False Reassurance & Interval Cancers Explained

A normal screening result is a narrow statement. It says that one specific test, looking at one specific part of the body, on one specific day, did not detect anything abnormal.

It does not say you do not have cancer. It does not cover organs the test did not look at. And it does not extend forward in time. Those three limits are ordinary properties of screening, not failures of it, but they are rarely said out loud when someone is handed a clear result and a sense of relief.

The risk is behavioral rather than technical. A person who finds a lump in May, remembers a normal mammogram in March, and decides to wait until next year's appointment has been harmed by a correct test result.

A cancer diagnosed after a normal screen and before the next scheduled one is called an interval cancer. Screening programs count them deliberately, because a program with zero interval cancers is not a possibility.

They arise three ways. The cancer was genuinely not visible at the time. It was technically present on the image but not identifiable prospectively among normal-looking tissue. Or it had not yet developed. Fast-growing tumors are overrepresented in this group for a simple reason: the faster a cancer grows, the less time it spends at a size that is detectable but not yet symptomatic.

Every screening test misses some cancers. The useful thing is knowing roughly how many.

Mammography. How much a mammogram detects drops as breast tissue gets denser. An AHRQ review of US registry data found that in women in their forties, mammograms caught roughly four in five cancers in non-dense breasts but closer to three in five in dense ones. The gap holds at older ages. The DENSE trial, run in Dutch women with extremely dense breasts, found a few interval cancers per thousand screens in the mammogram-only group. These are group averages from set populations and eras. Read them as scale, not as your own odds.

Stool testing. One round of FIT misses a substantial minority of the colorectal cancers present, which is exactly why it is repeated annually rather than done once. Stool DNA testing finds more cancers in one round. It also flags more people who turn out not to have cancer. And it is weaker than colonoscopy at finding precancerous polyps. Test brands and generations differ, so ask which test you had.

Colonoscopy. The most thorough colorectal option and still not perfect. Polyps behind folds or in poorly prepared segments can be missed, which is one concrete reason bowel preparation quality matters.

Low-dose CT. Small nodules can be obscured, and cancers can arise between annual scans.

None of this argues for skipping screening. For several cancers, screening programmes have been shown to lower deaths, and for some strategies they also prevent cancer by removing precancerous lesions. The size of that benefit differs by cancer, by test and by who is screened. It argues for holding a normal result as good news with a defined scope.

There is a second and larger gap. Recommended screening exists for only a handful of cancers: breast, cervical, colorectal, lung in eligible people, and prostate as a shared decision. There is no routine screening for pancreatic, ovarian, kidney, stomach, esophageal, uterine, brain, bladder, or blood cancers.

A person who is up to date on every recommended test has been checked for a small fraction of the cancers that exist. For all the others, noticing a symptom and getting it evaluated is the entire early-detection system.

This is the operative rule, and it is worth stating without qualification: a new, persistent, unexplained symptom is evaluated on its own, no matter how recently a screening test was normal.

Symptoms that warrant a call rather than a wait include a new breast lump or skin or nipple change, rectal bleeding or a persistent change in bowel habits, unexplained weight loss, blood in the urine, a cough or hoarseness lasting more than a few weeks, difficulty swallowing, a lump that does not resolve, unexplained persistent pain, and any vaginal bleeding after menopause.

Most of these turn out to have benign explanations. That is fine. The point is that the explanation should come from an evaluation rather than from arithmetic about when your last test was.

A useful sentence has three parts: what changed, when it changed, and what you want. For example, my mammogram in March was normal, this lump is new since then, and I would like diagnostic imaging of this area.

If you are told to wait and that does not sit right, two follow-up questions are reasonable. What specifically would change the plan. And can we note in my chart that I raised this concern today.

A normal screening result is worth having. It is not a shield, and it has an expiration date.

• NCI: Dense breasts, answers to commonly asked questions
• AHRQ evidence review: mammography performance by breast density
• USPSTF: Colorectal cancer screening recommendation
• NCI: What cancer screening statistics really tell us