If 2026 was the year of gas prices, 2027 will be the year of health-insurance premiums. The cost of employer-provided health coverage—the kind that 160 million Americans get—is expected to rise by as much as 11 percent, roughly double the average annual increase over the past decade. “These are the highest numbers we’ve ever seen,” Ellen Kelsay, the president and CEO of the Business Group on Health, an organization that lobbies on behalf of large employers, told me. “And the current projections are likely an understatement.”
Much of the increase will be passed on to employees—either directly, in the form of higher premiums, co-pays, and deductibles, or indirectly, in the form of reduced benefits, smaller raises, and slower hiring. The biggest reason this is all happening is deeply perverse: A law designed to end surprise medical billing has turned into a tool for squeezing even more money out of insurers and raising the cost of care for everyone else.
The upcoming spike is a case of a bad situation getting worse. Over the past decade, the cost of employer-sponsored health insurance has increased by 50 percent, much higher than overall inflation. A 2026 KFF survey found that the average yearly cost for a family of four was about $28,300, of which employees paid about $7,400. The reasons for this trend are well-documented: an aging workforce in need of more care, an increase in the prevalence of chronic illness, and a dramatic consolidation of the hospital sector.
Next year’s projected increase is being driven by three additional factors. The first is the mass adoption of GLP-1s. The share of Americans using these drugs for weight loss nearly quadrupled from 2024 to 2026. Even after discounts and rebates, the estimated cost per patient is about $7,000 to $9,000 a year. Projections from Marsh, a health-care-consulting firm, imply that GLP-1s will account for up to a quarter of next year’s rise in costs compared to the preexisting trend.
The second explanation appears to be AI. Over the past few years, providers have begun adopting AI tools that record and transcribe their meetings with patients. The ostensible justification for these systems is to give the doctor more time talking with the patient and less time filling out paperwork. But these products are designed to hunt for and maximize revenue from moments during an appointment that a doctor ordinarily might not bill for, such as noting a patient’s chronic disease during a routine checkup.
[Jenisha Watts: The impossible predicament of the uninsured]
The companies that make these tools are very open about this purpose. Ambience Healthcare urges doctors to use its products to “protect your bottom line with stronger documentation” and “capture every dollar earned—automatically.” Another company, Suki, claims that its AI scribes generate up to $6,566 in revenue per clinician per year by increasing the share of patient visits classified at more complex billing rates. Analyses from Marsh and the Business Group on Health imply that AI coding inflation will explain up to a quarter of next year’s extra health-insurance inflation.
The final and likely largest explanation for the upcoming cost spike is also the most surprising—and infuriating. Before 2022, if a patient visited an in-network hospital but received care from an out-of-network provider, their insurance would typically cover only the in-network rate, and the patient would receive an unexpected bill for the remaining amount. The No Surprises Act, which was passed in 2020 and came into effect two years later, made this practice illegal. The law imposed a seemingly simple solution: Patients in these situations would be responsible for only their portion of the in-network cost.
On the back end, however, the reform was more complicated. The out-of-network provider and insurer would be required to negotiate on any further payment among themselves. If the two sides couldn’t come to an agreement, they would enter a process known as independent dispute resolution, or IDR, in which both parties would submit their desired payment amount to a third-party arbitrator. Government agencies estimated that the law would give rise to only about 17,000 of these disputes a year, given that both sides would likely want to avoid the risk of having to accept the other’s number.
They were wrong. After the law took effect, health-care providers began submitting far more arbitration requests than expected, often demanding much higher compensation than any insurer would be willing to pay. The trend began slowly in 2022 and increased with each passing year. In 2025, providers filed 2.6 million disputes. (Technically, either side can initiate arbitration, but in practice, arbitration happens only once an insurer refuses to pay the amount requested, meaning that providers are the ones triggering the process more than 99 percent of the time.)
According to a report from Georgetown University’s Center on Health Insurance Reforms, half of those disputes originated from just two provider practices: Radiology Partners, which is majority-owned by private equity and venture capital, and TeamHealth, which is owned by the private-equity giant Blackstone. Another 27 percent came from HaloMD, a company that, as Stat has reported, was founded in 2022 specifically to help providers earn as much money as possible from IDR disputes. “This really isn’t about your average doctor trying to get a fair rate,” Jack Hoadley, a professor emeritus at Georgetown who co-authored the aforementioned report, told me. “It’s an active strategy by a few companies with a lot of money trying to game the system.” The volume of claims was 50 percent higher in the first half of 2026 than in the same period a year earlier, and experts expect it to rise even faster next year.
The entities taking advantage of the IDR process tend to point the finger at insurance companies. “The health insurance lobby’s false narrative that the IDR process has increased medical spending is not supported by the data,” a spokesperson for HaloMD told me in an email. A spokesperson for TeamHealth wrote to me, “TeamHealth supports the No Surprises Act and its independent dispute resolution process because it works. Patients are protected from surprise medical bills, and providers have an essential path to challenge insurers’ attempts to deny, delay, or lower fair reimbursement for the life-saving care we provide.” And Radiology Partners sent me a document arguing, “Congress thoughtfully designed the No Surprises Act to protect patients from surprise medical bills while preserving sustainable payment rates that protect patient access to care. Unfortunately, some insurers have disrupted that goal.”
[Annie Lowrey: Annoying people to death]
Since the law was enacted, providers have won about 85 percent of submitted disputes. In 2025, the median award they earned was more than four times the typical in-network cost for the service in question. (HaloMD notes on its website that it has a win rate of more than 88 percent and that the average reward is more than five times the qualifying payment amount.) In one case reviewed by The New York Times, a Dallas surgeon who had accepted a patient’s insurance was paid $1,843 for a prostate-removal operation; the surgeon’s assistant, who had not accepted the insurance, went through arbitration and made $50,456 for the same procedure. The Times investigation also revealed that a single plastic-surgery practice had submitted more than 6,000 arbitration claims, which had earned his practice as much as $440,000 for a single breast-reduction surgery. (The typical cost for breast-reduction surgery in the United States ranges from $9,000 to $15,000.)
Some providers argue that they win so often because insurance companies refuse to negotiate in good faith. An analysis by a group of physicians’ lobbies points out that in 8 percent of IDR disputes, insurers have submitted claims of $1 or less, and in about 25 percent of cases, the insurer didn’t submit a counter at all, resulting in an automatic win for the provider. But in a forthcoming analysis, Hoadley finds that 90 percent of insurer offers are either at or above the qualifying payment amount. When insurers don’t submit their own claim, that’s likely because they can’t keep up with the volume of submissions. An executive at the Health Care Service Corporation, a large health-insurance plan, recently told a conference audience that his company currently had “well over 400 people just working IDR disputes.”
Several recent papers by health-policy scholars offer a compelling explanation for why providers so consistently win such huge awards: The arbitrators have strong financial incentives to rule in their favor. IDR arbitrators are not public servants working at a federal agency; they are employees of a handful of for-profit companies and nonprofit organizations that are compensated based on the number of claims they process.
This creates perverse incentives. The arbitrators get paid only if providers submit cases, and providers will be more likely to submit cases if they expect to win. In many cases, providers get to choose which arbitration company will hear their case. “The more these arbitrators rule on behalf of providers, the more money they ultimately make,” Christopher Whaley, a health economist at Brown University, told me. “So what you end up with is a competition between arbitrators to be the most provider-friendly.”
The arbitration organizations that side most often with providers indeed turn out to receive the most submissions. Hoadley and his co-author, Kennah Watts, point out that in the first half of 2024, the arbitration company that received the most cases decided more than 90 percent of them in favor of providers; the entity with the least cases submitted, less than 1 percent of the total, decided in favor of providers only a third of the time. (Making matters even more ethically dubious, some private-equity firms have invested in both the arbitration companies that hear claims and the practices submitting the claims, according to the Private Equity Stakeholder Project.)
The costs of this unintended consequence are enormous. The New York Department of Civil Service, for example, has incurred more than $200 million in additional claim payments from IDR for the state-employee health plan, which the agency described as “a primary contributor to the nearly 10 percent increase in premium rates” in 2025. At the national level, Hoadley and Watts estimate that the combination of administrative costs and payments above the in-network rate added up to a total cost of about $1.6 billion a year from 2022 to 2024. In 2025, that number spiked to $16.6 billion. Experts expect it to be even higher in 2027 as more and more providers take advantage of a system that is basically handing out money. Projections by consulting firms imply that this could account for 25 to 40 percent of the extra rise in employer health-insurance costs next year. “I don’t think we’ve seen the worst of it yet—not even close,” Whaley told me. “This is a tidal wave that could totally wipe out health-care affordability if it isn’t fixed.”
[Toluse Olorunnipa: ‘We are looking at a massive crisis’]
Employers are already scrambling to deal with these rising costs. Disney recently announced that it would be dropping health coverage for working spouses with access to their own coverage next year. Starbucks is one of several major companies that are ending their coverage of GLP-1s for weight loss. Deloitte is rolling back its parental-leave and IVF-funding benefits for certain employees. But these tactics will go only so far. Higher premiums appear inevitable. According to KFF, employee premiums for family coverage increased by 9 percent in 2026, the fastest rate since 2010. If current projections are correct, that figure will be even higher next year, and likely will be accompanied by higher deductibles and co-pays.
The good news is that, unlike structural factors driving up health-care costs over the long term, the unintended consequences of the No Surprises Act should be susceptible to a legislative fix. (One House bill, for example, would scrap the IDR process and tie compensation to existing in-network rates.) The cost-of-living crisis in America is the result of a complex set of factors that will not all be easy to reverse. At the very least, Congress should start by solving the problems of its own making.