Rising Costs, Chronic Disease and AI: The Fight to Save U.S. Healthcare

In most industries, leaders can respond quickly when market conditions change. Within months, companies can shrink or expand their workforces, adopt innovative technologies, and reconfigure operations.

Healthcare lacks such flexibility. It takes a decade to train new physicians. Hospitals take years to plan, fund, and build — years longer than it takes for basic infrastructure in other industries.

With timelines like these, course correction in healthcare is inherently slow. Inaction or delays allow manageable threats to grow into crises. And by the time leaders move, it’s impossible to reverse the damage.

Two of the nation’s most pressing healthcare problems now face this reality. To make matters more challenging, we can’t fix one without solving the other.

The Affordability Cliff

Over the past 25 years, the nation’s total healthcare spending has climbed from $2 trillion to $5.3 trillion.

Businesses and the government have played “hot potato” with these rising costs.

To offset ever-higher premiums, employers slowed wage growth and switched to high-deductible health plans. In parallel, Medicare and Medicaid set payment increases well below the cost of delivering care, driving hospitals and physicians to make up the difference by charging higher rates to the privately insured.

The financial impact on families has been devastating. Half of Americans say they cannot afford their out-of-pocket expenses should they experience a major illness.

For everyone, financial challenges are mounting with no relief in sight. Insurance premiums are projected to rise by roughly 9% this year. In 2025, total U.S. medical costs rose more than 7% for the second consecutive year, pushing healthcare’s share of the economy to roughly 18%. Out-of-pocket spending by consumers climbed 7.2%, to exceed $500 billion, as demand for hospital care, prescription drugs, and physician services outpaced insurer projections.

Congressional action (and inaction) has amplified these pressures. December’s expiration of enhanced subsidies on the insurance exchanges is now driving double- and even triple-digit percentage premium increases for roughly 20 million enrollees. And beginning this year, another 8 to 10 million Americans could lose Medicaid coverage as new eligibility restrictions take effect.

Absent major intervention, healthcare spending is projected to exceed $7 trillion by the end of the decade, consuming more than one-fifth of the U.S. economy. At that point, small businesses will likely drop coverage for millions of employees. A major share of the federal spending will go toward paying off interest on the national debt. Funds, in turn, will be diverted from Medicare, Medicaid, and other healthcare programs.

And when the next recession begins (possibly within two years, according to historical analyses), the economic crisis will leave only one option: healthcare rationing.

To solve these financial issues without compromising the nation’s health, we will need to simultaneously address another major threat.

The Chronic Disease Epidemic

In the 21st century, the United States has experienced a scourge of chronic disease.

According to the Centers for Disease Control and Prevention, roughly 194 million U.S. adults now live with at least one chronic condition such as diabetes, hypertension, or heart failure. About 130 million people report multiple chronic diseases.

You might assume that if the healthcare system could prevent younger generations from developing these conditions, total healthcare spending would fall. But prevention alone will not offset the cumulative burden of chronic disease that’s already embedded in the American population.

To understand why, consider a single condition: diabetes. A patient newly diagnosed with diabetes can usually avoid serious, costly complications through lifestyle changes and relatively low-cost medications.

But when diabetes remains poorly controlled for a decade or more, biological damage accumulates. Each year, the risk of kidney failure or heart attack rises significantly. As a result, the annual cost of caring for a single patient with persistent, uncontrolled diabetes averages over $100,000 (four times more than someone who newly develops the disease). Thus, to offset the medical costs for an individual with a long history of diabetes, our nation would need to prevent four new cases, not just one.

Further complicating matters, effective chronic disease control requires substantial upfront investment, while the financial returns arrive years later.

That makes timing critical: the longer we wait, the fewer viable options remain. According to CDC estimates, acting now (through better prevention and management) could avert up to half of all heart attacks, strokes, cancers, and kidney failures, reducing national healthcare spending by $1 to $1.5 trillion annually. But if policymakers hold off, the required investment will be too large (and the payoff too delayed) to be politically or financially feasible.

Generative AI: The Missing Solution For Both Threats

Rising healthcare costs and chronic disease are not separate crises. They are conjoined. We cannot make healthcare affordable without making Americans healthier.

Generative AI tools like ChatGPT, Claude, and Gemini offer a credible path to reverse both threats.

To illustrate what GenAI makes possible, consider hypertension, the leading cause of stroke in the United States. Despite the availability of clear clinical guidelines and inexpensive medications to manage the disease, blood pressure remains uncontrolled in roughly half of Americans who have it.

So, rather than patients relying on three or four office visits a year (the current standard for chronic disease treatment), generative AI would analyze daily readings from home blood pressure monitors, detect early worsening trends, and prompt timely medication adjustments or outreach — before irreversible damage occurs. With GenAI, medical care moves from episodic to continuous.

Ultimately, preventing hundreds of thousands of heart attacks, strokes, and kidney failures (and avoiding vast medical costs) is a far more effective solution than rationing care. The combination of dedicated doctors, empowered patients, and generative AI can accomplish this more easily and reliably than any of the three alone.

We still have the choice and power to act. But the window is closing.

Robert Pearl, the author of “ChatGPT, MD,” teaches at both the Stanford University School of Medicine and the Stanford Graduate School of Business. He is a former CEO of The Permanente Medical Group.