This summer, a parasite most Americans had never heard of has sickened thousands. Since May 1, federal officials have counted more than 6,700 confirmed or suspected cases of cyclosporiasis across 34 states.

The intestinal illness is miserable: severe diarrhea, cramping, and fatigue that can last weeks. It is rarely fatal and can be treated with a common antibiotic once diagnosed. Yet the infection spread widely before health investigators identified even one major source of the infection. The deeper story is how late they saw this outbreak, which has been linked to contaminated lettuce.

Subscribe now

An ill-suited reporting network

Here’s what must happen for one infection to become a public health data point.

A person must become sick enough to seek care. The doctor must suspect cyclospora. Because it does not appear on a routine test, the doctor must then order the specific one that detects it. A laboratory must run that test and find the parasite. The result must reach the state. State officials must confirm the case and report it to the Centers for Disease Control and Prevention. Finally, someone at the state or CDC must recognize that a particular week’s numbers are outside the normal range.

Time is lost at every step. Symptoms can take a week or longer to appear, then fade and return before many patients seek care. Testing, reporting, and confirmation add further delays. Many weeks can pass between infection and the moment a case appears in a federal database.

This is the nation’s detection system working as designed: counting cases accurately, one patient at a time, after people become sick. It’s a fine system for measuring a stable problem but poorly suited for a dynamic one. While the signal moves slowly through clinics, laboratories, and government agencies, contaminated food remains on the shelf.

The federal government’s cuts

Over the past year, the federal government made this system less reliable.

Last July, state reporting to the CDC’s Foodborne Diseases Active Surveillance Network, or FoodNet, became optional for six of the eight pathogens it had actively tracked, including cyclospora. National case reporting continued through other systems. But FoodNet had spent nearly three decades building the detailed, population-based reference data that help officials distinguish a true spike from random variation.

Specialized expertise also shrank. According to the former leader of the CDC laboratory that responds to cyclospora outbreaks, the lab was reduced from 11 people to 3. Its scientists perform the genetic analysis that can reveal whether scattered cases came from the same source — critical evidence when produce has moved through a sprawling supply chain.

The federal government also delayed enforcement of a food-traceability rule designed to help investigators follow contaminated products rapidly through that chain. Its original compliance date was in January. Enforcement will now wait until July 2028.

These decisions did not create the outbreak. But federal officials reduced active tracking, diminished specialized expertise, and postponed faster tracing — and the result is an extraordinary surge.

Investigators are still searching for the source. The first patients fell ill in mid-May, and with every week the trail goes colder — memories of meals eaten a month ago begin to fade, and a smaller public health workforce means slower follow-up. The delay compounds itself: the longer a source goes unidentified, the harder it becomes to identify.

Surveillance tells health officials how large a problem is. When surveillance programs are cut, reported cases usually fall because fewer illnesses are counted, not because fewer people are sick.

That creates a dangerous feedback loop. A poorly measured problem begins to look too small to justify investment. The weakened system then produces even less evidence of harm. Cutting the capacity that documents a problem can erase the argument for restoring it. The United States is now entering that trap as biological risks become more complex.

A new detection strategy

The federal government should restore the surveillance capacity that was lost, rebuild specialized expertise, and implement modern food tracing. Those steps would improve our nation’s response to known threats such as cyclospora.

But even if this happened, the larger weakness would remain. Nearly every system that the United States depends on requires someone to name the threat in advance. A doctor must suspect cyclospora to order the right test. A laboratory must know what it is looking for. Surveillance systems usually use targeted tests selected before testing begins. An unexpected pathogen can therefore spread without producing a recognizable signal.

The United States needs an added layer of detection that scans broadly for unusual biological activity. Metagenomic sequencing can scan wastewater broadly for genetic material from many organisms at once. It will not find every pathogen in every sample but it can reveal changes that targeted tests miss and provide an earlier indication that something unexpected is spreading.

Such a detection strategy would help clinicians know the disease was spreading, epidemiologists know where to look, and food inspectors target their attention. Instead of waiting for sick patients to move one by one through doctors, labs, and health agencies, officials could act on the wastewater warning first.

Cyclospora has a known season, familiar sources, and an effective treatment. Yet thousands became sick before officials realized what was happening.

The lesson extends far beyond one parasite or one shipment of lettuce. The United States has built a public health system that sees biological threats mainly through patients who become sick enough, reach the right doctor who knows the right test to order, and receive the right test and results that make it into a government database.

That system often fails. And it is nearly always late.

The United States can build another layer in front of it, designed to recognize outbreaks before they are definitively identified. Instead, the federal government chose to undermine an already weak system.

This was foreseeable. And there will be more. This is a good time to start building better systems so public health officials can see outbreaks before thousands of Americans get needlessly sick.

In the second half of the Great AI Healthcare Reckoning podcast, we turned our attention to the potential for AI to either worsen existing health disparities or, in my view, harness its tremendous potential to actually reduce the gaps.

Watch the full episode below.

Thank you to Dr. Irene Papanicolas and Dr. Tom Tsai for their contributions throughout this series.

Please feel free to weigh in.

In 2024, the US health care sector employed nearly 18 million people. Politicians from both parties celebrate these numbers routinely. Health care is an economic engine, they say. A pillar of the middle class. A source of good jobs.

It is only half true: Health care does employ a lot of people. But that is nothing to celebrate.

Look at where the jobs are. While the number of physicians in the United States has doubled in the past five decades, the number of administrators has grown by more than 600 percent in that same time period. Billing departments, prior authorization teams, coding specialists, denial management units — these are among the fastest-growing features of American medicine. Other wealthy countries employ more doctors and nurses and far fewer administrators. The jobs the United States should be adding are clinical. The ones that have been added are too often administrative.

When Americans celebrate health care job growth without asking what those jobs actually do, we are measuring the wrong thing. This series has tried to focus attention on the right one: whether health care is becoming more affordable for the American people without sacrificing the medical excellence that saves lives.

There is no doubt that the American health care system is extraordinary at what it does best. When it comes to medical innovation — new cancer therapies, cardiac surgery, rare disease treatment — the United States leads the world. For the very sickest patients, with the most complex conditions, there is arguably no better place on earth to be treated.

That achievement is real. What drives it is the National Institutes of Health, academic medical centers, physician and scientist talent, and decades of public and private investment in basic and applied research. But the billing apparatus, prior authorization bureaucracy, market consolidation, and opaque pricing impose a tax on that achievement rather than enabling it. Affordability is not austerity. Affordability in health care comes from removing the waste that keeps patients from getting the care they need.

Look across the eight solutions in this series and you will notice something. Some rely primarily on markets: aggressive antitrust enforcement, expanding ambulatory surgery centers, value-based payments, and allowing nurse practitioners and pharmacists to practice at the top of their training. Some require government action: enforceable price limits, drug negotiation authority, standardized insurance rules, and administrative simplification. These do not map cleanly onto a party platform. That is intentional — and it is the point.

The political right’s instinct is that markets fix everything. But many health care markets are, in practice, government-protected monopolies. You cannot have a market solution where there is no market.

The left’s instinct is that the government should set the terms. But where real competition exists, it does lower costs and expand access. Neither instinct, applied rigidly, gets to affordability. Both, applied selectively and practically, will.

The only question worth asking about any reform proposal is simple: Does it make care more affordable for more Americans without sacrificing quality? If yes, pursue it. If not, set it aside.

So why does this matter right now, more than it did 10 years ago?

Two forces are bearing down simultaneously. The first is demographic. The number of Americans age 65 and older has already passed 60 million and is projected to approach 80 million by 2040. More older Americans means more chronic disease management, hospitalizations, and long-term care.

The second force is therapeutic. A new generation of cell and gene therapies is arriving — treatments that can cure previously untreatable diseases with a single dose, put certain cancers into lasting remission, and transform the lives of patients who once had few options. These are genuine breakthroughs. They are also often priced at $2 million or more per treatment.

Aging will increase demand. Breakthrough therapies will increase what medicine can do. Both are signs of progress. But progress becomes unaffordable if layered on top of waste.

If the nation arrives at that moment still carrying the existing inefficiency — administrative bloat, inflated prices, protected monopolies, incentives that reward volume over value — the system will not bend. It will break. Act now, while reform can still be thoughtful, or face a reckoning that is crisis-driven, blunt, and almost certainly worse.

The standard response to this kind of argument is patience. There is always another election. There is always a more politically favorable moment just ahead. And so the decision gets deferred — not because anyone concluded the problem wasn’t real, but because the immediate cost of acting always seems to exceed the immediate cost of waiting.

Yet, every year without antitrust enforcement is another year of consolidation that becomes harder to unwind. Every year without price reform is another year of contracts locked in at elevated rates. Every year without administrative simplification is another year in which more people are hired to navigate a system that should be simpler. The constituency for the status quo grows as more livelihoods come to depend on it. Delay compounds the problem and the consequences are concrete.

Rising costs price employers and workers out of insurance, swelling the ranks of the uninsured regardless of what laws dictate. For people who have insurance, premiums and out-of-pocket costs consuming household budgets grow, representing income not saved, not invested, not spent on a decent life — America getting poorer, household by household. On the government side, health care spending crowds out the public investment that actually produces long-term prosperity: a state spending 35 cents of every dollar on Medicaid spends fewer cents on universities and public infrastructure. The country that cannot control its health care costs is defunding its own future.

None of this requires waiting for Washington. States have more power over health care costs than is commonly understood. Governors do not need permission from Congress to repeal certificate-of-need laws, modernize scope-of-practice rules, enforce price transparency, or create affordability boards with teeth. Insurance market rules, Medicaid payment policy, scope-of-practice laws, and the oversight of hospital markets are all substantially state decisions. Indiana has shown what is possible by enacting price caps on hospital systems tied to Medicare rates, and other states should follow its lead.

Federal action remains necessary — on drug pricing, Medicare payment reform, and antitrust law. States cannot go it alone, but waiting for Washington to move first is a costly and unnecessary choice.

The interests that benefit from the status quo are organized and well-funded. The people who would benefit from reform are diffuse — spread across every ZIP code, every income level, every political affiliation. That asymmetry has sustained an unnecessarily expensive system for decades.

But the costs have grown too large to be quietly absorbed. Families feel them. Employers feel them. Governors feel them. The federal budget reflects them.

The question was never whether policy makers knew what to do. This series has hopefully put that argument to bed. The question — the only one that remains — is whether policy makers are finally willing to act.

We know that climate change harms health and the impact falls unequally. The late Pope Francis spoke about Earth as our common home and how it cries out to us because of the harms we have inflicted on her, noting we must “hear both the cry of the earth and the cry of the poor.”

I spoke about how the effects of climate change manifest in human health, and how we must intervene especially to protect the weak. A critical part of that is to build health systems that are resilient and equipped to serve all.

You can listen to my remarks here.

For this piece I’m pleased to be joined by my colleague Dr. Irene Papanicolas, the Provost’s Professor of Health Services, Policy, and Practice and Director of the Center for Health System Sustainability at the Brown School of Public Health.

Please feel free to weigh in.

Consider two patients with the same cancer diagnosis. They are prescribed the same drug — Keytruda, the world’s best-selling cancer medicine. One patient lives in Boston; the other in London. They don’t know it, but this changes everything.

The London patient pays nothing to their doctor — the National Health Service negotiates the price and covers the bill. The Boston patient, who is on Medicare, meanwhile, may pay up to $2,100 out of pocket every three weeks with no supplemental coverage. There is no annual cap.

This is because, in the United States, until recently it’s been illegal for Medicare to negotiate drug prices, and manufacturers can charge whatever they like without any justification.

Most other wealthy nations have made different choices. In the case of these two sick patients, the results speak for themselves.

Few sectors have done more to improve human health than the pharmaceutical industry. Drugs for HIV turned a death sentence into a manageable condition. Statins transformed cardiovascular disease. Immunotherapies and GLP-1s are reshaping care for cancer, obesity, and diabetes. Some of that foundational science was publicly funded, but the costly work of turning basic research into medicines is overwhelmingly private, and it deserves to be rewarded.

The current system does reward genuine breakthroughs. GLP-1s have generated tens of billions of dollars for their manufacturers. But it rewards patent gaming and copycat drugs just as generously. A company that stacks secondary patents to extend a monopoly by a decade earns the same protection as one that develops a genuinely transformative medicine. AbbVie, for instance, filed more than 130 patents on Humira, covering things such as the injector button and known dosages, and keeping cheaper biosimilars off the US market for several years after its core patent expired, while charging roughly $72,000 per year. That is the problem. Good policy should preserve strong incentives for real innovation — not game a system that passes the costs to patients and taxpayers.

The Trump administration correctly diagnosed that Americans pay two to three times what peer nations pay for the same drugs. But its “most favored nation” initiative, which would bring US drug prices in line with prices paid in comparable countries, hasn’t moved the needle. Drug companies keep raising prices, and 2025 had the highest number of brand-name drug list price increases ever recorded.

What has worked is Medicare drug price negotiation, established by the Inflation Reduction Act under the Biden administration and continued by the current one. The first round of 10 negotiated drugs produced price cuts of 38 to 79 percent and an estimated $6 billion in annual Medicare savings. Innovation didn’t stop. That is where serious reform starts.

Negotiate — and mean it

The Inflation Reduction Act proved the concept. When Medicare negotiates, prices come down — and the pharmaceutical industry survives. Critics worried that it would crush innovation. The first round of negotiations, completed in 2024 and capped at 10 drugs, gave a real-world test: All 10 manufacturers agreed to negotiate, prices fell, Medicare saved billions, and a rigorous analysis found the collective innovation impact amounted to less than one drug never being developed. The sky did not fall.

The problem is that the program is too narrow and too slow. Keytruda, the single-largest drug expenditure in Medicare, with $4.8 billion in allowed claims in 2023, cannot be negotiated until 2028 at the earliest. The 2025 budget legislation expanded the orphan drug exclusion, which shields drugs developed for rare diseases from price negotiation. While this carve-out makes sense in theory, it has been expanded to delay negotiation of Keytruda itself, a drug approved for 41 conditions, including some of the most common cancers in America. This narrows the program further at precisely the moment it should be growing. Meanwhile, a second round of 15 drugs will save an estimated $12 billion annually starting in 2027. The direction is right. The pace is not. Medicare is the largest drug buyer in the world. It should negotiate like it is — on more drugs, faster timelines, and with fewer loopholes.

When competition is permitted to work, it works. Stelara, a blockbuster immunotherapy, was one of the first drugs whose price was negotiated by Medicare. Then biosimilar competitors arrived and drove prices down by 90 percent, more than government negotiation had achieved. That is what markets do when they function. The problem is that ours largely don’t.

Pharmaceutical manufacturers game the patent system to extend monopolies well beyond what Congress intended. Drug manufacturer AbbVie built a wall of more than 130 patents around Humira — most of them having nothing to do with the underlying medicine. The result was that competition was delayed, and American patients paid accordingly.

Only 10 percent of biologics losing patent protection in the next decade have a biosimilar in development. Reforming patent abuse, accelerating biosimilar approval, and fixing the perverse incentives that lead pharmacy benefit managers to favor high-list branded drugs over cheaper alternatives would unleash competition the system is currently suppressing. Congress took a step in the right direction last year, but implementation is what matters now.

Lowering the cost of developing drugs matters, too. Clinical trials are extraordinarily expensive, and a meaningful share of that cost reflects regulatory requirements that have grown more burdensome without necessarily improving patient outcomes. The Food and Drug Administration could push for smarter trial design, adaptive approval pathways, and greater use of real-world evidence that could bring development costs down without compromising safety. Lower development costs also mean drugs don’t need to be as expensive to justify the investment.

We already know what drugs cost

There is a deeper problem that negotiation and competition alone don’t fully solve. Manufacturers set launch prices unilaterally, with no requirement to demonstrate that a new drug works better than what already exists, or that its price reflects its benefit. The Food and Drug Administration asks whether a drug is safe and effective. It does not ask whether a $200,000-a-year drug is worth $200,000 a year.

Many wealthy nations have some version of that question built into their systems, including the US. It has the Institute for Clinical and Economic Review, or ICER, an independent, privately funded agency that rigorously analyzes whether drug prices are justified by clinical benefit. Its work is credible and widely respected. It is also entirely advisory. No payer is required to consider it. Giving ICER’s analyses real weight in Medicare’s negotiating positions — not as a trigger for coverage denials, but as an anchor for what we are willing to pay — would help close the gap between what drugs cost and what they do for patients.

Expanding negotiation, fixing the patent system, and requiring that price reflect value at launch would not eliminate pharmaceutical innovation — they would redirect it toward the patients it is supposed to serve. The National Health Service in England pays half for the same drug as in the United States because their system decided the price had to be justified. Ours never asked.

As of Monday, the World Health Organization had reported nine hantavirus cases linked to the expedition cruise ship MV Hondius — with seven confirmed infections of the Andes strain and two probable infections. Three people are dead. The ship, refused entry by port after port, finally docked in the Spanish Canary Islands. The first positive identification of hantavirus infection among ship passengers was May 2, while the first death on the ship was April 11. As of April 24, 25 percent of the passengers had already disembarked.

The Andes strain is the one hantavirus capable of spreading from person to person. North America has its own strain, Sin Nombre (or Unnamed in Spanish), which is not known to spread between people. If there were an outbreak of the Andes strain in rural Nevada or even Washington, D.C., would health officials know? Almost certainly not. Not before it had spread far and wide.

Passengers began falling ill weeks before hantavirus was identified. The ship continued its voyage. Some passengers disembarked. They flew home — to Switzerland, to South Africa, to the United States. They went through airports. They returned to their families. By the time two laboratories confirmed the diagnosis, the passengers who carried the virus were already scattered across the globe.

This is a pattern that has been repeated. The original SARS virus in 2003 spread from Guangdong, China, to Hong Kong to 29 countries before it had a name. COVID-19 was probably seeding multiple continents before the first genome was published. By the time health officials understood what the world was facing, COVID was everywhere.

We only catch outbreaks in retrospect.

Subscribe now

The surveillance gap

Sin Nombre virus, carried by deer mice, has circulated quietly in the American West for decades. Last year, Betsy Arakawa — wife of actor Gene Hackman — died from hantavirus in her home in Santa Fe, briefly raising the alarm about a long-ignored threat.

The strain she likely had cannot spread person-to-person yet, but viruses evolve. And Andes itself could arrive here. An infected traveler with flu-like symptoms could spread it widely here before anyone suspects it.

Hantavirus became a nationally notifiable disease in 1995. When a doctor suspects it and actually tests a patient (not a guarantee for a rare disease) and if the patient tests positive, then the physician should file a report so the case can get counted.

The United States does not have continuous, pathogen-agnostic biological intelligence. It has a passive, symptom-triggered system that requires a sick person, an alert physician, and a correct diagnosis to register a single data point. That is what is called surveillance. It’s actually just case counting after the fact.

The Hondius timeline illustrates what this means. Passengers were sick. No system flagged it. The detection happened the old-fashioned way — sick people, worried doctors, weeks of elapsed time. If Andes transmitted as readily as influenza does, the passengers who scattered across the globe before the diagnosis was made would have seeded a large global outbreak. We are lucky that biology is on our side this time.

The United States built infrastructure, then cut it

The Centers for Disease Control and Prevention has lost more than 3,000 staff in the past 15 months. The United States has also withdrawn from the World Health Organization, eliminating capacities that would activate on day one of a novel emergence.

The Trump administration has been making cuts over the past 15 months, not because health threats have diminished but because of its desire to undo the work of the Biden administration. Our nation has built satellites and other surveillance systems because leaders decided that threat was worth watching. Biological threats deserve the same logic.

The CDC’s National Wastewater Surveillance System covers 145 million Americans weekly — but it is geographically incomplete, concentrated in urban counties, and designed around detecting a small number of well-known pathogens. Congress should expand it into other areas and support expanding its capabilities. A single outbreak response runs into the billions of dollars, as it did for outbreaks of Ebola and H1N1. The surveillance system costs a fraction of that.

Beyond wastewater, a true early warning system pairs environmental sampling with metagenomic sequencing capable of detecting pathogens before health officials know what to look for. What has been missing is the political will. This technology is here and ready for prime time. If it were in use today, health officials would know if the Andes strain was in the United States or not. Instead, we will need to wait for weeks to see if additional people test positive.

Hantavirus probably won’t be a pandemic, but that’s not the point

This outbreak will almost surely not become a pandemic. But let’s be honest about what we mean by near-miss: The way that pathogen spreads saved us. The system didn’t.

The Hondius timeline is identical to what we would face if a novel pathogen boarded the Hondius ship. Or arrived at the same airport where one of the ship’s passengers, who later died, arrived. Or emerged from the same rural reservoir. We keep watching outbreaks through the rear view mirror, reconstructing what happened after impact.

Three people are dead. Nine cases reported. Seven confirmed infections and two probable cases. A ship that was turned away from port after port has now docked in Tenerife — and dozens of passengers are already home, having passed through airports before anyone knew what they carried. The FIFA World Cup arrives on American soil in weeks. The Olympics follow. International travel has fully recovered, if not exceeded, pre-pandemic levels. This is the permanent baseline — a world of high-volume global mixing that our biosurveillance system, which is being actively dismantled, was not built to monitor.

The gap hantavirus is exposing is not about hantavirus. It is about every pathogen that hasn’t come out in the open, moving through the same blind spots, with no guarantee the next one’s biology will bail us out.

The technology to address this exists. Our leaders are choosing to ignore it.

I spoke to PBS NewsHour’s William Brangham on Monday evening on what you need to know.

Please feel free to weigh in.

American health care has a workforce problem — and it is driving up costs.

When nurse practitioners cannot practice independently, patients who need a routine primary care visit end up paying for more expensive care. When prescription renewals require a physician visit that an AI tool could handle, the system is more expensive, for no clinical benefit. The United States has spent decades restricting the supply of health care providers in the name of quality, and the result has been higher prices, worse distribution, and a workforce that isn’t organized around the needs of patients.

Two changes would go a long way toward fixing that: rethinking who is allowed to do what, and expanding the supply of providers at every level.

Subscribe now

Rethinking who is allowed to do what

American health care has largely refused to ask a key question: What level of training does a specific task actually require?

Managing a stable patient’s diabetes medications does not require the same skills as diagnosing a rare autoimmune condition. Performing a routine colonoscopy does not require the same training as performing cardiac surgery. A system that insists otherwise is wasteful.

The evidence is unambiguous. Nurse practitioners deliver care comparable in quality to physicians for many common conditions such as chronic disease management. Pharmacists, with years of post-graduate training focused precisely on medications already manage medication therapy, order labs, and treat common infections in many states — often through cooperative agreements with physicians. Each tier of this workforce is substantially cheaper than the one above — primary care physicians earn around $287,000 annually, nurse practitioners and pharmacists roughly half that — potentially generating substantial reductions in the cost of delivering care. Meanwhile, many of the tasks a registered nurse does can be easily managed by licensed practical nurses, medical assistants, and trained aides. Allowing registered nurses to focus on assessment, judgment, and complex care, while support roles absorb routine tasks, would reduce costs and make better use of a workforce in short supply.

Twenty-three states still impose physician supervision requirements on nurse practitioners, with no basis in evidence. Congress should establish a national minimum standard giving nurse practitioners the legal authority to diagnose, treat, and prescribe for a specific set of defined conditions without physician oversight — in every state, not just the ones that have already acted.

Some other countries take a more practical approach: The Netherlands ran a five-year national experiment with nurse practitioners and physician assistants independently performing endoscopies and minor surgical procedures, evaluated with input from the Dutch Royal Medical Association, and made independent nurse practitioner and physician assistant practice permanent law in 2018. The United Kingdom gave nurses full prescribing authority nearly two decades ago. Thirteen European countries now have nurse prescribing laws.

The United States should also identify technically straightforward and high-volume procedures, train non-physicians on validated surgical simulators for those procedures, and allow those who meet objective benchmarks to perform them, measuring outcomes to ensure the care is both high quality and safe. AI can now assess surgical skill from video footage with accuracy, making objective competency assessment scalable in ways that weren’t possible before. The question of which procedures can be safely delegated to well-trained, simulation-validated non-physicians is an empirical one and can be answered with high-quality studies, rather than assuming the answer is none. Aravind Eye Hospital in India built one of the world’s highest-volume, highest-quality cataract services program on a single principle: No one should do a job that someone with less training could do just as well. That principle is unremarkable in so many industries — but not in health care.

Technology has opened up new opportunities here. The average physician spends less than half of a 58-hour workweek with patients. The rest disappears into documentation, prior authorization, and administrative tasks. AI scribes, software that listens to patient visits and automatically generates clinical notes, are reclaiming that time in current real-world deployments. AI-assisted clinical decision support tools have reduced diagnostic errors by 16 percent in primary care settings. In January, Utah became the first state to authorize an AI platform to autonomously renew routine chronic medications — at $4 per renewal, with rigorous safety monitoring. Prescription renewals constitute roughly 80 percent of all medication activity. The cost savings implications are substantial.

But realizing those savings requires getting the payment environment right. Under fee-for-service payment, if AI makes a service more efficient but reimbursement stays the same, providers profit, while the system saves nothing. If reimbursement falls to reflect the efficiency, the incentive to adopt disappears. The only model that resolves this dilemma is value-based care, in which health systems accountable for total cost of care capture every efficiency gain as dollars saved: the AI prescription renewal or the nurse practitioner managing chronic disease follow-ups at half the physician cost. With value-based care, these reforms could lower costs for society.

Expanding the supply of providers

The United States faces a shortage of up to 86,000 physicians by 2036, with an aging population placing growing demands on a workforce already stretched thin. We need more providers at every level. Congress capped Medicare-funded residency slots in 1997, with only modest changes since. In 2025, nearly 47,000 medical school graduates applied for fewer than 38,000 positions, preventing qualified medical school graduates from completing their training. The bipartisan Resident Physician Shortage Reduction Act would add 14,000 slots over seven years. Congress has failed to pass it for years — it should finally act.

For nurses, the bottleneck is school capacity. In 2024-2025 alone, more than 80,000 qualified nursing school applicants were turned away due to limited faculty and clinical placement spots. Investment in faculty, simulation labs, and clinical placements — accelerated through health system partnerships with nursing schools — would expand the nursing pipeline.

The physician and nursing shortages are largely a consequence of bad policies. The tools to fix this problem exist. The rest is a choice.

Please feel free to weigh in.

Health insurance in America is supposed to protect people from financial catastrophe. Too often, it doesn’t. Nearly four in 10 insured Americans still report difficulty with health care costs — struggling to pay medical bills, taking on debt, or depleting savings when they get sick. That feeling of being unprotected is real, and it is widespread.

But there is a second failure, less visible and just as damaging: The design of American insurance actively drives up the cost of care. High deductibles don’t just expose patients to financial risk — they reshape how providers price services, how insurers negotiate with health care providers, and how the whole market functions. Without fixing insurance design, Americans will keep paying more and more for a system that protects them less and less.

Subscribe now

The standard response from policy makers is to demand better coverage: lower deductibles, stricter caps on out-of-pocket spending, more generous federal subsidies. Those instincts are right. But they don’t address why insurance has stopped working in the first place. There is a market-based critique of American insurance that deserves to be taken seriously. Insurance is designed to protect against unpredictable, high-cost events — the serious illness, the major surgery, the hospitalization no one can plan for. That is what car insurance does. We don’t run oil changes through car insurance, and there is a reason for that: Insuring small, routine, predictable expenses adds administrative overhead, distorts pricing, and ultimately makes everyone pay more. The argument that that kind of mistake has been made in health care — routing everything through the insurance apparatus, triggering claims processing, prior authorization, and utilization review for a $150 doctor visit — is not wrong.

In Germany, insurance also covers routine care — but costs are far lower than in America. The difference is that German routine care is cheap, and processing a German claim costs a fraction of what it costs here. The problem is not that routine care in America is insured. It is that it is insured at American prices, with American administrative complexity, and without the price discipline that makes the German model work.

That distinction points toward the real solution.

Until those underlying problems are fixed — and the earlier pieces in this series lay out how — there is a strong case for experimenting with routing more routine care outside of insurance. Direct-pay prices for common services such as primary care, routine labs and X-rays, or even minor surgeries are often substantially lower than in-network rates. At minimum, any direct payment for medically appropriate, physician-directed care for covered services should count toward a patient’s deductible — rewarding patients who find lower-cost options rather than penalizing them.

But that only holds if lower-income patients have a real way to pay for routine care. Here it is worth being honest about the status quo: High deductibles already deter lower-income Americans from routine care, and underfunded Health Savings Accounts offer little protection. The system is failing them now. Health Savings Accounts, reformed properly — made more generous, structured so that lower-income families actually have money in their accounts — combined with the lower prices that a thriving direct-pay market would produce, means these patients could be genuinely better off than they are today. Done right, expanding direct payment for routine care would put downward pressure on prices for everyone, insured and uninsured alike. The goal would be to lower the burden of costs, not just shift it to others.

For the complex, expensive, unpredictable care that insurance is actually designed for, there needs to be smarter design of insurance coverage. Today, patients typically pay the same regardless of where they get care — no reward for choosing a lower-cost facility, no signal that one provider charges three times another for identical quality. Two approaches could fix that.

Reference pricing sets a benchmark payment for a common service — say, $800 for an MRI. Patients who choose a $600 facility save money; those who prefer a $2,000 facility pay the difference. Research on California’s public employees showed large savings from this approach. Tiered networks work on the same principle but apply it more broadly: Insurers group physicians and hospitals into tiers based on the value they deliver — comparable quality at lower cost — and patients pay lower copays and deductibles when they choose higher-value providers.

For example, a patient choosing a high-quality, low-cost hospital (top tier) for a hip replacement might pay a $500 deductible, while the same procedure might have a $1,500 deductible at a place that charges more. The insurer is no longer neutral about where patients go — it is actively steering them toward better value and comparable quality. Both designs do something the current system almost never does: They present both quality and cost data to patients and they reward choosing wisely.

Making these approaches work requires two things the current system lacks. First, genuine price and quality transparency before the point of care — not just data posted somewhere online, but clear information at the moment a patient is choosing where to go. Second, protections against surprise balance billing, so patients who choose a lower-cost option don’t end up blindsided by a bill they didn’t anticipate.

None of this requires dismantling private insurance or imposing a single national model. It requires clarity about what insurance is for, seriousness about funding access to routine care for those who need it, and willingness to design the complex end of the system to actually reward better care at lower cost.

In earlier columns, I argued that we can lower health care costs by tackling excessive prices, restoring competition, improving payment models, and cutting administrative waste. But none of those reforms would fully succeed if the design of insurance continues to suppress competition, deter needed care, and leave people financially exposed when they get sick.

I’m a doctor. Here’s what opened my mind about the future of medical care.

For this piece I’m pleased to be joined by my colleague Dr. Irene Papanicolas, the Provost’s Professor of Health Services, Policy, and Practice and Director of the Center for Health System Sustainability at the Brown School of Public Health.

Please feel free to weigh in.

A person who has been on a blood thinner for years for a heart condition recently switched jobs and found his blood thinner was not covered by his new insurance. To get him the medication he needs, his doctor and her office spent hours filling out forms for a process called prior authorization, in which the physician has to make the case for why the treatment is necessary. The requests were initially denied but eventually approved. During the weeks it took to resolve, he never missed a dose because he was fortunate enough to have the resources to pay out of pocket, about $300 to $400 a month. His doctor estimated that the staff time spent on the prior authorization process cost her practice as much as the drug itself.

This story is not unusual. The experience most doctors have every day is not primarily about medicine. It is about paperwork.

Subscribe now

America spends up to $1 trillion a year on health care administration — billing, coding, prior authorizations, insurance negotiations, quality reporting, credentialing, and so forth. The instinctive defense of this reality is that a complex, multi-payer system requires complex administration. But is that true? Germany, the Netherlands, Switzerland, and Australia all have dozens of private insurers operating within their systems. These are not single-payer systems — patients have real choice, and insurers compete. But in these countries administrative costs are roughly half of those in the United States. In Germany, they constitute 4.3 percent of total health expenditures. In the Netherlands, it is 3.1 percent, and in Australia, 3 percent. In the United States, it is 7.8 percent.

And those figures count only insurer overhead. When adding the burden that fragmented insurance imposes on hospitals and practices, the total administrative spending approaches one dollar in four of health care spending in the United States. That is more than the United States spends on all of primary care. More than the United States spends on heart disease.

What separates other countries with multiple insurance models from the United States is how their insurers operate. In Germany, all insurers use the same billing system and coding rules, and a physician submits one type of claim regardless of which insurer the patient has. In the Netherlands, benefit designs are standardized nationally.

In the United States, every insurer does things differently — different forms, different billing codes, different prior authorization rules, different credentialing requirements. Physicians are often directly involved in this paperwork, and even when tasks are delegated, they must review, document, and justify care decisions. Every hospital and physician’s office has to hire staff to navigate all of them simultaneously. The result is a shadow workforce that exists purely to manage the paperwork, which increases the cost to the taxpayer and patients needing care. In Massachusetts, there are nearly as many health care administrative managers as there are physicians.

There are three key ways to reduce the administrative burden in the US health care system.

Fix prior authorization

Prior authorization exists, in theory, to prevent unnecessary care. In practice it has become one of the most despised features of American health care for patients and physicians alike. The delays are real and sometimes dangerous. But the deeper problem is the variation. A doctor prescribing a common medication fills out one form for one insurer, a different form for another, waits three days here and two weeks there. The American Medical Association found that the average physician practice completes 39 prior authorization requests per physician every week and spends 13 hours doing it. That is 13 hours of physician and staff time spent seeking permission to practice medicine rather than practicing it.

Instead, federal and state regulators should standardize the forms, mandate common timelines, and require automatic approval for routine, evidence-based treatments where clinical guidelines are clear. Then they should require that insurers operate under common rules about what requires authorization in the first place. Congress has been inching toward this for years but has not finished the job.

Build the health care equivalent of the banking clearinghouse

When you swipe a credit card, the transaction clears in seconds because the industry agreed on common standards decades ago. Health care claims work nothing like this. More than 9 billion insurance claims are filed every year; each costs between $12 and $19 to process and takes four to six weeks to pay, because every insurer runs a different system with different forms, rules, and workflows — adding both cost and delay. The solution is straightforward: a universal clearinghouse for submitting, processing, and paying claims. Providers submit once. Payment moves in days. David Cutler, a Harvard economist and an architect of this idea, estimates the savings at tens of billions of dollars annually — not by cutting care but by cutting the friction of paying for it, with benefits shared across providers, insurers, and ultimately patients and taxpayers.

Standardize the rules

The deepest fix would be to require all insurers to operate under a common set of rules. This includes standardized billing codes, a common credentialing system so physicians don’t re-credential separately with every insurer, and a single set of quality measures so providers aren’t reporting different measures to each payer. Physician practices currently spend more than $15 billion a year on measuring and reporting data on quality — roughly $40,000 per physician — for measures that fewer than a third of physicians believe actually reflect the quality of their care.

Administrative roles are deeply embedded across the health system, and reducing them is not costless. But we have built an enormous workforce devoted to managing complexity rather than delivering care. Reducing that complexity would free up both human and financial resources.

The bottom line

Reforms like these could save $200 billion or more annually. These savings would not require cutting benefits, rationing care, or reducing what doctors get paid but would rather be achieved by reducing what everyone pays for the paperwork. Physicians in Germany, the Netherlands, and Switzerland face far less administrative burden, with administrative costs that are half of ours, and report fewer demands that take them away from patient care. Their patients spend less time fighting with insurers. It’s time the United States requires private insurers to operate more rationally as well.

What happened to the person who paid for his medication out of pocket is what the system produces when every insurer makes its own rules. He got his medication because he could afford to pay out of pocket. Millions of Americans can’t — and for them, missing doses of blood thinners could lead to life-altering strokes or worse. The United States built this current system. It can build a better one.

The war with Iran has focused the world’s attention on nuclear weapons and oil. Those risks are real. But there is a third category of danger that the conflict has brought into sharp relief — one that the United States is far less prepared to handle than most people realize.

Iran likely has a biological weapons capability. This has been the view of the U.S. government for several years now, predating the second Trump administration.

The U.S. State Department’s most recent compliance report concludes that Iran “maintains flexibility to use, upon leadership demand, legitimate research underway … to produce lethal biological weapons agents.” The Office of the Director of National Intelligence similarly assesses that Iran “very likely aims to continue R&D of chemical and biological agents for offensive purposes.” These are unclassified assessments, sitting in government reports that too few people have read.

War makes an already serious problem harder. When command structures fracture, when facilities are struck and personnel scatter, the custody of dangerous materials unravels. Dangerous materials don’t secure themselves.

We have seen this before. After the Soviet Union collapsed, the United States spent years — and billions of dollars through the Nunn-Lugar program — scrambling to secure biological, chemical, and nuclear stockpiles that suddenly had no reliable custodian. That program worked because the post-Soviet states were willing partners.

Iran is neither stable nor a willing partner. Its biological capabilities sit within a network of military-affiliated universities, IRGC-linked research centers, and dual-use pharmaceutical and biotechnology institutions. Some of those facilities have already been targeted in strikes. Others may be. We do not have a reliable inventory of what biological materials were present at targeted sites, what was destroyed versus dispersed, or what risks the surrounding areas now face.

The Congressional Research Service has warned that ongoing military operations “may complicate onsite damage assessments and efforts to secure CBRN” — chemical, biological, radiological, and nuclear — materials.

In plain English: The U.S. may not know what it’s hit, what is now unguarded, or where it might go.

Biological agents pose a particular challenge. Unlike nuclear material, they don’t trigger radiation detectors at border crossings. Unlike chemical stockpiles, they can be small, portable, and capable of spreading on their own once released.

A vial doesn’t need a missile to become a weapon.

The consequences of a release — deliberate or accidental — would be unlike anything disrupted oil markets can produce. Covid-19 — which was not a weaponized pathogen and not optimized for lethality — erased tens of trillions of dollars from the global economy and reshaped daily life for years.

Here’s what a biological release in this conflict might actually look like: A strike damages a storage or research facility and materials are dispersed into the surrounding environment — an accident that no one intended nor immediately recognizes.

Or the chaos of a collapsing command structure creates an opportunity for a small quantity of a dangerous agent to fall in the hands of someone with motive and access. Even the government, facing existential military pressure, might calculate that asymmetric biological retaliation carries less risk than conventional defeat.

These are the scenarios that have worried experts for years. A deliberately engineered pathogen, released under any of these circumstances, could be far worse than anything disrupted oil markets can produce. Disrupted oil supplies are painful and recoverable. A biological event is not.

Which is why the most urgent question raised by this conflict is less about Iran than it is about us. The war in Iran has exposed a gap in American preparedness that predates this conflict — and will outlast it.

The United States does not have the biosurveillance infrastructure to detect a novel or engineered pathogen release quickly enough to contain it. The systems we rely on — syndromic surveillance, emergency department reporting, wastewater — are built to recognize diseases we already know. An engineered agent, or one we have never encountered, will likely be missed. It may circulate for days or weeks before the signal is strong enough to register. By the time it is showing up in emergency rooms, it has likely already seeded transmission chains across multiple cities. The window for containment isn’t just smaller — it may already be closed.

Beyond surveillance, our Strategic National Stockpile has documented gaps in the very countermeasures we would need — medical countermeasures against threat agents that have been on the U.S. concern list for decades. The research and manufacturing infrastructure to close those gaps quickly doesn’t currently exist at the scale this moment requires, and the workforce to deploy them has shrunk.

Congress should demand a full accounting of what we know about Iranian biological materials — where they are, and what has happened to them since strikes began. More importantly, Congress needs to treat biological preparedness as the national security priority it is: not as a subcategory of pandemic response, not as a line item that gets cut when budgets tighten, but as a core defense investment.

The world has spent 70 years building treaties, monitoring systems, and institutions to manage nuclear risk. For biological threats, we have a fraction of that infrastructure — and the current conflict is exposing exactly what that deficit costs.

A missile launch is visible. A vial leaving an unsecured facility is not.

For this piece I’m pleased to be joined by my colleague Dr. Tom Tsai, associate professor of health policy and management at Harvard T.H. Chan School of Public Health and co-director of the Healthcare Quality and Outcomes Lab at Harvard School of Public Health.

Please feel free to weigh in.

When a colleague, a primary care physician in North Carolina, joined a new kind of medical practice a few years ago, she wasn’t looking to reinvent how she practiced medicine. But what she found surprised her. Her previous practice looked like most primary care in America: a treadmill of 15-minute appointments, minimal support staff, and patients who fell through the cracks between visits. Today her patients have care coordinators who call them between appointments and catch problems before they become crises. Her highest-risk patients — the ones who used to cycle in and out of the emergency department — are healthier and are staying out of the hospital.

Her experience points to a promising idea that has been at the center of health care reform for more than a decade: Instead of paying doctors and hospitals for every test and procedure they perform, pay them for keeping patients healthy. In this model, called value-based care, doctors and hospitals are paid based on the health outcomes they achieve and the overall cost of caring for their patients, rather than simply for the volume of tests and procedures they order. Do that, the theory goes, and you align everyone’s incentives — doctors, hospitals, insurers, and patients — around what actually matters.

The idea is sound. The results, so far, have fallen short. The Congressional Budget Office found that the federal government’s major payment-reform programs — a central tool for implementing value-based care — cost $5.4 billion more than they saved between 2011 and 2020. These value-based care programs have not yet bent the cost curve.

But buried in the data is an encouraging sign: The programs that have actually worked share a common feature. They are often physician-led, and they hold doctors genuinely accountable for the total cost and quality of their patients’ care. When primary care teams are given both the tools and the real financial responsibility to manage health, they deliver better results — for patients and for the system.

Why hasn’t that spread more broadly? There are two main reasons.

Subscribe now

First, the doctors doing the work often see little of the savings they generate. Consulting companies and management firms that run these programs can capture 40 cents of every dollar saved — money that should be going to front-line care. Second, these programs often don’t cover the sickest and most vulnerable patients who need coordinated care the most.

Despite these limitations, accountable care can work. In 2022, Medicare’s flagship program resulted in savings of nearly $2 billion, with the biggest gains coming consistently from physician-led, primary care-centered organizations.

Still, given the comparatively small savings of value-based care, the temptation is to abandon this effort. That would be a mistake. Instead, health care officials need to get serious about making these programs better.

It begins by raising the stakes. Most value-based programs have been structured so that providers — doctors or hospitals — receive a bonus if they save money but face no penalty if costs run over. Programs with no downside risk are programs without real accountability. And without real accountability, the incentive to fundamentally change how care is delivered simply isn’t there. The Trump administration has taken a meaningful step here, eliminating these half-measures and launching new programs with longer contracts — up to 10 years — that give practices the time to invest in the kind of care coordination our colleague mentioned above has seen in her practice. Research shows it works: Programs with longer track records and stronger accountability have consistently delivered bigger savings.

It also means having more specialists lead many value-based care programs. Focusing on primary care is important, but it alone cannot bend the cost curve. Cardiologists, orthopedic surgeons, oncologists, radiologists — these and other specialists account for the majority of what Medicare spends, and for too long they have operated entirely outside any accountability for cost or outcomes.

New programs from the Centers for Medicare & Medicaid Services are beginning to change that, requiring specialists in areas like heart failure and chronic back pain to share in the financial consequences of their decisions, reducing payment to those who exceed spending targets. That’s the right direction.

The next step is extending it much further — where payers like CMS or private insurers set a target payment for a whole episode of care (like a joint replacement) so that every provider (in this case, the surgeon, the hospital, etc.) has an incentive to manage costs efficiently. Experts have estimated that bringing specialists into this accountability framework could save as much as $100 billion a year. That potential has barely been tapped.

Longer value-based care contracts between payers and providers also allow something that short-term thinking punishes: investing in patients’ long-term health. A doctor who prescribes an effective obesity medication today — one that could prevent diabetes, heart disease, and joint problems years from now — shouldn’t be penalized because the savings show up after the contract ends. Smarter payment design would reward exactly that kind of forward-looking care.

The ultimate goal is accountability — with hospitals, specialists, and primary care doctors all responsible for the total cost of a patient’s care while paying attention to good health outcomes. Maryland has been running an experiment along these lines for a decade, and sets a budget for how much money can be spent overall by hospitals in the upcoming year. The results show that even hospital systems can operate under spending constraints without sacrificing the quality of care. Cost discipline and good medicine are not in conflict. When every player is accountable, the incentives align much better.

None of this is easy. Moving to longer contracts and greater accountability will create anxiety among providers, particularly smaller practices and safety-net organizations serving low-income communities. Strong risk adjustment and guardrails against excessive volatility are essential.

But the status quo is not an option. Nearly one in five dollars is spent on health care. We cannot afford reforms that produce only incremental gains when Medicare’s trust fund may be depleted over the next 10 years.

Our colleague’s patients are healthier, and they’re staying out of the hospital. That’s not just better for them. It’s better for all of us. The question is whether we’re willing to build a system where far more patients can say the same.

For this piece I’m pleased to be joined by my colleague Dr. Irene Papanicolas, the Provost’s Professor of Health Services, Policy, and Practice and Director of the Center for Health System Sustainability at the Brown School of Public Health.

Please feel free to weigh in.

A patient undergoes a routine inpatient surgery at one hospital, and the bill comes back at $22,000. The same procedure, across town, costs half that. The clinical care is identical, but the difference isn’t quality. It’s pricing power.

In examples like this across the United States, too many health care markets aren’t working. Markets are supposed to discipline prices. In a functioning, competitive market, a health care provider shouldn’t be able to charge twice as much for the same service and get customers. But in many parts of American medicine, markets fail to tame excessive prices.

For decades, American health policy has rested on a simple assumption: If there is enough competition, prices will fall. In the last two pieces of this series, we’ve laid out how regulators and policy makers have failed to foster adequate competition in some communities and what they can do to bring it back. But beyond making markets work, there are other options, especially when policy makers find that breaking up large systems which are behaving badly may not be tenable. When prices stop reflecting value and instead reflect leverage, there is another approach that much of the world uses: price regulation.

Americans often recoil at the idea of price regulation, seeing it as heavy-handed or anti-market. But health care prices are already regulated. Medicare sets payment rates for seniors. Medicaid does the same for lower-income Americans. Several states, such as Vermont and Indiana, recently enacted legislation to cap or freeze hospital prices for commercially insured patients. The real question isn’t whether prices can be regulated — they clearly can. It’s why we continue to allow American families in noncompetitive markets to be subjected to a system in which prices reflect market power, not value..

Americans don’t use health care more than people in other wealthy countries. But they pay more when they use it. These high prices show up in different ways, from higher premiums for consumers to slower wage growth when employers put more money into health care and less into salaries, to growing pressure on state budgets and small businesses.

In most private contracts, prices are substantially higher than what Medicare pays providers. Health systems argue that they need these higher prices to compensate for underpayment from Medicaid or the uninsured. While there may be some truth to that argument, the real question is: How much is too much? In instances where markets are not functioning efficiently due to a lack of competition, what kind of prices should policy makers tolerate?

Recent analyses from RAND reveal that hospitals are paid, on average, roughly 2.5 times higher than what Medicare pays for the same services. But in some states, that average exceeds three times higher. And prices that are three times higher than Medicare rates are not rare; in parts of the country, they are the norm.

That reality suggests a straightforward, pragmatic reform. States could immediately cap commercial prices for all providers at 300 percent of Medicare, likely eliminating the most extreme excesses. The ceiling then could be gradually brought down over time (say, to 200 percent of Medicare over five years). Medicare rates were designed to sustain access and should be enough for a large number of providers to survive; twice that level should allow many well-run health systems to not merely survive, but thrive.

If this sounds radical, it really isn’t. In fact, the United States is the outlier when it comes to price regulation of private insurance markets. Most countries that rely on private insurance use price guardrails to make competition possible.

In Germany, multiple insurers compete, physicians operate independently, and patients have broad choice. But prices are shaped through negotiated national fee schedules. Hospitals are paid under structured frameworks rather than unconstrained local bargaining. Competition happens on quality and efficiency, and providers are prevented from exploiting market breakdowns to charge higher prices.

Switzerland, one of Europe’s most market-oriented systems, also relies heavily on private insurers and consumer choice. Yet prices are negotiated nationally with government intervention when negotiations fail. Not every service is priced the same way, but all operate within clear rules that limit how far prices can vary. Markets exist — but within boundaries. Other countries like the Netherlands also have private insurers operating under strong public oversight.

None of these systems is perfect. Each comes with trade-offs — controlling prices can mean less flexibility, more bureaucracy, or sometimes less innovation for truly groundbreaking treatments or services. But all begin from the same premise: Health care prices cannot be left entirely to local bargaining leverage in areas where there is inadequate competition to discipline prices. These international examples also underscore an important lesson: Regulating provider prices alone is not enough. In markets where insurers are concentrated, lower payments to providers won’t automatically translate into lower premiums for consumers because the dominant insurer can just pocket the savings. That is why many countries pair provider price limits with oversight of insurers, including insurance rate regulations.

Critics will argue that tying price caps to Medicare would destabilize health systems or threaten innovation. Those concerns deserve to be taken seriously. But a ceiling at 200 percent of Medicare prices still provides a generous margin above a payment system designed to ensure access nationwide.

Thousands of providers already deliver excellent care near those levels today. Some states are already moving in this direction, such as Vermont and Indiana, adopting price caps around this benchmark. More states, including Massachusetts, should do so to control costs. Failure to effectively run a financially sustainable system at twice Medicare rates suggests there are likely inefficiencies that can be rooted out.

Importantly, this approach would not require Washington to micromanage every medical service. States can adapt benchmarks to local conditions and phase them in gradually. The goal is not command-and-control pricing. It is establishing guardrails when markets fail to function effectively. When competition works, the government should step back. But when markets fail, refusing to act is a decision to let costs keep rising.

Markets need rules. And when markets fail to protect patients, policy should.

For this piece I’m pleased to be joined by my colleague Dr. Tom Tsai, associate professor of health policy and management at Harvard T.H. Chan School of Public Health and co-director of the Healthcare Quality and Outcomes Lab at Harvard School of Public Health.

Please feel free to weigh in.

In my last op-ed, I argued that in many communities, consolidation of health care providers has driven up prices by weakening competition. But preventing consolidation that does not provide benefits, or even breaking up consolidated systems that behave badly, is likely not enough. If we want health care markets to work — if we want prices to fall without sacrificing quality — the policies that shield established health care systems and stymie innovation must be removed.

For decades, federal and state rules have made it easier for large health systems to expand than for new competitors to emerge. Payment policies often reward consolidation through complex regulatory and financial structures that make it difficult for independent providers to thrive. Licensing laws can slow or prevent new facilities from opening. The Affordable Care Act banned an entire category of physician-owned hospitals from being created or expanded. Many of these policies were adopted with good intentions, but in today’s highly concentrated markets, they often protect dominant systems from competition.

The path forward will need to restore the patient-doctor relationship and allow delivery models that place doctors and patients back at the center of medical decision-making. If policy makers are serious about affordability, we must lower barriers to physician-led, patient-centered health care. Four reforms deserve immediate attention.

Adopt site-neutral payments

Today, Medicare — and often private insurers — pay dramatically different rates for the exact same service depending on where it is delivered. A routine cystoscopy, a simple urologic procedure in a physician’s office, is reimbursed at roughly $204 under Medicare. The same procedure performed in a hospital-owned outpatient clinic costs more than $540 — more than 2.5 times as much — even when the patient and procedure are identical.

These payment differentials were originally justified by modest differences in overhead and patient complexity. Over time, some hospitals took this as an incentive to acquire independent practices and reclassify routine care as hospital-based services. The result is predictable: The care doesn’t change but prices rise.

The Medicare Payment Advisory Commission has repeatedly recommended aligning payments across settings to eliminate these distortions. Site-neutral payments would dampen the incentive for private equity or hospital acquisitions that are not beneficial to patients and encourage independent practices and new competitors. When identical care is delivered, it should be paid the same — regardless of the logo on the building.

Repeal certificate-of-need laws

Certificate-of-need laws function to protect incumbent providers from competition. Created in the 1960s to prevent duplicative services, these laws now function primarily as barriers to entry. In states that retain them, new hospitals, expansions, or even the purchase of major medical equipment require lengthy regulatory approval — a process existing health systems can challenge, delay, or block.

Decades of research show that CON laws are associated with fewer hospitals, fewer ambulatory surgery centers, and in many instances higher prices. The Federal Trade Commission and the Department of Justice have repeatedly urged states to reconsider these laws because they suppress competition rather than promote it.

Markets cannot function efficiently if new competitors are prevented from entering them due to outdated regulations. States that are serious about affordability should eliminate CON laws and similar entry barriers.

Unlock the potential of ambulatory surgery centers

Ambulatory surgery centers are an excellent example of what is lost when entry barriers protect established health care systems. A rapidly growing share of surgical care can be delivered safely in ambulatory and and office-based settings at substantially lower cost (and far greater convenience for patients) than in hospital operating rooms. National studies consistently find that procedures performed in such centers cost 20 to 50 percent less than the same procedures in hospital outpatient departments, with comparable clinical outcomes.

Yet in many states — including Massachusetts — independent ambulatory surgical centers face significant regulatory hurdles, often through certificate-of-need programs that can substantially delay new entrants. As a result, Massachusetts has among the fewest ambulatory surgery centers per capita in the country, and many procedures that could be delivered safely in lower-cost settings are instead performed in higher-priced hospital facilities.

Expanding access to ambulatory surgery centers would not eliminate the need for full-service hospitals. It would simply allow routine surgical care to migrate to more efficient settings when clinically appropriate — lowering prices while preserving quality.

Allow physician-owned hospitals to compete again

The Affordable Care Act effectively froze the creation and expansion of most physician-owned hospitals, citing concerns that they would selectively treat healthier or more profitable patients. Subsequent research has largely failed to find this effect, and several studies suggest Medicare spending may be lower at certain physician-owned facilities compared with traditional hospitals.

What physician-owned hospitals clearly represent is an alternative ownership model — often smaller, more specialized, and more directly accountable to the clinicians delivering care. In some highly concentrated markets dominated by large health systems, prohibiting new physician-owned hospitals removes a potentially important source of competition.

Obviously not every physician-owned hospital is superior, and safeguards are necessary to prevent cherry-picking of patients. But the empirical evidence doesn’t support blanket restrictions against these types of hospitals — and if the goal is affordability and access while preserving quality, allowing physician-owned hospitals to enter and grow in markets would be good.

If Congress and state leaders adopted site-neutral payments, repealed certificate-of-need laws, reopened the door to physician-owned hospitals, and allowed independent ambulatory surgery centers to flourish, these changes would go a long way to improving competition in many health care markets. Prices would become more affordable. Quality would likely improve. Patients would have more choices. And care would be more centered in places patients and physicians find convenient. If state and federal policy makers did all this, the system would finally begin to work for patients.

Why is American health care so expensive? Because prices are too high.

Contrary to popular belief, Americans do not see doctors more often or stay in hospitals much longer than people in other wealthy countries. We do not undergo far more procedures or use vastly more care. Yet for the same office visits, the same surgeries, the same medicines, and the same scans, we routinely pay far higher prices — often two to three times as much. These prices, not excessive use, are the primary driver of America’s health care spending crisis.

So why are prices so high? In many communities, the answer is straightforward: There is no longer meaningful competition in health care markets.

Across the country, many health care markets have consolidated into a handful of dominant firms. In Massachusetts, two systems — Mass General Brigham and Beth Israel Lahey Health — control a large proportion of inpatient beds, and the trend extends far beyond hospitals. Insurers are consolidating into a small number of dominant firms. Private equity firms are acquiring physician practices, and large corporations are housing insurers, pharmacies, and clinics under one roof. Both vertical (combining entities that provide different types of services such as insurers and hospitals) and horizontal (combining organizations with similar services) consolidation has become a defining feature of the US health care system. While these arrangements can theoretically deliver efficiency, they often lead to fewer choices and higher prices — costs that ultimately fall on patients.

As competition erodes and prices rise, those costs are passed on to patients through higher premiums and out-of-pocket spending. In 2025, the typical American family of four was projected to pay close to $35,000 for health care — often reflecting the high price of care in markets with not enough competition.

To be clear, not every merger is harmful. In some cases, consolidation can lead to better care coordination, fewer unnecessary services, and lower administrative costs. But in other cases, mergers that promise “integration” end up delivering higher prices, fewer choices for patients, and potentially even lower quality of care. If health systems seek the benefits of consolidation, these health systems should be able to demonstrate with clear evidence that the merger will actually improve quality or reduce costs.

This situation reflects a series of policy choices that allowed health care markets to consolidate with little resistance. For years, antitrust enforcement in health care was weak, and regulators often treated nonprofit hospitals as “mission-driven” institutions exempt from normal market scrutiny. Policy makers, believing that larger systems would deliver better care at lower cost, encouraged consolidation. Too often, they were wrong — and prices rose instead.

Insurance markets followed a similar path. Lax oversight allowed insurers to grow larger without becoming better or more innovative, making it harder for new plans or alternative insurance designs to compete, ultimately leaving consumers with fewer choices. Some have argued that large insurers are needed to counterbalance large health systems. But that is not always the case. Two monopolies do not automatically create competition; they can instead entrench market positions.

If policy makers are serious about controlling health care spending, confronting high prices that are driven by a lack of competition must be a priority.

First, prior to major health care merger approvals, regulators should consistently demand comprehensive data analysis on the impact of the merger on prices, quality, and access. Too often these days the approvals are based on general financial projections or vague promises of integration.

Second, antitrust enforcement must go beyond reviewing new deals. In markets that are already highly concentrated, regulators should be willing to challenge past mergers that have demonstrably harmed consumers.

Third, where markets have become effectively monopolistic, policy makers should consider structural remedies, including divestitures or, when necessary, breaking up dominant systems to restore competition. Preventing consolidation would have been preferable, but in many cases that opportunity has passed.

Finally, nonprofit status should not confer special treatment from scrutiny around marketplace competition. Hospitals and insurers that behave like profit-maximizing firms should face a much higher bar for demonstrating community benefit, fair pricing, and transparency. Organizations that no longer meet that standard should not retain special protections.

Defenders of consolidation argue that scale is essential for stability. Some scale can help — but many health systems are well beyond that point. As experts have often noted, economies of scale can quickly turn into bureaucracy that dampen innovation and efficiency. Competition remains the most reliable engine for improving quality and controlling costs, and restoring it in health care is not about ideology. It is about fixing markets when they no longer work, giving people more choices, and easing the unsustainable burden that high health care prices continue to place on American families.

Look up your ZIP code to see how concentrated your local hospital market is here.

Subscribe now

Read the full series

When doctors get the diagnosis wrong, patients pay the price. Unfortunately, that is exactly what has happened with the American health care system: Lawmakers have misdiagnosed the problem, and the American people pay the price.

For millions of working families, the cost of the misdiagnosis is obvious: Premiums and deductibles now consume so much of household income that many families delay care, drain their savings, or make difficult choices between health care, rent, and child care. Small businesses drop coverage they can no longer afford, and states divert money from schools, roads, and public safety just to keep Medicaid afloat. The United States is on track to spend more than $7.7 trillion per year by 2032. That level of spending will further erode coverage, overwhelm public budgets, and force painful tradeoffs with other essential investments.

For decades, policy makers have treated rising health care costs as an unavoidable fact of modern life — the inevitable byproduct of an aging population, rapid medical innovation, and expanded coverage. That convenient story has allowed Washington to focus almost exclusively on subsidies — how big they are and where they come from. It’s like arguing over who should pay the restaurant bill while the price of the meal keeps climbing. But if you misdiagnose the disease, you cannot prescribe the right treatment.

Some of the increase in spending does reflect genuine progress. More Americans are living longer. Breakthrough treatments are saving lives that once would have been lost. But a large share of our cost explosion is the result of deliberate policy choices — choices that tolerate extreme prices, reward wasteful care, permit unproductive consolidation, and embed extraordinary inefficiency into the delivery of care.

The misdiagnosis is not just technical. It is ideological.

On the political right, the reflex is to blame government — and these days, especially the Affordable Care Act — for health care dysfunction. That story is simplistic. US health care spending was on an unsustainable trajectory long before the ACA, and while the law successfully expanded coverage, it did little to bend the cost curve. At best, the ACA slowed spending growth in the Medicare program, but it had limited impact on spending, positive or negative, among the commercially insured.

On the political left, the instinct is to blame “the market” and corporate profiteering. That story is also simplistic. In many American communities, there have been real breakdowns in the market. Patients largely cannot shop because they have too few choices. Prices are opaque. Competition is weak or nonexistent. In too many areas of America, what we call “the market” is actually a set of government-protected monopolies sustained by policy choices that reward consolidation and shield incumbents from competition.

While both sides argue, the status quo wins. Policy makers fail to take action to address why the health care costs are so high in the first place.

When the domestic argument hits a wall, the debate turns abroad. Liberals point to Denmark or the United Kingdom, two largely publicly funded systems with few out-of-pocket costs. Conservatives point to Singapore, which relies largely on health savings accounts and catastrophic coverage with more of a functioning market. The implication is that if we would simply import someone else’s system, our problems would disappear.

My mentor and friend, the late Uwe Reinhardt — one of the world’s great health economists — used to caution against that kind of thinking. During the 2016 presidential campaign, when Bernie Sanders frequently praised the Danish health care system, someone asked Reinhardt what he thought of it.

Defending the Danish system from criticisms by those who referred to it as socialized medicine, Reinhardt is purported to have responded “I will happily take the Danish health system, but you must also give me the Danish political system … and it would surely help if you gave me the Danish people.”

His comment was both witty and deeply insightful. Health care systems are embedded in political institutions, regulatory cultures, and social expectations that cannot simply be copied and pasted. What matters is not importing someone else’s model but fixing our own — using tools that fit American institutions and American values.

The United States can lower health care costs — or at least dramatically slow their growth — without sacrificing quality or access, if policy makers choose to focus on the right levers.

Over the next eight columns in the coming months, I will lay out (sometimes writing with a colleague) a practical reform agenda that could slow health care cost growth by hundreds of billions of dollars a year without cutting benefits, rationing care, or stifling innovation. The proposals target the actual drivers that affect spending: health care market consolidation and pricing, broken payment incentives, administrative waste, unnecessarily high drug costs, workforce restrictions, site-of-care distortions, and insurance design. None of these ideas are radical. All are achievable. Some require federal action; many can be implemented by states. And nearly every one will threaten a powerful political interest — which is precisely why they matter.

The real question facing the country is no longer how to pay for an expensive health care system. It is why we continue to tolerate one that is so unnecessarily expensive — and what we are finally willing to do about it.

It’s easy to feel like the United States is losing ground in public health. Policies and actions from federal health leaders have fractured trust, undermined science, and disrupted essential services. This past year brought funding cuts to core health programs, the spread of dubious science, and the use of food benefits and health care as bargaining chips in political negotiations. The country urgently needs a course correction before the story of public health in America becomes one of steady decline.

But as this tumultuous year draws to a close — and especially during this season of thanksgiving — it’s important to recognize the contributions that deserve gratitude. Some come from science and medicine delivering powerful new tools, others from institutions and individuals stepping up when clarity and leadership were needed, and still others from communities filling the gaps when federal systems fell short.

First, the nation is beginning to make progress in key areas of public health because biomedical science is delivering meaningful breakthroughs on some of our most difficult challenges. These advances are a powerful reminder that well-targeted innovations can drive large improvements in population health.

For the first time in decades, US obesity rates have flattened — and even begun to fall. A smaller share of Americans were obese in 2024 than in 2022, a remarkable shift largely driven by the growing use of GLP-1 medications. Once used mainly for diabetes, these drugs are now more widely available and also reduce cardiovascular risk, lower the likelihood of some cancers, and may even influence addiction and neurodegenerative diseases.

We are also seeing similar momentum in the opioid crisis. Provisional data on overdose deaths for 2024 show a drop to their lowest level since 2019, according to the Centers for Disease Control and Prevention, driven by expanded access to medication-assisted treatment such as naloxone and buprenorphine. Researchers are also developing nonaddictive pain treatments — an important step toward preventing future waves of dependency.

Meanwhile, CRISPR gene editing has given scientists the ability to adjust genes with extraordinary precision. This year, it was used to treat rare, previously untreatable genetic diseases and appears to have corrected a fatal disorder in a 5-month-old boy. Early research is exploring how CRISPR might help treat more common conditions, such as high cholesterol, and how to finance broad access to these emerging therapies so they reach the people who need them most. A similar shift is happening with GLP-1 drugs, and the Trump administration’s push to bring down their prices is an important step.

Second, for decades the federal government served as the cornerstone of health data and recommendations relied on by state and local health departments, clinicians, parents, and the public. Under Health and Human Services Secretary Robert F. Kennedy Jr., that system has become wildly dysfunctional — it’s no longer a reliable source of essential data and only rarely a source of credible guidance. Fortunately, medical societies, state leaders, and academic researchers have begun filling many of the gaps.

From Massachusetts to California, states have built collaboratives to coordinate vaccine recommendations, emergency planning, and disease surveillance — stepping in while the federal government ignores major public health threats and works to undermine vaccine confidence and access. And as federal dashboards for infectious diseases and mental health went dark, academic institutions and medical organizations such as the Vaccine Integrity Project provided transparent, scientifically grounded analysis.

My colleagues at the Brown University School of Public Health launched a weekly virus-tracking report that now reaches more than 10,000 subscribers and is increasingly used by local health departments seeking real-time data. Independent communicators have played a key role as well; through widely read Substacks and social media channels, public health voices like epidemiologist Katelyn Jetelina have kept much of the public informed in clear, accessible ways.

These efforts underscore the resilience of the US public health ecosystem and its ability to adapt — but they are no substitute for a high-functioning, highly trusted CDC.

Third, the federal government just reopened after the longest shutdown in US history — a self-inflicted debacle that left federal workers without paychecks and disrupted services millions rely on. Worse, the Trump administration tried to use the shutdown to cut food assistance through the Supplemental Nutrition Assistance Program.

In response, local institutions, grass-roots groups, and everyday Americans rallied to ensure people had enough to eat. In Rhode Island, veterans and volunteers collected nearly 2,000 pounds of food for the Providence Veterans Affairs hospital pantry. In Nebraska, bookstore employees organized a food drive that brought in a similar amount. In Pittsburgh, a school pantry helped dozens of families with food and hygiene products. Across the country, community food pantries saw an unexpected uptick in donations.

These efforts deserve thanks not only from those who received support but from the country as a whole. Many of the people who went without SNAP benefits or missed paychecks were our neighbors, friends, and family.

Now it’s time for Congress and the president to finally address the issue that helped spark the shutdown in the first place. Health care in the United States remains extraordinarily expensive, and the federal government and states must confront the high costs that make it so difficult to cover all Americans. It’s time to shine a bright spotlight on this fundamental challenge.

As this difficult year in public health draws to a close, it can seem quaint to take a moment to recognize what we’re thankful for, especially with so many challenges on the horizon, from soaring health care costs to rebuilding trust in science and public health. But practicing gratitude, whether in our personal lives or in public health, is not about feeling better during difficult times. It reminds us of our strengths and of what we can accomplish together.

Wishing you and your loved ones a happy Thanksgiving and a healthy holiday season.

Subscribe now