Transgenerational concordance in parent-to-child transmission of suicidal behaviour: a retrospective, nationwide, register-based cohort study of 4 419 642 individuals in Denmark.
Ranning A., Madsen T., Hawton K., Nordentoft M., Erlangsen A.
BACKGROUND: Suicidal behaviour runs in families, but the nature of transgenerational concordance needs elucidation. The aim of this study was to examine parent-to-child transmission by investigating whether presence and nature of parental suicidal behaviour was associated with suicidal behaviour in children. METHODS: We did a retrospective, nationwide, register-based cohort study in Demark using register data. We included all individuals born after 1953 who were 10 years or older and who were recorded as living in Denmark at some point between Jan 1, 1980, and Dec 31, 2016. Adults listed as living with their child at first registration in the Civil Registration System were considered as parents; later records of different legal parents allowed identification of potential step-parents. Self-reported ethnicity data were not available. Exposure to parental suicide attempt and suicide was identified using information from hospital contacts and causes of death from national registers. The examined outcomes were suicide attempt and death by suicide. We calculated incidence rate ratios (IRRs) and cumulative hazards for children's suicide attempt and suicide, taking into account type of parental suicidal behaviour, child's age of exposure, and sex. FINDINGS: In total, 4 419 642 individuals aged 10-63 years were observed during 1980-2016. Of these individuals, 150 222 (3·4%) were exposed to one or more parents with a suicide attempt, 31 564 (0·7%) to at least one parent who died by suicide, and 12 834 (0·3%) to both events. Individuals exposed to parental suicide attempt had higher rates of suicide attempt (IRR 2·72 [95% CI 2·33-3·17]) than individuals exposed to parental suicide (1·77 [1·50-2·09]) when compared with unexposed individuals. Higher rates of suicide were found for individuals exposed to parental suicide (IRR 3·18 [95% CI 2·84-3·58]) than for those exposed to parental suicide attempt (2·37 [2·19-2·57]). The cumulative hazard of suicide attempt was 0·07 for individuals exposed to parental suicide attempt, and the cumulative hazard of suicide was 0·009 for individuals exposed to parental suicide. Individuals exposed to parental suicide had higher odds of violent suicidal methods than those exposed to suicide attempt alone (odds ratio 2·0 [95% CI 1·7-2·3]). INTERPRETATION: A concordant pattern of higher rates of the same type of suicidal behaviour as the one of the parents was observed, including type of suicide method. Preventive, family-oriented interventions are warranted to mitigate familial transmission of risk, as are clinical considerations of familial exposure in risk assessment of patients. FUNDING: Mental Health Services, Capital Region of Denmark.