Most studies show about a 10% risk of suicide in a patient who has previously attempted suicide and that the risk is highest in the year following the attempt. Therefore, most assessments of patients after a suicide attempt focus on identification of risk factors for current and future suicide. Such assessments focus on known risk factors like male sex, older age in women, presence of a psychiatric disorder, and level of suicidal intent (considering things like “was there a well-thought out plan?”).
Now a new study (Runeson 2010), using data from linked databases in Sweden, has captured long-term data on the risk of successful suicide after a prior attempted suicide. In this study, 12% of patients committed suicide after a prior attempted suicide during a followup period of 21-31 years.
The most important finding, however, was that the risk varied significantly by the method used during the first attempted suicide. Whereas poisoning (overdoses) and cutting are far and away the most common methods of attempted suicide, the highest relative risk for successful suicide was for those whose index attempt was by hanging, strangulation, or suffocation/drowning and 87% of these suicides occurred within one year of the initial attempt. The risks were also higher for those whose index attempt involved firearms or explosives, gassing, or jumping from heights. Most successful suicides also used the same method that had been tried in the index attempted suicide.
The presence of a psychotic or affective disorder were additional independent risk factors for suicide. So a patient with such a severe psychiatric disorder and a suicide attempt by a method such as hanging, drowning or firearms would be at particularly high risk.
The implications of this study are that method of attempted suicide should be included in the overall risk assessment and that those patients deemed to be at the highest risk merit close followup, especially during the first year. The editorial accompanying the study (Hawton 2010) points out that the study was unable to determine whether the method used in the index attempt may have simply reflected the degree of lethality or intent on the part of the patient. It also notes that while these statistics accurately describe a population, it is still difficult to apply them in individual cases. Would you followup a patient who had attempted suicide by overdose any less rigorously than one who attempted suicide by hanging? Probably not. But you can expect the key points of this study to probably be incorporated into some existing suicide risk assessment tools.
Runeson B, Tidemalm D, Dahlin M, et al. Method of attempted suicide as predictor of subsequent successful suicide: national long term cohort study.
BMJ 2010; 341: c3222
Hawton K. Completed suicide after attempted suicide. BMJ 2010; 341:c3064