Comparison of Four Hospital Nurse Staffing Patterns for Wage Costs and Staffing Adequacy Using Computer Simulation.

Abstract

Decisions about staffing hospital nursing departments constitute critical operations decisions because they involve a large portion of the hospital's budget. Selection of a particular staffing pattern has often been a matter of personal preference or tradition rather than of scientific analysis. Managers in the business world have long used computer based decision support tools to assist them in making critical staffing and scheduling decisions. Nursing Service managers are now seeking to adapt this technology to the health care management arena in order to enhance the quality of their management decisions. In this study, computer simulation was used to examine differences in wage costs and in overstaffing and understaffing produced by each of four different hospital nurse staffing patterns. The four staffing patterns were: Fixed Distributed, Float Pool, Controlled Variable, and Skill Center staffing. This study employed a matched sample experimental design. Computer simulation was used to model each of three hospitals: a small community hospital, a large community, and a large Veterans Administration hospital. A sample size of 110 years of staffing pattern performance was obtained. The independent variable was staffing pattern. Dependent variables included mean yearly wage cost, number of excess nurses per year and the dollar cost of those nurses, number of understaffed shifts per year, and number of nurses short on an understaffed shift. It was found that Fixed Distributed staffing produced the highest wage costs, followed by Float Pool Staffing. There were no significant differences between Controlled Variable and Skill Staffing for mean yearly wage costs. Although statistically significant, the actual cost differences among the four staffing patterns were rather small, amounting to less than 4% of the total direct wage cost budget in the Large Community hospital, and less than 3% in the small community and VA hospitals. A similar pattern of statistically significant but relatively small sized differences were found for the dependent variables, incidence of overstaffing and understaffing. It was concluded that Fixed Distributed staffing performed poorly as compared with the three variable staffing patterns. However, there were no significant performance differences among the three variable staffing patterns for these dependent variables.