Cohort Profile: The Triple B Pregnancy Cohort Study: A longitudinal study of the relationship between alcohol, tobacco and other substance use during pregnancy and the health and well-being of Australian children and families

Abstract

The Triple B Pregnancy Cohort Study investigates the effects of parental alcohol, tobacco and other substance use on infant development and family functioning. The study (also known as: Bumps, Babies and Beyond), recruited two sub-samples: (i) a general antenatal clinic sample of pregnant women and their partners (n = 1534 women; 842 of their partners); and (ii) a smaller sample of pregnant women with diagnosed substance use disorders (SUD; n = 89 women only). Participants were recruited through public antenatal clinics attached to major hospitals and area health services in New South Wales (NSW) and Western Australia (WA). Of 4068 eligible women from the general antenatal clinics, 37.7% participated, with equivalent numbers for the SUD clinics being 198, 44.9%. There were 1453 and 65 live births from the two groups, respectively, with a total of 1414 and 65 mothers in the two groups, respectively. Data were also collected on 1264 (86.9%) of 1455 eligible partners of women recruited through the general antenatal clinics. The study collected repeated measures across pregnancy (trimesters 1, 2 and 3), and at 8 weeks and 12 months postnatally; retention at 12 months was 84.0% and 73.8% for mothers in the general antenatal and specialist SUD clinics, respectively. The data collected include demographic, parental, familial and infant factors, with a focus on parental substance use and mental health, parenting practices, familial functioning and infant development. Following pregnancy awareness, 42% of women consumed alcohol, 12% smoked tobacco and 4% used illicit drugs at some stage in pregnancy. Comprehensive assessments have been conducted with infants at 12 months to test numerous developmental domains, including cognitive, motor and language skills, along with measures of social and emotional functioning. Data access enquiries can be made to the principal investigator [delyse.hutchinson@deakin.edu.au]. In 2010, the Australian National Medical Health Research Council (NHMRC) funded this longitudinal pregnancy cohort study to address limitations in knowledge of the effects of parental substance use on infants and families. The Triple B Pregnancy Cohort Study (known as: Bumps, Babies and Beyond), includes two sub-samples: (i) a public antenatal clinic sample of pregnant women and their partners, and (ii) a smaller sample of pregnant women with diagnosed substance use disorders. Participants were recruited though public antenatal clinics at major hospitals and health services in New South Wales (NSW) and Western Australia (WA). The study aims to examine: the effects of parental substance use in pregnancy on infant development (e.g. cognitive, motor, language and socio-emotional development) and family functioning; the extent to which substance use is interrelated among couples; and the influence substance use may have on parents’ respective substance use patterns. The study aims to contribute to improved public health services. The study represents a significant investment in longitudinal research, which builds on a history of quality Australasian longitudinal research. The Developmental Origins of Health and Disease (DOHaD)1 framework posits that early life environmental exposures induce critical changes in development which have long-term impacts on health and disease risk. Consistent with this framework, clinical studies of high-risk samples of parents diagnosed with substance use disorders suggest that parental substance use has significant adverse impacts on infant and child development.2,3 For example, substance-dependent pregnant women and their infants have an increased risk of obstetric, fetal and neonatal complications. These include miscarriage, prematurity, low birthweight for gestational age, fetal alcohol spectrum disorders (FASD), neonatal abstinence syndrome (NAS) and longer-term deficits in children’s physical, cognitive, behavioural and emotional development.4–7 The primary limitation of clinical studies is that the data are often derived from small-scale cross-sectional surveys, so that the effects over time are not known and the findings are not generalizable to the population. Longitudinal studies provide insights into potential causes of health problems, as well as factors capable of preventing or moderating problems. Emerging evidence from general population-based longitudinal studies suggests that the effect of parental substance use can vary considerably as a function of parent gender, pattern and type of substance use, and the presence of associated socioeconomic, physical health and psychosocial risk factors.2,3,8–11 Parental substance misuse may impact adversely on children and families via direct and indirect pathways. For example, parental substance use can impact negatively on the quality of the marital/intimate partner relationship and the parent-child relationship by reducing family cohesion and increasing conflict and violence.12,13 However, there is a dearth of data on parental substance use in the prenatal period, particularly defined by timing of exposure during pregnancy (i.e. pre-pregnancy awareness and at each trimester following awareness), with most research focusing solely on maternal consumption and neonatal outcomes. Furthermore, many studies use poor measures of parental substance use (e.g. retrospective reports of use). Notably, few assess partner substance use and its impacts. As a result, there are major gaps in current knowledge about the extent to which perinatal (pre- and postnatal) parental substance use impacts on early child development and family functioning, and the mechanisms by which these influences occur, particularly for low/moderate levels of use which are the most frequent levels of substance use among Australian parents.14 This cohort protocol aims to: (i) describe the Triple B Cohort samples and study methodology; and (ii) provide key summary data on perinatal substance use patterns, along with findings from three recent publications. The Triple B Cohort Study is led by the National Drug and Alcohol Research Centre (NDARC) at the University of New South Wales (UNSW) Australia and the National Drug Research Institute (NDRI) at Curtin University, in collaboration with Deakin University, Sydney University, the University of Queensland, the Murdoch Childrens Research Institute and the University of Melbourne. The study has been supported by the NHMRC (2010–14), Australian Rotary Health (ARH; 2012–13) and the Foundation for Alcohol Research and Education (FARE; 2010–11). Additionally, PhD candidates on the project have been funded through ARH; the NDARC Education Trust (NET), Macquarie University and both the Australian Centre for Perinatal Science (ACPS) and NDARC at UNSW. Ethics approval was granted by relevant university, hospital and health services human research ethics committees. The study used a prospective cohort design. Pregnant women were recruited through general antenatal clinics (N = 1534: NSW = 1246; WA = 288) and specialist drug and alcohol antenatal clinics (N = 89; NSW = 59; WA = 30) in public hospitals and health services in NSW and WA between 2009 and 2013. Participating hospitals in NSW were the Royal Prince Alfred Hospital, Camperdown; the Royal Women’s Hospital, Randwick; and Liverpool Hospital, Liverpool. Participants in Western Australia were recruited through the King Edward Memorial Hospital, Subiaco. Pregnant women were invited to participate in the study by research officers who attended antenatal clinics at each hospital, across all days and months of the year, to represent (proportionally) all clinics operating at each recruitment site. A standardized script was used to describe the study to women. Eligibility criteria included: being pregnant; being over 15 years of age; having no major medical complications (mother or fetus); intention of mother or both parents to be the primary caregiver/s; being mentally able to complete assessments; possessing sufficient literacy in English; and informed consent. As outlined in Figures 1 and 2, 6597 pregnant women [255 from the specialist substance use disorder (SUD) cohort] were approached and informed about the study; 4266 (198 from the specialist SUD cohort) of these met eligibility criteria and were invited to participate. A total of 1534 (37.7%) and 89 (44.9%) women from the general and specialist cohorts, respectively, provided consent and completed at least one study measure. The participation rates of eligible women for individual hospitals ranged from 22.2% to 55.0%, with an overall participation rate of 44.0% for NSW and 25.0% for WA. Triple B Pregnancy Cohort Study: study flow diagram of mother, partner and infant participation rates (N = 1534 families) for the general antenatal sample. *The 8-week follow-up interview for partners was introduced after the pilot study. As such, 8-week data were unavailable for 60 participating partners, as it was not offered. **In some families, only infants were assessed at 12 months. Triple B Pregnancy Cohort Study. Study flow diagram of mother and infant participation rates (N = 88) for the specialist substance use disorder (SUD) antenatal sample. **In some families, only infants were assessed at 12 months; in total, 50 families remained in the cohort. Figures 1 and 2 present study flow diagrams of mother and infant participation rates, and of partners, separated by non-exposed and substance use-exposed groups, respectively. Of those who participated, 79 withdrew (six from the specialist SUD cohort), and a further 65 (18 from the specialist cohort) were lost to follow-up before giving birth (attrition rate: 7.8% and 31.5% for general and specialist SUD cohorts, respectively). The remaining 1414 mothers in the general cohort gave birth to a total of 1453 offspring which included 37 twin pairs and one set of triplets. In the specialist SUD cohort, 65 singletons were born. Parents of twins and triplets were asked to complete a separate survey about each child. Data were collected on 1264 (86.87%) of eligible partners from the general cohort, either directly (n = 842), or indirectly (n = 422) via maternal report. The characteristics of participating mothers and infants are presented in Table 1. Partner characteristics are presented in Table 2. Maternal and partner data were collected in trimester 3; data on the infant were collected at the 8-weeks postnatal follow-up and were derived from infant Blue Books completed at birth by hospital staff. Compared with the Australian population, the Triple B general antenatal cohort is similar to the Australian population on rates of employment (z = 1.40; P = 0.08), and the proportion of participants of Aboriginal or Torres Strait Islander origin (z = −0.53; P = 0.23). The sex distribution of infants was also similar to Australian population figures (z = −0.46; P = 0.32), although other infant characteristics differed from the general population, with longer gestation at birth (z = 13.8; P < 0.001; Somers’ d = 0.000; 95% CI −0.066, 0.068), higher birthweight (z = 3.82; P = 0.001; Somers’ d = 0.07; 95% CI −0.01, 0.13) and a higher proportion of twins/multiple births (z = 4.55; P < 0.001; Cohen's h = 0.12; 95% CI 0.042, 0.186). Mothers in the cohort were also older (z = 12.86; P < 0.001, Somers’ d = 0.11; 95% CI 0.05, 0.17), more socioeconomically advantaged (SEIFA; t1577 = 31.56; P < 0.001; Cohen’s d = 95% CI and the general population = P < 0.001; Cohen’s h = 95% CI In that there were more women (z = P < 0.001; Cohen’s h = 95% CI a higher proportion of women (z = P < 0.001; Cohen’s h = 95% CI and in (z = P < 0.001; Cohen’s h = 95% CI with general population and infant cohort characteristics and with Australian population data sample for each vary to and infant cohort characteristics and with Australian population data sample for each vary to Partner cohort characteristics and with Australian population data a sample for each vary to participating and partners, data are Partner cohort characteristics and with Australian population data a sample for each vary to participating and partners, data are the Triple B specialist SUD cohort higher levels of (z = P < 0.001; Cohen’s h = 95% CI higher of Aboriginal or Torres Strait Islander participants (z = P < 0.001; Cohen’s h = 95% CI higher of mothers (z = P = 0.001; Cohen’s h = 95% CI and (z = P = Somers’ d = 95% CI with the Australian population. Mothers were also (z = P = Somers’ d = 0.11; 95% CI 0.05, 0.17), = P < 0.001; Cohen’s h = 95% CI with more in (z = P < 0.001; Cohen’s h = 95% CI with the Australian population. There were no in (z = P = 0.08), the of births (z = P = 0.08), infant gestational (z = P = and infant sex distribution (z = P = between the specialist SUD cohort and the Australian population, although mothers in the specialist cohort were more socioeconomically advantaged = P < 0.001; Cohen’s d = 95% CI data were provided by of the 842 participating partners for mothers in the general antenatal cohort. In mothers in this partner characteristics for of the eligible partners who to participate. with Australian population data suggest that the proportion of partners of Aboriginal or Torres Strait Islander origin was in the general population (z = P = However, partners in the cohort to be older (z = P < 0.001; Somers’ d = 95% CI more = P < 0.001; Cohen’s h = 95% CI and more to be with the general population (z = P < 0.001; Cohen’s h = 95% CI In the rate of employment was higher among partners in the cohort (z = P < 0.001; Cohen’s h = 95% CI and there was a higher proportion of partners in with the general population (z = P = Cohen’s h = 95% CI were also conducted on the characteristics of partners from the general cohort as a function of data were via or indirectly via maternal report. These that the two partner not on employment in or = = P = partner for partners who and those who were on = = P = or Aboriginal or Torres Strait Islander origin = = P = participating partners were = P = Cohen’s d = 95% CI and higher completed = = P = 0.001; = 95% CI are in Table These trimester 1 to 12 trimester 2 weeks to trimester weeks to and an 8-week follow-up weeks A developmental follow-up at infant 12 months. Mothers were assessed at all time partners at trimester 8 weeks postnatal and the and infants at the 8-week and rates for eligible mothers and infants are presented in 1. and of the Triple B Pregnancy Cohort Study In women participation after trimester 1 or 2, pregnancy assessments were completed for and of the Triple B Pregnancy Cohort Study In women participation after trimester 1 or 2, pregnancy assessments were completed for across the of data for the general antenatal cohort has been low Of the mothers remaining in the cohort following withdrew or were lost to further that the total rate at 12 months from the cohort of 1534 was (i.e. 84.0% Of the infants included in the developmental data were collected from at the Of the 842 general antenatal cohort partners who directly in the withdrew or were lost to follow-up at 8 and a further at 12 months in a total retention rate of rates for mothers from the specialist SUD drug and alcohol antenatal clinics were higher for the general antenatal population = = P < 0.001; = 95% CI infant mothers remained in the study by 12 15 participants or were lost to a retention rate of from the cohort. Mothers in the both who withdrew or were lost to follow-up by 12 months were those who cohort = P < 0.001; Cohen’s d = 95% CI specialist cohort = Cohen’s d = 95% CI and higher rates of = = P < 0.001; = 95% CI In participants in the general cohort who withdrew or were lost to follow-up at 12 months through their for = P < 0.001; Cohen’s d = 95% CI and were more to be from WA from NSW = = P < 0.001; = 95% CI of infant data for both the general and the specialist SUD antenatal cohort who complete the development with those who no in sex = = P = = = P = birthweight = P = = P = = P = = P = and gestation at birth = = = = Table a of the measures included at each partner and infant measures at each partner and infant measures at each Data on alcohol and substance use in pregnancy and the postnatal are presented in Table of alcohol, tobacco and illicit substance use during pregnancy were in the pregnancy awareness, and considerably after pregnancy awareness in the general cohort 1 before = = P < = 95% CI = = P < 0.001; = 95% CI = = P < 0.001; = 95% CI For women in the specialist cohort, in alcohol and illicit substance use following pregnancy awareness were also in trimester 1 = = P < 0.001; = 95% CI = = P < 0.001; = 95% CI use, remained = = P = before and after pregnancy awareness in trimester Alcohol and other substance use in the Triple B Cohort during pregnancy and following is as and were introduced after the study not included for n = 8 women who no data as were of their pregnancy data were not for women who not were pregnant in trimester 1. Alcohol and other substance use in the Triple B Cohort during pregnancy and following is as and were introduced after the study not included for n = 8 women who no data as were of their pregnancy data were not for women who not were pregnant in trimester 1. following pregnancy awareness and during the of and of women consumed alcohol, and smoked and and consumed illicit drugs in the general and specialist groups, respectively. The of alcohol consumption in the sample was two in the specialist cohort and one in the cohort. Alcohol was with that of the general population for the two tobacco and illicit drug were in the general cohort higher for the specialist with the general population during pregnancy. A of have been on the Triple B the and of alcohol consumption by women before awareness of and in alcohol use following pregnancy and were in the early weeks of before pregnancy and respectively). the rate of was to be considerably higher the pregnancy is into associated with changes in alcohol use following pregnancy included of alcohol use pregnancy older maternal age, pregnancy and illicit substance were more to low or at low or levels were more to at the were to to The have to health and measures to to mothers and their In to the there is a dearth of prospective data on women by substance use disorders during the perinatal period, often to with recruitment and the of these women at this time is critical to perinatal services to maternal and infant development. prospective study on the SUD that these women psychosocial to their in and levels of perinatal and postnatal parenting the critical of and parenting for these high-risk pregnant women and their The study of it the most longitudinal of substance use in the perinatal to in Comprehensive during this knowledge of the impact of substance use on infant and critical risk and the study into substance use both before and after pregnancy a often in research. this is the study pregnant women recruited from a general antenatal clinic and a substance for substance use to be across a spectrum from to This of the psychosocial and physical risk factors from levels of substance the study is the to assess the impact of the substance use and mental health, both and on child health and family also the influence partners have on each substance of from infants 8 weeks and 12 and parents infant 12 provide a for research into factors individual in risk for substance use in and adverse effects of parental substance use on also that samples were not the potential of studies to developmental and the effect of pregnancy exposure to these there is potential for data with cohort For example, major of the Triple B Cohort assessments have been with the Australian Study and the Health Cohort Study This has the potential to an of cohorts, each focusing on prenatal and the Triple B Cohort provide and data on exposures in pregnancy. The and have a antenatal in trimester data across three and two respectively. The Triple B Study also key of substance use for which there are major public health, or it substance use, through pregnant women in for substance use problems, and low to alcohol and tobacco use, which are in the antenatal clinic sample on the low of illicit substance use in it is that Triple or other be able to the impacts of low to or use on also that the sample is for not research. the Triple B Study is a study conducted in two it was conducted in public hospitals and is not of of of families that use The and of data with other major and may for increased data and the potential to of in smaller A further potential limitation of the study to the and of from the that the general antenatal cohort in a of from the general population, and there was evidence to suggest that was higher among women in the cohort includes participants from a of and with substance use patterns, and overall was In as substance use in the general antenatal cohort during pregnancy was with Australian population A major focus of the study was on data in the antenatal as such, there is a longer between the 8-week an This the to early postnatal exposures may and development. that on some key developmental and for was assessed in the via recent retrospective also that rates for the study ranged from to these rates are with some recent longitudinal with recruitment in the limitation is that risk may of the the of the distribution is to be of the including substance use was via and is to potential In to address this participants were for during their trimester of to illicit substance between substance use and was that the provided was these the study of the effects of parental substance use on infants and families, which can direct health The can development of public health and early to parents to informed about substance use during the prenatal The also the health and of pregnant women by of substance in these may developmental in infant and family functioning in the The of this study can also on use of alcohol and other before in pregnancy and which may public health and on substance can be through the National Drug and Alcohol Research University of New South Wales can be to Data access is by the Research be with approval and and data The study protocol for research by the respective ethics with the research. in a The Triple B Pregnancy Cohort Study is a longitudinal pregnancy cohort on the impacts of parental alcohol and other drug use in pregnancy and on infant development and family functioning. The study recruited two sub-samples: (i) a general antenatal clinic sample of pregnant women and their partners (n = 1534 women; of their partners); and (ii) a smaller sample of pregnant women with diagnosed substance use disorders (n = 89 Participants were recruited in at antenatal clinics in and The sample has been through the gestational with assessments in 1, 2 and and at 8 weeks and 12 months postnatally; retention at 12 months was 84.0% and 73.8% for mothers in the general antenatal and substance use disorder clinics, respectively. The data collected include demographic, parental, familial and infant factors, with a focus on parental substance use and mental health, parenting practices, familial functioning and infant development. For on collaboration with the Triple B Cohort The research was funded by an Australian National Health and Medical Research Council (NHMRC) for to and and was supported by the National Drug and Alcohol Research Centre University of New South Wales NDARC and the National Drug Research Institute Curtin University, are funded by the Australian the and also from Australian Rotary the Foundation for Alcohol Research and and the Foundation for is supported by an NHMRC Research from the and is supported by an Australian Partner Research is supported by an Australian Research Council Research is supported by an NHMRC the NDARC and research and who with of the the hospitals and antenatal clinics for their with and the study participants and their families. to and also to the Research The Triple B Research includes the primary and of