Policymakers have renewed their focus on how the 340B drug discount program functions among “safety net” hospitals, particularly Disproportionate Share  Hospitals (DSH), which qualify for the 340B program because they provide a sufficient amount of inpatient care to Medicaid and low-income Medicare beneficiaries. In June, a leaked executive order from the Trump administration suggested that the program would start to tie the volume of discounted drugs to indigent patient volume. In July, the Centers for Medicare and Medicaid Services (CMS) proposed cuts to physician reimbursements for outpatient drugs covered under Medicare’s medical benefit (Part B) for 340B DSH hospitals. Tomorrow, October 11th, the House of Representatives’ Oversight and Investigations Subcommittee will hold a hearing titled, “Examining How Covered Entities Utilize the 340B Drug Pricing Program.”

Why choose this focus for the October hearing? 340B is a drug discount program that provides qualified hospitals with a 20 percent to 50 percent discount on the average manufacturer price of outpatient prescription drugs, and the opportunity to generate revenue from sales of 340B drugs to insured patients in order to fund safety-net care. While the program provides a vital source of revenue for some participants, there is increasing concern that DSH hospitals and their affiliated clinics and contract pharmacies are abusing the program by pursuing profits rather than ensuring patient access to needed care. DSH hospitals and their affiliates can behave in this way because—in contrast to the public health clinics that also qualify for 340B drug discounts—DSH hospitals and their affiliates are not required to report on how much revenue they generate from the sale of discounted drugs, nor demonstrate how they use revenues to enhance safety-net care.

Consequently, anecdotes supporting the safety-net dedication—or profiteering—of 340B participants fuel the current policy debate. Previous assessments of program functioning have largely focused on limited samples of cherry-picked 340B participants. Tomorrow’s hearing before the Oversight and Investigations Subcommittee features testimony from only three hospital systems participating in 340B (Johns Hopkins Hospital, Northside Health Systems, and Mission Health Systems, Inc).

In this post, we provide statistics on the safety-net engagement and financial stability of all 340B DSH hospitals in 2015. Our analysis suggests that anecdotes are not sufficient to understand the program; DSH hospitals that participate in 340B are in actuality quite diverse. Our results suggest the arguments of opponents and proponents of proposed reform might hold some merit. A priority for policymakers contemplating program reform should be more systematic and timely data on participating DSH hospitals’ 340B revenue and safety-net service provision.

Aggregate Statistics On 340B Participating DSH Hospitals

We analyzed administrative data from 2015 Medicare Cost Reports linked to the HRSA’s 340B Provider List. Together, these data are the most comprehensive, publicly available data on 340B participants. We compared the median and interquartile ranges—the range from the 25th to the 75th percentile—of safety-net engagement and financial stability of non-profit and public DSH 340B participants to the national median for public and non-profit hospitals in 2015 (Exhibit 1).

We found that in 2015 the median uncompensated care burden among the 1,042 340B participants in our data was 3.4 percent, which was only slightly higher than the median for all non-profit and public hospitals (3.0 percent). However, a quarter of 340B providers spent less than 1.9 percent of their budget on uncompensated care. Those in the top quartile spent twice as much as the national median (6.5 percent). This suggests a substantial amount of variation among participants.

The distribution was similar for the Medicaid share of total discharges: more than 25 percent of 340B hospitals cared for fewer Medicaid beneficiaries than the national median. For a quarter of these hospitals, 22.6 percent or more of their inpatients were covered by Medicaid.

Operating margin provides a measure of financial stability for hospitals. The underlying reason for allowing DSH hospitals to generate revenue off the program is to improve their financial stability. Looking at hospital operating margins, the median operating margin was lower among 340B hospitals (-1.6 percent) than among public and non-profit hospitals in the US (-0.8 percent). However, the interquartile range of 340B hospital operating margin ranged from -9.3 percent to 4.8 percent, suggesting that some 340B DSH participants operated at a substantial loss, while at least a quarter of participants operated with a comfortable margin.

Finally, we considered the DSH patient percentage—the criterion for entry into the 340B program. To qualify, hospitals must have a DSH patient percentage in excess of 11.75 percent. We found that the median 340B hospital’s DSH patient percentage exceeded the national median for public and non-profit hospitals, but the 25th percentile of the distribution was just above the threshold of 11.75 percent. This suggests that at least a quarter of hospitals just barely qualified for the program.

Upon further inspection, we found that many of the hospitals with the highest operating margins were also those that provided the least uncompensated care. The opposite was also true—the hospitals that provided the most uncompensated care also had the lowest operating margins. Furthermore, we found that there was little correlation between county-level uninsured rates, likely a good proxy of demand for health care by the uninsured, and the adjusted DSH patient percentage.

Implications For Reform Efforts

In 2015, the DSH hospitals participating in the 340B drug discount program exhibited varying financial stability and safety net care provision. Therefore, policymakers should exercise caution when drawing conclusions about the program or the likely effects of reform efforts based on the testimony or documented experience of a handful of participating DSH hospitals. Our results echo previous assessments, suggesting that the current breadth and diversity of DSH hospitals and their affiliates participating in 340B could be distorting the program away from its primary intent to serve patients with a limited ability to pay and contributing to wasteful spending.

In the absence of systematic assessments of program functioning, hospitals, and even some lawmakers, have argued forcefully against the current reform proposals, citing financial destabilization and cuts to services that could hurt patients who rely on the safety net. In contrast, the authors of the proposed reforms have suggested they offer benefits: targeting discounted drugs to low-income patients might improve pharmaceutical access for the uninsured, and reimbursement cuts could benefit Medicare patients by reducing their out-of-pocket spending on drugs covered under the medical benefit. Under current law, Medicare beneficiaries are required to pay 20 percent of Medicare’s reimbursement with no catastrophic limit, unless they are also covered by supplemental insurance. Our results suggest the arguments of both opponents and proponents of proposed reform might hold some merit.

Where do we go from here? We believe an important first step towards reorienting the program for the aid of patients who rely on the safety net is systematic, timely data collection. Currently, all hospitals are required to report on selected financial and service metrics in their annual Medicare cost reports. DSH hospitals are also required to report on their community benefit activities to the Internal Revenue Service to retain their tax-exempt status. Congress could build on these sunshine efforts, requiring hospitals to report on 340B drug volume, revenues generated, and safety-net services funded by these revenues to CMS and/or the IRS.

A second step is a discussion about what defines a safety-net medical provider in the 21st century. The DSH patient percentage is an antiquated measure harkening back to a time when hospitals largely provided inpatient care. It is by definition an inaccurate measure of safety-net care provision across inpatient and outpatient settings. The inadequacy of this metric is evidenced by the fact that 40 percent of all non-profit and public hospitals in our sample met the criteria for participation, yet it seems unlikely that such a high fraction of hospitals meet the Institute of Medicine’s definition of safety-net providers. A renewed focus on the care provided to uninsured and underinsured patients in both settings better fits the current organization of medical providers in the US. Congress should consider alternative metrics.

Exhibit 1: National Characteristics of 340B Providers, 2015

Source: 2015 hospital cost reports matched to the 340B provider list. The sample includes 2,639 non-profit and public general acute care hospitals in the 50 US states and the District of Columbia from Medicare hospital cost reports. We identified 1,042 hospitals that were currently participating in the 340B program according to the Health Resources and Services Administration’s 340B provider list. Uncompensated care is defined as the fraction of total hospital expenditures that are spent on charity care or hospital bad debt. The Medicaid share of total discharges is the fraction of total discharges that are covered by Medicaid. The hospital operating margin is defined as (operating revenue-less contractual allowances-operating expenditures)/(operating revenue-less contractual allowances). The operating margin was winsorized at the 5th and 95th percentiles to minimize the influence of outliers. The DSH patient percentage is defined as the sum of the Medicaid fraction of total utilization and the social security income eligible Medicare fraction of total Medicare utilization. This fraction is adjusted according to hospital size, and urban status.

Note: The authors are grateful for funding from the Commonwealth Fund. The opinions expressed are those of the authors, not the funder.