All Americans are worried about inflation. The prices for housing, clothes and food at the supermarket are eyepopping and threatening President Biden’s re-election. But there is one totally unexpected exception to inflation recently: health care.
For decades the growth in health care spending in the U.S. exceeded overall inflation — until relatively recently. With the exception of 2020 and the Covid cost spike, health care costs have remained at or below 18% of GDP since the enactment of the Affordable Care Act in 2010. This has been the longest stretch of no-cost-growth since at least 1965 and the enactment of Medicare and Medicaid. These overall numbers have been reinforced by reports that Medicare’s spending per person has been flat for more than a decade, and recent data showing that premiums for private employer-sponsored insurance have been increasing at 3.7% in the past decade, which is much slower than the 8.4% between 1999 and 2011.
The experts seem stumped. Recently the Congressional Budget Office acknowledged that their 2010 predictions significantly overestimated spending between 2010 and 2020, with Medicare and Medicaid spending in 2019 an unbelievable 17% lower than estimated. The CBO admitted they did not anticipate and could not account for the lower spending. Similarly, the New York Times claimed “no one knows why” per capita Medicare spending has not increased in 14 years.
All of this cost control also seems hard to reconcile with the CBO’s finding that the Center for Medicare and Medicaid Innovation (CMMI), an agency created by the ACA to try various programs to reduce Medicare costs, actually raised government expenditures by over $5 billion. Or that we went through a once-in-a-century pandemic that affected millions, all the while 10,000 baby boomers were enrolling in Medicare each day as they aged and got sicker.
But this is not really an inexplicable paradox. A few months ago, I received a call that I think helps explain a key part of why health care cost growth has been so much lower than projected.
The chair of orthopedics at the University of Pennsylvania approached me about using what I know about behavioral economics to induce surgeons to reduce the costs of hip and knee replacements and other surgical procedures. His goal was to cut 10% out of surgical costs. I was surprised that a chair of an orthopedics department — a man who pioneered hand transplants — was interested in cost savings.
When the ACA passed more than a decade ago, few orthopedists — indeed, few physicians — were interested in overall health care costs, much less assuming the initiative and responsibility for reducing them. During the ACA debate, the American Medical Association’s lobbying focused on getting higher physician payments by eliminating the Sustainable Growth Rate that each year threatened massive physician pay cuts.
Today, controlling costs is on every orthopedist’s mind, and they are not alone. The mindset of American physicians and other clinicians has changed, from ignoring costs to trying to cut them. Instead of figuring out more expensive medical tests and treatments, doctors are now asking whether a test or treatment will improve a patient’s health and how a service can be performed more efficiently by shifting where and how it is administered.
The catalyst for this change in attitude is payment-driven by the evolving shift from fee-for-service to value-based payment (VBP), from paying physicians to do more tests and treatments to paying them to improve health and prevent expensive disease exacerbations. In fact, CMMI was responsible for shifting Medicare from essentially no value-based payments — in which providers are responsible for quality and total cost of care — in 2012 to more than 30% of payments by 2016. This was the largest shift in Medicare payment in history and was a catalyst for change in the broader health care system.
However, the value-based payment change has been much slower than many of us hoped and predicted. Today, 60% of payments to physicians remain fee-for-service.
Nevertheless, value-based payment has been more influential than this number suggests. One reason is contagion or the spill-over effect. Several years ago, Medicare initiated a demonstration project of bundled payments for hip and knee replacements. Bundled payments are a single payment that covers all the costs associated with a procedure — the hospital costs, including operating, recovery, and hospital rooms; the artificial joint, and blood transfusions; the surgeon and anesthesiologist’ fees; post-procedure physical therapy and visits to the doctor; and complications for 90 days after the surgery. By keeping costs below the bundled payment amount, the hospital and doctors could keep the savings.
Suddenly incentivized financially to think about reducing costs, surgeons were more willing to negotiate lower costs for the artificial knee and hip implants, move surgery out of the hospital to lower cost ambulatory surgical centers, shift physical therapy out of expensive rehabilitation facilities to the patient’s home, and focus more attention on reducing costly complications like surgical site infections. And there were no negative impacts. Orthopedists did not start operating on more patients to make more money or take only healthy low-risk patients to lower costs and make more money.
Overall, in year one, there were savings, but they were not quite statistically significant: about 2%. More importantly, it turned out orthopedic surgeons did not change their practices to reduce costs only for Medicare patients in the bundled payment program. They also changed for other patients for whom they were not getting the financial incentives to save money. Saving money in the bundled payment program was also highly correlated with hospitals saving money for other joint replacement patients outside the bundled payment program. Indeed, it appears the program induced a change in the standard operating procedures for medical management of hospitals and surgeons regardless of how they were being paid. Their new medical “habits or standard practices” spilled over. (Importantly, the CBO and other evaluations of this bundled payment program used blinkers in their calculations of savings. They acknowledged the spill-over effects but failed to include them in calculating savings because they were only interested in the savings to the patients in the specific Medicare program, not overall health savings. This inevitably produces an underestimate of the financial and behavioral impact of the VBP program.)
Similarly, a decade ago a government program called MSSP ACOs — Medicare Shared Savings Program for Accountable Care Organizations — incentivized physicians to manage their patients, particularly their sickest patients, to preempt exacerbations that sent them to the emergency room and into hospital beds. Physician groups in this program began getting data on how they practiced medicine, who their sickest patients were, and began changing how they cared for the patients.
Better management of patients frequently leads to fewer emergency room visits and hospital admissions. For the past six years the program has saved money. Indeed, in 2022, 482 ACOs caring for nearly 11 million Medicare patients saved $1.8 billion. This constituted just under 3% of total health care costs for all these patients. Amazingly, the best performers were able to save 10% of costs or more. And, MSSP was also financially beneficial for physicians. Nearly two-thirds of ACOs earned bonus payments for their cost reductions. And it turned out that ACOs comprised of greater than 75% primary care physicians saved “more than twice as much” as the average — probably because they managed patients with more chronic diseases better.
As my orthopedist colleague demonstrates, it has taken time to shift physicians’ attitudes regarding costs and how they care for patients. But it is happening and accelerating. Now we need to rapidly expand these Medicare MSSP ACOS and bundled payments for surgery, so they are not 40% but become 60% or 70% of overall payment in the next three or four years. We also need to get more practices being paid through VBP. This may require mandating all practices participate. Just as importantly, VBP cannot only be a Medicare initiative. VBP needs to be expanded across all insurers. Finally, the VBP programs of United, Aetna, Humana and other insurers need to work in concert and with Medicare — use the same program design and incentive arrangements — so the physician and hospital incentives align and do not drive health care providers crazy with insignificant but time-consuming and wasteful variations.
Maybe if these changes are made we can see health care costs—except for drugs, whose costs I expect to rise — remain flat for the next decade. That would be nothing short of a miracle that, once again, the experts would have a hard time explaining.
Ezekiel J. Emanuel is an oncologist, vice provost for global initiatives, co-director of the Health Care Transformation Institute at the University of Pennsylvania, and author of several books, most recently “Which Country Has the World’s Best Health Care?” (Public Affairs Books, 2020).