1. CEBEI activity and the IY pooled data trial Bangor and Cardiff Annual Conferences 2017
Professor Judy Hutchings, OBE
Centre for Evidence Based Early Intervention, Bangor University

2. CEBEI mission
To bring evidence based interventions to public services particularly in the area of child antisocial and behavioural problems and to support intervention services to be effective in supporting disadvantaged children and families

3. Achieved through
Researching existing evidence based programmes in service settings
Developing and evaluating new interventions
Training staff in evidence based programmes
Dissemination through publications and conference presentations

4. completed studies being written up but not being presented today
School based small group Dina RCT trial – showed improved achievement of social problem solutions, prosocial solutions and teacher set social-emotional targets
The school readiness parent programme delivered by school staff to parents of newly enrolled children - Significant benefits over 6 months in observed academic and socio-emotional coaching, encouragement, praise and embedded in Powys and Flintshire

5. Current ongoing research projects – not presented today
Evaluating an adapted version of the web based parenting programme for classroom support staff - Anwen Rhys Jones – Mres student (and funding application made for a PhD to extend this project and develop Welsh medium materials
Supporting foster carers, a review literature on foster carer training and evaluation of delivery of support on Anglesey, Sue Layland – PhD student
The relationship between changes in parenting skills, maternal depression and child behaviour over an IY parent group, Jess Stewart - DClinPsych .

6. Current collaborative research projects –
England ADHD trial comparison of IY and NFPP with Southampton University,
Trials of Sinovuyo parenting programme in South Africa (and an adaptation trial in the Philippines) Oxford and Cape Town Universities
IY parenting outcomes - a pan-European pooled data trial led by Frances Gardner Oxford University. - A literature review effective components of parenting programmes
Other collaboration – membership of the WHO Violence Prevention Alliance – parenting sub-group, parenting for lifelong health.

7. Are parenting interventions for disruptive behaviour likely to affect social disparities in child outcomes? Pooling data from multiple intervention trials
The team – Prof Frances Gardner, Prof Stephen Scott, Dr Patty Leitjen, Prof Judy Hutchings, etc.
funded by the National Institute for Health Research (NIHR)’s Public Health Research Programme (Grant Ref Number ).

8. Parenting Interventions
Much known about the main effects of these interventions on child disruptive behavior
In society, these child outcomes are highly patterned by SES.
But it is much less clear if intervention outcomes are also patterned by SES

9. Moderators and equity effects
As interventions go to scale – we need to ask wider ‘equity’ questions about population effects of interventions on social disparities
Is there an ‘Inverse care’ effect for most disadvantaged families, whereby we help the most needy least? Access vs. effects.
Or, will these interventions narrow the gaps?

10. Inconsistent moderator findings
Two large reviews of parenting intervention predictor effects (567, 451 cites) (Lundahl, 2006; Reyno & McGrath, 2006)
Tested e.g., poverty, lone parent, parental depression
“more disadvantaged and distressed families benefit less”
Recent large trials give a different picture
“no moderator effects” or “more disadvantaged and distressed families can benefit more” (Gardner et al., 2009, 2010; Beauchaine et al., 2005).
Thus unclear what potential effects on social disparities

11. Problems with usual approaches to
studying moderators?
Individual trials
Well powered to test main effects
Not powered to test moderation effects (Brown, 2012)
Meta-analyses such as Cochrane reviews
More power and precision for main effects BUT...
Moderation only by trial level characteristics, eg by average level (per trial) of family characteristics
Means all variability within trials in participant characteristics is lost

12. Incredible Years Pooling Study
Who benefits from parenting interventions?
Combining data from near total sample of randomised trials of the Incredible Years parenting intervention in Europe.
CEBEI contributed to 5 of the trials, 2 in Wales, Birmingham, Portugal and Ireland
England
N = 112
N = 214
Portugal
N = 124
Sweden
N = 62
Wales
N = 153
N = 174
N = 76
Norway
N = 75
Ireland
N = 149
N = 103
N = 141
N = 161
Netherlands
N = 99
N = 156
Integrative Data Analysis
N = 1799

13. Incredible Years: What and Why?
Developed in USA, sessions, groups for parents of 3-8 year olds with disruptive behaviour;
Strong evidence base—
many independent RCTs in Europe;
In UK: widely disseminated in children’s health & family services

14. Primary project goal
To examine if the IY parenting intervention is less or more effective for reducing disruptive child behaviour in the most disadvantaged and most distressed families.
Research Questions
Is there a differential effect of the IY parenting intervention on reduced disruptive child behaviour:
by family socio-economic status?
by ethnic minority status?
by child and family problem severity?
by child characteristics - age, gender

15. Effects of IY parenting, per trial
Intervention effect size on child disruptive behavior
Cohen’s d (Eyberg Child Behavior Inventory)

16. Low income (SES)
No evidence to suggest IY effect on child disruptive behaviour varies by low vs average-high income (p=0.87).
No moderation effect.

17. No parent employed (SES)
- No evidence to suggest that IY effect varies by parent unemployment vs employed (p=0.42).
- No moderation effect.

18. Lone parent (SES)
- No evidence to suggest that IY effect varies by lone parent/ vs. partnered (p=0.99).
- No moderation effect.

19. Ethnic Minority
No evidence to suggest IY effect varies by ethnic minority vs. not (p=0.48).
Mostly UK, Holland, 7 trials

20. Parental depression Evidence of effect moderation (p=0.01)
- Children of parents with higher depression benefit more.

21. Baseline child disruptive behaviour problems (ECBI)
Evidence to suggest effect varies with baseline level of behaviour problems (p=0.015).
Children with higher scores at baseline benefit more

22. Child age
No evidence to suggest that IY effect varies with child age (p=0.95).
No moderation effect

23. Child gender
- Evidence to suggest that IY effect varies by gender (p=0.04).
- moderation effect whereby boys benefit more.

24. Conclusions
Children appear just as likely to benefit from IY, in terms of change in disruptive behaviour, if they:
Come from a low income, jobless or lone parent family, or are from an ethnic minority.
However, if the family has high levels of distress at outset, i.e. more severe child disruptive behaviour, or parent with high levels of depression, they are more likely to benefit.
No age effects - interesting, first well powered study on this; does not fit with thrust of neuroscience /early intervention work

25. For further information please visit our website
Thank you
For further information please visit our website