Health Issues

KFF designs, conducts and analyzes original public opinion and survey research on Americans’ attitudes, knowledge, and experiences with the health care system to help amplify the public’s voice in major national debates.

Forecasting the year ahead in health policy is always treacherous, because events intervene and screw up even the best predictions. I am sure lobbyists already have a full agenda of actions Congress might take that will affect the interests they represent. In health care, even pinprick changes in legislation can move billions of dollars. But my working theory is that the sharp partisan divide in Congress, and possibly even sharper disagreement on health care policy, can produce only small-ball actions on health next year. Plus, it’s an election year, when we don’t usually see major legislation. So, what should you watch for that will really matter for people, policy, and politics?

First and foremost is the role that health care affordability will play in the midterms. Assuming there is no deal on the enhanced ACA tax credits, spiking premium payments in the Marketplaces will become the national symbol for voters of concerns about their health care bills. Bad as the problem will be on its own, the public will feel it is much worse as media coverage and Democrats (as a political strategy) give far more people than directly experience the problem the impression that their premium payments could be doubling too, capitalizing on the very real concerns people have about affordability throughout the health system. Many polls will start to show that out-of-pocket payments for health care run even with, or eclipse, the cost of food and utilities as an economic concern. Democrats will prosecute the issue to the fullest, and Republicans will generally try to shift the subject and fight on other issues. Health care affordability will be in the spotlight, but how important it is as a vote-and-turnout driver remains to be determined.

Second, again assuming there’s no deal on the tax credits, we’ll see how enrollees actually respond in the Marketplaces. What share of the 24 million enrollees switch to cheaper high-deductible plans? Who are they? What is the impact of doing that on their financial security and health-seeking behavior? What happens to older and sicker enrollees who need better, more comprehensive coverage? How many millions choose to be uninsured in 2026 and who are they? What happens to the uninsured rate in the U.S.?

Third, states that have expanded Medicaid will be gearing up for Medicaid work requirements, which kick in in 2027. Red states may be looking for flexibility to implement the toughest possible requirements and reduce their Medicaid rolls and spending. Blue states will be looking for nooks and crannies in the law and the rules to lessen the impact in the hopes that Democrats seize control again in 2028 and reverse the requirements. Having implemented state welfare work requirements myself, I know there is always some ability to shade implementation one way or the other depending on the goals of a state. In our case in New Jersey, we had little interest in kicking people off welfare (and into homelessness or deep poverty, which we’d also have to address), but we did have a big interest in providing job training, childcare, transportation, and a pathway to jobs.

Fourth, after years of moderate increases (not as people experience it, but as we experts quantify these things), health costs will increase more sharply again. Employer premium increases may not touch double digits but could come close. The average cost of a family policy for employers could approach $30,000. It’s likely that cost sharing and deductibles will rise again after plateauing for several years. Cost pressures may lead to action in Congress on such measures as site-neutral payment for hospitals or restrictions on PBMs. Employers and public payers are increasingly skittish about the costs of GLP-1s for weight loss. It’s possible GLP-1s could turn from today’s dilemma to a technology-diffusion success story of sorts. Costs are coming down, pills are on the horizon, and payers are developing more sensible guidelines for their initial and long-term use. In a country where the dam breaks on every effective new medical technology and it’s rapidly disseminated at high costs, GLP-1s could emerge as a more balanced and sensible example of technology diffusion, albeit, like most new medical technologies, still one that increases costs.  

Fifth, the Trump administration has put pressure on drug prices through a variety of initiatives. Probably the most important is Medicare drug price negotiations, which began in the Biden years, but together they are putting pressure on drug prices. Sensing where the public is, President Trump has helped engineer a shift in the geopolitics of health care in which drug companies are no longer the NRA of health care and mostly untouchable (he obviously has not tried to engineer the same change for the NRA). Still, retail drug spending is only 9% of overall health spending (16% for employers) and there is growing awareness that hospitals gobble up the single largest share of the health care dollar.

Proposals to cap hospital prices or put hospitals on a budget seem to be coming back into fashion again but are non-starters with Republicans in charge, as of course is single payer. For 2026, we can therefore look generally for greater attention being paid to hospital prices and possibly greater action at the state level, where several states have established hospital cost targets with varying degrees of teeth. 

Sixth, people (and providers) really do hate prior authorization review. Will the administration’s voluntary effort to work with industry to streamline and pare back prior authorization result in any concrete relief for patients when it kicks in this year? Voluntary health care industry efforts have generally disappointed, as did the now-forgotten but once infamous industry Voluntary Effort (“the VE”) to reduce costs that was mounted to fend off Jimmy Carter’s aggressive cost-containment plan. The jury is out. 

Seventh, four in 10 Americans say they identify in some way with the MAHA agenda. We’ll learn more in the coming year about which tenets of this loose movement have staying power. Everyone is for exercise and for healthier food for kids (although I do still miss those Hostess Cupcakes and Twinkies that I had as a kid). But people who have serious illnesses may grow tired of the central MAHA notion that they are personally responsible for disease and may increasingly make the connection between their own need for medical care and cuts in Medicaid and ACA coverage. There is some overlap between MAHA and the anti-vax movement. Another thing to watch: will vaccination rates continue to decline as the administration chips away at universal vaccination and trust in vaccines?

Eighth, AI obviously already means a lot for physicians, researchers, hospitals, insurance companies and investors, with profound implications in some areas of medicine already, such as radiology. But when will AI start having real meaning for the things people care about most: their costs and their ability to navigate the health system? The implications and practical applications for consumers receive far less attention, as generally does the relatively low level of trust in AI consumers start with. Possibly, 2026 will be the year when attention shifts more seriously to the consumer side of AI in health. 

All told, it will be a big year ahead in health policy. And this list only just scratches the surface. There’s also Mifepristone, and family planning, everything not happening now in HIV, CDC, NIH and FDA, and CMS waivers and CMMI, the Medicare Advantage wars that never end, and price transparency, and more. But there probably won’t be a lot of significant action in Congress. Still, every incremental change is a big fight in health care, and hard won, and in a hyper-partisan Congress, 2026 will be no different. 

View all of Drew’s Beyond the Data Columns

When it comes to controlling health spending, the fundamental difference between our health system and those of other wealthy nations is that they each have a way to control total health spending, and then they leave health professionals and health care institutions relatively free, compared to us, to do their jobs with the resources they have. Those other countries spend a much smaller share of GDP on health care and some of them limit care just by spending less money on health, but generally, their health professionals maintain greater professional autonomy. They make less money but maintain high social status and respect. The United States, by contrast, has no mechanism to control overall health spending in our fragmented health system and then to compensate, we micromanage to control costs, making health care overly complex for patients and, too often, making the experience of providing it a desultory one for health professionals. It’s a primary reason why almost half of our doctors say they regret choosing medicine—not a good outcome no matter what you think of our health system or of doctors. 

The posterchild of how this happens in the U.S. is prior authorization review, the system by which insurance companies decide whether they will authorize payment for a procedure or a diagnostic test or a drug, or if they’ll authorize it in one setting or another. These days, prior authorization is becoming a duel between provider AI tools helping hospitals and large group practices out maneuver insurance company prior authorization, and insurance company AIs trying to weed out necessary from unnecessary care and promote “value” as well as, of course, insurance company profits. Everybody thinks they’re the good guy in the prior authorization dance. There’s the doctor who 99% of the time is looking out for the patient. There’s the insurance company, which thinks it’s on guard against unnecessary and unnecessarily costly care and bad apple doctors. And then there’s the patient, who may be waiting anxiously for the results of a crucial test or hoping that a procedure will bring relief.

Other than out-of-pocket costs, I would wager that prior authorization review is the one thing consumers find most burdensome in our current health system. I will test that out on one of KFF’s upcoming polls.

A proposal to eliminate prior authorization altogether could be the single most tangible and popular “health reform” idea a candidate could propose, Republican or Democrat. The speech would write itself: “I propose to get insurance companies out from between you and your doctor. No more hours on the phone trying to get your MRI approved. Your doctor will no longer need a team of administrators just to fight with your insurance company.”

It is, of course, not that simple. As with everything in health policy, there are tradeoffs, and addressing them makes things complicated quickly.

The most comprehensive analysis of eliminating prior authorization was conducted by Milliman in 2023 (Potential impacts on commercial costs and premiums related to the elimination of prior authorization requirements). It was a national study commissioned by the Blue Cross and Blue Shield Association in Massachusetts in the middle of ongoing fights there about meeting state cost targets. Boiling down a complex study, here are a few essentials:

• Milliman’s mid-range estimate for eliminating prior authorization says it would raise premiums by up to 4.8% for plans with a broad range of services. (Credit to Milliman for providing ranges where exact estimates cannot reliably be determined.)
• They worried that eliminating prior authorization could also have a “sentinel effect”—meaning services prior authorization deters just by being there would start to be provided.
• Patient cost sharing could also go up simply because people would be using more services.
• And yes, they concluded that there would also be a significant offsetting reduction in administrative costs (and hassles and anxiety, but Millman doesn’t go into that much).

These conclusions (and more in the analysis) add up to a daunting set of reasons why in our fragmented system, with no great way to control costs or limit unnecessary care, we seem to be stuck with prior authorization review. As frustrating as prior authorization can be for patients and health professionals, it’s hard to imagine anyone swallowing an almost 5% premium increase to abandon it, or risking the bad outcome from an unnecessary procedure an insurance company might sometimes catch. Still, traditional Medicare has operated almost entirely without any form of prior authorization review for decades and survived the experience as health care’s most popular and politically sacrosanct program. (Its competitor, Medicare Advantage, features prior authorization.)

Consider a hypothetical test of how much physicians value autonomy versus money and insurers really care about value versus the bottom line. Imagine a “deal” between insurers—or an insurer—and providers in which insurers eliminate prior authorization and in return, reduce payments to physicians by, say, 2% or 3%, to offset some of the expected increase in costs. In effect, physicians would trade income for greater professional autonomy more like what their counterparts have in some other countries, and insurers would get off their high horse about the loftier purposes of prior authorization to reduce administrative costs and bank certain savings, unless physicians make up lost revenue with greater volume. I don’t expect this somewhat fantastical experiment to happen.

The tradeoffs and potential costs of eliminating prior authorization have driven most payers in a different direction: don’t eliminate it, “do prior authorization smarter.” States have mounted a long list of prior authorization reforms to make it less onerous for patients and providers (9 states pass bills to fix prior authorization | American Medical Association). These include “gold carding”: essentially exempting physicians from prior authorization with a good track record of ordering needed services. Or, eliminating or restricting prior authorization for ongoing care for patients with a chronic illness. Take a patient with lifelong chronic migraines who has taken the same prophylactic drug for 20 years and treated the migraines with Sumatriptan for the same period of time. That patient and their provider probably can go with a longer interval of time before the next set of likely identical prescriptions are reviewed.

A trend will be circumscribing the role AI plays in reviews by requiring an actual physician to approve any review determination. That’s politically popular although, of course, physicians working for insurance companies issued many a denial long before AI was around.

Even traditional Medicare has now entered the “do it smarter” game. Recently CMMI announced WISeR, the Wasteful and Inappropriate Service Reduction Model, to test out prior authorization for selected services in six states. Services targeted include skin and tissue substitutes (controversy surrounding them motivated the demo), electrical nerve stimulator implants, cervical fusion, and more. Reportedly, a “gold carding” demo is also planned. It was perhaps surprising politics that CMS would move to one of health care’s least popular practices in politically sensitive traditional Medicare at this time, but it fits with the “do prior authorization smarter” trend. Earlier the administration and a group of insurance companies also announced a voluntary initiative to streamline prior authorization and make it less burdensome. A poll we did at the time showed that few have much faith that insurance companies will follow through on the initiative.

As these initiatives show, prior authorization is down but far from out, and we are almost certainly stuck with prior authorization with some modifications around the edges. It’s an example of the “small ball” featured in cost containment in health today. However, the impact of getting rid of prior authorization cannot only be measured by actuaries in premium increases. There are also the benefits to be found in reduced complexity, hassles, anxiety and administrative costs that limiting or eliminating prior authorization would have for patients and health professionals who have to struggle through the prior authorization maze. 

View all of Drew’s Beyond the Data Columns