Two Essays on the Impact of Healthcare Policies

Abstract

Since the late 1990s, opioids have been increasingly prescribed for pain treatment in the U.S. as a result of aggressive marketing by pharmaceutical companies, resulting in more than 450,000 opioid overdose deaths. In the same time period, several U.S. states have legalized medical marijuana, a drug that can also be used for pain relief. As a result, medical marijuana can be used as a substitute for opioids, potentially leading to a reduction in opioid prescriptions. On the other hand, marijuana use can lead to increased substance abuse, leading to a potential increase in opioid prescriptions. The lack of scientific and medical knowledge along with the uncertain regulatory environment vis-a-vis medical marijuana use also makes it possible that its legalization has no impact on opioid prescriptions. Using eleven years of claims data from a large health insurance company in the U.S., we study the effect of medical marijuana legalization on opioid prescriptions, leveraging the temporal variation in state-wise legalization. We find that, on average, opioid prescriptions decreased after medical marijuana legalization. We also find that the role of providers in reducing opioid prescriptions after legalization is more prominent than their corresponding role in increasing opioid prescriptions.

Despite the potential and promises of telehealth in improving access to care, particularly for those in areas suffering from lack of providers, the actual utilization of telehealth had been as low as below 1 percent of total healthcare before the Covid pandemic. Among the commonly identified challenges for this dim telehealth use, originating site restriction - where patients are required to be at the pre-approved locations for getting telehealth care - is often argued as one of major barriers for low telehealth utilization on patient sides. In the late 2010s, states started to lift this longstanding patient setting restriction within its Medicaid program. After the removal of originating site restrictions, Medicaid patients are allowed to be home or wherever they feel appropriate when getting telehealth cares. We exploit this unique state-level Medicaid policy change to understand how removing this restriction affect telehealth utilization. In this paper, we aim to answer two research questions. First, what is the effect of removing originating site restrictions on telehealth utilization in Medicaid? Second, how does removing originating site restrictions affect the geographical reach of telehealth providers? Using the newly released Medicaid analytic files (T-MSIS), we build a county-month panel of telehealth utilization and estimate policy effects employing a difference-in-differences model. We find that after removing the originating site restrictions, telehealth utilization increased across the number of claims, billing amount and Medicaid paid amount. Additionally, we find a larger increase in expenditure than in the number of telehealth claims, implying that cost of telehealth per claim became more expensive than before. Interestingly, our data reveals that most telehealth consultations continued to occur within the same county. Even with removal of originating site restrictions, the increase in telehealth utilization mostly come from patients and providers in the same county. This suggests that geographical proximity to providers remains a significant factor in telehealth visits, at least for behavioral health patients in Medicaid.