The effect of suicide bereavement on suicidal behaviour: a national cross-sectional survey of young adults in the UK

Abstract

BackgroundProviding support to people bereaved by suicide has become a key priority for suicide prevention strategies in England, Scotland, Wales, Northern Ireland, and many other developed countries. Stigma, social modelling, depression, heritability, and other factors are thought to enhance this group's risk of suicidality. Few studies have described the nature or magnitude of the effects of suicide bereavement on family, friends, and other contacts, or assessed the effectiveness of post-bereavement support. We aimed to test the hypothesis that young adults bereaved by the suicide of a close contact have an increased risk of suicidal thoughts and suicide attempts compared with those bereaved by other causes of sudden death of a close contact. MethodsWe sampled 631 898 staff and students on the email distribution lists of 37 UK higher education institutions in 2010. A mass email invited adults who had experienced a sudden bereavement to complete an online survey measuring a range of quantitative variables, with lifetime suicidal thoughts and lifetime suicide attempts as the primary outcomes. Inclusion criteria were age 18–40 years and sudden bereavement of a close contact since the age of 10 years. Multiple regression was used to compare those bereaved by suicide, those bereaved by accidental death, and those bereaved by sudden natural death (the baseline group), adjusting for covariates chosen a priori: age, sex, social class, family history of suicide (excluding index bereavement), past psychological problems, and kinship to the deceased (non-relative vs blood relative). FindingsOf 3686 bereaved adults fulfilling the inclusion criteria (from 4631 consenting to participate), 658 had been bereaved by suicide, 761 by sudden accidental death, and 2267 by sudden death due to natural causes. The sex ratio of men to women was 1:4. Because the denominator for the population of people exposed to sudden bereavement could not be estimated reliably, calculation of a response rate was not possible. Multiple logistic regression showed a non-significant excess risk of suicidal ideation (odds ratio 1·10, 95% CI 0·87–1·40) and a significant excess risk of suicide attempt (1·63, 1·06–2·50) in the group bereaved by suicide. Testing for an interaction between type of bereavement and kinship to the deceased found no statistically significant interactions. InterpretationThe adjusted findings suggest that young adults who have been exposed to suicide bereavement may be at increased risk of suicide attempts compared with young adults bereaved by other causes of sudden death, whether they were related to the deceased or not. Clinically, this finding suggests that inquiring about a history of suicide in unrelated close contacts should be added to a screen for family history of suicide within routine psychosocial assessments. From a public health policy perspective, the findings confirm the clinical assumption that this group are at increased risk for suicidal behaviour. Further work is needed to identify the nature and effectiveness of the support needed for this population. FundingALP is funded by a Medical Research Council Population Health Scientist Fellowship.