Children and Youth Services Review 91 (2018) 355–363

Contents lists available at ScienceDirect

Children and Youth Services Review

journal homepage: www.elsevier.com/locate/childyouth

Further analysis of the British Chinese Adoption Study (BCAS): Adult life T
events and experiences after international adoption
⁎
Margaret Granta,c, , Alan Rushtonb
a
Adoption and Fostering Alliance Scotland, United Kingdom
b
King's College London, United Kingdom
c
University of Stirling, Scotland, United Kingdom

A B S T R A C T

This paper seeks to contribute to debates about how people's adult lives unfold after experiencing childhood
adversity. It presents analysis from the British Chinese Adoption Study: a mixed methods follow-up study of
women, now aged in their 40s and early 50s, who spent their infant lives in Hong Kong orphanages and were
then adopted by families in the UK in the 1960s.
Sixty-eight women participated via questionnaires and face-to-face interviews. The paper draws on both
quantitative analysis (using standardised measures and systematically coded data on adult life events) and
qualitative analysis of interview data to identify the context and subjective meaning of the quantitative findings.
We found that most of the women lived largely positive, stable, well-supported lives although punctuated by
challenging periods. Using regression analysis, two variables were significantly associated with poorer mid-life
functioning over and above other potential influences: a) feeling unhappy about being adopted, and b) part-
nership adversity after age 25. No associations were found between childhood experiences and patterns of adult
adversity. Illustrations are given, based on the interviews, of the women's multi-faceted perspectives on the long-
term impact of being internationally adopted and on professional support.
We conclude that when early orphanage care (of reasonable quality) was followed by a good quality adoption,
most women were able to negotiate the majority of later difficulties successfully. The findings suggest two
important implications for understanding mid-life outcomes: a) that experiences in both childhood and adult-
hood should be taken into account and b) individuals' subjective views on being internationally adopted may
help explain divergent outcomes within groups with similar early experiences.

1. Introduction child to benefit from one-to-one care attuned to their individual needs.
This form of early adversity differs from others such as child abuse that
This paper examines how women's adult lives unfold after an aty- are commonly used to measure adverse childhood experiences (ACEs,
pical start in life: early separation from birth parents and care in Hong Hughes et al., 2017). The majority of previous research on ex-or-
Kong orphanages, followed by international adoption to the UK. phanage international adoption has been limited to the impact on
Although adoption usually occurs during early childhood, by perma- childhood and adolescence (Juffer & van IJzendoorn, 2005; McCall,
nently altering the child's family relationships it is an intervention with 2011; van IJzendoorn et al., 2011). Those follow-ups that have ex-
lifelong implications. Adoption legislation in the UK reflects the need tended into early adulthood have mostly, but not all, shown a raised
for a lifelong view, for example in England and Wales: ‘the paramount rate of poor mental health for a minority (for example, for contrasting
consideration of the court or adoption agency must be the child's wel- results see Cederblad, Höök, Irhammar, & Mercke, 1999 and Hjern,
fare, throughout his life’ (Adoption and Children Act 2002, part 1, Lindblad, & Vinnerljung, 2002). Mental health problems, such as an-
section 1 (2), emphasis added). For services involved in placing or xiety and mood disorders, have been found in a rare longitudinal study
supporting children and young people in alternative family care, un- to occur de novo in early adulthood, suggesting that the consequences
derstanding how adopted adults look back on their lives can help illu- of childhood adversity can appear many years later (van der Vegt et al.,
minate the long-term impact on individuals of practice and policy. 2009).
Early experience of orphanage care reduces the opportunities for a The more limited number of follow-up studies carried out beyond

⁎
Corresponding author.
E-mail address: [email protected] (M. Grant).

https://doi.org/10.1016/j.childyouth.2018.06.004
Received 25 February 2018; Received in revised form 3 June 2018; Accepted 3 June 2018
Available online 05 June 2018
0190-7409/ © 2018 Published by Elsevier Ltd.
M. Grant, A. Rushton Children and Youth Services Review 91 (2018) 355–363

early adulthood to mid-life have identified increased risks for poor participating in the study, 55 women (76%) were in relationships (no
mental health (Borczyskowski, Hjern, Lindblad, & Vinnerljung, 2006), significant differences were found on comparisons of current marital/
but one study has shown comparable outcomes to non-orphanage cohabiting status with the two NCDS groups) and 51 women (71%) had
samples (Storsbergen, Juffer, van Son, & Hart, 2010). Very poor out- at least one child; the average age at first becoming a parent was
comes have been found for a small minority across most samples. 31 years.
However, differences in sampling, in the extent of exposure to adver- Within the group, differences in pre-adoption experiences were not
sity, in follow-up timings and in outcome measures, combined with the found to be associated with mid-life functioning as variations in dura-
myriad potential influences on individuals over time, leave many tion of orphanage experience were small across the group. Women with
questions open. poorer mid-life functioning were more likely to report poorer quality
Among those studies that have extended to mid-life, most have adoptive parental care and/or more negative feelings about adoption
examined psycho-social outcomes with little data on the intervening (Rushton, Grant, Feast, & Simmonds, 2013). One in four of the BCAS
years, or have focused in depth on important but relatively narrow women sought professional help at some stage for depression, anxiety
areas of adult experiences such as dealing with racism (Grant, Rushton, or similar problems. This, however, was no greater than for the com-
& Simmonds, 2016). This leaves a gap in understanding the influence of parison group women.
other events in adult life. In this retrospective follow-up study of The women mostly recalled their adoptive experience positively:
women in their 40s and 50s, we argue that to gain a long-term per- 86% said they felt loved by their adoptive mother and the same pro-
spective on international adoption requires not only exploring child- portion felt wanted by their adoptive family. This was not true for all
hood experiences and mid-life outcomes but also how people deal with however. One in five women (20%; n = 14) felt unhappy or very un-
challenges and how their perspectives change in adulthood. happy about being adopted. The interview data also identified that
among women who on balance reported their adoptions positively,
1.1. The British Chinese Adoption Study: background to the adoptions struggles related to identity or previous incidents of race-based mis-
treatment were not uncommon. Where women recalled ‘low care’
In the late 1950s and 1960s a large number of migrants from the (Parental Bonding Instrument: Parker, Tupling, & Brown, 1979; Todd,
People's Republic of China entered Hong Kong and numerous children Boyce, Heath and Martin, 1994) by both adoptive parents or feeling
were ‘abandoned’ (perhaps better described as ‘left to be found’). unhappy about being adopted, these factors were statistically asso-
According to contemporaneously recorded files for each child, the ciated with poorer outcome (difference in means = 0.76, 95% CI 1.33
children were admitted to residential institutions, sometimes following to 0.19, p = 0.01; difference in means = 1.2, 95% CI 0.68 to 1.73,
hospital admission. The smallest institution cared for 65 children and p = 0.01 respectively) (Rushton, Grant, Feast, & Simmonds, 2013).
the largest had capacity for 450 children. The conditions were mate- The findings, which overall were more favourable than many would
rially adequate, including medical care, but the diet was restricted and have predicted given their early experiences, led to the consequent
repetitive. Inevitably a lack of personalised care and stimulation from a question of how the women arrived at these outcomes in mid-life. Were
consistent caregiver was lacking and this was likely to be a barrier to adverse adult life events associated with mid-life functioning? What
the development of secure, selective attachments. The orphanage en- proportions had or had not experienced adversity, of what type and
vironment was not regarded as global deprivation according to how did they respond? The current paper presents newly anaylsed data
Gunnar's (2001) rating of children's health, nutrition, stimulation and on adverse life events from age 17 onwards and the women's perspec-
relationship needs in contrast, for example to cohorts such as the tives on adoption and professional support.
English Romanian Adoption Study (Rutter et al., 2007). The lack of pre-
orphanage information on the children's records means that only very 1.3. Theory
rarely was anything known of the pregnancy and birth, or family ge-
netic factors. Many aspects of functioning have been found to be negatively af-
Intercountry adoption was put forward as a solution for children fected by early institutional care (Bos et al., 2011). If caregiving is
who could not be looked after in Hong Kong. Consequently 106 chil- unavailable, inconsistent and insensitive, or even abusive, this is likely
dren were brought to the UK via the Hong Kong Adoption Project of to affect a young child's stress levels which may alter psychological and
International Social Service. The prospective adoptive parents were neuro-biological functioning. In line with the evidence of de novo onset
recruited and assessed by the National Children's Homes (NCH, now of problems in adulthood cited earlier (van der Vegt et al., 2009), ‘la-
Action for Children) or Barnardo's. All were married couples and al- tent vulnerability theory’ suggests that children may react to early
though the majority were white British, a minority had one parent of adverse environments in ways that are potentially adaptive in the short
Chinese heritage. Mean age of the adoptive mothers at the date of term, but create vulnerability to future mental health problems
adoption was 33 years and fathers 35 years. The children were received (McCrory, Gerin, & Viding, 2017).
into socially and materially advantaged homes: mothers identified However, following adversity and when placed in a good adoptive
themselves currently as housewives and fathers' occupations ranged home, various restorative processes may potentially take place due to
from business and professional to skilled workers. the greatly increased availability of nurturing and personalised care and
greater predictability of everyday life (Woodhouse, Miah, & Rutter,
1.2. British Chinese Adoption Study: previous findings 2018). This may support the forming of fresh attachments, building
trust and self-esteem. Beneficial effects may also derive from beyond
The British Chinese Adoption Study was a follow-up into mid-life of the adoptive parents to positive relationships with other family mem-
a sample of girls raised in orphanages in Hong Kong and subsequently bers and the school environment (Brodzinsky, 1990). Conversely, a
transferred in infancy to adoptive family homes in the UK in the 1960s poor quality adoptive environment may compound any risks related to
(Feast, Grant, Rushton, & Simmonds, 2013; Rushton, Grant, Feast, & early adversity.
Simmonds, 2012; Rushton, Grant, Feast, & Simmonds, 2013). The early In examining mid-life functioning, we were influenced by theories
phases of the study focused on the links between childhood experiences of life span development (Sroufe, 2005), especially conceptualisations
and outcomes in mid-life (aged in their 40s and 50s). On measures of of mid-life as a ‘pivotal period’ (Lachman, Teshale, & Agrigoroaei,
psychological adjustment, self-esteem and general physical health, their 2015), indicating that at this life stage the women in our sample could
outcomes were comparable with age-matched non-adopted and do- be expected to have experienced a range of adverse life events: break-
mestically adopted peers drawn from the 1958 UK National Child De- downs of significant relationships, death or ill health of adoptive par-
velopment Study (NCDS; see Elliott & Vaitilingam, 2008). At the time of ents and parenting challenges. Life events research suggests that

356
M. Grant, A. Rushton Children and Youth Services Review 91 (2018) 355–363

exposure to difficult relationships (Brown & Harris, 1989) and stressful 3.1.2. Mid-life functioning
life events (Paykel, 2003) in adulthood are both associated with poorer A composite outcome index of mid-life (current) functioning had
psychological functioning. We questioned whether such experiences previously been constructed based on four continuous measures: life
would lead to worse outcomes among the ex-orphanage women. satisfaction (a 10-point scale from NCDS), self-esteem (Rosenberg,
However, we also appreciated that the vulnerable individual can act 1965), psychological distress (Malaise Inventory; Rutter, Tizard, &
upon the environment. They may learn to avoid stressors and possibly Whitmore, 1970) and state of mental health (GHQ12; Goldberg,
adopt a defensive stance against the challenges and demands of adult McDowell, & Newell, 1996). We standardised the four selected scales,
life (Rushton, 2014). combined them and then standardised the resulting composite to yield a
z-score, with lower scores representing fewer problems (internal con-
sistency = 0.76 Cronbach's alpha).
1.4. The aims of the present study
3.1.3. Feelings about adoption
Substantial research has focused on a risk and resilience model
Following a UK study of adopted adults of varying ages, we selected
following orphanage care and international adoption. Examining pro-
a short series of six questions related to adoption, including whether
cesses of positive adaptation in the face of adversity is as important as
they loved and felt loved by each of their adoptive parents (Triseliotis,
describing trajectories towards maladaptation. Given the relatively
Feast, & Kyle, 2005). Due to high overlap between items, we selected
positive outcomes for this group of internationally adopted women in
the item ‘I feel happy about being adopted’ to represent feelings about
mid-life found previously (Rushton, Grant, Feast, & Simmonds, 2013),
adoption (Feast, Grant, Rushton, & Simmonds, 2013). Responses on a 5-
here we explore the prevalence of adversity in adult life, whether this is
point Likert scale were dichotomised into happy/very happy or un-
associated with childhood experiences and the women's interpretation
happy/ambivalent.
of the impact of their early experiences across the lifespan. We raise
questions about the links between childhood experiences, adult life
3.2. Adult life events
events and mid-life functioning in the context of international adoption
following orphanage care.
3.2.1. Interviews
Our aim here was to select an established method for describing and
2. Method coding adult life events and experiences, including detailed examples of
specific adversities, their level and duration and individual responses to
2.1. Participants such stresses. The Adult Life Phase Interview (ALPHI), derived origin-
ally from a study of adult life events and depression (Bifulco,
The sample was derived from the International Social Service (ISS) Bernazzani, Moran, & Ball, 2000), enabled us to explore events and
records for 100 girls of Chinese heritage raised in orphanages in Hong changes systematically across five domains: partnerships; relationships
Kong and subsequently transferred in infancy to adoptive family homes with adoptive family members and close friends; parenting (where re-
in the UK in the 1960s (Feast, Grant, Rushton, & Simmonds, 2013). levant); education/employment and a general life events domain in-
Sixty-eight women participated via both questionnaire and face-to-face cluding health.
interviews. (Data from the previous analysis reported above includes an The interview starts by establishing the participant's age at major
additional four women who completed questionnaires only, and is changes during adult life: significant moves, start and end of partner-
based on n = 72; all analysis reported below is n = 68). Mean age was ships, birth/adoption of children, or deaths of loved ones. These events
48 years (SD = 2.4) at time of the interviews. are used to define adult life phases and as ‘anchor points’ throughout
The study was approved by the Research Ethics Committee at King's the interview to work out whether other experiences occurred before,
College, London (PNM 08/0927) and written informed consent was during or after these changes. The length and detail of any adversity is
obtained from participants. also explored, for example, how long before and/or after a relationship
breakup the person experienced stress and in what form. Chronic
stressors are recorded systematically by domain for each adult life
3. Measures
phase as well as at points of major transitions. Adversity is rated on a 4-
point scale within each domain (from 1 for marked adversity to 4 for
For this paper, qualitative data and resulting quantitative ratings on
little/no adversity) by trained interviewers using a standardised manual
adult life events (see Section 3.2) were newly analysed, then used in
(www.lifespancollection.org.uk/ALPHI%20measure.html).
conjunction with quantitative scales (Section 3.1) analysed previously.
Qualitative data on feelings about adoption and perspectives on pro-
3.2.2. Ratings and consistency checks
fessional support were also newly analysed (Section 3.3).
Following training with the ALPHI's originator (Bifulco, Bernazzani,
Moran, & Ball, 2000), each interviewer (including the first author of
3.1. Quantitative scales this paper) completed detailed ratings for their own interviews. They
alerted the research team at an early stage to any events/experiences
3.1.1. Orphanage care and adoption that did not fit into a typical pattern. Initial queries were discussed with
Variables related to orphanage care – duration of orphanage care the full research team to ensure consistency in relation to common and
and age at adoptive placement – were based on contemporaneously unusual life events. Each set of ratings was screened on submission by
recorded information in adoption files held by International Social the first author and any queries were resolved via discussion with a
Service UK (now Children and Families Across Borders). second research team interviewer, with reference to the verbatim in-
For adoptive family care, we used the Parental Bonding Instrument terview transcripts. Further information was sought from the original
(PBI; Parker, Tupling, & Brown, 1979), which records parental style interviewer if necessary. After the first author cleaned and entered the
during childhood and adolescence as recollected by the participant. We data, the ratings and queries were reviewed by the second author. The
selected a shortened 7-item version, which correlates highly with the two authors agreed a small number of changes (< 4%) to the original
original scale (r > 0.90; Todd et al., 1994). Following Collishaw et al. interviewer ratings to ensure consistency, in line with the ALPHI
(2007), we used a definition of ‘low care’ from adoptive parents (de- manual. All changes involved increasing the adversity rating, for ex-
fined as any score of 5 or below on the care subscale, range 0–9) to ample to reflect an accumulation of events occurring in quick succes-
dichotomise the results. sion. Table 1 shows examples of marked/moderate adversity and lower-

357
M. Grant, A. Rushton Children and Youth Services Review 91 (2018) 355–363

Table 1
Examples of marked/moderate and lower-level adversity.
Domain Marked/moderate adversity (examples) Some, little or no adversity (examples)

Partnerships:relationship with current partner Abuse; infidelity of partner; conflict or severe tension in Break-up of a relationship without serious distress; shorter
whether or not cohabiting/sexual relationship lasting more than six months; break-up of periods of dissatisfaction in relationship caused by lack of
relationship accompanied by serious distress or unhappiness. intimacy or support.
Social: interpersonal relationships (adoptive Death or serious physical or psychiatric ill-health of family or Less severe tension, relationship distress or effects of loss that
family, friends and ex-partners) close friends; crisis breakdown or severe tension in are temporary and/or resolved without leading to serious
relationships; harassment or serious disputes with ex- distress or unhappiness.
partners.
Parenthood: Birth/adopted children (whether or Infertility or major difficulties related to pregnancy or birth; Difficulties in conceiving, pregnancy or birth that are
not currently living with participant) serious difficulties with parenting or behaviour of child; death temporary and without serious consequences; less severe
or serious ill health of child. distress caused by children's behaviour, ill health or
interaction problems.
Material: Education and work Problems at work or with finding work; difficult work Less severe problems in same categories: at work or with
environment including discrimination; difficulties with finding work; difficult work environment including
finance or housing; struggles with education. discrimination; difficulties with finance or housing; struggles
with education.
Miscellaneous Serious/long term health difficulties or consequences of Health difficulties or consequences of disability that are less
disability. severe and/or short-lived.

level adversity for each domain. 3.4.2. Qualitative analysis

Next, we used the results of the quantitative analysis to identify gaps
3.3. Adult perspectives on early experiences, adoption and professional in the findings for further exploration via qualitative analysis (further
support details described in Section 4.5). The qualitative data enabled us to
investigate the process and context of these results as well as the wo-
In addition to the ALPHI, the semi-structured interviews included men's perspectives on key areas (Palinkas, 2014). We explored adverse
two sections about adoption, including questions on: being inter- events in two domains identified in the quantitative analysis as po-
nationally and transracially adopted, birth family, racism or negative tential influences on psychological functioning. We also examined and
stereotyping, talking about adoption to others, the impact of early ex- give illustrations from the interview data on the women's feelings about
periences, and the types of support adopted children might need. adoption and perspectives on professional support. We used a combi-
Examples include: ‘Can I start by asking, is being adopted something nation of frequency counting, thematic analysis and identification of
you feel comfortable talking about to other people?’; and ‘Can you tell diverse cases to demonstrate the breadth of experiences.
me about the positive aspects of being adopted from Hong Kong?’ and
‘Have there been any difficult aspects related to being adopted from
Hong Kong?’. 4. Results

3.4. Analytic procedures 4.1. Prevalence of adversity in adult life

The study aimed to investigate the relative influence of adult life In the analysis presented in this paper, adversity means marked or
events and experiences on mid-life functioning alongside other child- moderate levels of difficulties or challenges (as per Table 1). We started
hood influences. We used a mixed methods approach, involving two by checking for prevalence of adversity in each domain at any point in
main stages. adult life (Table 2).
Although the majority of the sample did not report major adversity,
elevated rates were found in the social domain (76%). This figure in-
3.4.1. Statistical analysis
cludes adverse life events or experiences related to adoptive family
First, by quantifying the number and severity of adverse adult life
members, friends and ex-partners: in other words, any social relation-
events, we examined whether within-group differences show any as-
ships except with their current partners or children (see Table 1 for
sociation with:
examples).
Based on the work undertaken in preparing and checking the ALPHI
a) recollections of childhood experiences
quantitative ratings, we had detected a potential difference between
b) mid-life functioning
how adversity in early adulthood was reported from later adversity. We
divided the adversity ratings for each domain into difficulties experi-
Binary variables (did versus did not experience marked or moderate
enced between age 17 (or earlier if the women left home earlier) and
adversity) were created from the ALPHI data for adversity in each do-
age 25 – a period defined as ‘emerging adulthood’ (Arnett, 2007) – and
main. Chi-squared and independent samples t-tests (with the unequal
variance assumption where appropriate) were used to test for associa-
Table 2
tions between variables related to orphanage and adoptive family care,
Prevalence of different types of adversity reported by adopted women across all
adverse adult life events and mid-life functioning. On the basis of these adult life phases.
tests, potential independent variables were chosen for entry in a linear
multivariable regression model. (An inclusive cut-off univariate p-value Domain Marked or moderate adversity None, little or mild adversity

of 0.07 was used at this stage in order to avoid missing any important n (%) n (%)
relationships in the multivariable model). The residuals from the fitted
model were examined for approximate normality and, as a sensitivity Partnerships 26 (38) 42 (62)
analysis, the regression was refitted with the ‘robust’ standard errors Social 52 (76) 16 (24)
Parenthood 19 (28) 49 (72)
option in Stata; this allows for unmet distributional assumptions. All
Material 12 (18) 56 (82)
statistical analyses were conducted using Stata version 14 (StataCorp, Miscellaneous 21 (31) 47 (69)
2015).

358
M. Grant, A. Rushton Children and Youth Services Review 91 (2018) 355–363

Table 3 between our adoption-related variables and the presence of adversity in

Prevalence of different types of moderate or marked adversity reported by early or mid-adulthood across each of the five domains.
adopted women in early adulthood and mid-adulthood.
Domain Marked or moderate Marked or moderate 4.3. Adult adversity and mid-life psychological functioning
adversity in early adversity in mid-
adulthood adulthood For each adult life domain, we used independent sample t-tests to
check whether marked/moderate adversity at any time, in early
n (%) n (%)
adulthood or in mid-adulthood were associated with scores on our mid-
Partnerships 7 (10) 19 (28) life psychological functioning index.
Social (adoptive family, 16 (24) 49 (72)
friends and ex-
4.3.1. Partnership domain
partners)
Parenthood 1 (1) 19 (28)
First we tested whether women who experienced partnership ad-
Material (education and 4 (6) 8 (12) versity at any point had poorer current well-being, and found no sig-
work) nificant association. Next we used our early adulthood and mid-adult-
Miscellaneous 6 (9) 18 (26) hood adversity variables. Although early adulthood partner adversity
did not reach conventional levels of statistical significance, there was
some evidence that more recent difficulties appeared to be associated
later. We created new binary variables for early adulthood (up to and
with outcome (b = 0.52, 95% CI −0.01 to 1.05, p = 0.053).
including age 25) and mid-adulthood (age 26 and over) for each do-
main. In addition, we considered the potential for testing the effects of
4.3.2. Social domain
proximal adversity (i.e. recent events, in the past year or so) but found
Ratings in the social domain in ALPHI take into account adversity
too few cases across the group for testing. Table 3 shows the patterns
related to (adoptive) family members, friends and ex-partners. In con-
across domains.
trast to the partner adversity results, in the social domain there was
The higher level of adversity experienced in mid-adulthood com-
some evidence that early adulthood adversity appeared to be associated
pared to earlier adulthood may be explained by two factors. First, mid-
with mid-life functioning (b = 0.55, 95% CI −0.02 to 1.11, p = 0.057)
adulthood covered a much longer period: on average 22 years (as mean
while adversity in mid-adulthood was not found to be related.
age of participants was 48 years). Second, some forms of adversity were
more likely to occur later: for example, ill health or death of parents,
4.3.3. Parenthood, material and miscellaneous domains
the breakdown of partnerships or marriages, or difficulties associated
Variables on adversity in each of the remaining three domains
with fertility or parenthood.
(parenthood, material and miscellaneous) were also tested. No sig-
nificant associations with mid-life functioning were identified. For
4.2. Early experiences and adult adversity parenthood, early adulthood adversity was discounted due to having
only one case by age 25.
Having assessed the extent of adult adversity in the sample, we then
explored whether childhood experiences were associated with patterns 4.4. Regression analyses: early experiences, adult events and mid-life
of adult adversity in each domain. We had two sources of data: vari- functioning
ables related to orphanage care (extracted from the original ISS files)
and variables related to growing up in their adoptive families (based on Of the five adult life events domains, only social adversity in early
questionnaires completed by participants prior to the interviews). adulthood and partnership adversity in mid-adulthood produced a
trend for statistical significance with mid-life functioning. Our next step
4.2.1. Orphanage care was to examine together all associations identified from the data ex-
We tested whether the orphanage variables (duration and age at exit tracted from adoption files, the questionnaire pack data and the inter-
in days) were associated with later adversity, using our binary adversity view data.
variables for early adulthood and mid-adulthood for each domain. No We had previously found that low care from adoptive parents, based
significant results at p < 0.05 were found using independent samples t- on the 7-item Parental Bonding Instrument scores (Collishaw et al.,
tests. 2007; Parker, Tupling, & Brown, 1979), was associated with poorer
This finding followed the pattern of our previous analysis, which functioning in mid-life. Participants with ‘low care’ scores reported
had identified no associations between age at adoption or duration of their adoptive parents' care during childhood and adolescence as
orphanage care against a range of adult mental or physical health lacking warmth, understanding and/or acceptance. Some parents' atti-
outcomes (Rushton, Grant, Feast, & Simmonds, 2013). One likely ex- tudes or behaviour were recalled as denigrating or humiliating the
planation is that lack of variation in duration of orphanage care child. In addition, feeling unhappy about being adopted was also as-
(compared to other studies) weakened the probability of finding an sociated with poorer functioning. No evidence was found that or-
effect. The majority of our sample were adopted close to the age of two phanage care variables (duration of care, age at entry or exit from or-
(Rushton, Grant, Feast, & Simmonds, 2013). phanage care and quality of orphanage care) predicted current
functioning in mid-life.
4.2.2. Adoptive care and feelings about adoption In sum, across all the analyses, four binary independent variables
Then we tested whether a) feeling unhappy about being adopted or held up as individually associated with outcome: low adoptive care,
b) low parental care during childhood were associated with later ad- feeling unhappy about being adopted, social adversity in early adult-
versity. Our previous analysis had identified that these two variables hood and partner adversity in mid-adulthood. These four variables were
were associated with psychological functioning in mid-life (Rushton, entered into a regression analysis, using our psychological functioning
Grant, Feast, & Simmonds, 2013). Among the 68 participants in the index as the (continuous) dependent outcome variable. Results are
current analysis, 13 women (19%) reported low care from adoptive provided in Table 4.
parents (i.e. care lacking in warmth, understanding and acceptance) Of the four independent variables, the significance held for two
and 14 women (21%) reported feeling unhappy about being adopted. associations with outcome: feeling unhappy (versus happy/ambivalent)
Eight women (12% of the sample) reported both. about being adopted (b = 1.02, 95% CI 0.43 to 1.61, p = 0.001) and
Using chi-squared tests, no significant associations were found partnership adversity since age 26 (b = 0.48, 95% CI 0.01 to 0.95,

359
M. Grant, A. Rushton Children and Youth Services Review 91 (2018) 355–363

Table 4 4.5.1.2. Social relationships. The ALPHI ‘social domain’ includes

Regression results for mid-life functioning. relationships with family members (parents and siblings) and friends.
Coefficient 95% CI P By separating out the ratings, we found that 51 women reported family-
related adversity, of whom 18 women also reported friends-related
Low care from adoptive parents 0.26 −0.33 to 0.86 0.378 adversity.
Feeling unhappy to be adopted 1.02 0.43 to 1.61 0.001
We identified two main patterns in the women's descriptions. The
Social adversity in early adulthood 0.30 0.21 to 0.81 0.243
Partner adversity in mid-adulthood 0.48 0.01 to 0.95 0.043
first was a reduction over time in family relationship functioning
stressors, reflecting either improved relationships or cutting-off of high-
N = 68; R-squared = 0.35; range of mid-life functioning index = −1.2 to 3.7 stress relationships. In the other direction, as participants entered mid-
with higher numbers indicating poorer functioning. life there was an increase in adversity in the form of losing parents or
dealing with other family/close friends' illnesses or health difficulties.
p = 0.043). Low care from adoptive parents and social adversity by age After removing all ratings based on bereavement or ill health of
25 were no longer significantly associated with outcome in this model. others, we found that the overall level of adversity reduced sub-
The histogram of residuals from the model gave no indication of stantially: family-related adversity decreased from n = 51 to n = 34
major departure from normality and re-fitting the regression with ro- and friends-related adversity decreased from n = 19 to n = 14.
bust option, the p-value for feeling unhappy about adoption remained
0.01 (b = 1.02, 95% CI 0.41 to 1.64) and p-value for partner adversity 4.5.2. Feelings about adoption
in mid-adulthood increased to 0.1 (b = 0.48, 95% CI −0.9 to 1.05). In our regression analysis, the women's feelings about adoption
remained associated with outcome, even after adult life events and
experiences were taken into account. By coding the interview data,
4.5. Qualitative analysis
three key themes were identified that related to the women's feelings
about adoption.
The regression analysis had identified two particular areas for fur-
First, although the majority of women indicated via the quantitative
ther exploration via the qualitative data: 1) relationships with partners
variable (see 3.1) that they viewed their adoptions positively or very
and/or family and friends in adulthood; and 2) feelings about adoption.
positively, there was a wide range of viewpoints and attendant emo-
To help understand the process of recovery from adverse events, we
tional responses. The diversity was much more notable in examining
added a third area: perspectives on professional support.
the narrative data than in looking at quantitative variables in isolation.
We approached the qualitative analysis with the following questions
Women who had, on the surface, similar experiences differed in whe-
in mind. First, in the ‘partnership’ and ‘social’ domains, what type of
ther they viewed their adoption largely positively, negatively or with
adverse events occurred, and how severe were these events? We used
some ambivalence.
the qualitative data from the relevant sections of our semi-structured
Second, feelings about adoption were multi-faceted, and related to a
interviews and the interviewers' quantitative ALPHI ratings of ‘sub-ca-
large number of factors. These included relationships with adoptive
tegories’ of adversity. We integrated and summarised the data from
parents, the alternative life the person imagined if they had not been
these two sources, with a focus on documenting frequencies of shared
adopted, the lack of information about their birth parents (and there-
experiences across the group.
fore lack of opportunities to search for and meet with birth family
Second, what contributed to the women's ‘feelings about adoption’?
members), the loss of connection with Hong Kong, the perceived ben-
For this question, we analysed data from two further sections of the
efits of being raised in the UK, or any combinations of these factors.
interview where the women reflected on various aspects of being in-
Each participant's feelings about adoption involved a process of
ternationally adopted. As this question required a more exploratory
weighing up such factors in relation to each other.
method, we used an inductive approach to code the data and develop
One participant description neatly captured the contradictions that
themes reflecting influences on feelings about adoption.
make forming a singular view difficult:
Third, in the context of relatively positive mid-life outcomes, what
role (if any) did professional support play? For this question, we se- Sometimes I do wonder what life would have been like [if I wasn't
lected a sub-sample of three women who had and three women who adopted], would it have been better? But then you sort of say to
had not sought professional help as examples to illustrate divergent yourself, well rather than saying that you have got to think on the
experiences. This method was in line with our aim of investigating positive side. I feel very eternally grateful to my parents for having
processes of adaptation as well as maladaptation. got me out of that home, but then I think to myself, why do I have to
By using the full range of data available, the qualitative analysis feel this feeling that I have got to be very grateful?
helped to provide context for a more rounded interpretation and ela-
Third, over time, the women in our study had experienced a number
boration of the quantitative findings (Fetters, Curry, & Creswell, 2013).
of important transitions, not least from dependent child of their adop-
tive parents to adults with independent lives and perspectives. At this
4.5.1. Adverse events in adult life: types and severity point in mid-life, their views on adoption were influenced by a range of
4.5.1.1. Partnerships. In total, 26 women experienced marked or other experiences, including, for many, having children of their own, as
moderate partner-related adversity lasting a minimum of six months well as the events related to partners and other relationships described
(Table 1), while 42 women reported no such experiences. Within the in the previous section.
former group, shorter-term adversity tended to reflect divorces or When asked about the positive and negative aspects of being
relationship breakdowns following overt conflict, sometimes adopted, another participant described life as a ‘learning curve’ and
involving disputes over children or finances, or infidelity. Six women concluded:
described longer-term problems in interacting with former partners,
You change – what you will think at one stage in your life, when you
mostly in relation to parenting arrangements.
look back on it you get it all into perspective, so…what I might have
The narrative data indicated that support from family and friends
thought was negative when I was younger I see as positive now,
helped women to manage challenges following the breakdown of long
'cause I think well, Mum and Dad had difficulties parenting, I've had
term or meaningful partnerships. There were a small number of ex-
difficulties parenting, none of us are perfect. It's not a negative I
ceptions, including two women who reported severe tension and con-
don't suppose, but it's just life isn't it?
flict with partners lasting several years. Both had voluntarily sought
mental health care. Feelings about adoption were often fluid. For some, troubling

360
M. Grant, A. Rushton Children and Youth Services Review 91 (2018) 355–363

questions had arisen or persisted in adulthood, related to birth family/

[C]ounselling is bloody brilliant for me! I know people talk about it,
Hong Kong, feelings about growing up in their adoptive family, or a
pooh-pooh it, and say ‘oh come and see me, pay me this much
combination of both. Conversely, other women who reported having
money and I'll talk to you’, but they don't understand the complexity
struggled during childhood, adolescence or early adulthood had come
of it, and I think you have to be tuned into it.
to a more positive view over time.
The third participant had received extensive previous professional
4.5.3. Perspectives on professional support support for poor mental health over a number of years, including par-
We explored individual decisions about and experiences of seeking enting support and, more recently, attending an adoption support
professional help, such as counselling, therapy or specialist adoption group. Discussing her current feelings and whether she would ever seek
support services. We selected six contrasting examples that highlight access to her adoption records, she concluded:
the broad range of perspectives within the group. Three women were
It's hard to describe to you really – I've basically been there, done
from among the 75% who reported they had not sought professional
that – I've closed the safe door on it now and it's just locked away
support; three women were from among the 25% who reported they
and I'm not interested in it anymore.
had sought professional support for problems such as anxiety or de-
pression. Looking beyond these six examples, completely contrasting views
Among the three women who had not sought professional help, the were not uncommon. For example, one woman felt that a counsellor
first participant had never felt concerned about her own mental health who was not adopted could never understand her, while another de-
but did speculate on whether a early lack of consistent, one-to-one scribed her frustration with professionals' assumptions that any later
parental care left a legacy. She thought, generally, people with such difficulties stemmed from her adoption. Although only a quarter of the
experience might need support as a result. She explained: women had sought professional help, there was a broad consensus
across the group that a) support should be available when needed and
I do think that having children or you look at other people's children
b) the need might arise at any point, not only during childhood.
you do wonder about the first 15 months of the child's life not
Potential triggers to seek support included fraught relationships with
having much of a family home - not attaching to a mother and all
adoptive family members and emotions related to their own experi-
that kind of thing so I wonder if that's made me a slightly detached
ences of being a parent, but also non adoption-specific experiences:
person in some ways or able to detach from people rather too easily.
divorce, bereavement, problems at work or in friendships.
A second participant described her attitude towards her adoption as
quite mixed, although her life in adulthood was settled and for the most
part enjoyable. She reflected on the experience of growing up visibly 5. Discussion
and transracially adopted, and how her perspective had changed since
childhood: What has been learned about adult life events and experiences in
this sample of internationally adopted women? Our mixed methods
[As a child] I often wished I didn't look Chinese and as I got older I
analysis identified three main findings. First, adversity in adulthood
just accepted it I suppose. … I didn't want to be Michael Jackson -
mostly occurred in the relationship domains, including relationship
you just accept yourself I think. I think that's my nature, I'm a very
functioning and dealing with death/ill health of family and friends.
contented person and I'm not curious. I accept things, that's why I
Second, although there was not widespread evidence of difficulties in
think I managed to get on so well, if you like.
forming or maintaining partnerships, adversity in this domain after age
In contrast, the third woman in this group stressed that her ap- 25 was associated with poorer mid-life functioning. Finally, feelings
proach had always been to view any difficulties in life as unrelated to about adoption were multi-faceted and dynamic, and negative feelings
being adopted. about adoption remained associated with poorer mid-life functioning
after other potential influences were taken into account.
I haven't ever felt it to be a problem because the way I've been
brought up - I've never had any problems and I've never, ever
thought problems have been to do with my adoption. I've totally
5.1. Adversity in adulthood
separated it … I thought anything that's happened it's just been
because of the life I am leading not because of adoption.
Both social adversity in early adulthood (ages 17–25 years) and
Among the three women who had sought professional support, all re- partner adversity in mid-adulthood (ages 26 – present) showed some
ported that experiences during childhood were a catalyst for seeking evidence of association with poorer current functioning as individual
support, but in different ways. The first participant had suffered from independent variables, but only partner adversity remained significant
depression in early adulthood, which she described as linked to her in our regression model. Although relationship problems with adoptive
feelings about being abandoned. With her adoptive parents' help, she parents and friends in early adulthood were reported as stressful, more
had sought professional counselling, and she had no subsequent ex- recent adversity in partnerships appeared to have a greater impact on
periences of poor mental health. While she still wondered about her functioning in mid-life. This trend may reflect a general shift from
origins, this was more ‘curiosity’ than an on-going preoccupation: growing up in one family to establishing a new family (of whatever
form) in adulthood, rather than a specific pattern for this adopted
As I am getting older I keep thinking … you know…when somebody
group.
says you are getting like your mother, physically, I think I don't
Our findings highlight that the influence of experiences in adult-
know what my mum would have looked like and what I will be like.
hood, as well as childhood, should be considered when examining mid-
… So every now and again I think that through, I guess in a light
life outcomes for internationally adopted adults. This is in line with
hearted way.
broader work on lifespan development that emphasises that psycholo-
The second participant recalled that her adoptive parents had gical growth in adulthood can compensate for the negative effects of
struggled to manage their own emotions and that this affected their childhood adversity on well-being in later life (Landes, Ardelt, Vaillant,
parenting abilities, although she was quick to point out that her (non- & Waldinger, 2014). More specifically, in relation to children who have
adopted) siblings also shared this perspective. Having sought counsel- spent time in orphanage care, it adds to the evidence of changes con-
ling during two particularly stressful periods in her personal and work tinuing well beyond childhood (Woodhouse, Miah, & Rutter, 2018).
life, she concluded:

361
M. Grant, A. Rushton Children and Youth Services Review 91 (2018) 355–363

5.2. Pre-adoption and adoptive experiences who did not; no issues of concern were identified (Feast, Grant,
Rushton, & Simmonds, 2013; Rushton, Grant, Feast, & Simmonds,
As with other studies of international adoption, we considered the 2013).
potential influence of orphanage care and/or age at adoption, but found A mid-life adoption follow-up is rare and a relatively high partici-
no association with prevalence of adversity in adult life. This adds to pation rate was achieved over a 50-year follow-up period (68% for this
our earlier analysis which found no association between differences in analysis). However, the reasonably small sample size restricts the
orphanage experiences and mid-life functioning (Rushton, Grant, Feast, ability to detect possibly important differences in statistical tests (Type
& Simmonds, 2013). One likely explanation is that lack of variation in II error).
orphanage care (compared to other studies, e.g. Rutter et al., 2007) Although it is usually quite difficult to generalise from qualitative
weakened the probability of finding an effect. The majority of our work, the data from these face-to-face interviews adds depth to un-
sample were adopted close to the age of two (mean age 23 months, SD derstanding. By combining methods, we were able to test for statistical
14 months, range 8 to 82 months). associations and then further explore the context and subjective
While low care from adoptive parents and feeling unhappy about meaning of our initial results.
being adopted were both found to be independently associated with As with all studies that rely primarily on one main source of data
poor mid-life functioning, only the latter remained significant in the and retrospective assessment of experiences, direction of effect and
final regression model. In other words, those women who reported low causality cannot be definitively determined. Moreover, the findings
care, yet on overall reflection at this point in time felt happy (or at least cannot necessarily be generalised to males or to children who have
not unhappy) about being adopted did not show raised risk for poor experienced more seriously depriving orphanage care, and only with
functioning in mid-life in this model. It is not possible to determine caution to other international adoption samples.
direction of causality, but we found previously that women who re-
ported low parental care gave specific and detailed descriptions of 6. Conclusions
hostile and negative parental behavior in their interviews (as opposed
to the broader items in the Parental Bonding Instrument on warmth, These data help to adjudicate on questions of life span experiences
acceptance and understanding). Our view, therefore, is that it was not and the effect on long-term outcomes by exploring the lives of adult
simply a factor of their current mood driving their feelings about women of Chinese heritage, spent largely in the UK in the past five
adoption. decades following early orphanage care in Hong Kong. The follow-up
A conclusion from our previous analysis was that a good adoption data did not identify numerous chains of negative experiences: many of
was associated with positive outcomes in most cases, whereas an un- the women lived largely positive, stable, well-supported lives punc-
satisfactory or unhappy adoption was associated with poorer mid-life tuated by challenging periods. However, the regression analysis showed
functioning. The current analysis has provided a more nuanced per- that feeling unhappy about being adopted and partnership adversity
spective. The findings suggest that when examining outcomes, along- since age 26 were both significantly associated with poorer mid-life
side externally measurable variables such as age at adoption, people's functioning over and above other potential influences. Our best inter-
perceptions of their adoptive experiences should be taken into account. pretation is that when early orphanage care is of reasonable quality,
and followed by a good (or mainly good) adoptive experience, adopted
5.2.1. Implications for practice women are able to negotiate the majority of adult life events success-
There has been a long-standing concern about the risks for long- fully and to draw on the social and financial resources available to them
term mental health problems for children who were disadvantaged in in adult life.
their early years. A central conclusion of the study is that, as a group, These findings underline that to understand long-term experiences
these women had not suffered excessive risk for serious adverse con- following international adoption, it is important to consider not only
sequences. This echoes the findings for women in a Dutch study of in- childhood experiences but events in adulthood too. In addition, ex-
ternationally adopted adults who experienced early orphanage care in ploring individuals' subjective views on being internationally adopted
Greece, also described as not severely depriving (Storsbergen, Juffer, may help explain divergent outcomes within groups with similar early
van Son, & Hart, 2010). experiences.
The qualitative analysis underlined the value some women placed
on counselling and psychological support. Help to understand and come Acknowledgements
to terms with negative experiences, specifically within the context of
international and transracial adoption, should be made available. The Sir Halley Stewart Trust funded the work undertaken for this
Counsellors need to take into account, when support is requested, the paper, following a grant from the Nuffield Foundation (grant number
ways people have found to negotiate some of the complexities related to CPF/36096) which funded the British Chinese Adoption Study. Julia
their early experiences and adoption using their own strengths and Feast OBE and Dr. John Simmonds OBE were co-contributers to the
personal support networks. British Chinese Adoption Study. Many thanks to our statistical advisor
Although the majority of the women had not sought professional Dr. Morven Leese and members of our Advisory Group.
support, many had questioned how their pre-adoption and adoptive
experiences might have affected them. An alternative model to ther- Conflict of interest
apeutic interventions is the provision of peer support via networks or
groups that offer the opportunity to meet other people who share the We wish to confirm that there are no known conflicts of interest
experience of being internationally adopted. Since the completion of associated with this publication.
this research, participants from this study, alongside others also
adopted from Hong Kong, have been active in setting up their own References
network.
Arnett, J. J. (2007). Emerging Adulthood: What is it, and what is it good for? Child
5.2.2. Strengths and limitations Development Perspectives, 1(2), 68–73.
Bifulco, A., Bernazzani, O., Moran, P. M., & Ball, C. (2000). Lifetime stressors and re-
The sample was fairly homogenous in that the girls were born on a current depression: Preliminary findings of the Adult Life Phase Interview (ALPhI).
similar date, shared the same location in their early lives and had a Social Psychiatry and Psychiatric Epidemiology, 35, 264–275.
limited span of ages at adoptive placement into the UK. Baseline data von Borczyskowski, A., Hjern, A., Lindblad, F., & Vinnerljung, B. (2006). Suicidal beha-
viour in national and international adult adoptees: A Swedish cohort study. Social
were compared for women who participated in this study and women

362
M. Grant, A. Rushton Children and Youth Services Review 91 (2018) 355–363

Psychiatry and Psychiatric Epidemiology, 41(2), 95–102. referrals of international adoptees: A meta-analysis. JAMA, 293(20), 2501–2515.
Bos, K., Zeanah, C. H., Fox, N. A., Drury, S. S., McLaughlin, K. A., & Nelson, C. A. (2011). Lachman, M. E., Teshale, S., & Agrigoroaei, S. (2015). Mid-life as a pivotal period in the
Psychiatric outcomes in young children with a history of institutionalization. Harvard life course balancing growth and decline at the crossroads of youth and old age.
Review of Psychiatry, 19(1), 15–24. International Journal of Behavioral Development, 39(1), 20–31.
Brodzinsky, D. M. (1990). A stress and coping model of adoption adjustment. In D. M. Landes, S. D., Ardelt, M., Vaillant, G. E., & Waldinger, R. J. (2014). Childhood adversity,
Brodzinsky, & M. D. Schechter (Eds.). The psychology of adoption (pp. 3–24). New midlife generativity, and later life well-being. The Journals of Gerontology. Series B,
York: Oxford University Press. Psychological Sciences and Social Sciences, 69(6), 942–952.
Brown, G. W., & Harris, T. O. (1989). Depression. In S. M. Monroe, K. Harkness, A. McCall, R. B. (2011). IX. Research, practice, and policy perspectives on issues of children
Simons, & M. E. Thase (Vol. Eds.), Life stress and symptoms of major depression. (189). without permanent parental care. Monographs of the Society for Research in Child
(pp. 168–175). The Journal of Nervous and Mental Disease. Development, 76(4), 223–272.
Cederblad, M., Höök, B., Irhammar, M., & Mercke, A. (1999). Mental health in interna- McCrory, E., Gerin, M., & Viding, E. (2017). Annual Research Review: Childhood mal-
tional adoptees as teenagers and young adults. An epidemiological study. Journal of treatment, latent vulnerability and the shift to preventive psychiatry – The con-
Child Psychology and Psychiatry, 40(8), 1239–1248. tribution of functional brain imaging. Journal of Child Psychology and Psychiatry,
Collishaw, S., Pickles, A., Messer, J., Rutter, M., Shearer, C., & Maughan, B. (2007). 58(4), 338–357.
Resilience to adult psychopathology following childhood maltreatment: Evidence Palinkas, L. A. (2014). Qualitative and mixed methods in mental health services and
from a community sample. Child Abuse and Neglect, 31, 211–229. implementation research. Journal of Clinical Child & Adolescent Psychology, 43(6),
van der Vegt, E. J., Tieman, W., van der Ende, J., Ferdinand, R. F., Verhulst, F. C., & 851–861.
Tiemeier, H. (2009). Impact of early childhood adversities on adult psychiatric dis- Parker, G., Tupling, H., & Brown, L. (1979). A parental bonding instrument. British
orders. Social Psychiatry and Psychiatric Epidemiology, 44(9), 724–731. Journal of Medical Psychology, 52, 1–10.
Elliott, J., & Vaitilingam, R. (2008). Now we are 50; key findings from the National Child Paykel, E. S. (2003). Life events and affective disorders. Acta Psychiatrica Scandinavica,
Development Study. A summary report. Centre for Longitudinal Studies. Institute of 108(s418), 61–66.
Education. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton
Feast, J., Grant, M., Rushton, A., & Simmonds, J. (2013). The British Chinese Adoption University Press.
Study: Planning a study of lifecourse and outcomes. European Journal of Social Work, Rushton, A. (2014). Early years adversity, adoption and adulthood: Conceptualising long-
16(3), 344–359. term outcomes. Adoption and Fostering, 38(4), 374–385.
Fetters, M. D., Curry, L. A., & Creswell, J. W. (2013). Achieving integration in mixed Rushton, A., Grant, M., Feast, J., & Simmonds, J. (2012). Assessing community con-
methods designs—Principles and practices. Health Services Research, 48(6pt2), nectedness and self-regard in a mid-life follow-up of British Chinese adoptions.
2134–2156. Adoption and Fostering, 36(3), 62–72.
Goldberg, D., McDowell, I., & Newell, C. (1996). General health questionnaire (GHQ), Rushton, A., Grant, M., Feast, J., & Simmonds, J. (2013). The British Chinese Adoption
measuring health: A guide to rating scales and questionnaires (2nd edn). New York: Study: Orphanage care, adoption and mid-life outcomes. Journal of Child Psychology
Oxford University Press. and Psychiatry, 54(11), 1215–1222.
Grant, M., Rushton, R., & Simmonds, J. (2016). Is early experience destiny? Review of Rutter, M., Beckett, C., Castle, J., Colvert, E., Kreppner, J., Mehta, M., & Sonuga-Barke, E.
research on long-term outcomes following international adoption with special re- (2007). Effects of profound early institutional deprivation: An overview of findings
ference to the British Chinese Adoption Study. The Scientific World Journal, 20 (16 from a UK longitudinal study of Romanian adoptees. The European Journal of
pages. ISSN: 2356-6140). Developmental Psychology, 4(3), 332–350.
Gunnar, M. R. (2001). Effects of early deprivation. In C. A. Nelson, & M. Luciana (Vol. Rutter, M., Tizard, J., & Whitmore, K. (1970). Education, health and behaviour. Malaise
Eds.), Handbook of developmental cognitive neuroscience. 2001. Handbook of develop- inventory. London: Longmans.
mental cognitive neuroscience (pp. 617–629). Cambridge, MA: MIT Press. Sroufe, L. A. (2005). Attachment and development: A prospective, longitudinal study
Hjern, A., Lindblad, F., & Vinnerljung, B. (2002). Suicide, psychiatric illness and social from birth to adulthood. Attachment & Human Development, 7(4), 349–367.
maladjustment in intercountry adoptees in Sweden: A cohort study. The Lancet, StataCorp (2015). Stata statistical software: Release 14. College Station, TX: StataCorp LP.
360(9331), 443–448. Storsbergen, H. E., Juffer, F., van Son, M. J. M., & Hart, H. T. (2010). Internationally
Hughes, K., Bellis, M. A., Hardcastle, K. A., Sethi, D., Butchart, A., Mikton, C., ... Dunne, adopted adults who did not suffer severe early deprivation: The role of appraisal of
M. P. (2017). The effect of multiple adverse childhood experiences on health: A adoption. Children and Youth Services Review, 32(2), 191–197.
systematic review and meta-analysis. The Lancet Public Health, 2(8), e356-e366. Todd, A. L., Boyce, P. M., Heath, A. C., & Martin, N. G. (1994). Shortened versions of the
van IJzendoorn M. H., Palacios J., Sonuga-Barke E. J. S., Gunnar M. R., Vorria P., McCall Interpersonal Sensitivity Measure, Parental Bonding Instrument and Intimate Bond
R. B., Le Mare L., Bakermans-Kranenburg M. J., Dobrova-Krol N. A. and Juffer F. Measure. Personality and Individual Differences, 16(2), 323–329.
(2011) Monographs of the society for research in child development, Vol. 76, 4, pages Triseliotis, J., Feast, J., & Kyle, F. (2005). The adoption triangle revisited. London: BAAF.
8–30. Woodhouse, S., Miah, A., & Rutter, M. (2018). A new look at the supposed risks of early
Juffer, F., & van IJzendoorn, M. H. (2005). Behaviour problems and mental health institutional rearing. Psychological Medicine, 48, 1), 1–10.

363