The yearly report of the percentage of gross domestic product (GDP) taken up by health care costs has been the linchpin for recurring rituals of angst and argument in many policymaking, political, and media circles in recent decades. That ratio is often cited as proof that U.S. health care costs are ‘skyrocketing’ or ‘out of control.’ Yet, as pointed out in a January 31 virtual seminar at the University of Pennsylvania, over the last 15 years, that same metric actually documents a widely unrecognized slowing in the rate of health care cost increases.

As she opened the “Has the U.S. Cured Health Care Cost Growth?” panel discussion at Penn’s Leonard Davis Institute of Health Economics (LDI), Executive Director and moderator Rachel M. Werner, MD, PhD, pointed out that since 2009, apart from two COVID-19 pandemic years, the health care GDP percentage has consistently hovered just below 18%.

“This is the longest stretch of no cost growth relative to GDP since at least 1965, when Medicare and Medicaid were created,” said Werner.

She also noted that that fact contradicts the physical reality of what many Americans experience with their own health care costs—about half of U.S. adults, even those with health insurance, say it is difficult to afford health care. Many say they engage in cost saving measures like delaying needed care or not filling the prescriptions they get from a doctor.

Paradox of the GDP Metric

Werner indicated the paradox between slowing health care GDP cost growth and the daily reality of U.S. patients underscores why we need to ask if the GDP metric is measuring the right aspects of health care spending.

The panel event brought together Werner and three other top academic health care experts: Melinda J.B. Buntin, PhD, Professor of Health Policy and Economics at the Johns Hopkins Bloomberg School of Public Health and Johns Hopkins Carey Business School; Amitabh Chandra, PhD, Professor of both Public Policy at the Harvard Kennedy School of Government and Business Administration at the Harvard Business School; and Ezekiel J. Emanuel, MD, PhD, LDI Senior Fellow and Professor of both Medical Ethics & Health Policy at the Perelman School of Medicine and Health Care Management at the Wharton School. Buntin was unable to attend but offered a statement and questions that were incorporated into the proceedings.

The group’s discussions ranged across an array of topics related to both what the drivers of the cost slowdown were and what the most effective cost reduction policies might be going forward. Woven throughout their comments was a concern about the value of the GDP percentage as a health care metric because:

• It doesn’t capture health outcomes or quality of care.
• It fails to account for the distribution of health care resources and access to care.
• It doesn’t reflect the efficiency of health care spending.

Real Meaning of the GDP Metric

“The economist in me realizes that lots of people like this measure because it’s very simple and allows for easy comparisons to other countries,” said Chandra. “But suppose the number goes up to 18.5% or down to 17%. Which is the best answer? I don’t even know the directionality. Are we getting more waste or more value at either of those changes? What I need is the ratio of health care spending to the health that spending created.”

Emanuel suggested that it is possible to pick a health care spending-to-GDP ratio number that would be appropriate for the U.S. and that number would be 12% of GDP, which is generally closer to the average amount that other Organization for Economic Cooperation and Development (OECD) countries spend.

“When we compare our health outcomes to every other country in the OECD, we’re certainly not doing well or getting our money’s worth,” said Emanuel. “The only place we’re really hitting it out of the park consistently, year in and year out, is in cancer. Otherwise, stroke numbers have gone up. If you look at hypertension, our control is getting worse. In diabetes, our control is getting worse. We haven’t made a dent in infant mortality and we’re way bad in maternal mortality and you could continue going down the list. When you look at disability-adjusted life years (DALYs), we’re going in the wrong direction. So, we may be spending 18% on health care but it clearly is not adding to better health in America.”

(DALYs are a measure of overall disease burden, expressed as the total number of years lost due to illness, disability, or premature death. It is used by global health organizations like the World Health Organization (WHO) to assess the impact of diseases and injuries on populations.)

Driving the Cost Slowdown

In the area of the drivers of the slowdown, the panelists pointed to forces including:

• Significant across-the-board reduction of per-beneficiary costs resulting from administrative price regulation in Medicare and Medicaid programs.
• Increases in the overall GDP that made the health costs segment apparently lower.
• The launch of the Affordable Care Act (ACA) exchanges.
• Value-based payment models.
• The under-recognized power of public controversy and debate around health care costs to motivate some health care stakeholders to independently change their practices in ways that lower costs.
• The high-deductible insurance plans that came into vogue with employers in the early 2000s, shifting substantial costs onto beneficiaries.

Emanuel, who was one of the architects of the Affordable Care Act that instituted them, pointed to the substantial impact of value-based payment plans in cost control.

The Value-Based Care Debate

“They haven’t been perfect, but the rollout and emphasis on these value-based programs have had a big effect, combined with the shift to Medicare Advantage and more capitation in that context,” said Emanuel. “Over time, we’ve seen less of an increase in emergency room utilization and an actual drop in hospital utilization and admissions. We’re down below 100 hospital admissions per 1,000 Americans, which is a historic low. That’s exactly what you would expect as value-based payment changes physician behavior, decreasing the utilization of the highest-cost parts of the system.”

Werner pointed out that when many people look at the success of the alternative payment models like value-based care, “they walk away with the impression that those programs haven’t worked as well as they were expected to.” She asked Emanuel, “How do you reconcile the power of alternative payment models’ ability to flatten the cost curve, but also their lack of success in meeting the goals they set out for themselves?”

“I’m not sure they set the goals themselves,” responded Emanuel. “I think the goals that are being used by the actuary of the Department of Health and Human Services (HHS) do cover things like the 3% savings in bundled payment for lower extremity surgeries and 3% savings over time in the Medicare Shared Savings Program (MSSP). But what that leaves out is a lot of spillover we have documented.”

Spillover refers to the power of the alternative programs’ operation and visibility to influence physicians and health care managers to either adopt new practices suggested by those programs or be more sensitive to how other costs savings in their sphere of operation could be improved—or as Emanuel characterized it, “effect a mental change on the provider side.”

Negative Side of High-Deductible Plans

In another area, Chandra, who serves on the Congressional Budget Office’s (CBO) panel of health advisers, pointed to the cost control effectiveness as well as the negative impact of high-deductible health plans.

“We know that when you put people into a high deductible health plan, spending falls 12%. Like overnight. It’s huge,” said Chandra. “If we put everybody in a high deductible plan, the share of GDP would absolutely fall. But all the empirical research proves that when you move people to those plans or expose them to co-payments, they cut back on exactly what we don’t want them to cut back on. And new research says many die because of that. It’s not that they might die. They do die. It’s an example of spending less and causally killing more people. So, expanding high deductibles would take us into a world where we’re reducing the share of GDP spent on health care, but we’re actually worse off as a society.”

According to Emanuel, one of the weakest areas of cost control has been in employer insurance. “Employers have largely been AWOL on the issue,” he said. “They bitch and moan about high prices, and that ‘health care is out of control’ but they have not really done very much in terms of adding their weight to cost control. Their silence has meant that commercial plan costs have gone up higher than Medicare and Medicaid. There are other factors, but I think that is a very important one that can’t be overlooked. Many employers have basically done nothing except introduce high-deductible plans, which now appear to have run their course for the last few years. Enrollment in them is flat at just under 33%.”

As the session approached its end, moderator Werner asked Emanuel and Chandra what their favorite idea would be for cutting costs in the system going forward.

High Cost of Administrative Complexity

Emanuel pointed to administrative costs throughout the heavily fragmented system. “We need to simplify,” he said. “One of the things that has plagued our system is complexity right from the start and we need to simplify and standardize not just our medical delivery, but the administrative operations of the system. We’re now spending about $1 trillion—or about 22% of total health care spend—on administrative costs. That could be streamlined but the incentive for the Department of Health and Human Services to do that is almost zero because of the trillion dollars, only $80 billion is the federal government spending. The rest is that of payers, providers, hospitals, and physicians and they can’t get their act together. There’s a lot of money there. About $250 billion in financial transactions claims—billings and payments that could be simplified.”

Chandra said his big answer to cutting costs in the future is a system that provides a basic level of care to every American. “It’s unlikely we’ll get a handle on costs or value until the country determines the minimum floor of health care services we need to give to every American. Most would be enrolled in this basic plan with basic coverage at birth by default. I think there could be a real bipartisan opportunity to start the conversation on this because it is a way to reduce spending and improve health. How would we structure this? I think Medicaid provides a great chassis for structuring this,” he said.

Emanuel disagreed and pointed to his own “scars” from his work on the ACA. “We put in that you’re going to have 10 essential benefits, and we kicked that to the National Academies of Sciences, Engineering, and Medicine to tell us what those the 10 benefits should be. We didn’t get an agreement. Then we did a survey and there was no agreement, and I don’t think there is ever going to be an agreement on this. I just don’t think it’s feasible,” he said.

End of Cost Increase Slowdowns

One thing that Emanuel and Chandra did agree on strongly is that we are now seeing the end of this long, historic period of stability in the health-percentage-of-GDP metric.

“It is clear that we are at the end,” said Chandra. “There is a new report from Altarum that looks back at health care spending in the end of 2024. In October, it grew 7.6% year over year, and in November, it grew 7%. So, we are now over 18% of GDP. At the same time, the CBO is predicting that health care spending is going to pick up.”

---

---