Results from a nationally representative sample of Australian pregnant women. Australian and New Zealand Journal of Obstetrics and Gynaecology 2013;53(4):347-352.

(2) Strouss L, Mackley A, Guillen U, Paul DA, Locke R. Complementary and Alternative Medicine use in women during pregnancy: do their healthcare providers know? BMC Complement Altern Med 2014 Mar 4;14(1):85-6882-14-85.

(3) Mitchell M. Risk, pregnancy and complementary and alternative medicine. Complementary Therapies in Clinical Practice 2010 5;16(2):109-113.

(4) Hall HG, Griffiths DL, McKenna LG. The use of complementary and alternative medicine by pregnant women: A literature review. Midwifery 2011 12;27(6):817-824.

(5) Adams J, Lui CW, Sibbritt D, Broom A, Wardle J, Homer C, et al. Women’s use of complementary and alternative medicine during pregnancy: a critical review of the literature. Birth 2009 Sep;36(3):237-245.

(6) Steel A, Adams J, Sibbritt D, Broom A, Gallois C, Frawley J. Determinants of women consulting with a complementary and alternative medicine practitioner for pregnancy-related health conditions. Women Health 2014;54(2):127-144.

(7) Thomson P, Jones J, Browne M, Leslie SJ. Why people seek complementary and alternative medicine before conventional medical treatment: A population based study. Complementary Therapies in Clinical Practice 2014;in press(0).

(8) Birthplace in England Collaborative Group, Brocklehurst P, Hardy P, Hollowell J, Linsell L, Macfarlane A, et al. Perinatal and maternal outcomes by planned place of birth for healthy women with low risk pregnancies: the Birthplace in England national prospective cohort study. BMJ 2011 Nov 23;343:d7400.

(9) PRN foundation. Netherlands Perinatal Registry. 2013.(10) De Geus M. Midwifery in the Netherlands. 2012.(11) Thomson P, Jones J, Evans JM, Leslie SL. Factors influencing the use of complementary and

alternative medicine and whether patients inform their primary care physician. Complement Ther Med 2012 0;20(1–2):45-53.

(12) Holst L, Wright D, Haavik S, Nordeng H. Safety and efficacy of herbal remedies in obstetrics—review and clinical implications. Midwifery 2011 2;27(1):80-86.

(13) Lim A, Cranswick N, South M. Adverse events associated with the use of complementary and alternative medicine in children. Arch Dis Child 2011 Mar;96(3):297-300.

(14) Steel A, Adams J, Sibbritt D, Broom A, Frawley J, Gallois C. Relationship between complementary and alternative medicine use and incidence of adverse birth outcomes: An examination of a nationally representative sample of 1835 Australian women. Midwifery 2014 Mar 29.

(15) Viljoen E, Visser J, Koen N, Musekiwa A. A systematic review and meta-analysis of the effect and safety of ginger in the treatment of pregnancy-associated nausea and vomiting. Nutr J 2014 Mar 19;13:20-2891-13-20.



(18) Herdman M, Gudex C, Lloyd A, Janssen M, Kind P, Parkin D, et al. Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Qual Life Res 2011 Dec;20(10):1727-1736.

R1R2R3R4R5R6R7R8R9


114 |

Chapter 6

(19) Huizink AC, Mulder EJ, Robles de Medina PG, Visser GH, Buitelaar JK. Is pregnancy anxiety a distinctive syndrome? Early Hum Dev 2004 Sep;79(2):81-91.

(20) WHO. Obesity and Overweight. 2014; Available at: http://www.who.int/mediacentre/factsheets/fs311/en/. Accessed 07/16, 2014.

(21) Health Council of the Netherlands. Towards an optimal use of folic acid. 2008;2008/02E.(22) Kotelchuck M. The Adequacy of Prenatal Care Utilization Index: Its US distribution and association

with low birthweight. Am J Public Health 1994;84(9):1486-1489.(23) Goldstein H, Browne W, Rasbash J. Multilevel modelling of medical data. Stat Med 2002 Nov

15;21(21):3291-3315.(24) Hall HG, McKenna LG, Griffiths DL. Midwives’ support for Complementary and Alternative

Medicine: A literature review. Women and Birth 2012 3;25(1):4-12.(25) Available at: http://www.cbs.nl/nl-NL/menu/themas/gezondheid-welzijn/publicaties/artikelen/

archief/2014/2014-4041-wm.htm.(26) Keriakos R, Bhatta SRC, Morris F, Mason S, Buckley S. Pelvic girdle pain during pregnancy and

puerperium. J Obstet Gynaecol 2011 10/01; 2014/08;31(7):572-580.(27) Close C, Sinclair M, Liddle SD, Madden E, McCullough JE, Hughes C. A systematic review

investigating the effectiveness of Complementary and Alternative Medicine (CAM) for the management of low back and/or pelvic pain (LBPP) in pregnancy. J Adv Nurs 2014 Mar 9.

(28) Wang SM, DeZinno P, Fermo L, William K, Caldwell-Andrews AA, Bravemen F, et al. Complementary and alternative medicine for low-back pain in pregnancy: a cross-sectional survey. J Altern Complement Med 2005 Jun;11(3):459-464.

(29) Thorne S, Paterson B, Russell C, Schultz A. Complementary/alternative medicine in chronic illness as informed self-care decision making. Int J Nurs Stud 2002 9;39(7):671-683.

(30) Al-Windi A. Determinants of complementary alternative medicine (CAM) use. Complement Ther Med 2004 0;12(2–3):99-111.

(31) Furber CM, Garrod D, Maloney E, Lovell K, McGowan L. A qualitative study of mild to moderate psychological distress during pregnancy. Int J Nurs Stud 2009 May;46(5):669-677.

(32) Bishop JL, Northstone K, Green JR, Thompson EA. The use of Complementary and Alternative Medicine in pregnancy: data from the Avon Longitudinal Study of Parents and Children (ALSPAC). Complement Ther Med 2011 Dec;19(6):303-310.

(33) Steel A, Adams J. Developing midwifery and complementary medicine collaboration: The potential of interprofessional education? Complementary Therapies in Clinical Practice 2012 11;18(4):261-264.

CHAPTER 7

General discussion and implications

R1R2R3R4R5R6R7R8R9


116 |

Chapter 7


| 117


7

The aim of this thesis is to contribute to knowledge of the prenatal healthcare use of pregnant women receiving primary midwifery care and on its determinants. Two perspectives were studied, namely the use of care offered within prenatal care programmes and ancillary care use. Prenatal care programmes are based on professional guidelines and mostly concern prevention. Ancillary care is care provided alongside care from maternal healthcare providers (in this thesis the main maternal healthcare provider is a primary care midwife). With respect to the use of care offered within prenatal care programmes, the following aims were formulated:1. To provide a systematic review of the evidence on the determinants of prenatal

healthcare use in high-income countries.2. To examine the determinants of inadequate prenatal healthcare use by low-

risk women in primary midwifery-led care in the Netherlands, and to determine whether these differ from those referred to prenatal secondary care.

3. To compare prenatal healthcare use in Belgium and the Netherlands with differently designed pregnancy care systems, as measured using the Content and Timing of care in Pregnancy (CTP) tool, and to identify their predisposing, enabling and need (pregnancy-related) determinants.

With respect to the use of ancillary care, the following aims were formulated:4. To compare GPs consultation rates, diagnoses and healthcare management for

pregnant women with those for non-pregnant women in the Netherlands.5. To examine the prevalence and the determinants of CAM use by low-risk pregnant

women in the Netherlands. This chapter presents the main findings. In addition, we will discuss the main findings, strengths and limitations of this thesis, followed by its implications for practice, education and further research.

MAIN FINDINGS

Use of care offered within prenatal care programmesAim 1 (Chapter 2): To provide a systematic review of the determinants of inadequate prenatal healthcare use in high-income countriesWe found that high quality evidence of the determinants of prenatal care use was limited, with only eight studies included in this review meeting the quality assessment criteria. These studies identified a number of individual determinants related to inadequate prenatal care use (i.e. entry after the first pregnancy trimester and/or an inadequate number of prenatal visits): low maternal age, low education level, unmarried status, ethnic minority status, planning care with a GP/midwife/midwifery team or hospital consultant compared to shared care, planning care in an urban teaching hospital compared to an urban non-

R1R2R3R4R5R6R7R8R9


118 |

Chapter 7

teaching hospital, not planning a place of delivery, lack of insurance, high parity, no previous premature birth, and late recognition of pregnancy. Contextual determinants associated with inadequate use were living in neighbourhoods with high rates of unemployment, being part of a single-parent family, having a medium-average family income, living in an area with low average education levels, and women reporting Canadian Aboriginal status. Regarding health behaviour, inadequate use was more likely among women who smoked during pregnancy.

Aim 2 (Chapter 3): To examine determinants of inadequate prenatal healthcare use by low-risk women in primary midwifery-led care in the Netherlands, and to determine whether these differ from those who are referred to prenatal secondary care.We found a prevalence of 24.7% for inadequate prenatal care use (i.e. the combination of care entry after twelve weeks gestation and an insufficient number of visits) among low-risk pregnant women in Dutch primary midwifery-led care. Overall, women of non-Western origin (compared to native Dutch women), unemployed women, women reporting chronic illnesses or disabilities, and women who did not use folic acid periconceptionally were more likely to use prenatal care inadequately. Women not referred to secondary care during pregnancy were more likely to use prenatal care inadequately if they intended to deliver in a hospital, if they did not use folic acid periconceptionally, and if they were exposed to cigarette smoke during pregnancy. Women referred to secondary care were more likely to use prenatal care inadequately if they reported chronic illnesses or disabilities, and did not use folic acid periconceptionally.

Aim 3 (Chapter 4): To compare prenatal healthcare use in Belgium and the Netherlands with differently designed pregnancy care systems, as measured by the Content and Timing of care in Pregnancy (CTP) tool, and to identify its predisposing, enabling and (pregnancy-related) need determinants. We found that women residing in the Netherlands used prenatal care adequately more often (58.3%) than women residing in Belgium (45.5%). Furthermore, 5.7% of Dutch women and 9.7% of Belgian women used prenatal care inadequately. In this study, adequacy of use of prenatal care was defined based on a combination of characteristics regarding entry to care and the content of care. However, regardless of country of residence, inadequate prenatal care content and timing were associated with lower education levels, unemployment, lower continuity of care provider and non-attendance at prenatal classes.


| 119


7

Use of ancillary care during pregnancyAim 4 (Chapter 5): To compare GP consultation rates, diagnoses and management for pregnant women and non-pregnant women in the Netherlands.We found that pregnant women contacted their GP on average 3.6 times, compared to 2.2 times for non-pregnant women of the same age range during their pregnancy. The most frequently recorded diagnoses were ‘pregnancy’ (40.8%, indicating that the pregnancy was not recorded in 59.2% of pregnant women) and ‘cystitis/urinary infection’ (8.7%) for pregnant women, and ‘cystitis/urinary infection’ (4.0%) and ‘general disease not otherwise specified’ (3.6 %) for non-pregnant women. The mean number of prescribed medications was lower in pregnant women than non-pregnant women (2.1 against 4.4). For pregnant women, the most frequently occurring referral indication concerned obstetric care, for non-pregnant women this concerned physiotherapy.

Aim 5 (Chapter 6): To examine the prevalence and determinants of use of complementary and alternative medicine (CAM) by low-risk pregnant women in the Netherlands.We found a prevalence of 9.4% for CAM use among low risk pregnant women. They were more likely to use CAM if they had supplementary healthcare insurance, if they rated their health as ‘bad/fair’, if they had reported chronic illnesses or handicaps, if they smoked during pregnancy and if they used alcohol during pregnancy.

DISCUSSION OF THE MAIN FINDINGS

Use of care offered within prenatal care programmesPrenatal care programmes are organized based on professional guidelines which are determined by evidence and consensus among professionals: adequacy of prenatal care use as measured in this thesis depends on definitions derived from these guidelines. Adequacy of use of care within prenatal care programmes was measured in this thesis according to two definitions. Chapter 3 used timing and the number of prenatal care visits controlling for gestational age to measure the adequacy of prenatal care. A revised version of the Kotelchuck index was constructed, according to the prenatal care guidelines of the Royal Dutch Organization of Midwives.1,2 In Chapter 4 ‘adequacy’ not only referred to initiation of care but also to receiving a minimal package of interventions and their timely application throughout the pregnancy.

In Chapter 3 we reported that a relatively high percentage (24.7%) of low-risk pregnant women in the Netherlands did not make timely use of care and/or did not receive the number of prenatal care visits recommended by the Dutch prenatal care guidelines for primary midwifery care. Using a different definition, as described in Chapter 4, we found that 58.3% of women used care as recommended in the guidelines on prenatal care programmes. We expected that the prevalence of adequate prenatal care use would be higher than we found

R1R2R3R4R5R6R7R8R9


120 |

Chapter 7

in these two studies because prenatal care is provided free of cost, and is safe, preventative and non-invasive. After all, use of prenatal care is an important determinant of maternal health.3,4 Timely and adequate use of prenatal care has been shown to be effective in reducing the likelihood of adverse pregnancy outcomes.3-6

Our findings raise the question of what the underlying mechanisms are for this inadequate use. Many reasons are mentioned in the literature from the perspective of pregnant women. Phillippi7 reported in a review that the most common barriers for women in the US to prenatal care access concern transportation problems, lack of motivation to obtain care,a lack of money, conflicting needs of existing children and structural barriers.b There may also be other barriers in the Netherlands, such as lack of familiarity with the maternal healthcare system, language problems and a lack of awareness of being pregnant. Though, in the Netherlands, research is lacking about the barriers and facilitators related to adequate prenatal care use.

Provider-related factors could contribute to inadequate prenatal care use, in addition to client-related factors. After all, it is midwives who are tasked with informing pregnant women about the prenatal care programme and inviting them for follow-up consultations. Perhaps our finding reflects how midwives adjust the number of visits to the preferences and wishes of pregnant women. Moreover, midwives themselves could experience barriers, which lead to fewer prenatal visits. However, to our knowledge, evidence of provider-related factors in relation to the adequacy of prenatal care use is unavailable.

The relatively high prevalence of inadequate prenatal care use we found could cast doubt on the evidence base for the number of prenatal cares visits within the Dutch prenatal care programme. However, scientific evidence regarding the optimal number of prenatal visits is lacking. In a Cochrane Review, Dowswell et al.8 compared the effects of prenatal care programmes with fewer visits for low-risk women with standard care. They found that a reduction of the number of prenatal visits to around eight did not lead to higher perinatal mortality compared to a regular prenatal care programme (caveat: the power of the studies included was limited).8 However, there was evidence that pregnant women were less satisfied with a lower number of prenatal visits due to too long a period being left between consultations. Reducing the number of prenatal visits in the Netherlands would not be desirable without rigorous evaluation of such changes and specifically monitoring perinatal mortality and morbidity rates for specific groups of women in specific regions (e.g. deprived areas).8-10 After all, a prenatal care programme is a complex intervention in which midwives are expected to adjust care to pregnant women related to the course of their pregnancies.1

a Lack of awareness of pregnancy, considering abortion, depression, hiding pregnancy, belief pregnancy is unnecessary or fear of medical procedures. 7 b Structural barriers consist of barriers related to clinics (e.g. location, hours, delay of initial appointment, not child-friendly facility or staff attitudes) and barriers related to providers (poor communication skills, cultural sensitivity, language issues or lack of a consistent provider). 7


| 121


7

The prevalences we found for inadequate prenatal healthcare use could reflect how midwives interpret the prenatal care programme guidelines. Implementation of guidelines into clinical practice is difficult. Gaps are often reported between research evidence and actual clinical practice.11

Determinants of inadequate useThe determinants we found for inadequate prenatal care use vary in the studies reported in this thesis, with many more determinants having been identified in the systematic review than in the other studies. Our systematic review (Chapter 2) found a wide range of determinants associated with inadequate prenatal healthcare use. However, due to internationally distinct maternal healthcare systems, different definitions of inadequate prenatal healthcare use and heterogeneous research populations (i.e. mixed high and low-risk pregnancies), it would not be legitimate to generalize these findings to primary midwifery care in the Netherlands, which is focused on low-risk pregnant women. Therefore, we performed a prospective cohort study in the Netherlands to discover whether the determinants are the same or different from countries with other maternal healthcare systems and settings (Chapter 3). We found that inadequate prenatal healthcare use is associated with a limited set of determinants in a low-risk setting. In Chapter 4 we found different determinants associated with healthcare use compared to the determinants found in Chapter 3. In both studies unemployed women were more likely to use healthcare inadequately. Unemployment (measured by occupational status) is linked to a wide range of health problems12 including low birth weight.12-14 Behaviour and psychosocial problems related to unemployment can be explanatory factors which contribute to the inadequate use of prenatal care.15

In the Dutch low-risk setting we found the following determinants to be associated with inadequate prenatal care use. Women of non-Western origin (compared to native Dutch women), unemployed women, women reporting chronic illnesses or disabilities, and women who did not adequately use folic acid periconceptionally were more likely to use prenatal care inadequately. De Graaf et al.10 showed that living in deprived areasc (partly determined by the population’s employment status) and being of non-Western origin are associated with higher perinatal mortality. The determinants we found overlap with the findings of De Graaf et al.10, which suggests that it would be worth exploring a possible relationship of inadequate use with perinatal mortality further.

c Deprived areas were selected using eighteen indicators consisting of SES, state of housing, safety problems, noise disturbance problems, pollution etc. (Ministry of Housing, Spatial Planning and the Environment).

R1R2R3R4R5R6R7R8R9


122 |

Chapter 7

Ancillary care useWe measured ancillary healthcare use, i.e. the use of care during pregnancy which is not part of a prenatal care programme, with respect to care provided by GPs and by CAM providers (Chapters 5 and 6). We found that pregnant women also substantially accessed GPs and CAM providers along with visits to midwives. On average, pregnant women visited GPs 3.6 times during pregnancy and the postpartum period, and almost ten percent of low-risk pregnant women visited a CAM provider. We can assume that in reality, ancillary care use will be higher because we did not measure the healthcare use of other care providers besides GPs and CAM providers (e.g. physiotherapists, dentists or psychologists etc.).

This thesis shows that ancillary care use exists and is substantial, which could lead to a series of problems inherent to fragmented care, such as conflicting advice from multiple healthcare professionals on pregnancy problems or complications, information loss, or even errors and the receipt of more interventions than are necessary.16 We assume that coordination of care can reduce these risks. At the moment, the Dutch government, professional organizations in primary and secondary maternity care, and maternal healthcare providers are heading towards integrated maternity care, specifically aimed at integrated care between primary care professionals (midwives) and secondary care professionals (gynaecologists).17 However, our research shows that there also has to be a focus on integrated care within primary care. Inter-professional collaboration is not easy to achieve and is a complex process.18 Schölmerich et al.19 analysed inter-professional coordination problems with regard to primary and secondary maternal care in the Netherlands. They identified five causes for poor coordination/cooperation: guidelines which do not facilitate shared care, a financial reimbursement system which does not provide incentives for cooperation, a lack of a shared maternity record system, non-proximity of care providers, and different perspectives on pregnancy and professionals speaking different ‘languages’.19 Although we cannot directly apply the knowledge of Schölmerich et al. to primary care givers, we can assume that the sources of problems mentioned can also underlie coordination problems within primary care.

Our results raise the questions why pregnant women use ancillary care in addition to regular prenatal care and which factors contribute to this use. We do not know whether this ancillary healthcare use is additional or substitutional. Additional ancillary care use can overburden the healthcare system, which could result in high costs to society. Substitutional ancillary care further highlights the need for inter-professional collaboration.

METHODOLOGICAL CONSIDERATIONS

We used data provided by midwives and GPs in this thesis. We also used data from questionnaires completed by pregnant women in primary midwifery practices. The GP data


| 123


7

were retrieved from the Netherlands Information Network of General Practice (LINH). LINH collects structured data about contacts, prescriptions and referrals. To ensure a high level of accuracy, GPs receive instructions and manuals about recording information as participants of the LINH. Unfortunately, possible explanatory variables of healthcare utilization could not be included in the analyses due to these data being unavailable. Therefore, we could not study the determinants of healthcare use in general practice.

The DELIVER study was used for primary midwifery care data. DELIVER is a Dutch acronym meaning ‘data primary care delivery’. This study was the first large-scale study to evaluate the quality and provision of primary midwifery care. Therefore, we were able to study a large cohort of low-risk pregnant women from twenty midwifery practices located across the Netherlands. These data were linked to those from the Netherlands Perinatal Register and to electronic client records kept by midwives, which created client and provider perspectives with regard to primary midwifery care. The DELIVER database was used in the studies described in Chapters 3, 4 and 6 of this thesis. In Chapters 3 and 4 the main outcome, i.e. prenatal healthcare use, was measured using the client records of midwives. This reduces the risk of errors related to recall bias. CAM use was measured in Chapter 6 with a questionnaire completed by the clients of midwifery practices. Clients were allowed to complete this questionnaire without the presence of a potentially judgmental healthcare provider, which probably decreased non-disclosure. The response rate for the questionnaire was 62 percent, which is an acceptable response rate20, though there may have still been some selection bias. In particular, clients from ethnic minority groups were underrepresented (17%, compared to 25% of the Dutch national female population aged between 15 and 45 in 2010)21, which could have led to selection bias.

With respect to the systematic review reported on in Chapter 2, we extracted data from three databases: PubMed, CINAHL and Embase. Furthermore, we did not restrict ourselves to studies published in English. This contributed to a broad and comprehensive search. However, we did not review grey literature and did not explore bibliographies, so we may have missed some relevant studies.

IMPLICATIONS

Implications for practiceWe found that many pregnant women visit a midwife less frequently than they should on the basis of professional guidelines or entry care after the first trimester. Professional organizations should be aware of this and evaluate their professional guidelines to ensure they remain current with respect to Dutch primary midwifery care. The perspectives of women and healthcare providers can help create a prenatal care programme which meets the needs of both. Furthermore, professional organizations need to consider whether

R1R2R3R4R5R6R7R8R9


124 |

Chapter 7

prenatal care programmes should be adjusted to specific groups of women who are more likely to use prenatal care inadequately.

Midwives need to be aware of our findings and should identify women who deviate from professional guidelines because this could be a potential determiner for safety risk.22 They should preferably accurately record the reasons women report for using prenatal care inadequately. Such a registration could contribute to knowledge of the barriers women experience in pregnancy care, which could result in interventions tailored to pregnant women at risk of using prenatal care inadequately. Moreover, this is also required because of statutory regulations. Furthermore, the determinants associated with inadequate healthcare use overlap with the determinants associated with higher perinatal mortality.10 Therefore, midwives have to be extra alert to these specific groups of women.

We found substantial use of ancillary care showing that coordination of care is needed. To further strengthen collaboration in primary care, proximity – i.e. joint housing of GPs and midwives – can be helpful. At the moment this is rarely done.23 Face-to-face contact between professionals using existing collaboration initiatives such as the Perinatal Care Partnerships (Verloskundig SamenwerkingsVerband) or Regional Support Frameworks (Regionale Ondersteunings Structuur) could also support primary care workers in developing teamwork and improving the continuity of care.24 Coordination could also be improved through an electronic patient record system capable of integrating client-centred information for primary (GPs and midwives) and secondary care providers. Furthermore, although there are joint guidelines between GPs and midwives, the participation of GPs in the development of new midwifery guidelines should be expanded.

We found that women frequently contact GPs in the pregnancy and postpartum period. Midwives are the main care providers for low-risk pregnant women,25 and must therefore actively ask pregnant women at every consultation if they have also visited another healthcare provider in order to coordinate care and to identify the problems pregnant women face. Midwives and GPs complement each other and should inform and/or involve each other when taking decisions about pregnant women.

Implications for educationWith respect to prenatal care programmes, student midwives should be aware of the fact that many women visit midwives less frequently than the Dutch midwifery guidelines recommend. During their training, students should learn how to motivate pregnant women to adequately use prenatal care. Teaching a client-centred approach could support this effectively.

With respect to ancillary care use, joint training of GPs and midwives should be organized to contribute to the integration of primary care. For example, midwives could learn from GPs and vice versa during their internships. Learning communities could also strengthen


| 125


7

collaboration. Furthermore, the important role GPs play for pregnant women should mean that GPs are trained to recognise and manage health problems during pregnancy and obstetric emergencies. Finally, knowledge of the aims and treatments of ancillary care should be included in midwifery training. Non-disclosure of CAM can happen for various reasons.26 Therefore, student midwives and midwives in general should be vigilant for CAM use.

Implications for researchIrrespective of the definition used, we found that many low-risk pregnant women visit a midwife less frequently than recommended by professional guidelines. Future research should focus on finding evidence for a prenatal care programme with an optimal number of prenatal visits. Cut-offs for definitions of inadequate prenatal care should also be considered in the development of this prenatal care programme. Additional research is also needed to improve our understanding of the healthcare needs of Dutch low-risk pregnant women by listening to these women and learning what barriers, facilitators and resistance they encounter related to prenatal care use. Research into the reasons, attitudes and beliefs of low-risk pregnant women with regard to entry and obtaining prenatal care will help overcome this knowledge gap. Knowledge of the facilitators, barriers, and resistance could also help care researchers develop interventions which could improve entry into and use of prenatal care.

Future research is also needed on whether midwives are able to comply with the existing prenatal care programme guidelines, and if not, why they choose to provide different care, as we also found that pregnant women frequently do not receive the interventions recommended by guidelines. Furthermore, we found that the determinants of inadequate healthcare use overlap with determinants associated with higher perinatal mortality and morbidity rates. Further research into this somewhat alarming issue is required.

More evidence is needed on the use of ancillary healthcare by pregnant women. This includes both research into the potential risks of ancillary care use and into whether this use is additional or substitutional care.

CONCLUSIONS

A relatively high percentage of pregnant women do not use the amount and content of prenatal care offered within prenatal care programmes. This inadequate prenatal healthcare use seems to be associated with a limited set of determinants in low-risk pregnant women: women of non-Western origin (compared to native-born Dutch women), unemployed women, women reporting chronic illnesses or handicaps, and women who do not use folic acid periconceptionally were more likely to use prenatal care inadequately.

R1R2R3R4R5R6R7R8R9


126 |

Chapter 7

Moreover, pregnant women use a considerable amount of ancillary care. In addition to visits to their main maternal healthcare provider, pregnant women in the Netherlands also visit their GP frequently (on average 3.6 times), and ten percent of them consult a CAM practitioner.

These findings on the use of care offered within prenatal care programmes and additional ancillary care could have considerable implications for prenatal care practice, policy, education and research. They offer potential routes to improving care for women with low-risk pregnancies and the outcomes of their pregnancies.


| 127


7

REFERENCES

(1) Royal Dutch Organisation of Midwives (Koninklijke Nederlandse Organisatie van Verloskundigen). Prenatal care. Recommendations for support, interaction and counseling (Prenatale begeleiding. Aanbevelingen voor ondersteuning, interactie en voorlichting). 2008; Available at: http://www.knov.nl/docs/uploads/Kwaliteit_Richtlijnen_PrenataleVKBegeleiding.pdf. Accessed 02/14, 2012.

(2) Dutch Association of Obstetrics and Gynecology (Nederlandse Vereniging voor Obstetrie en Gynaecologie). Dating of pregnancy, version 1.1. 2011.


(4) Partridge S, Balayla J, Holcroft CA, Abenhaim HA. Inadequate prenatal care utilization and risks of infant mortality and poor birth outcome: a retrospective analysis of 28,729,765 U.S. deliveries over 8 years. Am J Perinatol 2012 Nov;29(10):787-793.

(5) Raatikainen K, Heiskanen N, Heinonen S. Under-attending free antenatal care is associated with adverse pregnancy outcomes. BMC Public Health 2007;7:268.

(6) Blondel B, Marshall B. Poor antenatal care in 20 French districts: risk factors and pregnancy outcome. J Epidemiol Community Health 1998;52:501-506.

(7) Phillippi JC. Women’s perceptions of access to prenatal care in the United States: a literature review. J Midwifery Womens Health 2009 May-Jun;54(3):219-225.

(8) Dowswell T, Carroli G, Duley L, Gates S, Gulmezoglu AM, Khan-Neelofur D, et al. Alternative versus standard packages of antenatal care for low-risk pregnancy. Cochrane Database Syst Rev 2010 Oct 6;(10):CD000934. doi(10):CD000934.

(9) Poeran J, Denktas S, Birnie E, Bonsel GJ, Steegers EA. Urban perinatal health inequalities. J Matern Fetal Neonatal Med 2011 Apr;24(4):643-646.

(10) de Graaf JP, Ravelli AC, Wildschut HI, Denktas S, Voorham AJ, Bonsel GJ, et al. Perinatal outcomes in the four largest cities and in deprived neighbourhoods in The Netherlands. Ned Tijdschr Geneeskd 2008 Dec 13;152(50):2734-2740.

(11) Prior M, Guerin M, Grimmer-Somers K. The effectiveness of clinical guideline implementation strategies--a synthesis of systematic review findings. J Eval Clin Pract 2008 Oct;14(5):888-897.

(12) Adler NE, Newman K. Socioeconomic disparities in health: pathways and policies. Health Aff (Millwood) 2002 Mar-Apr;21(2):60-76.

(13) Simoes E, Kunz S, Munnich R, Schmahl FW. Association between maternal occupational status and utilization of antenatal care Study based on the perinatal survey of Baden-Wuerttemberg 1998-2003. Int Arch Occup Environ Health 2006 Jan;79(1):75-81.

(14) Ross CE, Mirowsky J. Does employment affect health? J Health Soc Behav 1995 Sep;36(3):230-243.

(15) Niedhammer I, Murrin C, O’Mahony D, Daly S, Morrison JJ, Kelleher CC, et al. Explanations for social inequalities in preterm delivery in the prospective Lifeways cohort in the Republic of Ireland. The European Journal of Public Health 2012 07/24;22(4):533-538.

(16) Metcalfe A, Grabowska K, Weller C, Tough SC. Impact of prenatal care provider on the use of ancillary health services during pregnancy. BMC Pregnancy Childbirth 2013 Mar 11;13:62.

(17) Dutch Health Care Inspectorate. Perinatal Care Partnerships: Improved acute care, shortcomings with regard to preventative care (Verloskundige samenwerkingsverbanden: acute zorg veiliger, preventie is blijven liggen). 2014.

(18) Brownie S, Thomas J, McAllister L, Groves M. Australian health reforms: enhancing interprofessional practice and competency within the health workforce. J Interprof Care 2014 May;28(3):252-253.

R1R2R3R4R5R6R7R8R9


128 |

Chapter 7

(19) Scholmerich VL, Posthumus AG, Ghorashi H, Waelput AJ, Groenewegen P, Denktas S. Improving interprofessional coordination in Dutch midwifery and obstetrics: a qualitative study. BMC Pregnancy Childbirth 2014 Apr 15;14:145.

(20) Draugalis JR, Coons SJ, Plaza CM. Best practices for survey research reports: a synopsis for authors and reviewers. Am J Pharm Educ 2008 Feb 15;72(1):11.


(22) Martijn L, Jacobs A, Amelink-Verburg M, Wentzel R, Buitendijk S, Wensing M. Adverse outcomes in maternity care for women with a low risk profile in The Netherlands: a case series analysis. BMC Pregnancy Childbirth 2013 Nov 29;13:219.

(23) Kenens RJ, Hofhuis H, Hingstman L. A first overview of mono- and multidisciplinairy collaborative initiatives in primary care. (Inventarisatie mono- en multidisciplinaire samenwerkingsverbanden in de eerste lijn: een eerste verkenning). 2006.

(24) Schafer W, Kroneman M, Boerma W, van den Berg M, Westert G, Deville W, et al. The Netherlands: health system review. Health Syst Transit 2010;12(1):v-xxvii, 1-228.

(25) Beentjes MM, Weersma RLS, Koch W, Offringa AK, Verduijn MM, Mensink PAJS, et al. Guideline Dutch College of General Practitioners, Pregnancy and Postpartumperiod (NHG-Standaard Zwangerschap en kraamperiode, tweede herziening). 2012.

(26) Thomson P, Jones J, Evans JM, Leslie SL. Factors influencing the use of complementary and alternative medicine and whether patients inform their primary care physician. Complement Ther Med 2012 0;20(1–2):45-53.

CHAPTER 8

Summary

R1R2R3R4R5R6R7R8R9


130 |

Chapter 8


| 131

Summary

8

This thesis aims to contribute to our knowledge of the prenatal healthcare use of pregnant women receiving primary midwifery care and of its determinants. We assessed two types of health service use, namely use of care offered within prenatal care programmes (Chapter 2, 3, and 4) and ancillary care use (Chapter 5 and 6). Prenatal care programmes comprise care determined by professionals and mostly concern prevention. Ancillary care is care provided alongside care from a principal maternal healthcare provider.

Chapter 1 is the general introduction to this thesis. Descriptions of pregnancy care, prenatal healthcare use and the determinants of prenatal healthcare use are provided. Prenatal care use is an important determinant of maternal and foetal health. Timely and adequate prenatal care has been shown to be effective in reducing the likelihood of adverse pregnancy outcomes. Therefore, better use of prenatal healthcare could contribute to the prevention of adverse perinatal outcomes. In addition to the use of care offered within prenatal care programmes, the determinants of prenatal care use are studied to identify the characteristics of those pregnant women who underuse prenatal care. Furthermore, the use of other types of care is studied, e.g. care from GPs and CAM practitioners. We used the model proposed by Andersen as a theoretical framework. This model categorizes the determinants of seeking and receiving professional care, such as prenatal care, as characteristics of the patients themselves and of their context.

The design of the Dutch pregnancy care system creates an opportunity to study the use and determinants of prenatal care use in low-risk pregnant women. Dutch pregnancy care is distinct from almost all other high-income countries. Pregnancy care is split into primary and secondary care, similar to the overall organization of the Dutch healthcare system. For most women with uncomplicated pregnancies, primary care midwives provide routine prenatal, intrapartum and postpartum care, and act as gatekeepers to secondary obstetric care.

With respect to use of care offered within prenatal care programmes, the following aims were formulated:1. To provide a systematic review of the determinants of late and/or inadequate use

of prenatal healthcare in high-income countries.2. To examine the determinants of inadequate prenatal healthcare use by low-

risk women in primary midwifery-led care in the Netherlands, and to determine whether these differ for women who are referred to secondary prenatal care.

3. To compare prenatal healthcare use in Belgium and the Netherlands with differently designed pregnancy care systems, as measured by the Content and Timing of care in Pregnancy (CTP) tool and to identify its predisposing, enabling and pregnancy-related determinants.

R1R2R3R4R5R6R7R8R9


132 |

Chapter 8

With regard to use of ancillary care, the following aims were formulated:4. To compare GP consultation rates, diagnoses and care management for pregnant

women with those for non-pregnant women in the Netherlands.5. To examine the prevalence and determinants for CAM use by low-risk pregnant

women in the Netherlands.

Chapter 2 concerns a systematic review of the determinants of late and/or inadequate prenatal healthcare use in high-income countries. Three databases were searched, eight high-quality studies were included. Two of the authors independently screened, read and assessed all the potential studies. A narrative synthesis was prepared, since a quantitative synthesis was not possible due to the heterogeneity of the included studies.

Low maternal age (<20 years), low education level (<9 years), unmarried status, ethnic minority status (widely differing operationalisations), receiving planned care from a GP/midwife/midwifery team or hospital consultant compared to shared care, receiving planned care in urban teaching hospitals compared to urban non-teaching hospitals, unplanned delivery location, not having insurance, high parity, no previous premature births and late recognition of pregnancy were all identified as individual determinants of inadequate use. Several contextual determinants were also associated with inadequate healthcare use or entering care late (after six months). These included living in deprived areas (60% non-white and 30% low income), living in areas with high rates of unemployment, single parent families, being part of a medium-average income family, living in an area with low average education levels, and reporting Canadian Aboriginal status. Regarding health behaviour, inadequate use was more likely among women who smoked during pregnancy. We found that the evidence in terms of high quality studies of the determinants of prenatal care use was limited: only eight studies met the criteria for inclusion. None of the studies included Dutch data. Although all the studies included here assessed prenatal healthcare use, they employed twelve different definitions. Standardization is urgently needed to be able to integrate the findings.

In Chapter 3 we studied the determinants of inadequate prenatal care use by low-risk women in primary midwifery-led care in the Netherlands. The data for this study were obtained from the DELIVER study, conducted by the Department of Midwifery Science of the VU University Medical Center Amsterdam. Data came from a pregnancy questionnaire (<34 weeks of gestation), electronic client records and data from the Netherlands Perinatal Registry. Inadequate prenatal care use was measured with a revised version of the Kotelchuck Index modified for the prenatal care guidelines of the Royal Dutch Organization of Midwives. The index combines the timing of entry to care and the number of visits.


| 133

Summary

8

We included 3,070 pregnant women who started prenatal care in a primary midwifery practice at the beginning of their pregnancies. A prevalence of inadequate use of 24.7% was found, and 24.7% of the women included were also referred to secondary obstetric care. Overall, our results showed that women of non-Western origin (compared to native Dutch women), unemployed women, women reporting chronic illnesses or disabilities, and women who did not use folic acid periconceptionally were more likely to use prenatal care inadequately. Pregnant women who were not referred during pregnancy were more likely to use prenatal care inadequately if they intended to deliver at a hospital, if they did not use folic acid periconceptionally or if they were exposed to cigarette smoke during pregnancy. Among those who were referred to secondary care, women reporting a chronic illness or disability and women who did not use folic acid periconceptionally were more likely to use prenatal care inadequately. This study found that a relatively high percentage (24.7%) of low-risk pregnant women did not use care in time and/or did not make as many prenatal visits as recommended in the prenatal care guidelines. Furthermore, we found that inadequate prenatal healthcare use was associated with a limited set of determinants in a low-risk setting. Our results can be used to target interventions to women who are at risk of inadequate prenatal healthcare use.

Chapter 4 studies the content and timing of care during pregnancy for low-risk pregnant Dutch women in primary midwifery care compared to a sample of low-risk pregnant Belgian women. The data on the Dutch women were drawn from the DELIVER database. The data on the Belgian women were obtained from an observational study conducted in the Brussels Metropolitan Region. Low-risk pregnant women residing in an urban area were eligible for inclusion. A pooled dataset was constructed to be able to adjust for pre-existing differences between the two populations. Prenatal care use was operationalized using the Content and Timing of Pregnancy tool. This tool considers the timing of the initiation of care and the number and timing of three specific interventions during pregnancy (blood screening, ultrasound and blood pressure measurement). We included 642 women, consisting of 321 Belgian and 321 Dutch women.

We found that women residing in Belgium used prenatal care inadequately more often (9.7%) than Dutch women (5.6%). Irrespective of the country they were from, inadequate prenatal care content and timing was associated with lower education level, unemployment, lower continuity of care provider and non-attendance at prenatal classes. To our knowledge, this study is the first cross-border study measuring prenatal healthcare use and its determinants. Despite the value of this study, more cross-border studies are required to examine other potential determinants of prenatal care use. Systematic and routine data collection on pregnant women is required.

R1R2R3R4R5R6R7R8R9


134 |

Chapter 8

Chapter 5 examined GP consultation rates, diagnoses and care management for pregnant women compared to non-pregnant women in the Netherlands. Longitudinal data from the Netherlands Information Network of General Practice was used. This register holds longitudinal data on the consultations, prescriptions and referrals for all the patients listed at 84 practices in the Netherlands in the period 2007–2009. We included 15,123 pregnant women and 102,564 non-pregnant women of the same age range (aged 15–45).

Pregnant women contacted their GP on average 3.6 times, compared to 2.2 times for non-pregnant women. The most frequently recorded diagnoses were ‘pregnancy’ and ‘cystitis/urinary infection’ for pregnant women, and ‘cystitis/urinary infection’ and ‘general disease not otherwise specified’ for non-pregnant women. The mean number of prescribed medications was lower in pregnant women compared to non-pregnant women (2.1 against 4.4). For pregnant women, the most frequent referral indication concerned obstetric care; for non-pregnant women this concerned physiotherapy. GP consultation rates in pregnancy and postpartum showed that GPs are important providers of care for pregnant women. Therefore, the involvement of GPs in collaborative care during pregnancy and postpartum should be reinforced.

Chapter 6 concerns the prevalence and determinants of use of CAM practitioners by low-risk pregnant women in the Netherlands. Longitudinal data from the DELIVER study was used. CAM use was measured using patient-reported data from the third DELIVER questionnaire. Women who reported at least one consultation with a CAM practitioner were defined as CAM users.

We found a CAM use prevalence of 9.4%. Low-risk pregnant women were more likely to visit a CAM practitioner if they had supplementary healthcare insurance, if they rated their health as ‘bad/fair’, if they had reported chronic illnesses or handicaps, if they smoked during pregnancy, and if they used alcohol during pregnancy. We concluded that CAM is relatively frequently used in a sample of low-risk pregnant women. The determinants we found in this study differ from those found in other studies: these included more heterogeneous pregnant populations. Maternal healthcare providers must become more aware of CAM practitioner use and incorporate this knowledge into daily practice, actively discussing this subject with pregnant women.

Chapter 7 summarizes and discusses the main findings of this thesis, and addresses its methodological considerations and the implications for practice, education and future research.

We found that a relatively high percentage of low-risk pregnant women did not use care in time and/or did not get the right amount of prenatal care as prescribed by midwifery guidelines. We discussed whether this could be related to barriers that pregnant women


| 135

Summary

8

perceive, or to barriers midwives experience, or to midwives adapting their care to the preferences of the pregnant women. Our results imply that future research should focus on finding evidence for a prenatal care programme which would determine the optimal number of prenatal visits and which would include the needs of pregnant women and the demands of healthcare providers with respect to the content of prenatal care.

Compared to the number of determinants we found in a systematic literature review on pregnant women in high-income countries, we found only a limited set of determinants associated with inadequate prenatal healthcare use in low-risk pregnant women in the Netherlands. These determinants overlap with previously identified determinants of higher perinatal mortality. This suggests that there could be an association between perinatal mortality and care use. Research is needed to disentangle this possible association.

Ancillary healthcare use through GPs and CAM providers is substantial. This shows a need to coordinate care to prevent loss of information, and the receipt of more interventions than medically necessary. Midwives are the central care providers for low-risk pregnant women. However, GPs also have an important task in the pregnancy and postpartum period, which exposes a need to strengthen collaboration between midwives and GPs, e.g. through the use of joint electronic files, joint training, joint guidelines and actual proximity (i.e. collocation) of GPs and midwives.

The findings on the use of care offered within prenatal care programmes and additional ancillary care could have considerable implications for practice, policy, education and research into prenatal care. They could help improve care for women with low-risk pregnancies and the outcomes of their pregnancies.

CHAPTER 9

Nederlandse samenvatting

R1R2R3R4R5R6R7R8R9


138 |

Chapter 9


| 139


9

Dit proefschrift beoogt bij te dragen aan de kennis over prenataal zorggebruik van zwangere vrouwen die eerstelijns verloskundige zorg ontvangen en de determinanten van prenataal zorggebruik. Het proefschrift gaat over twee soorten zorggebruik, te weten gebruik van zorg aangeboden binnen het prenatale zorgschema (hoofdstuk 2, 3 en 4) en aanvullend gebruik van zorg in de zwangerschap (hoofdstuk 5 en 6). Het prenatale zorgschema bestaat uit de zorg op basis van richtlijnen die zijn bepaald door professionals en betreft veelal preventie. Aanvullende zorg is zorg die zwangeren gebruiken naast het prenatale zorgschema.

Hoofdstuk 1 betreft de algemene inleiding van dit proefschrift. In dit hoofdstuk wordt uitleg gegeven over zwangerschapszorg, prenataal zorggebruik, en de determinanten van dit zorggebruik. Gebruik van prenatale zorg is een belangrijke determinant van foetale en maternale gezondheid. In het algemeen wordt verondersteld dat tijdige zorg en een adequaat aantal prenatale consulten bewezen effectief is om de kans op slechte zwangerschapsuitkomsten te verminderen. Naast het gebruik van prenatale zorg zijn de determinanten van dit zorggebruik onderzocht om kenmerken te identificeren van zwangeren die minder zorg gebruiken dan aanbevolen in de richtlijn. Ook is onderzocht of zwangere vrouwen, naast de zorg die ze krijgen van verloskundige zorgverleners, aanvullende zorg hebben gebruikt bij de huisarts en bij alternatieve zorgverleners. Het model van Andersen is als theoretisch kader gebruikt. Dit model verklaart het zorggebruik aan de hand van drie domeinen. In het model wordt een context gecreëerd waarin individuele karakteristieken en contextuele karakteristieken het zorggebruik beïnvloeden.

Het Nederlandse verloskundige systeem is anders georganiseerd dan in de meeste andere hoge inkomenslanden. Hierdoor is het mogelijk zorggebruik van prenatale zorg en de determinanten hiervan bij laag-risico zwangeren te onderzoeken. De gezondheidszorg in Nederland is opgesplitst in eerste- en tweedelijns zorg, dit geldt ook voor de verloskundige zorg. De eerstelijns verloskundige zorg voor laag-risico zwangeren wordt in de meeste gevallen verzorgd door een verloskundige. Zij heeft een poortwachtersfunctie ten aanzien tot de toegang tot de tweedelijns verloskundige zorg (gynaecologen).De volgende onderzoeksvragen worden beantwoord in deze thesis.Met betrekking tot het prenatale zorgschema:1. Wat zijn determinanten van te laat en/of ontoereikend prenataal zorggebruik in

geïndustrialiseerde landen?2. Wat is de prevalentie en wat zijn de determinanten van ontoereikend prenataal

zorggebruik van laag-risico zwangeren in de eerstelijns verloskundige praktijk in Nederland? Zijn er verschillen tussen vrouwen die verwezen zijn naar tweedelijns verloskundige zorg en vrouwen die niet verwezen zijn tijdens de zwangerschap?

3. Is er een verschil in prenataal zorggebruik tussen België en Nederland, landen met verschillend opgezette zorgsystemen, gemeten met het CTP meetinstrument

R1R2R3R4R5R6R7R8R9


140 |

Chapter 9

dat het moment van starten met prenatale zorg meet in combinatie met inhoudelijke zorgindicatoren? Wat zijn de predisponerende, faciliterende en zwangerschapgerelateerde determinanten hiervan?

Met betrekking tot het gebruik van aanvullende zorg:4. Is er verschil in het gebruik van huisartsenzorg (huisartsenconsultaties,

medicatievoorschriften en verwijzingen) tussen zwangere vrouwen en niet-zwangere vrouwen in Nederland?

5. Wat is de prevalentie en wat zijn de determinanten van het gebruik van zorg van laag-risico zwangeren verleend door alternatieve zorgverleners?

In Hoofdstuk 2 worden de uitkomsten van een systematische literatuurreview beschreven. Deze review vat de stand van zaken samen omtrent de determinanten van prenataal zorggebruik (laat in zorg komen of minder controles krijgen dan geadviseerd in richtlijnen) in geïndustrialiseerde landen. In drie databases werden studies gezocht. Potentiële studies werden door twee onderzoekers onafhankelijk van elkaar gelezen en beoordeeld. Uiteindelijk zijn acht studies geïncludeerd die van hoge kwaliteit waren, afkomstig uit vier verschillende landen (Verenigde Staten, Groot Brittannië, Finland en Canada). Er is een beschrijvende systematische review uitgevoerd omdat een meta-analyse niet mogelijk was vanwege de heterogeniteit van de geïncludeerde studies.

De volgende individuele determinanten hingen samen met ontoereikend prenataal zorggebruik: jonge leeftijd van de zwangere (<20 jaar), laag opleidingsniveau (<9 jaar onderwijs), ongehuwden/alleenstaande zwangeren, etnische minderheden (verschillend gedefinieerd), geplande zorg bij een huisarts/verloskundige/team verloskundigen of gynaecoloog in vergelijking met een multidisciplinair team, geplande zorg in een streekziekenhuis in vergelijking met een stadsziekenhuis, ongeplande plaats van bevalling, niet verzekerd zijn, hoge pariteit, geen premature bevalling in de anamnese en laat op de hoogte zijn van een zwangerschap. Contextuele determinanten die samenhingen met ontoereikend prenataal zorggebruik waren: wonen in een achterstandswijk (60% gekleurde bewoners en 30% laag inkomensniveau), wonen in buurten met hoge percentages werklozen, eenoudergezinnen, midden-inkomens en laag opgeleiden, en waar veel Canadese Aboriginals wonen. Daarnaast was roken een gezondheidsgedrag determinant die samenhing met ontoereikend prenataal zorggebruik. Het aantal onderzoeken met een hoge bewijskracht die determinanten van ontoereikend prenataal zorggebruik vaststellen is beperkt. Geen enkele studie omvatte Nederlandse gegevens. De acht geïncludeerde studies hanteerden 12 verschillende definities van ontoereikend prenataal zorggebruik. Het is noodzakelijk dat er gestandaardiseerde definities komen om resultaten beter te kunnen vergelijken en te integreren.


| 141


9

Hoofdstuk 3 beschrijft de determinanten van ontoereikend prenataal zorggebruik van laag risico zwangeren in de eerstelijns verloskundige zorg in Nederland. Data voor deze studie werden verkregen uit de DELIVER studie. We verzamelden gegevens uit een vragenlijst afgenomen bij zwangeren voor de 34e zwangerschapsweek, uit zwangerschapsdossiers en uit gegevens van de Landelijke Verloskundige Registratie. Ontoereikend prenataal zorggebruik is gemeten met behulp van een aangepaste Kotelchuck index. Deze index is aangepast aan de richtlijn Prenatale begeleiding van de Koninklijke Nederlandse Organisatie van Verloskundigen en combineert het instromen in verloskundige zorg en het aantal consulten in de zwangerschap.

We includeerden 3070 vrouwen die hun zorgtraject startten bij de eerstelijns verloskundige. Ontoereikend prenataal zorggebruik registreerden we bij 24,7% van alle vrouwen. Hetzelfde percentage vrouwen werd tijdens de zwangerschap verwezen naar tweedelijns zorg. Niet-Westerse vrouwen, werkloze vrouwen, vrouwen met chronische ziekten/handicaps en vrouwen die geen foliumzuur slikten gebruiken significant vaker ontoereikende prenatale zorg. Niet-verwezen vrouwen hadden vaker ontoereikend prenataal zorggebruik wanneer ze van plan waren te bevallen in het ziekenhuis, wanneer ze geen foliumzuur gebruikten en wanneer ze blootgesteld werden aan sigarettenrook. Verwezen vrouwen hadden vaker ontoereikend prenataal zorggebruik wanneer ze een chronische ziekte of handicap registreerden en wanneer ze geen foliumzuur slikten. In deze studie tonen wij aan dat een relatief grote groep vrouwen (24,7%) niet op tijd met prenatale zorg start en/of niet minder consulten krijgt dan geadviseerd in de standaard Prenatale Begeleiding. Daarnaast tonen we aan dat ontoereikend prenataal zorggebruik in een laag-risico setting gerelateerd is aan een beperkt aantal determinanten. De determinanten die we vonden kunnen worden gebruikt om interventies in te zetten op specifieke groepen vrouwen die vaker ontoereikend prenatale zorg gebruiken.

In hoofdstuk 4 is de inhoud en timing van prenatale zorg onderzocht. Dit is gedaan in een vergelijkende studie tussen een groep Nederlandse en Belgische laag-risico zwangere vrouwen. Gegevens werden verkregen uit de DELIVER studie en uit een Belgische studie uitgevoerd in Brussel. Laag-risico zwangeren uit een stedelijke omgeving werden geïncludeerd, en daarna werden de gegevens gecombineerd tot één gegevensbestand om zoveel mogelijk te kunnen corrigeren voor verschillen tussen de groepen. Prenataal zorggebruik werd geoperationaliseerd middels het “Content and Timing of Pregnancy (CTP)” meetinstrument. In dit meetinstrument wordt het instromen in verloskundige zorg gecombineerd met drie prenatale interventies (bloeddrukmetingen, echo’s en Hb-bepalingen). We includeerden 642 zwangere vrouwen.

Belgische vrouwen hadden vaker ontoereikend prenataal zorggebruik (9,7%) dan Nederlandse vrouwen (5,6%). Ongeacht het land van studie waren de volgende determinanten geassocieerd met ontoereikend prenataal zorggebruik: laag opleidingsniveau, werkeloosheid,

R1R2R3R4R5R6R7R8R9


142 |

Chapter 9

niet-continue verloskundige zorg en het niet gebruik maken van zwangerschapseducatie. Deze studie is naar ons idee de eerste grensoverschrijdende studie die prenataal zorggebruik en de determinanten hiervan onderzoekt. Meer grensoverschrijdende studies zijn nodig om andere potentiële determinanten van prenataal zorggebruik te onderzoeken. Om dit te bereiken moeten op een systematische manier data over zwangere vrouwen worden verzameld.

Hoofdstuk 5 beschrijft het gebruik van huisartsenzorg (huisartsenconsultaties en gestelde diagnoses, geneesmiddelenvoorschriften en verwijzingen van patiënten) tussen zwangere vrouwen en niet-zwangere vrouwen in Nederland. Longitudinale gegevens uit de LINH (Landelijk Informatie Netwerk Huisartsenzorg) studie werden hiervoor gebruikt. Gegevens over huisartsenconsultaties, geneesmiddelenvoorschriften en verwijzingen van patiënten uit 84 praktijken in Nederland in de periode van 2007-2009 werden verzameld. We includeerden 15.123 zwangere vrouwen en 102.564 niet-zwangere vrouwen van dezelfde leeftijd (15-45 jaar).

Zwangere vrouwen hadden gemiddeld 3,6 keer contact met de huisarts (telefonisch, huisbezoek, consult bij de huisartsenpraktijk), niet-zwangeren in hetzelfde tijdsbestek 2,2 keer. De meest gestelde diagnoses bij zwangere vrouwen waren ‘zwangerschap’ en ‘urineweginfectie’, terwijl dit bij niet-zwangeren ‘urineweginfectie’ en ‘algemene ziekte, niet nader gespecificeerd’ waren. Het gemiddeld aantal geneesmiddelenvoorschriften was lager voor zwangere vrouwen dan bij niet zwangere vrouwen (respectievelijk 2.1 en 4.4). Zwangere vrouwen werden door huisartsen het meest verwezen naar verloskundige zorgverleners en niet-zwangeren naar de fysiotherapeut. Wij tonen aan de huisarts een belangrijke zorgverlener is voor zwangere vrouwen. Daarom is het belangrijk dat de rol van huisartsen in de verloskundige keten versterkt wordt.

Hoofdstuk 6 beschrijft de prevalentie en determinanten van zorggebruik ten aanzien van alternatieve en complementaire zorgverleners. De onderzoekspopulatie bestond uit laag-risico zwangeren die onder zorg waren bij de eerstelijns verloskundige. Hiervoor werden data uit de DELIVER studie gebruikt. Gebruik van zorg bij een alternatieve zorgverlener werd gemeten middels een vraag uit de derde vragenlijst die postpartum werd ingevuld door deelnemende cliënten. Elke vrouw die tenminste één consult bij een alternatieve zorgverlener rapporteerde werd ingedeeld als gebruiker van zorg van een alternatieve zorgverlener.

We vonden dat 9,4% van de onderzoekspopulatie tenminste één keer een consult hadden bij een alternatieve zorgverlener. Determinanten van dit zorggebruik waren het hebben van een aanvullende zorgverzekering, het hebben van een chronische ziekte of handicap, nicotinegebruik en alcoholgebruik. We concluderen dat relatief veel zwangere vrouwen gebruik maken van alternatieve zorg. De determinanten die wij vonden wijken af van


| 143


9

eerdere studies; daarin werden meer heterogene populaties onderzocht. De belangrijkste aanbeveling uit deze studie is dat verloskundige zorgverleners op de hoogte dienen te zijn van dit zorggebruik en dat dit zorggebruik van zwangeren actief aan de orde gesteld dient te worden.

In Hoofdstuk 7 worden de belangrijkste bevindingen van dit proefschrift samengevat en besproken. Verder worden de methodologische aspecten, de implicaties voor praktijk, onderwijs en onderzoek besproken.

We vonden dat een relatief groot deel van de vrouwen die onder zorg zijn van de eerstelijns verloskundige niet voldoen aan de standaard betreffende het prenatale zorgschema. Dit betekent dat deze vrouwen later dan 12 weken in zorg komen en/of een ontoereikend aantal consulten heeft in het vervolg van de zwangerschap. We bediscussieerden deze uitkomst en vroegen ons af of dit ontoereikende zorggebruik gerelateerd zou kunnen zijn aan beperkingen die vrouwen ervaren ten aanzien van de verloskundige zorg of dat verloskundigen beperkingen ervaren ten aanzien van het uitvoeren van de standaard. Wellicht passen verloskundigen de zorg aan aan de behoefte van een zwangere. Onderzoek naar het optimale aantal consulten zou moeten worden uitgevoerd waarbij rekening wordt gehouden met de wensen van zwangeren, verloskundigen en waarbij uiteraard ook het effect op perinatale morbiditeit en mortaliteit wordt gemeten.

In een groep laag-risico zwangere vrouwen vonden wij een beperkter aantal determinanten die samenhingen met ontoereikend prenataal zorggebruik dan in de systematische literatuur review betreffende hoge inkomenslanden. Echter, de determinanten die wij vonden zijn deels gelijk aan de determinanten die samenhangen met hogere perinatale sterfte. Dit kan betekenen dat er een relatie is tussen prenataal zorggebruik en perinatale sterfte. Verder onderzoek is nodig om een mogelijke relatie op te sporen.

Het gebruik van aanvullend zorggebruik door zwangere vrouwen bij de huisarts was relatief hoog; ook maakte bijna 10% van de zwangere vrouwen gebruik van alternatieve zorg. Het is daarom noodzakelijk de zorg in de eerstelijn te coördineren om te voorkomen dat er informatie over een zwangere verdwijnt en om te zorgen dat er niet meer interventies worden ingezet dan medisch noodzakelijk. Verloskundigen hebben de regie over de zwangerschap, bevalling en het kraambed, maar ook huisartsen hebben een belangrijke taak tijdens de zwangerschap en het kraambed. Het is daarom belangrijk dat er een goede samenwerking bestaat tussen beide professies. Voorbeelden om de samenwerking tussen verloskundigen en huisartsen te verbeteren kunnen liggen in gezamenlijke huisvesting, gezamenlijke richtlijnen en een gezamenlijke digitale omgeving.

De uitkomsten zoals beschreven in dit proefschrift over het gebruik van zorg aangeboden binnen het prenatale zorgschema en het aanvullend gebruik van zorg in de zwangerschap hebben belangrijke consequenties voor de praktijk, beleid, onderwijs en onderzoek. Dit proefschrift kan bijdragen tot het verbeteren van de uitkomsten van zorg voor zwangere vrouwen en hun baby’s.

APPENDICES

Curriculum Vitae

Dankwoord

SHARE publications

R1R2R3R4R5R6R7R8R9


146 |

Appendices


| 147

Curriculum Vitae

A

CURRICULUM VITAE

Esthelle Idberga (Esther) de Jong was born on May 31st 1972 in Kampen, the Netherlands. In 1990 she received her pre-university secondary education degree (VWO) at the Vechtdalcollege, Hardenberg. After one year of studying psychology at the university of Utrecht she started a bachelor in nursing (Windesheim university of applied sciences), where she graduated in 1995, followed by a bachelor degree in midwifery in 1998 (Leuven university college). In the following year she started working as a midwife in secondary care at the Antonius Hospital in Sneek. In 1999, she moved to Groningen where she started working in primary midwifery care at ‘de groepspraktijk van verloskundigen’. Simultaneously, she started her Master’s programme Health Sciences at the university of Maastricht, where she obtained her Master’s degree in 2004/2005. In her masterthesis she investigated patient satisfaction in primary midwifery care in the northern part of the Netherlands.

After her Master’s, she started working at the midwifery academy Groningen as a lecturer in methodology and midwifery. From 2008 until 2015, she worked as a PhD student under the supervision of Prof. dr. S.A. Reijneveld and Prof. dr. F. Schellevis at the department of Health Sciences of the University Medical Center Groningen (UMCG). Her PhD project involved the use and determinants of prenatal healthcare services of pregnant women. At the same time, she continued working as a lecturer in education, focusing mainly on work related to the examination board of the Midwifery Academy Groningen-Amsterdam. In 2012 she became the chair of this board.

In august 2015 she will start as a researcher at the department of Midwifery Science Amsterdam next to her work as a lecturer in midwifery.

R1R2R3R4R5R6R7R8R9


148 |

Appendices


| 149

Dankwoord

A

DANKWOORD

Het volmaken en afronden van een promotietraject is soms een eenzame weg, maar juist door de aanwezigheid en adviezen van vele mensen lukte het keer op keer om vol goede moed door te gaan. De leden van mijn promotieteam hebben een enorme bijdrage hieraan geleverd. Nooit heb ik me afgevraagd of ik een vraag wel kon stellen en/of ik mijn hart wel kon luchten. Onze bijeenkomsten waren prettig en erg leerzaam voor mij. Ik wens elke promovendus zo’n team toe. Een aantal mensen wil ik in het bijzonder danken.

Professor S.A. Reijneveld. Beste Menno, jij hebt mij laten zien met weinig informatie precies de kern te kunnen weergeven. Dit is een kunst die jij als geen ander verstaat. Alhoewel ik graag de verloskunde wilde promoten in dit proefschrift was jij altijd degene die aangaf dat ik het in een breder perspectief moest plaatsen. Dank hiervoor! Hierdoor is het proefschrift ook een verbreding van mijn vak geworden en niet alleen een verdieping.

Professor F. Schellevis. Beste François, de eerste kennismaking met jou was tijdens een DELIVER-vergadering waarin ik mijn onderzoeksvoorstel presenteerde. Gelijk na de vergadering gaf je aan dat het ook mogelijk zou zijn om een grote huisartsendatabase te gebruiken in mijn onderzoek. Dit was zo’n goed idee dat jij ook gelijk maar mijn tweede promotor bent geworden. Elke keer als ik vond dat het proces te langzaam ging wist jij mij gerust te stellen. Dit heeft veel waarde gehad voor mij. Je inhoudelijke en wetenschappelijke expertise hebben veel kunnen bijdragen aan dit proefschrift. Veel dank hiervoor!

Dr. F. Baarveld. Beste Frank, onze samenwerking startte, lang voordat ik überhaupt nadacht over een promotie, tijdens het begeleiden van een groep studenten van de Verloskunde Academie Groningen bij hun afstudeerscriptie. Dit is de start geweest voor het idee van dit proefschrift. Jij was tijdens dit proces vooral waardevol in je praktische inbreng en ideeën. Jij hebt mij enorm geholpen in het wegwijs worden in de huisartsgeneeskunde. Ook wist jij altijd de obstakels die ik tegenkwam te relativeren en hielp jij mij met logistieke vragen. Het was een waar genoegen met je samen te werken.

Dr. D.E.M.C. Jansen. Lieve Danielle, deze promotie is het begin geweest van onze vriendschap. Het ene moment zaten we samen op kleuterstoeltjes te kijken naar een voorstelling van onze kinderen en zo was jij ineens mijn copromotor. Het leven kan soms een bijzondere wending nemen. Wij hebben elkaar veel gezien de afgelopen jaren. Elke bijeenkomst hebben we altijd eerst onze levens besproken, daarna kwamen we tot het onderzoek waar we eigenlijk altijd snel tot consensus kwamen. Ik heb het enorm gewaardeerd dat jij zoveel tijd voor mij hebt vrijgemaakt, altijd goede feedback gaf en dat je open stond voor elke vraag die ik

R1R2R3R4R5R6R7R8R9


150 |

Appendices

had. Geweldig! Als ik ook eens promovendi mag gaan begeleiden dan ben jij mijn lichtend voorbeeld. Ik hoop nog vaak samen met je te koffieleuten.

Esther Nieuwschepen-Ensing. Lieve Esther, al 24 jaar delen wij lief en leed. Onze vriendschap startte in 1991 op de eerste dag van de opleiding Verpleegkunde. Samen werden we verpleegkundige en daarna verloskundige. Ook werkten we in dezelfde regio. Onze kinderen werden geboren onder elkaars begeleiding. We zijn altijd bezig met het creëren van nieuwe ambities, jij meer in de managementhoek en ik in de wetenschap, met als centraal thema de verloskunde. Over dit onderwerp raken wij niet uitgesproken. Dit heeft ook geleid tot ideeën die ik kon opnemen in dit proefschrift. Dank hiervoor! Maar bovenal dank voor alle jaren vriendschap die ik met je deel.

Paul de Cock. Beste Paul, alhoewel je natuurlijk geen verloskundige bent begin je zo ondertussen wel het verloskundige gedachtengoed te begrijpen ;-). Probeer dit overigens ook maar niet al te goed want het is prettig dat jij vanuit een andere invalshoek onderzoek bekijkt. Dat is prettig en inspirerend. Dank hiervoor. Daarnaast weet jij moeilijke zaken goed te relativeren, dat was voor mij zeer prettig. Ik hoop, ondanks dat je de AVAG gaat verlaten, dat we nauw betrokken blijven bij elkaar.

Drs. Y. Beishuizen, R. Suierveld MSc, Drs. G.A.M. Vermeulen en Drs. M. Den Arend. Beste Yvonne en Reinskje, toen nog niemand structureel onderzoek uitvoerde op de VAG gaven jullie mij het vertrouwen dat ik als pionier dit zou kunnen gaan doen. Deze kans heb ik met beide handen aangegrepen. Alhoewel het eerst niet altijd makkelijk was als enige onderzoeker te werken bij de VAG bleven jullie mij altijd ondersteunen en stimuleren. Geweldig! Gea en Monique, de AVAG heeft deze promotie mogelijk gemaakt. Jullie zijn genereus geweest in het toekennen van tijd en faciliteiten. Hierdoor was het mogelijk te promoveren en ook nog parttime docent verloskunde te zijn. Dank hiervoor!

Mede-promovendi, lieve Dr. Monique, Carien, Dr. Agatha, Janneke, Linda, Catja, Trudy, Ruth en Myrte, wij hebben hetzelfde traject doorlopen en hebben dezelfde problemen ervaren. Het was zo fijn dit te kunnen delen en stoom te kunnen afblazen, waarna we natuurlijk met frisse moed weer verder gingen ;-). Agatha, het was enorm leuk en inspirerend om samen met jou ‘onze’ index te construeren. Monique, de tripjes samen naar Amsterdam (en Zuid-Afrika) waren altijd gezellig en ontspannen. Heerlijk! Catja en Carien, samen promoveren in Groningen is toch makkelijker. Fijn dat we altijd samen onze problemen konden bespreken. Jullie zijn er ook bijna. Het komt goed.


| 151

Dankwoord

A

Collega’s van de AVAG, werken bij de AVAG is een genot. Door de mogelijkheden maar bovenal door jullie! Ik zag jullie soms ploeteren om het onderwijs voor elkaar te krijgen terwijl ik maar ‘rustig’ achter mijn computer zat te onderzoeken. Geweldig dat jullie altijd veel interesse toonden. Het liefst zou ik jullie allemaal persoonlijk benoemen en bedanken maar dan wordt dit proefschrift dubbel zo dik.

Vrienden en familie, dit promotietraject duurde natuurlijk een ‘eeuwigheid’. Desondanks bleven jullie geïnteresseerd. Super! Twee kunstenaars uit onze vriendenkring hebben de mooie omslag van dit proefschrift gemaakt. Lieve Bert, wat geweldig dat jij de omslag hebt ontworpen. Na kunst in ons huis, heb jij ook een handtekening gezet op deze thesis, dank hiervoor! Lieve Ard, wat fijn dat jij zo makkelijk invloog en het ontwerp van de omslag afmaakte en hiervoor je foto’s ter beschikking stelde.

Lieve mama, jij stimuleerde me altijd om door te leren en mezelf te blijven ontwikkelen. Met jouw ondersteuning en hulp is dit gelukt!

Robert, jij hebt me altijd gestimuleerd en ondersteund. Als geen ander sta jij me bij in roerige tijden. Je bent een geweldige man. Samen kunnen wij de wereld aan!

152

Gea en Monique, de AVAG heeft deze promotie mogelijk gemaakt. Jullie zijn genereus

geweest in het toekennen van tijd en faciliteiten. Hierdoor was het mogelijk te promoveren en

ook nog parttime docent verloskunde te zijn. Dank hiervoor!

Mede-promovendi, lieve Dr. Monique, Carien, Agatha, Janneke, Linda, Catja, Trudy,

Ruth en Myrte (promovendus to be), wij hebben hetzelfde traject doorlopen en hebben

dezelfde problemen ervaren. Het was zo fijn dit te kunnen delen en stoom te kunnen afblazen,

waarna we natuurlijk met frisse moed weer verder gingen ;-)

Collega’s van de AVAG, werken bij de AVAG is een genot. Door de mogelijkheden maar

bovenal door jullie! Ik zag jullie soms ploeteren om het onderwijs voor elkaar te krijgen

terwijl ik maar “rustig” achter mijn computer zat te onderzoeken. Geweldig dat jullie altijd

veel interesse toonden. Het liefst zou ik jullie allemaal persoonlijk benoemen en bedanken

maar dan wordt dit proefschrift dubbel zo dik.

Vrienden en familie, dit promotietraject duurde natuurlijk een ‘eeuwigheid’. Desondanks

bleven jullie geïnteresseerd. Super! Twee kunstenaars uit onze vriendenkring hebben de

mooie omslag van dit proefschrift gemaakt. Lieve Ard, wat fijn dat jij makkelijk invloog en

het ontwerp afmaakte. Lieve Bert, wat geweldig dat jij de omslag van het proefschrift hebt

gemaakt. Na kunst in ons huis, heb jij ook een handtekening gezet op deze thesis, dank

hiervoor!

Robert, jij hebt me altijd gestimuleerd en ondersteund. Als geen ander weet jij me te

ondersteunen in roerige tijden. Je bent een geweldige man. Samen kunnen wij de wereld aan!

R1R2R3R4R5R6R7R8R9


152 |

Appendices


| 153

SHARE publications

A

SHARE PUBLICATIONS

Research Institute SHARE

This thesis is published within the Research Institute SHARE (Science in Healthy Ageing and healthcaRE) of the University Medical Center Groningen / University of Groningen.Further information regarding the institute and its research can be obtained from our internetsite: www.share.umcg.nl

More recent theses can be found in the list below.((co-) supervisors are between brackets)

2015

Peters LLTowards Tailored Elderly Care with self-assessment measures of frailty and case complexity(prof E Buskens, prof JPJ Slaets, dr H Boter)

Sulkers EPsychological adaptation to childhood cancer(prof R Sanderman, prof PF Roodbol, prof ESJM de Bont, dr J Fleer, dr WJE Tissing)

Febrianna SASkin problems related to Indonesian leather & shoe production and the use of footwear in Indonesia(prof PJ Coenraads, prof H Soebono, dr MLA Schuttelaar)

2014

Schneeberger CAsymptomatic bacteruiria and urinary tract infections in women: focus on diabetes mellitus and pregnancy(prof RP Stolk, prof JJHM Erwich, dr SE Geerlings)

Skorvanek, MFatigue, apathy and quality of life in patients with Parkinson’s disease(prof JW Groothoff, prof Z Gdovinova, dr JP van Dijk, dr J Rosenberger)

R1R2R3R4R5R6R7R8R9


154 |

Appendices

Kolvek GEtiology and prognosis of chronic kidney disease in children: Roma ethnicity and other risk factors(prof SAReiijneveld, prof L Podracka, dr JP van Dijk, dr J Rosenberger)

Mikula PHealth related quality of life in people with multiple sclerosis; the role of coping, social participation and self-esteem(prof JW Groothoff, prof Z Gdovinova, dr JP van Dijk, dr I Nagyova)

Amalia RImproving a school-based dental programme through a sociodental risk group approach(prof RMH Schaub, prof JW Groothoff, prof N Widyanti)

Christoffers WAHand eczema; interventions and contact allergies(prof PJ Coenraads, dr MLA Schuttelaar)

Troquete NACSTART-ing risk assessment and shared care planning in out-patient forensic psychiatry; results from a cluster randomized controlled trial(prof D Wiersma, prof RA Schoevers, dr RHS van den Brink)

Golea EFunctioning of young individuals with upper limb reduction deficiencies(prof CK van der Sluis, dr RM Bongers, dr HA Reinders-Messelink)

Nguyen HTMedication safety in Vietnamese hospitals; a focus on medication errors and safety culture(prof K Taxis, prof FM Haaijer-Ruskamp, prof JRBJ Brouwers, dr TD Nguyen)

Lehmann VSinglehood and partnerships in healthy people and childhood cancer survivors; a focus on satisfaction(prof M hagedoorn, prof R Sanderman, dr MA Tuinman)


| 155

SHARE publications

A

Jaarsma EASports participation and physical disabilities: taking the hurdle?!(prof JHB Geertzen, prof PU Dijkstra, dr R Dekker)

Ockenburg SL vanPsycholopgical states and physical fates; studying the role of psychosocial stress in the etiology of cardiovascular disease: a nomothetic versus an idiographic approach(prof JGM Rosmalen, prof P de Jonge, prof ROB Gans)

Beijers CGHMUnhealthy behaviors during pregnancy; who continues to smoke and consume alcohol, and is treatment of anxiety and depressive symptoms effective?(prof J Ormel, prof CLH Bockting, dr H Burger)

Kerdijk WStrategic choices in curriculum design to facilitate knowledge and competency development(prof J Cohen-Schotanus, prof JW Snoek, dr R Tio)

Spaans FHemopexin activity and extracellular ATP in the pathogenesis of preeclampsia(prof H van Goor, dr MM Faas, dr WW Bakker)

Brinksma ANutritional status in children with cancer(prof PF Roodbol, prof R Sanderman, prof ESJM de Bont, dr WJE Tissing)

Prihodova LPsychological and medical determinants of long-term patient outcomes; a specific focus on patients after kidney transplantation and with haemophilia(prof JW Groothoff, dr JP van Dijk, dr I Rajnicova-Nagyova, dr J Rosenberger)

Snippe EUnderstanding change in psychological treatments for depressive symptoms; the individual matters(prof R Sanderman, prof PMG Emmelkamp, dr MJ Schroevers, dr J Fleer)

R1R2R3R4R5R6R7R8R9


156 |

Appendices

Groen B Complications in diabetic pregnancy; role of immunology and Advanced Glycation End products(prof TP Links, prof PP van den Berg, dr MM Faas)

Visser LEarly detection and prevention of adolescent alcohol use; parenting and psychosocial factors(prof SA Reijneveld, dr AF de Winter)

Tovote KAAcceptance or challenge? Psychological treatments for depressive symptoms in patients with diabetes(prof R Sanderman, prof PMG Emmelkamp, prof TP Links)

Trippolini MEvaluation of functioning in workers with whiplash-associated disorders and back pain(prof MF Reneman, prof PU Dijkstra, prof JHB Geertzen)

Eriks-Hoogland IEShoulder impairment in persons with a spinal cord injury & associations with activities and participation(prof LHV van der Woude, porf G Stucki, prof MWM Post, dr S de Groot)

Suwantika AAEconomic evaluations of non-traditional vaccinations in middle-income countries: Indonesia as a reference case(prof MJ Postma, dr K Lestari)

Behanova MArea- and individual-level socioeconomic differences in health and health-risk behaviours; a comparison of Slovak and Dutch cities(prof SA Reijneveld, dr JP van Dijk, dr I Rajnicova-Nagyova, dr Z Katreniakova)

Dekker HTeaching and learning professionalism in medical education(prof J Cohen-Schotanus, prof T van der Molen, prof JW Snoek)


| 157

SHARE publications

A

Dontje MLDaily physical activity in patients with a chronic disease(prof CP van der Schans, prof RP Stolk)

Gefenaite GNewly introduced vaccines; effectiveness and determinants of acceptance(prof E Hak, prof RP Stolk)

Dagan MThe role of spousal supportive behaviors in couples’ adaptation to colorectal cancer(prof M Hagedoorn, prof R Sanderman)

Monteiro SPDriving-impairing medicines and traffic safety; patients’perspectives(prof JJ de Gier, dr L van Dijk)

Bredeweg SRunning related injuries(prof JHB Geertzen, dr J Zwerver)

Mahmood SISelection of medical students and their specialty choices(prof JCC Borleffs, dr RA Tio)

For more 2014 and earlier theses visit our website

University of Groningen On the use and determinants of prenatal healthcare services de ... ·...

Documents

Transcript of University of Groningen On the use and determinants of prenatal healthcare services de ... ·...