Last week, researchers at RAND released a study that made broad claims about the prices that private insurers pay hospitals. The RAND authors relied on severely limited data and questionable assumptions to draw far-reaching conclusions about the way hospitals are paid for patient care. And, despite these limitations, the authors have looked at a pond, and drawn conclusions about the ocean.
The authors suggest that Medicare payment rates are an appropriate substitute for the prices negotiated between hospitals and private insurance. However, it is widely acknowledged that Medicare reimburses below the cost of care. In 2017, hospitals received payment of only 87 cents for every dollar spent by hospitals caring for Medicare patients. The total Medicare payment shortfall was $53.9 billion in 2017, up from $37.8 billion in 2013. Adopting artificially low Medicare payment rates would strip vital health care resources from already strapped communities, seriously impeding access to care. Hospitals would not have the resources needed to keep their doors open, to invest in new technologies that improve the quality of care and care delivery, innovate and transform the way in which health care is delivered to improve health outcomes, and maintain the services communities need and expect.
There are also a number of other questions about the authors’ methodology. These include the authors’ blunt application of Medicare Severity Diagnosis-related Groups (MS-DRGs) and Ambulatory Payment Classifications (APCs); contradictions and lack of clarity in their description of their pricing simulation; and over-reliance on CMS star ratings as the only measure of hospital quality, when these ratings have been extensively demonstrated to not reflect the full range of care delivered in hospitals and use a flawed methodology that can lead to inaccurate and misleading comparisons of hospitals.
It is also important to note that the study’s conclusions should not be generalized to the national level – and in many cases even the state level. The authors themselves point out that the study’s key limitation is its small sample size – less than 5 percent of all covered lives in about half of all states, and just 2 percent of the 181 million Americans with employer-sponsored insurance nationally. Moreover, only a small proportion of hospitals are included in the data. For example, eight hospitals comprise the sample for inpatient prices in Wisconsin, out of 134 community hospitals in the state. The study’s entire sample represents just over $13 billion in hospital spending between 2015 and 2017, which was about 1 percent of total private hospital spending during that time.  It is also unclear which private plans were included, whether the data reflected different plan designs, and to the extent that the data came from self-selected employers, what employers’ risk pools looked like. There are several other limitations that the authors themselves have acknowledged: for a large number of hospitals, less than 11 claims were available for inpatient services and only their outpatient services were reported; the claims data could not distinguish between in-network and out-of-network services which could overstate the true private negotiated prices; possible inaccuracies in cross-walking the private claims data to Medicare provider numbers; including only facility and not professional claims although some private health plans might bundle both facility and professional claims thereby overstating private prices – the list goes on. To draw such far-reaching conclusions based on such limited and un-generalizable data is dubious at best.
Further, recent data refute the authors’ claim that hospital prices are driving spending growth. According to CMS, price growth for hospital care services was just 1.7 percent in 2017. This trend continued in December 2018, with year-over-year price growth still at 1.7 percent, according to the Altarum Center for Value in Health Care (“Altarum”). This was a lower rate of growth than all other categories of services, including physician and clinical services and prescription drugs. Altarum further notes that low hospital spending growth has been responsible for slower overall health spending growth. In fact, the RAND authors’ own analysis found an even lower rate of price growth for the period they examined – just 1.0 percent!
Hospitals and health systems are transforming how care is delivered to improve quality and make care more affordable for patients. Hospitals have invested hundreds of billions over the past decade (more than $62 billion in 2017 alone) in EHRs and other IT systems that support secure information sharing. These investments enable hospitals to participate in new payment models that incentivize quality and value rather than volume. And, these investments are a critical first step in bringing information to patients and their providers, to help improve care coordination and educate patients about the cost of care. These are steps hospitals are taking now toward better, more affordable care.
The AHA is committed to improving patients’ access to information on the price of their care. It’s important that individuals understand how much they will need to pay for their care, specifically their out-of-pocket costs. Yet hospitals, health systems and other providers do not always have access to detailed data on health plan benefit and beneficiary cost-sharing amounts; rather, insurers hold this information. We are encouraged by the growing ability for providers and insurers to work together to develop tools that they can use to help respond to patient pricing inquiries./p>
Aaron Wesolowski is AHA vice president of Policy Research, Analytics and Strategy.
 Underpayment by Medicare and Medicaid Fact Sheet - January 2019. American Hospital Association. https://www.aha.org/factsheet/2019-01-02-underpayment-medicare-and-medicaid-fact-sheet-january-2019
 AHA analysis of National Health Expenditure Accounts data. Office of the Actuary, Centers for Medicaid & Medicare Services (CMS). 2018. National Health Expenditure Accounts. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData/NationalHealthAccountsHistorical.html
 Since the state-based all payer claims database contains Medicare and Medicaid data for only two states, CO and NH, the $13 billion would be an even smaller percentage if Medicare and Medicaid national spending was included in the denominator. This shows, yet again that the sample size used in the study is extremely small.
 National Health Expenditure Data, 2017. CMS, Office of the Actuary, National Health Statistics Group. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData/index.html
 Health Sector Spending. Altarum. https://altarum.org/solution/health-sector-spending