Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Emotional and Behavioral Trajectories of 2 to 9 Years Old Children Born to Opioid-Dependent Mothers

Emotional and Behavioral Trajectories of 2 to 9 Years Old Children Born to Opioid-Dependent Mothers Maternal opioid use in pregnancy has increased dramatically. Knowledge about children’s longer-term emotional and behav- ioral development after prenatal opioid exposure is scarce. A regional sample of 89 opioid-exposed and 104 non-exposed comparison children were studied prospectively at ages 2, 4.5, and 9 years using the Strengths and Difficulties Questionnaire (SDQ) completed by primary caregivers. Across all childhood assessments, opioid-exposed children obtained significantly higher total difficulties scores than non-exposed comparison children. Growth curve modeling revealed that, relative to their same age peers, opioid-exposed children’s emotional and behavioral difficulties significantly worsened over time. Moreo - ver, fixed effects estimates showed that total difficulties trajectories were poorer for children subject to higher prenatal risk (Est = 1.78, 95% CI = [0.46, 3.09]) who were born to mothers with high levels of social adversity (1.11 [0.51, 1.71]), and were then raised in families characterized by high levels of psychosocial risk (1.94 [0.90, 2.98]) and unstable caregiving (1.91 [0.33, 3.48]). A complex set of pre- and postnatal processes contribute to opioid-exposed children’s emotional and behavioral development. Efforts to mitigate the long-term consequences of opioid use in pregnancy need to consider both children’s and their caregivers’ biopsychosocial risks. Keywords Opioids · Neonatal abstinence/opioid withdrawal syndrome (NAS/NOWS) · Developmental trajectories · Biopsychosocial risk Over the last 20 years, the use and abuse of licit and illicit increased dramatically (Bateman et  al., 2014; Patrick, opioids has skyrocketed globally (Sanlorenzo, Stark, & Davis, Lehmann, & Cooper, 2015; Winkelman, Villapiano, Patrick, 2018), including among women of reproductive- Kozhimannil, Davis, & Patrick, 2018). Methadone age (Bateman et al., 2014; CDC, 2017). As a consequence, maintenance treatment (MMT) provides advantages the number of infants prenatally exposed to opioids has over continued illicit drug use and is the most common management approach for opioid-dependent pregnant women (Jones, O’Grady, Malfi, & Tuten, 2008). However, * Julia Jaekel MMT involves chronic fetal opioid exposure placing the jjaekel@utk.edu infant at risk of neonatal opioid withdrawal syndrome Department of Child & Family Studies, University (NOWS) (Conradt, Crowell, & Lester, 2018). Longitudinal of Tennessee Knoxville, Knoxville, USA studies describing children’s longer-term development after Department of Psychology, University of Tennessee prenatal opioid exposure are scarce (Azuine et al., 2019; Knoxville, Knoxville, USA Kaltenbach et  al., 2018; Lee, Woodward, & Henderson, Department of Psychology, University of Warwick, Coventry, 2019; Levine & Woodward, 2018). UK One of the most critical developmental tasks for children School of Health Sciences, University of Canterbury, is to successfully regulate their behavior and emotions Christchurch, New Zealand (Fergusson, Horwood, & Ridder, 2005; Mischel et  al., School of Psychology, Speech and Hearing, University 2011). There is growing suggestion that prenatal opioid- of Canterbury, Christchurch, New Zealand exposure may impact children’s self-regulatory abilities, Department of Public Health, University of Tennessee and has consistently been associated with poorer behavioral Knoxville, Knoxville, USA Vol.:(0123456789) 1 3 444 Research on Child and Adolescent Psychopathology (2021) 49:443–457 control and attention-deficit hyperactivity disorder (ADHD) considering a wide range of pre- and postnatal factors as symptoms (Azuine et  al., 2019; Lee, Pritchard, Austin, potential confounders. Previous approaches have thus Henderson, & Woodward, 2020; Levine & Woodward, precluded opportunities to separately examine the roles 2017; Nygaard, Slinning, Moe, & Walhovd, 2016; Ornoy, of: a) other adverse prenatal exposures that co-occur with 2003). In addition, compared with their non-exposed fetal opioid exposure; b) maternal social background factors peers, children exposed prenatally to opioids may be at such as education and marital status that are associated with increased risk for emotional difficulties (Nygaard et  al., both opioid use and child outcomes; and, importantly, c) 2016), conduct problems and disruptive behaviors (Azuine the intervening effects of the quality and stability of the et al., 2019; Ornoy, 2003). However, follow-up studies of postnatal rearing environment in shaping child outcomes children exposed prenatally to opioids extending beyond (Azuine et  al., 2019; Crea et  al., 2008). The latter are infancy and early childhood are rare and methodologically especially important since although they may be correlated limited (Fill et al., 2018; Lee et al., 2019; Oei et al., 2017). with the independent variable of interest (prenatal opioid Specifically, existing studies typically report only cross- exposure) they occur after pregnancy and thus are more sectional outcomes at a single age of assessment. Of the few accurately viewed as intervening factors that may mediate longitudinal studies that do exist, these are confounded by or moderate associations between opioid exposure and child a reliance on small samples, high sample attrition (Nygaard outcomes. Furthermore, knowledge about the contributions et al., 2016), lack of tester blinding, no or poor measurement of the postnatal environment to children’s behavioral of other drug exposures (Crea, Barth, Guo, & Brooks, 2008; trajectories is critical in informing intervention efforts aimed Nygaard et al., 2016), or failure to include a comparison at supporting these and future cohorts of opioid-exposed group from the same population studied. This significantly children and families longer term. Specifically, it is critical limits our current understanding and interpretation of to identify variables in children’s environments that can available data. be modified through intervention, in order to facilitate Yet, we know from normative population studies that the resilience for this at-risk group. Most importantly, almost prevalence as well as the nature and severity of children’s no studies have examined the complex interplay of prenatal behavioral and emotional problems vary with age (Koumoula, exposures, social background, and postnatal environmental 2012; Lenze & Wetherell, 2011; Sasser, Kalvin, & Bierman, factors that may heighten or mitigate risk for infants exposed 2016; Spencer, Biederman, & Mick, 2007). For instance, to opioids. We propose that a biopsychosocial framework ADHD and conduct problems tend to have their peak onset and may help incorporate the multimodal and multidisciplinary highest prevalence in childhood, whereas emotional difficulties theoretical aspects mentioned above (Bolton & Gillett, generally peak later (Kessler et al., 2004). Anxiety problems, 2019). Conradt, Crowell, and Lester (2018), for instance, for example, can occur in childhood but have a median age of have proposed an integrated model of how different factors onset in early adolescence, while rates of depressive symp- such as socioeconomic status, maternal psychopathology toms typically rise dramatically during adolescence and peak and nutrition may alter opioid-exposed infants’ development in young adulthood (Kessler et al., 2004). prenatally, and they propose that effects could extend into the At present, it is not known whether prenatally opioid- postnatal period, for instance via mother-infant interactions exposed children follow similar behavioral trajectories as (Conradt et al., 2018). In line with this model, Fig. 1 shows a their non-exposed peers. Yet we can only really understand schematic depiction of the hypothetical pathways of interest the mechanisms underlying potential problems associated considered in this study. The potential importance of each of with prenatal opioid-exposure when accounting for these key processes or risk factors is also reviewed briefly normative developmental changes. Methodologically, below. this can be accomplished by longitudinally administering With regard to prenatal and neonatal exposures and risks, an established, standardized, and developmentally opioid use in pregnancy has been associated with maternal appropriate instrument such as the Strengths and Difficulties perinatal depression and psychiatric illness, and smoking Questionnaire (SDQ) (Goodman, 2001) to assess the during pregnancy, preterm birth, smaller head circumference development of children exposed to opioids as well as a at birth, and a longer hospital stay (Azuine et  al., 2019; representative regional comparison group. Conradt et  al., 2019; Davie-Gray, Moor, Spencer, & A further important issue that is vital to address the Woodward, 2013; Kelty & Preen, 2019; Woodward, longer-term consequences of the growing public health McPherson, & Volpe, 2018). Postnatally, the quality of problem of families affected by opioid use is to understand infants’ caregiving may be particularly important (Hatzis, the developmental mechanisms that might place children Dawe, Harnett, & Barlow, 2017; Konijnenberg, Lund, born to opioid dependent women at increased risk of & Melinder, 2015), but we know remarkably little about later problematic developmental trajectories. This is an whether variations in rearing environments are associated issue that has, to date, been given scant attention beyond with developmental variability after opioid-exposure. 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 445 Fig. 1 Schematic Model of the Hypothetical Pathways of Interest Children who were prenatally exposed to opioids have The aim of this study is to characterize the behavioral been shown to be at increased risk of exposure to psychosocial and emotional trajectories from age 2 to 9  years of a adversity and poor parental support (Hatzis et  al., 2017; regionally representative sample of prenatally opioid- Konijnenberg, Sarfi, & Melinder, 2016; Sarfi, Smith, Waal, exposed children relative to their non-exposed peers. & Sundet, 2011; Siqveland, Haabrekke, Wentzel-Larsen, & Utilizing a multi-level approach, we address key Moe, 2014). However, previous studies have not disentangled questions concerning predictive mechanisms of prenatal, the potential effects of prenatal and postnatal factors, but instead social background, and postnatal environmental risk. treated them collectively as confounders. What’s more, it has We formed three hypotheses: Compared with their rarely been investigated how social background risk indicators non-exposed peers, (1) children who were prenatally may be associated with combined prenatal and postnatal exposed to opioids will exhibit increased emotional and risks when studying children’s long-term development. behavioral difficulties at ages 2, 4.5, and 9 years, (2) their During pregnancy, there are complex mechanisms at play trajectories will remain stable or worsen with age, and between maternal factors such as pregnancy nutrition, (3) prenatal and social risks as well as postnatal rearing social disadvantage, and psychiatric risk that are correlated conditions will explain individual variations in children’s with pregnancy opioid use (Woodward et  al., 2018), but behavioral trajectories. their effects on long-term child development have not been investigated among opioid-exposed mother–child dyads. In addition, prenatal (i.e., biological) compared with postnatal Methods environmental factors may affect long-term trajectories via different pathways (Bendersky & Lewis, 1994; Jusiene, Participants Breidokiene, & Pakalniskiene, 2015), and their combined effects may potentially be additive, interactive, or both (Evans, Participants were assessed as part of a prospective longitudinal Li, & Whipple, 2013; Li, 2003; Sameroff & MacKenzie, 2003). study of two groups of children born between 2003 and 2008 at Sophisticated longitudinal study designs are timely and essential Christchurch Women’s Hospital, New Zealand (Davie-Gray et al., to assess the complex interplay of these biopsychosocial risks 2013; Levine & Woodward, 2017). Mothers were recruited during on opioid-exposed children’s long-term development. their third trimester or at birth. Exclusion criteria included very 1 3 446 Research on Child and Adolescent Psychopathology (2021) 49:443–457 preterm birth (≤ 32 weeks), congenital abnormality, HIV, fetal and perinatal medical treatment has been published alcohol syndrome, and non-English speaking. previously (Davie-Gray et al., 2013; Lee et al., 2020, 2019). The index group consisted of infants born to opioid- The comparison group consisted of randomly identified dependent mothers receiving MMT. Over the recruitment non-opioid-exposed infants born at the same hospital period, 120 mothers were eligible for inclusion in the study. over the same recruitment period. A total of 169 pregnant Of these, 99 mothers and 100 infants (including one set women were approached, with 108 mother and 110 infants of twins) were successfully recruited (84% of all eligible (including two sets of twins) successfully recruited. Regional opioid-exposed infants). Retention to age 9 years was 85%. census data showed that the socioeconomic profile of this Figure 1 provides an overview of recruitment and retention sample was representative of the families living in the region over the study period. Demographic characteristics are at the time (Statistics New Zealand, 2006). Retention to age presented in Table 1. A detailed description of recruitment 9 years was 90% (see Fig. 2 and Table 1). Table 1 Descriptive Participant Characteristics, Pregnancy to 18 Months Non-exposed Opioid-exposed Mean / % difference p Maternal Charac- teristics Age (years) 31.64 (5.44) 29.77 (5.26) -1.87 0.015 Pregnancy methadone - 62.63 (35.14) 62.63 < 0.001 dose (mg) Pregnancy depression 6.83 (4.83) 14.21 (6.13) 7.38 < 0.001 (EDPS) Pregnancy psychiatric 15.89 56.52 36.35 < 0.001 illness (%) Pregnancy nutrition 90.61 (25.31) 56.03 (20.81) -34.57 < 0.001 score Pregnancy smoking 1.39 (3.88) 13.10 (8.90) 11.71 < 0.001 (cigarettes/day) % minority ethnicity 17.76 25.00 7.24 0.212 % younger mother 5.61 3.26 -2.35 0.427 % single parent 9.35 48.91 39.56 < 0.001 % no formal educa- 19.63 80.43 60.80 < 0.001 tion qualification Depression at 5.26 (3.99) 9.79 (6.96) 4.53 < 0.001 18 months (EPDS) Infant Character- istics % female 52.34 42.39 -9.95 0.161 Gestational age 39.27 (1.70) 38.82 (1.69) -0.45 0.063 (weeks) Head circumference 0.23 (0.86) -0.24 (0.88) -0.47 < 0.001 at birth (z-score) Total days of mor- - 63.66 (41.02) 63.66 < 0.001 phine Total days in hospital 3.44 (2.29) 16.07 (12.14) 12.62 < 0.001 Primary caregiver 4.30 43.00 38.07 < 0.001 change (%) Composite Scores Prenatal risk (z-score) 0.34 (0.51) 2.18 (1.76) 1.84 < 0.001 Social risk (sum 0.57 (0.93) 1.37 (0.89) 0.79 < 0.001 score) Postnatal environ- -0.26 (0.48) 0.32 (0.63) 0.58 < 0.001 ment risk (z-score) Data are reported in Mean (SD), if not noted otherwise 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 447 Fig. 2 Study Flow Chart 1 3 448 Research on Child and Adolescent Psychopathology (2021) 49:443–457 α = 0.79) (Cox, Holden, & Sagovsky, 1987). Pregnancy Procedure psychiatric illness was included as a binary-coded variable (0 = no, 1 = yes). All variables were equally weighted and Participants were assessed at five time points: late pregnancy/ birth, 18 months, 2, 4.5, and 9 years. The Upper South B averaged, a higher total score indicated higher perinatal risk. Maternal Social Risk. A composite score was created Regional Ethics Committee, Canterbury, New Zealand (Ref: URB/07/10/042) approved the study protocol and the study as a summative index of five dichotomous maternal social risk indicators (i.e., single parent, minority ethnicity, no high was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later school degree, low socioeconomic status, mother < 21 years old) (Lee et al., 2020). amendments or comparable ethical standards. Written informed consent was obtained from biological mothers or Postnatal Rearing Environment Risk. A continuously scored index of postnatal caregiving quality was created by primary caregivers at each assessment point, with children also providing verbal assent at age 9 years. At each follow-up, averaging the following z-standardized variables: the Home Observation for Measurement of the Environment (HOME) participants received a $20 gift card as compensation for their involvement in the study. total score (α = 0.74; reverse coded) (Caldwell & Bradley, 1984; Totsika & Sylva, 2004), parenting stress measured using a 16-item scale assessing caregiver exposure to, for example, financial, housing, social support and relationship Measures stress (α = 0.68), maternal depression at 18 months (EPDS, α = 0.81) (Cox, Chapman, Murray, & Jones, 1996; Cox et al., Strengths and Difficulties Questionnaire (SDQ). Primary caregivers completed the SDQ at 2, 4.5 and 9 years during 1987) and 4.5  years (Composite International Diagnostic Interview, α = 0.89) (WHO, 1993), corporal punishment at structured parent interviews. The SDQ contains 25-items that measure 5 subscales: emotional symptoms, hyperactivity- 18 months (α = 0.72) (Straus, Hamby, Boney-McCoy, & Sug- arman, 1996), and maternal total illicit drug use between birth inattention, conduct problems, peer problems, and prosocial behavior (Goodman, 1997). Each subscale is scored on a and 4.5 years (see Table S1). Illicit drug use was continuously assessed at child age 18 months and 4.5 years based on ques- 3-point Likert-type scale (0 = not true, 1 = somewhat true, 2 = certainly true). A total difficulties score is calculated by tions from the Composite International Diagnostic Interview (WHO, 1993) about the use/abuse of a range of drugs includ- summing the scores of the four difficulties scales (Cronbach’s α = 0.62, 0.70, and 0.75 for the total scores respectively at each ing heroine, prescription opioids, cannabis, barbiturates, stim- ulants, hallucinogens inhalants and other illicit prescription age in this study). The SDQ is a widely used screening tool with good test–retest reliability, e.g., r = 0.72 for parent-rated drugs (Davie-Gray et al., 2013). The variable was binary coded (0 = no, 1 = yes) for inclusion in the risk score due to limited total difficulties scores over 4–6 months (Goodman, 2001). Maternal Pregnancy/Birth Characteristics. Detailed variance and non-normal distribution. All variables were equally weighted and averaged, a information about mothers’ mental health, nutrition, and social risks was collected via comprehensive structured higher total score indicated higher postnatal rearing envi- rd ronment risk. interviews administered in the late 3 trimester or at birth (Lee et al., 2020, 2019), see Table S1 for details. Primary Caregiver Changes. Any transitions or disruptions in children’s primary caregivers were documented Infant Characteristics. Detailed information about perina- tal characteristics (e.g., gestational age, head circumference at and coded as a binary variable (0 = no, 1 = yes). Preliminary analyses showed that primary caregiver changes indicated an birth) and infant medical treatment was extracted from hospital records. independent and specific risk that was not correlated with the more general and normally distributed index score of postnatal We created three risk index scores in order to include a diverse and large set of biopsychosocial risk variables in our rearing environment risk (Spearman’s rho = 0.12, p = 0.09). analyses, to control for potential multicollinearity between variables, and to increase external validity and reliability. Statistical Analyses Other Prenatal and Perinatal Risk Exposures. A continuously scored index of pre- and neonatal biological Descriptive analyses were conducted in SPSS 24 to examine group differences between opioid-exposed and risks was created by averaging the following z-standardized variables, thereby weighting their independent contributions non-exposed comparison children with regard to bio- logical and environmental factors as well as emotional equally: infant gestational age, head circumference at birth, 1-min APGAR score, and pregnancy nutrition (all reverse and behavioral difficulties. For the main analyses, a low number of data missing at random was imputed using coded), total days in hospital (as an index of postnatal illness severity), smoking during pregnancy, and maternal depres- the stochastic imputation method implemented in “mice 3.8.0” package in R (van Buuren et al., 2020). Imputation sion at birth (Edinburgh Postnatal Depression Scale (EPDS), 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 449 was confined to participants with at least two of the three subject to higher rates of psychiatric illness. Opioid- SDQ assessments available, resulting in a longitudinal exposed infants tended to have been born, on average, sample of 89 opioid-exposed and 104 non-exposed chil- to younger mothers than non-exposed children, although dren. Correlations between variables included in the rates of early motherhood (< 21  years) did not differ, prenatal biological and social risk, and postnatal rearing and they more often had single mothers without formal environment scores are presented in Table  2. Trajecto- educational qualifications (Table 1). Rates of minority ries of emotional and behavioral development from 2 to ethnicity were the same across groups. Accordingly, 9 years were assessed using linear mixed effects model the prenatal and social risk index scores were higher analysis in Stata 16. Trajectories of total difficulty scores in the opioid-exposed group. In addition, the postnatal over time were modeled as a function of age at assess- rearing environment risk index score was also higher, and ment (years) and group status (opioid-exposed vs. non- they had more changes in primary caregivers (i.e., care exposed), including both fixed and random effects, fol- disruptions). For instance, at ages 2, 4.5, and 9  years, lowing an established approach (Mangin, Horwood, & respectively, 80%, 69%, and 54% of the opioid-exposed Woodward, 2017). As part of a stepwise process, a quad- children were in the care of their biological mothers. For ratic term for age (age*age) was included in Model 2. In the non-exposed group, 100% of children were in the care Model 3, an interaction term (age*group) was added to of their biological mothers at ages 2 and 4.5 years, with test whether trajectories differed between groups. This 99% in their mothers’ care at age 9 years. model was then extended to assess the effects of child sex, Emotional and Behavioral Difficulties . Confirming prenatal biological and social risk, the postnatal rearing Hypothesis 1, prenatally opioid-exposed children had sig- environment, and primary caregiver changes on emotional nificantly higher total difficulties scores than their non- and behavioral development, including potential additive exposed peers across all childhood assessments, with the and interactive effects. Log-likelihood goodness of fit mean difference [95% confidence interval] increasing from tests were used to compare each growth curve model to 2.06 [0.61, 3.52] at age 2 years to 7.13 [5.30, 8.96] at age the previous one. 9 years (Table 3, Fig. 2). This was further confirmed by visual inspection of the data in Fig. 2, which suggests that Results opioid-exposed children’s difficulties appear to increase from ages 2 to 9  years, whereas the scores of children Descriptive Sample Characteristics. Indicators of in the non-exposed comparison group tended to remain prenatal and social background risk tended to be higher relatively stable. among children in the opioid-exposed group. On average, As part of a sensitivity analysis, we examined the opioid-exposed infants had lower head circumference domain-specific nature of opioid-exposed children’s z-scores at birth and spent more days in hospital than non- emotional and behavioral difficulties. Table 3 shows the exposed infants. There were however no group differences mean scores of the two study groups on each of the SDQ in infant sex, gestational age, and 1-min APGAR scores. subscales (emotional, conduct, inattention/hyperactivity, Opioid-using mothers had higher prenatal depression and and peer problems) as well as the total score by age at lower nutrition scores, reported higher average numbers assessment (2, 4.5, and 9 years), also see Fig. 3. Across of daily cigarettes smoked during pregnancy, and were all four domains, compared with their non-exposed Fig. 3 SDQ Total Difficulties Scores of Opioid-Exposed Versus Non-Exposed Children at 2, 4.5, and 9 Years 1 3 450 Research on Child and Adolescent Psychopathology (2021) 49:443–457 1 3 Table 2 Correlations Between Variables Included in the Neonatal Biological, Social, and Rearing Environment Risk Index Scores 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 1. infant gestational age - -0.09 -0.07 -0.28 -0.12 -0.22 -0.10 -0.05 0.01 -0.11 -0.08 0.05 0.19 -0.10 -0.08 -0.04 -0.05 -0.08 -0.41 -0.08 -0.17 2. head circumference - 0.00 -0.19 -0.17 -0.04 0.08 -0.17 0.03 -0.17 -0.20 -0.06 0.08 -0.02 -0.06 -0.17 -0.01 -0.16 -0.29 -0.12 -0.14 3. pregnancy nutrition - 0.05 -0.08 0.04 -0.19 -0.04 0.03 -0.18 -0.12 0.03 0.02 -0.05 -0.11 -0.03 0.07 -0.06 -0.13 -0.11 -0.06 4. total days in hospital - 0.44 0.05 -0.04 0.32 -0.03 0.44 0.51 -0.10 -0.33 0.24 0.29 0.32 -0.06 0.32 0.89 0.36 0.38 5. postnatal depression - 0.17 0.02 0.40 0.06 0.42 0.40 -0.01 -0.29 0.23 0.42 0.34 0.05 0.28 0.55 0.37 0.38 6. pregnancy psychiatr. illness - -0.01 0.17 -0.09 0.07 0.00 0.04 -0.11 0.21 0.24 0.06 0.04 0.12 0.21 0.01 0.21 7. pregnancy smoking - -0.02 0.19 0.19 0.06 0.14 -0.03 -0.04 -0.08 0.01 0.04 -0.20 0.13 0.11 -0.04 8. single parent - 0.08 0.47 0.47 0.22 -0.26 0.31 0.30 0.26 0.04 0.31 0.36 0.52 0.38 9. minority ethnicity - 0.12 0.26 0.20 -0.10 0.07 0.02 0.12 0.03 0.01 -0.04 0.47 0.12 10. no high school degree - 0.62 0.19 -0.44 0.28 0.33 0.33 -0.05 0.28 0.50 0.58 0.44 11. low SES - 0.15 -0.40 0.31 0.26 0.37 -0.01 0.37 0.55 0.59 0.47 12. mother < 21 years - -0.10 0.04 -0.09 -0.01 0.10 0.02 -0.07 0.25 0.06 13. HOME total score - -0.32 -0.27 -0.28 -0.10 -0.31 -0.36 -0.45 -0.61 14. parenting stress - 0.51 0.27 0.21 0.32 0.27 0.35 0.71 15. mat. depression, 18 m - 0.43 0.15 0.29 0.33 0.32 0.71 16. mat. depression, 4.5y - 0.05 0.22 0.36 0.28 0.58 17. corporal punishment - 0.15 0.00 0.23 0.47 18. total illicit drug use - 0.32 0.33 0.56 19. Prenatal Risk Exposure - 0.38 0.43 20. Social Risk Score - 0.57 21. Postnatal Rearing - Environment Risk Research on Child and Adolescent Psychopathology (2021) 49:443–457 451 Table 3 SDQ Subscale and Non-Exposed Opioid-Exposed Total Difficulties Scores of Opioid-Exposed Versus Non- SDQ Score n Mean (SD) n Mean (SD) Mean Difference (95% p Exposed Children at 2, 4.5, and CI) 9 Years Emotional Problems 2 y 107 1.36 (1.35) 92 1.45 (1.53) 0.09 (-0.31, 0.49) 0.666 4.5 y 103 1.27 (1.42) 87 1.90 (1.88) 0.64 (0.16, 1.11) 0.009 9 y 99 1.74 (1.78) 81 2.98 (2.39) 1.24 (0.61, 1.87) < 0.001 Conduct Problems 2 y 107 2.28 (1.91) 92 2.71 (2.09) 0.43 (-0.13, 0.99) 0.131 4.5 y 103 0.92 (1.11) 87 2.47 (1.90) 1.55 (1.09, 2.01) < 0.001 9 y 99 1.07 (1.54) 81 2.96 (2.42) 1.89 (1.28, 2.51) < 0.001 Hyperactivity/Inat- 2 y 107 3.48 (2.42) 92 4.50 (2.47) 1.02 (0.33, 1.70) 0.004 tention 4.5 y 103 2.44 (2.11) 87 4.29 (2.38) 1.85 (1.21, 2.49) < 0.001 9 y 99 2.65 (2.26) 81 5.17 (2.89) 2.53 (1.75, 3.30) < 0.001 Peer Problems 2 y 107 1.25 (1.50) 92 1.77 (1.79) 0.53 (0.07, 0.99) 0.025 4.5 y 103 0.93 (1.21) 87 1.59 (1.56) 0.65 (0.25, 1.06) 0.002 9 y 99 1.23 (1.47) 81 2.63 (2.06) 1.40 (0.86, 1.94) < 0.001 Total Difficulties 2 y 107 8.37 (4.80) 92 10.43 (5.60) 2.06 (0.61, 3.52) 0.006 4.5 y 103 5.56 (4.12) 87 10.25 (5.51) 4.69 (3.27, 6.07) < 0.001 9 y 99 6.66 (5.00) 81 13.79 (7.00) 7.13 (5.30, 8.96) < 0.001 same-age peers, there was an increase in opioid-exposed children in the opioid exposed group. However, by age children’s problems from 2 to 9  years. Specifically, 4.5 years, opioid-exposed children were found to have opioid-exposed children had higher inattention/ higher problem scores across all four SDQ domains, and hyperactivity (mean difference 1.02 [0.33, 1.70]) and these group differences further increased at age 9 years. peer problems (0.53 [0.07, 0.99]) than their peers at Trajectories of Difficulties. Linear mixed effects 2  years, but significant between group differences in growth curve models were run to examine the behavio- emotional and conduct problems were not detected ral trajectories of both study groups based on their SDQ despite the tendency for higher levels to be reported for total difficulties scores from 2 to 9 years of age, using Table 4 Multilevel Growth Curve Model Results of SDQ Total Difficulties Score Trajectories from 2 to 9 Years of Age (n = 193) Total Difficulties Model 1 Model 2 Model 3 Model 4 Model 5 Est (SE) Est (SE) Est (SE) Est (SE) Est (SE) Fixed effects Age 0.39 (0.25) -6.57 (1.31)*** -10.08 (1.48)*** -10.08 (1.48)*** -10.08 (1.48)*** Age - 1.74 (0.32)*** 1.74 (0.32) *** 1.74 (0.32)*** 1.74 (0.32)*** Group (opioid-exposed) 4.23 (0.58)*** 4.23 (0.58)*** -0.13 (1.05) -3.91 (1.15)** -4.61 (1.17)*** Interaction (group*age) - - 2.40 (0.48) *** 2.40 (0.48)*** 2.40 (0.48)*** Sex - - - -0.52 (0.52) -0.45 (0.51) Prenatal Risk - - - 2.03 (0.67)** 1.78 (0.67)** Social Risk - - - 1.18 (0.31)*** 1.11 (0.31)*** Postnatal Environmt. Risk - - - 1.56 (0.52)** 1.94 (0.53)*** Primary Caregiver Change - - - - 1.91 (0.80)* constant 1.94 (1.01) 7.74 (1.47)*** 14.12 (1.94)*** 20.41 (2.20)*** 19.09 (2.25)*** Random effects SD (Age) 2.21 (0.34)* 2.42 (0.30)* 2.11 (0.31)* 2.11 (0.31)* 2.11 (0.31)* SD (constant) 4.68 (0.75)* 5.16 (0.66)* 4.67 (0.67)* 3.99 (0.72)* 3.88 (0.73)* correlation (Age,constant) -0.72 (0.08)* -0.76 (0.06)* -0.70 (0.08)* -0.74 (0.08)* -0.74 (0.08)* Log-likelihood -1,769.61 -1,755.90 -1,744.16 -1,720.39 -1,717.62 for fixed effects *p < 0.05; **p < 0.01; ***p < 0.001; for random effects * marks a 95% confidence interval not including 0 1 3 452 Research on Child and Adolescent Psychopathology (2021) 49:443–457 the stepwise approach outlined above. Table  4 shows Finally, we examined possible interactive effects detailed results for five models that were subsequently between prenatal biological and social risk, as well as fitted to the data. In Model 1, trajectories of total difficul- with postnatal rearing environmental inf luences on chil- ties over time were modeled as a linear function of age at dren’s overall difficulties trajectories. None of these inter- assessment (years) and group status (opioid-exposed vs. action effects reached statistical significance. non-exposed). On average, opioid-exposed children had Thus Model 5, showing independent (i.e., additive) fixed total difficulties scores with coefficient estimates (CE) main effects, is considered the final model. Figure  4 shows pre- of 4.23 (95% CIs [3.09, 5.37]) points higher than their dicted total difficulties trajectories by group based on Model non-exposed peers at any given age (p < 0.001), while 5. Overall, opioid-exposed children’s scores were character- the fixed linear main effect of age at assessment was not ized by higher total difficulties and greater inter- and intra- significant. Child-specific random variability according individual variability from baseline (2 years) to 9 years of age. to age was significant with an estimated random-slope standard deviation of CE = 2.21 [1.63, 2.98]. The log- Discussion likelihood ratio chi-square test supported the use of the mixed effects Model 1 versus a simple linear model This is the first prospective analysis of the emotional and (χ (3) = 71.89, p < 0.001). behavioral development of an unselected regional sample To test for potential nonlinear effects of age on trajectories of opioid-exposed children from birth to 9 years, relative to over time, a quadratic term (age*age) was included in Model their same age typically developing peers. Using a hypoth- 2. As a result, the fixed linear (CE = -6.57 [-9.13, -4.01]) and esis-driven, multi-level analysis approach, we confirmed all quadratic effects of age (CE = 1.74 [1.11, 2.37]) were both three hypotheses we had formed a-priori. Specifically, com- significant with improved fit over Model 1, indicated by a pared with their non-exposed peers, (1) children who were reduced log-likelihood value (Table 3), and a significant log- prenatally exposed to opioids exhibited significantly higher likelihood ratio chi-square test (assumption Model 1 nested emotional and behavioral difficulties at 2, 4.5, and 9 years in Model 2: χ (1) = 27.42, p < 0.001). of age. Growth curve modeling further revealed that, (2) In Model 3, an interaction term (group*age) was introduced relative to their same age peers, opioid-exposed children’s to test whether behavioral trajectories differed between groups. difficulties trajectories worsened over time. Finally, results Again, model fit improved significantly (assumption Model 2 showed that (3) the index scores of prenatal risk, social risk, nested in Model 3: χ (1) = 23.48, p < 0.001) with the addition and postnatal rearing environment risk we had formed, as of this new parameter (CE = 2.40 [1.46, 3.34]). The significant well as changes in primary caregivers, independently and group x age interaction in Model 3 implies that opioid-exposed additively explained individual variations in children’s children’s total difficulties trajectories differed from non-exposed behavioral trajectories (see Fig. 4). children, with total difficulties scores worsening with age, and In addition to SDQ total difficulties scores, we also exam- thus confirming Hypothesis 2. ined the domain-specific nature of opioid-exposed children’s To test Hypothesis 3, Model 4 then included fixed emotional and behavioral difficulties in comparison to their effects of child sex, prenatal biological and social risk, same-age peers. An assessment of the existing literature as well as the postnatal rearing environment composite shows that prenatal opioid-exposure is consistently linked scores. While child sex did not predict total difficulties tra- with externalizing behavior problems, and in particular jectories, the previous effects of age and group remained behavioral dysregulation and attentional difficulties (Azuine significant and all three risk scores had additionally et  al., 2019; Levine & Woodward, 2017; Nygaard et al., significant effects. Thus, Hypothesis 3 was confirmed: 2016; Ornoy, 2003; Slinning, 2004; Sundelin Wahlsten & children with higher prenatal biological risk (CE = 2.03 Sarman, 2013). This is confirmed by our current results, [0.70, 3.35]), higher social background risk (CE = 1.18 with the rate of hyperactivity/inattention showing substantial [0.58, 1.79]), and higher postnatal rearing environment mean differences between groups among the SDQ subscales risk (CE = 1.56 [0.55, 2.57]) had higher total difficulties at each assessment. Similar increases in conduct problems scores. Again, model fit improved significantly (assump- with child age particularly from early childhood, were also tion Model 3 nested in Model 4: χ2(4) = 47.55, p < 0.001). observed amongst opioid-exposed children. Collectively, In Model 5, a fixed ee ff ct of having had any primary caregiver these findings suggest that these children are likely to be at changes on total difficulties trajectories was additionally added very high risk not just of more severe externalizing behavior (CE = 1.91 [0.33, 3.48]), and this further improved overall problems in childhood, but also a poorer prognosis longer model fit (assumption Model 4 nested in Model 5: χ 2(1) = 5.54, term. Finally, rates of peer (i.e., social) problems were also p < 0.019). Importantly, the group and group by age differences consistently elevated, and emotional problems of opioid- remained significant, even after accounting for related risk exposed children seemed to exacerbate over the course of factors, which also remained significant. childhood. Given the high rates of parental psychopathology 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 453 Fig. 4 Predicted (Model 5) Total Difficulties Scores Trajectories of Opioid-Exposed Versus Non-Exposed Children from 2 to 9 Years and family instability in this high-risk group, a question for with their caregiving quality, thereby directly and indirectly future study may be to examine whether early onset symp- affecting infants’ stress regulation and behavior (Conradt toms reflect attachment or social relationship related issues et  al., 2018). The second potential mechanism could be in early childhood. Further, assessments using standardized that prenatal opioid exposure alters the development of diagnostic measures of mental health disorders will also be dopaminergic reward-related circuits, which may in turn important. lead to dysregulated, hyperactive behavior (Sithisarn et al., According to the Developmental Origins of Health and 2017). What’s more, DOHaD studies have documented a Disease (DOHaD) model, unfavorable conditions in utero strong relationship between maternal depression and stress (e.g., opioid exposure, chronic stress) affect behavioral during pregnancy and children’s subsequent behavior functioning and health later in life via prenatal programing regulation trajectories, even after controlling for confounders (Barker, 2007). This includes teratogenic effects on the such as postnatal maternal depression (Henrichs & Van den developing central nervous system (Woodward et  al., Bergh, 2015). They also suggest that this relationship may 2018), potentially increasing mental health vulnerability. be continuously modulated by environmental experiences DOHaD also proposes that long-term risk is initially (Henrichs & Van den Bergh, 2015). Our novel findings of induced through adaptive responses (e.g., changes in the independent, additive contributions of prenatal risk, metabolism or tissue sensitivity to hormones) that the social risk, postnatal caregiving quality, and caregiver fetal organism makes to maternal cues about her health changes to predicting children’s emotional and behavioral or physical state, leading to altered organ development. problem trajectories add important information about Two mechanisms related to prenatal programming may the potential mechanisms placing infants born to opioid explain the behavior regulation difficulties seen in opioid- dependent mothers at increased psychopathological risk. For exposed children: The first suggests that opioids may alter instance, the non-exposed children’s trajectories displayed the hypothalamic–pituitary–adrenal (HPA) axis (Sithisarn in Fig. 4 show an expected small normative improvement et al., 2008, 2017; Slamberova, Riley, & Vathy, 2005; Taylor, in self-regulatory abilities from 2 to 4.5  years, but this Soong, Wu, Yee, & Szeto, 1997), which has been linked, pattern is less distinct among the opioid-exposed children. for instance, with ADHD (Ma, Chen, Chen, Liu, & Wang, This suggests that some risk factors may be at play that 2011). In addition, prenatal and postnatal stressors may dynamically alter the normative course of development have additive or interactive programming influences on fetal over time. Importantly, our final growth curve model HPA axis functioning, on top of prenatal opioid exposure showed that the additive effects of prenatal and social risks (Lester & Padbury, 2009). For instance, opioid-using remained significant even when effects of the postnatal mothers may have a higher risk for depressive symptoms rearing environment and primary caregiver changes were during and after pregnancy, and these could be associated added to the model. Thus, the adverse developmental 1 3 454 Research on Child and Adolescent Psychopathology (2021) 49:443–457 effects documented here cannot exclusively be attributed more process-oriented, longitudinal, and observation-based to the different postnatal environments that children grew assessment of reciprocal caregiver-child interactions would up in, but prenatal and social mechanisms related to help uncover the complex effects of social relationships on prenatal programming remain at play, even at later ages, trajectories in future studies. Relatedly, many predictors over and above the effects of in-utero opioid exposure. and the dependent variables were based on caregivers’ These results offer critically important information for reports, which can be biased (De Los Reyes & Kazdin, designing prevention and intervention services for current 2005; Van Roy, Groholt, Heyerdahl, & Clench-Aas, 2010). and future generations of opioid-using mothers and their What’s more, we were not able to include information offspring. Development is a complex multidirectional about other critically important domains that could explain process that can be described as a constant feedback loop mechanistic effects, such as (epi)genetics or stress response between biological and environmental factors, starting differences. The single stochastic imputation performed long before birth. Our findings suggest that intervention to maximize the utility of available data does not fully approaches that exclusively aim at modifying variables account for the variability structure of imputed and in children’s postnatal environments may not be able to missing values, and hence may have introduced downward effectively facilitate resilience for this at-risk group. Instead, bias in the estimates of standard errors of model estimates integrated biopsychosocial approaches that start during or (Carlin, Li, Greenwood, & Coffey, 2003; Rubin, 1987). before pregnancy and target opioid-using women’s mental In addition, statistical power to detect interaction effects and physical health may be warranted in order to increase between the different biopsychosocial risk scores was life chances for the next generation. limited. Future studies with larger sample sizes will be Strengths and Limitations. The findings presented here important in helping to replicate our study findings and to are from a prospective regional sample of opioid-exposed further evaluate whether the effects of pre- and postnatal children and an equally-sized group of non-exposed age- factors on child behavior are additive, interactive or both. matched controls. Compared to other longitudinal cohort A more in-depth examination of specific associations studies, our sample is characterized by very high sample between some of the risk exposures we have identified retention over time and age, a marker of excellent study and child outcomes may also be helpful, as this was quality. Alongside our opioid exposed group, measures beyond the scope of this integrated longitudinal analysis were also collected on a randomly identified, regionally of the contributions of a wide range of potential pre- and representative comparison group of non-exposed infants. This postnatal explanatory factors. helped avoid problems associated with the Flynn effect and Conclusion. Children who were exposed to opioids the use of normative cut-off points at very young ages, such in-utero remain at risk for emotional and behavioral as age 2. It also minimizes the likelihood that unintended and difficulties throughout the course of their childhood. In potentially unknown biases are introduced into the analysis. addition to the initial effects of opioids, a complex set of Assessments were rigorous and used internationally accepted, prenatal biological and postnatal environmental factors standardized measures at multiple times over the course of additively and dynamically contribute to substantial infancy and childhood. The rationale was hypotheses-driven variations in children’s individual developmental trajectories. and a sophisticated analysis approach helped uncover the Future studies and evidence-based efforts to mitigate the complex mechanisms underlying developmental outcomes long-term consequences of opioid use in pregnancy need to after in-utero opioid exposure. take children’s diverse biopsychosocial risks into account. However, the study also had some limitations. Our opioid- Supplementary Information The online version contains supplemen- exposed sample only included infants of mothers who were in tary material available at https: //doi.org/10.1007/s10802 -020-00766- w. MMT treatment. Untreated opioid dependence is extremely rare in New Zealand, and these women were prioritized for Acknowledgements The authors are grateful to Marie Goulden and substitution treatment which is free and accompanied by wrap- Carole Spencer for assistance with sample recruitment/retention and data collection. Most importantly, we would like to thank the families around psychiatric and obstetric care. Thus, results from these who kindly gave their time and support to this study. dyads may underestimate potential adverse effects of substance use on the child, and will have limited generalizability to Funding This study was funded by Cure Kids New Zealand (# 9518), infants of mothers who do not seek or receive substitution the New Zealand Lottery Grants Board (AP91256), Wayne Francis treatment when pregnant or when that care is unavailable. Trust, and the Health Research Council of New Zealand (#13/584). One of our aims was to assess different pre- and postnatal biopsychosocial risks and their contributions to Compliance with Ethical Standards opioid-exposed children’s development. As in any study, Conflict of Interest The authors declare that they have no conflicts of our analyses were limited by the variables that had been interest. assessed, and their respective distributions. For instance, a 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 455 Ethical Approval The Upper South B Regional Ethics Committee, women. Journal of Affective Disorders, 39(3), 185–189. https :// Canterbury, New Zealand (Ref: URB/07/10/042) approved the study doi.org/10.1016/0165-0327(96)00008 -0. protocol and the study was performed in accordance with the ethical Cox, J. L., Holden, J. M., & Sagovsky, R. (1987). Detection of postna- standards as laid down in the 1964 Declaration of Helsinki and its later tal nepression: Development of the 10-item Edinburgh Postnatal amendments or comparable ethical standards. Depression Scale. British Journal of Psychiatry, 150(6), 782–786. https ://doi.org/10.1192/bjp.150.6.782. Consent to Participate Written informed consent was obtained from Crea, T. M., Barth, R. P., Guo, S., & Brooks, D. (2008). Behavio- biological mothers or primary caregivers at each assessment point, with ral outcomes for substance-exposed adopted children: fourteen children also providing verbal assent at age 9 years. years postadoption. American Journal of Orthopsychiatry, 78(1), 11–19. https ://doi.org/10.1037/0002-9432.78.1.11. Davie-Gray, A., Moor, S., Spencer, C., & Woodward, L. J. (2013). Open Access This article is licensed under a Creative Commons Attri- Psychosocial characteristics and poly-drug use of pregnant bution 4.0 International License, which permits use, sharing, adapta- women enrolled in methadone maintenance treatment. Neuro- tion, distribution and reproduction in any medium or format, as long toxicology and Teratology, 38, 46–52. https ://doi.org/10.1016/j. as you give appropriate credit to the original author(s) and the source, ntt.2013.04.006. provide a link to the Creative Commons licence, and indicate if changes De Los Reyes, A., & Kazdin, A. E. (2005). Informant discrepan- were made. The images or other third party material in this article are cies in the assessment of childhood psychopathology: A critical included in the article’s Creative Commons licence, unless indicated review, theoretical framework, and recommendations for fur- otherwise in a credit line to the material. If material is not included in ther study. Psychological Bulletin, 131(4), 483–509. https ://doi. the article’s Creative Commons licence and your intended use is not org/10.1037/0033-2909.131.4.483. permitted by statutory regulation or exceeds the permitted use, you will Evans, G. W., Li, D., & Whipple, S. S. (2013). Cumulative risk and need to obtain permission directly from the copyright holder. To view a child development. Psychological Bulletin, 139(6), 1342–1396. copy of this licence, visit http://creativ ecommons .or g/licenses/b y/4.0/. https ://doi.org/10.1037/a0031 808. Fergusson, D. M., Horwood, J., & Ridder, E. M. (2005). Show me the child at seven: the consequences of conduct problems in child- hood for psychosocial functioning in adulthood. Journal of Child Psychology and Psychiatry, 46(8), 837–849. https ://doi.org/10.1 References 111/j.1469-7610.2004.00387 .x. Fill, M. A., Miller, A. M., Wilkinson, R. H., Warren, M. D., Dunn, J. Azuine, R. E., Ji, Y., Chang, H.-Y., Kim, Y., Ji, H., DiBari, J., & Wang, X. R., Schaffner, W., & Jones, T. F. (2018). Educational disabilities (2019). Prenatal risk factors and perinatal and postnatal outcomes among children born with neonatal abstinence syndrome. Pediat- associated with maternal opioid exposure in an urban, low-income, rics, 142(3). https ://doi.org/10.1542/peds.2018-0562 multiethnic US population. JAMA Network Open, 2(6), e196405– Goodman, R. (1997). The Strengths and Difficulties Question- e196405. https ://doi.org/10.1001/jaman etwor kopen .2019.6405. naire: a research note. Journal of Child Psychology Psychia- Barker, D. J. P. (2007). The origins of the developmental origins theory. try, 38(5), 581–586. Retrieved from http://onlinelibr ar y.wiley. Journal of Internal Medicine, 261(5), 412–417. https ://doi.org/ com/s t or e /10.1111/j.1469-7610.1997.tb015 45.x/ass e t / 10.1111/j.1365-2796.2007.01809 .x. j . 1 4 6 9 - 7 6 1 0 . 1 9 9 7 . t b 0 1 5 4 5 . x . p d f ? v = 1 & t = j d 8 0 k v 0 3 & s = Bateman, B. T., Hernandez-Diaz, S., Rathmell, J. P., Seeger, J. D., dcf7e e502e 577b1 3aa85 02a07 3fe05 e1295 dea8c Doherty, M., Fischer, M. A., & Huybrechts, K. F. (2014). Patterns Goodman, R. (2001). Psychometric properties of the Strengths and Dif- of opioid utilization in pregnancy in a large cohort of commercial ficulties Questionnaire (SDQ). Journal of the American Academy insurance beneficiaries in the United States. Anesthesiology, 120(5), of Child and Adolescent Psychiatry, 40, 1337–1345. 1216–1224. https ://doi.org/10.1097/aln.00000 00000 00017 2. Hatzis, D., Dawe, S., Harnett, P., & Barlow, J. (2017). Quality of car- Bendersky, M., & Lewis, M. (1994). Environmental risk, biological egiving in mothers with illicit substance use: A systematic review risk, and developmental outcome. Developmental Psychology, and meta-analysis. Substance Abuse: Research and Treatment, 30(4), 484–494. 11, https ://doi.org/10.1177/11782 21817 69403 8. Bolton, D., & Gillett, G. (2019). Biopsychosocial conditions of health Henrichs, J., & Van den Bergh, B. R. H. (2015). Perinatal developmen- and disease. The Biopsychosocial Model of Health and Disease: tal origins of self-regulation. In G. H. E. Gendolla, M. Tops, & S. New Philosophical and Scientific Developments (pp. 109–145). L. Koole (Eds.), Handbook of Biobehavioral Approaches to Self- Cham: Springer International Publishing. Regulation (pp. 349–370). New York, NY: Springer, New York. Caldwell, B. M., & Bradley, R. H. (1984). Home observation for meas- Jones, H. E., O’Grady, K. E., Malfi, D., & Tuten, M. (2008). Metha- urement of the environment. AR: Retrieved from Little Rock. done maintenance vs. methadone taper during pregnancy: Mater- Carlin, J. B., Li, N., Greenwood, P., & Coffey, C. (2003). Tools for nal and neonatal outcomes. The American Journal on Addictions, Analyzing Multiple Imputed Datasets. The Stata Journal, 3(3), 17(5), 372–386. https ://doi.org/10.1080/10550 49080 22662 76 226–244. https ://doi.org/10.1177/15368 67x03 00300 302. Jusiene, R., Breidokiene, R., & Pakalniskiene, V. (2015). Develop- CDC. (2017). Vital signs: prescription painkiller overdoses: a growing mental trajectories of mother reported regulatory problems from epidemic, especially among women. Retrieved 2/3/2018 toddlerhood to preschool age. Infant Behavior and Development, Conradt, E., Crowell, S. E., & Lester, B. M. (2018). Early life stress and 40, 84–94. https ://doi.org/10.1016/j.infbe h.2015.04.003. environmental influences on the neurodevelopment of children Kaltenbach, K., O’Grady, K. E., Heil, S. H., Salisbury, A. L., Coyle, with prenatal opioid exposure. Neurobiology of Stress, 9, 48–54. M. G., Fischer, G., & Jones, H. E. (2018). Prenatal exposure to https ://doi.org/10.1016/j.ynstr .2018.08.005. methadone or buprenorphine: Early childhood developmental out- Conradt, E., Flannery, T., Aschner, J. L., Annett, R. D., Croen, L. A., comes. Drug and Alcohol Dependence, 185, 40–49. https ://doi. Duarte, C. S., Lester, B. M. (2019). Prenatal Opioid Exposure: org/10.1016/j.druga lcdep .2017.11.030. Neurodevelopmental Consequences and Future Research Priori- Kelty, E., & Preen, D. B. (2019). Risk Factors Associated with the ties. Pediatrics, 144(3). https ://doi.org/10.1542/peds.2019-0128 Occurrence of Neonatal Opioid Withdrawal Syndrome: A Cox, J. L., Chapman, G., Murray, D., & Jones, P. (1996). Validation of Review. CNS Drugs, 33(11), 1113–1120. https://doi.or g/10.1007/ the Edinburgh postnatal depression scale (EPDS) in non-postnatal s4026 3-019-00681 -9. 1 3 456 Research on Child and Adolescent Psychopathology (2021) 49:443–457 Kessler, R. C., Abelson, J., Demler, O., Escobar, J. I., Gibbon, M., Ornoy, A. (2003). The impact of intrauterine exposure versus postnatal Guyer, M. E., & Zheng, H. (2004). Clinical calibration of DSM- environment in neurodevelopmental toxicity: long-term neurobe- IV diagnoses in the World Mental Health (WMH) version of the havioral studies in children at risk for developmental disorders. World Health Organization (WHO) Composite International Toxicology Letters, 140–141, 171–181. Diagnostic Interview (WMHCIDI). Int J Methods Psychiatr Res, Patrick, S. W., Davis, M. M., Lehmann, C. U., & Cooper, W. O. (2015). 13(2), 122–139. Increasing incidence and geographic distribution of neonatal Konijnenberg, C., Lund, I. O., & Melinder, A. (2015). Behavioural abstinence syndrome: United States 2009 to 2012. Journal of outcomes of four-year-old children prenatally exposed to Perinatology, 35(8), 650–655. https://doi.or g/10.1038/jp.2015.36. methadone or buprenorphine: a test of three risk models. Early Rubin, D. B. (1987). Multiple Imputation for Nonresponse in Surveys: Child Development and Care, 185(10), 1641–1657. https ://doi. John Wiley & Sons, Inc. org/10.1080/03004 430.2015.10165 06. Sameroff, A. J., & MacKenzie, M. J. (2003). Research strategies for Konijnenberg, C., Sar, fi M., & Melinder, A. (2016). Mother-child inter - capturing transactional models of development: The limits of the action and cognitive development in children prenatally exposed possible. Development and Psychopathology, 15(3), 613–640. to methadone or buprenorphine. Early Human Development, 101, https ://doi.org/10.1017/s0954 57940 30003 12. 91–97. https ://doi.org/10.1016/j.earlh umdev .2016.08.013. Sanlorenzo, L. A., Stark, A. R., & Patrick, S. W. (2018). Neonatal absti- Koumoula, A. (2012). The course of attention deficit hyperactivity nence syndrome: an update. Current opinion in pediatrics, 30(2), disorder (ADHD) over the life span. Psychiatriki, 23(Suppl 1), 182–186. https ://doi.org/10.1097/MOP.00000 00000 00058 9. 49–59. Sarfi, M., Smith, L., Waal, H., & Sundet, J. M. (2011). Risks and Lee, S. J., Pritchard, V. E., Austin, N. C., Henderson, J. M. T., & realities: Dyadic interaction between 6-month-old infants and Woodward, L. J. (2020). Health and neurodevelopment of chil- their mothers in opioid maintenance treatment. Infant Behavior dren born to opioid-dependent mothers at school entry. Journal and Development, 34(4), 578–589. https ://doi.or g/10.1016/j. of Developmental & Behavioral Pediatrics, 41(1), 48–57. https:// infbe h.2011.06.006. doi.org/10.1097/dbp.00000 00000 00071 1. Sasser, T. R., Kalvin, C. B., & Bierman, K. L. (2016). Developmental Lee, S. J., Woodward, L. J., & Henderson, J. M. T. (2019). Educa- trajectories of clinically significant attention-deficit/hyperactivity tional achievement at age 9.5 years of children born to mothers disorder (ADHD) symptoms from grade 3 through 12 in a high- maintained on methadone during pregnancy. PLoS One, 14(10), risk sample: Predictors and outcomes. Journal of Abnormal Psy- e0223685. https ://doi.org/10.1371/journ al.pone.02236 85 chology, 125(2), 207–219. https ://doi.org/10.1037/abn00 00112 . Lenze, E. J., & Wetherell, J. L. (2011). A lifespan view of anxiety Siqveland, T. S., Haabrekke, K., Wentzel-Larsen, T., & Moe, V. disorders. Dialogues Clin Neurosci, 13(4), 381–399. (2014). Patterns of mother-infant interaction from 3 to 12 months Lester, B. M., & Padbury, J. F. (2009). Third pathophysiology of pre- among dyads with substance abuse and psychiatric problems. natal cocaine exposure. Developmental Neuroscience, 31(1–2), Infant Behavior and Development, 37(4), 772–786. https ://doi. 23–35. https ://doi.org/10.1159/00020 7491.org/10.1016/j.infbe h.2014.09.003. Levine, T. A., & Woodward, L. J. (2017). Early inhibitory control Sithisarn, T., Bada, H. S., Dai, H., Reinhardt, C. R., Randall, D. C., and working memory abilities of children prenatally exposed to & Legan, S. J. (2008). Effects of perinatal oxycodone exposure methadone. Early Hum Dev, 116, 68–75. https://doi.or g/10.1016/j. on the response to CRH in late adolescent rats. Neurotoxicol- earlh umdev .2017.11.010. ogy and Teratology, 30(2), 118–124. https ://doi.org/10.1016/j. Levine, T. A., & Woodward, L. J. (2018). Early inhibitory control ntt.2007.12.010. and working memory abilities of children prenatally exposed to Sithisarn, T., Legan, S. J., Westgate, P. M., Wilson, M., Wellmann, methadone. Early Hum Dev, 116, 68–75. https://doi.or g/10.1016/j. K., Bada, H. S., & Barron, S. (2017). The Effects of Perinatal earlh umdev .2017.11.010. Oxycodone Exposure on Behavioral Outcome in a Rodent Model. Li, S.-C. (2003). Biocultural orchestration of developmental plasticity Front Pediatr, 5, 180. https ://doi.org/10.3389/fped.2017.00180 . across levels: The interplay of biology and culture in shaping the Slamberova, R., Riley, M. A., & Vathy, I. (2005). Cross-generational mind and behavior across the life span. Psychological Bulletin, effect of prenatal morphine exposure on neurobehavioral develop- 129(2), 171–194. https ://doi.org/10.1037/0033-2909.129.2.171. ment of rat pups. Physiological Research, 54(6), 655–660. Ma, L., Chen, Y. H., Chen, H., Liu, Y. Y., & Wang, Y. X. (2011). The Slinning, K. (2004). Foster placed children prenatally exposed to poly- function of hypothalamus-pituitary-adrenal axis in children with substances–attention-related problems at ages 2 and 4 1/2. Euro- ADHD. Brain Research, 1368, 159–162. https://doi.or g/10.1016/j. pean Child and Adolescent Psychiatry, 13(1), 19–27. https ://doi. brain res.2010.10.045.org/10.1007/s0078 7-004-0350-x. Mangin, K. S., Horwood, L. J., & Woodward, L. J. (2017). Cognitive Spencer, T. J., Biederman, J., & Mick, E. (2007). Attention-deficit/ development trajectories of very preterm and typically develop- hyperactivity disorder: diagnosis, lifespan, comorbidities, and ing children. Child Development, 88(1), 282–298. https ://doi. neurobiology. Ambulatory Pediatrics, 7(1 Suppl), 73–81. https :// org/10.1111/cdev.12585 . doi.org/10.1016/j.ambp.2006.07.006. Mischel, W., Ayduk, O., Berman, M. G., Casey, B. J., Gotlib, I. H., Statistics New Zealand. (2006). Census Regional Summary. New Jonides, J., & Shoda, Y. (2011). “Willpower” over the life span: Zealand. decomposing self-regulation. Social Cognitive and Affective Neu- Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. roscience, 6(2), 252–256. https ://doi.org/10.1093/scan/nsq08 1. (1996). The Revised Conflict Tactics Scales (CTS2): Development Nygaard, E., Slinning, K., Moe, V., & Walhovd, K. B. (2016). Behav- and Preliminary Psychometric Data. Journal of Family Issues, ior and Attention Problems in Eight-Year-Old Children with 17(3), 283–316. https ://doi.org/10.1177/01925 13960 17003 001. Prenatal Opiate and Poly-Substance Exposure: A Longitudinal Sundelin Wahlsten, V., & Sarman, I. (2013). Neurobehavioural Study. PLoS ONE, 11(6), e0158054. https ://doi.or g/10.1371/ development of preschool-age children born to addicted moth- journ al.pone.01580 54. ers given opiate maintenance treatment with buprenorphine dur- Oei, J. L., Melhuish, E., Uebel, H., Azzam, N., Breen, C., Burns, L., ing pregnancy. Acta Paediatrica, 102(5), 544–549. https ://doi. Wright, I. M. (2017). Neonatal abstinence syndrome and high org/10.1111/apa.12210. school performance. Pediatrics, 139(2). https ://doi.org/10.1542/ Taylor, C. C., Soong, Y. I., Wu, D., Yee, J. S., & Szeto, H. H. (1997). peds.2016-2651 Morphine stimulates adrenocorticotropin and cortisol release in 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 457 the late-term ovine fetus. Pediatric Research, 41(3), 411–415. Winkelman, T. N. A., Villapiano, N., Kozhimannil, K. B., Davis, M. https ://doi.org/10.1203/00006 450-19970 3000-00018 . M., & Patrick, S. W. (2018). Incidence and costs of neonatal absti- Totsika, V., & Sylva, K. (2004). The Home Observation for Meas- nence syndrome among infants with medicaid: 2004–2014. Pedi- urement of the Environment Revisited. Child and Adolescent atrics, 141(4). https ://doi.org/10.1542/peds.2017-3520 Mental Health, 9(1), 25–35. h t t p s : / / d o i . o r g / 1 0 . 1 0 4 6 / j . 1 4 7 5- Woodward, L. J., McPherson, C. C., & Volpe, J. J. (2018). Chapter 38 357X.2003.00073 .x. - Passive addiction and teratogenic effects. In J. J. Volpe, T. E. van Buuren, S., Groothuis-Oudshoorn, K., Vink, G., Schouten, R., Inder, B. T. Darras, L. S. de Vries, A. J. du Plessis, J. J. Neil, & Robitzsch, A., Doove, L., Gray, B. (2020). Package ‘mice’. J. M. Perlman (Eds.), Volpe’s Neurology of the Newborn (Sixth Van Roy, B., Groholt, B., Heyerdahl, S., & Clench-Aas, J. (2010). Edition) (pp. 1149–1189.e1120): Elsevier. Understanding discrepancies in parent-child reporting of emo- tional and behavioural problems: Effects of relational and socio- Publisher’s Note Springer Nature remains neutral with regard to demographic factors. Bmc Psychiatry, 10(1), 1–12. https ://doi. jurisdictional claims in published maps and institutional affiliations. org/10.1186/1471-244x-10-56. WHO. (1993). Composite International Diagnostic Interview (CIDI). Geneva, Switzerland: World Health Organization. 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Abnormal Child Psychology Springer Journals

Emotional and Behavioral Trajectories of 2 to 9 Years Old Children Born to Opioid-Dependent Mothers

Loading next page...
 
/lp/springer-journals/emotional-and-behavioral-trajectories-of-2-to-9-years-old-children-0Qi2nn5J14

References (84)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
ISSN
0091-0627
eISSN
2730-7174
DOI
10.1007/s10802-020-00766-w
Publisher site
See Article on Publisher Site

Abstract

Maternal opioid use in pregnancy has increased dramatically. Knowledge about children’s longer-term emotional and behav- ioral development after prenatal opioid exposure is scarce. A regional sample of 89 opioid-exposed and 104 non-exposed comparison children were studied prospectively at ages 2, 4.5, and 9 years using the Strengths and Difficulties Questionnaire (SDQ) completed by primary caregivers. Across all childhood assessments, opioid-exposed children obtained significantly higher total difficulties scores than non-exposed comparison children. Growth curve modeling revealed that, relative to their same age peers, opioid-exposed children’s emotional and behavioral difficulties significantly worsened over time. Moreo - ver, fixed effects estimates showed that total difficulties trajectories were poorer for children subject to higher prenatal risk (Est = 1.78, 95% CI = [0.46, 3.09]) who were born to mothers with high levels of social adversity (1.11 [0.51, 1.71]), and were then raised in families characterized by high levels of psychosocial risk (1.94 [0.90, 2.98]) and unstable caregiving (1.91 [0.33, 3.48]). A complex set of pre- and postnatal processes contribute to opioid-exposed children’s emotional and behavioral development. Efforts to mitigate the long-term consequences of opioid use in pregnancy need to consider both children’s and their caregivers’ biopsychosocial risks. Keywords Opioids · Neonatal abstinence/opioid withdrawal syndrome (NAS/NOWS) · Developmental trajectories · Biopsychosocial risk Over the last 20 years, the use and abuse of licit and illicit increased dramatically (Bateman et  al., 2014; Patrick, opioids has skyrocketed globally (Sanlorenzo, Stark, & Davis, Lehmann, & Cooper, 2015; Winkelman, Villapiano, Patrick, 2018), including among women of reproductive- Kozhimannil, Davis, & Patrick, 2018). Methadone age (Bateman et al., 2014; CDC, 2017). As a consequence, maintenance treatment (MMT) provides advantages the number of infants prenatally exposed to opioids has over continued illicit drug use and is the most common management approach for opioid-dependent pregnant women (Jones, O’Grady, Malfi, & Tuten, 2008). However, * Julia Jaekel MMT involves chronic fetal opioid exposure placing the jjaekel@utk.edu infant at risk of neonatal opioid withdrawal syndrome Department of Child & Family Studies, University (NOWS) (Conradt, Crowell, & Lester, 2018). Longitudinal of Tennessee Knoxville, Knoxville, USA studies describing children’s longer-term development after Department of Psychology, University of Tennessee prenatal opioid exposure are scarce (Azuine et al., 2019; Knoxville, Knoxville, USA Kaltenbach et  al., 2018; Lee, Woodward, & Henderson, Department of Psychology, University of Warwick, Coventry, 2019; Levine & Woodward, 2018). UK One of the most critical developmental tasks for children School of Health Sciences, University of Canterbury, is to successfully regulate their behavior and emotions Christchurch, New Zealand (Fergusson, Horwood, & Ridder, 2005; Mischel et  al., School of Psychology, Speech and Hearing, University 2011). There is growing suggestion that prenatal opioid- of Canterbury, Christchurch, New Zealand exposure may impact children’s self-regulatory abilities, Department of Public Health, University of Tennessee and has consistently been associated with poorer behavioral Knoxville, Knoxville, USA Vol.:(0123456789) 1 3 444 Research on Child and Adolescent Psychopathology (2021) 49:443–457 control and attention-deficit hyperactivity disorder (ADHD) considering a wide range of pre- and postnatal factors as symptoms (Azuine et  al., 2019; Lee, Pritchard, Austin, potential confounders. Previous approaches have thus Henderson, & Woodward, 2020; Levine & Woodward, precluded opportunities to separately examine the roles 2017; Nygaard, Slinning, Moe, & Walhovd, 2016; Ornoy, of: a) other adverse prenatal exposures that co-occur with 2003). In addition, compared with their non-exposed fetal opioid exposure; b) maternal social background factors peers, children exposed prenatally to opioids may be at such as education and marital status that are associated with increased risk for emotional difficulties (Nygaard et  al., both opioid use and child outcomes; and, importantly, c) 2016), conduct problems and disruptive behaviors (Azuine the intervening effects of the quality and stability of the et al., 2019; Ornoy, 2003). However, follow-up studies of postnatal rearing environment in shaping child outcomes children exposed prenatally to opioids extending beyond (Azuine et  al., 2019; Crea et  al., 2008). The latter are infancy and early childhood are rare and methodologically especially important since although they may be correlated limited (Fill et al., 2018; Lee et al., 2019; Oei et al., 2017). with the independent variable of interest (prenatal opioid Specifically, existing studies typically report only cross- exposure) they occur after pregnancy and thus are more sectional outcomes at a single age of assessment. Of the few accurately viewed as intervening factors that may mediate longitudinal studies that do exist, these are confounded by or moderate associations between opioid exposure and child a reliance on small samples, high sample attrition (Nygaard outcomes. Furthermore, knowledge about the contributions et al., 2016), lack of tester blinding, no or poor measurement of the postnatal environment to children’s behavioral of other drug exposures (Crea, Barth, Guo, & Brooks, 2008; trajectories is critical in informing intervention efforts aimed Nygaard et al., 2016), or failure to include a comparison at supporting these and future cohorts of opioid-exposed group from the same population studied. This significantly children and families longer term. Specifically, it is critical limits our current understanding and interpretation of to identify variables in children’s environments that can available data. be modified through intervention, in order to facilitate Yet, we know from normative population studies that the resilience for this at-risk group. Most importantly, almost prevalence as well as the nature and severity of children’s no studies have examined the complex interplay of prenatal behavioral and emotional problems vary with age (Koumoula, exposures, social background, and postnatal environmental 2012; Lenze & Wetherell, 2011; Sasser, Kalvin, & Bierman, factors that may heighten or mitigate risk for infants exposed 2016; Spencer, Biederman, & Mick, 2007). For instance, to opioids. We propose that a biopsychosocial framework ADHD and conduct problems tend to have their peak onset and may help incorporate the multimodal and multidisciplinary highest prevalence in childhood, whereas emotional difficulties theoretical aspects mentioned above (Bolton & Gillett, generally peak later (Kessler et al., 2004). Anxiety problems, 2019). Conradt, Crowell, and Lester (2018), for instance, for example, can occur in childhood but have a median age of have proposed an integrated model of how different factors onset in early adolescence, while rates of depressive symp- such as socioeconomic status, maternal psychopathology toms typically rise dramatically during adolescence and peak and nutrition may alter opioid-exposed infants’ development in young adulthood (Kessler et al., 2004). prenatally, and they propose that effects could extend into the At present, it is not known whether prenatally opioid- postnatal period, for instance via mother-infant interactions exposed children follow similar behavioral trajectories as (Conradt et al., 2018). In line with this model, Fig. 1 shows a their non-exposed peers. Yet we can only really understand schematic depiction of the hypothetical pathways of interest the mechanisms underlying potential problems associated considered in this study. The potential importance of each of with prenatal opioid-exposure when accounting for these key processes or risk factors is also reviewed briefly normative developmental changes. Methodologically, below. this can be accomplished by longitudinally administering With regard to prenatal and neonatal exposures and risks, an established, standardized, and developmentally opioid use in pregnancy has been associated with maternal appropriate instrument such as the Strengths and Difficulties perinatal depression and psychiatric illness, and smoking Questionnaire (SDQ) (Goodman, 2001) to assess the during pregnancy, preterm birth, smaller head circumference development of children exposed to opioids as well as a at birth, and a longer hospital stay (Azuine et  al., 2019; representative regional comparison group. Conradt et  al., 2019; Davie-Gray, Moor, Spencer, & A further important issue that is vital to address the Woodward, 2013; Kelty & Preen, 2019; Woodward, longer-term consequences of the growing public health McPherson, & Volpe, 2018). Postnatally, the quality of problem of families affected by opioid use is to understand infants’ caregiving may be particularly important (Hatzis, the developmental mechanisms that might place children Dawe, Harnett, & Barlow, 2017; Konijnenberg, Lund, born to opioid dependent women at increased risk of & Melinder, 2015), but we know remarkably little about later problematic developmental trajectories. This is an whether variations in rearing environments are associated issue that has, to date, been given scant attention beyond with developmental variability after opioid-exposure. 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 445 Fig. 1 Schematic Model of the Hypothetical Pathways of Interest Children who were prenatally exposed to opioids have The aim of this study is to characterize the behavioral been shown to be at increased risk of exposure to psychosocial and emotional trajectories from age 2 to 9  years of a adversity and poor parental support (Hatzis et  al., 2017; regionally representative sample of prenatally opioid- Konijnenberg, Sarfi, & Melinder, 2016; Sarfi, Smith, Waal, exposed children relative to their non-exposed peers. & Sundet, 2011; Siqveland, Haabrekke, Wentzel-Larsen, & Utilizing a multi-level approach, we address key Moe, 2014). However, previous studies have not disentangled questions concerning predictive mechanisms of prenatal, the potential effects of prenatal and postnatal factors, but instead social background, and postnatal environmental risk. treated them collectively as confounders. What’s more, it has We formed three hypotheses: Compared with their rarely been investigated how social background risk indicators non-exposed peers, (1) children who were prenatally may be associated with combined prenatal and postnatal exposed to opioids will exhibit increased emotional and risks when studying children’s long-term development. behavioral difficulties at ages 2, 4.5, and 9 years, (2) their During pregnancy, there are complex mechanisms at play trajectories will remain stable or worsen with age, and between maternal factors such as pregnancy nutrition, (3) prenatal and social risks as well as postnatal rearing social disadvantage, and psychiatric risk that are correlated conditions will explain individual variations in children’s with pregnancy opioid use (Woodward et  al., 2018), but behavioral trajectories. their effects on long-term child development have not been investigated among opioid-exposed mother–child dyads. In addition, prenatal (i.e., biological) compared with postnatal Methods environmental factors may affect long-term trajectories via different pathways (Bendersky & Lewis, 1994; Jusiene, Participants Breidokiene, & Pakalniskiene, 2015), and their combined effects may potentially be additive, interactive, or both (Evans, Participants were assessed as part of a prospective longitudinal Li, & Whipple, 2013; Li, 2003; Sameroff & MacKenzie, 2003). study of two groups of children born between 2003 and 2008 at Sophisticated longitudinal study designs are timely and essential Christchurch Women’s Hospital, New Zealand (Davie-Gray et al., to assess the complex interplay of these biopsychosocial risks 2013; Levine & Woodward, 2017). Mothers were recruited during on opioid-exposed children’s long-term development. their third trimester or at birth. Exclusion criteria included very 1 3 446 Research on Child and Adolescent Psychopathology (2021) 49:443–457 preterm birth (≤ 32 weeks), congenital abnormality, HIV, fetal and perinatal medical treatment has been published alcohol syndrome, and non-English speaking. previously (Davie-Gray et al., 2013; Lee et al., 2020, 2019). The index group consisted of infants born to opioid- The comparison group consisted of randomly identified dependent mothers receiving MMT. Over the recruitment non-opioid-exposed infants born at the same hospital period, 120 mothers were eligible for inclusion in the study. over the same recruitment period. A total of 169 pregnant Of these, 99 mothers and 100 infants (including one set women were approached, with 108 mother and 110 infants of twins) were successfully recruited (84% of all eligible (including two sets of twins) successfully recruited. Regional opioid-exposed infants). Retention to age 9 years was 85%. census data showed that the socioeconomic profile of this Figure 1 provides an overview of recruitment and retention sample was representative of the families living in the region over the study period. Demographic characteristics are at the time (Statistics New Zealand, 2006). Retention to age presented in Table 1. A detailed description of recruitment 9 years was 90% (see Fig. 2 and Table 1). Table 1 Descriptive Participant Characteristics, Pregnancy to 18 Months Non-exposed Opioid-exposed Mean / % difference p Maternal Charac- teristics Age (years) 31.64 (5.44) 29.77 (5.26) -1.87 0.015 Pregnancy methadone - 62.63 (35.14) 62.63 < 0.001 dose (mg) Pregnancy depression 6.83 (4.83) 14.21 (6.13) 7.38 < 0.001 (EDPS) Pregnancy psychiatric 15.89 56.52 36.35 < 0.001 illness (%) Pregnancy nutrition 90.61 (25.31) 56.03 (20.81) -34.57 < 0.001 score Pregnancy smoking 1.39 (3.88) 13.10 (8.90) 11.71 < 0.001 (cigarettes/day) % minority ethnicity 17.76 25.00 7.24 0.212 % younger mother 5.61 3.26 -2.35 0.427 % single parent 9.35 48.91 39.56 < 0.001 % no formal educa- 19.63 80.43 60.80 < 0.001 tion qualification Depression at 5.26 (3.99) 9.79 (6.96) 4.53 < 0.001 18 months (EPDS) Infant Character- istics % female 52.34 42.39 -9.95 0.161 Gestational age 39.27 (1.70) 38.82 (1.69) -0.45 0.063 (weeks) Head circumference 0.23 (0.86) -0.24 (0.88) -0.47 < 0.001 at birth (z-score) Total days of mor- - 63.66 (41.02) 63.66 < 0.001 phine Total days in hospital 3.44 (2.29) 16.07 (12.14) 12.62 < 0.001 Primary caregiver 4.30 43.00 38.07 < 0.001 change (%) Composite Scores Prenatal risk (z-score) 0.34 (0.51) 2.18 (1.76) 1.84 < 0.001 Social risk (sum 0.57 (0.93) 1.37 (0.89) 0.79 < 0.001 score) Postnatal environ- -0.26 (0.48) 0.32 (0.63) 0.58 < 0.001 ment risk (z-score) Data are reported in Mean (SD), if not noted otherwise 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 447 Fig. 2 Study Flow Chart 1 3 448 Research on Child and Adolescent Psychopathology (2021) 49:443–457 α = 0.79) (Cox, Holden, & Sagovsky, 1987). Pregnancy Procedure psychiatric illness was included as a binary-coded variable (0 = no, 1 = yes). All variables were equally weighted and Participants were assessed at five time points: late pregnancy/ birth, 18 months, 2, 4.5, and 9 years. The Upper South B averaged, a higher total score indicated higher perinatal risk. Maternal Social Risk. A composite score was created Regional Ethics Committee, Canterbury, New Zealand (Ref: URB/07/10/042) approved the study protocol and the study as a summative index of five dichotomous maternal social risk indicators (i.e., single parent, minority ethnicity, no high was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later school degree, low socioeconomic status, mother < 21 years old) (Lee et al., 2020). amendments or comparable ethical standards. Written informed consent was obtained from biological mothers or Postnatal Rearing Environment Risk. A continuously scored index of postnatal caregiving quality was created by primary caregivers at each assessment point, with children also providing verbal assent at age 9 years. At each follow-up, averaging the following z-standardized variables: the Home Observation for Measurement of the Environment (HOME) participants received a $20 gift card as compensation for their involvement in the study. total score (α = 0.74; reverse coded) (Caldwell & Bradley, 1984; Totsika & Sylva, 2004), parenting stress measured using a 16-item scale assessing caregiver exposure to, for example, financial, housing, social support and relationship Measures stress (α = 0.68), maternal depression at 18 months (EPDS, α = 0.81) (Cox, Chapman, Murray, & Jones, 1996; Cox et al., Strengths and Difficulties Questionnaire (SDQ). Primary caregivers completed the SDQ at 2, 4.5 and 9 years during 1987) and 4.5  years (Composite International Diagnostic Interview, α = 0.89) (WHO, 1993), corporal punishment at structured parent interviews. The SDQ contains 25-items that measure 5 subscales: emotional symptoms, hyperactivity- 18 months (α = 0.72) (Straus, Hamby, Boney-McCoy, & Sug- arman, 1996), and maternal total illicit drug use between birth inattention, conduct problems, peer problems, and prosocial behavior (Goodman, 1997). Each subscale is scored on a and 4.5 years (see Table S1). Illicit drug use was continuously assessed at child age 18 months and 4.5 years based on ques- 3-point Likert-type scale (0 = not true, 1 = somewhat true, 2 = certainly true). A total difficulties score is calculated by tions from the Composite International Diagnostic Interview (WHO, 1993) about the use/abuse of a range of drugs includ- summing the scores of the four difficulties scales (Cronbach’s α = 0.62, 0.70, and 0.75 for the total scores respectively at each ing heroine, prescription opioids, cannabis, barbiturates, stim- ulants, hallucinogens inhalants and other illicit prescription age in this study). The SDQ is a widely used screening tool with good test–retest reliability, e.g., r = 0.72 for parent-rated drugs (Davie-Gray et al., 2013). The variable was binary coded (0 = no, 1 = yes) for inclusion in the risk score due to limited total difficulties scores over 4–6 months (Goodman, 2001). Maternal Pregnancy/Birth Characteristics. Detailed variance and non-normal distribution. All variables were equally weighted and averaged, a information about mothers’ mental health, nutrition, and social risks was collected via comprehensive structured higher total score indicated higher postnatal rearing envi- rd ronment risk. interviews administered in the late 3 trimester or at birth (Lee et al., 2020, 2019), see Table S1 for details. Primary Caregiver Changes. Any transitions or disruptions in children’s primary caregivers were documented Infant Characteristics. Detailed information about perina- tal characteristics (e.g., gestational age, head circumference at and coded as a binary variable (0 = no, 1 = yes). Preliminary analyses showed that primary caregiver changes indicated an birth) and infant medical treatment was extracted from hospital records. independent and specific risk that was not correlated with the more general and normally distributed index score of postnatal We created three risk index scores in order to include a diverse and large set of biopsychosocial risk variables in our rearing environment risk (Spearman’s rho = 0.12, p = 0.09). analyses, to control for potential multicollinearity between variables, and to increase external validity and reliability. Statistical Analyses Other Prenatal and Perinatal Risk Exposures. A continuously scored index of pre- and neonatal biological Descriptive analyses were conducted in SPSS 24 to examine group differences between opioid-exposed and risks was created by averaging the following z-standardized variables, thereby weighting their independent contributions non-exposed comparison children with regard to bio- logical and environmental factors as well as emotional equally: infant gestational age, head circumference at birth, 1-min APGAR score, and pregnancy nutrition (all reverse and behavioral difficulties. For the main analyses, a low number of data missing at random was imputed using coded), total days in hospital (as an index of postnatal illness severity), smoking during pregnancy, and maternal depres- the stochastic imputation method implemented in “mice 3.8.0” package in R (van Buuren et al., 2020). Imputation sion at birth (Edinburgh Postnatal Depression Scale (EPDS), 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 449 was confined to participants with at least two of the three subject to higher rates of psychiatric illness. Opioid- SDQ assessments available, resulting in a longitudinal exposed infants tended to have been born, on average, sample of 89 opioid-exposed and 104 non-exposed chil- to younger mothers than non-exposed children, although dren. Correlations between variables included in the rates of early motherhood (< 21  years) did not differ, prenatal biological and social risk, and postnatal rearing and they more often had single mothers without formal environment scores are presented in Table  2. Trajecto- educational qualifications (Table 1). Rates of minority ries of emotional and behavioral development from 2 to ethnicity were the same across groups. Accordingly, 9 years were assessed using linear mixed effects model the prenatal and social risk index scores were higher analysis in Stata 16. Trajectories of total difficulty scores in the opioid-exposed group. In addition, the postnatal over time were modeled as a function of age at assess- rearing environment risk index score was also higher, and ment (years) and group status (opioid-exposed vs. non- they had more changes in primary caregivers (i.e., care exposed), including both fixed and random effects, fol- disruptions). For instance, at ages 2, 4.5, and 9  years, lowing an established approach (Mangin, Horwood, & respectively, 80%, 69%, and 54% of the opioid-exposed Woodward, 2017). As part of a stepwise process, a quad- children were in the care of their biological mothers. For ratic term for age (age*age) was included in Model 2. In the non-exposed group, 100% of children were in the care Model 3, an interaction term (age*group) was added to of their biological mothers at ages 2 and 4.5 years, with test whether trajectories differed between groups. This 99% in their mothers’ care at age 9 years. model was then extended to assess the effects of child sex, Emotional and Behavioral Difficulties . Confirming prenatal biological and social risk, the postnatal rearing Hypothesis 1, prenatally opioid-exposed children had sig- environment, and primary caregiver changes on emotional nificantly higher total difficulties scores than their non- and behavioral development, including potential additive exposed peers across all childhood assessments, with the and interactive effects. Log-likelihood goodness of fit mean difference [95% confidence interval] increasing from tests were used to compare each growth curve model to 2.06 [0.61, 3.52] at age 2 years to 7.13 [5.30, 8.96] at age the previous one. 9 years (Table 3, Fig. 2). This was further confirmed by visual inspection of the data in Fig. 2, which suggests that Results opioid-exposed children’s difficulties appear to increase from ages 2 to 9  years, whereas the scores of children Descriptive Sample Characteristics. Indicators of in the non-exposed comparison group tended to remain prenatal and social background risk tended to be higher relatively stable. among children in the opioid-exposed group. On average, As part of a sensitivity analysis, we examined the opioid-exposed infants had lower head circumference domain-specific nature of opioid-exposed children’s z-scores at birth and spent more days in hospital than non- emotional and behavioral difficulties. Table 3 shows the exposed infants. There were however no group differences mean scores of the two study groups on each of the SDQ in infant sex, gestational age, and 1-min APGAR scores. subscales (emotional, conduct, inattention/hyperactivity, Opioid-using mothers had higher prenatal depression and and peer problems) as well as the total score by age at lower nutrition scores, reported higher average numbers assessment (2, 4.5, and 9 years), also see Fig. 3. Across of daily cigarettes smoked during pregnancy, and were all four domains, compared with their non-exposed Fig. 3 SDQ Total Difficulties Scores of Opioid-Exposed Versus Non-Exposed Children at 2, 4.5, and 9 Years 1 3 450 Research on Child and Adolescent Psychopathology (2021) 49:443–457 1 3 Table 2 Correlations Between Variables Included in the Neonatal Biological, Social, and Rearing Environment Risk Index Scores 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 1. infant gestational age - -0.09 -0.07 -0.28 -0.12 -0.22 -0.10 -0.05 0.01 -0.11 -0.08 0.05 0.19 -0.10 -0.08 -0.04 -0.05 -0.08 -0.41 -0.08 -0.17 2. head circumference - 0.00 -0.19 -0.17 -0.04 0.08 -0.17 0.03 -0.17 -0.20 -0.06 0.08 -0.02 -0.06 -0.17 -0.01 -0.16 -0.29 -0.12 -0.14 3. pregnancy nutrition - 0.05 -0.08 0.04 -0.19 -0.04 0.03 -0.18 -0.12 0.03 0.02 -0.05 -0.11 -0.03 0.07 -0.06 -0.13 -0.11 -0.06 4. total days in hospital - 0.44 0.05 -0.04 0.32 -0.03 0.44 0.51 -0.10 -0.33 0.24 0.29 0.32 -0.06 0.32 0.89 0.36 0.38 5. postnatal depression - 0.17 0.02 0.40 0.06 0.42 0.40 -0.01 -0.29 0.23 0.42 0.34 0.05 0.28 0.55 0.37 0.38 6. pregnancy psychiatr. illness - -0.01 0.17 -0.09 0.07 0.00 0.04 -0.11 0.21 0.24 0.06 0.04 0.12 0.21 0.01 0.21 7. pregnancy smoking - -0.02 0.19 0.19 0.06 0.14 -0.03 -0.04 -0.08 0.01 0.04 -0.20 0.13 0.11 -0.04 8. single parent - 0.08 0.47 0.47 0.22 -0.26 0.31 0.30 0.26 0.04 0.31 0.36 0.52 0.38 9. minority ethnicity - 0.12 0.26 0.20 -0.10 0.07 0.02 0.12 0.03 0.01 -0.04 0.47 0.12 10. no high school degree - 0.62 0.19 -0.44 0.28 0.33 0.33 -0.05 0.28 0.50 0.58 0.44 11. low SES - 0.15 -0.40 0.31 0.26 0.37 -0.01 0.37 0.55 0.59 0.47 12. mother < 21 years - -0.10 0.04 -0.09 -0.01 0.10 0.02 -0.07 0.25 0.06 13. HOME total score - -0.32 -0.27 -0.28 -0.10 -0.31 -0.36 -0.45 -0.61 14. parenting stress - 0.51 0.27 0.21 0.32 0.27 0.35 0.71 15. mat. depression, 18 m - 0.43 0.15 0.29 0.33 0.32 0.71 16. mat. depression, 4.5y - 0.05 0.22 0.36 0.28 0.58 17. corporal punishment - 0.15 0.00 0.23 0.47 18. total illicit drug use - 0.32 0.33 0.56 19. Prenatal Risk Exposure - 0.38 0.43 20. Social Risk Score - 0.57 21. Postnatal Rearing - Environment Risk Research on Child and Adolescent Psychopathology (2021) 49:443–457 451 Table 3 SDQ Subscale and Non-Exposed Opioid-Exposed Total Difficulties Scores of Opioid-Exposed Versus Non- SDQ Score n Mean (SD) n Mean (SD) Mean Difference (95% p Exposed Children at 2, 4.5, and CI) 9 Years Emotional Problems 2 y 107 1.36 (1.35) 92 1.45 (1.53) 0.09 (-0.31, 0.49) 0.666 4.5 y 103 1.27 (1.42) 87 1.90 (1.88) 0.64 (0.16, 1.11) 0.009 9 y 99 1.74 (1.78) 81 2.98 (2.39) 1.24 (0.61, 1.87) < 0.001 Conduct Problems 2 y 107 2.28 (1.91) 92 2.71 (2.09) 0.43 (-0.13, 0.99) 0.131 4.5 y 103 0.92 (1.11) 87 2.47 (1.90) 1.55 (1.09, 2.01) < 0.001 9 y 99 1.07 (1.54) 81 2.96 (2.42) 1.89 (1.28, 2.51) < 0.001 Hyperactivity/Inat- 2 y 107 3.48 (2.42) 92 4.50 (2.47) 1.02 (0.33, 1.70) 0.004 tention 4.5 y 103 2.44 (2.11) 87 4.29 (2.38) 1.85 (1.21, 2.49) < 0.001 9 y 99 2.65 (2.26) 81 5.17 (2.89) 2.53 (1.75, 3.30) < 0.001 Peer Problems 2 y 107 1.25 (1.50) 92 1.77 (1.79) 0.53 (0.07, 0.99) 0.025 4.5 y 103 0.93 (1.21) 87 1.59 (1.56) 0.65 (0.25, 1.06) 0.002 9 y 99 1.23 (1.47) 81 2.63 (2.06) 1.40 (0.86, 1.94) < 0.001 Total Difficulties 2 y 107 8.37 (4.80) 92 10.43 (5.60) 2.06 (0.61, 3.52) 0.006 4.5 y 103 5.56 (4.12) 87 10.25 (5.51) 4.69 (3.27, 6.07) < 0.001 9 y 99 6.66 (5.00) 81 13.79 (7.00) 7.13 (5.30, 8.96) < 0.001 same-age peers, there was an increase in opioid-exposed children in the opioid exposed group. However, by age children’s problems from 2 to 9  years. Specifically, 4.5 years, opioid-exposed children were found to have opioid-exposed children had higher inattention/ higher problem scores across all four SDQ domains, and hyperactivity (mean difference 1.02 [0.33, 1.70]) and these group differences further increased at age 9 years. peer problems (0.53 [0.07, 0.99]) than their peers at Trajectories of Difficulties. Linear mixed effects 2  years, but significant between group differences in growth curve models were run to examine the behavio- emotional and conduct problems were not detected ral trajectories of both study groups based on their SDQ despite the tendency for higher levels to be reported for total difficulties scores from 2 to 9 years of age, using Table 4 Multilevel Growth Curve Model Results of SDQ Total Difficulties Score Trajectories from 2 to 9 Years of Age (n = 193) Total Difficulties Model 1 Model 2 Model 3 Model 4 Model 5 Est (SE) Est (SE) Est (SE) Est (SE) Est (SE) Fixed effects Age 0.39 (0.25) -6.57 (1.31)*** -10.08 (1.48)*** -10.08 (1.48)*** -10.08 (1.48)*** Age - 1.74 (0.32)*** 1.74 (0.32) *** 1.74 (0.32)*** 1.74 (0.32)*** Group (opioid-exposed) 4.23 (0.58)*** 4.23 (0.58)*** -0.13 (1.05) -3.91 (1.15)** -4.61 (1.17)*** Interaction (group*age) - - 2.40 (0.48) *** 2.40 (0.48)*** 2.40 (0.48)*** Sex - - - -0.52 (0.52) -0.45 (0.51) Prenatal Risk - - - 2.03 (0.67)** 1.78 (0.67)** Social Risk - - - 1.18 (0.31)*** 1.11 (0.31)*** Postnatal Environmt. Risk - - - 1.56 (0.52)** 1.94 (0.53)*** Primary Caregiver Change - - - - 1.91 (0.80)* constant 1.94 (1.01) 7.74 (1.47)*** 14.12 (1.94)*** 20.41 (2.20)*** 19.09 (2.25)*** Random effects SD (Age) 2.21 (0.34)* 2.42 (0.30)* 2.11 (0.31)* 2.11 (0.31)* 2.11 (0.31)* SD (constant) 4.68 (0.75)* 5.16 (0.66)* 4.67 (0.67)* 3.99 (0.72)* 3.88 (0.73)* correlation (Age,constant) -0.72 (0.08)* -0.76 (0.06)* -0.70 (0.08)* -0.74 (0.08)* -0.74 (0.08)* Log-likelihood -1,769.61 -1,755.90 -1,744.16 -1,720.39 -1,717.62 for fixed effects *p < 0.05; **p < 0.01; ***p < 0.001; for random effects * marks a 95% confidence interval not including 0 1 3 452 Research on Child and Adolescent Psychopathology (2021) 49:443–457 the stepwise approach outlined above. Table  4 shows Finally, we examined possible interactive effects detailed results for five models that were subsequently between prenatal biological and social risk, as well as fitted to the data. In Model 1, trajectories of total difficul- with postnatal rearing environmental inf luences on chil- ties over time were modeled as a linear function of age at dren’s overall difficulties trajectories. None of these inter- assessment (years) and group status (opioid-exposed vs. action effects reached statistical significance. non-exposed). On average, opioid-exposed children had Thus Model 5, showing independent (i.e., additive) fixed total difficulties scores with coefficient estimates (CE) main effects, is considered the final model. Figure  4 shows pre- of 4.23 (95% CIs [3.09, 5.37]) points higher than their dicted total difficulties trajectories by group based on Model non-exposed peers at any given age (p < 0.001), while 5. Overall, opioid-exposed children’s scores were character- the fixed linear main effect of age at assessment was not ized by higher total difficulties and greater inter- and intra- significant. Child-specific random variability according individual variability from baseline (2 years) to 9 years of age. to age was significant with an estimated random-slope standard deviation of CE = 2.21 [1.63, 2.98]. The log- Discussion likelihood ratio chi-square test supported the use of the mixed effects Model 1 versus a simple linear model This is the first prospective analysis of the emotional and (χ (3) = 71.89, p < 0.001). behavioral development of an unselected regional sample To test for potential nonlinear effects of age on trajectories of opioid-exposed children from birth to 9 years, relative to over time, a quadratic term (age*age) was included in Model their same age typically developing peers. Using a hypoth- 2. As a result, the fixed linear (CE = -6.57 [-9.13, -4.01]) and esis-driven, multi-level analysis approach, we confirmed all quadratic effects of age (CE = 1.74 [1.11, 2.37]) were both three hypotheses we had formed a-priori. Specifically, com- significant with improved fit over Model 1, indicated by a pared with their non-exposed peers, (1) children who were reduced log-likelihood value (Table 3), and a significant log- prenatally exposed to opioids exhibited significantly higher likelihood ratio chi-square test (assumption Model 1 nested emotional and behavioral difficulties at 2, 4.5, and 9 years in Model 2: χ (1) = 27.42, p < 0.001). of age. Growth curve modeling further revealed that, (2) In Model 3, an interaction term (group*age) was introduced relative to their same age peers, opioid-exposed children’s to test whether behavioral trajectories differed between groups. difficulties trajectories worsened over time. Finally, results Again, model fit improved significantly (assumption Model 2 showed that (3) the index scores of prenatal risk, social risk, nested in Model 3: χ (1) = 23.48, p < 0.001) with the addition and postnatal rearing environment risk we had formed, as of this new parameter (CE = 2.40 [1.46, 3.34]). The significant well as changes in primary caregivers, independently and group x age interaction in Model 3 implies that opioid-exposed additively explained individual variations in children’s children’s total difficulties trajectories differed from non-exposed behavioral trajectories (see Fig. 4). children, with total difficulties scores worsening with age, and In addition to SDQ total difficulties scores, we also exam- thus confirming Hypothesis 2. ined the domain-specific nature of opioid-exposed children’s To test Hypothesis 3, Model 4 then included fixed emotional and behavioral difficulties in comparison to their effects of child sex, prenatal biological and social risk, same-age peers. An assessment of the existing literature as well as the postnatal rearing environment composite shows that prenatal opioid-exposure is consistently linked scores. While child sex did not predict total difficulties tra- with externalizing behavior problems, and in particular jectories, the previous effects of age and group remained behavioral dysregulation and attentional difficulties (Azuine significant and all three risk scores had additionally et  al., 2019; Levine & Woodward, 2017; Nygaard et al., significant effects. Thus, Hypothesis 3 was confirmed: 2016; Ornoy, 2003; Slinning, 2004; Sundelin Wahlsten & children with higher prenatal biological risk (CE = 2.03 Sarman, 2013). This is confirmed by our current results, [0.70, 3.35]), higher social background risk (CE = 1.18 with the rate of hyperactivity/inattention showing substantial [0.58, 1.79]), and higher postnatal rearing environment mean differences between groups among the SDQ subscales risk (CE = 1.56 [0.55, 2.57]) had higher total difficulties at each assessment. Similar increases in conduct problems scores. Again, model fit improved significantly (assump- with child age particularly from early childhood, were also tion Model 3 nested in Model 4: χ2(4) = 47.55, p < 0.001). observed amongst opioid-exposed children. Collectively, In Model 5, a fixed ee ff ct of having had any primary caregiver these findings suggest that these children are likely to be at changes on total difficulties trajectories was additionally added very high risk not just of more severe externalizing behavior (CE = 1.91 [0.33, 3.48]), and this further improved overall problems in childhood, but also a poorer prognosis longer model fit (assumption Model 4 nested in Model 5: χ 2(1) = 5.54, term. Finally, rates of peer (i.e., social) problems were also p < 0.019). Importantly, the group and group by age differences consistently elevated, and emotional problems of opioid- remained significant, even after accounting for related risk exposed children seemed to exacerbate over the course of factors, which also remained significant. childhood. Given the high rates of parental psychopathology 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 453 Fig. 4 Predicted (Model 5) Total Difficulties Scores Trajectories of Opioid-Exposed Versus Non-Exposed Children from 2 to 9 Years and family instability in this high-risk group, a question for with their caregiving quality, thereby directly and indirectly future study may be to examine whether early onset symp- affecting infants’ stress regulation and behavior (Conradt toms reflect attachment or social relationship related issues et  al., 2018). The second potential mechanism could be in early childhood. Further, assessments using standardized that prenatal opioid exposure alters the development of diagnostic measures of mental health disorders will also be dopaminergic reward-related circuits, which may in turn important. lead to dysregulated, hyperactive behavior (Sithisarn et al., According to the Developmental Origins of Health and 2017). What’s more, DOHaD studies have documented a Disease (DOHaD) model, unfavorable conditions in utero strong relationship between maternal depression and stress (e.g., opioid exposure, chronic stress) affect behavioral during pregnancy and children’s subsequent behavior functioning and health later in life via prenatal programing regulation trajectories, even after controlling for confounders (Barker, 2007). This includes teratogenic effects on the such as postnatal maternal depression (Henrichs & Van den developing central nervous system (Woodward et  al., Bergh, 2015). They also suggest that this relationship may 2018), potentially increasing mental health vulnerability. be continuously modulated by environmental experiences DOHaD also proposes that long-term risk is initially (Henrichs & Van den Bergh, 2015). Our novel findings of induced through adaptive responses (e.g., changes in the independent, additive contributions of prenatal risk, metabolism or tissue sensitivity to hormones) that the social risk, postnatal caregiving quality, and caregiver fetal organism makes to maternal cues about her health changes to predicting children’s emotional and behavioral or physical state, leading to altered organ development. problem trajectories add important information about Two mechanisms related to prenatal programming may the potential mechanisms placing infants born to opioid explain the behavior regulation difficulties seen in opioid- dependent mothers at increased psychopathological risk. For exposed children: The first suggests that opioids may alter instance, the non-exposed children’s trajectories displayed the hypothalamic–pituitary–adrenal (HPA) axis (Sithisarn in Fig. 4 show an expected small normative improvement et al., 2008, 2017; Slamberova, Riley, & Vathy, 2005; Taylor, in self-regulatory abilities from 2 to 4.5  years, but this Soong, Wu, Yee, & Szeto, 1997), which has been linked, pattern is less distinct among the opioid-exposed children. for instance, with ADHD (Ma, Chen, Chen, Liu, & Wang, This suggests that some risk factors may be at play that 2011). In addition, prenatal and postnatal stressors may dynamically alter the normative course of development have additive or interactive programming influences on fetal over time. Importantly, our final growth curve model HPA axis functioning, on top of prenatal opioid exposure showed that the additive effects of prenatal and social risks (Lester & Padbury, 2009). For instance, opioid-using remained significant even when effects of the postnatal mothers may have a higher risk for depressive symptoms rearing environment and primary caregiver changes were during and after pregnancy, and these could be associated added to the model. Thus, the adverse developmental 1 3 454 Research on Child and Adolescent Psychopathology (2021) 49:443–457 effects documented here cannot exclusively be attributed more process-oriented, longitudinal, and observation-based to the different postnatal environments that children grew assessment of reciprocal caregiver-child interactions would up in, but prenatal and social mechanisms related to help uncover the complex effects of social relationships on prenatal programming remain at play, even at later ages, trajectories in future studies. Relatedly, many predictors over and above the effects of in-utero opioid exposure. and the dependent variables were based on caregivers’ These results offer critically important information for reports, which can be biased (De Los Reyes & Kazdin, designing prevention and intervention services for current 2005; Van Roy, Groholt, Heyerdahl, & Clench-Aas, 2010). and future generations of opioid-using mothers and their What’s more, we were not able to include information offspring. Development is a complex multidirectional about other critically important domains that could explain process that can be described as a constant feedback loop mechanistic effects, such as (epi)genetics or stress response between biological and environmental factors, starting differences. The single stochastic imputation performed long before birth. Our findings suggest that intervention to maximize the utility of available data does not fully approaches that exclusively aim at modifying variables account for the variability structure of imputed and in children’s postnatal environments may not be able to missing values, and hence may have introduced downward effectively facilitate resilience for this at-risk group. Instead, bias in the estimates of standard errors of model estimates integrated biopsychosocial approaches that start during or (Carlin, Li, Greenwood, & Coffey, 2003; Rubin, 1987). before pregnancy and target opioid-using women’s mental In addition, statistical power to detect interaction effects and physical health may be warranted in order to increase between the different biopsychosocial risk scores was life chances for the next generation. limited. Future studies with larger sample sizes will be Strengths and Limitations. The findings presented here important in helping to replicate our study findings and to are from a prospective regional sample of opioid-exposed further evaluate whether the effects of pre- and postnatal children and an equally-sized group of non-exposed age- factors on child behavior are additive, interactive or both. matched controls. Compared to other longitudinal cohort A more in-depth examination of specific associations studies, our sample is characterized by very high sample between some of the risk exposures we have identified retention over time and age, a marker of excellent study and child outcomes may also be helpful, as this was quality. Alongside our opioid exposed group, measures beyond the scope of this integrated longitudinal analysis were also collected on a randomly identified, regionally of the contributions of a wide range of potential pre- and representative comparison group of non-exposed infants. This postnatal explanatory factors. helped avoid problems associated with the Flynn effect and Conclusion. Children who were exposed to opioids the use of normative cut-off points at very young ages, such in-utero remain at risk for emotional and behavioral as age 2. It also minimizes the likelihood that unintended and difficulties throughout the course of their childhood. In potentially unknown biases are introduced into the analysis. addition to the initial effects of opioids, a complex set of Assessments were rigorous and used internationally accepted, prenatal biological and postnatal environmental factors standardized measures at multiple times over the course of additively and dynamically contribute to substantial infancy and childhood. The rationale was hypotheses-driven variations in children’s individual developmental trajectories. and a sophisticated analysis approach helped uncover the Future studies and evidence-based efforts to mitigate the complex mechanisms underlying developmental outcomes long-term consequences of opioid use in pregnancy need to after in-utero opioid exposure. take children’s diverse biopsychosocial risks into account. However, the study also had some limitations. Our opioid- Supplementary Information The online version contains supplemen- exposed sample only included infants of mothers who were in tary material available at https: //doi.org/10.1007/s10802 -020-00766- w. MMT treatment. Untreated opioid dependence is extremely rare in New Zealand, and these women were prioritized for Acknowledgements The authors are grateful to Marie Goulden and substitution treatment which is free and accompanied by wrap- Carole Spencer for assistance with sample recruitment/retention and data collection. Most importantly, we would like to thank the families around psychiatric and obstetric care. Thus, results from these who kindly gave their time and support to this study. dyads may underestimate potential adverse effects of substance use on the child, and will have limited generalizability to Funding This study was funded by Cure Kids New Zealand (# 9518), infants of mothers who do not seek or receive substitution the New Zealand Lottery Grants Board (AP91256), Wayne Francis treatment when pregnant or when that care is unavailable. Trust, and the Health Research Council of New Zealand (#13/584). One of our aims was to assess different pre- and postnatal biopsychosocial risks and their contributions to Compliance with Ethical Standards opioid-exposed children’s development. As in any study, Conflict of Interest The authors declare that they have no conflicts of our analyses were limited by the variables that had been interest. assessed, and their respective distributions. For instance, a 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 455 Ethical Approval The Upper South B Regional Ethics Committee, women. Journal of Affective Disorders, 39(3), 185–189. https :// Canterbury, New Zealand (Ref: URB/07/10/042) approved the study doi.org/10.1016/0165-0327(96)00008 -0. protocol and the study was performed in accordance with the ethical Cox, J. L., Holden, J. M., & Sagovsky, R. (1987). Detection of postna- standards as laid down in the 1964 Declaration of Helsinki and its later tal nepression: Development of the 10-item Edinburgh Postnatal amendments or comparable ethical standards. Depression Scale. British Journal of Psychiatry, 150(6), 782–786. https ://doi.org/10.1192/bjp.150.6.782. Consent to Participate Written informed consent was obtained from Crea, T. M., Barth, R. P., Guo, S., & Brooks, D. (2008). Behavio- biological mothers or primary caregivers at each assessment point, with ral outcomes for substance-exposed adopted children: fourteen children also providing verbal assent at age 9 years. years postadoption. American Journal of Orthopsychiatry, 78(1), 11–19. https ://doi.org/10.1037/0002-9432.78.1.11. Davie-Gray, A., Moor, S., Spencer, C., & Woodward, L. J. (2013). Open Access This article is licensed under a Creative Commons Attri- Psychosocial characteristics and poly-drug use of pregnant bution 4.0 International License, which permits use, sharing, adapta- women enrolled in methadone maintenance treatment. Neuro- tion, distribution and reproduction in any medium or format, as long toxicology and Teratology, 38, 46–52. https ://doi.org/10.1016/j. as you give appropriate credit to the original author(s) and the source, ntt.2013.04.006. provide a link to the Creative Commons licence, and indicate if changes De Los Reyes, A., & Kazdin, A. E. (2005). Informant discrepan- were made. The images or other third party material in this article are cies in the assessment of childhood psychopathology: A critical included in the article’s Creative Commons licence, unless indicated review, theoretical framework, and recommendations for fur- otherwise in a credit line to the material. If material is not included in ther study. Psychological Bulletin, 131(4), 483–509. https ://doi. the article’s Creative Commons licence and your intended use is not org/10.1037/0033-2909.131.4.483. permitted by statutory regulation or exceeds the permitted use, you will Evans, G. W., Li, D., & Whipple, S. S. (2013). Cumulative risk and need to obtain permission directly from the copyright holder. To view a child development. Psychological Bulletin, 139(6), 1342–1396. copy of this licence, visit http://creativ ecommons .or g/licenses/b y/4.0/. https ://doi.org/10.1037/a0031 808. Fergusson, D. M., Horwood, J., & Ridder, E. M. (2005). Show me the child at seven: the consequences of conduct problems in child- hood for psychosocial functioning in adulthood. Journal of Child Psychology and Psychiatry, 46(8), 837–849. https ://doi.org/10.1 References 111/j.1469-7610.2004.00387 .x. Fill, M. A., Miller, A. M., Wilkinson, R. H., Warren, M. D., Dunn, J. Azuine, R. E., Ji, Y., Chang, H.-Y., Kim, Y., Ji, H., DiBari, J., & Wang, X. R., Schaffner, W., & Jones, T. F. (2018). Educational disabilities (2019). Prenatal risk factors and perinatal and postnatal outcomes among children born with neonatal abstinence syndrome. Pediat- associated with maternal opioid exposure in an urban, low-income, rics, 142(3). https ://doi.org/10.1542/peds.2018-0562 multiethnic US population. JAMA Network Open, 2(6), e196405– Goodman, R. (1997). The Strengths and Difficulties Question- e196405. https ://doi.org/10.1001/jaman etwor kopen .2019.6405. naire: a research note. Journal of Child Psychology Psychia- Barker, D. J. P. (2007). The origins of the developmental origins theory. try, 38(5), 581–586. Retrieved from http://onlinelibr ar y.wiley. Journal of Internal Medicine, 261(5), 412–417. https ://doi.org/ com/s t or e /10.1111/j.1469-7610.1997.tb015 45.x/ass e t / 10.1111/j.1365-2796.2007.01809 .x. j . 1 4 6 9 - 7 6 1 0 . 1 9 9 7 . t b 0 1 5 4 5 . x . p d f ? v = 1 & t = j d 8 0 k v 0 3 & s = Bateman, B. T., Hernandez-Diaz, S., Rathmell, J. P., Seeger, J. D., dcf7e e502e 577b1 3aa85 02a07 3fe05 e1295 dea8c Doherty, M., Fischer, M. A., & Huybrechts, K. F. (2014). Patterns Goodman, R. (2001). Psychometric properties of the Strengths and Dif- of opioid utilization in pregnancy in a large cohort of commercial ficulties Questionnaire (SDQ). Journal of the American Academy insurance beneficiaries in the United States. Anesthesiology, 120(5), of Child and Adolescent Psychiatry, 40, 1337–1345. 1216–1224. https ://doi.org/10.1097/aln.00000 00000 00017 2. Hatzis, D., Dawe, S., Harnett, P., & Barlow, J. (2017). Quality of car- Bendersky, M., & Lewis, M. (1994). Environmental risk, biological egiving in mothers with illicit substance use: A systematic review risk, and developmental outcome. Developmental Psychology, and meta-analysis. Substance Abuse: Research and Treatment, 30(4), 484–494. 11, https ://doi.org/10.1177/11782 21817 69403 8. Bolton, D., & Gillett, G. (2019). Biopsychosocial conditions of health Henrichs, J., & Van den Bergh, B. R. H. (2015). Perinatal developmen- and disease. The Biopsychosocial Model of Health and Disease: tal origins of self-regulation. In G. H. E. Gendolla, M. Tops, & S. New Philosophical and Scientific Developments (pp. 109–145). L. Koole (Eds.), Handbook of Biobehavioral Approaches to Self- Cham: Springer International Publishing. Regulation (pp. 349–370). New York, NY: Springer, New York. Caldwell, B. M., & Bradley, R. H. (1984). Home observation for meas- Jones, H. E., O’Grady, K. E., Malfi, D., & Tuten, M. (2008). Metha- urement of the environment. AR: Retrieved from Little Rock. done maintenance vs. methadone taper during pregnancy: Mater- Carlin, J. B., Li, N., Greenwood, P., & Coffey, C. (2003). Tools for nal and neonatal outcomes. The American Journal on Addictions, Analyzing Multiple Imputed Datasets. The Stata Journal, 3(3), 17(5), 372–386. https ://doi.org/10.1080/10550 49080 22662 76 226–244. https ://doi.org/10.1177/15368 67x03 00300 302. Jusiene, R., Breidokiene, R., & Pakalniskiene, V. (2015). Develop- CDC. (2017). Vital signs: prescription painkiller overdoses: a growing mental trajectories of mother reported regulatory problems from epidemic, especially among women. Retrieved 2/3/2018 toddlerhood to preschool age. Infant Behavior and Development, Conradt, E., Crowell, S. E., & Lester, B. M. (2018). Early life stress and 40, 84–94. https ://doi.org/10.1016/j.infbe h.2015.04.003. environmental influences on the neurodevelopment of children Kaltenbach, K., O’Grady, K. E., Heil, S. H., Salisbury, A. L., Coyle, with prenatal opioid exposure. Neurobiology of Stress, 9, 48–54. M. G., Fischer, G., & Jones, H. E. (2018). Prenatal exposure to https ://doi.org/10.1016/j.ynstr .2018.08.005. methadone or buprenorphine: Early childhood developmental out- Conradt, E., Flannery, T., Aschner, J. L., Annett, R. D., Croen, L. A., comes. Drug and Alcohol Dependence, 185, 40–49. https ://doi. Duarte, C. S., Lester, B. M. (2019). Prenatal Opioid Exposure: org/10.1016/j.druga lcdep .2017.11.030. Neurodevelopmental Consequences and Future Research Priori- Kelty, E., & Preen, D. B. (2019). Risk Factors Associated with the ties. Pediatrics, 144(3). https ://doi.org/10.1542/peds.2019-0128 Occurrence of Neonatal Opioid Withdrawal Syndrome: A Cox, J. L., Chapman, G., Murray, D., & Jones, P. (1996). Validation of Review. CNS Drugs, 33(11), 1113–1120. https://doi.or g/10.1007/ the Edinburgh postnatal depression scale (EPDS) in non-postnatal s4026 3-019-00681 -9. 1 3 456 Research on Child and Adolescent Psychopathology (2021) 49:443–457 Kessler, R. C., Abelson, J., Demler, O., Escobar, J. I., Gibbon, M., Ornoy, A. (2003). The impact of intrauterine exposure versus postnatal Guyer, M. E., & Zheng, H. (2004). Clinical calibration of DSM- environment in neurodevelopmental toxicity: long-term neurobe- IV diagnoses in the World Mental Health (WMH) version of the havioral studies in children at risk for developmental disorders. World Health Organization (WHO) Composite International Toxicology Letters, 140–141, 171–181. Diagnostic Interview (WMHCIDI). Int J Methods Psychiatr Res, Patrick, S. W., Davis, M. M., Lehmann, C. U., & Cooper, W. O. (2015). 13(2), 122–139. Increasing incidence and geographic distribution of neonatal Konijnenberg, C., Lund, I. O., & Melinder, A. (2015). Behavioural abstinence syndrome: United States 2009 to 2012. Journal of outcomes of four-year-old children prenatally exposed to Perinatology, 35(8), 650–655. https://doi.or g/10.1038/jp.2015.36. methadone or buprenorphine: a test of three risk models. Early Rubin, D. B. (1987). Multiple Imputation for Nonresponse in Surveys: Child Development and Care, 185(10), 1641–1657. https ://doi. John Wiley & Sons, Inc. org/10.1080/03004 430.2015.10165 06. Sameroff, A. J., & MacKenzie, M. J. (2003). Research strategies for Konijnenberg, C., Sar, fi M., & Melinder, A. (2016). Mother-child inter - capturing transactional models of development: The limits of the action and cognitive development in children prenatally exposed possible. Development and Psychopathology, 15(3), 613–640. to methadone or buprenorphine. Early Human Development, 101, https ://doi.org/10.1017/s0954 57940 30003 12. 91–97. https ://doi.org/10.1016/j.earlh umdev .2016.08.013. Sanlorenzo, L. A., Stark, A. R., & Patrick, S. W. (2018). Neonatal absti- Koumoula, A. (2012). The course of attention deficit hyperactivity nence syndrome: an update. Current opinion in pediatrics, 30(2), disorder (ADHD) over the life span. Psychiatriki, 23(Suppl 1), 182–186. https ://doi.org/10.1097/MOP.00000 00000 00058 9. 49–59. Sarfi, M., Smith, L., Waal, H., & Sundet, J. M. (2011). Risks and Lee, S. J., Pritchard, V. E., Austin, N. C., Henderson, J. M. T., & realities: Dyadic interaction between 6-month-old infants and Woodward, L. J. (2020). Health and neurodevelopment of chil- their mothers in opioid maintenance treatment. Infant Behavior dren born to opioid-dependent mothers at school entry. Journal and Development, 34(4), 578–589. https ://doi.or g/10.1016/j. of Developmental & Behavioral Pediatrics, 41(1), 48–57. https:// infbe h.2011.06.006. doi.org/10.1097/dbp.00000 00000 00071 1. Sasser, T. R., Kalvin, C. B., & Bierman, K. L. (2016). Developmental Lee, S. J., Woodward, L. J., & Henderson, J. M. T. (2019). Educa- trajectories of clinically significant attention-deficit/hyperactivity tional achievement at age 9.5 years of children born to mothers disorder (ADHD) symptoms from grade 3 through 12 in a high- maintained on methadone during pregnancy. PLoS One, 14(10), risk sample: Predictors and outcomes. Journal of Abnormal Psy- e0223685. https ://doi.org/10.1371/journ al.pone.02236 85 chology, 125(2), 207–219. https ://doi.org/10.1037/abn00 00112 . Lenze, E. J., & Wetherell, J. L. (2011). A lifespan view of anxiety Siqveland, T. S., Haabrekke, K., Wentzel-Larsen, T., & Moe, V. disorders. Dialogues Clin Neurosci, 13(4), 381–399. (2014). Patterns of mother-infant interaction from 3 to 12 months Lester, B. M., & Padbury, J. F. (2009). Third pathophysiology of pre- among dyads with substance abuse and psychiatric problems. natal cocaine exposure. Developmental Neuroscience, 31(1–2), Infant Behavior and Development, 37(4), 772–786. https ://doi. 23–35. https ://doi.org/10.1159/00020 7491.org/10.1016/j.infbe h.2014.09.003. Levine, T. A., & Woodward, L. J. (2017). Early inhibitory control Sithisarn, T., Bada, H. S., Dai, H., Reinhardt, C. R., Randall, D. C., and working memory abilities of children prenatally exposed to & Legan, S. J. (2008). Effects of perinatal oxycodone exposure methadone. Early Hum Dev, 116, 68–75. https://doi.or g/10.1016/j. on the response to CRH in late adolescent rats. Neurotoxicol- earlh umdev .2017.11.010. ogy and Teratology, 30(2), 118–124. https ://doi.org/10.1016/j. Levine, T. A., & Woodward, L. J. (2018). Early inhibitory control ntt.2007.12.010. and working memory abilities of children prenatally exposed to Sithisarn, T., Legan, S. J., Westgate, P. M., Wilson, M., Wellmann, methadone. Early Hum Dev, 116, 68–75. https://doi.or g/10.1016/j. K., Bada, H. S., & Barron, S. (2017). The Effects of Perinatal earlh umdev .2017.11.010. Oxycodone Exposure on Behavioral Outcome in a Rodent Model. Li, S.-C. (2003). Biocultural orchestration of developmental plasticity Front Pediatr, 5, 180. https ://doi.org/10.3389/fped.2017.00180 . across levels: The interplay of biology and culture in shaping the Slamberova, R., Riley, M. A., & Vathy, I. (2005). Cross-generational mind and behavior across the life span. Psychological Bulletin, effect of prenatal morphine exposure on neurobehavioral develop- 129(2), 171–194. https ://doi.org/10.1037/0033-2909.129.2.171. ment of rat pups. Physiological Research, 54(6), 655–660. Ma, L., Chen, Y. H., Chen, H., Liu, Y. Y., & Wang, Y. X. (2011). The Slinning, K. (2004). Foster placed children prenatally exposed to poly- function of hypothalamus-pituitary-adrenal axis in children with substances–attention-related problems at ages 2 and 4 1/2. Euro- ADHD. Brain Research, 1368, 159–162. https://doi.or g/10.1016/j. pean Child and Adolescent Psychiatry, 13(1), 19–27. https ://doi. brain res.2010.10.045.org/10.1007/s0078 7-004-0350-x. Mangin, K. S., Horwood, L. J., & Woodward, L. J. (2017). Cognitive Spencer, T. J., Biederman, J., & Mick, E. (2007). Attention-deficit/ development trajectories of very preterm and typically develop- hyperactivity disorder: diagnosis, lifespan, comorbidities, and ing children. Child Development, 88(1), 282–298. https ://doi. neurobiology. Ambulatory Pediatrics, 7(1 Suppl), 73–81. https :// org/10.1111/cdev.12585 . doi.org/10.1016/j.ambp.2006.07.006. Mischel, W., Ayduk, O., Berman, M. G., Casey, B. J., Gotlib, I. H., Statistics New Zealand. (2006). Census Regional Summary. New Jonides, J., & Shoda, Y. (2011). “Willpower” over the life span: Zealand. decomposing self-regulation. Social Cognitive and Affective Neu- Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. roscience, 6(2), 252–256. https ://doi.org/10.1093/scan/nsq08 1. (1996). The Revised Conflict Tactics Scales (CTS2): Development Nygaard, E., Slinning, K., Moe, V., & Walhovd, K. B. (2016). Behav- and Preliminary Psychometric Data. Journal of Family Issues, ior and Attention Problems in Eight-Year-Old Children with 17(3), 283–316. https ://doi.org/10.1177/01925 13960 17003 001. Prenatal Opiate and Poly-Substance Exposure: A Longitudinal Sundelin Wahlsten, V., & Sarman, I. (2013). Neurobehavioural Study. PLoS ONE, 11(6), e0158054. https ://doi.or g/10.1371/ development of preschool-age children born to addicted moth- journ al.pone.01580 54. ers given opiate maintenance treatment with buprenorphine dur- Oei, J. L., Melhuish, E., Uebel, H., Azzam, N., Breen, C., Burns, L., ing pregnancy. Acta Paediatrica, 102(5), 544–549. https ://doi. Wright, I. M. (2017). Neonatal abstinence syndrome and high org/10.1111/apa.12210. school performance. Pediatrics, 139(2). https ://doi.org/10.1542/ Taylor, C. C., Soong, Y. I., Wu, D., Yee, J. S., & Szeto, H. H. (1997). peds.2016-2651 Morphine stimulates adrenocorticotropin and cortisol release in 1 3 Research on Child and Adolescent Psychopathology (2021) 49:443–457 457 the late-term ovine fetus. Pediatric Research, 41(3), 411–415. Winkelman, T. N. A., Villapiano, N., Kozhimannil, K. B., Davis, M. https ://doi.org/10.1203/00006 450-19970 3000-00018 . M., & Patrick, S. W. (2018). Incidence and costs of neonatal absti- Totsika, V., & Sylva, K. (2004). The Home Observation for Meas- nence syndrome among infants with medicaid: 2004–2014. Pedi- urement of the Environment Revisited. Child and Adolescent atrics, 141(4). https ://doi.org/10.1542/peds.2017-3520 Mental Health, 9(1), 25–35. h t t p s : / / d o i . o r g / 1 0 . 1 0 4 6 / j . 1 4 7 5- Woodward, L. J., McPherson, C. C., & Volpe, J. J. (2018). Chapter 38 357X.2003.00073 .x. - Passive addiction and teratogenic effects. In J. J. Volpe, T. E. van Buuren, S., Groothuis-Oudshoorn, K., Vink, G., Schouten, R., Inder, B. T. Darras, L. S. de Vries, A. J. du Plessis, J. J. Neil, & Robitzsch, A., Doove, L., Gray, B. (2020). Package ‘mice’. J. M. Perlman (Eds.), Volpe’s Neurology of the Newborn (Sixth Van Roy, B., Groholt, B., Heyerdahl, S., & Clench-Aas, J. (2010). Edition) (pp. 1149–1189.e1120): Elsevier. Understanding discrepancies in parent-child reporting of emo- tional and behavioural problems: Effects of relational and socio- Publisher’s Note Springer Nature remains neutral with regard to demographic factors. Bmc Psychiatry, 10(1), 1–12. https ://doi. jurisdictional claims in published maps and institutional affiliations. org/10.1186/1471-244x-10-56. WHO. (1993). Composite International Diagnostic Interview (CIDI). Geneva, Switzerland: World Health Organization. 1 3

Journal

Journal of Abnormal Child PsychologySpringer Journals

Published: Jan 12, 2021

There are no references for this article.