SUICIDE RISK ASSESSMENT
AND THE SUICIDAL
Lawrence Amsel and J. John Mann
Despite the advances in suicide research the prediction of suicide remains a difficult problem. The process approach to suicidal behaviour that we describe draws on insights from the “stress–diathesis model” that we have proposed and from advances in the field of risk-factor research. In this chapter we will review the prediction problem from a traditional statistical perspective, and examine how these new approaches may contribute to suicide-risk assessment. Where possible, we will illustrate these points using neurobiological risk factors associated with increased suicidal behaviours.
Although there are a number of well-established risk factors for suicidal behaviours, the prediction of suicide in individual cases remains difficult. A recent review article for the general clinician in the New England Journal of Medicine (Hirschfeld et al, 1997) on identifying suicidal patients illustrates this problem. Whereas the article mentions a number of risk factors, it does not give a single quantitative estimate for these risk factors, a practice that would be eschewed in a similar