Questions About Reasons for Living

To the Editor: There is currently considerable uncertainty about which forms of psychosocial and physical treatment are most effective for patients who deliberately harm themselves (1). A study by Kevin M. Mallone, M.D., et al. (2) ascribed “reasons for living” as a protective factor against suicidal acts in subjects with major depression and proposed that eliciting and increasing awareness of reasons for living in depressed patients merit further study as a suicide-prevention strategy. We welcome this original line of enquiry and endorse the authors’ recommendations for replication of their findings but contend that their conclusions may be premature and largely be the result of erroneous interpretation of the direction of causality and the inappropriate use of statistical tests.

The conclusions drawn by the authors appear to rest on three main findings. First, suicide attempters reported significantly greater subjective depression, hopelessness, and suicidal ideation than nonattempters. The significance of this difference was tested by use of the t test, which assumes that the data are normally distributed. However, in the authors’ Table 1, it is readily apparent that for the measures of hopelessness and suicidal ideation, the standard deviation multiplied by two is greater than the mean. This indicates that the mean is unlikely to be the center of the distribution, and for data that do not follow a normal distribution, appropriate nonparametric tests ought to be used (3).

Second, the depressed patients who had not attempted suicide scored significantly higher on items from the Reasons for Living Inventory than the attempters. Moreover, the total score on the Reasons for Living Inventory was significantly inversely correlated with scores for hopelessness, suicidal ideation, and subjective depression that were evaluated separately or as a composite measure of “clinical suicidality.” Third, objective measures of severity of depression and quantification of life events did not differentiate suicide attempters from nonattempters.

These findings seem to have led the authors to conclude that the nonattempters had a more optimistic mind set because they perceived (or had) more reasons to live or because inner restraints precluded suicide as an option. Although this conclusion appears intuitively appealing, the inverse correlation between reasons for living and clinical measures of suicidality could be equally due to the greater severity of subjectively perceived depression and hopelessness in attempters, resulting in the enumeration of fewer reasons to live and greater suicidal intent. Which of these explanations best fits the data awaits further clarification of the direction of causality, which may be possible with multivariate rather than univariate statistical analysis or a prospective study.