Questionnaire

Questions from the Centers for Disease Control and Prevention short ACE tool [30] and the Short Child Maltreatment Questionnaire [31] were used to retrospectively measure respondents’ exposure to ACEs < 18 years of age. ACEs were grouped into eleven types (see Additional file 1: Table S1) and respondents categorised as reporting 0, 1, 2–3 or ≥ 4 ACEs for analysis. Consistent with multiple ACE studies globally [3], for the purposes of this study the ACE count is used as an independent variable in order to examine a cumulative measure of childhood adversity and provide comparability to other ACE studies. The ACE tool has been validated as a reliable tool for retrospective assessment of adverse childhood experiences [32, 33]. However, we could not identify a tool validated for use with adults to retrospectively measure childhood community resilience assets or one that had been used in national surveys. Therefore, we used questions consistent with established resilience measures (Child and Youth Resilience Measure [34]) with the addition of measuring access to a trusted adult in childhood; a factor previously related to resilience and ACEs [22]. The community resilience assets measured were: knowing where to get help, having opportunities to apply one’s skills, being treated fairly, enjoying community culture, having supportive friends, having people to look up to, and having a trusted adult available (see Additional file 1: Table S1 for questions and responses categorised as having each asset). Common childhood conditions measured were asthma, allergies, headaches, and constipation/diarrhoea (as a single digestive conditions variable). Responses were dichotomised into never/rarely and sometimes/often for analysis. Self-rated overall health was reported as either excellent, very good, good, fair or poor and dichotomised into poor childhood health ‘yes’ (fair or poor) or ‘no’ (excellent, very good or good). High secondary school absenteeism was categorised as ‘yes’ (those missing > 20 days per year) and ‘no’ (missing ≤20 day per year).