bs_bs_banner

250

doi: 10.1111/ppe.12183

Study Design Article

The Odense Child Cohort: Aims, Design, and Cohort Profile Henriette Boye Kyhl,a Tina Kold Jensen,f Torben Barington,b,c Susanne Buhl,g Lene Annette Norberg,h Jan Stener Jørgensen,d Ditlev Frank Granhøj Jensen,e Henrik Thybo Christesen,a Ronald F. Lamont,d,i Steffen Husbya a

Hans Christian Andersen Children’s Hospital b

Department of Clinical Immunology

c

Odense Patient data Explorative Network (OPEN) d

Department of Gynecology and Obstetrics

e

Department of Budget and Data, Odense University Hospital

f

Department of Environmental Medicine, University of Southern Denmark

g

Odense City Government and Administration, Department for Child and Family, Centre for Health

h

Department of Youth and Adolescent, Elderly and Disabled, Odense City Government and Administration, Odense C, Denmark i

Division of Surgery, Northwick Park Institute for Medical Research Campus, University College London, London, UK

Abstract Background: The importance of the environment on the development of the fetus and infant throughout early life is increasingly recognised. To study such effects, biological samples and accurate data records are required. Based on multiple data collection from a healthy pregnant population, the Odense Childhood Cohort (OCC) study aims to provide new information about the environmental impact on child health by sequential follow-up to 18 years of age among children born between 2010 and 2012. Methods: A total of 2874 of 6707 pregnancies (43%) were recruited between January 2010 and December 2012. Three hundred seventy-four have since left the study, leaving 2500 active families. The non-participants act as controls contributing data through local registries. Biological material, questionnaires, and registry data were compiled. Anthropometric data and other physical data were collected. Results: Two thousand five hundred families actively participated in the study with 2549 children. Sixty-four per cent of the fathers and 60% and 58% of the mothers, respectively, donated a blood sample at 10 and 28 weeks of gestation. On average, 69% completed questionnaires, 78% of the children were regularly examined, and had a blood sample taken (46%). The participating pregnant women differed from the non-participants in several respects: age, body mass index, smoking, parity, education, and ethnicity. The infants were comparable with respect to gender and mode of delivery. Conclusions: The OCC provides material for in-depth analysis of environmental and genetic factors that are important for child health and disease. Registry data from non-participating women and infants are available which ensures a high degree of comparable data. Keywords: birth, child, cohort study, design. The Odense Child Cohort (OCC) is a joint research project in which children born during a 3-year period within the Municipality of Odense, Denmark will be monitored from fetal life to adulthood. The overall aim of the project was to obtain information on the importance of environmental factors on the fetus, Correspondence: Steffen Husby, Professor of Paediatrics, Hans Christian Andersen Children’s Hospital, Odense University Hospital, DK-5000 Odense C, Denmark. E-mail: [email protected]

infant, and child. Furthermore, the interaction between social, environmental, and biological determinants in health and disease will be studied. The aims of the study were: 1 To obtain accurate, validated information about known risk factors of lifestyle diseases that can be targeted for intervention and prevention; 2 To seek new physical, chemical, and social environmental factors that might contribute to the subsequent development of disease; 3 To disseminate the knowledge gained. © 2015 John Wiley & Sons Ltd Paediatric and Perinatal Epidemiology, 2015, 29, 250–258

The Odense Child Cohort This paper outlines the design, structure, and data collection incorporated into the OCC study to achieve its aims. Data obtained through the comprehensive collection of biological specimens and questionnaires from biological parents and their children (completed during pregnancy and at follow-up during infancy and childhood) can be linked with Danish National Databases to provide investigators with a unique opportunity for research. In addition, the Municipality of Odense provided a range of social data on both parents and children including information on socio-economic status and use of social services. The experience of other child cohorts was taken into consideration when planning the OCC.1–4 Adherence to the STROBE guidelines5 ensures robust methods of recruitment such as eligibility criteria, follow-up, and comparison to non-participants and outcome.

251

women to participate in the project: (i) midwifes, ultrasonographers, and obstetricians in the Obstetrics and Gynaecology Department of Odense University Hospital (OUH) were provided with details of the project and given instructions on how to encourage participation, (ii) posters and flyers advertising the OCC study were displayed in local pharmacies, libraries, and in general practitioners’ waiting rooms, (iii) the hospital sent out details of the study at the same time as information on ultrasound examinations, (iv) local and national media coverage of the project was asked for, and (v) as of 1 March 2012, district nurses and health visitors distributed written information on the study during their postnatal home visits.

Focus on high-risk pregnancies Methods Setting The OCC study was initiated in the Municipality of Odense in the Region of Southern Denmark which has a social distribution comparable to the rest of Denmark. The population comprises 192 000 inhabitants, [mainly ethnic Danish (85%)] including 49 963 women of reproductive age (15 to 49 years of age) providing 2100–2300 births per year (2010–13). Adolescent mothers (under 19 years of age) account for 0.53% of these births, and most of the women deliver in hospital with less than 1% of women delivering at home.

Enrolment Inclusion required a newly diagnosed pregnancy before 16 completed weeks of gestation between 1 January 2010 31 and December 2012 and resident status in the Municipality of Odense throughout pregnancy and childbirth. From 1 March 2012, criteria were expanded to allow inclusion at any time during pregnancy and up to the postnatal age of 2.5 months. Since recruitment took place over a 3-year period, some women had more than one eligible pregnancy. Women were excluded if they moved out of the municipality before birth. All pregnant women who met the inclusion criteria were informed about the study by project midwives or coordinator. A variety of methods were used to motivate eligible pregnant © 2015 John Wiley & Sons Ltd Paediatric and Perinatal Epidemiology, 2015, 29, 250–258

As of 1 May 2011, changes in the official Danish policy for pregnancy surveillance was changed and defined six high-risk pregnant women that had to be followed in special midwifery clinics at OUH: (i) body mass index (BMI) above 27 kg/m2, (ii) multiple gestation, (iii) history of complicated pregnancy, (iv) age below 20 years, (v) abuse by parents, and (vi) psychiatric conditions (depression, stress, or other disorders). We were therefore able to recruit pregnant women in groups (i)–(iv) before or immediately after their first midwifery consultation. Pregnant women in groups (v) and (vi) were contacted after their second midwifery consultation (normally between 18.–22. GA). This was a compromise in order to take into consideration the issues that these women have and the difficulties of recruiting them. The first blood sample was omitted for these women. In order to recruit a representative sample of pregnant women from the Municipality of Odense, special efforts were made to recruit participants from a non-western background. Between 1 August 2010 and 7 October 2013, data from non-participants were extracted anonymously from birth registries.

Sample size considerations The goal of the cohort was to elucidate research questions related to public health and common diseases with prevalence in the general population at or above 5–10%. We anticipated that such questions could be asked with an effective cohort sample size of

252

H. B. Kyhl et al.

2500–3000. In the sub-projects, a range of different research questions have been asked.

Collection of biological specimens and clinical measurements Parental examination and blood sampling Following enrolment, women were encouraged to donate a blood sample (a total of 75 mL of serum, plasma, and buffy coat) before 16 completed weeks of gestation. Fathers were requested to donate a blood sample (32.5 mL of serum, plasma, and buffy coat) or a hair sample (matchstick thickness) at any time before the birth. The next examination was arranged for 27 to 28 completed weeks of gestation. Pregnant women with suspected gestational diabetes because of glycosuria had an oral glucose tolerance test (OGTT) performed (group 1). The remainder were randomised into one of two groups: (i) those who had a fasting blood glucose carried out (group 2), and (ii) a control group matched by gestational age who did not have glycosuria but had an OGTT (group 3). This group was matched by weeks of gestation. The protocol for group 1 and 2 was established by the subproject Metabolic risk factors for polycystic ovarian syndrome (Table 1, iv). Serum and plasma were saved from all groups. Table 1. Studies already initiated using The OCC Bio Bank and Database Exposure to endocrine disrupting chemicals in utero and reproductive disorders in offspring.8–10 ii) Vitamin D and its hormonal and nutritional effects on pregnancy and infants.6,7 iii) The influence of social and biological factors on duration of breast feeding. iv) Metabolic risk factors for polycystic ovarian syndrome (PCOS). v) Development of language among Danish children. vi) Ethnic differences in the use of antenatal care/prenatal diagnosis. vii) The significance of epigenetics for the link between conditions in early life and health late in life. viii) Gene–environmental–social interactions in the development of attention deficit hyperactivity disorder. ix) Tracking of alcohol consumption during pregnancy and long-term effects on the mother, child, and family. x) Infections during childhood (0 to 3 years of age) xi) Oral health and mineralisation disturbances in child teeth. xii) Reference values in blood samples from healthy children. i)

An aliquot of urine was collected (10–12 mL) from all groups together with a 24-h urine collection (from which a 100 mL aliquot was saved to the bio bank) from groups 2 and 3. Table 2 provides an overview of the timing of sample collection. Data on pregnancy outcomes are provided by medical registers and charts.

Samples at birth Blood (serum, plasma, and buffy coat) from the placenta and umbilical cord was collected after delivery (67 mL). A 5-cm length of umbilical cord was collected and stored at +5°C then sent to the bio bank within 48 h to be stored at −80°C. If sufficient at birth, a small sample of neonatal hair (matchstick thickness) was collected. All the material was collected by midwives or social and health care workers on the maternity ward.

Child examinations and blood samples The children were and will be examined at 3 and 18 months and 3, 5, 7, 9, 12, 15, and 18 years of age. Weight, length, congenital malformations, blood pressure, skin fold thickness (triceps and subscapular), skin colour (a six-point scale from very white/red to black skin; Fitzpatrick Skin Type Classification), genital development (measurement from anus to the genitals) and craniotabes (a softening or thinning of the skull) were and will be recorded at each visit. All examinations and blood tests are voluntary and participants can refuse to submit with no consequences for their future participation in the project. On each occasion, parents are asked to give permission for a blood sample to be taken from their child. If permission is granted, a venous blood sample (1 Education level Lowerd Intermediatee Higherf Ethnic background Danish Western Non-western

0.03

0.53 0.03

– 0.01 0.21 0.33

– 0.06

– 0.33

Non-participants n = 4322 29 84 676 1430 1301 831 n = 4245 23.4 193 2533 494 1025 n = 4322 527 3795 n = 4322 1 1929 1608 785 – – – – – – – –

– (26–33) 1.9% 15.6% 33.1% 30.1% 19.2% – (21.1–26.8) 4.5% 59.7% 11.6% 24.1% – 12.2% 87.8% – (0–1) 44.6% 37.2% 18.2% b

71.0% 17.1% 7.5% c

81.3% 7.0% 11.6%

P valuea

The Odense Child Cohort: aims, design, and cohort profile.

The importance of the environment on the development of the fetus and infant throughout early life is increasingly recognised. To study such effects, ...
164KB Sizes 1 Downloads 11 Views