Background: A variety of models are used by hospitals, provincial governments, and departments of emergency medicine to "predict" the number of physician hours of coverage necessary to staff emergency departments. These models have arisen to meet specific requirements-some for the purpose of determining hourly rates of compensation, others to determine the amount of funding that will be provided to "purchase" physician coverage, and others to determine the number of hours of coverage necessary to maintain patient waits within "acceptable" limits. All such models have their strengths and weaknesses and have been criticized as not reflecting the "real" needs of any given department.
Objective: In the article that follows, a review of existing models is presented, annotating their strengths and weaknesses to derive the characteristics of an "ideal" workload model.
Conclusion: None of the models currently used to measure emergency department workload can be relied on to accurately predict the number of staffed hours necessary. Models that may achieve this objective are suggested.