Early Intervention for Foster Children

Philip A. Fisher, Ph.D.

Oregon Social Learning Center

Outcomes Outcomes at 6 mo at 6 mo

intervals thru intervals thru middle middle

childhoodchildhood

The Early Intervention Foster Care Study, 1999-

N= 117N= 117

PreschoolPreschool

AgedAged

Foster Foster

ChildrenChildren

Random Random AssignmentAssignment

MTFC-P MTFC-P interventionintervention

Regular foster Regular foster carecare

6060

5757

Community comparison group Community comparison group (60)(60)

Adrift in the foster care system

Birth 1yr 2yrrs 3yrs 4yrs 5yrs 6yrs 7yrs

Child 3

*

Child 2

Child 1

Overall sample characteristics

Mean age at first foster placement 3.4152

Age at first placement

Number of living transitions at study start Frequency Percent

1.00 17 14.5 2.00 27 23.1 3.00 21 17.9 4 or more 52 44.4

Valid

Total 117 100.0

Frequency Percent birth to 6 mons 3 2.6 6 mons to 12 mons 7 6.0 12 to 24 mons 8 6.8 24 to 36 mons 18 15.4 36 to 48 mons 37 31.6 48 mons or order 44 37.6

Valid

Total 117 100.0

When foster care works…

Child 4

Child 5

Birth 1yr 2yrrs 3yrs 4yrs 5yrs 6yrs 7yrs

Child 6

(adapted from Fisher, Burraston, & Pears, 2005)

1 2 3 4 5 6 7+

90

80

70

60

50

40

30

20

10

0

%

Number of placements

Percent of placement failures based on number of prior placements

HPA axis dysregulation associated with early life stress

0

0.2

0.4

0.6

0.8

1

wakeup midmorning bedtime

00.20.4

0.60.8

1

wakeup mid morning bedtime

typical daytime HPA activity

ug/

dl

typical

low daytime HPA activity

ug/

dl

chronically elevated daytime HPA activity

ug/

dl

0

0.2

0.4

0.6

0.8

1

wakeup midmorning bedtime

stress-induced ‘blunted’patterns

0

0.2

0.4

0.6

0.8

1

wakeup midmorning bedtime

00.20.4

0.60.8

1

wakeup mid morning bedtime

typical daytime HPA activity

ug/

dl

typical

low daytime HPA activity

ug/

dl

chronically elevated daytime HPA activity

ug/

dl

0

0.2

0.4

0.6

0.8

1

wakeup midmorning bedtime

stress-induced ‘blunted’patterns

00.20.4

0.60.8

1

wakeup mid morning bedtime

typical daytime HPA activity

ug/

dl

typical

low daytime HPA activity

ug/

dl

chronically elevated daytime HPA activity

ug/

dl

0

0.2

0.4

0.6

0.8

1

wakeup midmorning bedtime

00.20.4

0.60.8

1

wakeup mid morning bedtime

typical daytime HPA activity

ug/

dl

typical

low daytime HPA activity

ug/

dl

low daytime HPA activity

ug/

dl

chronically elevated daytime HPA activity

ug/

dl

chronically elevated daytime HPA activity

ug/

dl

0

0.2

0.4

0.6

0.8

1

wakeup midmorning bedtime

stress-induced ‘blunted’patterns

(downregulation via chronic stress)

Anxiety and affective disorders

Some, but not all, foster children show altered HPA axis function

Bruce, Fisher, Pears, & Levine (submitted)

HighAverageLow

Cortisol Classification

70.0%

60.0%

50.0%

40.0%

30.0%

20.0%

10.0%

0.0%

Perc

enta

ge o

f C

hild

ren

CC

FCGroup

HighAverageLow

Cortisol Classification

70.0%

60.0%

50.0%

40.0%

30.0%

20.0%

10.0%

0.0%

Perc

enta

ge o

f C

hild

ren

CC

FCGroup

Percentage of comparison and foster children with typical, low, and high cortisol patterns

0

10

20

30

40

50

60

70

80

90

Typical Low High

Cortisol pattern

Perc

enta

ge o

f ch

ildre

n

Comparison

Foster

Percentage of comparison and foster children with typical, low, and high cortisol patterns

0

10

20

30

40

50

60

70

80

90

Typical Low High

Cortisol pattern

Perc

enta

ge o

f ch

ildre

n

Comparison

Foster

Oregon Delaware

Dozier et al. (2007)

Neglect is the primary form of maltreatment associated with HPA dysregulation

Bruce, Fisher, Pears, & Levine (under review)

Relation between Neglect and Morning CortisolLevels

0

0.5

1

1.5

2

2.5

3

Morning Cortisol Level

Severi

ty o

f N

eg

lect

Low Average High

F(2, 114) = 4.27, p < .05

Severity neglect in low, average, and high morning cortisol groups

0

0.5

1

1.5

2

2.5

3

Morning Cortisol Level

Severi

ty o

f N

eg

lect

Low Average High

F(2, 114) = 4.27, p < .05

Prefrontal Cortex Regions

Inhibitory control

Working memory

Planning

Decision-making in context of rewards and consequences

ADHD Substance abuse

1.501.000.500.00-0.50-1.00-1.50-2.00

Executive Functioning

6.00

5.00

4.00

3.00

2.00

1.00

0.00

r = -.22, p = .02

# o

f p

lacem

en

ts

Association Between # of Prior Placements and Executive Functioning

Neurobiological vulnerabilities from early stress for foster children

HPA system

Severity of neglect

Dis

rupt

ed d

aily

rhy

thm

Prefrontal cortex

# of placements

Imp

aire

d ex

ecut

ive

fun

ctio

n•Neurobiologically informed interventions

•Interventions with neurobiological targets

Early intervention in foster care works

MTFC-P emphasizes 3 domains:

Child NeedsCaregiver-Child

RelationshipCase Management

Foster Parent Consultant

Family Therapist

‘Daily Report’ Caller Case Manager

Child Therapist Behavioral Skills Trainer

Child Psychiatrist

STAFF

Home Community Preschool/school

Contexts

5 Key Program Components

Foster parent support & consultation services Pre-placement training Weekly group meeting 24/7 on call support

Child treatment services

Parenting support for birth/adoptive families

Daily Report telephone check-in w/caregiver

Clearly specified staff roles & responsibilities

0

0.2

0.4

0.6

0.8

1

wakeup mid morning bedtime

Initial

3 Month

6 Month

HPA Plasticity

Therapeutic intervention

Group effects on morning cortisol levels across time for all children

MTFC-PReg foster care

Comm-unity Comp

Fisher, Gunnar, Dozier, Bruce, & Pears (2007), Annals NYAS

Intervention effects on executive functioning:Feedback negativity at Fz (prefrotnal center electrode site)

-12

-10

-8

-6

-4

-2

0

2

4

6

8

10

12

14

16

-200 -100 0 100 200 300 400 500 600 700 800

CC CorrectIncorrect

RFC CorrectIncorrect

Group:F(2, 31) = 1.80, nsInteraction:F(2, 31) = 5.11, p < .05

Bruce, Martin-McDermott, Fisher, & Fox (under review)

-12

-10

-8

-6

-4

-2

0

2

4

6

8

10

12

14

16

-200 -100 0 100 200 300 400 500 600 700 800

MTFC-P CorrectIncorrect

-12

-10

-8

-6

-4

-2

0

2

4

6

8

10

12

14

16

-200 -100 0 100 200 300 400 500 600 700 800

Month

Ca

reg

ive

r S

tre

ss

2 4 6 8 10 12 14

0.1

00

.15

0.2

00

.25

0.3

00

.35

RFCTFC

Fisher & Stoolmiller (in press)

Conditional probability of caregiver stress given child behavior problems

Caregiver stress levels are directly related to children’s cortisol levels Fisher & Stoolmiller (2007)

Month

Car

egiv

er S

tres

s

2 4 6 8 10 12 14

0.10

0.15

0.20

0.25

0.30

0.35

RFCMTFC-P

Month

Car

egiv

er S

tres

s

2 4 6 8 10 12 14

0.10

0.15

0.20

0.25

0.30

0.35

RFCMTFC-P

Caregiver stress

MTFC-P

Mo

rnin

g C

ort

isol

Change in attachment (% secure behavior)

0.6

0.7

0.8

0.9

T1 T2 T3 T4 T5

Time

Secu

re b

eh

avio

r

Reg Foster Care MTFC-P Community Comparison

Early intervention improves permanency outcomes

Birth 1yr 2yrrs 3yrs 4yrs 5yrs 6yrs 7yrs

Child 7

Child 8

Child 9

Intervention begins

Successful permanent placements for children with 4 or more prior placements at study start

Regular foster care 9 of 23 (39%) MTFC-P 23 of 29 (79%)

0

5

10

15

20

25

RFC MTFC-P

Perm

No Perm

Acknowledgements NIMH MTFC-P StudyKatherine Pears

Jackie BruceKristen Greenley

Hyoun Kim

NIDA Transdisciplinary Prevention Center

John ReidPatti Chamberlain

Leslie LeveDavid DeGarmo

Mike Stoolmiller Rebecca Fetrow

NIMH Early Experience, Stress Neurobiology &

Prevention Science Network

Megan Gunnar Paul PlotskyGig Levine

Steve SuomiNathan FoxSeth PollakMary Dozier

Charles Neal James Black

Mar Sanchez Delia Vasquez

Ron Barr_

NIH Institutes

NIMH, NIDA, NICHD