Changing the Schedule of Medical Benefits and the Effect on Primary Care Physician Billing: Quasi-Experimental Evidence from Alberta

We exploit a quasi-experiment in the province of Alberta, Canada, to identify how changes in the schedule of medical bene fits affected the provision of primary care services to patients with multiple co-morbidities. Speci cally, Alberta introduced a new fee code to compensate physicians for completing a comprehensive annual care plan (CACP) for qualifying patients. During the period of study, primary care physicians could practice in two settings: (i) solo practice; or (ii) primary care networks (i.e., team based care). This paper asks how the policy change affected physician-billing patterns and whether delivery structure affected physician- billing. Data come from Alberta's administrative physician claims data, covering the full population of Alberta and all services provided by primary care physicians, for one year before and two years after the policy change. We employ a difference-in-differences methodology and implement a set of robustness checks to control for confounding from other contemporaneous changes that may have occurred in Alberta as well as unobserved physician heterogeneity. Our results suggest the new fee code became the sixth most billed code in its first year (totalling $17.9 million), but was billed by only a small proportion of physicians (roughly 2% of physicians accounted for 20% of total billings). The fee code was disproportionately billed by physicians in team-based care (PCNs), and increased the billing of other complementary fee codes by 5%-10% (or roughly $80 million). The results suggest the unintended consequences of a well-intentioned policy can be costly.