It's mostly overutilization, stupid

The first study from a massive new database reclaims old ground in the long-running debate over the roots of higher health care spending in the U.S.

For decades, economists have struggled to explain why the U.S. spends far more than other industrialized nations on health care yet delivers worse outcomes. In recent years, their consensus concluded higher prices lay behind our higher spending, based on the undeniable fact that insurers, especially in the private sector, paid the highest prices in the world for nearly all products and services, whether drugs, hospital stays or physician visits.

Yet this focus on prices is a relatively recent phenomenon. Until the mid-2010s, most researchers and policymakers blamed excessive utilization.

Starting in the 1970s, a team of researchers at Dartmouth University used Medicare data to show wide variation in the amount of care delivered to elderly patients with similar medical conditions. Their findings, compiled in the Dartmouth Atlas of Health Care, suggested eliminating unnecessary use of tests, procedures and drugs in high-spending states and counties would bring U.S. spending closer to international norms.

A 2009 New Yorker article headlined “The Cost Conundrum” popularized their findings, and went on to become one of the most noted and quoted articles in health care journalism history. Dr. Atul Gawande, then the nation’s most influential health care journalist, used Dartmouth’s data to show that Medicare enrollees in McAllen, Texas, the nation’s top-spending Medicare region, received twice as much health care on average as beneficiaries in El Paso, a demographically similar region in the same state. Referencing a 2003 study by Elliott Fisher, one of the architects of the Dartmouth Atlas of Health, Gawande reported that overall:

“Patients in higher-spending regions received 60% more care than elsewhere. They got more frequent tests and procedures, more visits with specialists, and more frequent admission to hospitals. Yet they did no better than other patients, whether this was measured in terms of survival, their ability to function, or satisfaction with the care they received…

To make matters worse, Fisher found that patients in high-cost areas were actually less likely to receive low-cost preventive services, such as flu and pneumonia vaccines, faced longer waits at doctor and emergency-room visits, and were less likely to have a primary-care physician. They got more of the stuff that cost more, but not more of what they needed.”

Pushback against the overutilization thesis began even before Gawande wrote his seminal article. Critics pointed out the Dartmouth analysis relied solely on Medicare data, where, unlike prices paid by insurance companies, prices vary little across the country.

“It’s the prices, stupid,” the headline read on a 2003 article by a team of researchers led by Gerard Anderson of Johns Hopkins University. They based their analysis on an international comparison of prices and utilization patterns provided by the Organization of Economic Cooperation and Development, which includes every advanced industrial nation. The late Uwe Reinhardt of Princeton University, a leading health care economist and co-author of the article, popularized the phrase in numerous speeches to professional and patient advocacy groups working to make health care more affordable.

Their analysis was confirmed by a 2015 study, whose co-authors included Yale’s Zack Cooper and Carnegie Mellon’s Martin Gaynor, the former head of the Federal Trade Commission. Using U.S. hospital price and utilization data for public and private payers, they found:

“First, health care spending per privately insured beneficiary varies by a factor of three across the 306 hospital referral regions (HRRs) in the US. … Second, variation in providers’ transaction prices across regions is the primary driver of spending variation for the privately insured, whereas variation in the quantity of care provided across HRRs is the primary driver of Medicare spending variation. Consequently, extrapolating lessons on health spending from Medicare to the privately insured must be done with caution.”

The New York Times headlined its report on that study: “The experts were wrong about the best places for better and cheaper healthcare.” The findings, its journalists concluded, “are likely to force a rethinking of some conventional wisdom about health care.”

A series of studies conducted by the Rand Corporation over the past decade found prices paid by private insurers were on average 2 1/2 times what Medicare pays for the same services. Paid by different insurers also varied widely, even within the same market.

In a 2019 article eulogizing Reinhardt, Anderson and colleagues doubled down on their “it’s the prices, stupid” analysis:

“The conclusion that prices are the primary reason why the U.S. spends more on health care than any other country remains valid… On key measures of health care resources per capita (hospital beds, physicians, and nurses), the U.S. still provides significantly fewer resources compared to the OECD median country. Since the US is not consuming greater resources than other countries, the most logical factor is the higher prices paid in the U.S. Because the differential between what the public and private sectors pay for medical services has grown significantly in the past fifteen years, U.S. policy makers should focus on prices in the private sector.”

That’s precisely what happened during the Biden administration, which launched few new experiments to rein in overutilizaton. Experiments aimed at curbing overuse, dubbed value-based care, had shown little impact on use and mostly failed to save the government money (the major exception was Maryland’s all-payer pricing system for payers tied to global budgets for providers, which I helped analyze here and here). It focused instead on curbing health care prices for both the private and public payers.

Under Biden, the Department of Justice and Federal Trade Commission stepped up antitrust scrutiny of hospital, insurance and physician practice mergers. It included Medicare drug price negotiating authority in the 2022 Inflation Reduction Act, which it hopes will have spillover effects on private sector drug prices. The administration cancelled its “value-based insurance design” pilot project with the insurer-run Medicare Advantage plans, which wound up increasing payments to already the overpaid privatized plans.

Those actions have had little short-run effect. Health care prices continue to rise at a slightly faster rate than consumer prices and overall health care spending continues to grow at about the same rate as the general economy.

The return of overutilization

A new study that appeared in JAMA Health Forum last week suggests it may be time for another course adjustment, one that is the exact opposite of the hatchet job being carried out by the Trump administration with its proposed cuts to Medicaid and other public sector programs. Those cuts will inevitably throw people off the insurance rolls and force increases in public and private prices to pay for the newly uninsured’s health care costs, which will be made more expensive by their delaying care for their chronic diseases.

The new study’s bottom line? Drum roll please: Variation in the use of services — the way medicine is practiced between regions — accounts for nearly two-thirds of spending variation across the country. Price variation accounts for only a quarter.

How did Joseph Dieleman and colleagues at the University of Washington’s Institute for Health Metrics and Evaluation (IHME) arrive at that conclusion? They have compiled what appears to be the largest health care claims database in the U.S., which they are making freely available to public health researchers, public officials, journalists and individual citizens.

The database incorporates more than 40 billion claims paid by private insurers, Medicare, Medicaid, and individuals out of their own pockets between 2010 and 2019. It covers 3,110 counties, all but 40 in the U.S. It includes services delivered for 148 different health conditions; their prices; patient demographics; and which payer-type (Medicare, Medicaid, private insurance, and individual out-of-pocket) paid what price for each service.

It contains data from both traditional Medicare and Medicare Advantage plans; and from state Medicaid programs and their Medicaid Managed Care Organizations (the private insurers that now cover over 75% of state program beneficiaries). It drew private insurer claims from the Health Care Cost Institute, three private firms and the government’s Agency for Healthcare Research and Quality. The database includes nearly 77% of all health spending, leaving out only several large federal payers like the Veterans Administration, the Defense Department and the Indian Health Service.

The primary finding in the first study to use this massive database found that fully 65% of all spending variation was due to differences in utilization, not prices. Prices aren’t irrelevant, of course, especially when it comes to private insurer spending. But higher prices only accounted for 24% of the difference in cross-county comparisons, with disease prevalence (7%) and age (4%) accounting for the rest.

“Utilization is the primary factor leading to cross-boundary variation in spending,” Dieleman told me in an interview. The paper noted private insurance and out-of-pocket spending had more of their variation between counties based on service prices compared to Medicare, where it was almost entirely due to differences in utilization. “But wherever you look, utilization explains more than 50% of cross-county comparisons. This is really important,” he said.

Jon Skinner, a professor at Dartmouth long associated with the now defunct Dartmouth Atlas (its archives can still be accessed online), agreed. He co-authored a paper in 2022 that showed “quantity is higher for commercial payers where quantity is higher in Medicare,” he said. “Practice patterns are similar.” Skinner served as an advisor to the IHME project.

I contacted Gerard Anderson of Johns Hopkins for his thoughts on this latest study. He declined to comment.

Medicare Advantage gets more; the poor get less

There were several other findings in the study that policymakers ought to pay attention to. Among Medicare beneficiaries, those covered by Medicare Advantage plans used fewer services than those in traditional fee-for-service Medicare. Counties where the median MA coverage rate was in the 75th percentile (MA now covers around 51% of beneficiaries) used 4.9% fewer hospital inpatient services; 4.4% fewer nursing home visits; 2.7% fewer drugs; and 1.9% fewer physician visits compared to those in traditional Medicare.

There are at least three possible explanations for those differences. MA plans are enrolling healthier patients; MA plans are holding down care utilization through prior authorization and denials; or MA plans are managing care better than unmanaged traditional Medicare, which can be achieved through greater use of preventive medicine and intervening earlier with people at risk of developing the chronic disease that lead to costly hospitalizations and nursing home stays. Given the likelihood that it is probably some combination of the three in most areas of the country, policies seeking to eliminate the extra payments for MA plans should be carefully targeted to prevent the abuses, which have been documented in numerous recent studies and press accounts.

(The biggest abuse by Medicare Advantage plans is the use of home visits and physician coding to document illnesses among beneficiaries that aren’t being treated and wouldn’t show up in a claims database. That practice, which costs Medicare an estimated $80 billion annually in unnecessary payments to plans, is currently under investigation by the Justice Department, according to this morning’s Wall Street Journal.)

The IHME study also found people with higher household income made greater use of all health care services except emergency room and inpatient hospital care. Moreover, the higher the household income (and the likelihood of being well-insured) was “associated with an increase in prices for both physician and hospital services.”

Finally, people living in lower-income households with more serious health problems and less insurance coverage obtained fewer services. “Taken together these findings suggest a concerning inequity: the counties with the worst health, lowest insurance rates, and lowest incomes are receiving the least amount of health care,” the study concluded.

What path forward?

The best way to address both overutilization and high prices is something I will address in a subsequent post. For now, suffice it to say the political landscape for addressing either issue is dreadful.

It’s almost a given that the Biden era’s expanded subsidies for individual Affordable Care Act plans will not be renewed by the Republican Congress when they expire later this year. The GOP-run Congress has also made Medicaid its prime target for massive cuts to help pay for extending the Trump tax cuts for the wealthy and big corporations.

Either or both will raise the uninsured rate and lead to sharp increases in private insurance prices and premiums, which will eventually be passed along through higher prices for the 180 million Americans who are privately insured.