Sources of disagreement among clinicians' assessments of dangerousness in a psychiatric emergency room

Abstract

The literature on the prediction of dangerousness has focused primarily on the issue of predictive validity and the ramifications of the simple finding that “clinicians are vastly overrated” (Monahan, 1984) as predictors of future violence. Numerous authors have argued, with considerable justification, that the use of such assessments for the purpose of involuntary commitment is a questionable practice, given the poor validity of such predictions (Dix, 1983; Steadman, 1983). The field has been occupied with the challenge of providing empirical information useful for evaluating the ultimate justice of the practice of civil commitment. While numerous objections to the methodology of these studies have been raised (Hinton, 1983; Megaree, 1976; Monahan, 1981) there is, nonetheless, still general agreement that even well-designed studies will probably not find that such assessments yield very valid predictions in ordinary clinical settings (Monahan, 1984). Predicting a low base rate, highly context-dependent behavior, like violence, is a task that will, in all likelihood, not be simplistically depicted and/or easily routinized (Mulvey & Lidz, 1985a). This consensus, however, has not stopped both judicial and legislative authorities from concluding that assessments of dangerousness are still the responsibility of mental health professionals in a wide variety of clinical and forensic situations. This reality of public policy has led to the call for researchers to take a more “ameliorationist” position (Mulvey & Lidz, 1985b). Rather than trying to marshall evidence for or against the use of the prediction of dangerousness as a basis for involuntary treatment, the researchers are urged to focus on methods of improving how to perform this task.