What the letter grade on a cancer-screening recommendation actually means
A, B, C, D and I are not school marks. They encode how strong the evidence is, whether to screen at all — and in the US, whether insurance has to pay for it.
For how we live.
A, B, C, D and I are not school marks. They encode how strong the evidence is, whether to screen at all — and in the US, whether insurance has to pay for it.
A stool test at home and a colonoscopy are both recommended options with different trade-offs. The starting age fell to 45 because the disease is rising in younger adults.
US guidelines now let women aged 30 to 65 screen every five years with an HPV test. One major body would make that the primary test from age 25 — and self-collected samples work about as well.
A 200,000-woman trial found that screening caught more early-stage disease. It did not save lives — a result that explains why guidelines recommend against screening the general population.
Smoking is the leading modifiable cause, and the most common presenting sign is blood in the urine. There is no recommended screening test, which makes that sign the thing to act on.
The disease is lethal largely because it is found late. That is also why guidelines recommend against screening the general public — the evidence says it would do more harm than good.
Most skin cancers rarely kill; melanoma, about 1 percent of them, causes most of the deaths. There is no proven benefit from routine whole-body screening — which puts the focus on changing moles.
Two large randomised trials found low-dose CT screening cut lung-cancer deaths in heavy smokers. The eligibility rules are narrow, and the reason is written into the evidence.
US guidance recommends mammography from 40. A Canadian review of the same randomised evidence put the benefit at 0.27 fewer breast cancer deaths per 1,000 women aged 40-49 over ten years.
That is the 23-year result from the largest randomised trial. A 2026 Cochrane review of 789,086 men rates the mortality benefit moderate-certainty and the harms largely unmeasured.