@EdinUniUsher
www.ed.ac.uk/usher
Dr. Louise Marryat
Alexandra Blair
Professor John Frank
@ace_scot
@lmarryat
The Scottish ACE Study: exploring Adverse Childhood Experiences in Scotland
@EdinUniUsher
www.ed.ac.uk/usher
ACEs in Scotland?
• ScotPHN report, May 2016
• ‘although, data exists on various aspects of household dysfunction in Scotland, no published studies exist to date of the prevalence specifically of ACEs in the general population of Scotland’
• If 9% people in England have 4+ ACEs, Scotland is likely to be worse
• We know Scotland has poorer health outcomes than rest of UK which can’t be explained purely by deprivation…could ACEs be the answer?
@EdinUniUsher
www.ed.ac.uk/usher
Study 1AIM: to explore to what extent ACEs could be gathered using prospective cohort data, and what prevalence levels look like in this general population of Scottish children.
RESEARCH QUESTIONS:1. What are the levels of ACEs in the general
population of Scottish children?2. What factors predict a) having any ACEs
and b) having more ACEs in the general population?
3. To what extent does relative poverty account for the burden of ACEs in Scotland?
@EdinUniUsher
www.ed.ac.uk/usher
The Growing Up in Scotland study sample• Used Birth Cohort 1 – born 2004/5
• Data from sweeps 1-7 (10 months to age 8)
• National sample taken from child benefit records
• At sweep 1 there were 5,217 children recruited
• At Sweep 7 there were 3,456 children in the study (66% of sweep 1 children)
• Calculated ACE scores using data from every sweep
@EdinUniUsher
www.ed.ac.uk/usher
Data availability
Emotional abuse
Physical abuse
Sexual abuse
Neglect
Emotional neglect
Parents separated
or divorced
Witnessed domestic
abuse
Lived with substance misuser
Household member in
prison
Live with someone
with mental health
problems
@EdinUniUsher
www.ed.ac.uk/usher
Levels of ACEs by cohort
0 5 10 15 20 25 30 35 40
ACE9: Mental health problems
ACE6: Parents ever divorced or separated
ACE2: Physical abuse
ACE4: emotional neglect
ACE8: Drug or alcohol misuse
ACE7: Domestic violence
ACE10: Parent in prison
Wales KP GUS
@EdinUniUsher
www.ed.ac.uk/usher
Levels of ACEs
35 34.8
19.6
7.5
3
49.5
24.9
12.5
6.9 6.2
0
10
20
30
40
50
60
None 1 2 3 4 or more
GUS Kaiser Permanente
@EdinUniUsher
www.ed.ac.uk/usher
Children’s odds of having 3+ ACEs at age 8 were associated with…
Being male (O.R.=1.5)
Having a mother aged <20 at the birth of 1st child
(O.R.=2.1)
Living in a household in
lowest income band (O.R.=6.5)
Living in an urban area (O.R.=1.8)
@EdinUniUsher
www.ed.ac.uk/usher
ACE scores by income
8
26.633.8
42.252.8
10.8 3.2 1 0.5 1
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Bottom incomequintile
2nd quintile 3rd quintile 4th quintile Top incomequintile
None 1 2 3 4 or more
@EdinUniUsher
www.ed.ac.uk/usher
Income doesn’t explain everything however…• Calculated the Population Attributable Risk(PAR)
• ‘the proportion of the health outcome in an entire population, which is attributable to the exposure’
• Exposure = below relative poverty line; health outcome= 3+ ACEs
• Results suggest that in 22% of cases where children experience 3+ ACEs, this experience can be directly attributed to poverty
• Moving children above poverty line would decrease the proportion of children experiencing 3+ ACEs by c.7%.
@EdinUniUsher
www.ed.ac.uk/usher
Study 2 (currently under peer review)
Can community resources help mitigate the effects of household poverty on ACE incidence?
RESEARCH QUESTION:1. Is the relationship between household
poverty and the cumulative incidence of adverse childhood experiences modified by families’ access to community resources?
This study uses the same sample as study 1.
@EdinUniUsher
www.ed.ac.uk/usherObjectives
1. Effect modification: Assess whether the association between household income and 8-year cumulative ACE incidence is modified by households’ access to:
• Non-precarious housing
• Transportation services
• Breastfeeding counselling
• Childcare services
• A local public park or playpark (among urban residents)
2. Proportion eliminated: Assess the extent to which income inequalities in 8-year cumulative ACE incidence could be eliminated if access to the identified resources were available to all
@EdinUniUsher
www.ed.ac.uk/usher
Low income ACE incidenceHealth and
Social Outcomes*
Effect modifiersor “mitigating
factors” ?
The experience of low-income can vary according to the relative generosity of state investment in benefits, policies and resources (simplified DAG)
(Bambra & Eikemo 2009, Bambra 2011, O’Campo et al. 2015)
Background to study 2
* Not available in GUS cohort yet
@EdinUniUsher
www.ed.ac.uk/usherProportion of income-based inequality in 8-year cumulative ACEs (= >/3) eliminated if all families fully exposed to mediator(excess relative risk scale)
• Interpretation: The % of the relative income inequality in cumulative ACE incidence that could be eliminated by providing the amenity/service identified as a mediator (left column) to all families in the community (1. VanderWeele TJ. Explanation in Causal Inference: Methods for Mediation and Interaction. Oxford: Oxford University Press, 2015. 2. VanderWeele TJ. Mediation analysis: A practitioner’s guide. Ann Rev Public Health 2016; 37(1):17-32)
Mediators Proportion eliminated (%, 95% CI)
Park proximity -1.0 (-8.2, 5.5)
Housing 0.8 (-15.0, 13.0)
Transportation 20.8 (3.4, 41.0) * p< 0.05
Breastfeeding Education 7.5 ( -7.8, 24.0)
Childcare -20.0 (-39.0, 5.1)
@EdinUniUsher
www.ed.ac.uk/usher
How do we interpret these results?
Low income ACE incidence
Transportation
• Sense of control, self-efficacy
• Less stress and fatigue from compulsory walking
• Access to resources outside local-area
Syme 1996, Markovich 2011, Bostock 2001, Bambra 2007, Fairburn 2005, Dieterich 2013, Chapman 2004
Decisional latitude Empowerment Independence
@EdinUniUsher
www.ed.ac.uk/usher
Current available outcomes within the GUS data
@EdinUniUsher
www.ed.ac.uk/usherACE scores and health outcomes by age 8
0
5
10
15
20
25
30
35
Overweight orobese
AdditionalSupport needs
General healthFair or bad
Disability or LSI 2+ accidents orinjuries in past
year
None 1 2 3 4 or more
@EdinUniUsher
www.ed.ac.uk/usherSocial, emotional and behaviour development by
ACE scores
0
5
10
15
20
25
30
35
None 1 2 3 4+
*Not statistically significant
@EdinUniUsher
www.ed.ac.uk/usher
0
10
20
30
40
50
60
70
80
90
100
Supervised following court/hearing Parent expect child to stay on at schoolpost-16
0 1 2 3 4+
Other outcomes by ACE scores
@EdinUniUsher
www.ed.ac.uk/usher
Strengths and weaknesses of the study
Strengths Weaknesses
Response rates are high Possibly missing children with most ACEs
Prospective data more accurate? Is it experience or memory/impact that is important?
Rich background data on child and household
Two ACEs not able to assess –arguably most serious
Social desirability/fear increasing non-reporting?
Parents may not be aware of some instances of ACEs e.g. abuse
@EdinUniUsher
www.ed.ac.uk/usher
Discussion
• Proportions of children experiencing 1+ ACE higher than retrospective studies• Increase in parental
separation/recognition of mental health issues etc.?
• Poor recall in retrospective studies due to lack of early memory, mood at time of reporting, & ability to recall only what aware of at time
• Cohort differences• Time period measured
• Poverty and ACEs: supporting proportionate universalism?
@EdinUniUsher
www.ed.ac.uk/usher
Future work• Explore linked routine health data in relation to ACE scores
• Generational cycles of ACEs, and parental neurodevelopmental disorders
• Investigate resilience factors – supportive adults and family/school stability
• Prospective ACEs up to age 18
• Health, education and economic outcomes in adulthood
• Other suggestions??
@EdinUniUsher
www.ed.ac.uk/usher
Conclusions• Two-thirds of Scottish children have 1+ ACE by
age 8
• Compares unfavourably with previous studies
• Measurement differences make it hard to compare directly
• Clear that many Scottish children are experiencing far from ideal childhoods
• Few experiencing 4+ ACEs – the strong predictor of later negative outcomes
• Experience of ACEs was strongly associated with living in poverty
• Could better transport help support low income families to avoid ACEs?
@EdinUniUsher
www.ed.ac.uk/usher
Acknowledgements• Co-authors Professor John Frank and Alexandra Blair
• Children and families who participated, & continue to participate, in GUS
• The GUS team at ScotCen and CRFR, particularly Paul Bradshaw and Lesley Kelly
• Sara Dodds, Scottish Government
• Farr@Scotland who funded the work
@EdinUniUsher
www.ed.ac.uk/usher
Any questions or comments?