Article Text

Download PDFPDF

‘Joining the Dots: Linking Prenatal Drug Exposure to Childhood and Adolescence’ – research protocol of a population cohort study
  1. Kate Lawler1,
  2. Mithilesh Dronavalli1,2,
  3. Andrew Page2,
  4. Evelyn Lee3,4,
  5. Hannah Uebel1,5,
  6. Barbara Bajuk6,
  7. Lucinda Burns7,
  8. Michelle Dickson8,
  9. Charles Green9,
  10. Lauren Dicair10,
  11. John Eastwood11,12,
  12. Ju Lee Oei1,13
  1. 1School of Women's and Children's Health, Faculty of Medicine and Health, University of New South Wales, Randwick, New South Wales, Australia
  2. 2Translational Health Research Institute, Western Sydney University, Penrith, New South Wales, Australia
  3. 3Centre for Social Research in Health, University of New South Wales, Randwick, New South Wales, Australia
  4. 4Centre for Economic Impacts of Genomic Medicine, Macquarie University, North Ryde, New South Wales, Australia
  5. 5Sydney Children's Hospital Randwick, Randwick, New South Wales, Australia
  6. 6Critical Care Program, Sydney Children's Hospitals Network Randwick and Westmead, Westmead, New South Wales, Australia
  7. 7National Drug and Alcohol Research Centre, University of New South Wales, Randwick, New South Wales, Australia
  8. 8The Poche Centre for Indigenous Health, The University of Sydney Faculty of Medicine and Health, Sydney, New South Wales, Australia
  9. 9Alpha Maxx Healthcare, Memphis, Tennessee, USA
  10. 10Private Practice, Havertown, Pennsylvania, USA
  11. 11National Public Health Service, Te Whatu Ora–Health New Zealand, Dunedin, New Zealand
  12. 12School of Population Health, University of New South Wales, Randwick, New South Wales, Australia
  13. 13Department of Newborn Care, Royal Hospital for Women, Randwick, New South Wales, Australia
  1. Correspondence to Dr Ju Lee Oei; j.oei{at}unsw.edu.au

Abstract

Introduction Prenatal drug exposure (PDE) is one of the most important causes of child harm, but comprehensive information about the long-term outcomes of the families is difficult to ascertain. The Joining the Dots cohort study uses linked population data to understand the relationship between services, therapeutic interventions and outcomes of children with PDE.

Methods and analysis Information from routinely collected administrative databases was linked for all births registered in New South Wales (NSW), Australia between 1 July 2001 and 31 December 2020 (n=1 834 550). Outcomes for seven mutually exclusive groups of children with varying prenatal exposure to maternal substances of addiction, including smoking, alcohol, prescription/illicit drugs and neonatal abstinence syndrome will be assessed. Key exposure measures include maternal drug use type, maternal social demographics or social determinants of health, and maternal physical and mental health comorbidities. Key outcome measures will include child mortality, academic standardised testing results, rehospitalisation and maternal survival. Data analysis will be conducted using Stata V.18.0.

Ethics and dissemination Approvals were obtained from the NSW Population and Health Services Research Ethics Committee (29 June 2020; 2019/ETH12716) and the Australian Capital Territory Health Human Research Ethics Committee (11 October 2021; 2021-1231, 2021-1232, 2021-1233); and the Aboriginal Health and Medical Research Council (5 July 2022; 1824/21), and all Australian educational sectors: Board of Studies (government schools), Australian Independent Schools and Catholic Education Commission (D2014/120797). Data were released to researchers in September 2022. Results will be presented in peer-reviewed academic journals and at international conferences. Collaborative efforts from similar datasets in other countries are welcome.

  • Child Abuse
  • Epidemiology
  • Infant
  • Health services research

Data availability statement

No data are available. Data are only available through ethics approval and affiliation with the University of New South Wales.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

WHAT IS ALREADY KNOWN ON THIS TOPIC?

  • Children with prenatal drug exposure (PDE) are at higher risk of harm including death, poor physical and mental health, maltreatment and academic failure.

WHAT THIS STUDY HOPES TO ADD?

  • Joining the Dots is longest follow-up of the largest cohort of children with PDE using data-linkage, which is cost-effective and efficient to identify potential interventions.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY?

  • Joining the Dots will be a platform for collaboration with international datasets to enable global strategies for preventing harm to children with PDE.

Introduction

The Joining the Dots: Linking Neonatal Abstinence to Childhood and Adolescence cohort study (Joining the Dots) was established to investigate long-term outcomes of children who developed withdrawal or neonatal abstinence syndrome (NAS) after prenatal drug exposure (PDE). Over recent decades, maternal drug use, in particular opioids, has escalated worldwide, resulting in an up to 300% increase in the numbers of infants diagnosed at birth with NAS.1 NAS refers to the constellation of physiological manifestations in neonates associated with abrupt post-birth cessation of maternal transplacental drug transfer,2 most commonly from opioids. However, NAS may also be caused by non-opioid illicit substances and psychotropic medications, including antidepressants, anxiolytics and antipsychotics.3

Opioids were the first drugs related to NAS and opioid withdrawal syndrome in newborn infants was first described in 1875 as an almost certainly fatal condition.4 As recently as the 1950s, mortality rates for infants with NAS exceeded 30%.5 The long-term fates of this vulnerable and often hidden population of infants were, therefore, not under scrutiny. In the 1970s, efforts to standardise identification, diagnosis and treatment of NAS led to the development of clinical assessment tools. These tools, which include the Finnegan Neonatal Abstinence Severity Score—the most widely used NAS assessment tool today—led to rapid decline in fatalities from NAS.6 Newborn withdrawal has now become an uncommon direct cause of infant death and most infants with PDE will be discharged in a relatively healthy condition from hospital after birth.7

Unfortunately, longer-term outcomes, even beyond early childhood, of infants with PDE are unclear. They are often perceived as relatively healthy children once NAS resolves and seldom prioritised for long-term or intensive follow-up especially if they did not require treatment for NAS.8 Lack of follow-up is also compounded by difficulties in tracking this massive and often marginalised population of children. In Australia, for example, about 1–1.5% of children are estimated to be exposed to prenatal opioids,7 and as many as one in five children (~20 000) are prenatally exposed to other addictive agents, including alcohol, cigarettes, prescription medications and various illicit substances.9 Many of these children are also highly mobile, with up to 50% placed in out-of-home care (either temporarily or permanently) by the age of 5 years10 which makes tracking particularly difficult.

An increasing body of evidence now suggests that children with PDE are at risk of poorer outcomes than other children and that these problems persist into childhood.11 Notably, children with PDE are more likely to die and be rehospitalised.8 12–14 Those with a history of NAS are twice as likely to be developmentally delayed, and have difficulties with speech and language development, poorer school outcomes and increased incidence of mental health disorders.15–21 Additionally, PDE is more likely to be associated with confounders, including parental mental illness, low socioeconomic status and child protective service involvement22 which exacerbate risk of adverse outcomes. These confounders, along with the sheer number of children with PDE and difficulties in long-term follow-up, mean knowledge of effective intervention remains poor.

A critical need is evident for large-scale, contemporary data to enable analysis of long-term outcomes past infancy and illuminate protective factors to inform intervention and optimise outcomes. The Joining the Dots cohort study establishes a population dataset which tracks outcomes of children with PDE from birth to up to 20 years of age. The major objectives of Joining the Dots include the following:

  1. To investigate the association between PDE on a range of outcomes, including a child’s risk of mortality, mental and physical health, development, cognitive performance and behaviour.

  2. To investigate the lasting association of PDE on adolescents and young adults.

  3. To investigate the impact of environmental and socioeconomic factors, including rural residence, on the incidence of PDE and outcomes for children with PDE.

  4. To investigate the impact of child protective service interventions, such as out-of-home care (including foster care and relative/kinship care), on the aforementioned outcomes.

  5. To quantify the economic burden of PDE in Australia.

Methods and analysis

Study design

Joining the Dots employs a retrospective, population-based cohort design. The Joining the Dots cohort has been established using linked administrative data from a variety of databases (table 1). The primary database is the New South Wales (NSW) Perinatal Data Collection (PDC), which provides maternal and infant information from all births in NSW from 1 July 2001 to 31 December 2020 (n=1 834 550). The Centre for Health Record Linkage23 conducted record linkage between the PDC and multiple datasets using probabilistic matching of information by personal identifiers (ie, name, date of birth, addresses and hospital record numbers). A unique project person number (PPN) was created for each individual identified in the linkage, and this PPN was assigned to all datasets, replacing identifying information. De-identified databases were provided to the investigators for analysis.

Table 1

Summary of databases

In the case of differences or inconsistencies in information across databases due to overlap (ie, demographic information), information in the PDC supersedes that of other databases. The exception to this is in studies which involve data from the National Assessment Program Literacy and Numeracy (NAPLAN) database, where Australian First Nations status is ascertained from the NAPLAN database instead of the PDC. This is because the PDC identifies only maternal Australian First Nations status, whereas the NAPLAN database identifies Australian First Nations status by maternal and paternal status.

Cohort selection

The Joining the Dots cohort was based on diagnostic information contained in the NSW Admitted Patient Data Collection (APDC) database, which provided information on all hospital separations (discharges, transfers and deaths) in public and private hospitals in and outside NSW to NSW residents from 1 July 2001 to 31 December 2021. The APDC contained diagnoses contributing to episodes of care, coded according to the International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Australian Modification. Other information was obtained from linked datasets, including: NSW PDC–birth details; NSW Registry of Births, Deaths and Marriages and Cause of Death Unit Record File–death details; NSW Mental Health Ambulatory Data Collection (maternal outpatient mental health diagnoses); and NSW Family and Community Services Dataset–KiDS Data Collection (out-of-home care information).

Investigators divided the cohort into seven mutually exclusive groups (figure 1) based on their history of prenatal exposure to maternal substances of addiction:

  1. No known history of PDE either from infant or maternal records (n=1 622 813).

  2. A history of PDE (n=211 737), including:

    1. Maternal smoking only (n=188 571).

    2. Maternal alcohol abuse/dependence only (n=2806).

    3. Maternal smoking and alcohol abuse/dependence (n=1891).

    4. Maternal drug abuse/dependence and/or smoking and/or alcohol abuse/dependence, but child not diagnosed with PDE in infancy (n=11 183).

    5. Child diagnosed with PDE in infancy, but not NAS (P04.4, n=1268).

    6. Child diagnosed with NAS in infancy (P96.1, n=6018).

Figure 1

Cohort selection and grouping. ICD-10-AM, International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Australian Modification; NAS, neonatal abstinence syndrome; PDE, prenatal drug exposure.

Outcome measurement

Individual record linkage between available databases enables measurement of a variety of exposures and outcomes including physical and mental health indicators, socioeconomic health determinants, involvement with child protective services and academic standardised testing results. The information available from each database is summarised in table 1.

Key exposure measures for the Joining the Dots Study include maternal drug use type, maternal social demographics or social determinants of health, and maternal physical and mental health comorbidities. Key outcome measures include death, academic standardised testing results, hospitalisation and maternal survival. The definition of follow-up for this cohort differs between databases. Cognitive function is routinely followed up through results of standardised academic testing in year 3 (ages 8–9), year 5 (ages 10–11), year 7 (ages 12–13) and year 9 (ages 14–15) of schooling, while other follow-up occurs episodically at specific events, including hospitalisation, emergency department presentations, mental health ambulatory care, entry into out-of-home care and death.

Data analysis plan

Multiple projects will be conducted as part of Joining the Dots, including but not limited to: impacts of out-of-home care on mortality for children with PDE, impacts of out-of-home care on school outcomes for children with PDE, analysis of hospitalisation and associated costs for children with PDE, and impacts of remoteness and socioeconomic disadvantage on outcomes for children with PDE. Detailed data analysis plans will be created for each project, based on the project’s specific objectives and hypotheses. Thus, this descriptive paper presents a general overview of common approaches to data analysis across Joining the Dots. Data analysis will be conducted using established statistical programs, for example, Stata V.18.0. Complete case analysis will be used, and data will not be imputed. Directed acyclic graphs (DAGs) will define the hypothesised causal pathways of PDE on short-term and long-term outcomes. To assess mortality and other time-to-event outcomes, Poisson regression and survival analysis will be used. To investigate the extent of which a mediator (eg, an intervention such as out-of-home care) influences or accounts for the association between PDE and a specified outcome (eg, mortality, school outcomes, rehospitalisation), causal mediation analyses and/or effect modification analyses will be conducted. The economic burden of hospital costs associated with PDE will be calculated based on the Australian Refined Diagnosis Related Group cost, which captures the mean direct and overhead costs associated with each episode of care, and generalised linear models will be used to estimate the differences in crude and adjusted mean hospital costs for children with PDE versus children without PDE.

Confounding by indication will be accounted for in each analysis through adjustment of variables related to possible confounders identified in DAGs for that particular study. Confounders accounted for across Joining the Dots Studies include (but are not limited to): calendar year of birth; Socioeconomic Index for Areas Index of Relative Socioeconomic Disadvantage (a measure of social disadvantage in a small contained geographical area); maternal age <20 years; serious mental illness in the mother; Australian First Nations heritage; PDE where PDE is not the exposure of interest; and out-of-home care (as a mediator or exposure modifier).

Strengths and limitations

The Joining the Dots cohort study has several strengths. The large-scale study will provide quantitative information on long-term outcomes for a vulnerable population spanning over a considerable period in a cost-effective and time-efficient manner. As this is a birth cohort, data exist for all children born between 2001 and 2020. The use of routinely collected information from administrative databases mitigates common barriers to participation in randomised control trials, such as time constraints and language and cultural restrictions, reducing selection and attrition bias.

Some key limitations exist in the Joining the Dots Study. The quality of the data and diversity of variables used are dependent on the compliance of the administrator or clinician entering the data—a common limitation for observational studies.24 Babies born later in the study period have reduced follow-up as this dataset has consecutive years of follow-up. Some patient information is more detailed if they have had more contact with the healthcare system, and vice versa. Importantly, vulnerable populations may be under-represented in some datasets due to barriers to accessing healthcare resources and reduced school attendance for standardised NAPLAN testing. Finally, probabilistic methods used in record linkage carry a risk of false positive linkages. For our study, a low (0.5%) false positive linkage rate is noted by CHeReL.

Conclusion

To date, the Joining the Dots cohort study is the largest and longest follow-up of children with PDE—a vulnerable cohort requiring extensive follow-up in order to tailor interventions. Record linkage has facilitated large-scale follow-up (up to 20 years) for this population in a time and cost-effective manner. Results from this study will be used to inform policies and best practice models for this group of children and their families, to facilitate collaboration with other researchers using similar databases and to drive research into developing cost-effective and pragmatic interventions to support children, families and care providers.

Data availability statement

No data are available. Data are only available through ethics approval and affiliation with the University of New South Wales.

Ethics statements

Patient consent for publication

Ethics approval

Ethics approvals were obtained from the New South Wales Population and Health Services Research Ethics Committee (29 June 2020; 2019ETH12716) and the Australian Capital Territory Health Human Research Ethics Committee (11 October 2021; 2021-1231, 2021-1232, 2021-1233); and the Aboriginal Health and Medical Research Council (5 July 2022; 1824/21), and all Australian educational sectors: Board of Studies (government schools), Australian Independent Schools and Catholic Education Commission (D2014/120797). Data were released to the researchers in September 2022 and stored in a password-protected computer hosted by the University of New South Wales. Results will be presented in peer-reviewed academic journals and at international conferences. Collaborative efforts from similar datasets in other countries are welcome.

References

Footnotes

  • X @Kate_Lawler1, @Drtruthmd, @expchuck

  • Contributors KL cleaned the data with MDr, co-developed the methodology, completed statistical analysis and wrote the manuscript. MDr merged the databases, cleaned the data, co-developed the methodology, provided statistical supervision to Joining the Dots team members, and revised and approved the final manuscript to be submitted. AP co-developed the methodology, provided statistical expertise and assisted with statistical interpretation, contributed intellectual context, and revised and approved the final manuscript to be submitted. EL co-developed the methodology, contributed intellectual context, and revised and approved the final manuscript to be submitted. HU co-developed the methodology, completed statistical analysis, and revised and approved the final manuscript to be submitted. BB contributed intellectual context and revised and approved the final manuscript to be submitted. LB contributed intellectual context and revised and approved the final manuscript to be submitted. MDi contributed intellectual context and revised and approved the final manuscript to be submitted. CG contributed intellectual context and revised and approved the final manuscript to be submitted. LD contributed intellectual context and revised and approved the final manuscript to be submitted. JE contributed intellectual context and revised and approved the final manuscript to be submitted. JLO developed the project idea, obtained ethics approval and linkage data, co-developed the methodology, and revised and approved the final manuscript to be submitted.

  • Funding All phases of the Joining the Dots Study were supported by the Mindgardens Neuroscience Network (grant number: N/A); Australian Red Cross (grant number: N/A); Alpha Maxx Healthcare (grant number: N/A); the Centre for Research Excellence for Integrated Health and Social Care (CREHSCI), funded by the National Health and Medical Research Council (grant number APP1198477); and University of Sydney (grant number: N/A).

  • Competing interests JLO submitted proposals leading to awarding of funding support (Sphere Mindgardens Neuroscience Network, Australian Red Cross, Alpha Maxx Healthcare, National Health and Medical Research Council grant, Medical Research Future Fund and Ross Foundation) that was used to finance data linkage and statistical support, but had no influence on data analysis or publication. JLO has also received support for research and travel from Mallinkrodt, but this was not related to the current manuscript. CG is associated with Alpha Maxx Healthcare, but his link had no bearing on data analysis or publication. There is no relevant financial disclosure to be made for any of the authors in this manuscript.

  • Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.