Cancer screening: is it really as simple as it sounds?

This is my personal take on the evidence, not medical advice.

Recent debates in the UK have pushed back against the comfortable assumption that screening the whole population for cancer is always a good idea. As an epidemiologist, this is a question I have wrestled with for the best part of 40 years, because the maths are not as simple as most people believe.

A good screening programme, in theory, is something like a blood test that picks up cancers early, before they have spread, so they can be treated with drugs or surgery to reduce the risk of death. For that to work, the test has to be sensitive, meaning it catches most cancers and misses very few, and specific, meaning it does not throw up false positives, where the test looks abnormal for some other reason but there is no cancer there at all. This is exactly where the trouble starts, whether we are talking about prostate cancer, which has been in the news lately, or, more surprisingly, breast cancer and the mammograms already built into the UK screening service.

Take prostate cancer. Many countries screen routinely using a blood test called PSA, or prostate specific antigen. If it comes back high, men are referred on for scans, biopsies and sometimes surgery. The UK has held back from doing this across the whole population because the harms are judged to outweigh the benefits. The National Screening Committee looked hard at the evidence and concluded that the benefit is small: for every thousand men screened, roughly two fewer will die of prostate cancer over the following 15 to 20 years. Those two lives matter enormously, especially if you are one of them. But no trial has shown that screening improves overall survival, largely because many of these men die of something else first.

Now look at the other side of the ledger. For every thousand men screened, around 20 will be told they have a prostate cancer that was borderline and mild, and would very probably never have troubled them in their lifetime. All 20 carry the anxiety and fear of a cancer they assume could kill them. And about 12 of them will go on to have major treatment, surgery or radiotherapy, with no real benefit, sometimes leaving them with incontinence or sexual problems for the rest of their lives. A tiny benefit set against a great deal of harm, from telling healthy men they have cancer, is the real reason this country has not adopted it. In its most recent review the committee went only as far as recommending two yearly PSA testing for the small group of men who carry a high risk BRCA2 gene variant, while still rejecting screening for everyone else.

Breast cancer is the intriguing counterexample, because here the UK has adopted screening. Yet the mammography programme is far from settled. Researchers, most prominently at the Nordic Cochrane Centre, have argued that the benefit in lives saved is modest compared with the number of women given false positive results, or diagnosed and treated for cancers that would never have harmed them. For some conditions, in other words, screening simply is not good enough to justify offering it to everyone.

Share

The newest hope was a single blood test that could detect any cancer years before it caused symptoms. A very large NHS trial set out to test exactly that. The results, reported in 2026, were sobering. The test failed to meet its main goal of catching enough cancers early to cut the rate of advanced disease, and so far it has not been shown to save lives. Independent experts concluded there is no evidence base to justify rolling it out to the whole population, and warned it could pull scarce scanning and biopsy resources away from people who already have symptoms.

So where does this leave us? My own view, and it is the direction the evidence points, is that we are better off spending our very limited NHS resources on encouraging people to be aware of their own bodies and to come forward early, and on building a faster, more efficient service to treat cancers when they are caught early, rather than on screening millions of people who will never develop these cancers. But as always in science, the picture keeps shifting, so we have to keep watching the evidence.