“Catch it early” has been the rallying cry of public health campaigns for decades — and the intuition is so strong that questioning it sounds almost callous. But screening data accumulated over the past thirty years has revealed that earlier isn’t reliably better, and in some conditions it’s actively worse. The story is more nuanced than the awareness ribbons suggest, and patients deserve the unsanitized version.
Lead-time bias makes screening look more effective than it is
If a disease is going to kill someone at age 70 either way, detecting it at 60 instead of 65 makes survival statistics look better — five extra years of “survival” — without changing when the person actually dies. This is lead-time bias, and it inflates the apparent benefit of screening programs in ways that took decades to untangle. Length-time bias compounds the problem: screening preferentially catches slow-growing tumors that were unlikely to cause harm anyway, while fast, aggressive cancers often appear between screenings. The result is screening programs that look successful in survival statistics but show little or no improvement in overall mortality when measured properly. Mammography, PSA testing, and thyroid ultrasound all illustrate this gap.
Some cancers don’t need to be found
Autopsy studies on people who died of unrelated causes routinely find indolent prostate cancers, small thyroid nodules, and ductal carcinoma in situ that never produced symptoms. These are real cancers under the microscope but not real threats biologically. When screening picks them up, the patient becomes a cancer patient — with all the surgery, radiation, hormone therapy, anxiety, insurance consequences, and employment implications that label carries — for a disease that wouldn’t have hurt them. South Korea’s thyroid cancer screening program is the textbook case: incidence increased fifteen-fold while mortality stayed flat. Tens of thousands of thyroidectomies were performed on cancers that would never have surfaced. The patients now live with lifelong hormone replacement and a cancer history that affects everything from life insurance to job prospects.
The right screens are the ones with mortality evidence
Some screening programs genuinely save lives. Colonoscopy in average-risk adults reduces colon cancer mortality through both detection and polyp removal. Cervical cancer screening has dramatically cut deaths from a once-common cancer. Lung cancer screening with low-dose CT in heavy smokers shows clear benefit in randomized trials. The pattern in successful screens is the same: randomized data showing reduced disease-specific and all-cause mortality, not just earlier detection or longer “survival.” When a screen hasn’t crossed that bar — and many widely promoted screens haven’t — the conversation should be cautious. The default isn’t “more is better”; it’s “what does the evidence actually show?”
Bottom line
The phrase “early detection saves lives” is true for some conditions and misleading for others. Patients deserve to know which is which, what the harm rates of follow-up procedures look like, and what overdiagnosis means in their specific case. A clinician who can walk through the number-needed-to-screen, the false-positive cascade, and the realistic benefit is doing better medicine than one who treats every screen as obviously worthwhile. Sometimes the right answer is yes, sometimes no, and sometimes wait.
Leave a Reply