A new study published in the February 2019 issue of Health Affairs examined growth in hospital and physician services from 2007-14. The study focused on growth in prices for total inpatient care, total outpatient care, and four high-volume services: cesarean section, vaginal delivery, hospital-based outpatient colonoscopy, and knee replacement. The authors used data from Health Care Cost Institute (“HCCI”), which include claims for individuals with employer-sponsored insurance from Aetna, Humana, and UnitedHealthcare – three of the five largest insurers in the United States. All told, HCCI data capture health spending on approximately 28% of Americans with employer-sponsored insurance.
According to the HCCI data, hospital prices accounted for the majority of the cost of care, and hospital prices grew at a faster pace than physician prices for inpatient and outpatient hospital-based care from 2007-14. More specifically, hospital prices for inpatient and outpatient hospital-based care during that time grew by 42% and 25%, respectively. By comparison, physician prices grew at a pace of only 18% and 6%, respectively.
The authors performed several analyses to demonstrate the robustness of their results. For example, they restricted their analyses to episodes occurring at hospitals that did not employ physicians, but rather had outside community physicians performing services. They did so to illustrate that there were no systematic differences in results between hospitals that employed physicians and those that did not. Additionally, the authors analyzed the HCCI data separately for episodes that occurred in counties where the HCCI insurers had above- or below-median shares of covered lives. This screen permitted the authors to determine whether their results were robust in areas where those insurers were likely to have more or less bargaining power. The authors found that in areas where the insurers had above-median shares of covered lives, 87% of the growth in inpatient prices and 91% of the growth in outpatient prices were driven by growth in hospital prices. However, in areas where the HCI insurers had below-median shares of covered lives, growth in hospital prices was an even more significant driver of inpatient and outpatient prices, at 91% and 99%, respectively.
Based on their empirical findings, the authors identified four potential policy approaches to stemming the growth of hospital prices. First, they called for tighter antitrust enforcement, citing studies that draw a link between hospital mergers and increased hospital prices. They called on state and federal regulators to review proposed mergers more vigorously, consider tougher remedies, and block mergers that could raise prices. Second, they suggested that regulating hospital payments – e.g., setting inpatient prices at 120% of Medicare – would decrease spending. Third, they noted that the use of reference pricing has led consumers to access lower-priced care and led to modest reductions in provider prices. Finally, they noted that physicians in vertically-integrated settings often refer patients to more expensive locations. The authors suggested that these effects could be lessened by incentivizing physicians to refer to more cost-effective settings and through greater antitrust scrutiny of vertical integration.
Shortly after its publication, the American Hospital Association (“AHA”) issued a response to the study. Not surprisingly, the AHA took issue with the authors’ methodology and conclusions. Specifically, it noted that the HCCI data on which the study is based does not include data from Blue Cross and Blue Shield plans, which dominate many health insurance markets throughout the country. According to the AHA, the HCCI data sample represents just 13.5% of Americans with any sort of health coverage, including Medicare and Medicaid. The AHA also noted that hospitals are forced to manage a number of significant costs – including, for example, regulatory compliance, which costs an average-sized community hospital approximately $1,200 per admission to support compliance with the regulations of just four federal agencies. These burdensome costs, however, do not apply to the physicians included in the study, making for an inapposite comparison. Finally, the AHA notes that a hospital’s cost of care includes expensive inputs (e.g., drugs, medical devices, and health information technology) which hospitals have a limited ability to control and that do not comparably impact physicians.
* * *
In addition to the points raised by the AHA, two other considerations are salient. First, the authors of the study contend that more rigorous antitrust enforcement is necessary in order to curb increasing hospital prices. However, it is unclear whether the hospitals observed in the study were creatures of merger and/or whether they have market power such that they could command supracompetitive reimbursement rates from health insurers. While the authors accounted for potential indicia of relative bargaining power on the payor side (i.e., whether the insurers’ shares were above or below the median within a given geography), it is not self-evident that they controlled for market power on the hospital side. Whether the observed hospitals operate in markets that would be deemed “highly concentrated” within the meaning of the 2010 FTC/DOJ Horizontal Merger Guidelines could shed some light on the issue. Without that information, it is difficult to gauge whether tougher antitrust enforcement – which could serve as a check on market concentration and a hospital’s post-merger bargaining clout – would have any bearing on the study’s findings.
Second, the authors suggest that vertical integration – whereby physicians tend to refer patients to care settings that may be more expensive than other available options – is a driver of rising hospital prices. However, a counterargument can be made that an integrated healthcare system with a well-organized provider network can not only enhance quality and reduce utilization, it can also reduce the overall cost of care. Such were the findings of another recent study by researchers with the Healthcare Financial Management Association, Leavitt Partners, and McManis Consulting (“HFMA study”). The authors of the HFMA study determined that markets with less integration – and with apparently more provider competition – actually observed higher prices. On the other hand, lower-cost markets tended to be marked by consolidation, leaving between two and four health systems with comprehensive geographic coverage competing in the market. Moreover, physicians in lower-cost markets tended to be either employed by health systems or be in close alignment with a system. Accordingly, the authors of the HFMA study posited that the question of “what type” of competition may be more important than “how much” competition, concluding that integration can lead to less costly care.
* * *
The Health Affairs study raises important considerations about rising healthcare costs. The AHA’s response and the HFMA study, however, underscore that these issues are complex and that there are valid counterarguments to the policy debate that hospital pricing studies inevitably generate.