SUICIDE - SOME ANSWERS

Lifetime Risk for Suicide in Depression

There is a clear association between suicide and depression. New data on depression using more current diagnostic approaches, and new approaches to statistical analysis allow for the more exact specification of lifetime risk. One review using a statistical modeling approach to correct for the problem of limited follow-up periods estimated that if samples of depressed individuals were followed until all were deceased, lifetime suicide risk in depression would be about 6% (Inskip, Harris, & Barraclough, 1998). However, this review covered studies which were based largely on inpatient samples, which may overestimate lifetime suicide risk compared to the full population of individuals with depressive disorders. Best estimates are that suicide rates among those who had previously been treated for a depressive disorder as inpatients are about twice as high (4.1%) as those who had been treated as outpatients (2%). Those treated as inpatients following suicide ideation or suicide attempts, are about three times as likely to die by suicide (6%) as those who were only treated as outpatients (Bostwick & Pancratz, in press). Together these results suggest there may be a hierarchy of risk based on a patient's treatment history. Given that patients with a hospitalization history, particularly when suicidal, have a much elevated suicide risk compared to both outpatients and non-patients, the clinical decision to hospitalize in and of itself may be a useful indicator of increased lifetime suicide risk. A third study reported dramatic gender differences in lifetime risk of suicide in depression, where risk for men is estimated to be about 7% versus 1% for women (Blair-West, et al., 1999).

These estimates are far smaller than those found in earlier reports. An often-quoted risk is that about 15% of patients with major depression will eventually die by suicide (Guze & Robins, 1970). This rate was further reinforced by a large-scale review of 30 studies (Goodwin and Jamison, 1990). These data derive largely from studies of severely depressed inpatients, whereas the majority of depressed people are treated as outpatients or not treated at all. People who have had depression of a severe enough intensity to warrant inpatient treatment may be at higher risk of suicide than those with less severe symptomatology. These estimates reflect only the most severely ill segment of the population suffering from depression. Although these estimates are useful in highlighting the seriousness and lethal consequences of untreated depression, these same estimates do not carry much credibility if they are paired with more current prevalence rates of depression. In fact, applying the 15% estimate of suicide risk to the estimated rates of depression for various age groups in the U.S., the overall suicide rate would be at least four times the current rate (see Blair-West, Mellsop, & Eyeson-Annan, 1997).

Most of the studies used for the estimation of suicide risk followed up patients for only a few years. Because suicide risk has been found to be greater soon after hospitalization and early in the course of a diagnosed illness, subjects who die in studies with short follow-up periods are more likely to die from suicide than any other cause. In addition, suicide is over-represented as a cause of death among younger persons, when psychiatric disorder is often first diagnosed (Inskip et al., 1998). This results in higher relative rates of suicide, since there are few other causes of death occurring for this age group. Thus, following up patients for only a few years after treatment results in high estimates for the lifetime risk of suicide for all persons with an affective disorder.

There are different approaches to the definition and measurement of "suicide risk" (Bostwick & Pancrantz, in press). Both Guze and Robins and Goodwin and Jamison employed proportionate mortality to estimate suicide rates. Proportionate mortality is calculated by dividing the number subjects who have died of suicide by the number of subjects who have died from all causes during the follow-up period. If all of the subjects in every study were followed throughout their life span, proportionate mortality would correctly estimate the probability of suicide. However, if subjects are not followed for their entire lives, proportionate mortality method will overestimate the risk of suicide. A second approach to calculating suicide risk is case fatality prevalence. This approach compares the number of persons who have died by suicide to the number of persons originally in the cohort being followed. A third approach to calculating suicide risk is a standardized mortality ratio, calculated by dividing the observed mortality of a cohort by the expected mortality of an age- and gender-matched cohort representative of the general population.

Increased precision in reporting characteristics of depressed persons, such as their history of suicidal behavior and health service utilization, will help move the field ahead in understanding more specific risk factors and targeted opportunities for treatment and prevention of suicide among persons with affective disorders.

References

Blair-West, G. W., Mellsop, G. W., Eyeson-Annan, M. L. (1997). Down-rating lifetime suicide risk in major depression. Acta Psychiatrica Scandinavica, 95, 259-263.

Blair-West, G. W., Cantor, C. H., Mellsop, G. W., & Eyeson-Annon, M. L. (1999). Lifetimes suicide risk in major depression: Sex and age determinants. Journal of Affective Disorders, 55, 171-178.

Bostwick, J. M., & Pancrantz, S. (in press). Affective disorders and suicide risk: A re-examination. American Journal of Psychiatry.

Boyd, J.H. & Weissman, M.M. (1981). Epidemiology of affective disorders: A re-examination and future directions. Archives of General Psychiatry, 38, 1039-1046.

Goodwin, F. K., & Jamison, K. R. (1990). Manic-Depressive Illness. Oxford University Press: New York.

Guze, S. B., & Robins, E. (1970). Suicide and affective disorders. British Journal of Psychiatry, 117, 437-438.

Inskip, H. M., Harris, E. C., Barracough, B. (1998). Lifetime risk of suicide for affective disorder, alcoholism, and schizophrenia. British Journal of Psychiatry, 172, 35-37.