Staffing of inpatient psychiatric units and assault by patients

BACKGROUND: Assault by a patient is a serious problem for patients, staff, and hospitals. There is disagreement in the literature about the relationship between the number of staff on psychiatric units and the assault rate. OBJECTIVE: The purpose of this study was to determine the relationship between the number of and characteristics of nursing staff and the occurrence of assault on inpatientpsychiatric units. STUDY DESIGN: This was a comparative study of assaults over a 6-month time period. The primary unit of analysis was the unit shift. RESULTS: When comparing shifts with an assault to shis without an assault, the statistical results for each of the four dependent variables (number of registered nurses, number of'staff, patient/staff'ratio, and staff gender) were not significant. There was a significant fp < .001) interaction among three factors: assault status, unit, and time. Although the number of registered nurses (RNs) was not significantly related to assault occurrence, examination of individual units by time of day (shift) revealed surprising and important results. Some units had more assaults when there were fewer RNs, and some units had more assaults when there were more RNs. CONCLUSION: The difference between shifts with an assault and shifts without an assault depended on the particular unit and the particular time of day. The findings cannot be generalized across units and must be examined in detail at the unit level A most important finding in need of further investigation is the inconsistent pattern of number of RNs related to assault occurrence both among units and within each unit.