This study examined specific risk factors discriminating psychiatric inpatients who were never suicidal from those who reported one or more suicide attempts, and compared characteristics discriminating first-time attempters from first-time and multiple repeaters (recent attempters with one or more previous attempts).

716 inpatients admitted to the Division of Psychiatry of a University Hospital were evaluated for their suicidal history, psychiatric diagnosis, current and past stressors, and other clinical and sociodemographic risk factors.

Of the 358 attempters, 86 were first-time attempters, 219 first-time repeaters, and 53 multiple repeaters. There were significant differences between the nonattempters and the attempters in sociodemographic variables, DSM-IV-TR diagnosis (attempters were diagnosed more frequently with eating disorders, and less frequently with DSM-IV Axis II disorders than nonattempters), and clinical history. First-time attempters had fewer lifespan stressors (M=1.86; SD=0.90) than repeaters (first-time repeaters: M=2.26 [SD=0.89]; multiple repeaters: M=2.28 [SD=1.13]). Furthermore, repeaters were more likely to have personality disorders (ORs=2.71-3.83), low social support (OR=4.96-6.88), and previous hospitalizations (ORs=37.23-556.69) than first-time attempters.

In repeaters, stressors may have a specific effect and a cumulative effect. They can predispose to attempts at an earlier age and, through a process of “kindling” start a “suicide career.” Alternately, an accumulation of stressors can produce a “last straw” effect. In repeaters hospitalization can be seen as another factor contributing to the kindling phenomenon, and brief intensive care in hospital settings must be weighed accurately for efficacy and for its possible negative effects.