W. Gardner et al., CLINICAL VERSUS ACTUARIAL PREDICTIONS OF VIOLENCE IN PATIENTS WITH MENTAL ILLNESSES, Journal of consulting and clinical psychology, 64(3), 1996, pp. 602-609
This study compared the accuracy of an actuarial procedure for the pre
diction of community violence by patients with mental illnesses with t
he accuracy of clinicians' ratings of concern about patients' violence
. Data came from a study in which patients were followed in the commun
ity for 6 months after having been seen in a psychiatric emergency roo
m. Accuracy of actuarial prediction was estimated retrospectively, wit
h a statistical correction for capitalization on chance. Actuarial pre
diction had lower rates of false-positive and false-negative errors th
an clinical prediction. The seriousness of the violence correctly iden
tified by the actuarial predictor (the true positives) was similar to
the seriousness identified by clinicians. Actuarial predictions based
only on patients' histories of violence were more accurate than clinic
al predictions, as were actuarial predictions that did not use informa
tion about histories.