Utilising genetically informed research designs to better understand family processes and child development: implications for adoption and foster care focused interventions

Abstract

Understanding the interplay between genetic factors and family environmental processes (e.g. interparental relationship quality, positive vs negative parenting practices) and children’s mental health (e.g. anxiety, depression, conduct problems, ADHD) in the contexts of adoption and foster care research and practice is critical for effective prevention and intervention programme development. While evidence highlights the importance of family relationship processes (e.g. interparental relationship quality, parent‒child relationship quality) for the mental health and well-being of children in adoption and foster care, there is relatively limited evidence of effective interventions specifically for these families. Additionally, family-based interventions not specific to the context of adoption and foster care typically show small to medium effects, and even where interventions are efficacious, not all children benefit. One explanation for why interventions may not work well for some is that responses to an intervention may be influenced by an individual’s genetic make-up. Alternatively, the targets of family relationship level interventions (e.g. parenting processes) may not always affect the specific environment ‘trigger’ deemed salient to specific child/adolescent outcomes. This article summarises how genetically informed research designs can help disentangle genetic from environmental processes underlying psychopathology outcomes for children, and how this evidence can provide improved insights into the development of more effective preventive intervention targets for adoptive and foster families. We discuss current difficulties in translating behavioural genetics research to prevention science and provide recommendations to bridge the gap between behavioural genetics research and prevention science, with lessons for adoption and foster care research and practice.

Keywords

Quantitative genetics genetically-informed research designs family child mental health intervention adoption foster care prevention science

Introduction

Children and young people in adoption and foster care are at elevated risk of developing multiple poor outcomes, including internalising problems (e.g. depression, anxiety), externalising problems (e.g. conduct problems, aggression), substance misuse, cognitive impairments (e.g. poor language development), negative peer relationships and reduced academic attainment (Fisher, 2015). Understanding the processes that impact on poor child outcomes is crucial for developing and providing efficacious intervention and prevention services aimed at improving outcomes for children and young people. Evidence highlights the relevance of positive rearing environments (e.g. positive interparental and parent‒child relationships) for the well-being of children in adoption and foster care (Harold, Leve and Sellers, 2017; Hyde, et al., 2016). However, fundamental questions remain as to the efficacy of ‘environmental’ interventions (i.e. those that target family relationship processes, such as parenting practices) aimed at children in those contexts. In particular, the relative role of genetic factors passed on from birth parents to children and how these biologically sourced factors influence (and are influenced by) the rearing environments experienced by adopted children and children in foster care remain poorly understood by academics, practitioners, policy-makers and parents/carers working to improve outcomes for children and families. This article aims to clarify evidence relating to the interplay between genetic and environmental (adoptive parent/carer caregiving) factors and psychopathology outcomes for children, presented with a specific UK practice and policy focus.

Adoption and fostering: the current UK context

The UK has recently experienced two major challenges with regards to looked after children, with substantive implications for children in a foster care and adoption context: (1) the increase in the number of children placed in care; and (2) the breakdown of adoption and foster care placements, meaning more children are placed in long-term care. Recent legislative recommendations stating that adoption orders should only be instigated against the wishes of the birth parent as a last resort (The Supreme Court, 2013) have been linked to a reduction in the number of children placed for adoption through the social care system in the UK (Simmonds, 2016). This has resulted in an increase in the number of children remaining in care relative to those being placed for adoption. At 31 March 2018, there were 75,420 looked after children in England alone ‒ a rise of 11% since 2012 when the figure was 67,050 (Department for Education, 2012) ‒ with the majority of these children in foster placements (55,200; 73% of looked after children; Department for Education, 2018).

Conversely, only 3820 looked after children ceased to be looked after due to adoption in 2018. This represents a fall of 28% from the peak figure of 5360 in 2015‒2016 (Department for Education, 2018; 2017), demonstrating the decrease in adoption placements as a result of the change in legal ruling. In addition, it is estimated that approximately 3% to 4% of children adopted in the UK are returned to care after an adoption order is granted (Triseliotis, 2002; Wijedasa and Selwyn, 2017). However, estimates range from 10% to 50% when additional factors (including age of the child at placement, specific learning or developmental difficulties, specific challenging behaviours) are considered (see also Selwyn, Wijedasa and Meakings, 2014). Furthermore, it is estimated that only 68% of looked after children remained in the same placement for one year, while 32% had two or more placements (e.g. moving between foster placements) within one year (Department for Education, 2016), demonstrating instability of placements for children in care. Thus, more children are remaining in care, highlighting the need for support for this vulnerable population.

While the number of adoption orders in the UK is decreasing, it is important to have effective interventions for families in the context of adoption and foster care to reduce placement breakdown (thus preventing the number of children in care further increasing) and to improve outcomes for children. However, there is relatively limited evidence of effective interventions specifically for adoptive or foster families and more robust evidence-led interventions (or modifications of existing ones) are needed.

Broadly speaking, we know that family-based interventions (not specific to looked after children) can reduce rates of child psychopathology (Chamberlain, et al., 2008; Leve, et al., 2012; Tolan and Dodge, 2005; Weisz, et al., 2005). Such interventions have primarily focused on the parent‒child relationship, with an emphasis on promoting positive parenting practices as a key family process mechanism leading to improvements in child outcomes (e.g. Conduct Problems Prevention Research Group, 2002; Eddy and Chamberlain, 2000; Martinez and Forgatch, 2001; Webster-Stratton and Herman, 2008). However, these interventions typically show small to medium effects in terms of improving parenting and/or child welfare. Further, not all children benefit or show sustained effects (i.e. not all children demonstrate long-term improved outcomes as a result of intervention).

One explanation for why interventions may not work for some children is that underlying genetic predispositions can affect children’s responses to the environment (e.g. genetic predispositions leading to children responding more negatively to a certain parenting behaviour) and therefore influence their responses to interventions targeting these environments (e.g. responding negatively to interventions that target this parenting behaviour; Reiss and Leve, 2007; Reiss, et al., 2000; van IJzendoorn and Bakermans-Kranenburg, 2015). Quantitative behavioural genetics research can be used to disentangle genetic and environmental influences such as parenting on child outcomes and examine how genetic predispositions and the rearing environment can interact to shape child outcomes. Therefore, findings from such research designs can be informative for intervention and prevention strategies (i.e. by highlighting aspects of the rearing environment that are important intervention targets, and when interventions may or may not be appropriate) and thus help improve the efficacy of strategies for at-risk children and families.

This article will outline how genetically-informed research designs can improve understanding of the interplay between genetic and environmental processes that impact child and adolescent psychopathology (e.g. depression, anxiety, aggression, conduct problems, ADHD) and, in turn, how this evidence can provide improved insights into the development of effective preventive intervention targets, with a specific focus on adoptive and foster care families. We discuss current difficulties in translating behavioural genetics research to prevention efforts and provide recommendations to bridge the gap between traditional behavioural genetics research and adoption and foster care focused prevention and intervention research and practice.

Understanding the interplay between genetic factors, family environmental processes and child psychopathology

Family environmental processes, such as the quality of the interparental relationship or the consistency of positive vs negative parent–child interaction quality, are recognised as significant influences on children’s emotional and behavioural development (Harold, Leve and Sellers, 2017). However, despite advances in understanding the interplay between family level processes (e.g. parenting) and child mental health (e.g. conduct problems), one of the limitations of past research is that the vast majority of it has mainly relied on studies involving biologically related parents/carers and children; consequently it is not possible to unambiguously separate environmental (e.g. parenting) from shared genetic effects (i.e. genes passed on from genetically related/birth parents to their children/offspring). Although molecular genetics research has provided evidence linking specific genetic processes to child psychopathology (see Thapar and Harold, 2014), research suggests that genetic factors are unlikely to fully explain variation in child developmental outcomes; rather such variation is more likely explained as a result of the specific interplay between genetic factors and family environmental experiences (Caspi, et al., 2003; Henry, Boivin and Tarabulsy, 2015; Kim-Cohen, et al., 2006). It is therefore important to use genetically informed research designs that provide the opportunity to examine associations between family-level processes (e.g. the interparental relationship, parenting) and child outcomes (e.g. depression, conduct problems) with a focus on gene-environment interplay, i.e. relative genetic and environmental contributions and their interaction in explaining child outcomes. Further, without the careful implementation of genetically informed research designs, it is impossible to contend with a fundamental challenge to interpreting any association between an index of rearing environmental experience (e.g. negative parenting practices) and child psychopathology (e.g. conduct disorder) in that associations derived from biologically related parents and children may be confounded by common genes passed on from parents to children that influence both the rearing environmental factor and the outcomes observed in children (see Harold, Leve and Sellers, 2017).

There are three primary ways in which genetic factors passed on from biologically related parents to children/offspring can influence associations between parental behaviour and child outcomes (see examples in Table 1). First, in standard research designs, where parents and children are genetically related, the examination of associations between postnatal environmental factors (e.g. parenting practices) and child outcomes (e.g. conduct problems) may be confounded by shared genes passed on from birth parents to their children, affecting the strength of associations between parenting behaviours and child behaviours (referred to as passive gene–environment correlation, or rGE; Jaffee and Price, 2007). An adoption-at-birth design allows the examination of associations between parenting behaviours and child behaviours where the confound of passive rGE is removed (when children are placed with a non-relative adoptive family). Second, it is recognised that parenting responses to a child may be a reaction to genetically influenced attributes in the child (i.e. child-on-parent effects, known as evocative gene–environment correlation, or rGE; Ge, et al., 1996). The adoption design (see below) provides unique insight into genetic vs environmental contributions to child psychopathology, as it allows the effects of genetically influenced attributes in the child (measured directly or through birth parent information) to be assessed relative to genetically unrelated caregiver (adoptive parent) responses to child behaviour (Ge, et al., 1996). Research examining evocative rGE therefore allows examination of child effects on parenting, and the influence of child-evoked parenting on subsequent child developmental outcomes (see Harold, et al., 2013a). Third, gene-environment interaction (GxE) refers to the interaction between genes and the environment in influencing child outcomes; specifically, how environmental influences (e.g. parenting) may moderate (change the direction or magnitude of) genetic effects on child outcomes and vice-versa (i.e. a child’s genes may moderate the effect of the rearing environment on child outcomes; Leve, et al., 2017; Reiss, Leve and Neiderhiser, 2013).

Table 1.

Description and examples of how heritable characteristics provide insight into the relationship between parenting and child adjustment.

Heritable characteristic	Definition	Example
Passive rGE	Associations between parents and children may be attributable to environment or shared genes	The association between hostile parenting and child disruptive behaviour may be due to hostile parenting acting as an environmental influence on child behaviour (i.e. children may act in a disruptive manner in direct response to hostile parenting, e.g. by modelling hostile behaviours), or associations may be due to the same genes underlying both hostile parenting and disruptive child behaviour. Passive rGE refers to the concept that the association between hostile parenting and child conduct problems may be better explained by shared genes between parents and children (although research suggests that this is not the case; Harold, et al., 2011).
Evocative rGE	Children’s genetically influenced behaviour can evoke specific responses from parents	In the context of an adoption-at-birth design, genetic risk for child impulsivity (indicated by birth parent psychopathology) predicts child impulsivity which in turn predicts subsequent hostile parenting in adoptive parents (e.g. Harold, et al., 2013a)
GxE	Children’s genes interact with their environment to influence behaviour (child responses to environment differ depending on genetic risk and vice-versa)	Structured parenting can reduce behaviour problems in children with high genetic risk (indicated by birth parent psychopathology e.g. alcohol/drug use, antisociality), but can increase behaviour problems in children with low genetic risk (e.g. Leve, et al., 2009).

Note: rGE = gene-environment correlation, GxE = gene-environment interaction.

There are two forms of gene-environment interactions that are important for intervention and prevention science. The first is the ‘diathesis-stress’ model of GxE, where psychopathology results from inherited risk (diathesis) that occurs under particular environmental risks (stressors). Examples of ‘diathesis-stress’ GxE is evident from twin studies (e.g. Kendler, et al., 1995) and adoption studies (e.g. Cadoret, et al., 1995; Leve, et al., 2010). The second form of GxE which has been more recently specified is ‘differential susceptibility’, whereby an individual is differentially sensitive/susceptible to high levels of both positive and negative rearing environments: inherited risks increase susceptibility to the environment, resulting in more positive developmental outcomes in more positive conditions (e.g. warm, nurturing parenting) and more negative developmental outcomes from more negative ones (e.g. poor parent monitoring; Brody, et al., 2013; Dick, et al., 2011). Studies examining GxE therefore illustrate how specific environments may have positive or negative effects for some individuals depending on their genetic susceptibility (see Leve, et al., 2013). This provides evidence for targeted interventions that are informed by biological risk and demonstrates that environments such as parenting (i.e. intervention target) can interact with heritable traits to improve child outcomes. Indeed, based on a meta-analysis of 22 randomised controlled trials, van IJzendoorn and Bakermans-Kranenburg (2015) reported that the combined effect sizes for interventions targeted at those at genetic risk were significant and large (i.e. effect size r = .33), whereas those who were not at genetic risk did not show significant improvements after interventions (i.e. effect size r = .06). This supports the hypothesis of differential susceptibility and suggests that even in the absence of overall efficacy, interventions may have a large impact on a subgroup of genetically susceptible individuals (van IJzendoorn and Bakermans-Kranenburg, 2015).

Understanding gene-environment interplay (GxE, passive and evocative rGE) therefore has important implications for the development of interventions and the evaluation of their efficacy, highlighting where parents’ responses to child behaviour may be altered through intervention and how targeting specific rearing environments (e.g. parenting behaviours) may or may not improve child outcomes, depending on the child’s genetic make-up. It is also necessary to employ research designs that can identify environmental risks which are independent of common genetic influences if we are to better understand what malleable environmental factors may be targeted to reduce poor outcomes for children (Harold, Leve and Sellers, 2017). Understanding relative genetic and environmental contributions to child outcomes is of particular relevance for the development of efficacious interventions specific to families in which parents and children are genetically unrelated (i.e. specific to an adoption and foster care context). For example, understanding how the rearing environment can affect child outcomes among genetically unrelated parents and children can translate to a potential intervention target for such individuals in an adoption and foster care context. While relatively limited interventions specifically target adoptive and foster families, we provide examples of the evaluation of two such interventions later in this article, the findings of which demonstrate that adapting the rearing environment can improve outcomes for children among these genetically unrelated parent/carer--child groupings.

Quantitative behavioural genetics research designs: a practice focused primer

Quantitative behavioural genetic research designs identify genetic and environmental contributions to behaviour by examining the variation in genetic relatedness between family members (see Figure 1). The most commonly used design is the twin study which examines similarities between monozygotic (MZ) twins, who share 100% of their genes, and dizygotic twins (DZ) who share, on average, 50% of their genes. Greater similarity between MZ compared to DZ twins indicates genetic influences. Where MZ and DZ twin pairs share a trait to an equal extent within the pair, this is indicative of environmental factors (Thapar, et al., 2007).

Figure 1.

A summary of traditional behavioural genetic research designs (Harold, Leve and Sellers, 2017).

The extended family design examines associations between siblings who differ in their genetic relatedness (D’Onofrio, et al., 2013): full siblings (sharing, on average, 50% of their genes), half-siblings (sharing, on average, 25% of their genes) and step-siblings (sharing no genes). If associations for a particular trait are stronger between full sibling pairs than half- or step-sibling pairs, this would be indicative of genetic influence. Conversely, where associations are similar between the different sibling pair types, this would suggest environmental factors as influences on children’s outcomes.

The Children of Twins (CoT) design is an extension of the ‘extended family’ design and takes advantage of the fact that the children of MZ twins are equally genetically related to their parents as they are to the twin’s sibling (i.e. their uncle/aunt), but they do not typically share an environment with the parent’s twin sibling (D’Onofrio, et al., 2007). In MZ twin families, if the correlation between parent‒child is greater than the MZ uncle/aunt‒child correlation (avuncular correlation), this is indicative of environmental influences. If the parent‒child correlation is similar to the MZ uncle/aunt‒child (avuncular) correlation, this suggests genetic influences. The comparison between avuncular correlations (uncle/aunt‒child) between MZ and DZ families provides insights into the nature of familial effects: if MZ avuncular correlations are larger than DZ avuncular correlations, then genetic factors are implied.

Studies of siblings reared apart can also examine the relative roles of genes and the environment for child development (Bouchard, et al., 1990; Leve, et al., 2017; Mednick, Gabrielli and Hutchings, 1984; Pedersen, et al., 1991; Rutter, et al., 2001). Siblings reared apart do not share a rearing environment, but as part of the study design genetic similarities are controlled for, so any differences between siblings are ascribed to (different) rearing environments. Studies of siblings reared apart can include MZ twins: as MZ twins share 100% of their genes, any differences between siblings are ascribed to differences in rearing environment. Studies of siblings raised apart can also compare adopted children (who are reared by biologically unrelated parents) with their biological siblings who remain with their birth parent(s) (e.g. Kendler, et al., 2016; Sorensen, et al., 1989). This method provides insights into how different environmental influences can affect child outcomes where children share genes and can be used to infer what the outcomes for children may have been had they not been adopted (Harold, Leve and Sellers, 2017).

The adoption design can also be used to examine environmental influences on children. Adopted children who are placed in non-relative placements are genetically unrelated to their rearing parents (removing the confound of passive rGE), so associations between adoptive parents and the adopted child are attributed to environmental processes (e.g. Leve, et al., 2013; Rhea, et al., 2013). A full adoption design also includes birth parents, providing the opportunity to examine genetic influences: where children are adopted at birth, associations between birth parents and adopted children can only be attributed to genetic factors (and specific to the birth mother: intrauterine influences). In addition, where a full adoption design is longitudinal, evocative effects (i.e. genetically influenced attributes in the child that ‘evoke’ specific rearing environment responses; evocative rGE; Ge, et al., 1996) can also be examined. Therefore, in addition to allowing the examination of rearing environmental influences on child development, the full adoption design also provides insight into how child behaviour (which is in part attributable to genetic influences) can influence/evoke specific parenting behaviours in genetically unrelated (adoptive) parents.

Artificial Reproductive Technologies (ARTs) provide the opportunity to examine associations between parents and children who are genetically related or genetically unrelated to either or both of their rearing parents (‘adoption at conception’; Harold, et al., 2012). Through in-vitro fertilisation (IVF), children can be genetically related to both rearing parents (homologous IVF), the rearing mother but not father (sperm donation), the rearing father but not mother (egg donation) or neither parent (embryo donation). In an additional group (gestational surrogacy), children are genetically related to both parents but the prenatal environment is provided by a surrogate, thus allowing the examination of prenatal influences separate from genetic influences. Associations between genetically related parent‒child dyads but not between genetically unrelated parent‒child dyads indicate genetic influences. Associations between genetically unrelated parent‒child dyads indicate environmental influences (see Harold, et al., 2013b; Thapar, et al., 2009).

Unlike the above study designs, which examine associations between parents/carers and children who differ in their degree of genetic relatedness, molecular genetic studies focus on specific measured genes assayed from DNA samples. While there is increasing evidence of interactions between measured genetic factors and environmental exposures (GxE) for child outcomes (Belsky, Suppli and Israel, 2014; Moffitt, Caspi and Rutter, 2005; 2006), more research is required to better understand how specific genetic susceptibilities work with specific environmental exposures (Thapar and Harold, 2014). In addition, molecular genetic studies typically require very large sample sizes (to provide adequate statistical power) that may limit the ability to acquire environmental process information (psychometrically robust measures of the environment) which would allow effective examination of gene‒environment interplay (Harold, Leve and Sellers, 2017). Furthermore, effect sizes in molecular genetic studies tend to be quite small (Thapar and Harold, 2014). Consequently, here we focus primarily on the importance of quantitative behavioural genetic research for prevention science.

Converting findings from quantitative behaviour genetics research to front-line practice: an update on the evidence

As summarised, it is well established that both genetic and environmental factors contribute to child development. Adoption and twin studies have demonstrated genetic influences on multiple outcomes, including cognitive ability, language development, depression, anxiety, ADHD, school achievement and other outcomes (Haworth, et al., 2010; Leve, et al., 2013; Thapar, et al., 2007; Thapar and Rice, 2006). There is also evidence of genetic factors contributing to long-term outcomes including the intergenerational transmission of poor mental health outcomes: some twin and adoption studies provide evidence of strong genetic influences on antisocial behaviours (Arseneault, et al., 2003; Bornovalova, et al., 2014), with a recent sibling study suggesting that intergenerational transmission of anxiety may be accounted for by genetic confounding (Bekkhus, et al., 2017). However, environmental influences are also recognised as important for child development, with meta-analyses of twin and adoption studies finding evidence of both genetic and environmental contributions to child psychopathology, including depression, anxiety, conduct disorder and broader internalising and externalising symptoms (Burt, 2009).

A recent study of siblings reared apart provides evidence of environmental influences of risk for child substance abuse, with children living with their biological parents being at increased risk compared to their adopted siblings (Kendler, et al., 2016). CoT studies examining intergenerational transmission find that the transmission of anxiety and depression is primarily attributed to environmental influences (Eley, et al., 2015; Natsuaki, et al., 2014; Silberg, Maes and Eaves, 2010). However, findings are complex with evidence highlighting that associations may differ depending on outcomes: associations between parental depression and child depression are accounted for by environmental factors, whereas associations between parental depression and child conduct problems are accounted for by both genetic and environmental factors (Silberg, Maes and Eaves, 2010; Singh, et al., 2011). Building on this evidence, a fundamental problem for intervention programme development using traditional and some quantitative behavioural genetic research designs (e.g. twin designs) is the inability to unambiguously disentangle genetic influences underlying the associations between environmental processes and child outcomes (passive and evocative rGE, implications for testing GxE).

Removing the confound of passive rGE

Some quantitative genetic research designs are able to directly address the confound of passive rGE by employing samples of parents and children who are not biologically related: adoption studies and studies of children conceived via IVF. Evidence from such study designs demonstrates the importance of a range of specific family processes as environmental factors that impact on child outcomes. Specifically, interparental conflict and poor parenting practices have been identified as important risks for child outcomes, with interparental conflict predicting child ADHD symptoms (Harold, et al., 2013a), child disruptive behaviours (Bornovalova, et al., 2014), child sleep problems (Mannering, et al., 2011) and adolescent delinquency (Burt, et al., 2007) via disrupted parenting practices (see Harold, Leve and Sellers, 2017). Evidence further highlights that the interplay between interparental conflict and poor parenting practices may extend beyond a traditional focus on the mother–child relationship, with very recent evidence highlighting the importance of both mother and father parenting practices in the context of interparental conflict and child outcomes (Harold, et al., 2013b; Rhoades, et al., 2012; Stover, et al., 2012).

Examining the relevance of evocative rGE

As noted earlier, evocative rGE examines how genetically inﬂuenced child characteristics may evoke specific patterns of response, such as parental hostility (Ge, et al., 1996). This is of interest to intervention research as particular environmental processes can be identified and supported to reduce the impact of child-driven effects (Luthar and Brown, 2007). A relatively small adoption sample provided the first example of evocative rGE in the field of developmental science (Ge, et al., 1996), finding that birth mother psychopathology was associated with disrupted child behaviour, which, in turn, was associated with adoptive mother hostility. More recently, larger adoption studies have advanced understanding of evocative rGE, identifying genetically influenced child characteristics that evoke negative maternal and paternal parenting practices (Elam, et al., 2014; Fearon, et al., 2015; Hajal et al., 2015; Harold, et al., 2013a). Furthermore, adoption studies have also demonstrated how child-evoked negative parenting behaviours can in turn increase children’s negative behaviours (Elam, et al., 2014; Harold, et al., 2013a), highlighting such parenting as a mechanism for continuity (and increase) in negative child behaviours over time. Twin studies have also demonstrated evocative processes, evidencing the effect of children’s genetically influenced characteristics on parenting behaviour (Klahr and Burt, 2014) and negative family relationships (Feinberg, et al., 2005; Neiderhiser, Marceau and Reiss, 2013; Reiss, et al., 2000). These illustrative examples show how quantitative behavioural genetics studies can be used to demonstrate child-on-parent effects and how child-evoked parenting can influence long-term child development. Furthermore, this research provides information for potential intervention pathways that would not be evident from studies that are not genetically informed. Findings suggest that sensitively informing parents that children can inherit specific behaviours and helping parents to become ‘resilient’ to potential child-evoked effects may interrupt the processes through which heritable traits and harsh parenting responses may increase long-term child behaviour problems.

Exploring gene-environment (GxE) interaction

Evidence from quantitative behavioural genetics studies suggests that the impact of specific family processes (including interparental conflict, negative parenting/hostility and maltreatment) on child behaviour problems may differ as a function of children at high vs low genetic risk (e.g. Cadoret, et al., 1995; Jaffee, et al., 2005; Rhoades, et al., 2011; Rice, et al., 2006; Schermerhorn, D’Onofrio and Turkheimer, 2011). Rather than being vulnerable to specific risk environments, children may be differentially susceptible to certain types of family environments as a function of their own genetic make-up (Hyde, et al., 2016; Leve, et al., 2009). According to a recent study, positive parenting buffered the impact of genetic risk, reducing early callous-unemotional behaviours in children at high genetic risk (Hyde, et al., 2016). Furthermore, a study using an adoption design (Leve, et al., 2009) found that specific parenting strategies differentially affected child behaviour problems depending on child genetic risk: structured parenting (providing clear instructions and structure for child activities) decreased child behaviour problems in children at high genetic risk but was associated with increased child behaviour problems where children were at low genetic risk. Conversely, positive reinforcement benefited children regardless of such risk. These results indicate that the interventions that promote structured parenting may only be beneficial for children at high genetic risk but that alternative parenting techniques may be more beneficial for children where the genetic risk is low. Furthermore, evidence from quantitative behavioural genetics studies also suggests that specific family processes can also be differentially influenced by child genetic risk. In an adoption study, birth mother externalising problems (a marker of child genetic risk) predicted adoptive mother negativity but only in the context of adoptive parent interparental hostility (Fearon, et al., 2015). This provides evidence of evocative rGE interacting with specific features of the rearing environment (interparental relationship hostility) and associated impacts on parenting.

Overall, genetically informed studies highlight the importance of the family environmental processes (e.g. interparental conflict, maternal and paternal parenting) for child outcomes whether parents are biologically related to their children or not (e.g. in the contexts of adoption and foster care). This underscores the importance of the rearing environment for intervention targets. Findings from gene-environment interaction focused research have important implications for the development of interventions (Collins and Varmus, 2015); they suggest the benefits for a range of child outcomes of more precise targeting of interventions to a child’s specific characteristics. Evidence of evocative gene-environment correlations is also informative for interventions: findings can identify areas where parents may be affected by genetically influenced child behaviours (i.e. increasing awareness of child effects on parenting) and targeting such processes to support parents to become resilient to these child effects and promote more resilient/adaptive rearing environments and, ultimately, improve outcomes for children. As an illustrative example to this hypothesis, we provide a brief synopsis of two interventions focusing on primary family process/environmental factors reviewed in this article (parenting practices) specific to foster care and adoption contexts.

Examples of interventions in the context of adoption and foster care

Treatment Foster Care Oregon (TFCO)

TFCO (formerly known as Multidimensional Treatment Foster Care, MTFC) is a US-based intervention designed for foster carers of children and adolescents who have experienced maltreatment and are at risk for delinquency (Chamberlain, 2003; Fisher and Stoolmiller, 2008). In TFCO, foster carers are provided with intense parent skills training (e.g. providing support, mentoring, supervision and consistent limit setting). TFCO has been shown to improve outcomes (e.g. depression, delinquency, psychosis, teenage pregnancies) for both children and adolescents in the US and Sweden (Leve, et al., 2012; Poulton, et al., 2014; Westermark, Hansson and Olsson, 2011). One study has evaluated the effectiveness of this intervention in the UK, finding that improvements in adjustment were only evident for those with antisocial behaviour (Green, et al., 2014), although limitations due to sample constraints reduced the statistical power of this study to detect overall group differences (Harold and DeGarmo, 2014).

AdOpt Parenting Programme

A recently implemented programme in the UK aimed at adoptive parents and children is the AdOpt parenting programme (National Implementation Service, NIS; www.evidencebasedinterventions.org.uk/about/national-implementation-service), adapted from a US intervention (KEEP; Keeping Foster & Kinship Carers Supported; Price, et al., 2009) for adoptive parents post-legal order with children aged three to eight years. The programme is designed as a preventive intervention to help parents understand the often complex needs of their adopted children and to support positive parenting techniques, with the aim of enhancing positive behaviours in children. This parenting programme has been evaluated in the UK and demonstrated improvements in parenting behaviours, in addition to reductions in child total problems and conduct problems (Harold, et al., 2017). It did not demonstrate improvements in child emotional, hyperactivity or peer problems, nor did it improve prosocial behaviours. The evaluation of the AdOpt parenting programme suggests that it is suitable as a universal intervention for the specified population (adoptive families, post legal-order), impacting on parenting behaviours, child total difficulties and conduct problems. However, evidence derived from quantitative behavioural genetics studies may help illuminate the mechanisms and processes that affect other mental health difficulties that were not evidenced to be influenced by the intervention (e.g. emotional, peer and hyperactivity problems), so furthering understanding of intervention targets specific to these difficulties in the context of genetic risk.

Challenges in translating research from quantitative behavioural genetics to prevention science

Quantitative behavioural genetics research has identified the role of specific family environmental factors relative to underlying genetic susceptibility in explaining variation in multiple child developmental outcomes. These research designs help address limitations of research that is not genetically informed by examining gene-environment interplay, specifically passive rGE, evocative rGE and GxE. This evidence can inform prevention science by providing an evidence base for more precise intervention targets (what to target and for whom) based on genetically influenced child characteristics. However, there are a number of existing challenges to translating quantitative behavioural genetics to prevention science.

1. Environments are multifaceted and dynamic

Many family-based intervention studies focus on a specific collection of ‘environmental’ targets. For example, the AdOpt parenting programme targets a range of parenting behaviours including parental support, sensitivity and warmth, as well as monitoring and limit setting (Harold, Leve and Sellers, 2017). In contrast, quantitative genetics research often focuses on unidimensional constructs rather than on multiple measures of parenting (as one index of family environmental influence). This has limited translation of behavioural genetics research to prevention science. More recently, quantitative behavioural genetics research has begun to examine how multiple aspects of the family environment can influence child behaviour (Leve, et al., 2017). This more closely aligns with intervention studies that target multiple environmental processes (see Harold and Sellers, 2018).

In addition, employing a longitudinal research design is advantageous. Longitudinal research provides the opportunity to examine modifiable, mediating mechanisms – a central component of preventive interventions. Studies that remove the confound of passive rGE and examine evocative rGE may be more readily translated to prevention studies (Ge, et al., 1996; O’Connor, et al., 1998). Specifically, examination of rGE provides the opportunity for two targets for prevention science: (1) examining evocative rGE can identify genetically influenced child behaviours that can evoke negative responses from parents, enabling interventions to target environmental responses to genetically influenced traits, for example promoting resilient parenting for specific child evoked characteristics; and (2) removing the confound of passive rGE allows the identification of specific environmental influences on children’s development, enabling interventions to promote individual strengths to reduce adverse responses to specific environmental risks, e.g. promoting child resilience to adverse environments. Furthermore, examining GxE can identify aspects of the environment that can be targeted by intervention to offset genetic risk, allowing programmes to be tailored to individuals depending on their genetic susceptibility (van IJzendoorn and Bakermans-Kranenburg, 2015). However, only a small number of studies have examined GxE or rGE (passive and evocative) on child outcomes longitudinally (see Harold, Leve and Sellers, 2017). It is therefore important for future research to look at how aspects of the environment affect child outcomes at a later time-point and how this relationship may vary as a function of genetic risk.

2. Promoting positive rearing environments in intervention studies as compared to measuring negative rearing environments in research studies

Preventive interventions focus on enhancing positive rearing environments (e.g. building parenting skills and promoting positive environmental change) to prevent negative child outcomes: for example, the AdOpt parenting programme focuses on promoting positive parenting strategies (Harold, Leve and Sellers, 2017). In contrast, many genetically informed studies focus on environmental risks (e.g. hostile parent–child relationships, harsh parenting practices) and how genetic risks interact with these to affect child outcomes (e.g. Harold, et al., 2013a; Leve, et al., 2009; Rhoades, et al., 2011). This makes it challenging to translate findings from behavioural genetic research to prevention and intervention contexts. Studies that examine GxE and rGE scrutinise positive, strength-based environments (e.g. parent‒child warmth, interparental satisfaction, etc.) and evidence how these positive environments can offset genetic risk (or promote child strengths) would more closely align to prevention science efforts (e.g. Ganiban, et al., 2007; Leve, et al., 2009; Neiderhiser, et al., 2007).

3. The importance of employing accurate statistical approaches to examining questions and interpreting findings

Another limitation of translating quantitative behavioural genetics research to intervention development and prevention science is that different statistical approaches are typically used by the two disciplines. Quantitative behavioural genetics studies typically employ correlational approaches (to examine associations between variables or groups) whereas, conversely, intervention studies tend to compare mean/variance scores between groups (i.e. comparing mean scores between a group that has received an intervention and a group that has not). These different statistical approaches typically employed by these respective disciplines (i.e. mean level differences between groups vs correlations between variables) have hindered the translation of findings to practice contexts. For example, a common misconception in applying genetic research to intervention development is that heritable behaviours are not modifiable (and would therefore be unsuitable intervention targets). Quantitative genetic studies examine correlations between family members who differ in their degree of genetic relatedness (e.g. MZ or DZ twins) to calculate heritability (the proportion of variance in behaviours attributable to genetic factors). However, this heritability is not equivalent to non-malleability (i.e. impenetrable to change). Indeed, the misconception that heritability equates to non-malleability remains in research and practice circles, despite multiple quantitative genetic studies (e.g. Leve, et al., 2009) demonstrating that positive rearing environments can interact with heritable behaviours to offset genetic risk and improve outcomes for children (thus providing evidence to suggest that heritable behaviours are malleable via environmental processes; see also Leve, et al., 2010). This underlines the need for better clarification and interpretation of findings from quantitative behavioural genetics studies in order to strengthen the translation from research to intervention/prevention practice.

4. The relevance of sample characteristics to the interpretation of substantive findings

Another challenge to translating research from quantitative genetics to prevention studies is that intervention research is typically conducted with high-risk samples. This is in contrast to the majority of quantitative genetics research that has been conducted with low-risk ones. Existing GxE research demonstrates that genetic and environmental influences on behaviour vary as a function of risk (Rutter, 2006). Therefore, sample characteristics need to be considered in translation efforts to ensure that intervention implications map on to research findings. In addition, examining GxE and rGE processes in risk-based samples would also improve translation efforts (e.g. Jaffee, et al., 2005). Further, interpretation of the ‘environment’ relative to prevention science targets needs to be responsive to the complexities of family, community and wider environmental impacts on outcomes.

5. Bridging the gap between genetically-informed research and preventive interventions

Despite the limitations and challenges of translating genetically informed research to prevention science and the implementation of family-based interventions in the context of adoption and foster care, there is significant potential for translation between disciplines. Leve and colleagues (2017) outline how translational efforts can be made from genetically informed research to preventive intervention development. Specifically, they demarcate steps for both quantitative genetics research and for preventive intervention research. Quantitative behavioural genetics researchers should conduct studies that: (1) specify a theory of change; (2) examine the role of genetic and environmental influences on outcomes by employing robust measures that map onto preventive intervention targets; and (3) are replicable and demonstrate robust effects. Alongside such practical steps, prevention science should: (1) identify an intervention that maps onto a theory of change specified in quantitative genetics studies; (2) ensure that the intervention targets the specified environmental mechanism, and that there is overlap with quantitative genetic studies regarding the measurement of this mechanism; and (3) employ designs and samples that identify individuals for whom and conditions under which the intervention is most effective (Leve, et al., 2017). These steps will allow quantitative genetics studies to map onto an intervention mechanism of change, thus intervention can be appropriately modified, taking into account inherited characteristics to provide more precise mechanisms of change.

Summary and recommendations

It is well established that family environmental processes are important for child outcomes, whether parents/carers and children are genetically related or not. Thus, the need to improve interventions for adoption and foster families is self-evident given that the children involved are at elevated risk of psychopathology and other related outcomes (e.g. reduced academic attainment, adverse intergenerational transmission processes). The evidence base from the quantitative behavioural genetics research discussed in this article is, therefore, highly relevant as it can substantively inform intervention and prevention studies. These research designs can generate insights into the role of specific mechanisms underlying specified child outcomes and can inform why specific environments may be moderated by inherited child characteristics. Where genetically informed studies are longitudinal, they can also generate knowledge of dynamic processes that may illuminate how adaptation may vary during specific developmental periods (e.g. infancy, childhood and adolescence). This knowledge can then be integrated into prevention science strategies aimed at increasing the efficacy of family-based interventions targeting improved child outcomes. Additionally, quantitative behaviour genetic research designs can provide insights into how a child’s inherited propensities may affect the efficacy of interventions; what works well for one child may not work for another. These insights can allow interventions to be tailored to specific risks and therefore improve specified child outcomes. In some countries (e.g. specific US states), the administrative processes leading to adoption and foster care placement decisions are highly visible and lend themselves to an offer of intervention by social (state) or health agencies at or even before the time of placement (see Leve, et al., 2012). This creates real opportunities to offer evidenced-based interventions to families at high risk of intergenerational challenge, particularly at the time of placement. However, there are also risks to the ever-delicate balance between the needs of parental caregivers and children. The language of scientific research and associated literatures can potentially be experienced as stigmatising and critical of a group of parents and even would-be parental caregivers who are already far more heavily scrutinised than most biological parents. Interventions in this field need to be carefully pre-piloted and scrutinised for potential unwanted effects, both on the families recruited and also on readiness to offer adoption and foster care placement. Given the emerging importance of evocative rGE effects, primary care health professionals will often be the first in line to recognise family stressors and related mechanisms in such families. Interventions to help health professionals be more aware that parental and child distress may be best addressed through family-based interventions are likely to require considerable adjustment of referral routes in order to improve programme alignment and targeted child outcomes.

Notwithstanding these important caveats, opportunity is at hand to integrate and translate quantitative behavioural genetics research into prevention science efforts to provide a robust evidence base for practice, and to promote efficacious and individually targeted supports to help children and their families.

Footnotes

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Ruth Sellers and Gordon Harold were supported by the Economic and Social Research Council (ESRC) project grant awards (ES/N003098/1 and ES/L014718/1 respectively). Leslie Leve was supported by the National Institutes of Health, USA (1/UG3/ODO23389).

Ruth Sellers is an Andrew and Virginia Rudd Senior Lecturer in Child Development and Community Mental Health, University of Sussex.

Amelia Smith is a postgraduate student in the same department.

Leslie D Leve is a Professor at the Prevention Science Institute, University of Oregon.

Elizabeth Nixon is an Assistant Professor in Developmental Psychology based in the School of Psychology, Trinity College Dublin.

Tam Chipawe Caneis a Lecturer in Social Work, School of Education and Social Work, University of Sussex, UK.

Jackie Cassell is Head of the Department of Primary Care and Public Health and Director of Research and Knowledge Exchange, Brighton and Sussex Medical School.

Gordon Harold is Andrew and Virginia Rudd Chair, and Professor of Psychology, University of Sussex, and Adjunct Professor, Psychology, Trinity College Dublin.

References

Arseneault

Moffitt

Caspi

, et al. (2003) Strong genetic effects on cross‐situational antisocial behaviour among 5‐year‐old children according to mothers, teachers, examiner‐observers, and twins’ self‐reports. Journal of Child Psychology and Psychiatry 44(6): 832–848.

Bekkhus

Lee

Nordhagen

, et al. (2017) Re-examining the link between prenatal maternal anxiety and child emotional difficulties, using a sibling design. International Journal of Epidemiology. doi: 10.1093/ije/dyx186

Belsky

Suppli

Israel

(2014) Gene-environment interaction research in psychiatric epidemiology: a framework and implications for study design. Social Psychiatry and Psychiatric Epidemiology 49(10): 1525–1529.

Bornovalova

Cummings

Hunt

, et al. (2014) Understanding the relative contributions of direct environmental effects and passive genotype-environment correlations in the association between familial risk factors and child disruptive behavior disorders. Psychological Medicine 44(4): 831–844.

Bouchard

Lykken

McGue

, et al. (1990) Sources of human psychological differences: the Minnesota study of twins reared apart. Science 250: 223–228.

Brody

Chen

, et al. (2013) Supportive family environments, genes that confer sensitivity, and allostatic load among rural African American emerging adults: a prospective analysis. Journal of Family Psychology 27(1): 22–29.

Burt

(2009) Rethinking environmental contributions to child and adolescent psychopathology: a meta-analysis of shared environmental influences. Psychological Bulletin 135: 608–637.

Burt

McGue

Krueger

Iacono

(2007) Environmental contributions to adolescent delinquency: a fresh look at the shared environment. Journal of Abnormal Child Psychology 35(5): 787–800.

Cadoret

Yates

Woodworth

Stewart

(1995) Genetic-environmental interaction in the genesis of aggressivity and conduct disorders. Archives of General Psychiatry 52(11): 916–924.

10.

Caspi

Sugden

Moffitt

, et al. (2003) Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science 301(5631): 386–389.

11.

Chamberlain

(2003) Treating Chronic Juvenile Offenders: Advances made through the Oregon Multidimensional Treatment Foster Care model. Washington, DC: American Psychological Association.

12.

Chamberlain

Price

Leve

Laurent

, et al. (2008) Prevention of behavior problems for children in foster care: Outcomes and mediation effects. Prevention Science 9(1): 17–27.

13.

Collins

Varmus

(2015) A new initiative on precision medicine. New England Journal of Medicine 372(9): 793–795.

14.

Conduct Problems Prevention Research Group (2002) Evaluation of the first 3 years of the Fast Track prevention trial with children at high risk for adolescent conduct problems. Journal of Abnormal Child Psychology 30(1): 19–35.

15.

Department for Education (2012) Children looked after in England (including adoption and care leavers) year ending 31 March 2012. Available at: www.gov.uk/government/uploads/system/uploads/attachment_data/file/219210/sfr20-2012v2.pdf

16.

Department for Education (2016) Children looked after in England including adoption: 2015 to 2016. Available at: www.gov.uk/government/statistics/children-looked-after-in-england-including-adoption-2015-to-2016

17.

Department for Education (2017) Children looked after in England (including adoption) year ending 31 March 2017. Available at: www.gov.uk/government/uploads/system/uploads/system/uploads/attachment_data/file/664995/SFR50_2017-Children_looked_after_in_England.pdf

18.

Department for Education (2018) Children looked after in England (including adoption) year ending 31 March 2018. Available at: www.gov.uk/government/uploads/system/uploads/attachment_data/file/757922/Children_looked_after_in_England_2018_Text_revised.pdf

19.

D’Onofrio

Lahey

Turkheimer

Lichtenstein

(2013) Critical need for family-based, quasi-experimental designs in integrating genetic and social science research. American Journal of Public Health 103(S1): S46–S55.

20.

D’Onofrio

Slutske

Turkheimer

, et al. (2007) Intergenerational transmission of childhood conduct problems: a Children of Twins Study. Archives of General Psychiatry 64(7): 820–829.

21.

Dick

Meyers

Latendresse

, et al. (2011) CHRM2, Parental monitoring, and adolescent externalizing behavior evidence for gene-environment interaction. Psychological Science 22(4): 481–489.

22.

Eddy

Chamberlain

(2000) Family management and deviant peer association as mediators of the impact of treatment condition on youth antisocial behavior. Journal of Consulting and Clinical Psychology 68(5): 857–863.

23.

Elam

Harold

Neiderhiser

, et al. (2014) Adoptive parent hostility and children’s peer behavior problems: examining the role of genetically informed child attributes on adoptive parent behavior. Developmental Psychology 50(5): 1543–1552.

24.

Eley

McAdams

Rijsdijk

, et al. (2015) The intergenerational transmission of anxiety: a children-of-twins study. American Journal of Psychiatry 172(7): 630–637.

25.

Fearon

Reiss

Leve

, et al. (2015) Child-evoked maternal negativity from 9 to 27 months: evidence of gene-environment correlation and its moderation by marital distress. Development and Psychopathology 27(401): 1251–1265.

26.

Feinberg

Reiss

Neiderhiser

Hetherington

(2005) Differential association of family subsystem negativity on siblings' maladjustment: using behavior genetic methods to test process theory. Journal of Family Psychology 19(4): 601–610.

27.

Fisher

(2015) Review: Adoption, fostering, and the needs of looked‐after and adopted children. Child and Adolescent Mental Health 20(1): 5–12.

28.

Fisher

Stoolmiller

(2008) Intervention effects on foster parent stress: associations with children’s cortisol levels. Development and Psychopathology 20(3): 1003–1021.

29.

Ganiban

Spotts

Lichtenstein

, et al. (2007) Can genetic factors explain the spillover of warmth and negativity across family relationships? Twin Research and Human Genetics 10(2): 299–313.

30.

Conger

Cadoret

, et al. (1996) The developmental interface between nature and nurture: a mutual influence model of child antisocial behavior and parent behaviors. Developmental Psychology 32(4): 574–589.

31.

Green

Biehal

Roberts

, et al. (2014) Multidimensional Treatment Foster Care for Adolescents in English care: randomised trial and observational cohort evaluation. The British Journal of Psychiatry 204(3): 214–221.

32.

Hajal

Neiderhiser

Moore

, et al. (2015) Angry responses to infant challenges: parent, marital, and child genetic factors associated with harsh parenting. Child Development 86(1): 80–93.

33.

Harold

DeGarmo

(2014) Concerns regarding an evaluation of MTFC-A for adolescents in English care. The British Journal of Psychiatry 205(6): 498.

34.

Harold

Elam

Lewis

, et al. (2012) Interparental conflict, parent psychopathology, hostile parenting, and child antisocial behavior: examining the role of maternal versus paternal influences using a novel genetically sensitive research design. Development and Psychopathology 24(4): 1283–1295.

35.

Harold

Hampden-Thompson

Rodic

Sellers

(2017) An Evaluation of the AdOpt Parenting Programme. London: Department for Education.

36.

Harold

Leve

Barrett

, et al. (2013a) Biological and rearing mother influences on child ADHD symptoms: revisiting the developmental interface between nature and nurture. Journal of Child Psychology and Psychiatry 54(10): 1038–1046.

37.

Harold

Leve

Elam

, et al. (2013b) The nature of nurture: disentangling passive genotype–environment correlation from family relationship influences on children's externalizing problems. Journal of Family Psychology 27(1): 12–21.

38.

Harold

Leve

Sellers

(2017) How can genetically informed research help inform the next generation of interparental and parenting interventions? Child Development 88(2): 446–458.

39.

Harold

Rice

Hay

, et al. (2011) Familial transmission of depression and antisocial behavior symptoms: disentangling the contribution of inherited and environmental factors and testing the mediating role of parenting. Psychological Medicine 41(06): 1175–1185.

40.

Harold

Sellers

(2018) Annual Research Review: Interparental conflict and youth psychopathology: an evidence review and practice focused update. Journal of Child Psychology and Psychiatry 59(4): 374–402.

41.

Haworth

Wright

Luciano

, et al. (2010) The heritability of general cognitive ability increases linearly from childhood to young adulthood. Molecular Psychiatry 15(11): 1112–1120.

42.

Henry

Boivin

Tarabulsy

(2015) Gene-environment interplay in parenting young children. In: Horwitz

Neiderhiser

(eds) Gene-Environment Interplay in Interpersonal Relationships across the Lifespan. New York: Springer, pp. 13–55.

43.

Hyde

Waller

Trentacosta

, et al. (2016) Heritable and nonheritable pathways to early callous-unemotional behaviors. American Journal of Psychiatry 173(9): 903–910.

44.

Jaffee

Price

(2007) Gene–environment correlations: a review of the evidence and implications for prevention of mental illness. Molecular Psychiatry 12(5): 432.

45.

Jaffee

Caspi

Moffitt TE , et al. (2005). Nature × nurture: Genetic vulnerabilities interact with physical maltreatment to promote conduct problems. Development and Psychopathology 17(01): 67–84.

46.

Kendler

Ohlsson

Sundquist

(2016) The rearing environment and risk for drug abuse: a Swedish national high-risk adopted and not adopted co-sibling control study. Psychological Medicine 46(7): 1359–1366.

47.

Kendler

Walters

Neale

, et al. (1995) The structure of the genetic and environmental risk factors for six major psychiatric disorders in women: phobia, generalized anxiety disorder, panic disorder, bulimia, major depression, and alcoholism. Archives of General Psychiatry 52(5): 374–383.

48.

Kim-Cohen

Caspi

Taylor

, et al. (2006) MAOA, maltreatment, and gene–environment interaction predicting children's mental health: new evidence and a meta-analysis. Molecular Psychiatry 11(10): 903–913.

49.

Klahr

Burt

(2014) Elucidating the etiology of individual differences in parenting: a meta-analysis of behavioral genetic research. Psychological Bulletin 140(2): 544–586.

50.

Leve

Harold

Chamberlain

, et al. (2012) Practitioner review: children in foster care -- vulnerabilities and evidence‐based interventions that promote resilience processes. Journal of Child Psychology and Psychiatry 53(12): 1197–1211.

51.

Leve

Harold

, et al. (2010) Refining intervention targets in family-based research: lessons from quantitative behavioral genetics. Perspectives on Psychological Science 5(5): 516–526.

52.

Leve

Harold

, et al. (2009) Structured parenting of toddlers at high versus low genetic risk: two pathways to child problems. Journal of the American Academy of Child & Adolescent Psychiatry 48(11): 1102–1109.

53.

Leve

Neiderhiser

Harold

, et al. (2017) Naturalistic experimental designs as tools for understanding the role of genes and the environment in prevention research. Prevention Science 19(1): 68–78.

54.

Leve

Neiderhiser

Shaw

, et al. (2013) The early growth and development study: a prospective adoption study of child behavior from birth through middle childhood. Twin Research and Human Genetics 16(1): 412–423.

55.

Luthar

Brown

(2007) Maximizing resilience through diverse levels of inquiry: prevailing paradigms, possibilities, and priorities for the future. Development and Psychopathology 19(3): 931–955.

56.

Mannering

Harold

Leve

, et al. (2011) Longitudinal associations between marital instability and child sleep problems across infancy and toddlerhood in adoptive families. Child Development 82(4): 1252–1266.

57.

Martinez

Jr. and Forgatch

(2001) Preventing problems with boys’ noncompliance: effects of a parent training intervention for divorcing mothers. Journal of Consulting and Clinical Psychology 69(3): 416–428.

58.

Mednick

Gabrielli

Hutchings

(1984) Genetic influences in criminal convictions: evidence from an adoption cohort. Science 224: 891–894.

59.

Moffitt

Caspi

Rutter

(2005) Strategy for investigating interactions between measured genes and measured environments. Archives of General Psychiatry 62: 473–481.

60.

Moffitt

Caspi

Rutter

(2006) Measured gene-environment interactions in psychopathology: concepts, research strategies, and implications for research, intervention, and public understanding of genetics. Perspectives on Psychological Science 1(1): 5–27.

61.

Natsuaki

Shaw

Neiderhiser

, et al. (2014) Raised by depressed parents: Is it an environmental risk? Clinical Child and Family Psychology Review 17(4): 357–367.

62.

Neiderhiser

Marceau

Reiss

(2013) Four factors for the initiation of substance use by young adulthood: a 10-year follow-up twin and sibling study of marital conflict, monitoring, siblings, and peers. Development and Psychopathology 25(01): 133–149.

63.

Neiderhiser

Reiss

Lichtenstein

, et al. (2007) Father‒adolescent relationships and the role of genotype-environment correlation. Journal of Family Psychology 21(4): 560–571.

64.

O’Connor

Deater-Deckard

Fulker

, et al. (1998) Genotype–environment correlations in late childhood and early adolescence: antisocial behavioral problems and coercive parenting. Developmental Psychology 34(5): 970–981.

65.

Pedersen

McClearn

Plomin

, et al. (1991) The Swedish adoption twin study of aging: an update. Acta Geneticae Medicae et Gemellologiae: Twin Research 40(1): 7–20.

66.

Poulton

van Ryzin

Harold

, et al. (2014) Effects of multidimensional treatment foster care on psychotic symptoms in girls. Journal of the American Academy of Child and Adolescent Psychiatry 53(12): 1279–1287.

67.

Price

Chamberlain

Landsverk

Reid

(2009) KEEP foster‐parent training intervention: model description and effectiveness. Child & Family Social Work 14(2): 233–242.

68.

Reiss

Leve

(2007) Genetic expression outside the skin: clues to mechanisms of Genotype x Environment interaction. Development and Psychopathology 19(4): 1005–1027.

69.

Reiss

Leve

Neiderhiser

(2013) How genes and the social environment moderate each other. American Journal of Public Health 103(S1): S111–S121.

70.

Reiss

Neiderhiser

Hetherington

Plomin

(2000) The Relationship Code: Deciphering genetic and social patterns in adolescent development. Cambridge, MA: Harvard University Press.

71.

Rhea

Bricker

Corley

, et al. (2013) Design, utility, and history of the Colorado Adoption Project: examples involving adjustment interactions. Adoption Quarterly 16(1): 17–39.

72.

Rhoades

Leve

Harold

, et al. (2011) Longitudinal pathways from marital hostility to child anger during toddlerhood: genetic susceptibility and indirect effects via harsh parenting. Journal of Family Psychology 25(2): 282–291.

73.

Rhoades

Leve

Harold

, et al. (2012) Marital hostility and child sleep problems: direct and indirect associations via hostile parenting. Journal of Family Psychology 26(4): 488–498.

74.

Rice

Harold

Shelton

Thapar

(2006) Family conflict interacts with genetic liability in predicting childhood and adolescent depression. Journal of the American Academy of Child & Adolescent Psychiatry 45(7): 841–848.

75.

Rutter

(2006) Genes and Behavior: Nature–nurture interplay explained. Malden, MA: Blackwell.

76.

Rutter

Pickles

Murray

Eaves

(2001) Testing hypotheses on specific environmental causal effects on behavior. Psychological Bulletin 127(3): 291–324.

77.

Schermerhorn

D’Onofrio

Turkheimer

(2011) A genetically informed study of associations between family functioning and child psychosocial adjustment. Developmental Psychology 47(3): 707–725.

78.

Selwyn

Wijedasa and Meakings

(2014) Beyond the Adoption Order: Challenges, interventions and adoption disruption. London: Department for Education.

79.

Silberg

Maes

Eaves

(2010) Genetic and environmental influences on the transmission of parental depression to children’s depression and conduct disturbance: an extended Children of Twins study. Journal of Child Psychology and Psychiatry 51(6): 734–744.

80.

Simmonds

(2016) Family placement: continuity and discontinuity over time. In: Barratt

Lobatto

(eds) Surviving and Thriving in Care and Beyond: Personal and professional perspectives. London: Karnac Books, pp. 3–20.

81.

Singh

D’Onofrio

Slutske

, et al. (2011) Parental depression and offspring psychopathology: a Children of Twins study. Psychological Medicine 41(7): 1385–1395.

82.

Sorensen

Price

Stunkard

Schulsinger

(1989) Genetics of obesity in adult adoptees and their biological siblings. British Medical Journal 298: 87–90.

83.

Stover

Connell

Leve

, et al. (2012) Fathering and mothering in the family system: linking marital hostility and aggression in adopted toddlers. Journal of Child Psychology and Psychiatry 53(4): 401–409.

84.

Thapar

Harold

(2014) Editorial perspective: Why is there such a mismatch between traditional heritability estimates and molecular genetic findings for behavioural traits? Journal of Child Psychology and Psychiatry 55(10): 1088–1091.

85.

Thapar

Harold

Rice

, et al. (2007) The contribution of gene-environment interaction to psychopathology. Development and Psychopathology 19(4): 989–1004.

86.

Thapar

Rice

(2006) Twin studies in pediatric depression. Child and Adolescent Psychiatric Clinics 15(4): 869–881.

87.

Thapar

Rice

Hay

, et al. (2009) Prenatal smoking might not cause attention-deficit/hyperactivity disorder: evidence from a novel design. Biological Psychiatry 66(8): 722–727.

88.

The Supreme Court (2013) Judgment: In the Matter of B (A Child) FC. Available at: www.supremecourt.uk/cases/docs/uksc-2013-0022-judgment.pdf

89.

Tolan

Dodge

(2005) Children’s mental health as a primary care and concern: a system for comprehensive support and service. American Psychologist 60: 601–614.

90.

Triseliotis

(2002) Long‐term foster care or adoption? The evidence examined. Child & Family Social Work 7(1): 23–33.

91.

van IJzendoorn

Bakermans-Kranenburg

(2015) Genetic differential susceptibility on trial: meta-analytic support from randomised controlled experiments. Development & Psychopathology 27(1): 151–162.

92.

Webster-Stratton

Herman

(2008) The impact of parent behavior-management training on child depressive symptoms. Journal of Counseling Psychology 55(4): 473–484.

93.

Weisz

Sandler

Durlak

Anton

(2005) Promoting and protecting youth mental health through evidence based prevention and treatment. American Psychologist 60: 628–648.

94.

Westermark

Hansson

Olsson

(2011) Multidimensional treatment foster care (MTFC): results from an independent replication. Journal of Family Therapy 33(1): 20–41.

95.

Wijedasa

Selwyn

(2017) Examining rates and risk factors for post-order adoption disruption in England and Wales through survival analyses. Children and Youth Services Review 83: 179–189.