Oversight: put the right expert on the problem

On a carrier flight deck, rank does not decide who is right. The Berkeley researchers who studied Navy flight operations in the 1980s found crews that handed a decision to whoever knew the most about the problem in front of them, whatever that person's stripes. Karl Weick and Kathleen Sutcliffe later named the principle deference to expertise, and the AHRQ patient safety primer sums it up for health care: decisions go to the people who know the work best, often frontline staff, rather than to the most senior. In a hospital, the bedside nurse who has watched a patient all night may know more than the attending who saw the chart at seven, and a reliable hospital acts on that.

"High reliability organizations have a mechanism in place to identify individuals with the greatest relevant expertise, and defer to their judgement for critical decisions." Dr. Scott, from the original SCOTT Protocol

The step Dr. Scott calls Oversight asks you to build that mechanism around your own health, and it works in two directions. The first points outward. When a problem is serious, unclear or unusual, get the person who knows the most about that particular problem onto it, even if that means asking for a second look. The second direction points inward, and people skip it. Nobody in the exam room knows more than you do about what your body felt like last Tuesday, or what you will and won't put up with to feel better.

The right expert on the problem

Most of the time, the clinician you already see is the right expert. Oversight earns its keep when the problem is hard, because hard problems are where diagnosis goes wrong. A 2015 National Academies committee concluded that most people will experience at least one diagnostic error in their lifetime, sometimes with devastating consequences.

How large is that problem in the United States? A 2023 estimate published in BMJ Quality & Safety by David Newman-Toker and colleagues, puts the toll at about 795,000 Americans a year who die or are left permanently disabled because a dangerous disease was misdiagnosed. Five conditions (stroke, sepsis, pneumonia, blood clots in the veins and lungs, and lung cancer) account for close to 40 percent of that harm. The figure is modeled from hospital discharge records and cancer registries rather than counted case by case, so treat it as an estimate with a wide range rather than a count.

What a second look finds

The study quoted most often on second opinions comes from Mayo Clinic. Researchers there pulled the records of 286 patients whom primary care clinicians had referred to the clinic's General Internal Medicine Division in 2009 and 2010, then compared the diagnosis written on each referral with the diagnosis the patient left with. The results, published by Van Such and colleagues in 2017, look like this.

Read the study for what it is. It covers one academic center, a few hundred people and two years, and these were patients whose problems were complicated enough that someone had already decided they needed to go to Mayo. It does not tell you that a routine diagnosis from your own doctor is wrong one time in five. What it does show is that on hard cases, a second set of expert eyes usually changes the picture, mostly by sharpening it. The cases that ended with a distinctly different diagnosis also cost significantly more, one more reason to get the answer right early.

Mayo's own news release said "as many as 88 percent" of these patients went home with a new or refined diagnosis. That's accurate arithmetic, but switch the chart to "As headlined" and you can see what the single number hides.

When a second opinion is worth asking for

You don't need a second opinion on a sinus infection. These are the situations where it usually pays to widen the circle:

• A serious or rare diagnosis, before treatment starts. The days between a diagnosis and the first treatment are the cheapest moment to check it.
• Major surgery that isn't an emergency.
• A treatment plan that is risky, complicated or hard to undo.
• A diagnosis that doesn't fit what you are living with, or symptoms that keep coming back after treatment that should have worked.
• The sense, after more than one visit, that nobody is really listening. That one counts too.

Cost is often what stops people. If you have Medicare, check the rules for your plan: Medicare Part B covers a second opinion in some cases for surgery that is medically necessary but not an emergency. You pay 20 percent of the Medicare-approved amount, tests the second doctor orders are covered when they are medically necessary, and Medicare also covers a third opinion if the first two disagree. If you have a Medicare Advantage plan or private insurance, the rules can differ, so call the number on your card and ask before you book.

The expert on you

Deference to expertise has a second half that is easy to miss: sometimes the person who knows the most is the one in the bed. Dr. Scott put it plainly in her original slide.

"No physician is an expert in what it's like to be you or how you define wellbeing." Dr. Scott, from the original SCOTT Protocol

Your clinician knows disease. You know your baseline: how much energy is normal for you in February, when the swelling started, what the new pill did to your sleep, which symptom scares you and which one you have learned to live with. Her slide goes on: "As an expert in you, you have the ability to bring relevant, timely information to your provider, which allows them to provide the best possible care." That information is often the deciding clue in a diagnosis, and it only exists in one place.

Researchers are starting to document this. A 2026 study of cancer care teams that interviewed patients and clinicians about high reliability habits found that patients "developed expertise throughout treatment, thereby supporting system safety." The same authors noted that much of the monitoring fell to patients and their families. If you are going to carry that load anyway, carry it on purpose.

Owning your expertise also means knowing where it stops. You are the authority on what you feel and what you value. You are not the authority on what a troponin level means. Oversight asks you to bring the first kind of knowledge into the room, clearly and early, so the person with the second kind can use it.

Say it out loud, and say it plainly

Expertise only helps if it reaches the person making the decision. Aviation learned that the hard way. On March 27, 1977, a KLM 747 began its takeoff roll in fog at Los Rodeos airport on Tenerife while a Pan Am 747 was still taxiing on the same runway. The KLM flight engineer, listening to the radio, suspected the runway wasn't clear. He said so, but he said it as a question.

17:06:11 brakes off17:06:47 impact

1. 17:06:11 The KLM captain releases the brakes and starts the takeoff roll.
2. 17:06:20 Tower “Stand by for takeoff … I will call you.”
3. 17:06:25 Tower to Pan Am “Report runway clear.”
4. 17:06:32 KLM flight engineer “Is he not clear, then?”
5. 17:06:34 KLM captain “What do you say?”
6. 17:06:34 KLM flight engineer “Is he not clear, that Pan American?”
7. 17:06:35 KLM captain “Oh, yes.” (emphatically)
8. 17:06:47 Impact with the Pan Am 747 still on the runway in fog.

The collision killed 583 people, still the worst accident in aviation history. The engineer had the right worry and voiced it twice; the captain brushed it aside, and the engineer let it go. Read the full story in our field note.

Twenty-one months later, United Airlines Flight 173 ran out of fuel and crashed near Portland, Oregon, after circling for about an hour while the captain worked on a landing gear problem. The National Transportation Safety Board's report named, as a contributing cause, "the failure of the other two flight crewmembers to fully comprehend the criticality of the fuel state or to successfully communicate their concern to the captain" (NTSB-AAR-79-7). The crew had noticed; they hadn't made the captain hear it. Six months later, NASA brought airlines and researchers to San Francisco for a workshop on resource management on the flight deck. The training that grew out of that work, now called crew resource management, teaches junior crew members to state a concern directly and teaches captains to invite it.

You are a crew member in every appointment. If something worries you, state it as a concern, not a hint: "I'm worried this pain is my heart. What makes you sure it isn't?" If you leave without understanding the plan, the plan has a hole in it. The Institute for Healthcare Improvement's Ask Me 3 program boils the minimum down to three questions every patient should be able to answer before leaving:

Who is on your crew?

High reliability organizations know before the emergency who holds which job. Most families work it out in the hospital hallway. Take two minutes now. The seats below are the ones a reliable personal system needs filled, including the clinician who coordinates your care and the person legally named to speak for you if you can't. The National Institute on Aging describes that last seat, the health care proxy, as someone who should know your values and wishes, named in a legal document called a durable power of attorney for health care.

An empty seat is not a failure. It's a hole you now know about, which is the whole point of the exercise. Fill the proxy seat first if it is open, then put the card where your family can find it, next to your one-page health brief.

Oversight is the step that keeps you from facing a hard problem alone, or with the wrong help. The next step is about what happens when, despite all of that, something still goes wrong.

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.