Patient acuity indicators as predictors of pharmacy workload

Abstract

The feasibility of using two workload indicators from a nursing patient-classification system as a means of predicting pharmacy workload was studied. Frequency data for 13 pharmacy distribution and clinical activities were recorded daily for 28 consecutive days and compared with daily data for acuity of patient illness and number of standard hours of nursing care required on nine nursing units. The strength of the associations between (1) pharmacy workload and patient acuity and (2) pharmacy workload and standard hours of nursing care was determined by linear regression analysis. Both same-day and one-day-lagged analyses were performed; the one-day-lagged analyses looked at pharmacy workload on a given day in relation to nursing workload on the previous day. A total of 252 observations were available for analysis. Pharmacy workload and same-day standard hours of nursing care were correlated most strongly, although all of the analyses yielded large coefficients of correlation. Analysis of data from individual nursing stations yielded smaller coefficients of correlation, especially for the one-day-lagged analyses. At least 73% of the variance in pharmacy workload could be attributed to variance in nursing workload. The index of patient acuity of illness and the number of standard hours of nursing care are good predictors of pharmacy workload of the same and the following days; the potential exists to use these nursing workload indicators in determining pharmacy staffing requirements.