A clean bill of health is a snapshot

The phrase comes from the sea. A bill of health was a certificate given to a ship’s master on leaving port, stating whether infectious disease was aboard or in the port of departure, to be shown at the next port of arrival. A clean one meant nothing was known to be wrong on the day the ship sailed. About the voyage ahead, it said nothing at all.

That is a fair description of an annual physical. You sit on the crinkly paper, the cuff squeezes, a few tubes of blood go off to the lab, and a week later a portal message tells you everything looks good. The message is accurate. It describes the port you just left.

Dr. Carol J. Scott, an emergency physician who has spent her career seeing what happens when that voyage goes badly, wrote the warning into the first step of her protocol:

“A ‘clean bill of health’ can create a false sense of security.” Dr. Scott, from the original SCOTT Protocol

She made the longer case in a HuffPost column, “A Clean Bill of Health Can Be Deadly: The Power of S.M.A.R.T. Thinking”, first published in 2015. This note picks up the same thread: why a good report can mislead, what it can and can’t tell you, and how to use one without being lulled by it.

One reading, one afternoon

Blood pressure shows the problem most clearly because it never sits still. It climbs on the stairs, drops in your sleep, and answers to the coffee you drank in the car on the way over. A clinic reading is one sample of a number that moves all day.

For some people, that sample flatters them. Researchers call it masked hypertension: a reading that isn’t elevated in the clinic, and high blood pressure when a portable monitor follows you through an ordinary day. Pooling a New York study that did exactly that with national survey data, Wang and colleagues estimated in 2017 that 12.3% of American adults with a non-elevated clinic reading, no heart disease and no blood pressure medicine had masked hypertension. That works out to roughly 17.1 million people whose checkup said “fine” about the one number that wasn’t.

Guideline writers know readings mislead. The U.S. Preventive Services Task Force’s 2021 recommendation on hypertension screening, still the current statement as of October 2026, tells clinicians to confirm a high office reading with measurements taken outside the clinic before starting treatment. That protects people whose pressure jumps in the exam room. Masked hypertension runs the other direction, and a single visit can’t catch it, because the visit is the one moment it is hiding.

Never told

Masked hypertension is a special case. The bigger gap is plainer than that: a great many people with a measurable problem have simply never been told.

Using the current definition of high blood pressure (130/80 or higher, or taking medication for it), national exams from 2021 to 2023 found that only 59.2% of U.S. adults with the condition knew they had it. With prediabetes the gap is wider still. More than 115 million American adults have it, and the CDC estimates that roughly 8 in 10 of them don’t know. Another 11 million adults have diabetes that has never been diagnosed, more than a quarter of everyone who has the disease.

None of these conditions introduces itself. The National Institute of Diabetes and Digestive and Kidney Diseases says plainly that many people with type 2 diabetes have no symptoms, or symptoms so mild they go unnoticed. You can feel wonderful while the damage accumulates, which is what makes “I feel fine” and “the doctor said I’m fine” such comfortable sentences to live inside.

What a reassuring visit can leave out

Four published estimates, each drawn as dots out of 100. The groups being counted differ, so read the line under the number.

12 of every 100

U.S. adults whose clinic blood pressure was not elevated, with no heart disease and no blood pressure medicine

About 12 in 100 had masked hypertension: a reading that looked fine in the office and ran high the rest of the day. National estimate from 2005 to 2010 data.

• High blood pressure outside the clinic
• No masked hypertension

Show the numbers

Figure	Out of 100	Counted among	Source
High blood pressure outside the clinic	12	U.S. adults whose clinic blood pressure was not elevated, with no heart disease and no blood pressure medicine	Wang et al., 2017
Never told they have it	41	U.S. adults with high blood pressure (130/80 or higher, or taking medication for it)	NCHS Data Brief 511
Don’t know they have it	80	U.S. adults with prediabetes	CDC, 2026
Experience a diagnostic error in a year	5	U.S. adults who get outpatient care	Singh et al., 2014

Dots are rounded to whole numbers: 12.3% shows as 12 dots, 40.8% as 41, 5.08% as 5.

Sources: Wang et al., American Journal of Epidemiology, 2017; NCHS Data Brief 511, 2024; CDC; Singh et al., BMJ Quality & Safety, 2014.

Even a careful visit can miss

The visit has limits of its own. Clinicians make diagnoses under time pressure, with partial information, working from whatever the patient remembers to mention. In 2014, Hardeep Singh and colleagues combined three large studies of U.S. clinic patients and estimated an outpatient diagnostic error rate of 5.08%, about 12 million adults a year, and judged that roughly half of those errors could be harmful. Their own summary was blunt: diagnostic errors affect at least 1 in 20 American adults.

A year later, a National Academies committee reviewing the evidence concluded that “most people will experience at least one diagnostic error in their lifetime, sometimes with devastating consequences.”

Read that as arithmetic rather than an accusation. A short appointment once a year is a narrow window onto twelve months of living, and the person on the other side of the desk can only work with what comes through it.

What a checkup can and can’t tell you

None of this makes the annual visit less worth keeping. It makes it worth understanding. Roughly, the visit is strong on the left and blind on the right:

The left column is real value. The right column is where the protocol hands work back to you, and it’s the part the phrase “clean bill of health” quietly lets you forget.

Read your health as a trend

High reliability organizations have a name for paying attention to what is happening now instead of trusting the last report: sensitivity to operations. A control room doesn’t check the reactor once a year and call it safe until next October. For one person, the equivalent is modest. It mostly comes down to keeping your numbers in sequence and asking better questions about them.

Line the numbers up

Ask for your last few years of results side by side, or copy them onto one page yourself. Imagine a fasting glucose of 91, then 95, then 99 across three annual visits. Under the American Diabetes Association’s 2026 Standards of Care, prediabetes starts at 100 mg/dL, so each result is normal and each visit ends with good news. Lined up, the three numbers describe a direction, and direction is exactly what a yearly verdict leaves out.

Blood pressure works the same way, with the advantage that you can measure it yourself. An upper-arm cuff at home, used on a schedule your clinician agrees with, turns one number a year into dozens. The illustration below shows the difference that makes.

One person, two ways of watching

Two years of a healthy-feeling adult's systolic blood pressure. Switch between what three yearly checkups show and what a weekly home reading shows. Illustration with simulated readings.

Three office visits Weekly home log

Three checkups, three reasonable-looking numbers, until the third one isn't.

Show the numbers

Period	Average home reading (mmHg)
Months 1–3	120
Months 4–6	120
Months 7–9	123
Months 10–12	126
Months 13–15	129
Months 16–18	136
Months 19–21	139
Months 22–24	140

Office visit	Reading (mmHg)
Month 1	123
Month 13	128
Month 24	147

Thresholds from the 2025 AHA/ACC high blood pressure guideline (stage 1 is 130–139 systolic; stage 2 is 140 or higher), summarized by the American College of Cardiology. Readings are simulated for illustration.

Write down your baseline while you’re well

Your baseline is what normal looks like for you on an ordinary week: resting blood pressure, weight, how many flights of stairs leave you winded, how you sleep, what your energy is like at three in the afternoon. Collecting it is dull. It becomes very valuable the first time something shifts and a clinician asks whether it’s new, because you will have an answer that isn’t a guess.

Change the question you ask

“Am I okay?” invites a yes. It’s a kind question to ask a busy doctor and a poor one for learning anything. Try questions that point at the right-hand column:

• Which of my numbers is moving, even if it’s still in the normal range?
• Is anything I’m due for still undone?
• Should I be checking my blood pressure at home, and what reading would make you want to see me sooner?
• Given my family’s history, is there anything you would screen for earlier? The CDC suggests exactly this kind of conversation, for instance asking about earlier colonoscopy if a parent or sibling had colorectal cancer before 50.

Keep the good news, and put a date on it

Nothing here argues for distrusting a good report. Distrust is exhausting, and most good news is true. The point is to file it correctly: a clean bill of health is a dated document about a single afternoon, and the useful question is what has changed since that date.

“A healthy past doesn’t ensure a system free from failure.” Dr. Scott, from the original SCOTT Protocol

The ship’s certificate was never meant to promise a safe crossing. Somebody still had to watch the weather.

Put it to work

This note belongs to the first step of the SCOTT Protocol, self-awareness, which borrows the high reliability habit of treating small signals as information rather than proof that all is well. Three tools in the field kit turn it into something you can do this week.

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.