Access to medical care and how it differs for various patients is a critical health policy issue. While existing work has examined how clinic structure can influence overall productivity, less research has explored how provider mix within a clinic relates to access for different patients (e.g., Medicaid enrollees and the commercially insured). Using providers of varying skill types may facilitate cheaper service delivery and thereby decrease the likelihood that lower-paying patients (Medicaid) are turned away. However, this may crucially depend on prevailing scope of practice regulations, which can impede the substitution of labor between provider types.

We use data from a simulated patient study where trained field staff - randomly assigned to commercial, Medicaid, or self-pay - called primary care offices requesting the first available new patient appointment for either routine care or an urgent health concern. The study was conducted in ten states (Arkansas, Georgia, Iowa, Illinois, Massachusetts, Montana, New Jersey, Oregon, Pennsylvania, and Texas).

 For private and Medicaid insurance, we describe the association between the availability of new patient appointments and the size of a physician office by both the number of physicians and mid-level providers. To test the robustness of our findings, we then estimate within-office models (i.e., clinics receiving both a private and Medicaid call) since all clinics within this subset indicated Medicaid managed care or PCCM participation in a pre-experimental non-deceptive call. For a complementary analysis, we use the self-pay experimental arm to also examine the relationships between provider mix and the quoted visit price for those granted an appointment. Importantly, we partition our sample to explore heterogeneity in these results by the prevailing state-based scope of practice laws relevant to a given clinic. Our regulatory groupings are as follows: liberal (IA, MT, OR), moderate (AR, MA, NJ) and restrictive (GA, IL, TX, PA) policy states.

 In all regulatory environments, mid-level providers are commonly used; however, their implications for access and care costs markedly vary across the three policy groups. Clinics with more non-physician staff are associated with better access for Medicaid patients (as much as a 25 percentage point increase in the probability of receiving an appointment) and lower prices for a visit (as much as $45 cheaper on average), but these relationships are only found in states granting full practice autonomy to these mid-level providers. A more diverse provider mix has no perceptible association with provider willingness to accept a new Medicaid patient or the costs of providing services within more restrictive policy states.