Why high reliability matters

When Dr. Carol J. Scott first wrote about her protocol, she began with a question: what is health? Her answer opened with a plain claim.

"Health is our most important asset. It impacts our jobs, our relationships, and even our ability to enjoy life itself." Dr. Scott, from her original essay "What is Health?"

Then she pointed at the oddity in how most of us treat that asset. Think about how you look after money. A retirement account gets quarterly statements, an adviser, a rebalancing schedule and a plan for the bad year that will eventually come. A house gets an inspection before you buy it and insurance afterward. Your health, which pays for both, gets an annual checkup if you're diligent and a quiet hope if you're busy. It slips, as Dr. Scott wrote, "supplanted by the obligations of daily life."

That gap says less about willpower than about design, and design problems have known fixes. The best ones come from places where one slip can kill hundreds of people: cockpits, nuclear control rooms, the flight deck of an aircraft carrier. Researchers call them high reliability organizations. The habits that keep them safe work at a kitchen table too, and the numbers on this page show why they're needed there.

Most of the damage is done slowly

Dr. Scott's essay drew a line between two kinds of illness. A broken leg, a sprained ankle or a bad infection can take a few months out of your life, but most people recover. The illnesses that reshape a whole life usually don't arrive from outside. They come, she wrote, from "avoidable failures in the physiological machinery" of the body, failures that build "over the course of months or years." The name for the result is chronic disease.

It no longer waits for old age. In 2023, three in four American adults reported at least one of twelve common chronic conditions and more than half reported two or more, according to a CDC analysis of national survey data. Even among adults 18 to 34, six in ten had one.

The slowness is the cruel part, and it's also the hopeful part. Blood pressure that creeps up over a decade, or blood sugar that drifts into the prediabetic range, shows up in readings long before it shows up as symptoms. More than 115 million American adults have prediabetes, and roughly 8 in 10 of them don't know it. A failure that takes years to develop can be caught. A sudden one rarely gives you that chance.

High blood pressure is the clearest case. Take 100 American adults who have it, counting from 130/80, the threshold the 2025 AHA/ACC guideline uses for stage 1. According to the National Center for Health Statistics, about 59 of them have been told so by a clinician, about 51 are being treated and only about 21 have it under control. Nothing about the other 79 feels different on an ordinary morning. A slow failure makes no noise until the stroke or the heart attack, and by then the warning time is gone.

What actually ends American lives

The same pattern shows up at the end of the line. When the National Center for Health Statistics tallied deaths in 2024, heart disease and cancer alone accounted for more than 1.3 million of them. Eight of the ten leading causes were chronic diseases: conditions that develop over years and, for the most part, show up in numbers before they show up as pain.

Many of those deaths came at the end of long lives, and no method prevents all of them. But when the CDC compared each state's death rates with those of the three best-performing states, it estimated that about a third of deaths before age 80 from heart disease (34%) and stroke (33%) were potentially preventable. That analysis used 2008 to 2010 data; the share is a rough guide, not a forecast. Still, it puts a number on what slow failure costs.

Trying harder is not a safety system

When something goes wrong in health care, the instinct is to find the person who slipped. The safety researcher James Reason called that the "person approach," and in a 2000 paper in the BMJ he argued that it misreads how harm happens. Every complex system has layers of defense, and every layer has holes. Most days the holes don't line up; harm gets through on the day they do.

"We cannot change the human condition, but we can change the conditions under which humans work." James Reason, "Human error: models and management," BMJ, 2000

High reliability organizations run their whole day on that idea. In the 1980s, researchers at the University of California, Berkeley, went looking for operations that handle dangerous technology for years without a catastrophe, and they studied Navy carrier flight decks, air traffic control and a utility that ran a nuclear plant. What they found was less a rulebook than a set of habits. People there assume something is about to go wrong, watch what is actually happening rather than what the last report said, distrust the comfortable explanation, hand each decision to whoever knows most about it and practice recovering. The Agency for Healthcare Research and Quality describes the result as "a condition of persistent mindfulness." The full history is on the origins page.

Your own care runs through exactly the kind of system Reason described. In a single year you might see a primary care doctor, two specialists, a pharmacist and an urgent care clinic, each with its own records and its own slice of the picture. Every handoff between them is another slice of cheese. Diagnosis is one place where the holes show. By the best available estimate, about 795,000 Americans a year die or are permanently disabled after a dangerous disease is misdiagnosed, and in 2015 the National Academies concluded that "most people will experience at least one diagnostic error in their lifetime, sometimes with devastating consequences."

These are modeled estimates, not head counts, and the ranges are wide. The point survives the uncertainty: a large share of serious harm comes from problems that were present, and often detectable, before anyone named them correctly. You are the one constant in your own care. You attend every appointment, you remember what the last doctor said and you notice when something has changed, which makes you a layer of defense whether or not anyone has told you so.

Handoffs carry their own risk. In a study of 400 patients sent home from the medical service of one academic hospital, 19% had an adverse event within about three weeks, most often involving a medication, and about a third of those events were judged preventable. A fresh expert look works in the other direction. At Mayo Clinic, only 12% of 286 patients referred by primary care doctors left with the same diagnosis they arrived with, and 21% left with a distinctly different one. That was one center and a group of patients sent on for more advanced care, so the figure won't hold everywhere. It still shows how often a second, expert look changes the picture.

For cancer, timing decides a great deal

With cancer, the defense that matters most is finding it early. National Cancer Institute figures for people diagnosed from 2016 to 2022 show the same shape for every common cancer: survival is high when the disease is found while it's still localized, and it falls steeply once the cancer has spread. Colorectal cancer is the starkest case, with five-year survival of 91% when localized and 17% at the distant stage.

Screening exists to find cancers in people who feel fine, while they're still early, and it depends on something unglamorous: a regular place to get care. In 2023, 71.5% of adults with a usual source of care were up to date on colorectal cancer screening, compared with 41.1% of those without one, according to national survey data published by the CDC. Among people under 65 with no insurance, the figure was 23.8%.

In reliability terms, a usual source of care is a monitoring layer: someone who knows your history, tracks what's due and notices when a number moves. About one adult in ten doesn't have one, and among adults 18 to 34 it's about one in six. To see which screenings the U.S. Preventive Services Task Force recommends for your age and sex as of October 2026, use the screening timeline.

Where the defenses are thinnest

Reliability science has a useful habit for reading uncomfortable numbers: it examines the system before it judges the person. That habit matters most when the numbers describe Black Americans.

In 2024, life expectancy at birth was 74.8 years for non-Hispanic Black Americans and 78.9 years for non-Hispanic white Americans, a gap of 4.1 years, according to the 2024 U.S. life tables. Nearly six in ten Black adults (58.0%) have high blood pressure under the current 130/80 definition, the highest share of any group. The maternal mortality rate for Black women in 2024 was 44.8 deaths per 100,000 live births; for white women it was 14.2.

It would be easy to read those figures as a story about personal choices, and the evidence doesn't support that reading. In 2023, 85.6% of Black women ages 50 to 74 were up to date on breast cancer screening, above the 80.0% for all women that age, according to the same CDC survey analysis. Yet the American Cancer Society's 2025 report on cancer in Black Americans found that Black women are 9% less likely than white women to be diagnosed with cancer and 10% more likely to die of it, with death rates for breast, colorectal, cervical and liver cancer running 40 to 50% higher among Black Americans than white Americans.

Through a reliability lens, a gap like that is information about the system, not a verdict on the people in it. It says the layers have more holes for some patients than for others. Closing them is the health system's job. Until it does, an extra layer you control is worth more, not less: knowing your own numbers, keeping your own records, asking for the reasoning behind a decision and bringing someone with you to the appointments that matter.

What high reliability looks like for one person

Dr. Scott's original site described the SCOTT Protocol as a personal adaptation of high reliability science, the set of tools that aviation developed to reduce the risk of critical failures and that nuclear power and other fields took up. Each of its five steps borrows one habit from those operations and points it at a single body.

Read the numbers on this page again with the steps beside them. The chronic conditions that become common by midlife are what self-awareness is for: knowing in advance where your own body is most likely to fail, from your family history to your risk factors. Blood pressure that drifts for years without a symptom is the case for consistent monitoring, which treats your health as a trend instead of a once-a-year snapshot. Missed diagnoses and the screening gap are both problems of oversight, of having the right expert watching and a regular place where someone knows you.

The last two steps deal with what happens after something goes wrong, which it eventually will. Tenacity means expecting the side effect, the missed test or the bad month, and planning the recovery before you need it. Taking action closes the gap between knowing and doing: the colonoscopy you book this week rather than next year, the cuff you actually use.

Ordinary actions add up. In two long-running studies of U.S. health professionals, women who kept all five low-risk habits the researchers measured, from never smoking to eating a high-quality diet, were projected at age 50 to live 14 years longer than women who kept none; for men the gain was 12 years. Those studies were observational and their participants mostly white, so the exact numbers won't transfer to everyone. The direction is hard to argue with.

None of this asks you to become a doctor. It asks you to run the most complicated system you will ever manage with the same discipline a flight crew brings to a routine takeoff. A good place to start is the reliability check, which takes about five minutes and shows which of your five defenses is thinnest.

Last updated October 16, 2026. Guidance cited here was current on that date; clinical guidelines change, so check with your own clinician. See our editorial policy.