Modeling intensive care unit census.

A mathematical model revealing the relationships between bed capacity, average patient admission rate, average patient length of stay, utilization rate, and overfill rate in intensive care units is developed and explained. Mathematical model predictions are compared to predictions of two kinds of discrete event intensive care unit simulations and to data from a variety of real intensive care units. This methodology applies to any intensive care unit or hospital ward. There is no significant (p < 0.05) difference between measured utilization and overfill rates assessed in actual intensive care units, the rates obtained by discrete event simulations, and the rates predicted by the intensive care unit model. The intensive care unit census model can enhance rational determination of intensive care unit bed and staff requirements.