Philip A Fisher, PhD, University of Oregon & OSLC

INTERVENTIONS FOR 0-5 YEAR OLD CHILDREN: NEW DEVELOPMENTS

STRESS!

SOLUTIONS FOR STRESS

birth 6 years 14 years

Experience Shapes Brain Architecture by Over-Production Followed by Pruning

(700 synapses formed per second in the early years)

Toxic Stress Changes Brain Architecture

Normal

Toxic stress

Prefrontal Cortex andHippocampus

Typical neuron— many connections

Damaged neuron— fewer connections

Sources: Radley et al. (2004)

Bock et al. (2005)

STRESS AND CORTISOL

SOLUTIONS FOR STRESS: MTFC-P

MTFC-P foster kids have more normal cortisol levels

Mechanisms:

Reduces foster parent stress associated with child behavior problems

Foster parents more able to tolerate children’s misbehavior

Children feel safer during major transition periods

ABILITY TO MAKE USE OF FEEDBACK

Walker, H. (1994). Social skills in school-age children and youth: Issues and best practices in assessment and intervention. Topics in Language Disorders, 14(3), 70-82

Effects of Foster Care Interventions:

ERP Feedback Negativity study using a color flanker task

Color Flanker Task: 4 Combinations

Green

Red

Congruent Incongruent

PILOT STUDY of intervention effects attention to feedbackFeedback 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

Community CorrectIncorrect

CorrectIncorrect

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

Bruce, Martin-McDermott, Fisher, & Fox (2009)-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

-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

Intervention CorrectIncorrect

Limitations: No pre-intervention measures, small subsample of overall RCT sample, variation in how long intervention group receiving services

Reg Foster Care

Figure 1. Replication of intervention effects on FRN from randomized clinical trial of foster care intervention. Grand average waveforms for the feedback-related negativity (FRN) at Fz for correct (gray line) and incorrect trials (black line). Note. The feedback stimuli were presented at 0 ms, and the baseline window was at –150 to –50 ms relative to the feedback stimuli.

-15

-10

-5

0

5

10

15

-150 0 150 300 450 600 750

-15

-10

-5

0

5

10

15

-150 0 150 300 450 600 750

-15

-10

-5

0

5

10

15

-150 0 150 300 450 600 750

-15

-10

-5

0

5

10

15

-150 0 150 300 450 600 750

Reg Foster CareIntervention

Pre Intervention

FRN FRN

FRN

FRN

Post Intervention

COMPENSATORY PROCESS INTERVENTIONS

Implicit Assumption: Rather than focusing on the affected neural systems, it may be more effective to recruit alternative systems and develop competencies based on these systems

Pros: May be effective with individuals who do not show responsiveness to other interventions; some evidence from neurorehabilitation literature

Cons: Largely untested

PATTERNS OF BRAIN ACTIVATION DURING A GO-NO GO TASK FOR

COMMUNITY AND FOSTER CHILDRENNormal control (n=8) Foster care (n=8)

Fisher, Bruce, et al., in press

“One perspective on individual differences in memory capacity views variation in terms of the number of 'slots' that are available for short-term storage. However, apparent capacity differences might also be explained by variations in the efficiency with which information is selected to fill this limited workspace. A useful analogy for understanding the difference between these two ideas is the difference between the space that is available in an exclusive nightclub and the effectiveness of the bouncer who grants admission. From this perspective, high-capacity individuals may have a better bouncer rather than a larger nightclub...”

Awh & Vogel, 2008 Nature Neuroscience

PREVENTING PLACEMENT FAILURE

3/29/2011 5

Who Disrupts?Chamberlain, Price, Reid, Landsverk, Stoolmiller, and Fisher (2006)

Number of behaviorproblems per day

After 6, every additionalbehavior problem s probability of disruption by 25%

7

8

9

10

11

1 2 3 4 5 6

Probability of failed placement by condition

Project KEEP

MTFC-P

20.00

17.00

16.00

15.00

13.00

12.00

11.00

10.00

9.00

8.00

7.00

6.00

5.00

4.00

3.00

2.00

1.00

.00

TC # of pre KEEP study placements

0.70000

0.60000

0.50000

0.40000

0.30000

0.20000

0.10000

0.00000

Me

anP

red

icte

dp

rob

ab

ilit

y

TXControl

GROUP

MTFC-P

MTFC-P

20.00

17.00

16.00

15.00

13.00

12.00

11.00

10.00

9.00

8.00

7.00

6.00

5.00

4.00

3.00

2.00

1.00

.00

TC # of pre KEEP study placements

0.70000

0.60000

0.50000

0.40000

0.30000

0.20000

0.10000

0.00000

Me

anP

red

icte

dp

rob

ab

ilit

y

TXControl

GROUP

20.00

17.00

16.00

15.00

13.00

12.00

11.00

10.00

9.00

8.00

7.00

6.00

5.00

4.00

3.00

2.00

1.00

.00

TC # of pre KEEP study placements

0.70000

0.60000

0.50000

0.40000

0.30000

0.20000

0.10000

0.00000

Me

anP

red

icte

dp

rob

ab

ilit

y

TXControl

GROUP

MTFC-PRe

plic

ated

find

ings

PDR Problem Behaviors

Fitt

ed

Log

Haz

ard

of P

lace

me

nt D

isru

ptio

n

0 5 10 15 20 25

-10

12

RFC

PDR Problem Behaviors

Fitt

ed

Log

Haz

ard

of P

lace

me

nt D

isru

ptio

n

0 5 10 15 20 25-1

01

2

TFC

POSSIBLE MECHANISM: MTFC MITIGATES THE EFFECTS OF PROBLEM BEHAVIOR ON

RISK FOR DISRUPTED PLACEMENT

Fisher et al. (in press)

COST SAVINGS!

SAMPLE FOR ECONOMIC ANALYSIS

117 children enrolled in the RCT of MTFC-P, including n=57 intervention and n=60 regular foster care

Children were between 3-6 years old at enrollment and entering a new placement

Placement instability subsample: Children entering study with 4 or more prior placements (RFC=23, MTFC-P=29)

COST EFFECTIVENESS METHODOLOGY

MTFC-P intervention costs. The total cost of MTFC-P was estimated from clinical trial records and study staff estimates. Study accounting records provided payroll costs, cost of facilities and overhead, and information on purchases of goods and services. Study staff members estimated the time to complete each intervention task. The cost of staff supervision, time to develop treatment plans for the child and family, and staff training were also included.

RFC costs. Staff members from the child welfare agency provided estimates of time spent delivering RFC services. When agency-reported costs for delivering RFC services were not available, published unit costs were used.

Nonprotocol health and social services : For services provided outside of MTFC-P or RFC programs, comprehensive profiles of usual care services were created from a carregiver-report survey designed for this study.

Estimate of the Cost of Running the EIFC Program

for 9 Months from the Agency Perspective

Component of EIFC Program Cost Notes Intensive Pre-service Training for Foster Parents

$337 12 families, one staff member

Ongoing Consultation and Support for Foster Parents

Daily Telephone Calls 9974 12 families, one call per day 5 days per week for 6 months

Weekly Foster Parent Support Groups

2601 One meeting per week for 6 months

24 Hour Crisis On Call 640 2-3 calls per week, some include visits Home and Community Visits 6390 2 per month per family Individual Therapy 13,293 1.5 hours per week Therapeutic Play Group 2160 One meeting per week for 6 months Develop Individual Treatment Plan 197 Medication Management 2335 Initial visit, plus every 6 weeks, 50%

of children receive Parent Training for Permanent Placement Families

22,511 12 families, 1.5 individual sessions per week for 6 months

Staff Supervision 8955 Weekly meeting, plus individual EIFC Staff Training 503 Amortized over 5 years Total Cost for 6 months $69,897 Total Cost per child $6138 Assume one child per family

$6138

Type of Cost

RFC

N = 60

MTFC-P

N = 57

1. MTFC-P Intervention NA 6168

(1760)

2. Total RFC Services Cost 21 253

(8515)

16 003

(7256)**

Cost of Negative Transitions 406

(552)

317

(422)

Room and Board 10 204

(4820)

10 159

(4598)

All Other Services 10 643

(3889)

5527

(2899)**

3. Nonprotocol Service Cost 8749

(5644)

5058

(3506)**

4. Total Foster Care Costs (sum of

items 2 and 3)

21 253

(8515)

22 171

(8754)

5. Total Public Agency Cost (sum of

items 1 and 4)

30 090

(11 014)

27 204

(9939)**Tab

le 1

. Una

djus

ted

mea

n (S

D)

cost

s (2

008

US

D)

at 2

4 m

o.

post

-ran

dom

izat

ion,

full

sam

ple

Abbreviations: RFC, regular foster care; MTFC-P, Multidimensional Treatment Foster Care for Preschoolers.Significance tests adjusted for differences in gender, number of placements prior to start of study, and baseline severity.*P < .05; **P < .005.

Type of Cost

RFC

N = 23

MTFC-P

N = 29

1. MTFC-P Intervention NA 6097

(1916)

2. Total RFC Services Cost 27 113

(5973)

17 577

(8094)**

Cost of Negative Transitions 493

(603)

420

(476)

Room and Board 14 717

(3200)

12 101

(5062)**

All Other Services 11 903

(2588)

5057

(2876)**

3. Nonprotocol Service Cost 8948

(5471)

5921

(3901)*

4. Total Foster Care Costs (sum of

items 2 and 3)

27 113

(5973)

23 674

(9747)

5. Total Public Agency Cost (sum of

items 1 and 4)

36 061

(9160)

29 595

(11 243)*

Tab

le 1

. Una

djus

ted

mea

n (S

D)

cost

s (2

008

US

D)

at 2

4 m

o. p

ost-

rand

omiz

atoi

n fo

r in

stab

ility

su

bsam

ple

Abbreviations: RFC, regular foster care; MTFC-P, Multidimensional Treatment Foster Care for Preschoolers.Significance tests adjusted for differences in gender, number of placements prior to start of study, and baseline severity.*P < .05; **P < .005.

COST EFFECTIVENESS PLANE

Higher cost, worse outcome

Lower cost, worse outcome

Higher cost, better outcome

Lower cost, better outcome

-$10

,000

-$5,

000

$0$5

,000

$10,

000

Incr

em

en

tal t

ota

l co

sts

Incremental proportion of permanent placements (TFC-RFC)

I

IVIII

II

NEGATIVE TRANSITIONS WHILE IN CARE

Better outcome, higher cost

Better outcome, lower cost

Worse outcome, higher cost

Worse outcome, lower cost

-$10

,000

-$5,0

00

$0

$5

,000

$1

0,0

00

-$15

,000

-$20

,000

Incre

me

nta

l to

tal co

sts

$ (

TF

C-R

FC

)

-2 -1 0 1 2

Incremental weighted negative transitions (TFC-RFC)1000 replications, missing values imputed using MICE, adjusted for differences in gender, number of prior placements, and severity

Full sample (N=117)Better outcome, higher cost

Better outcome, lower cost

Worse outcome, higher cost

Worse outcome, lower cost

-$10

,000

-$5,0

00

$0

$5

,000

$1

0,0

00

-$15

,000

-$20

,000

-2 -1 0 1 2

Incremental weighted negative transitions (TFC-RFC)1000 replications, missing values imputed using MICE, adjusted for differences in gender, number of prior placements, and severity

Instability sample (N=52)

Permanent Placements

Worse outcome, higher cost

Worse outcome, lower cost

Better outcome, higher cost

Better outcome, lower cost

-$10

,000

-$5,0

00

$0

$5

,000

$1

0,0

00

-$15

,000

-$20

,000

Incr

em

en

tal t

ota

l co

sts

$ (

TF

C-R

FC

)

-1 0 1

Incremental proportion of permanent placements (TFC-RFC)1000 replications, missing values imputed using MICE, adjusted for differences in gender, number of prior placements, and severity

Full sample (N=117)Worse outcome, higher cost

Worse outcome, lower cost

Better outcome, higher cost

Better outcome, lower cost

-$10

,000

-$5,0

00

$0

$5

,000

$1

0,0

00

-$15

,000

-$20

,000

-1 0 1

Incremental proportion of permanent placements (TFC-RFC)1000 replications, missing values imputed using MICE, adjusted for differences in gender, number of prior placements, and severity

Instability sample (N=52)

NEW DIRECTIONS

MTFC-P

KEEP-P

KEEP-P with playgroups(San Diego)

KEEP-P for adoptive parents

(England)

KITS

KIT S randomized trials for foster

children, DD, low income,

KITS implementations

in Baltimore, Australia

Video Feedback

Prenatal drug and alcohol

exposed infants

Foster care biological parent

visitations