Suicide risk in relation to socioeconomic, demographic, psychiatric, and familial factors: a national register-based study of all suicides in Denmark, 1981-1997.

Suicide risk was addressed in relation to the joint effect of factors regarding family structure, socioeconomics, demographics, mental illness, and family history of suicide and mental illness, as well as gender differences in risk factors. Data were drawn from four national Danish longitudinal registers. Subjects were all 21,169 persons who committed suicide in 1981-1997 and 423,128 live comparison subjects matched for age, gender, and calendar time of suicide by using a nested case-control design. The effect of risk factors was estimated through conditional logistic regression. The interaction of gender with the risk factors was examined by using the log likelihood ratio test. The population attributable risk was calculated. Of the risk factors examined in the study, a history of hospitalization for psychiatric disorder was associated with the highest odds ratio and the highest attributable risk for suicide. Cohabiting or single marital status, unemployment, low income, retirement, disability, sickness-related absence from work, and a family history of suicide and/or psychiatric disorders were also significant risk factors for suicide. Moreover, these factors had different effects in male and female subjects. A psychiatric disorder was more likely to increase suicide risk in female than in male subjects. Being single was associated with higher suicide risk in male subjects, and having a young child with lower suicide risk in female subjects. Unemployment and low income had stronger effects on suicide in male subjects. Living in an urban area was associated with higher suicide risk in female subjects and a lower risk in male subjects. A family history of suicide raised suicide risk slightly more in female than in male subjects. Suicide risk is strongly associated with mental illness, unemployment, low income, marital status, and family history of suicide. The effect of most risk factors differs significantly by gender.