From the time Kelly was a toddler, Anders was her protector. When she was just 15 months old, a serious illness damaged her kidneys. Doctors warned the family that one day she might need a transplant. Decades later, when Kelly turned 30 and her kidneys began to fail, Anders didn’t hesitate, he immediately volunteered to donate one of his.

The surgery was successful.

The next morning, Anders visited Kelly and told her donating his kidney had been the best day of his life. But hours later, he began experiencing severe pain and vomiting. His pain medication was changed, and warning signs went unnoticed. When his mother returned to check on him, Anders’ hand was cold, his lips were blue, and he wasn’t breathing.

Anders fell into a coma and died nine days later. The family was initially told his heart had simply stopped. Only after pushing for answers did they learn the truth: a cascade of preventable failures, inadequate monitoring and medication management, had taken the life of a healthy young man who had just saved his sister.

Stories like Anders’ are not rare tragedies. They are symptoms of a systemic failure. And the most maddening part? We already know exactly how to prevent them.

The number hiding in plain sight

For more than two decades, patient safety experts have warned that preventable medical harm is one of the most urgent public health crises in America. Research suggests medical errors contribute to roughly 250,000 deaths each year in the United States, placing them behind only heart disease and cancer as a cause of death. Globally, the toll may reach 3 million deaths annually.

But here is what rarely gets stated plainly: if every hospital in this country implemented all of the evidence-based practices that researchers and clinicians have already identified and validated, we could reduce that death toll from approximately 200,000 a year to as few as 20,000, a 90% reduction. Not through new drugs or breakthrough science. Through

protocols that exist today, posted on our website, available to any hospital administrator who cares to look.

That is not an aspirational goal. It is a quantifiable, achievable outcome that we are choosing, collectively, not to pursue.

In 1999, the Institute of Medicine’s landmark report To Err Is Human shocked the nation by estimating that 44,000 to 98,000 Americans were dying each year from preventable medical errors. That report was meant to ignite transformation. It promised accountability, transparency, and systemic change. By 2011, an OIG report showed we were losing 200,000 patients a year. Twenty-five years after that first alarm, families like the Pedersons are still paying the price.

Why the aviation comparison misses the point

When people talk about patient safety, they inevitably reach for the aviation analogy. Airlines transformed their safety culture; why can’t hospitals? It’s a fair comparison as far as it goes, but understanding why it breaks down reveals the real problem.

First, when a plane goes down, it dominates the news cycle for days. Medical errors kill the equivalent of two fully-loaded passenger jets every single day in America, and it barely registers. The absence of a single catastrophic, visible event means there is no public outrage, no pressure campaign, no congressional hearing.

Second, when a plane crashes, the pilots die too. That brutal alignment of incentives — skin in the game — drove aviation to make safety non-negotiable. When a patient dies from a preventable error, the doctors and nurses go home. That is not a criticism of healthcare workers, most of whom entered the profession to heal people. It is a structural reality: the system does not force those who design and deliver care to bear the consequences of its failures in the same visceral way.

Third, passengers can choose not to fly. That market pressure gave airlines a powerful financial incentive to fix safety problems fast. Patients who need hospital care have no such choice. They come because they must, which means hospitals face no equivalent consumer penalty for unsafe outcomes.

The lesson from aviation is not simply that bold safety goals work, though they do. The lesson is that healthcare lacks the self-correcting mechanisms that forced aviation to change. Which means those mechanisms have to be built deliberately, from the outside in.

What actually works: the CHOC model

We know this can be done because we have done it.

When I was asked to chair the quality committee at Children’s Hospital of Orange County, we brought in all 20 of the evidence-based practices the Patient Safety Movement Foundation had identified; standardized protocols for the specific, known causes of preventable harm: failure to rescue, medication errors, hospital-acquired infections, sepsis, communication breakdowns, venous thromboembolisms, falls, and diagnostic errors. The result: zero preventable deaths for more than six years straight, with serious harms dramatically reduced as well.

But the clinical protocols alone were not what made it work. The turning point was a governance decision.

When I joined the quality committee, I noticed that the hospital measured itself primarily against peers on a narrow set of Medicare “never events.” As long as they were slightly below the peer average, leadership felt comfortable. I asked a simple question: why isn’t the goal zero? To their credit, they agreed — and shortly after, the faculty suggested and the board voted to tie one-third of faculty bonuses to achieving zero preventable harm.

What happened next taught me everything about how institutional change actually occurs. Instead of hoping for zero, the hospital started planning for zero. Staff pulled every one of our evidence-based practices, color-coded their compliance — green for what they were doing, yellow and red for what they weren’t — and built a mitigation plan to close every gap. They didn’t achieve zero the first year. Instead of retreating, they doubled down. And then they did achieve it, year after year.

The formula is straightforward: board-level attention, incentives aligned with the goal, and a clear evidence-based roadmap. When those three things come together, patient safety stops being a values statement and becomes a self-governing system.

I cannot think of more than five hospitals in this country that have implemented all 20 evidence-based practices. Most are doing three or four.

The policy fix that almost happened

The structural barriers to change are real, but they are not immovable. Between 2021 and 2024, I served on President Biden’s President’s Council of Advisors on Science and Technology, where we produced a detailed patient safety report outlining exactly what the federal government could do, without new legislation, to drive adoption of evidence-based practices across the healthcare system. The report was released in September 2023 with significant attention and real momentum. Then October 7th happened, and it was overtaken by events.

As the Biden administration wound down, executive orders were drafted that could have used the levers of CMS reimbursement to create meaningful incentives for hospitals to act. CMS itself asked the White House to hold off, promising to implement the recommendations administratively. When the transition came, those recommendations were not implemented.

It is a familiar story in this space. I have spent years walking the halls of Congress, meeting with senators and members from both parties who express genuine alarm about preventable patient deaths, and then watching hospital industry lobbyists arrive to warn of financial burdens and regulatory overreach until the political will evaporates. The American Hospital Association is good at its job to the detriment of its members and patients.

The carrot, the stick, and the fix

What the PCAST report proposed, and what I believe remains the most viable path forward, is a reimbursement reform with real teeth.

Under the current system, a hospital can perform a hip replacement, cause a patient to die from a medication overdose, and still collect full payment for the original surgery. There is no financial consequence for the failure; in some cases, the complications generate additional billable care.

The proposal flips that logic. If a hospital has implemented all evidence-based practices and a patient is still harmed, the hospital continues to receive full reimbursement — for the original procedure and for any secondary care required to address the harm. We are not asking for perfection. We are asking hospitals to show up and do the work.

But if a hospital has not implemented the evidence-based practices and a patient is harmed, reimbursement stops, not just for the harm-related care, but for the original procedure as well. And the converse, if a patient is injured and the hospital had implemented the evidence-based practices to avoid that harm, then the hospital should be paid fully for everything, including the harm-related care. The argument to hospitals is simple: just dress for the game. We will worry about the scoreboard later.

I believe if that single reimbursement change were made, we would see rapid, widespread adoption of evidence-based practices within a few years, like we saw at the hospital that I chaired its quality committee. The hospitals that insist implementation is too costly would discover that not implementing is far more expensive.

The cost of inaction

Every statistic about patient safety hides a human story.

A mother who never leaves the hospital after childbirth. A father who dies from a missed diagnosis. A child who never gets the chance to grow up.

These tragedies ripple far beyond hospital walls. Families are shattered. Communities lose loved ones. Healthcare workers carry the emotional weight of errors they never intended to make. Preventable harm is not just a clinical issue. It is a moral one — and an economic one that our system has decided, through inaction, to keep paying.

Some critics argue that implementing evidence-based practices won’t get us to zero. I would like to have that conversation after every hospital has actually tried. If we fall short of zero, we will still be far below 200,000 deaths a year. That alone would represent one of the greatest public health achievements in American history.

Hospitals should not be performing elective procedures until they have implemented every evidence-based practice available to them. Healthcare leaders, policymakers, insurers, and regulators must treat patient safety with the same urgency we apply to pandemics and national security threats.

Behind every number in this crisis is someone like Anders Pederson — a healthy young man who walked into a hospital to give his sister a kidney and never walked out. The third leading cause of death in America is not inevitable. It is preventable. We have known that for 25 years. The question is whether we finally intend to act like it.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.