Lifetime substance use disorder and suicidal behaviour: a population-based familial co-aggregation register study

Abstract Substance use disorders (SUD) are associated with suicidal behaviour, but less is known regarding the extent to which familial factors influence these associations across different types of SUD. We performed a national register study including 4,209,615 individuals born in Sweden 1958–1999 and followed 1973–2020. Exposure was life-time SUD diagnosis, including alcohol use disorder (AUD), drug use disorder (DUD) or dual diagnosis (AUD + DUD). DUD was further classified into substance type (e.g., cannabis, opioids, sedatives or stimulants). Outcome was life-time suicidal event, operationalized as either suicide attempt or death by suicide. Log-linear regression models estimating risk ratios (RR) were fitted with adjustment for sex, birth year, socioeconomic factors and psychiatric co-morbidity. Of 258,046 individuals with SUD, 23.0% ( n = 59,436) had any life-time suicidal event compared to 1.5% ( n = 59,784) in 3,951,569 persons without SUD. SUD was associated with increased risk of any suicidal event (Adjusted RR: 6.4 [6.3–6.5]), suicide attempt (6.9 [6.8–7.0]) and suicide death (3.3 [3.2–3.5]). Elevated suicide risk was observed across all SUD, with greatest risk in those with AUD + DUD. Larger RR suicidal events were observed in relatives of individuals with SUD as genetic similarity increased (e.g., full siblings: 2.6 [2.6–2.7]; half siblings: 1.5 [1.5–1.6]). This pattern was observed for all types of SUD, with highest RR in individuals with AUD + DUD relatives (e.g., full siblings: 3.5 [3.4–3.6]; half siblings: 1.7 [1.7–1.8]). Our results suggest that SUD of all substances and suicidality share familial factors, with the clinical implication that family history of SUD constitutes a risk factor for suicidal behaviour.