Family Environment and Attentiondeficit/Hyperactivity

bs_bs_banner Child: care, health and development

Original Article doi:10.1111/cch.12112

Family environment and attention-deﬁcit/ hyperactivity disorder in adopted children: associations with family cohesion and adaptability

T. M. Crea,* K. Chan† and R. P. Barth‡ *Graduate School of Social Work, Boston College, Chestnut Hill, MA, USA †School of Social Welfare, SUNY Albany, Albany, NY, USA, and ‡School of Social Work, University of Maryland, Baltimore, MD, USA

Accepted for publication 7 September 2013

Abstract Background Positive family environments are crucial in promoting children’s emotional and behavioural well-being, and may also buffer development of attention-deficit/hyperactivity disorder (ADHD). ADHD is highly heritable, but psychosocial factors in the family environment, particularly family cohesion and communication, may mediate genetic predispositions. The purpose of the current study is to examine the mediating influence of the adoptive family environment between pre-adoptive risk factors and youths’ ADHD symptomatology at 14 years post adoption. Methods The data used in this study were obtained from the fourth wave of the California Long-Range Adoption Study (CLAS) (n = 449). Using structural equation modelling (SEM), family sense of coherence and family adaptability were tested as possible mediators between environmental and biological predictors and ADHD symptomatology. Predictors included birthweight, gender, age at adoption, adoption from foster care, transracial adoption status, Keywords ADHD, adoption, ethnicity and having a previous diagnosis of ADHD. attention-deficit/ Results Results show that, while adoption from foster care is negatively associated with family hyperactivity disorder, family environment, SEM functioning, higher family cohesion and adaptability mediate this influence on children’s ADHD symptomatology. Older age of adoption directly predicts greater ADHD symptoms with no Correspondence: mediating influence of the family environment. Thomas M. Crea, Conclusions The mediating influence of the family environment between children’s risk factors Graduate School of Social Work, Boston College, 140 and ADHD symptoms suggests that family intervention strategies may be helpful in improving Commonwealth Ave., adopted children’s outcomes. Once children are adopted, targeting family communication patterns McGuinn 302, Chestnut Hill, MA 02467, USA and dynamics may be an additional part of developing an evidence-based, post-adoption services E-mail: [email protected] toolkit.

ing family conﬂict (Jaycox & Repetti 1993), family cohesion Introduction (Moos & Moos 1994) and family coherence and adaptabil- Positive family environments are crucial in promoting chi- ity (Antonovsky & Sourani 1988), and generally refers to ldren’s emotional and behavioural well-being, and may also families where members are mutually supportive within a buffer development of attention-deﬁcit/hyperactivity disorder predictable and nurturing environment. In regards to ADHD, (ADHD) (Mulligan et al. 2011). The term ‘family environ- research has shown that, while ADHD has a strong genetic ment’ encompasses multiple, closely related concepts, includ- component (Burt 2009), psychosocial adversity in the family

© 2013 John Wiley & Sons Ltd 853 854 T.M. Crea et al. environment may trigger an underlying predisposition disentangle whether ADHD is the cause or result of factors (Biederman 2005). within the family environment (Howe 2010), clearly the envi- ADHD is characterized by inattention (daydreaming, diffi- ronment plays an important role in the course of developing culty focusing on a single task, distractibility) and hyperactivity symptomatology. (excessive talking, fidgeting, restlessness; Biederman 2005). It is often difficult to determine whether these symptoms are the Family environmental contributors to ADHD causes or consequences of a dysfunctional family environment Several studies have shown higher rates of family dysfunction in (Howe 2010), as biological traits – such as shared ADHD families with children diagnosed with ADHD, particularly between parents and children – contribute to environmental related to problems in communication, relationships and conditions that reinforce children’s symptoms (Cadoret et al. problem solving (Cunningham & Boyle 2002), higher family 1995; Leve et al. 2010b). Given the absence of genetic links in an conflict and lower levels of organization (Foley 2010; Mulligan adoptive family, the influence of a chaotic family environment et al. 2011). Some of these problems may be related to parental related to parental psychopathology (Biederman et al. 2002; psychopathology, or poor parenting practices (Cunningham & Schroeder & Kelly 2009), has, in theory, been isolated (Cadoret Boyle 2002; Deault 2010) possibly related to parental ADHD et al. 1995; Leve et al. 2010b). Yet, while data are scarce, esti- symptoms (Daley 2006). mates of ADHD prevalence in adopted children may be much Among adoptive families, Cadoret and colleagues (1995) higher than the general population: 21.8% (Simmel et al. 2001) found that children with an antisocial birth parent – who also compared with 3–8% of school-aged children (Foley 2010), and experience negative adoptive family interactions – are at greater 5.29% worldwide (Polanczyk et al. 2007). The purpose of the risk for aggressive behaviours. Research by Leve and colleagues current study is to examine the relationship of the family envi- (2010b) showed that adoptive mothers’ affective states moder- ronment, above and beyond pre-adoptive risk factors, with ated the relationship between birth parents’ and adopted chi- ADHD symptomatology among 449 adopted youths. ldren’s externalizing problems. Similarly, Tully and colleagues (2008) found that maternal depression significantly predicted The influence of the family environment children’s depression and externalizing behaviour problems for both groups, regardless of adoption status. These studies seem Measures of family environment are strongly associated with to confirm the importance of the family environment on chi- children’s behaviour problems, beyond biological and other ldren’s well-being above and beyond biological similarities psychosocial risk factors (Lucia & Breslau 2006). Higher family between parents and children, a finding particularly relevant for cohesion has been associated with lower internalizing and exter- this current study of adopted children. nalizing behaviours in children (Henderson et al. 2003), and lower rates of depression and anxiety among adolescents (Burt Research questions et al. 1988). In a study of children adopted from orphanages in the Soviet Union (McGuiness et al. 2005), only low birthweight This study uses two validated, related measures of the family and family cohesion significantly predicted variation in chi- environment – family adaptability and family sense of coher- ldren’s behavioural problems. ence (Antonovsky & Sourani 1988) – and is guided by two Studies of the family environment’s influence on ADHD have research questions: (1) To what extent does family cohesion shown similar patterns to other child well-being indicators mediate the relationship between key biological and environ- (Biederman 2005), but with a strong genetic component mental predictors, such as birthweight, age at adoption, foster (Lifford et al. 2008; Leve et al. 2010a). In a review of 20 twin care status, and ADHD symptomatology for adopted children? studies, Faraone and colleagues (2005) found that shared genes And (2) To what extent does family adaptability mediate the accounted for 76% of ADHD heritability. Some researchers relationship between these biological and environmental pre- posit that adverse environments may trigger the underlying dictors, and ADHD symptomatology? condition (Biederman 2005) such that a gene–environment Methods interaction exists (Leve et al. 2010b). Burt and colleagues (2003), for example, found that parent–child conflict is a risk Sample factor for several externalizing disorders in children, including ADHD, but that co-morbidity is likely related to psychopathol- The data used in this study were obtained from the California ogy within the family environment. Thus, while it is difficult to Long-Range Adoption Study (CLAS). Analysis for this study

Table 1. Respondent characteristics and scale Full sample (N = 449) Scale means means* FSOC FAS CPRS-R n or M (SD) % M (SD) or r M (SD) or r M (SD) or r Respondent Mother 362 80.6 135.7 (18.5) 55.3 (9.8) 22.5 (18.4) Father 77 17.1 133.6 (19.9) 54.8 (11.2) 20.0 (15.9) Marital statusa Married/cohabitating 387 86.2 136.2 (17.5) 55.5 (9.7) 21.6 (17.7) Divorced/separated 36 8.0 129.9 (23.0) 52.9 (11.1) 25.7 (19.0) Single/widowed 25 5.6 126.4 (25.4) 51.8 (10.0) 24.1 (21.2) Respondent ethnicity White 364 81.1 134.8 (18.8) 54.9 (10.2) 22.1 (18.1) Hispanic 22 4.9 136.6 (18.9) 57.2 (9.7) 19.8 (16.1) Other 15 3.3 133.2 (21.8) 54.3 (10.9) 28.1 (21.6) Education High school, vocational school or 143 31.8 135.1 (19.0) 55.9 (10.0) 21.0 (18.5) community college 4-year college or above 302 67.3 135.0 (18.4) 54.8 (10.0) 22.6 (17.8) Employment status Full-time 201 44.8 135.0 (19.8) 55.1 (10.5) 22.2 (17.6) Part-time 124 27.6 135.0 (17.2) 55.4 (9.5) 19.4 (17.5) Retired 42 9.4 132.0 (18.8) 53.9 (10.7) 23.0 (17.5) Not employed 73 16.3 137.8 (16.6) 55.3 (8.8) 25.6 (19.5) No. of children No. biologicalc 0.7 (1.1) – r =−0.042 r =−0.046 r = 0.105 No. adopted 1.7 (1.0) – r =−0.028 r = 0.013 r =−0.002 FSOC (α=0.90)b,c 135.1 (18.7) – – r = 0.815 r =−0.543 FAS (α=0.88)c 55.1 (10.0) – – – r =−0.511 CPRS-R (α=0.97) 22.1 (18.0) –

*Percentages may not total 100% because of missing data. aP < 0.05 for FSOC; bP < 0.05 for FAS; cP < 0.05 for CPRS-R. CPRS-R, Conner’s Parent Rating Scale-Revised; FAS, family adaptability scale; FSOC, family sense of coherence.

was conducted on the ﬁnal wave of these data collected for Most respondents were mothers (80.6%), married or children adopted in California between July 1988 and June cohabitating (86.2%), white (81.1%), had a 4-year college 1989. Respondents were adoptive parents instructed to com- degree or higher (67.3%), and nearly half were employed full plete the survey for only one adopted child. Data collection for time (44.8%; see Table 1). The mean number of biological chil- this study began in the summer of 2003 and was completed by dren in the family was 0.7 (SD = 1.1) and the mean number of 2004, approximately 14 years after the children were adopted. adopted children was 1.7 (SD = 1.0). For child characteristics Questionnaires were mailed to 956 adoptive parents for which (see Table 2), children were nearly evenly split between males researchers had contact information from the original study and females. On average, children were 3.4 years old (SD = 2.5) population (n = 1219), asking for information about various at the time of adoption, and 16.7 years old (SD = 2.5) at the time aspects of adoption, including child characteristics and out- of the study. comes for adopted children. Parents completed questionnaires for 469 children, representing 49.0% of the mailed question- Measures naires. Of these, two respondents had missing information on their adoptive child’s ADHD diagnosis, and an additional 18 Demographic variables had missing information on a current assessment of inattention and hyperactivity, captured by the Conner’s Parent Rating Respondent characteristics included respondent type (adoptive Scale-Revised (CPRS-R), the dependent variable of the study. mother/father); marital status (married/cohabitating; divorced/ These cases were removed for a ﬁnal study sample of n = 449. separated; single/widowed); respondent ethnicity (White/

Table 2. Child characteristics and scale Full sample (N = 449) Scale means means* FSOC FAS CPRS-R n or M (SD) % M (SD) or r M (SD) or r M (SD) or r Genderc Female 218 48.6 135.6 (18.5) 55.1 (10.0) 19.9 (17.6) Male 231 51.4 134.6 (18.9) 55.2 (10.1) 24.2 (18.2) Mean age (years) 16.7 (2.5) – – – – Mean age at Wave 1 3.4 (2.5) – – – – Age of child at Wave 1 (categorized)b,c 0–1.9 143 31.8 136.3 (18.7) 55.8 (9.8) 20.1 (18.1) 2–2.9 117 26.1 137.5 (16.8) 57.0 (8.7) 18.1 (14.9) 3–5.9 86 19.2 132.9 (20.00 52.7 (10.5) 27.9 (18.6) 6 or over 55 12.2 130.5 (19.6) 53.0 (10.7) 27.5 (21.0) ADHD diagnosisa,b,c Yes 149 33.2 126.1 (19.7) 50.3 (10.5) 37.7 (16.4) No 300 66.8 139.8 (16.3) 57.5 (8.9) 14.4 (13.1) Birthweight (ounces) 111.5 (21.4) – r =−0.001 r = 0.055 r = 0.002 Ethnicity White 247 55.0 134.0 (20.0) 54.4 (10.6) 22.6 (18.4) Latino/a 80 17.8 137.1 (17.1) 56.8 (9.3) 22.4 (15.9) Asian 38 8.5 138.9 (14.9) 57.7 (8.2) 15.7 (16.0) Black 28 6.2 127.7 (18.0) 52.1 (10.0) 25.0 (19.2) Other 5 1.1 134.4 (20.0) 54.0 (8.0) 15.6 (10.4) Transracial (vs. no) 131 29.2 135.7 (16.7) 55.9 (9.0) 20.8 (16.8) How adopteda,b,c Independently 220 49.0 136.8 (17.7) 56.3 (9.5) 19.5 (17.2) Public agency (foster care) 133 29.6 131.1 (20.2) 52.9 (10.7) 27.9 (19.3) Private agency 48 10.7 138.1 (17.1) 55.8 (8.6) 19.1 (15.6)

*Percentages may not total 100% because of missing data. aP < 0.05 for FSOC; bP < 0.05 for FAS; cP < 0.05 for CPRS-R. ADHD, attention-deﬁcit/hyperactivity disorder; CPRS-R, Conner’s Parent Rating Scale-Revised; FAS, family adaptability scale; FSOC, family sense of coherence.

Latino/Other); education (high school/vocational school/ explainable and predictable; that resources exist to meet community college/4-year college or above); employment demands; and that demands pose worthwhile challenges to the status (full-time/part-time/retired/not employed); and the family (Antonovsky & Sourani 1988). The FSOC has been used number of children in the home (for biological and adopted across a variety of settings and populations, including children’s children). Child characteristics included gender (male/female); health (Eriksson & Lindstrom 2006) and the psychosocial age (in years); having ever received a diagnosis of ADHD (yes/ adjustment of adopted children (Ji et al. 2010). Previous studies no, by parental report; no other information was available suggest that internal consistency for this scale is high regarding by whom this diagnosis was given, or when); (Antonovsky & Sourani 1988). Results on the reliability of this birthweight (ounces); ethnicity (White/Latino/Asian/Black/ scale suggest good internal consistency in the current sample Other); whether the adoption were transracial (yes/no); and (α=0.90). how the child was initially adopted (independently of an agency; through a private agency; through a public child welfare agency, i.e. foster care). Measures also included three standard- Family adaptability scale (FAS) ized instruments, described below. The FAS consists of 10 Likert scale items, each ranging from 1 to 7, and measures families’ satisfaction with the ﬁt between individual family members and the family, and between the family Family sense of coherence (FSOC) and the larger community. The internal consistency of the FAS The FSOC consists of 26 Likert scale items, each ranging from 1 is fairly high in previous studies (Antonovsky & Sourani 1988), to 7, and measures the extent to which the family is structured, and performs well in the current sample (α=0.88). Evidence

Child’s Child’s Child’s Gender Race Birthweight ADHD Diagnosis

Age at Adoption

FSOC

CPRS-R

FAS Adoption from Foster Care

Number of Number of Family College- Biological Adopted Marital Educated Children Children Status Parents

Figure 1. Initial theoretical model. ADHD, attention-deficit/hyperactivity disorder; CPRS-R, Conner’s Parent Rating Scale-Revised; FAS, family adaptability scale; FSOC, family sense of coherence. suggests that the constructs measured by the FSOC and FAS, though the CPRS-R has been used by clinicians to help with while conceptually distinct, are highly correlated. making a diagnosis of ADHD, for the purposes of this study, the diagnosis of ADHD was made temporally prior to collection of these data, and thus should be treated as a predictor. The first Conner’s Parent Rating Scale-Revised (CPRS-R) path model included the following predictors as exogenous The CPRS-R is designed to assess ADHD symptomatology in variables: family marital status; adoptive parents’ educational children (Conners et al. 1998). The CPRS-R consists of 27 attainment; child’s gender; child’s race (Other vs. White); Likert scale items, ranging from 0 to 3, with overall scores birthweight in ounces; the numbers of biological and adoptive ranging from 0 to 81. Raw scores of 15 and higher or standard- children in the family respectively; age at adoption; adoption ized T scores over 65 are indicative of a clinical ADHD diagnosis from foster care; and having a diagnosis of ADHD prior to the (Conners 1997). The mean raw score for the current sample is study. Family coherence (FSOC) and family adaptability (FAS) 22.1 (18.0) indicating, on average, high ADHD symptomatol- were specified as mediators in this analysis. Based on the results ogy. The CPRS-R has demonstrated high internal consistency of the first model (see Fig. 1), the researchers modified the with Cronbach’s alpha ranging from 0.88 to 0.96 (Conners et al. specification of the analysis by removing all non-significant 1998). In the current study, the CPRS-R demonstrated excellent paths (Letzter-Pouw & Werner 2012) and formulated compet- reliability (α=0.97). ing models to assess and compare their overall goodness of fit, meaningfulness for interpretation, and explanatory power. A final model is presented in Fig. 2. Analysis

Multiple imputation using SAS 9.2 was performed on items Results from each of the mediating measurements separately (FSOC, FAS), using responses to estimate imputations for missing The initial model, which included all predictors, with the FSOC values with individual items (Yuan 2000; Acock 2005). Even and the FAS as mediators, yielded poor goodness of ﬁt statistics

ADHD Diagnosis

-0.37** 0.48*** -0.37** FSOC 0.85*** -0.27**

Age at Adoption 0.11** CPRS-R 0.65***

0.84** -0.11** -0.12* FAS

-0.14* Adoption from Foster Care

Figure 2. Final path model with standardized estimates. *P < 0.05, **P < 0.01, ***P < 0.001. ADHD, attention-deﬁcit/hyperactivity disorder; CPRS-R, Conner’s Parent Rating Scale-Revised; FAS, family adaptability scale; FSOC, family sense of coherence.

[χ2 = 254.44, d.f. = 1, P < 0.00001, root mean square error of initial model and the final model. Both family coherence and approximation (RMSEA) = 0.78, 90% confidence interval for family adaptability were in turn significantly predicted by RMSEA = (<0.68; 0.88), P-value for test of Close Fit (RMSEA < ADHD and foster status, but not age at adoption. 0.0001) = 0.95, normed fit index (NFI) = 0.66, non-normed fit In the final structural model (see Fig. 2), measurement errors index (NNFI) =−23.17, confirmatory fit index (CFI) = 0.63, criti- from the FSOC and FAS were also allowed to correlate, based on cal n = 8.48, standardized root mean square residuals (RMR) = the high degree of shared variance. Family coherence and family 13.86, goodness of fit index (GFI) = 0.91, adjusted GFI index adaptability were specified as predictors of the CPRS-R. The (AGFI) =−5.96]. Results from bivariate analyses indicated that specified final model appears to have a good overall fit with the birthweight, ethnicity and transracial adoption status were not data [χ2 = 0.75, d.f. = 3, P = 0.86, RMSEA < 0.001, 90% confi- significantly associated with any standardized measures (FSOC, dence interval for RMSEA = (<0.001; 0.049), P-value for test FAS, CPRS-R) in this study. Sensitivity testing was conducted on of Close Fit (RMSEA < 0.05) = 0.95, NFI = 1.00, NNFI = 1.02, the original data as well as the imputed values for these variables, CFI = 1.00, critical n = 5118.17, standardized RMR = 0.0092, and the results indicated that these factors remained non- GFI = 1.00, AGFI = 0.99]. significant in predicting both the mediators (FAS and FSOC) or For our structural regression model (see Fig. 2), FSOC and the final dependent variable (CPRS-R). From the evidence of the family adaptability emerged as direct predictors which buffers analysis, birthweight, ethnicity and transracial adoption status the effect of hyperactivity as measured by the CPRS-R (FSOC, were dropped in the specification for the final model. direct effect =−0.27; FAS, direct effect =−0.11). The results also support the hypotheses that they are mediators to hyperactivity, from the effects of both a diagnosis of ADHD and FSOC and FAS adoption from foster care (see Table 3). The results support Results suggest that four of the predictors (family coherence, the hypotheses that family coherence and adaptability may family adaptability, diagnosis of ADHD, age at adoption) were buffer the deleterious impact of foster status and a diagnosis statistically significant in predicting hyperactivity as measured of ADHD on children’s hyperactivity. Factors such as by the CPRS-R when controlling for all other variables in the birthweight, transracial adoption status and ethnicity did not

Table 3. Standardized direct, indirect and total effects for endogenous colleagues (1995) who also found that adverse family environ- variables ments, and particularly low family cohesion, predicted higher Endogenous Predictor Direct Indirect Total rates of ADHD. The current study adds to this body of work by variable variables effects effects effects r2 demonstrating that the risk of ADHD symptomatology posed CPRS-R FSOC −0.27** NA −0.27 0.42 to children adopted from foster care may be buffered by strong FAS −0.11** NA −0.11 ADHD diagnosis 0.48*** 0.14 0.62 family cohesion and adaptability. Adoption from foster NA 0.05 0.05 In contrast, older age at the time of adoption emerged as a care direct risk factor for increased ADHD symptoms, and family Age at adoption 0.11** NA 0.11 environment did not appear to mediate this risk. This finding is FSOC ADHD diagnosis −0.37** NA −0.37 0.15 Foster status −0.12* NA −0.12 somewhat contrary to past research on adoptees which found FAS ADHD diagnosis −0.37** NA −0.37 0.16 that family cohesion – and birthweight, which did not emerge as − − Foster status 0.14* NA 0.14 a statistically significant effect in this study, directly or indirectly *P < 0.05, **P < 0.01, ***P < 0.001. – accounted for the greatest variation in children’s behaviours, ADHD, attention-deficit/hyperactivity disorder; CPRS-R, Conner’s Parent above and beyond the age of the child (McGuiness et al. 2005). Rating Scale-Revised; FAS, family adaptability scale; FSOC, family sense of coherence. Older adopted children are more generally at highest risk for emotional and behavioural problems (Berry & Barth 1990), but the lack of research specific to older adopted children’s ADHD emerge as significant predictors either for the final endog- makes it difficult to contextualize the results of the current enous variable (CPRS-R) or for the mediators (FSOC and study. It is possible that, as children get older, ADHD symp- FAS). tomatology becomes less responsive to changes in the family environment upon adoption. Such a finding may suggest a developmental window for intervening in adopted children’s Discussion and implications ADHD symptoms, perhaps through early intervention. A strength of the current study is the ability to examine long- The association between family structure and ADHD par- term outcomes for adopted children involved in CLAS at 14 tially confirmed earlier research which found that having a non- years post adoption. Results suggest the importance of the intact structure contributed to ADHD symptoms (Hurtig et al. family environment on adopted children’s outcomes and 2005; Forssman et al. 2009). In this sample, married couples support the gene–environment model of ADHD aetiology scored higher in family coherence than divorced/separated or (Leve et al. 2010b). Contrary to earlier research (Mick et al. single/widowed couples. Yet, no significant direct or indirect 2002; McGuiness et al. 2005), this study found no significant relationships emerged between marital status of the parents and relationships between birthweight and ADHD symptomatol- the ADHD symptoms of a child. ogy. Rather, family coherence and adaptability emerged as the strongest predictors. Limitations The negative correlations between ADHD symptomatology and family coherence and adaptability support earlier research While this study draws data from a longitudinal investigation, it concerning these relationships (Schroeder & Kelly 2009; is limited in its use of only one wave of study for which the Mulligan et al. 2011). Importantly, compared with FSOC scale measures of interest were available. Thus, it is not possible to norms, both mothers and fathers in the current study scored establish with certainty the directionality of effects between higher than those in Antonovsky and Sourani’s (1988) original family environment and ADHD. No information was available study of FSOC (M = 128.63, SD = 33.35 for husbands; related to the child’s medication status for ADHD, a variable M = 130.85, SD = 33.99 for wives). The current sample thus is that may have important implications for child and family func- more optimistic about their family coherence. tioning. The response rate of 49.0% for Wave 4 of CLAS repre- The family environment emerged as a significant mediator sents 38.5% of the Wave 1 sample; this sample may be biased between adoption from foster care and ADHD symptomatol- such that families with higher cohesion and adaptability may ogy. Earlier work by Rutter and colleagues (1975) found that have been more likely to respond. The sample is also primarily adoption from foster care constituted one of several indicators White, well-educated and middle class, such that the findings of risk which, in aggregate, predict greater psychopathology in may not be generalizable to more diverse populations or to children. This work was later confirmed by Biederman and earlier CLAS waves. We conducted an attrition analysis and

© 2013 John Wiley & Sons Ltd, Child: care, health and development, 40, 6, 853–862 860 T.M. Crea et al. found that those in Wave 4 tended to have a higher income at tative approach prior to children’s adoption. Once children Wave2(P < 0.001) and a higher level of education at Wave 3 are adopted, targeting family communication patterns and (P < 0.001). Another limitation is the sole reliance on adoptive perceptions of the family environment may be an additional parent reports of ADHD symptomatology. While this sample of part of developing an evidence-based, post-adoption services adopted children provides a unique opportunity to examine the toolkit. relationship between family environment and ADHD in adoptive families, no non-adoptive families were available for comparison. Key messages

• Research shows that psychosocial factors in the family Conclusion environment, particularly family cohesion and communi- The significant mediating influence of the family environment cation, may mediate genetic predispositions to ADHD. suggests that family intervention strategies may be helpful in • This study of adopted children found that family cohesion improving adopted children’s outcomes. Several studies have and adaptability mediated the relationship between pre- shown that parents of children with ADHD frequently display adoptive risk factors and ADHD symptomatology at 14 elements of authoritarian parenting, versus authoritative par- years post adoption. enting (e.g. Hinshaw et al. 1997; Hurt et al. 2007), and that • These findings suggest that family intervention strategies, positive parenting styles are associated with children’s self- focused on family communication patterns and dynamics, regulation and attention (Eisenberg et al. 2005; Kawabata et al. may be helpful in improving adopted children’s outcomes 2012). Some have argued for early intervention parent training and building an evidence base for post-adoption services. programmes targeted towards pre-school-aged children (Daley et al. 2009), and at least one study suggests that positive parenting training may help decrease young children’s behaviour References problems (Bor et al. 2002). Yet, other research indicates that parent involvement pro- Acock, A. C. (2005) Working with missing values. Journal of Marriage grammes are more helpful in reducing internalizing behaviours, and Family, 67, 1012–1028. and less so with externalizing issues and ADHD (Corcoran & Antonovsky, A. & Sourani, T. (1988) Family sense of coherence and Dattalo 2006). Deault (2010) notes the lack of attention given to family adaptation. Journal of Marriage and the Family, 50, 79–92. Barth, R. P. & Miller, J. M. (2000) Building effective post-adoption parental factors that promote resilience in children, as opposed services: what is the empirical foundation? Family Relations, 49, to dysfunction. This lack of research may contribute to a relative 447–455. absence of early intervention approaches to addressing ADHD Berry, M. & Barth, R. P. (1990) A study of disrupted adoptive in early childhood (Sonuga-Barke et al. 2011), an approach that placements of adolescents. Child Welfare, 69, 213–214. may be worthwhile given the environmental influences seen to Biederman, J. (2005) Attention-deficit/hyperactivity disorder: a be prevalent in its aetiology (Sonuga-Barke & Halperin 2010) or selective overview. Biological Psychiatry, 57, 1215–1220. as triggers to an underlying predisposition (Leve et al. 2010a). Biederman, J., Milberger, S., Faraone, S. V., Kiely, K., Guite, J., Mick, One promising approach to treating ADHD in young children E., Ablon, S., Warburton, R. & Reed, E. (1995) Family- environment risk factors for attention-deficit hyperactivity may be The Incredible Years (TIY) parent and child training disorder. Archives of General Psychiatry, 52, 464–470. programme which shows promise in fostering parenting Biederman, J., Faraone, S. V. & Monuteaux, M. C. (2002) Impact of changes in mothers and positive behavioural changes in chil- exposure to parental attention-deficit/hyperactivity disorder on dren (Webster-Stratton et al. 2011). clinical features and dysfunction in the offspring. Psychological A related issue for adoptive families is the lack of post- Medicine, 32, 817–827. adoption services available to help families avert, or resolve, Bor, W., Sanders, M. R. & Markie-Dadds, C. (2002) The effects of the crises that may occur. Scholars and advocates have highlighted Triple P-Positive Parenting Program on preschool children with co-occurring disruptive behavior and attentional/hyperactive this important lack of services for several years (Barth & Miller difficulties. Journal of Abnormal Child Psychology, 30, 571–587. 2000), while little has changed in the adoption field in providing Burt, C. E., Cohen, L. H. & Bjorck, J. P. (1988) Perceived family supports to adoptive families (Smith 2010). Early intervention environment as a moderator of young adolescents’ life stress approaches that target cognitive development for young chil- adjustment. American Journal of Community Psychology, 16, dren in care (e.g. Nelson et al. 2007) may constitute a preven- 101–122.

Burt, S. A. (2009) Rethinking environmental contributions to child (2009) Independent contributions of cognitive functioning and and adolescent psychopathology: a meta-analysis of shared social risk factors to symptoms of ADHD in two Nordic environmental influences. Psychological Bulletin, 135, 608–637. populations-based cohorts. Developmental Neuropsychology, 34, Burt, S. A., Krueger, R. F., McGue, M. & Iacono, W. (2003) 721–735. Parent–child conflict and the comorbidity among childhood Henderson, A. D., Sayger, T. V. & Horne, A. M. (2003) Mothers and externalizing disorders. Archives of General Psychiatry, 60, sons: a look at the relationship between child behavior problems, 505–513. marital satisfaction, maternal depression, and family cohesion. The Cadoret, R. J., Yates, W. R., Troughton, E., Woodworth, G. & Stewart, Family Journal: Counseling and Therapy for Couples and Families, M. A. (1995) Genetic-environmental interaction in the genesis of 11, 33–41. aggressivity and conduct disorders. Archives of General Psychiatry, Hinshaw, S. P., Zupan, B. A., Simmel, C., Nigg, J. T. & Melnick, S. 52, 916–924. (1997) Peer status in boys with and without attention-deficit Conners, C. K. (1997) Conners’ Parent Rating Scales Revised-Short hyperactivity disorder: predictions from overt and covert antisocial Form (CPRS-R: S). Multi-Health Systems, North Tonawanda, NY, behavior, social isolation, and authoritative parenting beliefs. Child USA. Development, 68, 880–896. Conners, C. K., Sitarenios, G., Parker, J. D. A. & Epstein, J. N. (1998) Howe, D. (2010) ADHD and its comorbidity: an example of The revised Conner’s Parent Rating Scale (CPRS-R): factor gene–environment interaction and its implications for child and structure, reliability, and criterion validity. Journal of Abnormal family social work. Child and Family Social Work, 15, 265–275. Child Psychology, 26, 257–269. Hurt, E. A., Hoza, B. & Pelham, W. E. (2007) Parenting, family Corcoran, J. & Dattalo, P. (2006) Parent involvement in treatment for loneliness, and peer functioning in boys with attention-deficit/ ADHD: a meta-analysis of the published studies. Research on Social hyperactivity disorder. Journal of Abnormal Child Psychology, 35, Work Practice, 16, 561–570. 543–555. Cunningham, C. E. & Boyle, M. H. (2002) Preschoolers at risk for Hurtig, T., Taanila, A., Ebeling, H., Miettunen, J. & Moilanen, I. attention-deficit hyperactivity disorder: family, parenting, and (2005) Attention and behavioural problems of Finnish adolescents behavioral correlates. Journal of Abnormal Child Psychology, 30, may be related to the family environment. European Child and 555–569. Adolescent Psychiatry, 14, 471–478. Daley, D. (2006) Attention deficit hyperactivity disorder: a review of Jaycox, L. H. & Repetti, R. L. (1993) Conflict in families and the the essential facts. Child: Care, Health and Development, 32, psychological adjustment of preadolescent children. Journal of 193–204. Family Psychology, 7, 344–355. Daley, D., Jones, K., Hutchings, J. & Thompson, M. (2009) Attention Ji, J., Brooks, D., Barth, R. P. & Kim, H. (2010) Beyond preadoptive deficit hyperactivity disorder in preschool children: current risk: the impact of adoptive family environment on adopted findings, recommended interventions and future directions. Child: youth’s psychosocial adjustment. American Journal of Health, Care, and Development, 35, 754–766. Orthopsychiatry, 80, 432–442. Deault, L. C. (2010) A systematic review of parenting in relation to Kawabata, Y., Tseng, W. L. & Gau, S. S. (2012) Symptoms of the development of comorbidities and functional impairments in attention-deficit/hyperactivity disorder and social and school children with attention-deficit/hyperactivity disorder (ADHD). adjustment: the moderating roles of age and parenting. Journal of Child Psychiatry and Human Development, 41, 168–192. Abnormal Child Psychology, 40, 177–188. doi: 10.1007/s10802-011- Eisenberg, N., Zhou, Q., Spinrad, T. L., Valiente, C., Fabes, R. A. & 9556-9. Liew, J. (2005) Relations among positive parenting, children’s Letzter-Pouw, S. & Werner, P. (2012) The relationship between loss of effortful control and externalizing problems: a three-wave parents in the Holocaust, intrusive memories, and distress among longitudinal study. Child Development, 76, 1055–1071. child survivors. American Journal of Orthopsychiatry, 82, 201–208. Eriksson, M. & Lindstrom, B. (2006) Antonovsky’s sense of Leve, L. D., Harold, G. T., Ge, X., Neiderhiser, J. M. & Patterson, G. coherence scale and the relation with health: a systematic (2010a) Refining intervention targets in family-based research: review. Journal of Epidemiological Community Health, 60, lessons from quantitative behavioral genetics. Perspectives on 376–381. Psychological Science, 5, 516–526. Faraone, S. V., Perlis, R. H., Doyle, A. E., Smoller, J. W., Goralnick, Leve, L. D., Neiderhiser, J. M., Scaramella, L. V. & Reiss, D. (2010b) J. J., Holmgren, M. A. & Sklar, P. (2005) Molecular genetics of The Early Growth and Development Study: using the prospective attention-deficit/hyperactivity disorder. Biological Psychiatry, adoption design to examine genotype–environment interplay. 57, 1313–1323. Behavioral Genetics, 40, 306–314. Foley, M. (2010) A comparison of family adversity and family Lifford, K. J., Harold, G. T. & Thapar, A. (2008) Parent–child dysfunction in families of children with attention deficit relationships and ADHD symptoms: a longitudinal analysis. hyperactivity disorder (ADHD) and families of children without Journal of Abnormal Child Psychology, 36, 285–296. ADHD. Journal for Specialists in Pediatric Nursing, 16, 39–49. Lucia, V. C. & Breslau, N. (2006) Family cohesion and children’s Forssman, L., Bohlin, G., Lundervold, A. J., Taanila, A., Heiervang, E., behavior problems: a longitudinal investigation. Psychiatry Loo, S., Jarvelin, M. R., Smalley, S., Moilanen, I. & Rodriguez, A. Research, 141, 141–149.

McGuiness, T. M., Ryan, R. & Robinson, C. B. (2005) Protective children with and without ADHD. Journal of Child and Family influences of families for children adopted from the former Soviet Studies, 18, 227–235. Union. Journal of Nursing Scholarship, 37, 216–221. Simmel, C., Brooks, D., Barth, R. P. & Hinshaw, S. P. (2001) Mick, E., Biederman, J., Faraone, S. V., Sayer, J. & Kleinman, S. (2002) Externalizing symptomatology among adoptive youth: prevalence Impact of low birth weight on attention-deficit/hyperactivity and preadoption risk factors. Journal of Abnormal Child Psychology, disorder. Journal of Developmental and Behavioral Pediatrics, 23, 29, 57–69. 16–22. Smith, S. L. (2010) Keeping the Promise: The Critical Need for Moos, R. H. & Moos, B. S. (1994) Family and Environment Scale Post-adoption Services to Enable Children and Families to Manual: Development, Applications, Research, 3rd edn. Consulting Succeed. Evan B. Donaldson Adoption Institute, New York, NY, Psychologists Press, Palo Alto, CA, USA. USA. Mulligan, A., Anney, R., Butler, L., O’Regan, M., Richardson, T., Sonuga-Barke, E. J. S. & Halperin, J. M. (2010) Developmental Tulewicz, E. M., Fitzgerald, M. & Gill, M. (2011) Home phenotypes and causal pathways in attention deficit/hyperactivity environment: association with hyperactivity/impulsivity in disorder: potential targets for early intervention? Journal of Child children with ADHD and their non-ADHD siblings. Child: Care, Psychology and Psychiatry, 51, 368–389. Health and Development, 39, 202–212. Sonuga-Barke, E. J. S., Koerting, J., Smith, E., McCann, D. C. & Nelson, C. A., Zeanah, C. H., Fox, N. A., Marshall, P. J., Smyke, A. T. Thompson, M. (2011) Early detection and intervention for & Guthrie, D. (2007) Cognitive recovery in socially deprived young attention-deficit/hyperactivity disorder. Expert Review of children: the Bucharest Early Intervention Project. Science, 318, Neurotherapeutics, 11, 557–563. 1937–1940. Tully, E. C., Iacono, W. G. & McGue, M. (2008) An adoption study of Polanczyk, G., de Lima, M. S., Horta, B. L., Biederman, J. & Rohde, L. parental depression as an environmental liability for adolescent A. (2007) The worldwide prevalence of ADHD: as systematic depression and childhood disruptive disorders. American Journal of review and metaregression analysis. American Journal of Psychiatry, Psychiatry, 165, 1148–1154. 164, 942–948. Webster-Stratton, C. H., Reid, M. J. & Beauchaine, T. (2011) Rutter, M., Cox, A., Tupling, C., Berger, M. & Yule, W. (1975) Combining parent and child training for young children with Attainment and adjustment in two geographical areas: 1 – the ADHD. Journal of Clinical Child and Adolescent Psychiatry, 40, prevalence of psychiatric disorders. British Journal of Psychiatry, 191–203. 126, 493–509. Yuan, Y. C. (2000) Multiple imputation for missing data: concepts Schroeder, V. M. & Kelly, M. L. (2009) Associations between family and new development (Version 9.0). Rockville, MD, USA: environment, parenting practices, and executive functioning of SAS Institute.