Job Market Paper

• The role of cost-sharing amounts in the take-up of biosimilars

Abstract: Seniors in the United States receive health insurance benefits through Medicare. The Medicare drug benefit, known as Medicare Part D, uses commercial insurers to administer the benefit. The federal government bears the risk of high-cost drugs, including biologics. This mechanism reduces the adverse selection problem but exacerbates the moral hazard problem. So insurers are insufficiently sensitive to drug prices. I study how the moral hazard created by Medicare affects coverage choices made by participating insurers. In particular, I examine how subsidies affect coverage choices when cheaper products become available. Using data on formularies for the universe of US health insurance plans, I employ a triple difference approach that compares before and after, Medicare and non-Medicare, and drugs with and without low-cost entrants. I find that when low-cost products (“biosimilars”) become available, non-Medicare plans reduce coverage of biologics by 20% more than Medicare plans. If Medicare Part D plans emulated non-Medicare commercial plans, the federal government could save $300 million annually. Patients would save $15-20 million. These savings represent more than 4% of spending on biological drugs with a biosimilar available. I find that a policy change implemented in 2018 had a weak positive impact on the take-up of biosimilars in Medicare Part D, while a 2022 law might have a more substantial effect.

Working Papers

• Strategic Patient Dropping: Evidence from the Dialysis Quality Incentive Program

(with Paul J. Eliason, Benjamin Heebsh, Riley J. League, Ryan C. McDevitt, James W. Roberts)

Abstract: Federal and private health insurance plans are increasingly adopting pay for performance contractual arrangements. The effects of these programs are not well understood. We study the federally mandated program of this type in the U.S., the Quality Incentive Program (QIP) that incentivizes dialysis facilities to provide high-quality care. Facilities receive scores based on a set of performance measures and face a reduction of their Medicare reimbursements of up to 2% if their scores are too low. We find that patients who will lower facilities' scores are 24% more likely to switch facilities in any month, and 20% more likely to switch facilities after being discharged from a hospital. This is consistent which anecdotal evidence of dialysis facilities blackballing certain types of patients or not accepting them once discharged from a hospital. We also find evidence that penalty inducing patients are more likely to switch to facilities which are worse in terms of several quality of care indicators such as mortality and hospitalization rates, suggesting that patient switches are involuntary.

• Provision of Transplant Education for Patients Starting Dialysis: Disparities Persist

(with Vincenzo Villani, Gabriel Butler, Paul J. Eliason, James W. Roberts, Nicole DePasquale, Christine Park, Lisa M McElroy, Ryan C. McDevitt), submitted