Originally Authored by Emily J. Helder, Elizabeth Mulder & Marjorie Linder Gunnoe in Child Neuropsychology



Abstract

Most existing research on children adopted internationally has focused on those adopted as infants and toddlers. The current study longitudinally tracked several outcomes, including cognitive, behavioral, emotional, attachment, and family functioning, in 25 children who had been internationally adopted at school age (M = 7.7 years old at adoption, SD = 3.4, range = 4–15 years). We examined the incidence of clinically significant impairments, significant change in outcomes over the three study points, and variables that predicted outcomes over time. Clinically significant impairments in sustained attention, full-scale intelligence, reading, language, executive functioning, externalizing problems, and parenting stress were common, with language and executive functioning impairments present at higher levels in the current study compared with past research focusing on children adopted as infants and toddlers. Over the three study points, significant improvements across most cognitive areas and attachment functioning were observed, though significant worsening in executive functioning and internalizing problems was present. Adoptive family-specific variables, such as greater maternal education, smaller family size, a parenting approach that encouraged age-expected behaviors, home schooling, and being the sole adopted child in the family were associated with greater improvement across several cognitive outcomes. In contrast, decreased parenting stress was predicted by having multiple adopted children and smaller family sizes were associated with greater difficulties with executive functioning. Child-specific variables were also linked to outcomes, with girls displaying worse attachment and poorer cognitive performance and with less time in orphanage care resulting in greater adoption success. implications for future research and clinical applications are discussed.

Keywords: International adoption; Outcome; School age: Longitudinal; Cognitive


Introduction

Internationally adopted children, especially those adopted from orphanage care, are at increased risk for difficulties in a variety of domains compared with nonadopted samples (Juffer et al., 2011). Most studies examining these children find that age at adoption is a strong predictor of outcome, with older ages of adoption associated with greater incidence of difficulties (Julian, 2013). However, the vast majority of research has focused on children adopted during infancy and toddlerhood, thus the need for studies that examine outcomes in children adopted at older ages has been highlighted (McCall, 2011). This is especially relevant given the recent demographic shifts in international adoption in the United States, with approximately 30% of children adopted at age 5 or older (U.S. Department of State, 2013). The current study sought to contribute toward filling this gap in the literature by longitudinally examining children who had been internationally adopted at older ages, tracking a comprehensive range of outcomes including cognitive skills, behavioral and emotional adjustment, attachment, and adoptive family functioning. Child-specific and adoptive-family-specific variables were also assessed in order to examine which factors would predict improvement in outcomes over time.

Outcomes in Internationally Adopted Children

Outcomes among internationally adopted children adopted during infancy and toddlerhood have been extensively described, including studies that examine cognitive outcomes (Scott, Roberts, & Glennen, 2011; van Ijzendoorn, Juffer, & Poelhuis, 2005), emotional and behavioral adjustment (Juffer & van Ijzendoorn, 2005), attachment (i.e., McGoron et al., 2012), and adoptive family functioning (i.e., Judge, 2003).

Research on cognitive outcomes tends to describe significant delays and impairments at the time of adoption (Dalen & Theie, 2012; Jacobs, Miller, & Tirella, 2010; Jacobs et al., 2010; Scott, 2009) with improvement observed over the first several years in the adoptive home (Beckett, Castle, Rutter, & Sonuga-Barke, 2010; Glennen, 2007). Despite generally good recovery of skills when examining international adoptees as a whole, a significant minority of international adoptees tend to display longer lasting cognitive difficulties in several key areas, including intelligence (Beckett et al., 2010), sustained attention (Behen, Helder, Rothermel, Solomon, & Chugani, 2008; Pollak et al., 2010), language (Helder, Behen, Wilson, Muzik, & Chugani, 2013; Scott, 2009), memory (Pollak et al., 2010), and academic skills (Beckett et al., 2007; Miller, Chan, Tirella, & Perrin, 2009).

Research examining emotional and behavioral adjustment has largely focused on externalizing and internalizing difficulties along with behavioral indicators of executive functioning (Juffer & van Ijzendoorn 2005; Merz & McCall, 2010; Merz, McCall, & Groza, 2013; Wiik et al., 2011). Findings from these studies suggest that internationally adopted children demonstrate more externalizing and internalizing difficulties than nonadopted children but in many ways are quite similar in their behavioral and emotional adjustment to children adopted domestically with one exception. Internationally adopted children are more likely than other adoptee groups to display symptoms of attention deficit/hyperactivity disorder (ADHD), with incidence estimates averaging 20–30% depending on study characteristics (Stevens et al., 2009; Wiik et al., 2011). In contrast to cognitive outcomes, several cross-sectional studies have suggested that emotional and behavioral difficulties in international adoptees may actually worsen or become more prominent with longer time in the adoptive home (Gunnar, Van Dulmen, & the International Adoption Project Team, 2007; Merz & McCall, 2010), though this may also be a reflection of greater difficulties as these children enter adolescence.

Social and adoptive family outcomes, such as the quality of the attachment and parenting stress, have also been investigated (reviewed in Bakermans-Kranenburg et al., 2011). Although many internationally adopted children form a secure attachment with their adoptive family, children who have been internationally adopted are at increased risk for insecure attachments and disinhibited attachment behaviors, sometimes referred to as indiscriminate friendliness (Bakermans-Kranenburg et al., 2011). In fact, some studies have found that disinhibited attachment behaviors are quite persistent, even after many years in the adoptive home (Rutter et al., 2007). High levels of parenting stress have also been reported in families who have internationally adopted, compared with what is expected in the general population (Miller et al., 2009) and are often associated with the level of behavioral difficulties in the adopted child (Judge, 2003).

Predictors of Outcomes

Research examining variables that affect outcomes among internationally adopted children have generally found that factors related to the child more strongly and consistently predict outcomes than factors related to the adoptive home environment (Castle, Beckett, Rutter, & Sonuga-Barke, 2010; Kumsta, Rutter, Steven, & Sonuga-Barke, 2010).

Older age at adoption, often highly correlated with longer time spent in orphanage care, is one of the strongest predictors of negative outcome across many studies assessing a variety of outcomes (Colvert et al., 2008; Gunnar et al., 2007, 2012; Jacobs et al., 2010; Pollak et al., 2010; Wiik et al., 2011). Research examining age at adoption suggests that neglect and deprivation that occur during sensitive periods may developmentally program the brain in atypical ways, negatively affecting later development (Julian, 2013). Past research investigating the link between age at adoption and outcomes have suggested a threshold-type relationship with depriving experiences lasting greater than 6 months (Stevens et al., 2008), 18 months (Merz & McCall, 2010), or 24 months (Gunnar et al., 2007) linked to greater risk for difficulties. This variety in thresholds reported is due to the interaction between age at adoption and severity of deprivation, such that more severely depriving circumstances, such as Romanian orphanages (Stevens et al., 2008) create a lower age threshold for difficulties in comparison to settings that are psychosocially depriving but provide for basic care needs (i.e., Merz & McCall, 2010).

Additionally, longer time in the adoptive home (and thus older age at testing) has also been associated with greater emotional and behavioral difficulties in several crosssectional studies (Gunnar et al., 2007; Merz & McCall, 2010). The quality of the preadoptive environment also seems to be a consistent predictor, with poorer quality settings (i.e., orphanages generally of poorer quality than foster care settings) resulting in worse outcomes especially when children spend greater time in such settings (Loman et al., 2013; Merz & McCall, 2010; McDermott, Westerlund, Zeanah, Nelson, & Fox, 2012; Roy & Rutter, 2006).

Other child-specific factors have also been examined, but they less consistently predict outcomes. For example, some studies find that adoption from Russia and/or Eastern European is associated with more significant difficulties (Barcons-Castel, Fornieles-Deu, & Costas-Moragas, 2011; Gunnar et al., 2012; Lindblad, Weitoft, & Hjern, 2010) while other studies do not find an association between the child’s birth country and outcome (Jacobs et al., 2010; Wiik et al., 2011). Some studies find that boys display more difficulty with emotional and behavioral adjustment (Gunnar et al., 2012; Miller et al., 2009; Roy, Rutter, & Pickles, 2004; Sonuga-Barke & Rubia, 2008; Wiik et al., 2011) and other studies report no gender differences in those same outcomes (Barcons-Castel et al., 2011; Colvert et al., 2008; Merz & McCall, 2010).

Child-specific factors that are consistently unrelated to outcomes across many studies are birth weight and other prenatal risk factors (Kreppner, O’Connor, Rutter, & Romanian Adoptees Study Team, 2001; Loman et al., 2013; Merz & McCall, 2010; Miller et al., 2009; Pollak et al., 2010; Roy et al., 2004). This suggests that outcomes in internationally adopted children are more robustly related to their postnatal experiences of deprivation than to prenatal risk factors.

Much less research has been dedicated to examining adoptive-family-specific factors that predict outcomes. Adoptive-family-demographic predictors, such as parental age, education, and socioeconomic status, are often reported to be unrelated or weakly related to outcomes in internationally adopted children (Beckett et al., 2007; Kreppner et al., 2001). Variables related to the number and type of siblings have yielded conflicting results, with some studies reporting no impact of family size, sibling composition (i.e., other adopted children in home versus not), or sibling spacing on outcomes (Castle et al., 2009; Raaska et al., 2013) and others reporting that families with biological children of the adoptive parent are at increased risk for adoption disruption and lower satisfaction (Beckett, Groothues, & O’Connor, 1998, 1999; Boer, Versluis-den Bieman, & Verhulst, 1994). No research currently exists regarding the impact of school type (traditional vs. home schooled) on outcome in internationally adopted children.

A more recent topic in the popular adoption press is the relationship between adoptive parenting approaches and outcomes, though very little systematic research currently exists. One such study found that parenting and the child’s age at adoption significantly interacted, with authoritarian parenting associated with worse behavioral outcomes in children internationally adopted at younger ages but more positive behavioral outcomes for children internationally adopted at older ages (Audet & Le Mare, 2011). Additionally, another study found that high emotional sensitivity and attunement on the part of the adoptive parent may be linked with positive outcomes related to attachment in children internationally adopted as infants and toddlers (Garvin, Tarullo, Van Ryzin, & Gunnar, 2012). No research has been conducted to assess the impact of parenting approaches that encourage regression/dependence behaviors (e.g., placing food directly in the child’s mouth, co-sleeping, etc.) on outcomes such as attachment or emotional and behavioral adjustment. This is especially relevant given that these type of behaviors are often advised in popular press articles and books for adoptive parents (e.g., McCreight, 2002). Current researchers in the field of international adoption have called for a greater emphasis on adoptive family factors as predictors of outcome, especially as related to family structure and attempts made by adoptive families to remediate initial difficulties that their children experience (Juffer et al., 2011; McCall, 2011).

Current Study

The current study builds on this existing literature by longitudinally examining a range of predictors and outcomes in children who have been internationally adopted at the age of 4 years or older. Cognitive outcomes (including intelligence, verbal and visual memory, receptive and expressive language, sustained attention, and academic skills), behavioral and emotional outcomes (including externalizing and internalizing problems as well as behavioral indicators of executive functioning), and social and adoptive family outcomes (including disturbances in attachment, parenting stress, and overall adoption success) were assessed yearly for 3 years. A number of predictors of these outcomes were also measured. These included child-specific factors, such as length of time in orphanage care and gender, as well as adoptive-family factors, including parental education, family size, sibling composition, school type, and initial parenting approach.

The first aim of the study was to examine the incidence of clinically significant difficulties across a variety of outcomes. Comparisons could then be made with previous literature regarding whether difficulties were similarly common in older international adoptees compared to children adopted as infants and toddlers. It was hypothesized that the difficulties with the highest incidence in the current study would be consistent with what has been reported in other studies with internationally adopted children, specifically difficulties with intelligence, attention, language, and externalizing behaviors such as hyperactivity.

The second aim of the study was to examine whether there was significant improvement and/or worsening in outcomes over the 3 years of the study. Based on previous research, it was hypothesized that cognitive outcomes and attachment would show significant improvement across the three study visits. It was also hypothesized that behavioral and emotional difficulties as well as adoptive family outcomes, such as parenting stress and adoption success would either remain consistent or worsen over time. This was based on several cross-sectional studies suggesting greater difficulties in these areas correlating with longer time in the adoptive home (Gunnar et al., 2007; Merz & McCall, 2010).

The final aim of the current study was to identify child and adoptive-family predictors that would be associated with positive and/or negative change in outcomes over time. Based on previous research that highlights age at adoption as a strong predictor of outcome, it was hypothesized that child-specific factors would more consistently and robustly predict change in outcomes than adoptive family factors (e.g., maternal education). Research on adoptive family factors is fairly limited, especially in examining how these factors might be related to cognitive outcomes, thus, adoptive family factors were examined in an exploratory fashion.

Methods

Particpants

Participants were recruited through several local adoption agencies and through local organizations that provide social opportunities and postadoptive support (e.g., classes, training) for families who have adopted internationally. Potential participants who expressed interest in the study underwent phone screening to ensure that they met the following criteria: (a) adopted internationally within the last three years, (b) 4 years or older at the time of adoption, (c) under the age of 18 years old at the beginning of the study, (d) no known evidence of alcohol or drug exposure prenatally based on parent and physician report, (e) no medical condition that might independently affect the child’s performance on cognitive testing (i.e., epilepsy, history of head injury, Down Syndrome).

Twenty-five children met the above criteria and were included in the study. At the beginning of the study, participants’ mean age was 8.93 years (SD = 3.31) and they had been in the adoptive home an average of 15.12 months (SD = 9.68, ranging from 5 to 30 months). The sample included 10 boys and 15 girls, all of whom were right handed. Participants had been adopted from a variety of countries, including China (n = 8), Russia (n = 6), Ethiopia (n = 5), Haiti (n = 2), Ghana (n = 2), Columbia (n = 1), and Ukraine (n = 1) and had a mean age at adoption of 7.67 years (SD = 3.42, ranging from 4–15 years). Regarding preadoptive care, 76% had spent some time with biological parents or relatives (range 1–115 months, median = 49 months), 92% had experienced orphanage care (range 3–145 months, median = 38 months), and 24% had spent time in foster care settings (range 13–103 months, median = 34.5 months).

For all but 2 participants, adoptive families consisted of two-parent homes and 88% of adoptive mothers and 82% of adoptive fathers had college degrees. Including the child enrolled in the study, the number of children in adoptive families ranged from 2–7 (mean = 4.12 ± 1.48). Seventy-six percent of participants had at least one other sibling who had been adopted internationally or domestically (but was not necessarily a biological sibling of participant) while 24% of participants had siblings that were all biological children of the adoptive parent. Eighteen of the participants (72%) were enrolled in traditional school settings while 7 (28%) were home schooled by the adoptive parent(s).

Procedure

Participation in the study involved yearly visits for 3 consecutive years. All children were accompanied to study visits by at least one adoptive parent (n = 24 adoptive mother accompanied participant) and informed consent and assent were obtained. Study visits consisted of three components: a semi-structured interview with the adoptive parent, cognitive testing, and structured rating forms completed by the adoptive parent.

The semi-structured interview gathered information about preadoptive care, medical history, academic placements, use of services, and attachment. Cognitive testing completed with the child included assessment of intelligence, sustained attention, verbal and visual memory, expressive and receptive language, and academic skills. Structured rating forms completed by the adoptive parent were utilized to assess executive functioning, behavioral and emotional adjustment of the child, parenting stress, and the overall success of the adoption. Parent interview, child cognitive testing, and parent-structured ratings were repeated at each of the three study visits, with the exception of preadoptive care information which was collected only at the first study visit. Additionally, a structured rating form for parents that assessed the parenting approach used by adoptive parent during the initial 6 months postadoption (Appendix A) was also administered at the first study visit only.

All 25 participants are included in data presented from Study Visit #1. Twenty-one participants returned for Study Visit #2 (mean time elapsed between Study Visits 1 and 2 = 11.5 months, SD = 1.79 months). Two participants joined the study late and were not yet due for their Study Visit #3 at the time of manuscript preparation. Thus, data for Study Visit #3 includes 19 participants (mean time elapsed between Study Visits #2 and #3 = 12.1 months, SD = 1.5 months). There were no significant differences between those who completed all three study visits and those who did not with regard to any predictor variables, including gender, χ 2 = 2.43, p = .27, time spent in orphanage care, t(23) = 1.03, p = .31, maternal education, t(23) = −0.74, p = .47, sibling structure, χ 2 = 1.50, p = .31, family size, t(23) = 1.19, p = .25, school type, χ 2 = 1.85, p = .24, or parenting approach, t(23) = ‒1.28, p = .21.

Measures

Predictors

All predictor variables were measured at Study Visit #1 through parent report during the semi-structured interview and structured parent-rating forms. Time in adoptive home was operationalized as the months since the child entered the United States at Study Visit #1. Time in orphanage was operationalized as the total number of months that the child spent in orphanage care, which excluded time spent in other contexts (i.e., biological family). Maternal education represents the number of years of formal schooling completed by the adoptive mother. Sibling composition referred to the presence or absence of other adopted children in the family besides the participant (1 = presence of multiple adopted children in the family and 2 = participant in study is only adopted child, with siblings being biological children of adoptive parent). Family size was operationalized as the number of children in the family, including the participant. School type was dichotomous and coded as 1 = traditional school setting (including public & private schooling) and 2 = home schooling.

Parenting approach referred to the degree to which adoptive parents used techniques that encouraged regression/dependence behaviors in their newly adopted children as opposed to techniques that encouraged more age-expected behaviors. This was assessed utilizing a five-item parent-rated instrument that was created for the study (see Appendix A). This measure presented parenting behaviors on a continuum, with the end points of this continuum representing the differing parenting strategies. For example, “Make sure all nice things come only from parents” was contrasted with “Permit child to accept sweets, presents, hugs from relatives, friends, and teachers.” At Study Visit #1, parents were asked to rate their parenting approach during the first 6 months after the child came home on a 5-point continuum for each of the five items, resulting in a possible score range of 5–25. A total score was calculated by adding the ratings across the five items, with lower scores indicating a more regressive/dependence parenting approach (M = 19.08, SD = 4.63). Examination of the scale indicated good internal consistency (Cronbach’s α = .79).

Outcomes

Cognitive Outcomes

Intelligence was operationalized as the full-scale intelligence score at each study visit on the Wechsler Intelligence Scale for Children-IV (WISCIV; Wechsler, 2003). Scores were standardized based on age-based norms (M = 100, SD = 15). Given past research suggesting that the d’ (Detectability) score correlates strongly with parent ratings of attention (Epstein et al., 2003), the T-score for Detectability from the Conner’s Continuous Performance Test-II (CCPT-II; Conners, 2006) was used to operationalize sustained attention. Higher T-scores for Detectability reflect more difficulty distinguishing between targets and nontargets. Conners (2006) provides the following guidelines for interpretation of T-scores: < 40 very good performance, 40–44 good performance, 45–54 within the average range, 55–59 mildly atypical, 60–64 moderately atypical, > 65 markedly atypical. Verbal and visual memory were assessed utilizing the Verbal Memory Index and Visual Memory Index of the Wide Range Assessment of Memory and Learning-II (WRAML-II; Sheslow & Adams, 2003). Scores were standardized based on age-based norms (M = 100, SD = 15). Expressive language was operationalized as performance on the Formulating Sentences subtest of the Clinical Evaluation of Language Fundamentals-IV (CELF-IV; Semel, Wiig, & Secord, 2003) while receptive language was assessed with the Concepts and Following Directions subtest of the CELF-IV. Age-standardized subtests scaled scores were calculated (M = 10, SD = 3) for both subtests. Performance on the Wide Range Achievement Test-4 (WRAT4) Word Reading, Spelling, and Math Computation subtests (Wilkinson & Robertson, 2006) were used to operationalize academic achievement. Age-based standardized scores were utilized (M = 100, SD = 15).

Behavioral & Emotion Outcomes

Parent-rated executive dysfunction was assessed utilizing the Global Executive Composite from the Behavior Rating Inventory of Executive Function (BRIEF; Gioia, Isquith, Guy, & Kenworthy, 2000). This summary score includes clinical scales assessing behavioral regulation and metacognitive aspects of executive functioning. It has a mean of 50 and a standard deviation of 10, with higher scores representing greater executive functioning difficulties. The Internalizing Problems Composite and Externalizing Problems Composite from the Parent Rating Scales of the Behavior Assessment System for Children-2 (BASC-2; Reynolds & Kamphaus, 2004) were used to assess internalizing and externalizing, respectively. The Internalizing Problems Composite is composed of items from the Anxiety, Depression, and Somatization subscales and the Externalizing Problems Composite is composed of items from the Conduct Problems, Aggression, and Hyperactivity subscales. Scores for both scales have a mean of 50 and a standard deviation of 10, with higher scores representing greater problems.

Social & Adoptive Family Outcomes

Three measures of social and adoptive family outcomes were included, two were derived from parent report on structured rating forms and the other was obtained in the semi-structured interview portion of the visit. The presence of attachment disturbances was assessed with the Disturbances of Attachment Interview (DAI; Smyke & Zeanah, 1999), which was administered yearly as a part of the semi-structured interview portion of the study visit. This 13-item semi-structured interview addresses a range of attachment-related behaviors relevant to internationally adopted populations. This includes questions addressing whether the child seeks comfort when distressed, is overly friendly with strangers and displays hypervigilant attachment behaviors. Each item is rated by the interviewer on a scale of 0–2 based on parent response to interview probes and a total score is formed by adding the ratings together for items (possible range 0–26). Higher scores indicate a greater disturbance in attachment relationships. Research using this instrument and comparing it with other attachment interviews has demonstrated the measure to have good reliability and validity (Zeanah, Smyke, & Dumitrescu, 2002), though no inter-rater reliability data is available for the current study.

Frustration and stress surrounding the parent-child relationship were examined through the Relational Frustration Index from the Parent Relationship Questionnaire (PRQ; Kamphaus & Reynolds, 2006). This 12-item scale addresses both overall parenting stress as well as stress related to specific difficult situations (e.g., arguments). The scale has a mean of 50 and a standard deviation of 10, with higher scores representing a greater amount of parenting stress.

Success of the adoption was examined using the Parent-Rated Adoption Success Scale (PASS), a 10-item questionnaire specifically designed for the study (see Appendix B), which both directly and indirectly assessed success. Option choices were presented on a 5-point Likert scale, ranging from “strongly disagree” to “strongly agree” and a total score was calculated by summing responses to all items (resulting in a possible total score range of 10–50), with higher scores indicated greater parent perceived success (see Table 2 for means and standard deviations across study visits). The internal consistency of the scale was good for all 3 years of the study (Cronbach’s α = .87, .87, and .89).

Data Analysis

In order to examine the incidence of clinically significant impairment on cognitive measures, the percent of the sample with scores < 1.5 standard deviations from the normal population mean was examined (i.e., for full-scale intelligence, scores < 78 were considered impaired). This cut-off was chosen in order to allow for comparison with previous research examining incidence of cognitive difficulties in internationally adopted children (Beckett et al., 2010; Behen et al., 2008; Gunnar et al., 2012; Kreppner et al., 2001) and because the manual for the sustained attention measure suggests use of a 1.5 standard deviation cut-off as well (Conners, 2006).

In order to examine the incidence of clinically significant elevations on measures related to emotional and behavioral adjustment and parenting stress, the percent of the sample with scores > 1.5 standard deviations from the mean are also reported (e.g., for internalizing problems, scores > 65 were considered elevated). This choice of cut-off was used based on interpretation guidelines outlined in the manuals for the BRIEF (Gioia et al., 2000), BASC-2 (Reynolds & Kamphaus, 2004), and PRQ (Kamphaus & Reynolds, 2006) and because it allows for comparison with previous literature on international adoptees that utilized similar cut-offs (i.e., Merz & McCall, 2010).

Tests for the significance of a proportion were conducted to examine whether incidence of impairment in the current study differed significantly from the standardization sample of each measure. Also, one sample t-tests were also used to examine whether mean performance on outcomes differed from published norms. Incidence of clinically significant difficulties and comparison between performance and standardization norms were not able to be calculated for attachment disturbances and adoption success, as the measures utilized in the current study did not have published normative data.

One-way repeated-measures analyses of variance (ANOVAs) were utilized to examine change in each outcomes over time. A Mauchly’s test was used to test for violations to sphericity assumptions. The Huynh-Feltd corrected df and p values are reported for tests where significant violations of sphericity were found. When the ANOVA was significant, significant main effects (i.e., change from Study Visit #1 to #2, or from Study Visit #2 to #3) were examined using the Sidak correction for multiple comparisons.

In order to examine which predictors were related to change over time in specific outcomes, multiple regressions were conducted for each outcome with the Study Visit #3 value serving as the dependent variable and predictor variables entered in three blocks. The first block included two covariates, the Study Visit #1 value for the particular outcome and the time in the adoptive home at Study Visit #1. The second block consisted of child-specific factors: time spent in orphanage care and gender. Age at adoption was highly correlated with time spent in orphanage care for the current sample (r = .575, p < .01) but had no significant bivariate correlations with outcomes, thus time in orphanage (rather than age at adoption) was used to avoid the impact of multicollinearity. The third block consisted of adoptive family characteristics: maternal education, sibling composition, family size, school type, and parenting approach.

It is important to note that results from this last set of analyses examining predictors of change over time must be viewed with caution given the small sample size relative to number of predictors (i.e., there is a significant risk of Type II error). However, since relatively few studies have examined adoptive family predictors (especially sibling composition, family size, school type, and parenting approach) and no longitudinal studies have been conducted examining older adoptees, data are presented to guide future research in these areas.

Results

Incidence of Clinically Significant Difficulties

Table 1 presents the percentage of participants falling below cut-offs set for clinically significant impairments in cognition and clinically significant elevations for behavioral, emotional, and adoptive-family outcomes. Percentages that were significantly higher in the current study compared with those reported in the standardization sample are noted.

Across all 3 years of the study, the cognitive impairments that consistently occurred at a rate higher than would be expected were in the areas of full-scale intelligence (44%, 38%, 32%), sustained attention (40%, 30%, 21%), receptive language (68%, 38%, 42%), expressive language (68%, 57%, 32%), and reading (44%, 29%, 21%). Parent-rated executive functioning (16%, 30%, 37%) and externalizing problems (24%, 25%, 32%) were the most common impairments in emotional and behavioral functioning and were present at a higher level than was observed in the standardization sample across study visits.

Table 1. Percent of Sample Scoring in the Clinically Impaired or Elevated Range on Outcomes across the Three Study Visits
Note. *Percentage below cut-off in current sample differs from percentage in standardization sample at p < .05.

Changes in Outcomes Over Time

Table 2 presents means and standard deviations for all outcomes across each study visit. Significant differences between the current sample mean at each study visit and the standardization sample are noted. Significant improvement or worsening over time is also examined.

Table 2. Descriptive Data and Examination of Change in Study Outcomes across the Three Study Visits
Notes. Means sharing a common subscript are not statistically different at p < .05 for post hoc tests.
* Mean performance for study participants differs significantly from standardization sample at p < .05.

Cognitive Outcomes

One-way repeated-measures ANOVAs indicated significant improvement in full-scale intelligence, F(2, 36) = 7.10, p = .003, ηp 2 = .28, verbal memory, F(2, 36) = 6.34, p = .004, visual memory, F(1.5, 27.5) = 3.52, p = .05, receptive language, F(2, 36) = 4.54, p = .02, and reading, F(1.5, 26.5) = 7.4, p = .006, with post hoc comparisons revealing significant improvement when comparing Study Visit #1 to Study Visit #3.

Improvement was also indicated for math skills, F(2, 36) = 4.03, p = .03, with post hoc tests revealing significant improvement from Study Visit #2 to Study Visit #3. Finally, improvement was observed in spelling, F(2, 36) = 13.78, p < .001, with post hoc tests revealing better performance in Study Visits #2 and #3 as compared to Study Visit #1. Statistically significant improvement was not observed for sustained attention, F(2, 36) = 1.67, p = .20, or for expressive language, F(2, 36) = 2.70, p = .08.

Behavioral and Emotional Outcomes

One-way repeated-measures ANOVAs were significant for executive functioning, F(2, 36) = 5.45, p = .009, and internalizing problems, F(2, 36) = 5.39, p = .009, with post hoc tests revealing worsening parent ratings in both areas from Study Visit #1 to Study Visit #2. Post hoc tests did not reveal significant differences between Study Visit #2 and Study Visit #3 for these variables. The one-way repeated-measures ANOVA was not significant for externalizing problems, F(2, 36) = 1.23, p = .30, though it is important to note that, while significant change over time was not observed, externalizing problems were a common area of difficulty across all 3 years of the study, especially in comparison to internalizing problems.

Social and Adoptive Family Outcomes

The one-way repeated-measures ANOVA was significant for attachment disturbances, F(2, 36) = 4.91, p = .01, with post hoc tests revealing a significant reduction in problematic attachment behaviors when comparing Study Visit #1 to Study Visit #3. The ANOVAs for parenting stress, F(2, 36) = 0.55, p = .59, and adoption success, F(2, 36) = 2.76, p = .08, did not indicate a significant change over the course of the study.

Predictors of Change in Outcome Over Time

As was mentioned earlier, the analyses examining predictors of change in outcomes over time were exploratory, given the risk for Type II error associated with the small sample size in the context of nine predictors. Thus, data are presented for both statistically significant and nonsignificant trends identified in the data analysis.

Cognitive Outcomes

Multiple regression analyses were used to examine the impact of child and adoptive family predictors on change in cognitive outcomes over time. The regression examining full-scale intelligence at Study Visit #3 was significant (Table 3) when all three blocks of predictors were included, R2 = .98, F(9, 18) = 40.11, p < .001. After controlling for full-scale IQ at Study Visit #1 and time in adoptive home, higher maternal education (β = 0.46, p < .001), a sibling composition where the participant was the only adopted child (β = 0.36, p < .001), and a parenting approach that encouraged greater age-expected behaviors (β = 0.46, p < .001) significantly predicted greater improvements in full-scale intelligence scores.

The regression analysis examining predictors of change in verbal memory (Table 4) was also significant, R2 = .90, F(9, 18) = 9.35, p = .001, with male gender (β = −0.45, p = .02), greater maternal education (β = 0.32, p = .05), smaller family size (β = −0.37, p = .05), and a parenting approach encouraging age-expected behaviors (β = 0.44, p = .006) serving as significant predictors of greater improvement in verbal memory, after controlling for time in adoptive home and verbal memory performance at Study Visit #1.

The regression analysis examining change for visual memory was significant (data not shown), R2 = .77, F(9, 18) = 3.30, p = .05, with a nonsignificant trend for greater maternal education (β = 0.41, p = .08) predicting greater improvement in visual memory, after controlling for time in adoptive home and Study Visit #1 performance.

The regression analysis examining change in receptive language performance was significant (data not shown), R2 = .87, F(9, 18) = 6.85, p = .004, with a home-schooling school type predicting greater improvement in receptive language performance at Study Visit #3 (β = 0.42, p = .05), over and above control variables.

The regression analysis examining change in reading scores revealed a nonsignificant trend (data not shown), R2 = .87, F(9, 18) = 6.85, p = .10, with a trend for greater maternal education (β = 0.49, p = .06) predicting greater improvement in reading scores, after controlling for time in adoptive home and reading scores at Study Visit #1. The regression analysis examining change in math scores also revealed a nonsignificant trend (data not shown), R2 = .75, F(9, 18) = 2.92, p = .06, with bigger family size predicting (β = −0.61, p = .04) lower math scores, after controlling for time in adoptive home and math scores at Study Visit #1.

Regressions examining change for sustained attention, R2 = .49, F(9, 18) = 0.96, p = .53, expressive language, R2 = .46, F(9, 18) = 0.85, p = .59, and spelling, R2 = .59, F(9, 18) = 1.43, p = .30, were not significant and no statistically significant predictors were identified after controlling for time in adoptive home and performance at Study Visit #1 (data not shown).

Table 3. Multiple Regression Predicting Full-Scale IQ at Study Visit #3
Notes. B = unstandardized coefficient, β = standardized coefficient; Sibling Structure: 1 = other adopted siblings, 2 = siblings are biological children of adopted parent; School Type: 1 = traditional, 2 = home-schooled; Parenting Approach: low scores = regressive/dependence, high scores = age-matched expectations. ƭp < .10. *p < .05. **p < .01. ***p < .001.
Table 4. Multiple Regression Predicting Verbal Memory at Study Visit #3
Notes. B = unstandardized coefficient, β = standardized coefficient; Sibling Structure: 1 = other adopted siblings, 2 = siblings are biological children of adopted parent; School Type: 1 = traditional, 2 = home-schooled; Parenting Approach: low scores = regressive/dependence, high scores = age-matched expectations. ƭ p < .10. *p < .05. **p < .01. ***p < .001.

Behavioral and Emotional Outcomes

The regression examining change in executive functioning was significant (data not shown), R-squared = .77, F(9, 18) = 3.35, p = .04, after controlling for time in adoptive home and Study Visit #1 executive functioning, smaller family size significantly predicted worsening executive functioning ratings (β = ‒1.05, p = .006) and there was a nonsignificant trend for girls to show worsening executive functioning (β = 0.35, p = .10).

The regressions examining change in internalizing problems, R-squared = .80, F(9, 18) = 4.10, p = .02, and externalizing problems, R-squared = .83, F(9, 18) = 4.95, p = .01, were both significant (data not shown). After control variables were taken into account, no predictor variables were statistically significant.

Social and Adoptive Family Outcomes

The regression analysis examining change in attachment disturbances was significant (data not shown), R-squared = .77, F(9, 18) = 3.30, p = .05, with girls displaying significantly worsening attachment disturbances (β = 0.49, p = .04) and a nonsignificant trend for participants residing in families comprised of multiple adopted children displaying more attachment disturbances over time (β = −0.38, p = .09), after control variables were taken into account.

The regression analysis examining change in parenting stress was significant (data not shown), R-squared = .86, F(9, 18) = 6.34, p = .006, with sibling composition characterized by the participant having no other adopted siblings (β = 0.39, p = .04) serving as a predictor worsening parenting stress.

The regression analysis examining change in adoption success ratings was significant (data not shown), R-squared = .87, F(9, 18) = 6.97, p = .004. Less time in orphanage care was a significant predictor of improvements in adoption success (β = −0.34, p = .02) prior to the addition of adoptive family predictors (Block 3) into the model. Additionally, there was a nonsignificant trend for boys to have greater improvement in adoption success ratings (β = −0.33, p = .08) once all predictors and control variables were entered into the model.

Discussion

The current study sought to evaluate a comprehensive range of outcomes over time in children who have been adopted internationally at older ages. This included evaluating the percentage of children scoring in the impaired or clinically elevated range across these outcomes, examining significant changes in the outcomes over a 3- year period, and identifying child and adoptive family variables that predicted this change.

Incidence of Clinically Significant Difficulties

Examination of clinically significant impairments revealed a relatively high incidence of impairments for cognitive outcomes (32–68%, depending on domain) at the first study visit, when children had been in their adoptive home for a little over a year, on average. In contrast to cognitive outcomes, clinically significant difficulties in child behavior and emotional adjustment and parenting stress were less commonly reported at the first study visit (12–24%, depending on domain). High levels of cognitive impairment have been reported by previous studies examining children in the time shortly after international adoption or foster care placement (Dalen & Theie, 2012; Nelson et al., 2007; Scott, 2009), though the current study builds on this evidence by examining cognitive areas not often studied (i.e., verbal and visual memory), by doing direct cognitive testing rather than utilizing parent and teacher report of cognitive domains (Merz et al., 2013; Tan, Loker, Dedrick, & Marfo, 2012), and by limiting the sample to children adopted at older ages.

By the third study visit, participants had been in their adoptive home for an average of three and a half years. Examination of incidence of clinically significant impairments revealed a clearer pattern of strengths and weaknesses, with full-scale intelligence, sustained attention, receptive and expressive language, and reading emerging as common areas of cognitive impairment that were present at a significantly higher level than the standardization sample. Behavioral indicators of executive dysfunction and externalizing problems were the most common behavioral and emotional difficulties and were also present at higher levels than the standardization sample. Parenting stress was also clinically elevated in a larger number of families as compared to what would be expected in the standardization sample of the measure that was utilized.

Intelligence, sustained attention, and academic skills are commonly reported deficits in internationally adopted children and the incidence of participants scoring in the impaired range in the current study is comparable to studies composed of children internationally adopted at younger ages (Beckett et al., 2010; Behen et al., 2008; Gunnar et al., 2012; Kreppner et al., 2001; Merz & McCall, 2010). Specifically, Beckett et al. and Kreppner et al. examined the same sample of children, all of whom were adopted from Romania before 42 months of age, finding impairments in 20–30% of their sample when age at adoption exceeded 6 months of age. Those studies that examined children from less depriving circumstances (i.e., Gunnar et al., 2012) also found similar levels of impairment (28% for attention problems) but at older age cutoffs (adopted after 24 months of age). This consistency between the current study and past literature examining children adopted at younger ages suggests that these deficits are not more common in children internationally adopted at older ages.

In contrast, the high incidence of clinically significant receptive and expressive language difficulties, even by the third study visit, is greater than what is reported in studies (utilizing similar cut-off scores) that examined children internationally adopted before the age of 3 years (Glennen & Masters, 2002; Glennen & Bright, 2005; Jacobs et al., 2010). This may be due to several factors, including children adopted at older ages having a more difficult time acquiring a new language (i.e., Silverberg & Samuel, 2004) or longer duration of exposure to severely depriving preadoption circumstances having a more permanent effect on the development of language regions in the brain (i.e., Helder et al., 2013). At this point, it is unclear whether these language difficulties will be longlasting deficits resistant to improvement or whether older international adoptees just need a longer time to catch-up to age-expected levels of English proficiency as compared to children internationally adopted at younger ages. Additional follow-up of this sample is planned and would directly address this.

Similar to language difficulties, incidence of parent-rated executive dysfunction was still higher at the third study visit in comparison to past research with internationally adopted children examined at younger ages (Jacobs et al., 2010; Merz et al., 2013). For example, Jacobs et al. utilized the same measure as the current study and reported an incidence of 11% with clinically elevated scores in their sample of children internationally adopted before 2 years of age, compared to the 32% in our study. In contrast to executive functioning, incidence of clinically significant externalizing and internalizing problems were fairly consistent with past literature on children internationally adopted at younger ages that used similar measures (i.e., Merz & McCall, 2010; Miller et al., 2009). Merz and McCall presents incidence estimates for their sample that included children who had been as old as 5 years at the time of adoption. Comparison between the portion of their sample adopted above 18 months of age and the current study sample revealed very similar externalizing estimates between studies and slightly lower incidence of clinically significant internalizing problems in the current sample. Regarding parenting stress, comparisons between studies examining children internationally adopted at younger ages and the current study yielded conflicting patterns. Specifically, clinically significant parenting stress was reported to be both higher (51%; Miller et al., 2009) and lower (10%; Judge, 2003) than what was found in the current study. Some of this discrepancy may be due to the use of different measures of parenting stress and/or reporting incidence of difficulties from different subscales even when the same measure was utilized.

To summarize, children adopted internationally at older ages are most at risk for impairments in intelligence, sustained attention, receptive and expressive language, reading, executive dysfunction, and externalizing problems. Additionally, their parents are more likely to have clinically elevated levels of parenting stress compared with the general population. Compared with internationally adopted children who were adopted during infancy and toddlerhood, those adopted at older ages are more likely to have impairments in receptive and expressive language and executive dysfunction. The incidence of impairments in other areas of cognitive, behavioral, and emotional functioning were comparable with past research, suggesting that older international adoptees were not more likely to have impairments than children adopted internationally at younger ages across other domains.

Changes in Outcomes Over Time

As was hypothesized, examination of change in cognitive functioning over the three study visits revealed significant improvements across almost every cognitive performance measured. On average, mean cognitive scores improved roughly three quarters of a standard deviation from the first study visit to the third study visit. This improvement is likely attributed to entry into resource-rich adoptive home and school environments. In contrast to this generally positive change in most cognitive domains, sustained attention did not significantly improve. This is consistent with past research that finds stability in inattention and overactivity symptoms over time (Sonuga-Barke & Rubia, 2008; Stevens et al., 2008). Expressive language skills were also not significantly improved, though they demonstrated a nonsignificant trend towards improvement.

Past studies with children adopted internationally at younger ages have reported improvement over time in cognitive areas such as visual reception (Jacobs et al., 2010), fine motor skills (Dalen & Theie, 2012; Jacobs et al., 2010), communication skills (Dalen & Theie, 2012), and intelligence (Beckett et al., 2010). The current study builds on this past research because it includes direct cognitive testing rather than parent questionnaires (in contrast to Dalen & Theie, 2012), includes children adopted from countries with less severe deprivation (in comparison to Beckett et al., 2010), includes a comprehensive set of cognitive skills and focuses on children adopted at older ages.

In contrast to improvements in cognitive outcomes, worsening of several behavior and emotional outcomes (behavioral indicators of executive dysfunction and internalizing problems) were noted across the three study visits. This is also consistent with hypotheses and builds on past cross-sectional studies that found worse behavioral and emotional functioning correlated with longer time in the adoptive home (i.e., Gunnar et al., 2007; Merz & McCall, 2010) and one longitudinal study that found worsening emotional difficulties over time in children adopted from Romania (Colvert et al., 2008). The lack of significant change over time for externalizing problems is consistent with research that found no decrease in conduct problems for international adoptees over two years in the adoptive home (Kreppner et al., 2001). These findings have important clinical implications as they suggest a need for long-term support for international adoptees and their families as new difficulties may emerge or existing difficulties may worsen as the child spends a longer time in the adoptive home and enters adolescence.

With regard to social and adoptive family outcomes, only attachment demonstrated significant improvement over time. This finding is somewhat in contrast to research supporting persistence of disinhibited attachment behaviors (i.e., indiscriminate friendliness) in international adoptees after adoption (i.e., Rutter et al., 2007). Of note, the current study assessed attachment using a different measure that focused on both disinhibited attachment behaviors as well as other characteristics of attachment (i.e., hypervigilance, expressed interest in back and forth interactions with parent). However, preliminary analyses with the current study data that examined items assessing disinhibited attachment behaviors separately also found continued improvement across study visits. Adoptive family outcomes did not show significant change over time, though there was a nonsignificant trend for worsening in parent’s ratings of adoption success over time. Very little research exists on these outcomes and even fewer longitudinal studies exist for comparison. However, our findings do seem consistent with Castle et al.’s (2009) report that parent’s negative evaluation of their internationally adopted child worsened from ages 6–11. There are several possible explanations for parents’ worsening evaluations of children’s symptomology and the trend toward poorer ratings of adoption success over time. These include an actual increase in children’s symptoms, parents’ improved ability to evaluate the severity of symptoms (as they observe their children in a variety of social contexts), parents’ willingness to report symptoms to the research team, a change in parents’ perception of the severity of the symptoms following the initial adjustment period, or, more likely, some combination of these.

Predictors of Change in Outcomes Over Time

Examination of child and adoptive family predictors of change in outcomes over time revealed several important findings. In contrast to what was hypothesized, both child-specific factors and adoptive-family-specific factors were strong predictors of change in outcomes. Shorter duration of orphanage care was a significant predictor of higher adoption success ratings; however, it did not serve as a significant predictor of any other outcomes. Initially this seems in contrast to research studies that have found age at adoption and/or time in orphanage care to be a robust predictor of outcome in internationally adopted samples (see Julian, 2013, for review). Our finding may be due to the restriction of range in time spent in orphanage care, as children with brief orphanage stays were less likely to be included in the current study due to our focus on children adopted at older ages. Alternatively, this lack of significant findings for duration of deprivation could be viewed as support for studies that have argued for a threshold effect for duration of deprivation, with impairments absent or minimal in children internationally adopted before a certain age, but impairments present for children adopted after a certain age cut-off (Gunnar et al., 2007; Merz & McCall, 2010; Stevens et al., 2008). It may be that once children are adopted over a certain age threshold, the impact of increasing lengths of deprivation does not have an appreciable impact on outcomes. Our study also found that boys had better outcomes in several areas, including significantly better verbal memory and attachment as well as trends for better executive functioning and adoption success. This finding is at odds with other international adoption literature, which tends to report worse outcomes for boys when gender differences are present (Gunnar et al., 2012; Miller et al., 2009; Roy et al., 2004; Sonuga-Barke & Rubia, 2008; Wiik et al., 2011). It may be that girls adopted at school age or later are at increased risk compared with girls adopted in infancy and toddlerhood. This may correspond with research reporting higher incidence of emotional difficulties in nonadopted girls during and after puberty (i.e., Parker & Brotchie, 2010), especially given that this relationship is thought to be mediated by the effects of stress, something that international adoptees often experience at higher levels both before and after adoption.

Adoptive family variables were also consistent predictors of change across outcomes, though the direction of the effect for one predictor (sibling structure) differed for cognitive outcomes versus other outcomes. Specifically, greater maternal education predicted greater improvement in intelligence and verbal memory and trends for greater improvement in visual memory and reading performances. A parenting approach that encouraged age-expected behaviors in the initial months after adoption (as opposed to a more regressive/dependence-encouraging approach) predicted greater improvement in intelligence and verbal memory. Families that home schooled had children with significantly greater improvement in receptive language performance. Smaller family sizes were predictive of greater improvement in cognitive outcomes for verbal memory and math but worsening outcomes for behavioral and emotional outcomes, including executive functioning. Being the only adopted child in the family was associated with greater improvement in intelligence scores and a trend for improved attachment, but these families also reported worsening levels of parenting stress.

The importance of greater maternal education as a predictor of positive outcomes in the current study is consistent with research done by Wiik et al. (2011) that found higher parent education predictive of fewer symptoms of ADHD in internationally adopted children. However, it is in contrast to nonsignificant relationships between parental education and a variety of outcomes within a longitudinal study of children adopted from Romania (Beckett et al., 2007; Kreppner et al., 2001). In terms of severity of deprivation, the current sample is more similar to the sample of Wiik et al. Thus, it may be that maternal education serves as a stronger predictor for children adopted from less depriving circumstances and/or for children adopted at older ages. Also of note, most previous studies have examined maternal education as it relates to behavioral and emotional outcomes rather than to cognitive outcomes, where the current study finds a relationship.

Sibling composition has been previously studied as it relates to adoption success and adoption disruption in both international and domestic adoption samples (i.e., Barth & Brooks, 1997). Most research has found that the presence of other adopted children in the family is associated with higher ratings of adoption success, fewer adoption disruptions, and fewer behavioral difficulties for the internationally adopted child (Barth & Brooks, 1997; Beckett et al., 1998, 1999; Boer et al., 1994). Consistent with this pattern, the current study found that parents with multiple adopted children reported lessening parenting stress over time. However, no research has examined sibling composition as it relates to cognitive outcomes or attachment, where in the current study we find the opposite pattern from parenting stress. It may be that families with one adopted child see a greater disparity in cognitive performance or attachment between their adopted child and biological child(ren), resulting in more resources being allocated to address challenges and higher expectations being communicated to the child. Future research would be needed to replicate this finding and to examine some of the mechanisms by which sibling composition impacts cognitive functioning and attachment over time.

Family size has rarely been examined as a predictor of outcome in international adoption, though one study did find that greater family size was associated with higher parenting stress when assessed six months after international adoption (Viana & Welsh, 2010). Family size has not been previously examined as it relates to cognitive outcomes, and, similar to sibling composition, it may be that smaller families have more time and resources to devote to addressing cognitive challenges. This may explain why smaller family size would be associated with greater improvement in several cognitive outcomes in the current study.

The use of parenting approaches that encourage regression/dependence behaviors in the initial months after adoption to promote attachment and positive outcomes are at times recommended by social workers during the adoption process and are referred to in popular press articles and books for prospective adoptive parents (i.e., McCreight, 2002). However, most of these recommendations have not been evaluated in a systematic way to examine whether they are predictive of attachment or other positive outcomes. The current study found that a parenting approach that encouraged behaviors that were generally age appropriate for the child at the time of adoption (i.e., permitted child to receive nice things from friends and teachers, encouraged participation in settings where child was cared for by nonparents, confronted misbehavior directly) was linked with greater improvements in cognitive outcomes and unrelated to changes in behavior, emotional adjustment, and social/attachment behaviors. This finding suggests that using a parenting approach that encourages more regressive/dependence behavior may not be linked with many of the claimed outcomes, at least not for children adopted at older ages. Although tentative given the small sample size, this is the first research to examine this concept and future research is warranted given the pervasiveness of this type of recommendation to adoptive parents.

The positive outcome associated with use of home-school settings as opposed to traditional schooling should also be followed up by future studies. Given that homeschool environments were only associated with one of the cognitive outcomes and given the small sample size of the current study, it is less certain whether this predictor would be replicated by other samples in future studies.

Strengths and Limitations

The current study adds to the existing literature on outcomes in internationally adopted children for several reasons. First, and foremost, it is the only existing longitudinal study focusing on outcomes in children adopted at older ages. Given that children adopted at older ages are becoming a larger segment of the internationally adopted population, this type of study is especially important for making prognostic judgments in clinical and education settings and in identifying at-risk children and families when allocating intervention resources. Second, due to the longitudinal design and the comprehensive set of assessment tools utilized, this study builds on past research that was either cross-sectional or used parent-report measures to assess cognitive functioning. Third, because participants were adopted from a broad range of countries that provided care that was generally less globally depriving than what was provided in Romanian orphanages, this study allows for the examination of generalizability of findings for some of the best existing longitudinal studies on international adoptees (i.e., English and Romanian Adoptees studies & the Bucharest Early Intervention Project studies). Last, the current study examined several adoptive family factors, such as parenting approach, that have not been previously studied.

Given the desire to conduct highly comprehensive assessments for each child, the resulting small sample size is a significant limitation for the current study. This is especially the case for the third hypothesis that examined predictors of change in outcomes over time, which may have resulted in a lack of significant findings when a predictor truly did have a significant effect. Additionally, the sample was self-selected, which may have resulted in the inclusion of children with greater level of difficulties given that families with concerns may be more likely to participate. However, the current study provides results that future research with larger sample sizes, chosen randomly, can use for hypothesis generation. A second limitation is that there was some variability with regard to length of time in the adoptive home when children joined the study. This was statistically controlled for in some analyses but was not possible for all hypotheses (i.e., examining incidence of clinically significant difficulties). Ideally, the sample would have all entered the study when they had been in their adoptive home for an equivalent amount of time. Some attempt was made to control this as participants were limited to those adopted within the last three years. Additionally, had the sample been identified at the time of adoption, additional data could have been collected regarding the child’s status upon adoption. Finally, the current study assessed behavioral and emotional functioning as well as adoptive family functioning solely through parent report, rather than utilizing self-report or additional raters outside the immediate family (i.e., teachers). Additional follow-up with the current sample is planned and examination of self-report of emotional and behavioral adjustment will be included.

References

Audet, K., & Le Mare, L. (2011). Mitigating effects of the adoptive caregiving environment on inattention/overactivity in children adopted from Romanian orphanages. International Journal of Behavioral Development, 35, 107–115. doi:10.1177/0165025410373313

Bakermans-Kranenburg, M. J., Steele, H., Zeanah, C. H., Muhamedrahimov, R. J., Vorria, P., Dobrova-Krol, N. A., … Gunnar, M. (2011). III. Attachment and emotional development in institutional care: Characteristics and catch-up. Monographs of the Society for Research in Child Development, 76, 62–91. doi:10.1111/j.1540-5834.2011.00628.x

Barcons-Castel, N., Fornieles-Deu, A., & Costas-Moragas, C. (2011). International adoption: Assessment of adaptive and maladaptive behavior of adopted minors in Spain. The Spanish Journal of Psychology, 14, 123–132. doi:10.5209/rev_SJOP.2011.v14.n1.10

Barth, R. P., & Brooks, D. (1997). A longitudinal study of family structure and size and adoption outcomes. Adoption Quarterly, 1, 29–56. doi:10.1300/J145v01n01_03

Beckett, C., Castle, J., Rutter, M., & Sonuga-Barke, E. J. (2010). VI. Institutional deprivation, specific cognitive functions, and scholastic achievement: English and Romanian adoptee (ERA) study findings. Monographs of the Society for Research in Child Development, 75, 125–142. doi:10.1111/j.1540-5834.2010.00553.x

Beckett, C., Groothues, C., & O’Connor, T. (1999). The role of sibling group structure on adoption outcomes. In A. Mullender (Ed.), We are family: Sibling relationships in placement and beyond (pp. 83–95). London: BAAF.

Beckett, C., Groothues, C., O’Connor, T. G., & the ERA Study Team. (1998). Adopting from Romania: The role of siblings in adjustment. Adoption & Fostering, 22, 25–34. doi:10.1177/030857599802200205

Beckett, C., Maughan, B., Rutter, M., Castle, J., Colvert, E., Groothues, C., … Sonuga-Barke, E. J. (2011). Scholastic attainment following severe early institutional deprivation: A study of children adopted from Romania. Journal of Abnormal Child Psychology, 35, 1063–1073. doi:10.1007/s10802-007-9155-y

Behen, M. E., Helder, E., Rothermel, R., Solomon, K., & Chugani, H. T. (2008). Incidence of specific absolute neurocognitive impairment in globally intact children with histories of early severe deprivation. Child Neuropsychology, 14, 453–469. doi:10.1080/09297040802244136

Boer, F., Versluis-den Bieman, H. J., & Verhulst, F. C. (1994). International adoption of children with siblings: Behavioral outcomes. American Journal of Orthopsychiatry, 64, 252–262. doi:10.1037/h0079528

Castle, J., Beckett, C., Rutter, M., & Sonuga-Barke, E. J. (2010). VIII. Postadoption environmental features. Monographs of the Society for Research in Child Development, 75, 167–186. doi:10.1111/j.1540-5834.2010.00555.x

Castle, J., Groothues, C., Beckett, C., Colvert, E., Hawkins, A., Kreppner, J., … Rutter, M. (2009). Parents’ evaluation of adoption success: A follow-up study of intercountry and domestic adoptions. American Journal of Orthopsychiatry, 79, 522–531. doi:10.1037/a0017262

Colvert, E., Rutter, M., Kreppner, J., Beckett, C., Castle, J., Groothues, C., … Sonuga-Barke, E. J. (2008). Do theory of mind and executive function deficits underlie the adverse outcomes associated with profound early deprivation? Findings from the English and Romanian adoptees study. Journal of Abnormal Child Psychology, 36, 1057–1068. doi:10.1007/s10802-008-9232-x

Conners, C. K. (2006). Conners’ continuous performance test. Toronto, ON: Multi-Health Systems.

Dalen, M., & Theie, S. (2012). Internationally adopted children from non-European countries: General development during the first two years in the adoptive family. Scientific World Journal, 2012, 1–9.

Epstein, J. N., Erkanli, A., Conners, C. K., Klaric, J., Costello, J. E., & Angold, A. (2003). Relations between continuous performance test performance measures and ADHD behaviors. Journal of Abnormal Child Psychology, 31, 543–554. doi:10.1023/A:1025405216339

Garvin, M. C., Tarullo, A. R., Van Ryzin, M., & Gunnar, M. R. (2012). Postadoption parenting and socioemotional development in postinstitutionalized children. Development & Psychopathology, 24, 35–48. doi:10.1017/S0954579411000642

Gioia, G. A., Isquith, P. K., Guy, S. C., & Kenworthy, L. (2000). Behavior rating inventory of executive function. Lutz, FL: Psychological Assessment Resources.

Glennen, S. L. (2007). Predicting language outcomes for internationally adopted children. Journal of Speech, Language, and Hearing Research, 50, 529–548. doi:10.1044/1092-4388(2007/036)

Glennen, S. L., & Bright, B. J. (2005). Five years later: Language in school-age internationally adopted children. Seminars in Speech and Language, 26, 86–101. doi:10.1055/s-2005-864219

Glennen, S. L., & Masters, M. (2002). Typical and atypical language development in infants and toddlers adopted from Eastern Europe. American Journal of Speech and Language Pathology, 11, 417–433. doi:10.1044/1058-0360(2002/045)

Gunnar, M. R., VanDulmen, M. H., & the International Adoption Project Team. (2007). Behavior problems in postinstitutionalized internationally adopted children. Development and Psychopathology, 19, 129–148. doi:10.1017/S0954579407070071

Gunnar, M. R., Wenner, J. A., Thomas, K. M., Glatt, C. E., McKenna, M. C., & Clark, A. G. (2012). The brain-derived neurotrophic factor Val66Met polymorphism moderates early deprivation effects on attention problems. Development and Psychopathology, 24, 1215–1223. doi:10.1017/S095457941200065X

Helder, E. J., Behen, M. E., Wilson, B., Muzik, O., & Chugani, H. T. (2013). Language difficulties in children adopted internationally: Neuropsychological and functional neural correlates. Child Neuropsychology. Advance online publication. doi:10.1080/09297049.2013.819846

Jacobs, E., Miller, L. C., & Tirella, L. G. (2010). Developmental and behavioral performance of internationally adopted preschoolers: A pilot study. Child Psychiatry Human Development, 41, 15–29. doi:10.1007/s10578-009-0149-6

Judge, S. (2003). Determinants of parental stress in families adopting children from Eastern Europe. Family Relations, 52, 241–248. doi:10.1111/j.1741-3729.2003.00241.x

Juffer, F., Palacios, J., Le Mare, L., Sonuga-Barke, E., Tieman, W., Bakermans-Kranenburg, M. J., … Verhulst, F. C. (2011). II. Development of adopted children with histories of early adversity. Monographs of the Society for Research in Child Development, 76, 31–61. doi:10.1111/j.1540-5834.2011.00627.x

Juffer, F., & van Ijzendoorn, M. H. (2005). Behavior problems and mental health referrals of international adoptees: A meta-analysis. Journal of the American Medical Association, 293, 2501–2515. doi:10.1001/jama.293.20.2501

Julian, M. M. (2013). Age at adoption from institutional care as a window into the lasting effects of early experiences. Clinical Child and Family Psychology Review, 16, 101–145. doi:10.1007/s10567-013-0130-6

Kamphaus, R. W., & Reynolds, C. R. (2006). Parenting relationship questionnaire. Minneapolis, MN: Pearson.

Kreppner, J. M., O’Connor, T. G., Rutter, M., & the English and Romanian Adoptees Study Team. (2001). Can inattention/overactivity be an institutional deprivation syndrome? Journal of Abnormal Child Psychology, 29, 513–528. doi:10.1023/A:1012229209190

Kumsta, R., Rutter, M., Steven, S., & Sonuga-Barke, E. J. (2010). IX. Risk, causation, mediation, and moderation. Monographs of the Society for Research in Child Development, 75, 187–211. doi:10.1111/j.1540-5834.2010.00556.x

Lindblad, F., Weitoft, G. R., & Hjern, A. (2010). ADHD in international adoptees: A national cohort study. European Child Adolescent Psychiatry, 19, 37–44. doi:10.1007/s00787-009-0038-3

Loman, M. M., Johnson, A. E., Westerlund, A., Pollak, S. D., Nelson, C. A., & Gunnar, M. R. (2013). The effect of early deprivation on executive attention in middle childhood. Journal of Child Psychology and Psychiatry, 54, 37–45. doi:10.1111/j.1469-7610.2012.02602.x

McCall, R. B. (2011). IX. Research, practice, and policy perspectives on issues of children without permanent parental care. Monographs of the Society for Research in Child Development, 76, 223–272. doi:10.1111/j.1540-5834.2011.00634.x

McCreight, B. (2002). Parenting your adopted older child: How to overcome the unique challenges and raise a happy and healthy child. Oakland, CA: New Harbinger Publications.

McDermott, J. M., Westerlund, A., Zeanah, C. H., Nelson, C. A., & Fox, N. A. (2012). Early adversity and neural correlates of executive function: Implications for academic adjustment. Developmental Cognitive Neuroscience, 2, S59–S66. doi:10.1016/j.dcn.2011.09.008

McGoron, L., Gleason, M. M., Smyke, A. T., Drury, S. S., Nelson, C. A., Gregas, M. C., … Zeanah, C. H. (2012). Recovering from early deprivation: Attachment mediates effects of caregiving on psychopathology. Journal of the American Academy of Child & Adolescent Psychiatry, 51, 683–693. doi:10.1016/j.jaac.2012.05.004

Merz, E. C., & McCall, R. B. (2010). Behavior problems in children adopted from psychosocially depriving institutions. Journal of Abnormal Child Psychology, 38, 459–470. doi:10.1007/s10802-009-9383-4

Merz, E. C., McCall, R. B., & Groza, V. (2013). Parent-reported executive functioning in postinstitutionalized children: A follow-up study. Journal of Clinical Child & Adolescent Psychology. Advance online publication. doi:10.1080/15374416.2013.764826

Miller, L., Chan, W., Tirella, L., & Perrin, E. (2009). Outcomes of children adopted from Eastern Europe. International Journal of Behavioral Development, 33, 289–298. doi:10.1177/0165025408098026

Nelson, C. A., Zeanah, C. H., Fox, N. A., Marshall, P. J., Smyke, A. T., & Guthrie, D. (2007). Cognitive recovery in socially deprived young children: The Bucharest Early Intervention Project. Science, 318, 1937–1940. doi:10.1126/science.1143921

Parker, G., & Brotchie, H. (2010). Gender differences in depression. International Review of Psychiatry, 22, 429–436. doi:10.3109/09540261.2010.492391

Pollak, S. D., Nelson, C. A., Schlaak, M. F., Roeber, B. J., Wewerka, S. S., Wiik, K. L., … Gunnar, M. R. (2010). Neurodevelopmental effects of early deprivation in postinstitutionalized children. Child Development, 81, 224–236. doi:10.1111/j.1467-8624.2009.01391.x

Raaska, H., Elovainio, M., Sinkkonenm, J., Stolt, S., Jalonen, I., Matomäki, J., … Lapinleimu, H. (2013). Adopted children’s language difficulties and their relation to symptoms of reactive attachment disorder: FinAdo study. Journal of Applied Developmental Psychology, 34, 152–160. doi:10.1016/j.appdev.2012.12.003

Reynolds, C. R., & Kamphaus, R. W. (2004). Behavior assessment system for children (2nd ed.). Minneapolis, MN: Pearson.

Roy, P., & Rutter, M. (2006). Institutional care: Associations between inattention and early reading performance. Journal of Child Psychology and Psychiatry, 47, 480–487. doi:10.1111/j.1469- 7610.2005.01526.x

Roy, P., Rutter, M., & Pickles, A. (2004). Institutional care: Associations between overactivity and lack of selectivity in social relationships. Journal of Child Psychology and Psychiatry, 45, 866–873. doi:10.1111/j.1469-7610.2004.00278.x

Rutter, M., Colvert, E., Kreppner, J., Beckett, C., Castle, J., Groothues, C., … Sonuga-Barke, E. J. S. (2007). Early adolescent outcomes for institutionally-deprived and non-deprived adoptees. I: Disinhibited attachment. Journal of Child Psychology and Psychiatry, 48, 17–30. doi:10.1111/j.1469-7610.2006.01688.x

Scott, K. A. (2009). Language outcomes of school-aged internationally adopted children: A systematic review of the literature. Topics in Language Disorders, 29, 65–81. doi:10.1097/TLD.0b013e3181977b4c

Scott, K. A., Roberts, J. A., & Glennen, S. (2011). How well do children who are internationally adopted acquire language? A meta-analysis. Journal of Speech, Language, and Hearing Research, 54, 1153–1169. doi:10.1044/1092-4388(2010/10-0075)

Semel, E., Wiig, E. H., & Secord, W. A. (2003). Clinical evaluation of language fundamentals (4th ed.). San Antonio, TX: Pearson.

Sheslow, D., & Adams, W. (2003). Wide range assessment of memory and learning (2nd ed.). Lutz, FL: Psychological Assessment Resources.

Silverberg, S., & Samuel, A. G. (2004). The effect of age of second language acquisition on the representation and processing of second language words. Journal of Memory and Language, 51, 381–398. doi:10.1016/j.jml.2004.05.003

Smyke, A. T., & Zeanah, C. H. (1999). Disturbances of attachment interview. Unpublished manuscript.

Sonuga-Barke, E. J., & Rubia, K. (2008). Inattentive/overactive children with histories of profound institutional deprivation compared with standard ADHD cases: A brief report. Child: Care, Health and Development, 34, 596–602.

Stevens, S. E., Kumsta, R., Kreppner, J. M., Brookes, K. J., Rutter, M., & Sonuga-Barke, E. J. (2009). Dopamine transporter gene polymorphism moderates the effects of severe deprivation on ADHD symptoms: Developmental continuities in gene–environment interplay. American Journal of Medical Genetics Part B: Neuropsychiatric Genetics, 150B, 753–761. doi:10.1002/ajmg.b.31010

Stevens, S. E., Sonuga-Barke, E. J., Kreppner, J. M., Beckett, C., Castle, J., Colvert, E., … Rutter, M. (2008). Inattention/overactivity following early severe institutional deprivation: Presentation and associations in early adolescence. Journal of Abnormal Child Psychology, 36, 385–398. doi:10.1007/s10802-007-9185-5

Tan, T. X., Loker, T., Dedrick, R. F., & Marfo, K. (2012). Second-first language acquisition: Analysis of expressive language skills in a sample of girls adopted from China. Journal of Child Language, 39, 365–382. doi:10.1017/S0305000911000109

U.S. Department of State. (2013). Intercountry adoption statistics. Retrieved from http://adoption.state.gov/about_us/statistics.php

van IJzendoorn, M. H., Juffer, F., & Poelhuis, C. W. (2005). Adoption and cognitive development: A meta-analytic comparison of adopted and non-adopted children’s IQ and school performance. Psychological Bulletin, 131, 301–316. doi:10.1037/0033-2909.131.2.301

Viana, A. G., & Welsh, J. A. (2010). Correlates and predictors of parenting stress among internationally adopting mothers: A longitudinal investigation. International Journal of Behavioral Development, 34, 363–373. doi:10.1177/0165025409339403

Wechsler, D. (2003). Wechsler intelligence scale for children, 4th edition (WISC-IV). San Antonio, TX: Pearson.

Wiik, K. L., Loman, M. M., Van Ryzin, M. J., Armstrong, J. M., Essex, M. J., Pollak, S. D., & Gunnar, M. R. (2011). Behavioral and emotional symptoms of post-institutionalized children in middle childhood. The Journal of Child Psychology and Psychiatry, 52, 56–63. doi:10.1111/j.1469-7610.2010.02294.x

Wilkinson, G. S., & Robertson, G. J. (2006). Wide range achievement test – 4. Lutz, FL: Psychological Assessment Resources.

Zeanah, C. H., Smyke, A. T., & Dumitrescu, A. (2002). Attachment disturbances in young children. II: Indiscriminate behavior and institutional care. Journal of the American Academy of Child and Adolescent Psychiatry, 41, 983–989. doi:10.1097/00004583-200208000-00017

Appendices

Apendix A – Parenting Approach Scale

Instructions: Parenting books on adoption recommend various approaches to help adopted children bond and adjust. Researchers do not know which of these approaches work best. We have paired some of these approaches and represented them as continuums. Please mark only ONE oval on each continuum to describe your approach to parenting the first 6 months your child was with you

Apendix B – Parent-Rated Adoption Success Scale

Leave a Reply