Cohort Profile: The Amsterdam Born Children and their Development (ABCD) Study

Abstract

In 2000, a study among pregnant women in Amsterdam revealed large ethnic disparities in birth outcomes1 that were consistent with previous findings in both The Netherlands and other Western countries.2–7 So far, conventional risk factors such as low maternal socio-economic status (SES), smoking or body mass index (BMI) could not explain these disparities. These results, in conjunction with increasing evidence of the long-term implications of adverse pregnancy outcomes (fetal origins of disease hypothesis),8 formed the basis of the Amsterdam Born Children and their Development (ABCD) study, which was established in 2003. This large prospective population-based cohort study examines the association between maternal lifestyle, medical, psychosocial and environmental conditions during pregnancy and children’s health at birth as well as in later life. Specific attention is paid to ethnic disparities. Consequently, the study has detailed measurements of ethnic background, including country of birth of the pregnant woman herself, country of birth of her parents and ethnic identity. The study is a collaboration between the Public Health Service of Amsterdam and the two university medical centres in Amsterdam [Academic Medical Center (AMC) and VU University Medical Center (VUmc)], and has been implemented in cooperation with hospitals and midwives, Youth Health Care (YHC) centres, primary schools and the University of Tilburg. The main objectives of the ABCD study are: However, a fourth objective has been added over the past few years, following mounting evidence that postnatal conditions also may have important consequences for health and disease in later life:9 to investigate children’s health at birth as well as in later life, and ethnic disparities therein; to investigate maternal lifestyle, medical, psychosocial and environmental conditions during pregnancy, and ethnic disparities therein; to investigate the extent to which (ethnic disparities in) maternal conditions during pregnancy explain (ethnic disparities in) children’s health at birth as well as in later life. iv. to investigate early-life conditions, and ethnic disparities therein, and the extent to which these conditions explain children’s health in later life. The pregnancy and early-life conditions and outcomes of interest in the ABCD study are broadly defined into the research areas shown in Figure 1. The main maternal and environmental risk factors concern nutrition and metabolism, psychosocial stress, substance abuse, medical conditions and air pollution. The main outcome parameters concern birth outcomes and infant growth and, when the children are older, obesity and metabolic syndrome, cognitive development, psychosocial health and nutrition. More information is available on our website: http://www.abcd-study.nl. Main research areas of the ABCD study Results of the study will be used to improve prenatal care and to develop culture-specific intervention strategies to promote the health of all children from the earliest age possible. Between January 2003 and March 2004, all pregnant women living in Amsterdam were asked to participate in the ABCD study during their first prenatal visit to an obstetric care provider (general practitioner, midwife or gynaecologist). Altogether, 12 373 women were approached—by estimate, ≥99% of the target population. According to Dutch law, all pregnant women, including illegal immigrants and asylum-seekers, are entitled to receive prenatal care, which is free of charge if costs are a problem. For all of the women approached, the care provider completed a registration form which included personal data such as name, address and date of birth. Based on this information, a questionnaire covering socio-demographic characteristics, obstetric history, lifestyles and psychosocial conditions was sent to the pregnant women within 2 weeks, to be filled out at home and returned to the Public Health Service by prepaid mail. A reminder was sent 2 weeks later. The questionnaire included an informed consent sheet the women could use to grant permission for follow-up of their infants at the age of 3 months and every 5 years thereafter, and for the perusal of their medical files. In addition, women were invited to participate in the ABCD biomarker study. For this, an extra blood sample (10-ml EDTA and 9-ml serum) was taken during routine blood collection for screening purposes following the first antenatal check-up. Of the 12 373 women approached, 8266 women filled out the pregnancy questionnaire (response rate: 67%). Of this group, 7050 women granted permission for follow-up (85%) and 7043 women granted permission for perusal of her and her child’s medical files (85%). A total of 4389 women (53%) participated in the biomarker study. To enhance participation among foreign-born women, two supportive measures were taken: (i) a Turkish, Arabic or English translation was provided to women born in Turkey, Morocco or other non-Dutch-speaking countries and (ii) the possibility of completing the questionnaire orally was offered to women who were illiterate or had reading difficulties. Although the participation rates among foreign-born women (42–64%) were still lower than the participation rate among Dutch-born women (77%), they were comparable with response rates observed in other population-based multi-ethnic studies in The Netherlands.10–12 To investigate the degree of selection bias resulting from this ethnic non-response, we conducted a non-response analysis by probabilistic medical record linkage with The Netherlands Perinatal Registry; results indicated that selection bias was minimal.13Table 1 shows the socio-demographic characteristics of the participants according to country of birth. The questionnaire focused primarily on the mother; however, paternal information was available for country of birth (58.0% Dutch, 6.5% Surinamese, 4.6% Turkish, 8.5% Moroccan, 13.1% other non-Western, 6.8% other Western and 2.5% unknown), employment status (81.2% employed) and height (mean 180.9 cm, standard deviation 11.5). Response rates and maternal characteristics in the study population according to country of birtha aNW: non-Western; W: Western. Response rates and maternal characteristics in the study population according to country of birtha aNW: non-Western; W: Western. Approval of the study was obtained from the Central Committee on Research Involving Human Subjects in The Netherlands, the medical ethics review committees of the participating hospitals and the Registration Committee of the Municipality of Amsterdam. Figure 2 shows the structure of the ABCD study. Of the 8266 respondents, 7863 women gave birth to viable singleton infants and 132 women gave birth to viable multiples. The remaining mothers either experienced a miscarriage or fetal death (n = 92) or were lost to follow-up (n = 179) with no registered birth and no information on miscarriage or fetal death available from the care provider. By law, all children born in Amsterdam after 24 weeks’ gestation (either stillborn or live-born) must be registered at the municipality’s registry office, after which a ‘mutation report’ is sent to the Public Health Service’s YHC department. Between the 4th and 7th day after delivery, YHC nurses visit the live-born infants and their parents to obtain a blood sample to screen for congenital metabolic disorders. At this time they also record the date of delivery, infant sex, birth weight and gestational age [based on ultrasound or, if unavailable (<10%), on the timing of the last menstrual period] as provided by the obstetric care provider. Structure of the ABCD study. Attrition in phase III due to withdrawal, infant or maternal death and unknown address or emigration Three months after giving birth, the mothers who gave permission for follow-up (6735 mothers of singletons, 119 mothers of multiples) received a questionnaire concerning the course of their pregnancy and delivery; the baby’s health, development and growth; and the mother’s lifestyle during and after pregnancy (Phase II). Again, the questionnaire was available in multiple languages to increase the response in specific ethnic groups. A total of 5218 mothers (5131 mothers of singletons, 87 mothers of multiples) filled out this questionnaire (response rate: 76%). Compared with the initial sample, this group was older, more highly educated, of lower parity, and more often of Dutch origin. However, we found no effect of attrition bias in our first follow-up study, which investigated the association of maternal stress and emotional problems with infant crying.14 After the 3-month questionnaire, follow-up measurements were planned for singleton infants only. Until the children are 4 years old, specially trained YHC nurses conduct an average of 14 standardized routine measurements to monitor their growth and feeding patterns. Of the 6735 women who gave birth to live-born singleton infants, 6575 women gave permission to collect growth data. Data collection is ongoing, and so far, growth data (including feeding type and duration) of 5273 ABCD children have been digitized. At the current stage, 6161 of the 6735 mothers who gave permission for follow-up are being approached for the 5-year follow-up measurement of their child (Phase III, 2008–10). Attrition in this follow-up number is due to withdrawal, infant or maternal death and loss-to-follow-up as a result of an unknown address or emigration. Details of the Phase III follow-up are given below (‘What has been measured?’ section). In short, mothers receive two questionnaires about the child’s health, development and behaviour, with one of them to be filled out by the child’s teacher. In addition, the children are invited to participate in the ABCD health check, which takes place at school. Efforts to enhance participation among all women and children, regardless of ethnicity and education, include the use of translated questionnaires, an information leaflet, and a specially developed cartoon that explains the health check measurements to the children (Figure 3). Also, women from ethnic minority groups who do not respond within a month are approached by phone by trained students who explain the study in the women’s own language. Data collection is ongoing, and preliminary results show a response rate of 66% for the general questionnaire (n = 4060 out of the 6161 mothers approached) and a response rate of 58% for the teacher questionnaire. Of the participating mothers, 94% gave permission for the health check. Further follow-up measurements will take place every 5 years (in 2013 at age 10, in 2018 at age 15, etc.) until adulthood. Information for participants: a specially developed cartoon (in Dutch) that explains the measurements to the 5-year-old children. Reprinted with the permission of Floor de Goede/ComicHouse.nl. Translations: (1) Exciting: ABCD measurements at school today. What’s going to happen? (2) A cold spray to numb your finger … ; (3) then a tiny prick you can hardly feel … ; (4) weight … ; (5) height … and waist; (6) heart rate; (7) a little balloon around your arm to measure your blood pressure … ; (8) and at the end, some tests on the computer and on paper and (9) done! Heading home with a small gift and your ABCD diploma … Table 2 provides an overview of the measurements in the three subsequent phases of the study (details are available from the authors upon request). In Phase I, the maternal micronutrient status (including vitamins A, D, and B12; folate; iron; Zn; Mg; Ca and n-3 and n-6 fatty acids), metabolic function (including cholesterol, triglycerides and thyroid hormone levels) and cortisol levels were measured in blood samples collected at a median of 13 weeks’ gestation [inter-quartile range (IQR) 12–14 weeks]. The pregnancy questionnaire was filled out at an average of 16 weeks’ gestation (IQR 14–18 weeks) and included ethnicity-related and socio-demographic variables, lifestyle, psychosocial conditions and obstetric/medical information. Whenever possible, validated questionnaires were included, such as the CES-D questionnaire for depression15 and the Job Content Questionnaire for job strain.16 Furthermore, residential information was used to model individual exposure to traffic-related air pollution (in collaboration with the Institute of Risk Assessment Sciences at Utrecht University). Information on birth outcomes was obtained from YHC, as mentioned previously. Supplementary information concerning complications during pregnancy and neonatal outcomes was obtained by probabilistic medical record linkage with The Netherlands Perinatal Registry.13 General overview of data collected in the ABCD study SDQ: Strengths and Difficulties Questionnaire; NICU: neonatal intensive care unit; CITO: Central Institute for Test Development; ANT: Amsterdam Neuropsychological Tasks. General overview of data collected in the ABCD study SDQ: Strengths and Difficulties Questionnaire; NICU: neonatal intensive care unit; CITO: Central Institute for Test Development; ANT: Amsterdam Neuropsychological Tasks. The questionnaire filled out by mothers in Phase II concerned the course of their pregnancy and delivery, their physical and psychosocial health and the development and health of their child. The questionnaire was filled out at the average infant age of 13 weeks (IQR 12–13 weeks). Follow-up data from YHC included infant nutrition (i.e. duration of breast-feeding and/or bottle-feeding, time of introduction of solid foods) and infant and child weight, length/height and head circumference at an average of eight time points in the first 14 months and six time points between the ages of 14 months and 4 years. In the current phase (Phase III), participating mothers are sent a questionnaire ~2 weeks after their ABCD child’s fifth birthday. In addition to items about the child’s health, development and behaviour, this general questionnaire contains items about family socio-demographics, maternal lifestyle and psychosocial conditions and family history of medical conditions. The questionnaire is accompanied by a questionnaire about the child’s performance, behaviour and CITO (Central Institute for Test Development) index for elementary education, to be filled out by the child’s teacher. In both questionnaires, validated questions and scales were used wherever possible (e.g. the Strengths and Difficulties Questionnaire for behaviour).17 The general questionnaire includes an informed consent sheet for granting permission for the child’s participation in the ABCD health check. Two weeks before the health check, mothers are notified by letter and receive an additional self-administered food frequency questionnaire (FFQ). The FFQ was developed by TNO Food (Zeist, The Netherlands). It consists of 71 food items for which the frequency of consumption and portion size are to be reported. A validation study comparing the FFQ with the gold standard of doubly labelled water showed that the questionnaire is a valid instrument for estimating mean energy intake in a group of 4- to 6-year-old children and for ranking subjects according to their energy intake; the Pearson correlation coefficient between energy intake and energy expenditure was 0.62.18 For children living in Amsterdam, the health check is held at the child’s primary school. Children and parents who have moved outside of Amsterdam and children enrolled in small schools with no space available for the health check are invited to a central location during the weekend and holidays. The health check itself consists of a fasting capillary blood sample for assessing glycaemic control and lipid profile (glucose, C-peptide, total cholesterol, high-density lipoprotein, low-density lipoprotein, triglycerides) as well as fatty acid profile, physical measurements and a cognitive test battery. The physical measurements include anthropometric measurements (height, weight, waist and hip circumference), body composition measures (fat mass and fat-free mass) by bioelectrical impedance analysis, blood pressure in supine and sitting positions and cardiovascular function measures (heart rate, heart rate variability and pre-ejection period) in supine and sitting positions using the VU Ambulatory Monitoring System.19 Cognitive function is examined using four tasks from the Amsterdam Neuropsychological Tasks (ANT) programme. These tasks estimate sensory-motor speed, visuomotor skills and response selection.20 A full list of publications is available on our website (http://www.abcd-study.nl). The main findings to date are summarized below. Data collection for Phase III is ongoing, and papers are in preparation. Factors found to be related to adverse pregnancy outcomes (stillbirth or miscarriage, low birth weight or preterm birth) include having high levels of thyroid-stimulating hormone,21 folate depletion,22 an adverse fatty acid profile,23 smoking, overweight/obesity, high blood pressure and a low educational level,24,25 living in a poor neighbourhood26 and having work-related stress.27 Maternal stress and mood disorders, including work-related stress as well as pregnancy-related anxiety or depression, were associated with continuing to smoke during pregnancy28 and an increased risk of having an excessive crying infant.14 Any association was found neither between maternal stress and mood disorders and preeclampsia or gestational hypertension, nor between maternal physical activity and these outcomes.29,30 Maternal pre-pregnancy BMI was associated with infant BMI at age 14 months.31 In the above-mentioned pregnancy outcomes, large differences were observed between ethnic groups. Particularly among women from Surinamese, Antillean and Ghanaian background, low birth weights and preterm births were more prevalent.24,25 Ethnic disparities were also observed in the start of prenatal care (i.e. the timing of the first visit to a care provider),32 use of folic acid supplements,33 maternal thyroid function34 and the previously mentioned risk factors.24,25,27,35 The observed disparities in birth weight and preterm birth were explained in part by these risk factors and their accumulation.24,25 The study’s main strengths relate to its unique design and mode of operation. It was implemented during routine care, which allowed for detailed measurement in early pregnancy and, currently, for detailed follow-up of children’s growth and development. Collecting maternal blood samples in conjunction with routine blood collection for screening purposes made it possible to use biomarkers to measure status in early for a more objective measure of and have the of not from culture-specific a for our A of this design to the in exposure in pregnancy, which we not However, our to on early pregnancy was on evidence that the of development is at this In addition, are to be (e.g. women with the conditions going on The response rate of the study Phase was which is lower for large population-based pregnancy in the such as the and for the study in The Netherlands Particularly in an response rates to large studies are our response rate may be measures were taken to enhance of ethnic minority women in questionnaires and completing them orally with trained mentioned participation could not be this in a degree of selection the current response rate is 66% (Phase III, preliminary we to more of an to include all ethnic minority groups in the the study more trained who can the ethnic minority women with the questionnaires in their own language. However, the costs to do so are and in The Netherlands are not to studies at the that is So far, the study has with a of to a small group of and In this the of the obstetric care YHC, and also the primary schools have been to the data the study has been a between the Public Health Service and the university medical centres from the it both and health This is an important in of the study’s to health care and Results of the study are to health care obstetric care and on a basis and for to a prenatal educational for pregnant women which on smoking and psychosocial The ABCD study collaboration and the interest of A of for in using data for research purposes is available at http://www.abcd-study.nl. In we a research that include information on the background, research questions and a for and Approval of a upon the and the of the are the of research in then may be or to are established a which includes delivery, for and to the ABCD For more information, our website or one of the at The ABCD study has been by from The Netherlands for Health Research and Development The Netherlands and Center for the additional provided by the University of research for the 5-year the Institute for Public Health and the blood blood and and ABCD health check for children living outside of The authors all participating mothers and their children for their time and and are to all obstetric care YHC and primary schools for in the of the ABCD study. The authors data for of