Risk of suicide and spouse's psychiatric illness or suicide: nested case-control studyBMJ 2003; 327 doi: https://doi.org/10.1136/bmj.327.7422.1025 (Published 30 October 2003) Cite this as: BMJ 2003;327:1025
- Esben Agerbo (), assistant professor1
- Accepted 6 August 2003
Suicides cluster in families with histories of psychiatric disorders and suicides.1 2 Genetic and environmental factors may play a role in the familial aggregation of suicides but are inseparable in most studies. Because married couples are usually genetically unrelated, studying them can identify environmental factors and means of protection. Your spouse dying or your spouse having a psychiatric disorder is stressful; mortality is increased in the surviving spouse.3 I investigated the association between a spouse's psychiatric illness, suicide, and other causes of death and own suicide.
Participants, methods, and results
I got data by linking population based registers using unique personal identification numbers, which are assigned to all people living in Denmark. Until 1993, suicide was defined as ICD-8 (international classification of diseases, 8th revision) codes E950-E959; after 1994, ICD-10 codes X60-X84. I matched each person aged 25 to 60 years who had committed suicide during 1982-97 to a random subsample of 20 people stratified by sex and year of birth. I only enrolled people who had been living in Denmark for the past two years. I identified all spouses and children who were living with these people on 31 December two years before the suicide. I got admission and discharge dates and diagnoses from the Danish psychiatric central register, which has monitored all facilities for inpatients since 1969.1 2 I analysed the data with conditional logistic regression.
I identified 9011 suicides, 180 220 controls, and 111 172 spouses (table). People whose spouse had ever been admitted with a psychiatric disorder were at greater risk of committing suicide, particularly if the first admission had been recent (risk ratio 5.09; 95% confidence interval 3.53 to 7.35). People whose spouse had died, especially by suicide (21.69; 11.10 to 42.37), were also at a greater risk of committing suicide. An adjusted analysis found weaker associations, but a spouse's suicide remained indicative of own suicide (P = 0.01).
A greater risk of committing suicide was associated with a spouse who had been admitted to hospital with a psychiatric disorder or had died, particularly if the cause was suicide. Both spouses having been admitted was not associated with any additional effect.
Risk of suicide was particularly great in people whose spouse had been first admitted within the previous two years, which advocates a causal relationship. A severe mental illness can have an impact on other family members' social life, leisure time, and economy. Shared environmental factors may put cohabiting partners at risk of the same diseases.4 That the effect of assortative mating, measured here as both spouses previously admitted, however, did not further increase the suicide risk is striking. This finding suggests that behavioural traits associated with some psychiatric disorders, or that the admission itself, increase the suicide risk in spouses.
Research based on routine registers has limitations—for example, data about episodes of illness that did not lead to admission and attempted suicides are unavailable.2 Conjugal bereavement has an impact on mortality among surviving spouses,3 and bereavement due to suicide increases own risk of suicide more than bereavement after other modes of death. Suicide may induce particularly difficult grief,5 and suicide of a spouse increases the other spouse's awareness of suicide as a possible means to end grief. Assortative mating for suicidal behaviour is less likely because preference for mating between people with psychiatric illness did not increase the risk of suicide.
I thank Majella Byrnes and Preben Bo Mortensen, National Centre for Register-based Research.
Contributors EA is the sole contributor.
Funding Stanley Medical Research Institute, Danish National Research Foundation.
Competing interests None declared.
Ethical approval Not needed.