Just over a decade ago, celebrated surgeon and writer Dr. Atul Gawande penned a New Yorker article that called out certain areas of the country for their high Medicare spending.
But despite the public lashing some regions endured, the most recent data from the Dartmouth Atlas show little has changed since then with respect to the communities atop the Medicare spending tower and those at the bottom. Miami was highest among hospital referral regions in the latest price-adjusted data, from 2017, followed by two smaller cities that were also near the top in 2010: Munster, Ind., and Monroe, La. Grand Junction, Colo., Anchorage, Alaska, and Honolulu had among the lowest spending in both years.
“It’s not like someone wakes up one day and says, ‘I’m upset that somebody said I’m a high-spending provider or hospital. I’m going to turn that around,’ ” said Dr. Robert Wachter, chair of the department of medicine at the University of California at San Francisco. “Even if they did, it’s just very hard to do.”
One important thing has changed over the years, however. Between 2003 and 2010, the average compounded growth rate in per-beneficiary Medicare spending was 6.6%. For 2010 to 2017, it was just 1.4%.
That spending slowdown stems from a number of factors, although the weight of each is the subject of debate. A number of Affordable Care Act provisions took effect in 2010, the year the law was enacted. Electronic health record adoption, which helps doctors see which tests and procedures patients have already had, has proliferated since then. And there’s been rising awareness around the problem of overuse.
“Everybody is getting a little bit better compared to where they were, compared to the path they were on,” said Len Nichols, a nonresident director with the Urban Institute and professor emeritus at George Mason University.
One of the most expensive medical technologies is proton beam therapy, which uses gigantic machines to deliver radiation for cancer treatment.
Because of the price tag, only about 30 centers perform proton therapy in the U.S. Two are within 20 miles of one another in Miami.
That’s just one example in a city chock-full of hospitals and doctors. Miami’s unusually high medical spending has been an open secret for decades. The region’s price-adjusted Medicare spending per beneficiary was $13,199 in 2017, compared with $10,317 nationally. Even today, there’s no sign that’s about to budge, said Steven Ullman, a professor at the University of Miami and director of its Center for Health Management and Policy. “It makes life interesting down here,” he said.
Dartmouth Atlas data adjusts for factors like age and regional cost differences to put the focus on a metric its team believes is most important: utilization.
A lot of the regional differences come down to available supply in an area, said Kristen Bronner, the Atlas’ managing editor. If there are a lot of hospital beds or physicians in an area, that supply will get used, which is a big part of what drives higher Medicare spending, she said. Per a rough generalization, Bronner said utilization drives two-thirds of spending differences, while price drives one-third.
Providers in high-cost spending areas aren’t eager to discuss the subject. Several of Miami’s leading health systems did not respond to requests for comment or turned down repeated interview requests, including Baptist Health South Florida, the University of Miami Health System and HCA East Florida.
It’s more than just supply that drives Miami’s high spending, Ullman explained. There’s also the frenzied pace of provider consolidation. When health systems gain market share, it gives them free rein to drive up prices. Decades ago Miami had a wealth of independent hospitals. Today, there are “must have” systems with prices reflecting that, Ullman said. A high rate of fly-by-night Medicare fraudsters also drives up spending there, he said.
Another factor that keeps the needle from moving: Providers are heavily saturated in fee-for-service contracts, so they have incentives to run more tests and procedures.
“We do a lot down here per patient,” Ullman said.
Sticky practice patterns
Part of what makes the rankings so sticky is they’re a reflection of deeply ingrained cultural practices, influenced by generations of local physicians learning the trade.
Consider the difference in practice patterns from coast to coast, Nichols said. In Northern California, the gestalt of avoiding hospitalization has been the norm for decades. In Miami, by contrast, your cousin might be hospitalized for a few days after feeling dizzy and falling down, he said.
“Norms really matter, and they matter more in the art of medicine than the science of medicine,” Nichols said. “It’s still largely an art.”
While spending is largely driven by provider practice patterns, patients could play a role, too, Bronner said. In Florida, it’s possible patients are demanding more treatment, especially near the end of life.
UCSF’s Wachter likens an area’s healthcare spending trends to voting patterns. Neither are likely to change in a five- or even 10-year period.
One area that did see some movement between 2010 and 2017: McAllen, Texas. The southern Texas city was the main subject of Gawande’s New Yorker article. McAllen was the highest Medicare spender by hospital referral region in 2010, but had slid to eighth-highest in 2017. McAllen’s per-beneficiary spending went from $13,842 in 2010 to $12,792 in 2017.
Like their peers in Miami, hospitals in McAllen didn’t respond to interview requests. Wachter said physicians there weren’t happy about the attention they got following the article, which highlighted their duplicative technologies and unscrupulous practice patterns. The article went so far as to imply unethical behavior and profiteering, Wachter said.
“That doesn’t feel good,” he said. “I know they were shamed. I know they were trying to reel it in.”
A big spending influencer that isn’t accounted for in the Dartmouth Atlas data, however, is the percentage of an area’s residents who rely on Medicare versus commercial insurance, Nichols said. McAllen has a lot of Medicare and Medicaid beneficiaries, which means providers have had to become adept at making money under those programs, he said.
Conversely, where Medicare spending is low, some research has shown commercial spending tends to be high. That means low Medicare spending areas in Dartmouth Atlas’ data aren’t off the hook.
A changing Grand Junction
Healthcare leaders in Grand Junction, Colo., were actually surprised to see their region still showing up as among the lowest Medicare spenders in 2017.
Grand Junction, nestled in the heart of Colorado’s wine country, has changed a lot over the past decade. Namely, a much higher proportion of the area’s doctors are now employed by the local hospitals than in the past, which has put a damper on collaboration among providers, said Dr. Gregory Reicks, chief medical officer of the Mesa County Physicians IPA.
Patrick Gordon, CEO of Rocky Mountain Health Plans, a UnitedHealthcare subsidiary based in Grand Junction, put it more bluntly: “It’s just an outright competitive market,” he said. “Active, open, aggressive hospital competition.
Building out programs to garner revenue and market share.”
Grand Junction has two hospitals: St. Mary’s Hospital, which is owned by SCL Health, and Community Hospital, part of the Colorado West Healthcare System. Both declined to comment.
Grand Junction’s Medicare spending per beneficiary grew from $6,993 in 2010 to $7,594 in 2017, which is still far below the national average.
Regardless, when asked why Grand Junction still performs better on utilization, Reicks and Gordon pointed to the same thing: a robust end-of-life care system.
About two decades ago, providers in the community joined together to launch a not-for-profit hospice provider called HopeWest. Since then, HopeWest has enjoyed broad utilization across the community. Reicks said that’s lowered Medicare end-of-life spending and has meant fewer people dying in hospitals.
Not only that, the region’s providers and health plans launched one of the country’s first health information exchanges, Quality Health Network, about 17 years ago, Reicks said. Doctors use it to check what tests or procedures patients have already gotten from other providers.
But despite the low Medicare spending, Reicks and Gordon agreed that high commercial spending is still a problem in the isolated region, where some local providers set prices using a degree of monopoly power.
In Anchorage, a long-standing lack of access to Medicare providers, especially primary-care physicians, could contribute to the low per-capita spending there, said Jared Kosin, CEO of the Alaska State Hospital and Nursing Home Association. Anchorage had the second-lowest spending in both 2010 and 2020. “It’s a pretty challenging market,” Kosin said. “Access to care is something we’re always dealing with.”
Why overall spending growth slowed
For his part, Nichols believes the ACA is responsible for more than half of the decline in Medicare spending since 2010.
Within the ACA, the biggest source of Medicare savings was its reduction to marketbasket updates on hospitals. Marketbasket updates, which help determine hospital reimbursement, are estimates of the expenses hospitals shoulder, such as wages and utilities. The ACA held that those updates be reduced by an amount equal to productivity. In practice, that meant facilities were forced to become as much or more efficient than the economy as a whole, Nichols said.
Medicare payment reforms rolled out by CMS’ Center for Medicare and Medicaid Innovation likely also helped some, but the ACA’s crafters relied in part on the marketbasket reduction to pay for the expansions in Medicaid and subsidized private coverage.
“Hospital guys were looking at the world and seeing that Washington had actually committed to getting this coverage expansion to pay for itself,” Nichols said, “and that was through taxes on the wealthy and Medicare savings, which comes out of Medicare spending.”
Paul Hughes-Cromwick, an independent healthcare consultant, agreed that the ACA deserves credit, calling it “the least told story in America about health policy.” But like the decline in smoking rates, there are many contributing factors. The Choosing Wisely campaign flags low-value care. Insurers more tightly manage prior authorization. The 2009 stimulus bill gave providers money for adopting electronic health record platforms.
“Each one thing? I don’t know. But together, yes,” Hughes-Cromwick said.