Title: How do Children Fare in Care: A Longitudinal Study of Outcomes in Foster Care Family Permanence and
1How do Children Fare in Care A Longitudinal
Study of Outcomes in Foster CareFamily
Permanence and Foster Care Session 22
- Dr Elizabeth Fernandez
- School of Social Sciences and International
Relations - University of New South Wales
- Australia
- E.Fernandez_at_unsw.edu.au
- International Society for Child Indicators
Inaugural Conference - Chapin Hall Center for Children
- Chicago, Illinois
- June 26-28, 2007
2Childrens Conceptions of Fostering
- What is a foster home? "Places of refuge where
people can stay, where you get looked after.
Would you call this home a foster home? "No,
this is my house. (009, male, 11 years old) - Would you call yourself a foster child? "Not
really, I wouldn't call myself one cause my carer
treats me as her own son, even though she's
already got one and she's already got a daughter
but she treats me as her own son. And I really
adore her. - No a normal kid like everyone else. Because here
it is like a real family. Some parents dont care
about children, thats why I came into foster
care. (005, male, 12 years old)
3Childrens Conceptions of Fostering (contd)
- A person who acts like your mum and dad, I
havent got my mum and dad or my brother or the
pets that I had before. Thats why its not the
same. Youre in somebody elses house and its
not your real mum and dad but its the person
thats looking after you for the moment.
(female, 8 years old) - Sort of because they miss their mum and they
really want to go back to her, and they wont be
able to see her for a long, long time, so they
act differently because of this. But once they
settle in they get fine, and then um they just
forget about it and start moving on. (male, 9
years old)
4The impact of the care experience on childrens
wellbeing
- Children come to the foster family setting as an
already at risk group - In relation to children's wellbeing in care
research has identified a range of concerns
including - Instability in care placements (Barber
Delfabbro, 2003 Fernandez, 1999 Pecora,
Williams, Kesler Herings, 2003 Ryan and Testa,
2004) - Inability of care systems to ensure optimal
educational outcomes (Jackson, 2000Dobel-Ober,
Lawrence, Berridge Sinclair, 2003 Rosenfeld
Richman, 2003 Zetlin, Weinberg Kimm, 2003)
5Previous Research Highlights that
- Children in care are significantly more likely to
exhibit psychological problems than children in
the general population (Anderson et al. 2004
Clausen et al. 1998 Barber Delfabbro, 2003
Minnis Devine, 2001 Meltzer et al. 2003). - Evidence of indications of emotional and
behavioural concerns in the clinical range
(McMillen et al. 2004 Callaghan et al. 2003
Garland et al. 2003 Stahmer et al. 2005 - Children who experience multiple changes in
caregivers tend to develop elevated emotional and
behavioural problems (Stanley et al. 2005) - Children with externalising behaviours are
particularly vulnerable to placement breakdown
(Stijker et al. 2005 Newton et al. 2000 Mc
Auley Trew, 2000). - Children leaving the care system are
disproportionately likely to experience
homelessness, unemployment, depression and become
involved in problematic activity (Courtney
Dworsky, 2006 Stein, 1994). -
6Previous Research Highlights (cont)
- Studies undertaken internationally point to the
educational deficits which children bring to the
care experience (Jackson, 2001 Harker et al.
2004 Mitic Rimer, 2002 Zetlin et al. 2003) - Low educational attainment, poor attendance,
overrepresentation in school exclusion, frequent
school changes ,low completion rates and high
unemployment among those who age out of the
system are noted (Blome, 1997 Harker et al.
2004 Jackson, 2001 Pecora et al. 2006 Stein,
1994) - Limited supports are available in schools to
respond to the emotional and behavioural
difficulties children present (Elliot, 2002
Remsbery, 2003 )
7The impact of the care experience on childrens
wellbeing
- Current research points to limitations of cross
sectional studies in capturing developmental
sequences - Increasing recognition of the need to give a
central place to the voices of children in
research and practice (Gilligan, 2002 Newman,
2003) - Limited research that views outcome from
different participants in the foster care process
(Courtney, 2000 Kelly Gilligan, 2002)
8The Research Aims
- To document the needs and experiences of children
in care from the perspective of their carers,
case workers, birth parents and children
themselves - To analyse the perceived adjustment and
psychosocial functioning of children over the
study period and document placement and
developmental outcomes - To establish how the children and their carers
experience social work support, and the impact of
professional input on outcomes - To explore childrens perceptions of their
developing relationships with foster families,
and their established relationships with their
birth family and significant others
9Data Collection
- Interviews were carried out at 4 months after
entry to care and 18-24 month intervals
thereafter - Child interviews (8 18 yrs)
- Caseworkers (of children of all ages)
- Foster/adoptive carers (of children of all ages)
- Birth parents (of children of all ages)
10Interviews Explored
- Conceptions of fostering
- Childs placement history
- Reasons for entry to care
- Relationship with the carer and foster siblings
- Relationship with the birth family
- Schooling experience
- Physical health
- Emotional and behavioural development
- Identity and self image
- Relationship with caseworker and agency
11Data Collection (contd)
- Measures used in the study
- Looking after children AAR subscales (completed
by caseworkers and children) - Achenbach CBCL (completed by carers)
- Achenbach TRF (completed by teachers)
- Hare self esteem scale (completed by children)
- Interpersonal parent and peer attachment scale
(completed by children) - Attachment styles questionnaire (completed by
carers) - Foster care alliance scale (completed by children
and carers) - Strengths and Difficulties Questionnaire
(completed by children)
12About the Children
- 59 children participated in the study
- Boys 52
- Girls 48
- Ages ranged from 3 to 15
- 12 years was the most frequently occurring age
- Children are from Barnardos Find-a-Family Centre,
an integrated service for permanent family care
and adoption for hard to place children
requiring long-term placement. Many have
multiple failed placements prior to their
Find-a-Family placement and, almost none have
adoption care plans on entry
13Primary Reasons for Care
- Factors related to the Birth Mothers
- Drug or alcohol dependence (29)
- Mental health 22 cases
- Physical abuse and neglect 36 of cases
- Factors related to the Birth Father
- Abandonment 39 of cases.
- Physical or sexual abuse or neglectful parenting
39 - Domestic violence (22)
- Child factors
- Emotional and behavioural factors (25) including
antisocial behaviour or severe emotional
disturbance such as depression
14Care History
- Respondents have been in care for 8.2 years on
average and have been in Barnardos care for the
majority of that time (6.3 years). Respondents
have had on average 4.4 placements in total. - Males have spent longer in care on average (9
years) than have females (7.4 years) - The majority (71) of children are in non
relative foster families and 19 are adopted
15Change of Placement and Childrens Responses
-
- Many of the children interviewed had multiple
carers over time. Childrens placements ranged
from 2 to 7 foster homes, an irony that is
presented by one of the children when asked why
they thought they were in care - Ah, cause we've moved around a lot (male,
11yrs) - Every kid should get a prize for having the
courage to move. Something like a hundred
dollars (male,13 yrs) - Below another child advise around an issue that
many of the children have experienced, that is
being moved because of challenging behaviours. - Dont chuck temper tantrums. And that's it(
male,13 yrs) -
16Change of Placement and Childrens Responses
(cont)
- Many children were aware that they would
eventually find a permanent foster placement,
even though they were not sure how long their
present placement would last - (SIGH) well if I am very very, extremely good I
might stay here and this might be my forever
family but if um, if this isnt a good place I
will have to move, which I dont want to
(female, 8 yrs) - (Until) I'm old enough to move out into a flat
(female, 11yrs)
17Change of Schools
- Three quarters of the children had experienced
at least one - change in schooling since their separation from
their birth family - More than half of the children had had three or
more changes - Heaps, probably about 5 or 6 times. I think I
get stupider every time I have to move
(female,14 yrs) - Ive been to thousands of schools...about, 5 or
6. I don't know (male, 11yrs)) - When asked to evaluate how they were doing at
school, most children attempted to assess their
own abilities. - can't hardly readand plus I'm year 5 going in
year 6can't even hardly read or do neat
writing (female, 10 yrs) - Um, playing and English. I'm not so good at my
maths (female, 11yrs) - Hand writing everything. Not everything in the
world thoughI'm good at mostly everything
(female, 8yrs) - Q. what are you like compared to your
classmates? - Um, middle of the class. I mean, we got graded
and it goes up to A's and I'm in the(middle)
(male, 12yrs) -
18(No Transcript)
19Foster Parent Cohesion
- Forty-eight per cent of respondents indicated
they got on 'very well' both with their foster
mother and their foster father. - All but one respondent indicated that they got on
with their foster mother very well or quite
well - Almost 9 out of 10 respondents were positive
about their relationship with their foster
father, rating 'very well' or 'quite well - Just over 1 in 10 respondents indicated that they
got on with their foster father 'not very well' - Eighty-six per cent of respondents were positive
about their relationship with their foster
sibling - However nearly 1 in 10 respondents indicated that
they got on with their foster sibling 'not very
well'
20Childrens perceptions of cohesion
- The relationships with the foster mothers
remained very positive, especially amongst boys
and younger children - Children who had a stronger level of maternal
attachment were more likely to sustain highly
cohesive relationships within the foster families
- The higher the cohesion with the foster mother
the higher the cohesion with the foster father
(r0.37, n40, p0.021) - Age was significantly related to cohesion with
the foster father (r0.5, p0.01) such that older
children were less likely to report getting on
very well with the foster father -
21Childrens perceptions of cohesion (contd)
- The childs cohesion with other children from the
foster family, was significantly related to the
childs number of placements - Children who got on very well with the children
of the foster family had significantly fewer
placements than children who did not get on very
well (p 0.018)
22Cohesion (cont)
- She's (female carer), understanding, she's nice.
She helps me with lots of things. Shes just a
very kind person she's got a nice heartsometimes
I give a hug to (carer) before I go to bedI
don't know what's the most important thing. When
I go shopping with her, just me and her by
ourselves we just talk you know. I like how we
have time by ourselves sometimes (female, 17yrs) - She, umm, she'll spend lots of time on me and
she's really nice, andshe helps me with things
when I need help...she always has the right
advice to tell me...cause they treat me like Im
part of the family so I think I am (female,
10yrs)
23(No Transcript)
24Frequency of Contact
- Birth Mothers and siblings were the most
frequently contacted family members - One child in 5 had contact with his or her birth
mother at least fortnightly - Nearly three-quarters of children (72) saw their
birth mother at least once every 3 months - A quarter had no contact at all
- Just over half (56) of the children had no
contact at all with their birth father - 28 saw their father between once a month and
every few months or holidays
25Current Contact with Family of Origin
- Q. How do you feel when you see her? (Birth
Mother) - HappyMmm, I dunno I just have this feelingMmm,
nice, mmm happy, mmm that's about it (006, male,
11yrs) - Um, the fact that I'm happy (007, female,
11yrs) - Im always missing my mum. It doesn't happen
that much now, cause I see her every month
(male, 13yrs) - I ask her a lot of questions, I ask her what was
she like when she was little, what was I like
when I was littleJust to see herShe realises
what she's done and she tried to change but
that's her (female, 17yrs) - Most of the time she doesn't turn upShe might
turn up every second time.She's heaps nice. And
she never laid a hand on me (male, 14yrs)
26Current Contact with Family of Origin (cont)
- Compared to Interview 1, the only significant
change was an increase in childrens desire to
see their fathers - Many of the children expressed that they never
see their birth fathers, they did however appear
to be interested in seeing them and establishing
a connection - I'd like to see him (father) a lot more, heaps
and heaps and heaps more times, it makes me feel
happy(female, 8yrs) - I dont have a real dad, I never did. I only
have false dads (female, 8yrs) - Ive never had a first dad (male, 11yrs)
27Childrens Self Esteem
- Childrens self esteem was assessed using the
Hare self esteem scale. Includes peer
self-esteem, home self esteem and school
self-esteem and a total score - Girls and boys both had an average of 82
- Peer self esteem was negatively correlated with
total number of placements, (r -0.42, p0.05) so
that the more placements children had the lower
their peer self esteem - Age at entry to care was also found to be related
to global self esteem (r0.37, p0.05). That
is, children who went into care at an older age
had higher self esteem at interview 2
28Hare Self Esteem Scores, including gender
breakdown
- Girls were found to have remained stable from
Interview 2 to Interview 3 on all the subscales
and the total self esteem score - Boys however had significantly higher home self
esteem scores and total self esteem scores at
Interview 3 compared to interview 2. - This finding is encouraging given the small
sample sizes and indicates that boys responded
positively to the foster home environment.
29Self esteem and Childrens care history
- From the childrens interviews it was apparent
that being in care affected their self esteem.
However, the children did also compare themselves
to their peers for some reassurance. - Its like we're second hand kids unless that's
how all kids feel who are my age (female,
12yrs) - When I see my friends with their parents I see
nothing different...it just seems the same, like
Ive got play stations and Nintendos, and being
allowed to play and going to friends houses as
well (male, 13yrs) - Some people in my class don't even have a dad.
And I get lots of stuff (female, 10yrs)
30Looking After Children Assessment Action
Records
31(No Transcript)
32Relationship skills
- At Interview 3, 11 skills were reported by more
than 75 per cent of the children, an increase of
six skills from Interview 1 and four from
Interview 2 - At Interview 3 more children were reporting
feelings of trust, confidence amongst their peers
and less demanding with their carers - Several behaviours did not improve very much from
Interview 2 to 3 including comfort others who
are upset, considerate of others feelings,
not getting into fights, sharing, ability to
make friends
33LAC Scores and Cohesion
- A positive correlation exists between the number
of reported relationship skills and the level of
cohesion with the foster mother (r0.42, n42,
p0.006) and the foster father (r0.38, n42,
p0.014). - Cohesion also relates to some LAC subscales
- The greater the foster mother cohesion the fewer
relationship problems with carers (Carer
Subscale r-0.41, n42, p0.006) - The greater the foster father cohesion the fewer
relationship problems with carers (Carer
Subscale r-0.36, n42, p0.020), the fewer the
conduct problems - The nature of the relationship with the foster
father at interview 1 appears to have had an
important developmental influence on the
children, so if there was very good cohesion the
child increased their relationship skills by
interview 2 3
34(No Transcript)
35(No Transcript)
36Anxiety problemsChildren self assessment
- Worrying a lot was the most frequently reported
symptom over a third of the children reported
this - Nearly a quarter of children reported
anxiety-related somatic complaints - Females appeared to have more anxiety problems on
average than did males (t(40)-1.95, n40,
p0.058) - Just under a third of children experienced
multiple anxiety symptoms at the time of the
third interview - Improvements at interview 3 were in relation to
reduced feelings of misery and sadness, fears and
sleeping difficulties
37Child Behaviour Checklist
- In the present study the CBCL 4-18 was used
- This is an observational measure for children
aged 4 to 18 (Achenbach 1991) which assesses 113
problem behaviours to provide information on 3
overall problem scores - Internalising Problems inhibited or
over-controlled behaviour (I, II and III) - Externalising Problems antisocial or
under-controlled behaviour e.g., delinquency or
aggression (IV and V) - Total Problems Scale all mental health problems
reported by parents or adolescents
38Child Behaviour Checklist (contd)
- 8 further subscales
- 1.Somatic Complaints chronic physical complaints
without known cause or medically verified basis. - 2.Withdrawn shyness and social isolation.
- 3.Anxious/Depressed feelings of loneliness,
sadness, being unloved, worthlessness, anxiety
and general fears. - 4.Delinquent Behaviour breaking rules and norms
set by parents and communities (e.g., lying,
swearing, stealing or truancy). - 5.Attention Problems difficulty concentrating
and sitting still, and impaired school
performance. - 6.Aggressive Behaviour bullying, teasing, temper
tantrums and fighting. - 7.Social Problems impaired peer relationships.
- 8.Thought Problems strange behaviour or ideas,
obsessions.
39Children aged 4-17 years in clinical range of
problems on CBCL, compared to the Mental Health
of Young People in Australia (MHYPA) Survey
(n3870)
Comparisons are made with the findings of the
Australian governments mental health of young
people in Australia (2000), based on a national
representative sample
40Carer Ratings on the Achenbach Child Behaviour
Checklist interview 1
- 43.4 of the children were in the clinical range
for number of total problems, - 35.8 for internalising problems
- 34.0 for externalising problems
- Clinical rate for Total Problems is three times
the Australian community sample - Internalising and externalising problems exceeded
the MHYPA community norms
41Carer Ratings on the Achenbach Child Behaviour
Checklist interview 2
- Between 7.5 and 28 demonstrated clinically
significant problems on the subscales - Attention problems, social problems, delinquent
behaviour, anxiety and depression rated in the
clinical range. - 38 of children were in the clinical range of
total problems - 22 for internalising problems
- 37 for externalising problems
42Comparison data of scores from Interviews 1 to 2
- Significant decreases detected between carer
ratings at Interview 1 and 2 on the internalising
scores (t2.07, df 50, plt0.05) and the anxiety
and depression subscale (t2.01, df 50, plt 0.05) - Fewer children fell into the clinical range of
total problems at the second interview - Ratings remained above the Australian normative
data on all subscales total problems and
externalising problems - Internalising problems had dropped.
43CBCL Teacher Report Form (TRF)
- Teachers of children in care were asked to
complete the Achenbach teachers check-list, a
companion to the child behaviour checklist - The instrument is norm referenced and assesses
key problem sub-scales and overall problem scores
- The TRF also includes an Adaptive Functioning
Scale which include 5 ratings over two subscales
on the childs positive attributes as displayed
at School
44TRF (contd)
- Academic Performance
- teachers ratings of the childs performance in
academic subjects - Adaptive Functioning
- Four adaptive characteristics and the sum of the
four characteristics - How hard the child is working
- How appropriately he/she is behaving
- How much he/she is learning
- How happy he/she is
45Who were included in this part of the study?
- The TRF was completed for children aged between 5
and 17, with an average age of 11.1 years (sd 3.1
years) - Additionally each childs main teacher completed
a checklist for another child in the class,
matched for age and sex but who resides in a
birth family
46Table T-scores for TRF Problems at Assessment 1
for Care and Control Groups
47Table (contd)
48Children in care
- The problem subscale scores have a minimum of 50,
and a clinical cut off of 64 - The maximum scores for the children varied from
67 (withdrawn) to 91 (for aggressive behaviour) - The average scores ranged from 52.6 (somatic
complaints) to 58.63 (social problems) - the highest average scores for girls was social
problems (mean 59.65) and for boys, aggression
(mean 58.48) - There were 14 children in the clinical range for
the summary scores for internalising problems,
(greater than 63 on the teacher ratings), 21
with externalising problems and 177 over
threshold on total problems
49Control Group Children
- Compared to the children in care only two
significant differences were detected - Firstly the children in care had higher t-scores
on aggressive behaviour (means 58.2 for care
and 54.3 control P0.013) - The care group had higher t-scores for
externalising problems (means 56.7 for care,
52.1 for control, p0.019) - The control group had high level of children in
the clinical range of scores for internalising
problems - 25of the children in the control group had
scores which fell in the clinical range for
internalising problems
50Figure 11 Percentage of children in the clinical
range for TRF summary scales compared to the
sample of matched children, not in care
51Table T-scores for adaptive functioning scales
for children in care and control group
A high score is indicative of more adaptive
functioning
52Adaptive Functioning Scales (TRF)
- Children in care
- children in care had the highest average score
for happiness and the lowest for behaving
appropriately - By gender, girls had their highest average
ratings for working hard (mean 44.85) - and the boys, being happy (46.35) or working hard
(46.5) - The highest percentiles in the scales for this
group ranged from 73rd percentile (behaving
appropriately) to the 93rd percentile (learning) - Control group
- The control group childrens percentile means
varied from a low of 30.31 for learning, to a
high of 37.40 for working hard
53Figure 13 Adaptive scale percentiles for
children in care and the matched children, not in
care for academic performance and the sum of the
adaptive scales
54Comparisons between the care and control samples
- Girls in care, appear to function at a lower
average percentile to their non care peers - Boys samples appear more evenly matched
- In relation to the subscale behaving
appropriately children in care have lower
ratings
55Comparisons between the groups at assessment 2
- Both groups demonstrated significant changes in
their TRF problem scores from the first
assessments - With regard to the summary scales, both groups
showed significant reductions in the ratings - In the subscales, the care group changed in six
areas, as opposed to 4 areas in the control group - The strongest changes for the control group
surrounded the internalising cluster - The care children showed most change in the
externalising clusters
56Summary
- Both the children in care and control group had
arrange of problems detected - Evidence of a greater prevalence of problems in
the care group - The high prevalence of internalising problems
amongst the control group - At the second assessment there were no
differences between the two groups on the problem
subscales, which, in a restorative program is a
positive finding
57Summary (contd)
- On the adaptive functioning scales, children in
care showed significant improvements across all
subscales - Children in care were functioning near to the
50th percentile, based on the normal population - Control group showed some significant gains but
without the same breadth or magnitude - Some of this change may be attributed to the
effects of restorative care and the Barnardos
intervention
58Comparisons between the groups at assessment 2
(contd)
- the teachers not blind to the status of the two
children that they were assessing - factors need to be considered when interpreting
these data - Some of these changes might be apportioned to
different rating styles of the teachers at the
different time points, but this is a constant for
the two groups. - Some may be natural maturation although the age
related norms should also account for this
effect. - Finally some of the effects may be attributed to
the effects of the Barnardos intervention
59Teacher Carer Ratings on the Achenbach Rating
Scales
- There was significant agreement between the
carers and teachers on the ratings of children - They were more likely to agree on the
identification of children below clinical
threshold problems than above - Carers would rate more children at above
threshold levels than teachers, especially with
regard to externalising and total problems
60Teacher Carer Ratings on the Achenbach Rating
Scales (contd)
- This difference may be an indication that
teachers were more conservative than the carers - Have a better idea of the normal range of
behaviour - The structure of the classroom provides
sufficient guidelines for the children to behave
within non clinical levels - In the absence of such structure, the carers
observe more problematic spectrum
61Summary and implications
- Children had high levels of psychological need
- Problems with attention, social interactions,
anxiety, aggression approximate estimates from
other studies - Findings underline the importance of recognising
emotional and behavioural difficulties
experienced by children in care early and
identifying their impact on carers. - Vulnerabilities and strengths of children based
on gender and age, and the need for
individualised and differential responses from
carers and caseworkers. - Recognise adversities which threaten Childrens
wellbeing in care such as, maltreatment
histories, disrupted attachments and interrupted
schooling - Externalising behaviours have been cited as a
strong predictor of placement breakdown
(Leathers, 2002) - Strengthen professional decision making to ensure
children are less likely to move
62Summary and implications (cont)
- Monitor children at increased risk of instability
in care - Support children at risk of psychological
difficulties with therapeutic services - Support carers in enhancing their relationship
with troubled children - Skill foster parents in approaches needed for the
sensitive management of childrens emotional and
behavioural problems - Support carers and teachers through professional
training to identify and work with vulnerable
children to prevent the development of
psychopathology
63Focus on Strengths and Competencies of Children
in Care
- Childrens sense of happiness improved overtime
is a positive finding implying placement in care
provided a route to rehabilitative intervention
for children with maltreating histories - Permanent care afforded a context to develop a
more secure base - Being in care offered a pathway into restorative
services - School environment and the educational process
can potentially offer structure, boundaries and
security to the children in care systems
64Focus on Strengths and Competencies of Children
in Care (cont)
- The importance of school in offering children
positive role models and benefits of routines
and rituals for children experiencing upheaval
and adversity in their environment
(Gilligan,1998) - Research attributes positive school experience
and achievement to happy adult behaviour
(Rutter, Quinton et al. 1993) - Instability of care placements and the
difficulties involved in starting new schools on
a regular basis present significant challenges to
children in care
65Resilience enhancing interventions
- Develop strategic interventions that promote
childrens strengths and competence - Foster childrens relationship building skills
- Support carers in acknowledging and reinforcing
childrens prosocial behaviours - Develop co-ordinated multidisciplinary response
to address overlapping domains of need, such as
education and mental health