back to top

We Don't Know If Cancer Screening Actually "Saves Lives", Experts Argue

While some types of screening can reduce deaths from a specific type of cancer, they don't necessarily save lives overall, according to an analysis in the British Medical Journal.

Posted on

This week in the British Medical Journal, a group of experts argue patients are being misled by claims that cancer screening "saves lives".

Rui Vieira / PA WIRE

In the analysis, Vinay Prasad, an assistant professor of medicine, Jeanne Lenzer, a former senior clinical policy analyst, and David Newman, a professor of emergency medicine, call for higher standards of evidence for cancer screening.

They say screening should be measured against the total number of deaths, rather than just deaths from one specific type of cancer.

They also say the risks associated with screening need to be better communicated to patients to allow them to make their own minds up about whether the benefits are worth the possible negative impacts.

Saying that screening "save lives" is misleading, Prasad told BuzzFeed News, because dying in general "has not been proven to be lower".

Screening means testing healthy people for early signs of cancer.

There are three screening programmes in the UK at the moment: bowel cancer, breast cancer, and cervical cancer screening. Men over 50 can ask for a prostate cancer test, but there is no national screening programme for it because the test is not reliable enough, according to Cancer Research UK.

But there's a difference between "saving lives" and lowering rates of cancer.

What the authors are arguing is that while deaths from the specific cancers may be reduced with some kinds of screening, we don't know if the number of overall deaths are reduced.

Epidemiologst Gilbert Welch, who was not involved in the analysis but whose work was cited, told BuzzFeed News that he agrees with the authors.

In his book Less Medicine, More Health (Beacon Press, 2015), Welch writes:

Screening is often promoted as "saving lives." I bet when most people hear that language they think it means that screening helps people live longer. That's what most of my patients think—and that's what I would have thought before I started to study cancer screening.

Most trials measure how screening changes the death rate from the cancer being screened for.

Disease-specific mortality is usually what trials are measured on. And lots of screening methods do decrease the number of deaths from that specific cancer. But when you widen the net and look at all deaths over the same length of time, it's often a different story.

Screening often reduces the death rate from a specific cancer.

One example is the Minnesota Colon Cancer Control Study. It followed 50,000 people over 15 years, and found that by screening for blood in the stool, deaths from colon cancer could be reduced by 33%.

In 2013, 30 years after the initial screening, the study's researchers published a follow-up. They found that in the screened group, 2% of people had died from colon cancer. In the unscreened group, 3% had (as is illustrated in the graph below). So the screening worked, right?

But when you look at overall deaths, they can remain the same.

Well, if you look at how many people in each group died, for any reason, over the same time, the numbers are exactly the same. You can see it on the graph below: The two curves, one for the screened group, one for the unscreened group, are on top of one another.

Screening might "save lives" from one disease, but not "save lives" overall.

Screening doesn't help us deal with all kinds of cancer.

Some people who are screened will still die from cancer, because there are some cancers that can't be stopped. In his book, Welch compares dealing with different cancers to trying to keep different animals in a pen:


The turtles aren't going anywhere. They are the indolent, nonlethal cancers. The rabbits are ready to hop out at any time. They are the potentially lethal cancers, cancers that might be stopped by early treatment. Then there are the birds. Quite simply: they are already gone. They are the most aggressive cancers, the ones that have already spread by the time they are detectable, the ones that are beyond cure.

Screening can only help with the rabbits. The turtles don't need help; the birds can't be helped.

Screening can lead to overdiagnosis and unnecessary treatment.

One way screening causes harm is by detecting cancers that would have never done damage, because those patients end up going through medical procedures that are unnecessary and potentially bad for them.

This is how some types of cancer can counterintuitively end up causing more damage if they are caught early on. It is the turtles – cancers that were never really going anywhere in the first place – that cause the problem of overdiagnosis.

In their BMJ analysis, Prasad and his co-authors use the example of PSA, a test for prostate cancer. They say PSA tests lead to lots of false positives, which contribute to 1 million people undergoing prostate biopsies every year. This can have bad consequences:

Prostate biopsies are associated with serious harms, including admission to hospital and death. Moreover, men diagnosed with prostate cancer are more likely to have a heart attack or commit suicide in the year after diagnosis or to die of complications of treatment for cancers that may never have caused symptoms.

So what should be done about it?

To find out whether screening really does "save lives", Prasad and colleagues write, "we need trials that are 10 times larger" and better suited to measuring all deaths, not just cancer-specific ones.

But in the meantime, Prasad and his colleagues say the risks of screening need to be better communicated to patients to help them make up their own minds.

"Screening is a discussion," he told BuzzFeed News. "It is not a box to check off at a certain age."

Kelly Oakes is science editor for BuzzFeed and is based in London.

Contact Kelly Oakes at kelly.oakes@buzzfeed.com.

Got a confidential tip? Submit it here.