An integrated approach to estimating suicide risk

It is ironic that if we had a perfect predictive instrument we would not be able to recognize it because it could never be validated by its critical outcome criterion. Though some exceptions could occur, we would be obliged to take all available measures to prevent a suicidal outcome in cases where suicide was predicted. After the crisis we could have no way of knowing with certainty whether the person would have suicided or not. Even if we accepted the reality that people are not either 0% or 100% likely to suicide, and developed a perfect scale to estimate degree of risk, we would still be unable to validate it in individual cases. If it indicated "moderate" risk of 2.5-5.0%, for example, and no intervention were offered, we would have to observe one suicide in every 20-40 persons assessed at this level of risk to demonstrate its validity. The key to assessment is obtaining information, primarily regarding present or anticipated pain and the threshold of pain tolerance in the individual involved. Since different persons communicate in a variety of ways--verbal, nonverbal, symbolic, metaphoric, etc., eclecticism in approach is essential. For some clinicians communication will be facilitated most by one style; for others, a different method would be most effective. Thus, the "best" approach is the one that works best given the unique characteristics of the persons involved and under the conditions existing at the time. My own bias is that every assessment, whatever the approach, must include some form of direct inquiry regarding suicidal intent, and that the final decision in this regard must be a subjective and intuitive judgment. Contrary to possible assumptions in the legal world, accurate assessment does not necessarily mean safety. It can serve as a guide to the degree of risk that may be involved in a treatment program, but even low risk management measures may have an adverse outcome without implications of negligence or carelessness. There has been no mention here of biological markers of suicide, which are of much current interest but still in an investigational stage. Similarly, rational suicide has not been mentioned, though our aging population and the status of AIDS are making this issue progressively more important. The principles involved in assessment of risk are the same as with other forms of suicide, however. Finally, we can only presume that more precise assessment will operate to reduce suicidal deaths.(ABSTRACT TRUNCATED AT 400 WORDS)