Measuring adverse outcomes in inpatient psychiatry: the reliability of nurse recall.

Nurses have been used in studies to report on quality-of-care issues and may be the most feasible sources of information about adverse outcomes. To date, however, the use of nurses to recall adverse outcomes of inpatient psychiatric care has not been examined systematically. The purpose of this study was to determine the reliability of psychiatric nurses' recall of physical restraint episode(s) (PRE) and assaultive/aggressive episode(s) (A/AE) and to describe the factors that influence recall. The study design was prospective and retrospective, and used a convenience sample of 36 nurses from a psychiatric hospital located in a large eastern city. Despite differences in nurse characteristics, perceptions of shift conditions, and organizational characteristics, nurse recall of PRE and A/AE was reliable overall. The majority of nurses actually recalled the exact number of events they had prospectively reported. Furthermore, the period for recall (2 vs. 4 weeks) did not significantly affect the reliability of nurse recall. These results suggest that psychiatric nurses are reliable informants of adverse outcomes. Lacking readily accessible and systematically reported data on PRE and A/AE, nurses may be the most feasible sources of information for these outcomes of inpatient psychiatric treatment. Future research should replicate these findings in other psychiatric treatment settings.