Abstract
This study examines the accuracy of long-term clinical predictions of dangerousness among psychiatric inpatients and explores factors influencing the levels of such accuracy. Hospital and state criminal history records of all psychiatric patients (N = 31) for whom, during a four-year period, treatment staff pursued extended civil commitments based on dangerousness under the Postcertification for the Imminently Dangerous statute (California Welfare and Institutions Code § 5300) were reviewed. A matched control group consisted of 31 patients who had been placed on 14-Day Certifications for Dangerousness to Others, but who were not subsequently placed on 180-Day Postcertifications. Sixty-one percent of patients in the postcertification group engaged in documented physically assaultive behavior during the extended one- to five-year follow-up period, compared with 26 percent of patients in the matched control group, suggesting that inclusion in the extended commitment group was indicative of greater long-term potential for assault. Differences in assaultiveness did not emerge during the first year of followup, but became clear and significant over subsequent years. Accuracy of prediction differed as a function of patient ethnic group.
Footnotes
-
Drs. Zeiss, Fenn, and Yesavage are affiliated with the Veterans Affairs Palo Alto Health Care System. Dr. Tanke is affiliated with Applied Decision Research, Mountain View, CA. This work was supported by the Department of Veterans Affairs. Portions of this paper were presented as work in progress at the 1989 and 1990 annual meetings of the American Academy of Psychiatry and the Law.
- Copyright © 1996, The American Academy of Psychiatry and the Law





