showed a high concordance of attempted suicide in a surviving twin with suicide in monozygotic twins (38 percent), but not in dizygotic co-twins (0 percent3). A large Australian study of almost 3000 twin pairs confirmed these findings (Statham et al., 1998). If a monozygotic twin attempted suicide, his/her co-twin had a 17.5-fold increased risk of having made an attempt. In controlling for other risk factors for suicide, such as mood disorder, substance abuse, trauma, personality problems, and life events, a family history of a suicide attempt still conveyed a 4-fold increased risk for the co-twin making an attempt. All studies of suicide in twins discussed here were carried out in twins who were raised together, with ostensibly a shared environment, thereby controlling for environmental effects. To date, there are no published findings for suicide for twins raised apart, which is another way to examine this issue. The attempts among twins did not cluster in time, making imitation a less likely explanation.

In this study, genetic modeling showed that 45 percent of the variance for suicidal thoughts and behavior was genetic, which suggests a continuity between ideation and attempt (Statham et al., 1998). Glowinski et al. (2001) studied 3416 Missouri female adolescent twins and found that genetic and shared environmental influences together accounted for 35 percent to 75 percent of the variance in risk. The twin/cotwin suicide attempt odds ratio was 5.6 (95 percent confidence interval [CI] 1.75–17.8) for monozygotic twins and 4.0 (95 percent CI 1.1–14.7) for dizygotic twins after controlling for other psychiatric risk factors.

Family Studies

Several family studies have compared the risk for suicide or suicide attempt in the first degree relatives of individuals who have completed suicide, compared to the rate among relatives of control probands. Although the methodology varies to some degree, the results consistently point to a 4-fold increase in risk among relatives of suicide probands compared to the relatives of controls. Tsuang (1983) reported that the rates of suicide in the relatives of patients who completed suicide were higher than the rates among the relatives of patients who did not complete suicide. This suggests that something other than mental disorder is being transmitted to increase the familial transmission of suicide. Similarly, Egeland and Sussex (1985) found that the rate of suicide in the pedigrees of the Old Order Amish showed a marked degree of clustering


This very low rate may be due to reporting bias.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement