Background: Difficulty maintaining physician staffing in emergency departments (EDs) prompted the government of Ontario to offer alternate funding arrangements (AFAs) to replace fee-for-service remuneration for physicians working in EDs. Objective: To analyze the effect of AFAs on physician staffing and practice patterns. Methods: We obtained Ontario Health Insurance Program fee-for-service and shadow-billing records for all physician services provided in EDs one year before and one year after

more »

... mplementation of an ED AFA. Only sites with reliable billing data were retained. Physicians were assigned to small/rural, community or teaching hospital groups based on their billing claims. For each hospital type, and all hospitals combined, we compared the pre- and post-AFA periods in terms of the number of physicians working regularly in the ED and their workload. As a possible unintended consequence of AFAs, we also compared physicians' involvement in primary care. Results: Overall, 76.2% of eligible hospitals adopted an ED AFA, of which 49 (42.6%) were included in our study (16 small/rural, 27 community and 6 teaching hospitals). In the post-AFA period, the number of physicians working in EDs increased by 7, from 674 to 681, representing a 1.0% increase overall in the workforce (p = 0.84). The change varied by hospital type, from a 5.8% increase in teaching hospitals to a 2.2% decrease in community hospitals, though none was significant. In the post-AFA period, the number of physicians working a moderate number of days per month increased from 190 to 214, representing a 3.2% absolute increase (p = 0.39), and the number working few (<5) or many (>10) days per month decreased. Post-AFA, the number of physicians working in EDs who also provided primary care services decreased by 1.7%, from 544 to 535 (p = 0.10). Conclusion: Emergency department AFAs have been widely adopted in Ontario, but have not been associated with substantial changes in the overall physician workforce in EDs. However, trends toward increased physician numbers were seen in small/rural and teaching hospitals. There was little evidence of any adverse effects on the provision of primary care services by physicians.