Family dynamics and self-harm and suicidality in children and adolescents: a systematic review and meta-analysis

Family dynamics are implicated in self-harm and suicidality among children and adolescents. However, whether negative family dynamics confer a prospective risk and positive family dynamics confer a protective effect is not understood. To address this research gap, we aimed to summarise the prospective, longitudinal evidence examining the relationship between family dynamics, self-harm, and suicidality (ie, suicidal thoughts and behaviours) during childhood and adolescence.

In this systematic review and meta-analysis, we searched MEDLINE, Embase, PsycINFO, and ERIC, as well as CINAHL, without language restrictions, from the date of database inception to May 20, 2025. Observational studies were eligible if they prospectively followed up a cohort of children and adolescents (age <20 years) from a general population, community, or school showing either no elevated risk or typical development of self-harm and suicidality for a minimum of 12 months. Eligible studies had to have included prospectively measured family dynamics, including positive (eg, emotionally supportive) and negative (eg, harsh, aversive, or punitive) parenting behaviours and family functioning (eg, family cohesion) or dysfunction (eg, discord or conflict) in childhood or adolescence. Extracted data consisted of study-level information and characteristics, participant characteristics, descriptions of study measures, and study results. Extracted data were entered into Covidence for consensus. The primary outcomes were self-harm (ie, thoughts of non-suicidal self-harm or self-injury, and non-suicidal self-harm or self-injury) and suicidality (ie, suicidal ideation and suicide attempt) over any reporting period. We examined individual and combined outcomes using a random-effects model. We assessed study quality using a modified Newcastle-Ottawa scale. This study is registered with PROSPERO (CRD42023434804).

We screened 3860 articles and retained 38 studies, of which all 38 contributed to the narrative synthesis and 24 to the quantitative analyses. Altogether, the studies comprised 101 879 children and adolescents. Most study samples were from the USA (12 [32%]) or China (11 [29%]), with exposure and outcome ascertainment 12 months apart (25 [66%] studies) from age 10 years to 19 years (36 [95%]). Data stratified by sex, gender, race, or ethnicity were not consistently available. Female participants constituted a larger proportion of the samples than did male participants (ranges 42-100% vs 28-58%). Exposure to negative parenting was associated with an increased likelihood of combined self-harm and suicidal ideation (OR 1·29 [95% CI 1·15-1·46]) and non-suicidal self-harm or self-injury (1·46 [1·25-1·71]), but not suicidal ideation (1·07 [0·92-1·24]). Negative parenting practices and continuously measured self-harm and suicidality were not significantly associated. Positive parenting practices were not associated with suicidal ideation or with combined self-harm and suicidal ideation. Family dysfunction was longitudinally associated with an increased combined likelihood of self-harm and suicidality (OR 1·29 [95% CI 1·13-1·48]) and non-specific self-harm (1·70 [1·10-2·63]), but not suicide attempt (1·24 [0·93-1·66]). The overall rating of study quality was moderate (mean 6·5 of 10·0 stars [SD 1·29]).

Negative parenting practices and family dysfunction seem to precede self-harm and suicidality among children and adolescents. Reducing negative family dynamics could alleviate these severe mental health concerns in the short term and assisting families to minimise early-life exposure to these dynamics could prevent the onset of self-harm and suicidality.

Frederick Banting and Charles Best Canada Graduate Scholarship Doctoral Award Program and Canada Research Chairs Program.