VIOLENCE, VICTIMS, AND A THERAPIST'S DUTY TO PROTECT (CALIFORNIA).

Abstract

Issues regarding confidentiality and potential dangerousness, when an individual is seen in outpatient psychotherapy, are addressed in this study which was derived from a controversial California Supreme Court decision. The Tarasoff holding imposed a duty to protect upon psychotherapists where the therapist knew, or should have known, that there was a treat of violence to a third party. This case appeared to pit society's interest in the prevention of violence against its interest in effective psychotherapy (through a postulated cost to the psychotherapy relationship of a duty to warn involving a breaching of confidentiality). A two-pronged research procedure consisting of a survey of 278 psychotherapists and a structured interview with 61 psychiatric outpatients was utilized to collect data regarding the decision-making process that a therapist employs when confronted with a potentially aggressive patient. It was found that, in general, dangerous outpatients both entered psychotherapy voluntarily and were potentially violent before entering treatment. Such individuals did not frequently meet the legal criteria for commitment and were reported to be more dangerous toward intimates than strangers. The formation of a therapeutic alliance did not appear to be a necessary condition for the patient to reveal dangerousness. Findings from therapist responses revealed that they infrequently responded to potential violence by committing the patient to a hospital or by notifying the police. Their most common responses were more frequent psycotherapy and voluntary hospitalization. Warnings were reported as an intermediate level response. Outpatient respondents indicated that their they considered it appropriate for a therapist to use information from therapy sessions to warn others in the case of serious threats. Confidentiality did not appear to be a premier concern, with few outpatients reporting that they would abandon treatment if confidentiality were broken during a homicidal emergency. Finally, warnings may not significantly impair a therapy alliance nor create an appreciable level of psychotherapy attrition. This finding is suggestive that, from a social policy perspective, a warning may be an efficacious procedure and may not generate the commonly predicted thwarting of effective treatment and ultimate reduction in public safety.