- Email: [email protected]
Predictors of late initiation for prenatal care in a metropolitan region in Belgium. A cohort study
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
Available online at www.sciencedirect.com
Public Health journal homepage: www.elsevier.com/puhe
Original Research
Predictors of late initiation for prenatal care in a metropolitan region in Belgium. A cohort study M. Fobelets a,*, K. Beeckman b,c, A. Hoogewys d,e, M. Embo d, R. Buyl f, K. Putman a,g a
I-CHER (Interuniversity Centre for Health Economics Research), Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Belgium b Department of Medical Sociology and Health Sciences, Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Brussel, Belgium c Department of Nursing and Midwifery, Nursing and Midwifery Research Group, Universitair Ziekenhuis Brussel, Brussel, Belgium d Department of Midwifery, University College Arteveldehogeschool Ghent, Gent, Belgium e Department of Pedagogy of the Young Child, University College Arteveldehogeschool Ghent, Gent, Belgium f Department of Biostatistics and Medical Informatics, Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Belgium g School of Health, University of Central Lancashire, Preston, Lancashire, United Kingdom
article info
abstract
Article history:
Objectives: Timely initiation of prenatal care (PNC) in the first pregnancy trimester allows
Received 10 February 2014
prevention, identification and treatment of risk factors. However, not all women initiate
Received in revised form
PNC timely, especially women in a deprived situation. The aim of this study was to mea-
2 February 2015
sure the prevalence of late initiation, defined as initiation after 14 weeks of gestational age.
Accepted 12 March 2015
Secondly the authors wanted to identify predictors for late PNC onset.
Available online 21 April 2015
Study design: Observational cohort study. Methods: Pregnant women (n ¼ 1750) were recruited in all four hospitals in Ghent (Belgium),
Keywords:
a metropolitan region. A socio-economic deprivation ranking was measured by using a
Prenatal care (MeSH)
General Deprivation Index (GDI), which consists of six criteria to assess a socio-economic
Pregnancy (MeSH)
situation as deprived. A univariate analysis and a forward conditional multivariate logis-
Delivery of health care (MeSH)
tic regression model were used analysing the association between deprivation and the
Health services accessibility (MeSH)
likelihood to initiate PNC late.
Initiation of care
Results: 1115 women were included of whom 6.1% (n ¼ 68) initiated PNC late. A foreign
Deprivation
maternal country of birth (OR 2.10; 95% CI 1.15e3.83) and a total GDI 3 (OR 4.40; 95% CI 2.36e8.21) were good predictors for late initiation. More specifically, the GDI criteria education (OR 4.02; 95% CI 2.00e8.08) and unemployment (OR 2.40; 95% CI 1.17e4.90) were significantly associated with higher likelihood for late initiation.
* Corresponding author. Laarbeeklaan 103, 1090 Jette, Belgium. Tel.: þ32 477 47 20; fax: þ32 477 47 11. E-mail address: [email protected] (M. Fobelets). http://dx.doi.org/10.1016/j.puhe.2015.03.008 0033-3506/© 2015 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
649
Conclusions: A small group of women initiates PNC late. Vulnerable groups, at risk for late initiation can be identified through assessing their deprivation status. Priority for additional support should be given to women with low educational attainment or women in uncertain employment situations. © 2015 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
Introduction Late initiation of prenatal care (PNC) is known to have a negative impact on both mother and child. Initiation of PNC in the first trimester allows prevention, identification and treatment of risk factors.1,2 Previous studies demonstrated that young women,3e5 immigrants,4,6,7 women with a low income,7,8 low education,3,5 living in a temporary accommodation4 are more likely to initiate PNC late. Other factors described are (grande) multiparity,3,4,9 poor language proficiency,3,4 unplanned pregnancy,3,8 not having a regular obstetrician,7 not being married,5 smoking5,6 and living in a temporary accommodation.4 Some of these variables related to late onset of PNC, such as income and education, are also associated as indicators for measuring deprivation. Deprivation can be seen as the consequence of a lack of resources and opportunities.10 The term deprivation is broader than poverty. It does not only reflect the lack of financial resources but also the lack of opportunities, both in monetary and non-monetary terms. Previous research has shown that deprivation is associated with poorer health.11,12 Deprivation can be measured on two levels: individual and neighbourhood level. The assessment of neighbourhood deprivation is commonly used in studies to evaluate the association between perinatal health and deprived neighbourhoods.13e15 Living in a deprived area is associated with higher prevalence of adverse pregnancy outcomes. A limitation of the assessment of neighbourhood deprivation is the lack of correspondence to the individual socio-economic status between neighbourhood residents.13 Residents of a deprived neighbourhood tend to be more deprived compared to ones living in a non-deprived area, but this does not count for all residents. Therefore, the authors have chosen to measure deprivation by using the ‘General Deprivation Index' (GDI) as a first attempt to measure deprivation at individual level. The GDI is based on six items of the EU-SILC (European Union Statistics and Income and Living Conditions) and specifically developed by the Belgian Child and Family agency (Kind en Gezin) to assess deprivation in families with children.16,17 Low accessibility of health services has a positive impact on the timing of the first PNC visit.18 In Belgium, almost all residents (>99%) dispose of an obligatory healthcare insurance. Several policies have been set up to support social deprived women in their access to the health system in Belgium. Vulnerable women are exempt from out-of-pocket payments for services in community centers. These policies have a positive effect on the accessibility of PNC. It is estimated that only one percent of pregnant women do not receive any PNC in Belgium.19 The Belgian PNC guideline includes recommendations on the timing and number of visits
and diagnostic tests such as ultrasound screening, blood analysis, etc. Timely initiation of PNC is important for an adequate PNC follow-up. The latter allows an accurate pregnancy dating, medical history assessment, and identification and treatment of maternal and neonatal risk factors.7,20,21 Furthermore, timely initiation entails health education about substance abuse, nutrition and supplements advice to avoid complications such as prematurity and birth defects.3,22,23 In Belgium, a first PNC visit should take place in the first trimester, up to 14 weeks.19 The aim of this study was to measure the prevalence of late initiation and to identify predictors for late PNC onset based on a sample of women in a metropolitan region in Belgium. Predictors were identified based on several independent variables namely: age, maternal country of birth, parity and six socio-economic variables for the assessment of general deprivation: income, education, employment status, child(ren)’s development opportunities, housing and self-perceived general health.
Methods Data collection Women were consecutively recruited on the maternity units in all four hospitals in Ghent, one teaching hospital and three general hospitals. Ghent is the third largest city in Belgium with ± 250,000 residents. The recruitment period was from April 2011 until June 2011. All women who gave birth during that period were invited to participate in the study. Ethical approval was obtained from the four participating hospitals. Data were collected about the timing of PNC initiation by a short interview during their stay at the maternity unit. Data on the general characteristics of the women and the results of the GDI were collected by public health nurses, employed by the Child and Family agency (Kind en Gezin) in Flanders.17 This agency aims to actively contribute to the well-being of young children and their families by providing preventive medical, psychosocial and parenting/pedagogic services and organises the follow-up services postnatal in terms of health screening. Both datasets were pooled and afterwards anonymised by a third party.
Outcome variables The primary outcome variable in this study was the time of initiation of PNC. Late initiation of PNC was defined as initiation after 14 weeks of gestational age, in accordance with the definition of European Perinatal Health Report.21
650
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
The following independent variables were considered: age, maternal country of birth, parity and socio-economic deprivation score. The first variable age, was divided into four categories: 15e24, 25e29, 30e34 and 35 years. Maternal country of birth was divided into two groups: Belgian and foreign. Parity was divided into primiparous and multiparous women. A socio-economic deprivation assessment was performed, resulting in a GDI. Deprivation was defined as: ‘a condition where people experience limitations in participating in highly valuated commodities such as education, employment, and housing. Not as a one-off event, but as a sustained condition which occurs in various fields, both material and immaterial', in accordance to the definition used by the Child and Health agency.17 A GDI consists of six criteria: income, education, employment status, child(-ren)’s development, housing and self-perceived general health. Each criterion was assessed by a public health nurse during postnatal home visits, using a standardised procedure. These criteria are based on the topics of the EU-SILC survey (European Union Statistics and Income and Living Conditions) and especially developed by the Belgian Child and Family agency to assess deprivation in families with children.16,17,25 Each GDI criterion was scored binary, one for deprived and zero for not deprived. A GDI was calculated as the sum of the individual criteria. A GDI score of three or higher was assessed as socio-economical deprived. Income was evaluated deprived if the monthly salary is: irregular, below the minimum wage, an unemployment benefit and/or a minimum ‘substance’ income.16,17,25 The evaluation of education was based on the education of both parents. Education was scored deprived if at least one of the parents does not have a High School diploma or a diploma of a special education programme. Employment status was scored based on the employment status of both parents. If both parents were unemployed, working in precarious employment or a sheltered workshop, the GDI was evaluated deprived for the criteria employment. In case of single parent situation, the employment status of the single parent was considered. Children's development was assessed deprived if the community nurses experience difficulties of the parent(s) in raising and/or taking care of the child(-ren), also if the other children are irregularly or not attending nursery school. In case of doubts about the parent child interaction, a scale for assessing a problematic child raising situation (POS-scale: Problematische OpvoedingsSituatie Schaal) was used to investigate the risks for child abuse. Housing was evaluated as deprived if the house is unsafe, unhealthy, inadequate (lack of space in relation to the family size), lack of conveniences in the house or the environment and homeless families (no house, refuge, asylum centre, squat, etc). The last criteria, self-perceived general health, women were asked about their health status as the health status of the other family members. Self-perceived general health was defined as deprived if a child and/or one of the parents and/or other (resident) family member(s) have a poor health status such as a chronic disorder or disability and/or experience barriers to access the healthcare system.
Statistical analysis First, univariate statistical analyses were conducted using a Chi-square test for all (categorical) predictor variables, to assess the association between late initiation and the above mentioned independent variables. All non-significant variables (P-value > 0.05) were eliminated for further analysis. Variables with a statistic significance of P 0.05 were analysed further by performing a multivariate logistic regression. GDI was analysed by using the total score (Model 1) and the individual score on each sub item (Model 2). In both analyses, a forward conditional model was chosen. The results are presented by prevalences, percentages, P-values, odds ratios (and 95% confidence intervals). Hosmer and Lemeshow test was used to evaluate the goodness-of-fit for this model. The analysis was performed by using IBM SPSS Statistics version 20.
Results A total of 1750 women delivered during the three months of recruitment. All women consented to participate in the study. Cases of women with incomplete patient files and missing birth numbers were unable to be linked between both datasets and therefore 321 cases (18.3%) were excluded. Another 314 cases (17.9%) were excluded as a consequence of missing values. Finally they included a total of 1115 women (63.7%) in this study. Characteristics of the participating women are presented in Table 1. Of the 1115 women, 68 initiated PNC late (6.1%). The largest group of women was between 30 and 34 years old (37.0%). Of all women, 13.3% was between 15 and 24 years old. This age category was found to be most likely to initiate PNC late (16.8%). Maternal characteristics show that 78.2% of the sample were Belgian women, 3.9% of them initiated PNC late compared to 13.9% of women with a foreign maternal country of birth. The study sample consisted of 44.8% primigravidae and 55.2% multiparae. The difference between both groups regarding late initiation of PNC was negligible. Regarding the GDI, 11.4% were assessed positive for three or more criteria on the GDI and were therefore found to be deprived. A total of 22% of the cases with a total GDI 3 initiated PNC late. For all six criteria of the GDI, the deprived group seemed to initiate more frequently PNC late compared to non-deprived women. The univariate analysis showed that parity was not associated with late initiation of PNC (P > 0.05) and therefore excluded for further analyses. The following variables were found to be statistically significant in de univariate analysis and therefore included for further analysis: age, maternal country of birth, GDI total and individual score. The first forward conditional model (Model 1) is presented in Table 2. This model showed that a positive GDI (3 criteria positive) was a good predictive variable for late initiation with an OR 4.40 (95% CI 2.36e8.21). A foreign maternal country of birth was a predictive value for late initiation with an odds ratio of 2.10 (95% CI 1.15e3.83). No statistical significance was found between late onset of PNC and maternal age.
651
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
Table 1 e Maternal characteristics. Study sample (n ¼ 1115)
Variable
Initiation PNC < 15 weeks gestational age n (%) Age (years) 15e24 25e29 30e34 35 Maternal country of birth Belgian Foreign Parity Primiparae Multiparae General Deprivation Index <3 positive criteria 3 positive criteria Criteria General Deprivation Index Children's development Deprived Not deprived Education Deprived Not deprived Self-perceived general health Deprived Not deprived Housing Deprived Not deprived Income Deprived Not deprived Employment status Deprived Not deprived a
124 376 397 150
(83.2) (95.9) (96.1) (93.2)
25 16 16 11
(16.8) (4.1) (3.9) (6.8)
<0.001a
837 (96.1) 210 (86.1)
34 (3.9) 34 (13.9)
<0.001a
474 (94.8) 573 (93.2)
26 (5.2) 42 (6.8)
NS
948 (96.0) 99 (78.0)
40 (4.0) 28 (22.0)
<0.001a
33 (76.7) 1014 (94.6)
10 (23.3) 58 (5.4)
<0.001a
154 (80.6) 893 (96.6)
37 (19.4) 31 (3.4)
<0.001a
45 (83.3) 1002 (95.3)
9 (16.7) 59 (5.6)
0.001a
76 (83.5) 971 (94.8)
15 (16.5) 53 (5.2)
<0.001a
96 (79.3) 951 (95.7)
25 (20.7) 43 (4.3)
<0.001a
112 (78.9) 935 (96.1)
30 (21.1) 38 (3.9)
<0.001a
P-value derived from chi square and statistically significant (P 0.05) and subjected to the multivariate logistic regression analysis.
The results of Model 2 are shown in Table 3. In this model, all six criteria of the GDI were individually included in the analysis. Two criteria seemed to be good predictors for late initiation of PNC, namely education and employment status.
Table 2 e Model 1: multivariate logistic regression estimating late initiation based on women's characteristics and the total General Deprivation Index score. Variable Age (years) 15e24 (ref) 25e29 30e34 35 Maternal country of birth Belgian (ref) Foreign General Deprivation Index No deprivation (ref) Deprived a
P-value late initiation PNC
Initiation PNC 15 weeks gestational age n (%)
Odds ratio
95% CI
P-value
NS NS NS
2.10
1.15e3.83
0.016a
4.40
2.36e8.21
<0.001a
P-value 0.05 and statistically significant (Goodness of fit, Hosmer and Lemeshow test P ¼ 0.392).
Women who did not finish high school are significantly more likely to initiate PNC late compared to the others with an OR 4.02 (95% CI 2.00e8.08). Unemployed women are more likely to initiate PNC late with an OR 2.40 (95% CI 1.17e4.90).
Discussion This study aimed to measure the prevalence of late initiation of PNC in Ghent. The second aim was to gain more insight in the predictors for late onset of PNC. Despite the fact that only 1% of all Belgian women receive no PNC, 6.1% of all women in this study sample initiated PNC after 14 weeks gestational age. This result is comparable to the last published data for Belgium.7 A recent study, performed in Brussels (Belgium) showed a prevalence of late PNC initiation of 5.1%. This study is the only source for recent data, since national data on the onset of PNC is not systematically collected in Belgium. Consequently, the last European Perinatal Health Report did not present data of late initiation of PNC for Belgium.24 The results of the first model showed that deprivation measured by the GDI and maternal country of birth are good
652
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
Table 3 e Model 2: multivariate logistic regression estimating late initiation based on women's characteristics and individual score on each sub item of the total General Deprivation Index. Variable Age (years) 15e24 (ref) 25e29 30e34 35 Maternal country of birth Belgian (ref) Foreign General Deprivation Index Deprived for income Deprived for health Status Deprived for employment Status Deprived for housing Deprived for low stimulation level Deprived for educational level a
Odds ratio
95% CI
P-value
NS NS NS
NS Yes Yes Yes Yes Yes Yes
vs vs vs vs vs vs
No No No No No No
2.40
1.17e4.90
4.02
2.00e8.08
NS NS 0.017a NS NS <0.001a
P-value 0.05 and statistically significant (Goodness of fit, Hosmer and Lemeshow test P ¼ 1.000).
predictors for late PNC initiation. The association between late onset of PNC and maternal country of birth has been published before.4,6,7 Also, the GDI showed a good predictive value for late initiation. A comparison of the results with the literature was not possible since the GDI has been developed by the local Child and Family agency. Although the GDI is not validated, the tool has been used for several years by the public health nurses and seems to have a good validity for measuring deprivation. Moreover, the 6 GDI criteria cover most predominant factors influencing socio-economic status and are based on the topics of the EU-SILC survey. The results of household deprivation assessment could be considered as representative for the Ghent area. A percentage of 11.4% was assessed deprived in the study sample which corresponds to the 11.3% described on the National Website of Local Statistics of Flanders area.26 By using the GDI, public health nurses are able to identify vulnerable families and enables them to provide additional support and refer these families to other instances for social support. In contrast to the first model, the second model showed no statistical significance between late initiation of PNC and maternal country of birth, whilst the first model showed two significant predictors: GDI and maternal country of birth. In the second model, each of the six GDI criteria was evaluated separately, which resulted in a significant result for two criteria, namely education and working status. The results were in accordance with previously published studies identifying the association between low educational level, unemployment and late initiation of PNC.3,5,8 No significance was found for maternal country of birth in the second model. Employment status and education seem to have a high collinearity with maternal country of birth, which explains the non-significance for maternal country of birth in the second model. All 1750 women who delivered in one of the four maternity hospitals of Ghent participated in this study. This is a unique data sample as it represents all women who delivered in this city during the recruitment period since homebirths are very rare in Flanders (<1%).27 Research has been conducted to identify risk factors for late onset of PNC, but studies on the association between socio-economic deprivation measured
on individual level and late initiation of PNC are lacking. Perinatal morbidity and mortality studies have been conducted in the past to assess the association with deprivation. In previous research, authors have chosen to evaluate deprivation by using a Social Index to measure neighbourhood deprivation.13,14 As mentioned previously, the limitation of these studies is the lack of individual data on the socioeconomic status. People living in a deprived area aren't systematically living in a deprived situation. Therefore, this study is a first incentive to measure deprivation on individual level and its association between late PNC onset. There are some negative aspects that need be recognised when assessing deprivation on individual level only. For example, previous research showed that neighbourhood deprivation is associated with the level of accessibility of health services, which was not considered in this study.28 On the other hand, a recent study in Ghent on geographical accessibility, based on the distance from medical services, showed that the distance to services was a factor with only a small impact.29 Other factors such as the patients' knowledge, perceptions about the organisation of care and the quality of care were found to be larger barriers on accessibility to care. There are some limitations in this study that need to be addressed. To conduct this study, the authors limited to the variables included in the database of the Child and Family agency. Therefore, previously described predictive variables for late PNC initiation such as marital status, local language ability, having a regular obstetrician, etc. could not be included.3,4,7 Parity was not found to be statistical significant in this analysis. Previous research has shown that grande multiparae are more likely to initiate PNC late.4 Only 1.2% of all women in this study sample delivered more than four times. Therefore they could not make a distinction between multiparae and grande multiparae. This study was focused on the identification of women initiating PNC late. Starting PNC in the first pregnancy trimester, up to 14 weeks of gestational age, is primordial for the health of mother and child and such predictive variables should be identified to reach the maximum amount of pregnant women in time. This article emphasized the importance
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
of identifying vulnerable women to reduce the risk of late PNC initiation. Accessibility to care for all pregnant women and promotion of early initiation of PNC, certainly in deprived situations, should serve as one of the keystones of preventive health programmes on national as well as local level. Further studies are recommended to assess the influence of deprivation on late initiation of PNC. The development of a simple tool to measure household deprivation without the necessity of home visits would be helpful in daily practice. The assessment of deprivation during a visit to primary care services, based on a few short questions would be valuable to identify vulnerable groups for late initiation of PNC. A deprivation assessment during primary care visits should obligatory be performed during the anamnesis. Ideally, the identification of the vulnerable group for late PNC onset should take place before conception and information should automatically be synchronised between healthcare providers. To conclude, a small group of women initiates PNC late. Women, at risk for late onset can be recognised by evaluating their deprivation status. Additional attention and assistance for women with a low educational level or in an uncertain employment status should serve as one of the keystones when prioritising support programmes for vulnerable women. The identification of more variables such as marital status, local language ability, having a regular healthcare provider, etc. would be beneficial for the development of a new more refined deprivation index to be used in primary care. This deprivation index should be specifically designed for healthcare workers to identify women preconceptional in a deprived situation at risk for late initiation.
2.
3.
4.
5.
6.
7.
8.
9.
10. 11.
12.
Author statements Acknowledgements
13.
We thank the Belgian Child and Family agency (Kind en Gezin) for making their data available. 14.
Ethical approval Overall ethical approval was obtained from the central ethics committee of ‘Universitair Ziekenhuis Gent' (B670201110296, March 2011) following the approval of the local ethics committees from all participating hospitals.
15.
Funding
16.
None declared.
Competing interests
17.
None declared. 18.
references
1. Bergsjø P, Villar J. Scientific basis for the content of routine antenatal care. II. Power to eliminate or alleviate adverse born
19.
653
outcomes; some special conditions and examinations. Acta Obstet Gynecol Scand 1997;76:15e25. National Institute for Health and Care Excellence (NICE). Antenatal care. Routine antenatal care for the healthy pregnant women. Available at: http://www.nice.org.uk/nicemedia/ live/11947/40115/40115.pdf; 2010 [last accessed 16 January 2014]. Alderliesten ME, Vrijkotte TGM, van der Wal MF, Bonsel GJ. Late start of antenatal care among ethnic minorities in a large cohort of pregnant women. BJOG 2007;114:1232e9. Creswell JA, Hatherall B, Morris J, Jamal A, Renton A. Predictors of the timing of initiation of antenatal care in an ethnically diverse urban cohort in the UK. BMC Pregnancy Childbirth 2012;13:103. http://dx.doi.org/10.1186/1471-2393-13103. Hueston WJ, Gilbert GE, Davis L, Sturgill V. Delayed prenatal care and the risk for low birth weight delivery. J Community Health 2003;28:199e208. 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;109:265e73. Beeckman K, Louckx F, Putman K. Predisposing, enabling and pregnancy-related determinants of late initiation of prenatal care. Matern Child Health J 2010;15:1067e75. Johnson AA, El-Khorazaty MN, Hatcher BJ, Wingrove BK, Milligan R, Harris C, et al. Determinants of late prenatal care initiation by African American women in Washington, DC. Matern Child Health J 2003;7(2):103e14. Baker EC, Rajasingam D. Using trust databases to identify predictors of late booking for antenatal care within the UK. Public Health 2012;126(2):112e6. Nolan B, Whelan CT. Resources, deprivation and poverty. Oxford: Clarendon Press; 1996. Stafford M, Marmot M. Neighbourhood deprivation and health: does it affect us all equally? Int J Epidemiol 2003;32:357e66. Van Lenthe FJ, Schrijvers CT, Droomers M, Joung IM, Louwman MJ, Mackenbach JP. Investigating explanations of socio-economic inequalities in health: the Dutch GLOBE study. Eur J Public Health 2004;14(1):63e70. Poeran J, Maas AFG, Birnie E, Denktas S, Steegers EAP, Bonsel GJ. Social deprivation and adverse perinatal outcomes among Western and non-Western pregnant women in a Dutch urban population. Soc Sci Med 2013;83:42e9. Timmermans S, Bonsel GJ, Steegers-Theunissen RP, Mackenbach JP, Steyerberg EW, Raat H, et al. Individual accumulation of heterogeneous risks explains perinatal inequalities within deprived neighbourhoods. Eur J Epidemiol 2011;26(2):165e80. Messer LC, Laraia BA, Kaufman JS, Eyster J, Holzman C, Culhane J, et al. The development of a standardized neighborhood deprivation index. J Urban Health 2006;83(6):1041e62. European Comission. Eurostat: income and living conditions. Available at: [last accessed 30 August 2013] http://epp. eurostat.ec.europa.eu/portal/page/portal/income_social_ inclusion_living_conditions/introduction; 2012. Studiedienst van de Vlaamse Regering. Kansarmoede-index van Kind en Gezin (ID: 259). Available at: [last accessed 16 January 2014] http://aps.vlaanderen.be/sgml/largereeksen/259.htm; 2005. Charreire H, Combier E. Poor prenatal care in an urban area: a geographic analysis. Health Place 2009;15:412e9. Federaal Kenniscentrum voor de gezondheidszorg (KCE). Nationale richtlijn prenatale zorg: een basis voor een klinisch pad voor de opvolging van zwangerschappen. Available at: [last accessed 16 January 2014] https://kce.fgov.be/sites/default/ files/page_documents/d20041027313.pdf; 2004.
654
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
20. Nothnagle M, Marchi K, Egerter S, Braveman P. Risk factors for late or no prenatal care following medicaid expansions in California. Matern Child Health J 2000;4:4. 21. Sarnoff R, Adams E. Racial and ethnic disparities in the discordance between women's assessment of the timing of their prenatal care entry and the first trimester standard. Matern Child Health J 2001;5:3. 22. 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;111:S53e65. 23. Colley G, Johnson CH, Morrow B, Gaffield ME, Ahluwalia I. Prevalence of selected maternal and infant characteristics, pregnancy risk assessment monitoring system (PRAMS), 1997. MMWR CDC Surveill Summ 1999;48(5):1e37. 24. EURO-PERISTAT Project with SCPE and EUROCAT. European perinatal health Report. The health care of pregnant women and babies in Europe in 2010. Available at: [last accessed 16 January 2014] http://www.europeristat.com/images/European% 20Perinatal%20Health%20Report_2010.pdf; 2010.
25. Kind en Gezin. Kansarmoederegistratie binnen Kind en Gezin. Available at: [last accessed 16 January 2014] www. kindengezin.be/img/kansarmoederegistratie-toelichting.doc; 2011. 26. Studiedienst van de Vlaamse Regering. Lokale statistieken. Available at: [last accessed 16 January 2014] http://aps. vlaanderen.be/lokaal/lokale_statistieken.htm; 2014. 27. Cammu H, Martens E, Martens G, Van Mol C, Jacquemyn Y. SPE perinatale activiteiten in Vlaanderen 2011. Available at: [last accessed 16 January 2014] http://www.zorg-en-gezondheid. be/uploadedFiles/NLsite_v2/Cijfers/Cijfers_over_geboorte_ en_bevalling/SPE_jaarrapport%202011.pdf; 2010. 28. Law M, Wilson K, Eyles J, Elliott S, Jerrett M, Moffat T, et al. Meeting health need, accessing health care: the role of neighbourhood. Health Place 2005;11:367e77. 29. Willems S, Peersman W, De Maeyer P, Buylaert W, De Maeseneer J, De Paepe P. The impact of neighborhood deprivation on patients' unscheduled out-of-hours healthcare seeking behavior: a cross-sectional study. BMC Fam Pract 2013;14:136. http://dx.doi.org/10.1186/1471-2296-14-136.
Available online at www.sciencedirect.com
Public Health journal homepage: www.elsevier.com/puhe
Original Research
Predictors of late initiation for prenatal care in a metropolitan region in Belgium. A cohort study M. Fobelets a,*, K. Beeckman b,c, A. Hoogewys d,e, M. Embo d, R. Buyl f, K. Putman a,g a
I-CHER (Interuniversity Centre for Health Economics Research), Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Belgium b Department of Medical Sociology and Health Sciences, Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Brussel, Belgium c Department of Nursing and Midwifery, Nursing and Midwifery Research Group, Universitair Ziekenhuis Brussel, Brussel, Belgium d Department of Midwifery, University College Arteveldehogeschool Ghent, Gent, Belgium e Department of Pedagogy of the Young Child, University College Arteveldehogeschool Ghent, Gent, Belgium f Department of Biostatistics and Medical Informatics, Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Belgium g School of Health, University of Central Lancashire, Preston, Lancashire, United Kingdom
article info
abstract
Article history:
Objectives: Timely initiation of prenatal care (PNC) in the first pregnancy trimester allows
Received 10 February 2014
prevention, identification and treatment of risk factors. However, not all women initiate
Received in revised form
PNC timely, especially women in a deprived situation. The aim of this study was to mea-
2 February 2015
sure the prevalence of late initiation, defined as initiation after 14 weeks of gestational age.
Accepted 12 March 2015
Secondly the authors wanted to identify predictors for late PNC onset.
Available online 21 April 2015
Study design: Observational cohort study. Methods: Pregnant women (n ¼ 1750) were recruited in all four hospitals in Ghent (Belgium),
Keywords:
a metropolitan region. A socio-economic deprivation ranking was measured by using a
Prenatal care (MeSH)
General Deprivation Index (GDI), which consists of six criteria to assess a socio-economic
Pregnancy (MeSH)
situation as deprived. A univariate analysis and a forward conditional multivariate logis-
Delivery of health care (MeSH)
tic regression model were used analysing the association between deprivation and the
Health services accessibility (MeSH)
likelihood to initiate PNC late.
Initiation of care
Results: 1115 women were included of whom 6.1% (n ¼ 68) initiated PNC late. A foreign
Deprivation
maternal country of birth (OR 2.10; 95% CI 1.15e3.83) and a total GDI 3 (OR 4.40; 95% CI 2.36e8.21) were good predictors for late initiation. More specifically, the GDI criteria education (OR 4.02; 95% CI 2.00e8.08) and unemployment (OR 2.40; 95% CI 1.17e4.90) were significantly associated with higher likelihood for late initiation.
* Corresponding author. Laarbeeklaan 103, 1090 Jette, Belgium. Tel.: þ32 477 47 20; fax: þ32 477 47 11. E-mail address: [email protected] (M. Fobelets). http://dx.doi.org/10.1016/j.puhe.2015.03.008 0033-3506/© 2015 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
649
Conclusions: A small group of women initiates PNC late. Vulnerable groups, at risk for late initiation can be identified through assessing their deprivation status. Priority for additional support should be given to women with low educational attainment or women in uncertain employment situations. © 2015 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
Introduction Late initiation of prenatal care (PNC) is known to have a negative impact on both mother and child. Initiation of PNC in the first trimester allows prevention, identification and treatment of risk factors.1,2 Previous studies demonstrated that young women,3e5 immigrants,4,6,7 women with a low income,7,8 low education,3,5 living in a temporary accommodation4 are more likely to initiate PNC late. Other factors described are (grande) multiparity,3,4,9 poor language proficiency,3,4 unplanned pregnancy,3,8 not having a regular obstetrician,7 not being married,5 smoking5,6 and living in a temporary accommodation.4 Some of these variables related to late onset of PNC, such as income and education, are also associated as indicators for measuring deprivation. Deprivation can be seen as the consequence of a lack of resources and opportunities.10 The term deprivation is broader than poverty. It does not only reflect the lack of financial resources but also the lack of opportunities, both in monetary and non-monetary terms. Previous research has shown that deprivation is associated with poorer health.11,12 Deprivation can be measured on two levels: individual and neighbourhood level. The assessment of neighbourhood deprivation is commonly used in studies to evaluate the association between perinatal health and deprived neighbourhoods.13e15 Living in a deprived area is associated with higher prevalence of adverse pregnancy outcomes. A limitation of the assessment of neighbourhood deprivation is the lack of correspondence to the individual socio-economic status between neighbourhood residents.13 Residents of a deprived neighbourhood tend to be more deprived compared to ones living in a non-deprived area, but this does not count for all residents. Therefore, the authors have chosen to measure deprivation by using the ‘General Deprivation Index' (GDI) as a first attempt to measure deprivation at individual level. The GDI is based on six items of the EU-SILC (European Union Statistics and Income and Living Conditions) and specifically developed by the Belgian Child and Family agency (Kind en Gezin) to assess deprivation in families with children.16,17 Low accessibility of health services has a positive impact on the timing of the first PNC visit.18 In Belgium, almost all residents (>99%) dispose of an obligatory healthcare insurance. Several policies have been set up to support social deprived women in their access to the health system in Belgium. Vulnerable women are exempt from out-of-pocket payments for services in community centers. These policies have a positive effect on the accessibility of PNC. It is estimated that only one percent of pregnant women do not receive any PNC in Belgium.19 The Belgian PNC guideline includes recommendations on the timing and number of visits
and diagnostic tests such as ultrasound screening, blood analysis, etc. Timely initiation of PNC is important for an adequate PNC follow-up. The latter allows an accurate pregnancy dating, medical history assessment, and identification and treatment of maternal and neonatal risk factors.7,20,21 Furthermore, timely initiation entails health education about substance abuse, nutrition and supplements advice to avoid complications such as prematurity and birth defects.3,22,23 In Belgium, a first PNC visit should take place in the first trimester, up to 14 weeks.19 The aim of this study was to measure the prevalence of late initiation and to identify predictors for late PNC onset based on a sample of women in a metropolitan region in Belgium. Predictors were identified based on several independent variables namely: age, maternal country of birth, parity and six socio-economic variables for the assessment of general deprivation: income, education, employment status, child(ren)’s development opportunities, housing and self-perceived general health.
Methods Data collection Women were consecutively recruited on the maternity units in all four hospitals in Ghent, one teaching hospital and three general hospitals. Ghent is the third largest city in Belgium with ± 250,000 residents. The recruitment period was from April 2011 until June 2011. All women who gave birth during that period were invited to participate in the study. Ethical approval was obtained from the four participating hospitals. Data were collected about the timing of PNC initiation by a short interview during their stay at the maternity unit. Data on the general characteristics of the women and the results of the GDI were collected by public health nurses, employed by the Child and Family agency (Kind en Gezin) in Flanders.17 This agency aims to actively contribute to the well-being of young children and their families by providing preventive medical, psychosocial and parenting/pedagogic services and organises the follow-up services postnatal in terms of health screening. Both datasets were pooled and afterwards anonymised by a third party.
Outcome variables The primary outcome variable in this study was the time of initiation of PNC. Late initiation of PNC was defined as initiation after 14 weeks of gestational age, in accordance with the definition of European Perinatal Health Report.21
650
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
The following independent variables were considered: age, maternal country of birth, parity and socio-economic deprivation score. The first variable age, was divided into four categories: 15e24, 25e29, 30e34 and 35 years. Maternal country of birth was divided into two groups: Belgian and foreign. Parity was divided into primiparous and multiparous women. A socio-economic deprivation assessment was performed, resulting in a GDI. Deprivation was defined as: ‘a condition where people experience limitations in participating in highly valuated commodities such as education, employment, and housing. Not as a one-off event, but as a sustained condition which occurs in various fields, both material and immaterial', in accordance to the definition used by the Child and Health agency.17 A GDI consists of six criteria: income, education, employment status, child(-ren)’s development, housing and self-perceived general health. Each criterion was assessed by a public health nurse during postnatal home visits, using a standardised procedure. These criteria are based on the topics of the EU-SILC survey (European Union Statistics and Income and Living Conditions) and especially developed by the Belgian Child and Family agency to assess deprivation in families with children.16,17,25 Each GDI criterion was scored binary, one for deprived and zero for not deprived. A GDI was calculated as the sum of the individual criteria. A GDI score of three or higher was assessed as socio-economical deprived. Income was evaluated deprived if the monthly salary is: irregular, below the minimum wage, an unemployment benefit and/or a minimum ‘substance’ income.16,17,25 The evaluation of education was based on the education of both parents. Education was scored deprived if at least one of the parents does not have a High School diploma or a diploma of a special education programme. Employment status was scored based on the employment status of both parents. If both parents were unemployed, working in precarious employment or a sheltered workshop, the GDI was evaluated deprived for the criteria employment. In case of single parent situation, the employment status of the single parent was considered. Children's development was assessed deprived if the community nurses experience difficulties of the parent(s) in raising and/or taking care of the child(-ren), also if the other children are irregularly or not attending nursery school. In case of doubts about the parent child interaction, a scale for assessing a problematic child raising situation (POS-scale: Problematische OpvoedingsSituatie Schaal) was used to investigate the risks for child abuse. Housing was evaluated as deprived if the house is unsafe, unhealthy, inadequate (lack of space in relation to the family size), lack of conveniences in the house or the environment and homeless families (no house, refuge, asylum centre, squat, etc). The last criteria, self-perceived general health, women were asked about their health status as the health status of the other family members. Self-perceived general health was defined as deprived if a child and/or one of the parents and/or other (resident) family member(s) have a poor health status such as a chronic disorder or disability and/or experience barriers to access the healthcare system.
Statistical analysis First, univariate statistical analyses were conducted using a Chi-square test for all (categorical) predictor variables, to assess the association between late initiation and the above mentioned independent variables. All non-significant variables (P-value > 0.05) were eliminated for further analysis. Variables with a statistic significance of P 0.05 were analysed further by performing a multivariate logistic regression. GDI was analysed by using the total score (Model 1) and the individual score on each sub item (Model 2). In both analyses, a forward conditional model was chosen. The results are presented by prevalences, percentages, P-values, odds ratios (and 95% confidence intervals). Hosmer and Lemeshow test was used to evaluate the goodness-of-fit for this model. The analysis was performed by using IBM SPSS Statistics version 20.
Results A total of 1750 women delivered during the three months of recruitment. All women consented to participate in the study. Cases of women with incomplete patient files and missing birth numbers were unable to be linked between both datasets and therefore 321 cases (18.3%) were excluded. Another 314 cases (17.9%) were excluded as a consequence of missing values. Finally they included a total of 1115 women (63.7%) in this study. Characteristics of the participating women are presented in Table 1. Of the 1115 women, 68 initiated PNC late (6.1%). The largest group of women was between 30 and 34 years old (37.0%). Of all women, 13.3% was between 15 and 24 years old. This age category was found to be most likely to initiate PNC late (16.8%). Maternal characteristics show that 78.2% of the sample were Belgian women, 3.9% of them initiated PNC late compared to 13.9% of women with a foreign maternal country of birth. The study sample consisted of 44.8% primigravidae and 55.2% multiparae. The difference between both groups regarding late initiation of PNC was negligible. Regarding the GDI, 11.4% were assessed positive for three or more criteria on the GDI and were therefore found to be deprived. A total of 22% of the cases with a total GDI 3 initiated PNC late. For all six criteria of the GDI, the deprived group seemed to initiate more frequently PNC late compared to non-deprived women. The univariate analysis showed that parity was not associated with late initiation of PNC (P > 0.05) and therefore excluded for further analyses. The following variables were found to be statistically significant in de univariate analysis and therefore included for further analysis: age, maternal country of birth, GDI total and individual score. The first forward conditional model (Model 1) is presented in Table 2. This model showed that a positive GDI (3 criteria positive) was a good predictive variable for late initiation with an OR 4.40 (95% CI 2.36e8.21). A foreign maternal country of birth was a predictive value for late initiation with an odds ratio of 2.10 (95% CI 1.15e3.83). No statistical significance was found between late onset of PNC and maternal age.
651
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
Table 1 e Maternal characteristics. Study sample (n ¼ 1115)
Variable
Initiation PNC < 15 weeks gestational age n (%) Age (years) 15e24 25e29 30e34 35 Maternal country of birth Belgian Foreign Parity Primiparae Multiparae General Deprivation Index <3 positive criteria 3 positive criteria Criteria General Deprivation Index Children's development Deprived Not deprived Education Deprived Not deprived Self-perceived general health Deprived Not deprived Housing Deprived Not deprived Income Deprived Not deprived Employment status Deprived Not deprived a
124 376 397 150
(83.2) (95.9) (96.1) (93.2)
25 16 16 11
(16.8) (4.1) (3.9) (6.8)
<0.001a
837 (96.1) 210 (86.1)
34 (3.9) 34 (13.9)
<0.001a
474 (94.8) 573 (93.2)
26 (5.2) 42 (6.8)
NS
948 (96.0) 99 (78.0)
40 (4.0) 28 (22.0)
<0.001a
33 (76.7) 1014 (94.6)
10 (23.3) 58 (5.4)
<0.001a
154 (80.6) 893 (96.6)
37 (19.4) 31 (3.4)
<0.001a
45 (83.3) 1002 (95.3)
9 (16.7) 59 (5.6)
0.001a
76 (83.5) 971 (94.8)
15 (16.5) 53 (5.2)
<0.001a
96 (79.3) 951 (95.7)
25 (20.7) 43 (4.3)
<0.001a
112 (78.9) 935 (96.1)
30 (21.1) 38 (3.9)
<0.001a
P-value derived from chi square and statistically significant (P 0.05) and subjected to the multivariate logistic regression analysis.
The results of Model 2 are shown in Table 3. In this model, all six criteria of the GDI were individually included in the analysis. Two criteria seemed to be good predictors for late initiation of PNC, namely education and employment status.
Table 2 e Model 1: multivariate logistic regression estimating late initiation based on women's characteristics and the total General Deprivation Index score. Variable Age (years) 15e24 (ref) 25e29 30e34 35 Maternal country of birth Belgian (ref) Foreign General Deprivation Index No deprivation (ref) Deprived a
P-value late initiation PNC
Initiation PNC 15 weeks gestational age n (%)
Odds ratio
95% CI
P-value
NS NS NS
2.10
1.15e3.83
0.016a
4.40
2.36e8.21
<0.001a
P-value 0.05 and statistically significant (Goodness of fit, Hosmer and Lemeshow test P ¼ 0.392).
Women who did not finish high school are significantly more likely to initiate PNC late compared to the others with an OR 4.02 (95% CI 2.00e8.08). Unemployed women are more likely to initiate PNC late with an OR 2.40 (95% CI 1.17e4.90).
Discussion This study aimed to measure the prevalence of late initiation of PNC in Ghent. The second aim was to gain more insight in the predictors for late onset of PNC. Despite the fact that only 1% of all Belgian women receive no PNC, 6.1% of all women in this study sample initiated PNC after 14 weeks gestational age. This result is comparable to the last published data for Belgium.7 A recent study, performed in Brussels (Belgium) showed a prevalence of late PNC initiation of 5.1%. This study is the only source for recent data, since national data on the onset of PNC is not systematically collected in Belgium. Consequently, the last European Perinatal Health Report did not present data of late initiation of PNC for Belgium.24 The results of the first model showed that deprivation measured by the GDI and maternal country of birth are good
652
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
Table 3 e Model 2: multivariate logistic regression estimating late initiation based on women's characteristics and individual score on each sub item of the total General Deprivation Index. Variable Age (years) 15e24 (ref) 25e29 30e34 35 Maternal country of birth Belgian (ref) Foreign General Deprivation Index Deprived for income Deprived for health Status Deprived for employment Status Deprived for housing Deprived for low stimulation level Deprived for educational level a
Odds ratio
95% CI
P-value
NS NS NS
NS Yes Yes Yes Yes Yes Yes
vs vs vs vs vs vs
No No No No No No
2.40
1.17e4.90
4.02
2.00e8.08
NS NS 0.017a NS NS <0.001a
P-value 0.05 and statistically significant (Goodness of fit, Hosmer and Lemeshow test P ¼ 1.000).
predictors for late PNC initiation. The association between late onset of PNC and maternal country of birth has been published before.4,6,7 Also, the GDI showed a good predictive value for late initiation. A comparison of the results with the literature was not possible since the GDI has been developed by the local Child and Family agency. Although the GDI is not validated, the tool has been used for several years by the public health nurses and seems to have a good validity for measuring deprivation. Moreover, the 6 GDI criteria cover most predominant factors influencing socio-economic status and are based on the topics of the EU-SILC survey. The results of household deprivation assessment could be considered as representative for the Ghent area. A percentage of 11.4% was assessed deprived in the study sample which corresponds to the 11.3% described on the National Website of Local Statistics of Flanders area.26 By using the GDI, public health nurses are able to identify vulnerable families and enables them to provide additional support and refer these families to other instances for social support. In contrast to the first model, the second model showed no statistical significance between late initiation of PNC and maternal country of birth, whilst the first model showed two significant predictors: GDI and maternal country of birth. In the second model, each of the six GDI criteria was evaluated separately, which resulted in a significant result for two criteria, namely education and working status. The results were in accordance with previously published studies identifying the association between low educational level, unemployment and late initiation of PNC.3,5,8 No significance was found for maternal country of birth in the second model. Employment status and education seem to have a high collinearity with maternal country of birth, which explains the non-significance for maternal country of birth in the second model. All 1750 women who delivered in one of the four maternity hospitals of Ghent participated in this study. This is a unique data sample as it represents all women who delivered in this city during the recruitment period since homebirths are very rare in Flanders (<1%).27 Research has been conducted to identify risk factors for late onset of PNC, but studies on the association between socio-economic deprivation measured
on individual level and late initiation of PNC are lacking. Perinatal morbidity and mortality studies have been conducted in the past to assess the association with deprivation. In previous research, authors have chosen to evaluate deprivation by using a Social Index to measure neighbourhood deprivation.13,14 As mentioned previously, the limitation of these studies is the lack of individual data on the socioeconomic status. People living in a deprived area aren't systematically living in a deprived situation. Therefore, this study is a first incentive to measure deprivation on individual level and its association between late PNC onset. There are some negative aspects that need be recognised when assessing deprivation on individual level only. For example, previous research showed that neighbourhood deprivation is associated with the level of accessibility of health services, which was not considered in this study.28 On the other hand, a recent study in Ghent on geographical accessibility, based on the distance from medical services, showed that the distance to services was a factor with only a small impact.29 Other factors such as the patients' knowledge, perceptions about the organisation of care and the quality of care were found to be larger barriers on accessibility to care. There are some limitations in this study that need to be addressed. To conduct this study, the authors limited to the variables included in the database of the Child and Family agency. Therefore, previously described predictive variables for late PNC initiation such as marital status, local language ability, having a regular obstetrician, etc. could not be included.3,4,7 Parity was not found to be statistical significant in this analysis. Previous research has shown that grande multiparae are more likely to initiate PNC late.4 Only 1.2% of all women in this study sample delivered more than four times. Therefore they could not make a distinction between multiparae and grande multiparae. This study was focused on the identification of women initiating PNC late. Starting PNC in the first pregnancy trimester, up to 14 weeks of gestational age, is primordial for the health of mother and child and such predictive variables should be identified to reach the maximum amount of pregnant women in time. This article emphasized the importance
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
of identifying vulnerable women to reduce the risk of late PNC initiation. Accessibility to care for all pregnant women and promotion of early initiation of PNC, certainly in deprived situations, should serve as one of the keystones of preventive health programmes on national as well as local level. Further studies are recommended to assess the influence of deprivation on late initiation of PNC. The development of a simple tool to measure household deprivation without the necessity of home visits would be helpful in daily practice. The assessment of deprivation during a visit to primary care services, based on a few short questions would be valuable to identify vulnerable groups for late initiation of PNC. A deprivation assessment during primary care visits should obligatory be performed during the anamnesis. Ideally, the identification of the vulnerable group for late PNC onset should take place before conception and information should automatically be synchronised between healthcare providers. To conclude, a small group of women initiates PNC late. Women, at risk for late onset can be recognised by evaluating their deprivation status. Additional attention and assistance for women with a low educational level or in an uncertain employment status should serve as one of the keystones when prioritising support programmes for vulnerable women. The identification of more variables such as marital status, local language ability, having a regular healthcare provider, etc. would be beneficial for the development of a new more refined deprivation index to be used in primary care. This deprivation index should be specifically designed for healthcare workers to identify women preconceptional in a deprived situation at risk for late initiation.
2.
3.
4.
5.
6.
7.
8.
9.
10. 11.
12.
Author statements Acknowledgements
13.
We thank the Belgian Child and Family agency (Kind en Gezin) for making their data available. 14.
Ethical approval Overall ethical approval was obtained from the central ethics committee of ‘Universitair Ziekenhuis Gent' (B670201110296, March 2011) following the approval of the local ethics committees from all participating hospitals.
15.
Funding
16.
None declared.
Competing interests
17.
None declared. 18.
references
1. Bergsjø P, Villar J. Scientific basis for the content of routine antenatal care. II. Power to eliminate or alleviate adverse born
19.
653
outcomes; some special conditions and examinations. Acta Obstet Gynecol Scand 1997;76:15e25. National Institute for Health and Care Excellence (NICE). Antenatal care. Routine antenatal care for the healthy pregnant women. Available at: http://www.nice.org.uk/nicemedia/ live/11947/40115/40115.pdf; 2010 [last accessed 16 January 2014]. Alderliesten ME, Vrijkotte TGM, van der Wal MF, Bonsel GJ. Late start of antenatal care among ethnic minorities in a large cohort of pregnant women. BJOG 2007;114:1232e9. Creswell JA, Hatherall B, Morris J, Jamal A, Renton A. Predictors of the timing of initiation of antenatal care in an ethnically diverse urban cohort in the UK. BMC Pregnancy Childbirth 2012;13:103. http://dx.doi.org/10.1186/1471-2393-13103. Hueston WJ, Gilbert GE, Davis L, Sturgill V. Delayed prenatal care and the risk for low birth weight delivery. J Community Health 2003;28:199e208. 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;109:265e73. Beeckman K, Louckx F, Putman K. Predisposing, enabling and pregnancy-related determinants of late initiation of prenatal care. Matern Child Health J 2010;15:1067e75. Johnson AA, El-Khorazaty MN, Hatcher BJ, Wingrove BK, Milligan R, Harris C, et al. Determinants of late prenatal care initiation by African American women in Washington, DC. Matern Child Health J 2003;7(2):103e14. Baker EC, Rajasingam D. Using trust databases to identify predictors of late booking for antenatal care within the UK. Public Health 2012;126(2):112e6. Nolan B, Whelan CT. Resources, deprivation and poverty. Oxford: Clarendon Press; 1996. Stafford M, Marmot M. Neighbourhood deprivation and health: does it affect us all equally? Int J Epidemiol 2003;32:357e66. Van Lenthe FJ, Schrijvers CT, Droomers M, Joung IM, Louwman MJ, Mackenbach JP. Investigating explanations of socio-economic inequalities in health: the Dutch GLOBE study. Eur J Public Health 2004;14(1):63e70. Poeran J, Maas AFG, Birnie E, Denktas S, Steegers EAP, Bonsel GJ. Social deprivation and adverse perinatal outcomes among Western and non-Western pregnant women in a Dutch urban population. Soc Sci Med 2013;83:42e9. Timmermans S, Bonsel GJ, Steegers-Theunissen RP, Mackenbach JP, Steyerberg EW, Raat H, et al. Individual accumulation of heterogeneous risks explains perinatal inequalities within deprived neighbourhoods. Eur J Epidemiol 2011;26(2):165e80. Messer LC, Laraia BA, Kaufman JS, Eyster J, Holzman C, Culhane J, et al. The development of a standardized neighborhood deprivation index. J Urban Health 2006;83(6):1041e62. European Comission. Eurostat: income and living conditions. Available at: [last accessed 30 August 2013] http://epp. eurostat.ec.europa.eu/portal/page/portal/income_social_ inclusion_living_conditions/introduction; 2012. Studiedienst van de Vlaamse Regering. Kansarmoede-index van Kind en Gezin (ID: 259). Available at: [last accessed 16 January 2014] http://aps.vlaanderen.be/sgml/largereeksen/259.htm; 2005. Charreire H, Combier E. Poor prenatal care in an urban area: a geographic analysis. Health Place 2009;15:412e9. Federaal Kenniscentrum voor de gezondheidszorg (KCE). Nationale richtlijn prenatale zorg: een basis voor een klinisch pad voor de opvolging van zwangerschappen. Available at: [last accessed 16 January 2014] https://kce.fgov.be/sites/default/ files/page_documents/d20041027313.pdf; 2004.
654
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 6 4 8 e6 5 4
20. Nothnagle M, Marchi K, Egerter S, Braveman P. Risk factors for late or no prenatal care following medicaid expansions in California. Matern Child Health J 2000;4:4. 21. Sarnoff R, Adams E. Racial and ethnic disparities in the discordance between women's assessment of the timing of their prenatal care entry and the first trimester standard. Matern Child Health J 2001;5:3. 22. 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;111:S53e65. 23. Colley G, Johnson CH, Morrow B, Gaffield ME, Ahluwalia I. Prevalence of selected maternal and infant characteristics, pregnancy risk assessment monitoring system (PRAMS), 1997. MMWR CDC Surveill Summ 1999;48(5):1e37. 24. EURO-PERISTAT Project with SCPE and EUROCAT. European perinatal health Report. The health care of pregnant women and babies in Europe in 2010. Available at: [last accessed 16 January 2014] http://www.europeristat.com/images/European% 20Perinatal%20Health%20Report_2010.pdf; 2010.
25. Kind en Gezin. Kansarmoederegistratie binnen Kind en Gezin. Available at: [last accessed 16 January 2014] www. kindengezin.be/img/kansarmoederegistratie-toelichting.doc; 2011. 26. Studiedienst van de Vlaamse Regering. Lokale statistieken. Available at: [last accessed 16 January 2014] http://aps. vlaanderen.be/lokaal/lokale_statistieken.htm; 2014. 27. Cammu H, Martens E, Martens G, Van Mol C, Jacquemyn Y. SPE perinatale activiteiten in Vlaanderen 2011. Available at: [last accessed 16 January 2014] http://www.zorg-en-gezondheid. be/uploadedFiles/NLsite_v2/Cijfers/Cijfers_over_geboorte_ en_bevalling/SPE_jaarrapport%202011.pdf; 2010. 28. Law M, Wilson K, Eyles J, Elliott S, Jerrett M, Moffat T, et al. Meeting health need, accessing health care: the role of neighbourhood. Health Place 2005;11:367e77. 29. Willems S, Peersman W, De Maeyer P, Buylaert W, De Maeseneer J, De Paepe P. The impact of neighborhood deprivation on patients' unscheduled out-of-hours healthcare seeking behavior: a cross-sectional study. BMC Fam Pract 2013;14:136. http://dx.doi.org/10.1186/1471-2296-14-136.