Research into parenting programmes:

evidence-based policy or what?

Stephen ScottProfessor of Child Health and Behaviour,

King’s College London

Director of Research, National Academy for Parenting Practitioners

DCSF Research conference 9 February 2010

Good Childhood Findings

• Many positive aspects of child life today:

–Better physical health, better homes, more communication

–More tolerant, more concerned environment

• BUT–UNICEF on GB: more out of education 15-19, income inequality

–Rise in emotional & behavioural problems in last 30 years

Oppositional & defiant

Blamed by parents

Disliked by siblings

Gets into fights

Rejected by peers

Low self esteem

Hard to control

Poor school achievements

Blames others

Stealing and truanting

Deviant peer group

Antisocial attitude

Career offender

Unemployed

Drug misuse

0

5

10

15

5 years 8 years 11 years 14 years 17 years

Escape

1/5

1/5

1/5

4/5

4/5

4/5

4/5

Continuity of anti-social behaviour from age 5 to 17. Source: Scott 2002

% of allchildren

No past antisocial behaviour

1/51/5

10% 10% 10% 10%

Total extra cost to age 28(Scott, Knapp et al 2001, BMJ )

10,400

24,300

70,000

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

No problems Conduct

problems

Conduct

disorder

Mea

n c

ost

£ 1

998

9. What is the pathway to high public cost, A or B?

Early risk factors

Socioeconomic status

Male AReading ability High CostParenting by adulthood

B B

AntisocialBehaviour at 10

Zone of impact of parenting

programmes

• Abuse–and it amelioration in Looked After Children

• Inequalities in attainment

•Socialisation and Antisocial behaviour

Clinical trial (Scott, Spender et al 2001, BMJ)

• 141 children age 3-7 referred to CAMHS

• severe, persistent antisocial behaviour (worst 1%)

• ‘Incredible Years’ parenting programme:

– videotapes shown in group, 3 wks each of • Play

• praise & rewards

• setting limits

• handling misbehaviour

1.82 1.87

1.9

0.75 0.75

Pre Post 1 Year

0

0.5

1

1.5

2

Waiting List Parent Group

PACS Interview Score

Child Antisocial Behaviour

Long-term follow up

Follow up 2005-07 7-10 years later of 94 children now aged 10-17 (mean 13)

Intention to treat, 74 allocated to IY, 20 controls

Parent SDQ total(p<0.003)

10

12

14

16

18

20

22

IY

Controls

Youth report home beh(p<0.038)

0

0.5

1

1.5

2

2.5

3

3.5

4

IY

Controls

“An ounce of action is worth a ton of theory”

Friedrich Engels, 1860

TRAINER

PARENTINGPRACTITIONER

PARENT

CHILD

Feedback loops of research contribution

Page 18

The National Academy for Parenting Practitioners

Training: over 4,000 practitioners trained in evidence-based programmes this year alone

Research: extensive programme researching what we do and finding out what works and how

Dissemination: events with stakeholders explaining what we are about, workshops for practitioners on new ways of working. Website with latest information on what works www.parentingacademy.org

Page 19

Research

1. Parenting Programme Evaluation Tool & Training evaluation studies

2. Trials• FFT• SPOKES• High Need• Fostered children• Callous-unemotional children

3. Measures of Parenting & Cost-effectiveness

Example of Evaluation of a programme

“Supporting Parents of children with autism”

Element 1: Target population• Aims; Assist parents in managing children’s

behaviour, increase parental knowledge about autism

Strengths: • Well described characteristics of the target

population, fairly well specified expected outcomes, appropriate programme classification: targeted and specialist

Weaknesses • Lack of appropriate intake need assessment• Lack of appropriate measures of programme

expected outcomes

Self-rating: 4 Academy Rating : 3

Element 2:Content and DeliveryStrengths• Evidence of some theoretical framework (behavioural, cognitive)• Flexibility in delivering sessions & wide range

of attractive resources

Weaknesses• Lack of major key ‘autism specific’ theories

(theory of mind, developmental milestones)• Poor content (lacks consideration of

communication difficulties)• Poor balance between didactic component &

active skill training, e.g. limited in role-plays and group exercises.

Self-rating: 3 Academy Rating : 1

Implications for training & workforce development

1st Training Offer: Parenting groups implemented

1st Training offer: Implementation significantly correlated with practitioners’ level of qualification

Implications for training & workforce development

* Does not include SF

Implications for training & workforce development

1st Training Offer: Implementation significantly correlated with perceived relevance of the training

Implications for training & workforce development

Stage 2: Parenting groups implemented

* Phase 1

*

(1) Use evidence-based programmes

Effect sizes av 0.6 with EB progs (Barlow et al 2002) Effects 0-0.2 otherwise

Eg Fort Bragg, Homestart,

Oxford Home Visiting (Weisz et al 1998) 0

0.1

0.2

0.3

0.4

0.5

0.6

not EB Ev-Based

(2) Target accuratelyImprovement in antisocial behaviour, according to initial

severity level (Scott 2005)

0

0.5

1

1.5

2

2.5

3

3.5

Mean

Z sc

ore

1st 2nd 3rd 4th

Initial severity by quartile

Before

Follow up

Clinical trial worst 2%

SPOKES prevention trial targeting worst 20%

PALS prevention targeting by geographical area

(3) Be engaging!

Percent initial enrolment (prevention) depends on trust & liking, and:

• Accessible• Available• Affordable• At convenient time

(after Pugh 1997)

05

1015202530354045

50

Ave Best

(4) Keep attendance up

9

9.5

10

10.5

11

11.5

12

12.5

13

Before Follow Up

Control

All allocated

Attend 5+

Changes in Sensitive Responding according to number of sessionsAttended in PALS trial(Scott, O’Connor & Futh inPress)

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

child

agg

ress

ion

impr

ovem

ent

lowest lower third middle third upper third

Skill of professional in delivering programme

Child outcome and professional skill

(5) Quality, quality, quality(Scott, Carby and Rendu 2007)

(6) Build in evaluation and review

• For some areas, do Randomized Controlled Trials to answer questions eg how many sessions are enough, which programmes have which effects, etc– Not to do so is to set out to sea without a map– This is not a luxury, it is prudent and cost- saving

• For all areas, collect pre-post data on effectiveness, measure whole population – otherwise you don’t know who you are excluding (already done as SATS)

ThinkThink

Feel Do