Early-life protective factors for adolescent self-harm and suicidality: a longitudinal cohort study in Australia

Suicide and self-harm are serious health concerns and can emerge in childhood and adolescence. Research into childhood self-harm and suicidality and prevention has focused predominately on risk factors, whereas protective factors remain understudied. We aimed to explore the type and timing of early-life (age 6-13 years) factors that are protective against self-harm and suicidality in adolescence (age 14-15 years).

In this cohort study, we used four predictor waves (waves 4-7) and one outcome wave (wave 8) of the Longitudinal Study of Australian Children. Children in the cohort were born in 2003-04 (wave 1) and have been surveyed every 2 years. Predictors were aspects of childhood wellbeing that were drawn from an existing multidimensional participatory child wellbeing framework (the Nest framework), which used domains of valued, loved, and safe; material basics; healthy; learning; and participating. Dimensions of childhood wellbeing were coded by a previously developed indicator dashboard. Each dimension was operationalised with multiple subdomains, which were the key predictors. We used logistic regressions to model the degree to which these wellbeing dimensions, measured across four timepoints in middle childhood and pre-adolescence (age 6-13 years), predicted adolescent (age 14-15 years) self-harm ideation and behaviour and suicidal ideation and attempts in the past 12 months. We involved people with related lived experiences in the study design and implementation.

3044 children were in the wave 4-8 analytical sample: 1570 (51·6%) male participants and 1474 (48·4%) female participants. 3034 (99·7%) children in the cohort were born in Australia, but 858 (28·2%) children in the wave 4-8 analytical sample had at least one parent born overseas and 76 (2·5%) children were Aboriginal or Torres Strait Islander Australians. Across all ages, most children had most of their wellbeing needs met on individual subdomains of wellbeing. No variable showed consistent associations across ages and across the four outcome variables. The strongest protective factors against self-harm ideation were a good state of mind (odds ratio 0·55, 95% CI 0·42-0·71) and sense of belonging (0·60, 0·46-0·77) at age 12-13 years. For self-harm behaviour, relationships with parents (0·45, 0·21-0·97) and involvement in the community (0·53, 0·32-0·89) at age 10-11 years were most protective. For suicidal ideation, the strongest protective factors were sense of belonging at age 12-13 years (0·57, 0·42-0·78) and school satisfaction at age 10-11 years (0·32, 0·12-0·89). For suicide attempts, protective factors included school safety (0·59, 0·39-0·89) at age 6-7 years and school satisfaction at age 10-11 years (0·47, 0·26-0·85).

We identified various childhood wellbeing factors that relate to adolescent self-harm and suicidality across the four predictor waves, suggesting that the protective factors for self-harm and suicide change as children grow and develop. The range of protective factors and their changing importance in childhood years suggest that timing of policy interventions should be considered carefully. For instance, interventions targeting school safety likely have long-term value when targeting children in primary school years.

Australian National University-Australian Capital Territory Government Wellbeing Framework and National Health and Medical Research Council.