Essays on Health Care Accessibility

Abstract

This dissertation comprises three empirical papers on health policy issues related to the accessibility of health care. The first two papers focus on regulatory issues that influence the delivery of health care by nurse practitioners. The third paper studies the effect of improved public transportation on health care accessibility. Nurse practitioners (NPs) are an increasingly important part of the health care workforce, particularly in primary care. NPs’ authority to practice without physician oversight is regulated by state-level scope of practice restrictions. To the extent that these regulatory restrictions prevent NPs from practicing to their full abilities and capacity, these restrictions could create inefficiencies in the primary care delivery system. In Chapter 1, I study what happens when states relax their scope of practice restrictions. Using a novel dataset of claims and electronic health records data paired with a difference-in-differences research design, I quantify the effects of relaxed scope of practice laws on 1) workload and patient allocation between NPs and physicians; 2) NPs’ autonomy in their day-to-day provision of care; and 3) the quality of care provided by NPs. I find that relaxing scope of practice restrictions for NPs does not change the volume or allocation of patients to NPs nor the quality of care provided by NPs. However, I do find evidence that NPs practice more autonomously following scope of practice policy changes. Given the lower reimbursement that NPs typically receive, these findings suggest that allowing NPs to practice without physician oversight is likely to reduce health care spending, without harming the quality of care for patients. In Chapter 2, I examine another important policy issue related to NPs: a billing custom known as “incident-to” billing. Incident-to billing occurs when services provided by an NP are billed under a supervising physician’s national provider identifier. When incident-to billing occurs, the services provided by the NP are attributed to the supervising physician in widely-used sources of data. Physician practices have a financial incentive to bill NP-rendered visits incident-to, but face a trade-off, since it requires the physician to directly supervise the NP. Data reflecting the prevalence of incident-to billing is rare. I present the first claims-based, national estimates of NP incident-to billing frequency. I rely on a novel source of claims data that has the unique ability to observe both the rendering provider and the billing provider for all patient visits. I find that nearly half of visits rendered by an NP were billed incident-to a physician in a sample of practices. The prevalence of incident-to billing varied considerably by practice, and was skewed at the extremes. Incident-to billing was more common in reduced scope of practice states, in urban areas, and at single-site practices, but was not associated with more complex care being provided. These findings bolster the policy proposal to end incident-to billing or modify information required on claims, so as to observe care provided by the full range of primary care providers. In the final chapter of this dissertation, I pivot to examine health care accessibility from a patient perspective. Specifically, I consider the role of public transportation in patients’ access to health care. Patients often cite transportation barriers as a reason they miss their medical appointments, but it is not known whether improved public transit leads to fewer missed appointments. I examine the frequency of “no-shows” at a large health system before and after the opening of a large light rail line that significantly improved the area’s public transit availability. I use a difference-in-differences research design in conjunction with a multi-year dataset generated from electronic health records, reflecting over two million scheduled appointments. I find a small but statistically significant decline in no-shows after the light rail opened, with larger declines observed among Medicaid patients. These results suggest investments in public transit infrastructure can improve access to care and result in downstream health benefits for the community, and especially low-income patients.