Don

1
Dont forget the basics School

• Malmö, September 26, 2013
• Bo Vinnerljung, professor
• Social Work, Stockholm University
• bo.vinnerljung_at_socarb.su.se

2

• Interdisciplinary research, eg.
• Anders Hjern, pediatrics/epidemiology
• Marie Berlin, sociology
• Karl Gauffin, public health
• Emma Björkenstam, epidemiology
• Eva Tideman, psychology
• Marie Sallnäs, social work
• Bo Vinnerljung, social work

3
Anders Hjern
4
Menue

• Crash course in epidemiological analysis
• Results from national cohort studies based on
  register data
• Results from intervention studies
• Relevance for work with anti-social adolescents

5
What is risk?

• Increased/reduced probability
• for something specific For what?
• a relative concept Compared to whom/what?
• a quantitative concept How much?
• The comparison group has always Relative Risk
  (RR) 1
• RR 2 is 100 more
• RR gt2-3 are high excess risks (over-risks)
• Risk does not say anything about how many
• 3 of all girls in Sweden become mothers before
  age 20
• 0,75 of all boys become fathers before age 20
• The risk of teenage parenthood is 4 times higher
  for girls compared to boys (RR 4)

6
To adjust for background factors (confounders)

• If the distribution of these background factors
  were the same in the groups that we compare are
  there still any excess risks?
• Examples sex, parental education
• Adjustments are done in multivariate analyses
  each background factors unique association with
  the outcome (all other variables being equal)

7
Nordic National registers

• Data bases covering the entire population.
  Purpose statistics and research.
• Individual data, based on a personal ID-number
  that follows every resident from
  birth/immigration to death
• Each national register is regulated by law,
  Strict secrecy rules.
• All residents have a right to know the
  information about them in the national registers
  but they do have the right to erase information
  about themselves in the registers.
• Registers may be used/linked for research after
  approval from an ethical board. Data are
  anonymized - individuals cannot be identified.
• No cases of abuse or leaks.

8
Some medical national registers in the Nordic
countries
Register Denmark Finland Norway Sweden
Cause of death 1943 1969 1951 1952
Medical Birth 1973 1987 1967 1973
Hospital discharge 1977 1987 2008 1987
Cancer 1943 1952 1952 1958
Prescribed Drugs 1995 1995 2004 2005
9
Epidemiology, exploratory research How many?
Changes over time? Who? Patterns of care? Care
leavers experiences. Prevalence of eg, mental and
somatic health problems
Longitudinal studies Risk factors, protective
factors? Risk mechanisms, mediators, moderators?
Clinical trials
10
Does school matter for all children

• .. in a longitudinal perspective?
• Yes, school performance is a powerful predictor
  of future psychosocial problems
• regardless of SES (socioeconomic background,
  class), at least in Sweden

11
Grades final year of primary school (age 16)

• Low/incomplete grades lowest 1/6 in the country
  lt (M-1 SD)
• Compared to all the rest (5/6)

12
Suicide attempts
SES for parents at age 10
Vinnerljung et al, 2010
13
Links between childhood factors and drug abuse
after age 20

• All born in Sweden 1973-88, alive at age 16
• N1,4 million. Follow-up to 2008 (age 20-35).
• Indication of drug abuse
• Drug related death or
• Hospitalization with drug abuse diagnosis (not
  intox) or
• Convicted for drug related crime
• 3 (42.000)
• Gauffin et al, 2012

14
Background factor RR (all else being equal)
Mothers SES --
Mother born outside Europa 1,7
Mother single 1,5
Mother a teenage mother --
Mother did not work --
Mother disability pension --
Mother lived on welfare 1,4
Mother serious mental health problems 1,3
Father serious mental health problems --
Mother substance abuse 1,4
Father substance abuse 1,6
Mother substantial criminality 1,3
Father substantial criminality 1,5
Grew up in a city vs in a rural community 1,7
Man (sex) 2,4
No/incomplete/very low school grades at age 16 4,1
15
Poor grades/school performance have links to eg.

• Low cognitive capacity (0.65 with IQ at
  conscription)
• Other individual traits, eg. working memory
• Behavioral problems but this is not a one-way
  street
• Mental health problems
• Poor support from home, adverse childhood etc
• School related factors, poor peer status in
  school

16
What does this mean for children in societal
care?They do poorly in schoolSocial services
do a poor job with their school/education (but
not SiS) School failure is added to other
risk factors
17
National cohort study/register data

• Analyses of datasets with all born 1972-1981.
  Linking 10 national registers.
• Follow-up from age 16 to 2005 (age 24-33)
• 930.000 persons, including
• 7.000 children that grew up in foster care,
  average time in care 12 years
• Vinnerljung et al, 2010 Berlin et al, 2011

18

• No/incomplete/low grades Above average
• Boys
• Normal group 22 41
• In home care before teens 55 15
• Children from welfare families 57 13
• Grown up in foster care 60
  11
• Girls
• Normal group 11
  60
• In home care before teens 37
  27
• Children from welfare families 40 25
• Grown up in foster care 43 23
• Grown up in foster care gt 5 years in care (M 11
  years in care), left care after age 17. All
  persons with disability pension at age 23 were
  excluded from the analysis

19
What factors increase the risk for poor grades
among foster children?

• Sex/gender
• Birth year --
• Age at placement --
• Time in care --
• Mother mental health problems --
• Mother substance abuse --
• Father mental health problems --
• Far substance abuse --

20
Negative expectations...

• Results from trials suggest that social workers,
  teachers and foster parents commonly
• underestimate the cognitive capacity of foster
  children
• have pessimistic expectations of school
  performance
• Same results in Danish and British studies

21
Foster children Five-fold disadvantaged by the
care and education system

• High risks of school failure
• Lower grades than peers with same IQ
• Lower education than peers with same IQ
• Lower education than peers with same grades
• Lower chances of secondary education if they had
  poor grades (eg. fewer use adult education)

22

• Children growing up in foster care have high
  risks for future
• Suicide RR 6.4
• Suicide attempts RR 6.2
• Serious mental health problems RR 5.0
• Drug abuse RR 6.8
• Alcohol abuse RR 4.9
• Serious criminality RR 7.5
• Teenage parenthood RR 3.8
• Welfare dependency RR 9.8
• (adjusted for sex and birth year)

23
(No Transcript)
24
(No Transcript)
25
Analyses of foster children only

• No/low grades the only substantial risk factor
  for all negative outcomes (except sex)
• Parental pathology no/weak association with
  outcomes
• No associations between age at placement or time
  in care
• Abscence of school failure the only strong factor
  that predicted good outcomes (except gender)
  regardless of how
  good outcome was defined

26
Conclusions

• Children who fail at school are a high risk group
  for future psychosocial problems regardless of
  socioeconomic background
• School failure seems to be the strongest risk
  mechanism for foster childrens long term
  development (same for other children from the
  social marginal)
• A determinant for foster childrens future
• The good news school performance is a variable
  risk factor . (in contrast to sex, age,
  genes, experiences from early childhood etc)

27
The Helsingborg trial - SkolFam

• 25 foster children age 7-12 were tested with
  standardized psychological and educational
  instruments
• Results were used to
• access available support from school
• tailor individual educational support and
  interventions
• advice teachers/schools, foster parents - and the
  children
• This was done by an educational psychologist and
  a special education teacher.
• Re-tests after 24 months to evaluate the program

28
Results after first measurements

• Normal cognitive capacity (average lower than
  peers, same as international adoptees).
• 75 were substantially underachieving in school
• Large knowledge gaps were common
• Most foster parents, social workers and teachers
  had low/pessimistic expectations on the
  childrens school performance.
• Lots of pseudo-psychological explanations to why
  the children did poorly in school
• Pathologizing of children was common, amateur
  diagnoses

29
Results after 2 years

• Clinical knowledge says that the foster
  childrens school performance should have
  deteriorated (larger gaps over time between them
  and normal population peers)
• Wisc IQ Total (plt0,001)
• Wisc IQ Performance (plt0,01)
• Wisc IQ Verbal (plt 0,05)
• Spelling (DLS)
• Reading speed
• Word Chains
• Magnes Math diagnosis n.s.
• Tideman et al, 2011

30
Replication in Norrköping 2008-11

• 21 children, two years between tests
• Working memory training was added for children
  with poor numeracy skills
• WISC same results as in Helsingborg
• Literacy similar results to Helsingborg
• But Maths.
• Tordön et al, submitted

31
Math scores related to working memory training
Without WM training With WM training
Total
32

• Case study Jens, age 11
• At start of intervention
• Personal assistent in school, concentration
  problems, suspected ADHD, suspected learning
  disability
• IK 70, working memory 62 (WISC)
• Maths, stanine 1
• Working memory training
• Re-start in math
• After two years
• IK 86 (23) , working memory 99 (60 )
• Math stanine 5
• No concentration problems
• Likes school

33
Paired Reading trial

• We trained foster parents to use a special
  reading tecnique for 80 children age 8-12
• Reading 20 min/day, 3 days/week
• 16 weeks (pre-post evaluation, std instruments)
• Replication of a English trial

34
Results

• The average child increased her/his reading age
  with almost a year (11 months), after 4 months of
  Paired Reading ( in England 12 months)
• Improved scores on the WISC vocabulary scale
• Improved Total IQ

35

• Interventions younger children
• School perform ? Anti-soc behavior
• ? Anti-soc behavior School perform
• --------------------------------------------------
  ---------------
• Interventions anti-social adolescents
• ? Anti-soc behavior School perform
• School perform ? Anti-soc behavior

36

• Does improved literacy/numeracy skills have a
  value, even if this is not related to reduced
  anti-social behavior in the short run?
• Yes it increases the chance that youth can
  benefit from turning points later in life
  (hypothesis).

37
Turning points

• Far more decisive than treatment
• Examples partner, parenthood, getting tired of
  crime/substance abuse, health concerns, getting a
  job, religon
• Today, inclusion in the labor market usually
  requires school/education (learning a trade)
• The way out,the escape from disadvantage.

38
What to do?

• A minimum standard for school/education in
  residential care for anti-social adolsecents
• At least
• Assessment of literacy and numeracy skills
• Identify knowledge gaps in maths, re-start
• Start literacy and numeracy training, high
  quality programs (evidence based methods).
• Cognitive tests for assessing potential - but.

39

• Results from cognitive tests (IQ)
• of children from the social marginal
• are often instable, and may change over time

40
References

• Jablonska B, Lindberg L, Lindblad F, Rasmussen F,
  Östberg V Hjern A (2009) School performance and
  hospital admissions due to self-inflicted
  injuries. Int Journal of Epidemiology, 38,
  1334-1341.
• Björkenstam C, Björkenstam E, Ljung R,
  Vinnerljung B Tuvblad C (2013). Suicidal
  behavior among delinquent former child welfare
  clients. European Child Adolescent Psychiatry,
  22, 349-355.
• Tideman E, Vinnerljung B, Hintze K Isaksson AA
  (2011). Improving foster childrens school
  achievements Promising results from a Swedish
  intensive study. Adoption Fostering, 35, 44-56.
  
• Berlin M, Vinnerljung B Hjern A (2011). School
  performance in primary school and psychosocial
  problems in young adulthood among care leavers
  from long term foster care. Children and Youth
  Services Review, 33, 2489-2487.
• Vinnerljung B Hjern A (2011). Cognitive,
  educational and self-support outcomes of
  long-term foster care versus adoption. A Swedish
  national cohort study. Children and Youth
  Services Review, 33, 1902-1910.
• Forsman H Vinnerljung B (2012) Interventions
  aiming to improve school achievements of children
  in out-of-home care a scoping review. Children
  and Youth Services Review. 34, 1084-1091.
• Gauffin K, Vinnerljung B, Fridell M, Hesse M
  Hjern A (2013). Childhood socioeconomic status,
  school failure, and drug abuse a Swedish
  national cohort study. Addiction, 108, 1441-1449.