Consequences of the 340B Drug Pricing Program

Articles Referenced: February 8, 2018: Consequences of the 340B Drug Pricing Program

The authors reply:

As noted in our limitations and reiterated by Drs. Ross, Li, Bhatt, and Orlowski our findings pertain directly to hospitals near the eligibility threshold, may not generalize to hospitals with higher DSH percentages, and do not exclude beneficial use of drug discounts by individual hospitals. Nevertheless, our findings—based on analysis of hospitals accounting for >50% of Medicare spending on care in public/non-profit general acute hospitals—constitute strong evidence of hospital responses that are inconsistent with 340B goals and thus support Program reform. For example, our findings support limiting eligibility to hospitals serving higher shares of underserved patients and, moreover, policies that benefit the underserved directly, without distorting incentives to provide drugs.

We reject the notion that low-income Medicare beneficiaries are not a population intended to benefit from the 340B Program. Approximately 40% of fee-for-service Medicare beneficiaries either have Medicaid coverage, which reimburses providers at lower rates for care not covered by Medicare, or are exposed to substantial cost-sharing because they lack supplemental coverage or receive partial assistance.^1,2 Also, major investments to serve other underserved populations should have spillover effects on low-income Medicare beneficiaries. While we agree that 340B-eligible hospitals may treat more low-income patients or provide more uncompensated care than ineligible hospitals, such differences do not support conclusions about Program effects. Eligible hospitals by definition disproportionately serve disadvantaged populations, so differences in hospital contributions to low-income populations are expected in the absence of the Program.

A strength of our study’s design is that it circumvents confounding from other causes of hospital-physician consolidation to which Drs. Ross and Li refer. Ours is not the only study to suggest the 340B Program has shifted cancer care to hospital-owned facilities or encouraged integration with oncology practices serving affluent communities.^3,4

Drs. Bhatt and Orlowski’s comments are taken directly from a critique of our study by the American Hospital Association.² We addressed these comments elsewhere and encourage readers to read our response² and appendix.⁵ The specific methodological points raised about our implementation of the regression discontinuity are minor and discussed in our updated appendix; none are of great consequence, let alone “undermine” our findings. In short, our conclusions remained unaltered.