No abstract

A total of 2,611 calls to 14 helplines were monitored to observe helper behaviors and caller characteristics and changes during the calls. The relationship between intervention characteristics and call outcomes are reported for 1,431 crisis calls. Empathy and respect, as well as factor-analytically derived scales of supportive approach and good contact and collaborative problem solving were significantly related to positive outcomes, but not active listening. We recommend recruitment of helpers with these characteristics, development of standardized training in those methods that are empirically shown to be effective, and the need for research relating short-term outcomes to long-term effects.

To determine the relative effectiveness of telephone intervention styles with suicidal callers, researchers listened unobtrusively to 617 calls by suicidal persons at two suicide prevention centers and categorized all 66,953 responses by the 110 volunteer helpers according to a reliable 20-category checklist. Outcome measures showed observer evaluations of decreased depressive mood from the beginning to the end in 14% of calls, decreased suicidal urgency ratings from the beginning to the end in 27% of calls, and reaching a contract in 68% of calls, of which 54% of contracts were upheld according to follow-up data. Within the context of relatively directive interventions, a greater proportion of "Rogerian" nondirective responses was related to significantly more decreases in depression. Reduction in urgency and reaching a contract were related to greater use of Rogerian response categories only with nonchronic callers.

Models of telephone crisis intervention in suicide prevention and best practices were developed from a literature review and surveys of crisis centers. We monitored 2,611 calls to 14 centers using reliable behavioral ratings to compare actual interventions with the models. Active listening and collaborative problem-solving models describe help provided. Centers vary greatly in the nature of interventions and their quality according to predetermined criteria. Helpers do not systematically assess suicide risk. Some lives may have been saved but occasionally unacceptable responses occur. Recommendations include the need for quality assurance, development of standardized practices and research relating intervention processes to outcomes.