“Blood test finds 50 types of cancer” – we had taken this route before |
Sloppy, incomplete, scruffy coverage of screening tests is one of the most clearly identified problems in health reporting – something HealthNewsReview.org has exposed countless times over the past 15 years.
In a press release 22 days ago for public relations – from a “health care company whose mission is to detect cancer early” – study results were touted that, according to the company, the “The ability to detect more than 50 types of cancer with a single blood draw could transform cancer screening as a complement to existing screenings. ”
There’s a lot to unzip from this type of claim, and I haven’t seen any coverage that did this. There may have been some that I haven’t seen.
This week, The Guardian has drawn the ire of many watchers on social media and the email lists I follow.
The Guardian reported, “It correctly identified when cancer was present in 51.5% of cases across all stages of the disease, and only 0.5% of the cases misdiagnosed cancer.” Numbers mean to the patient. The test’s 51.5% sensitivity – its ability to find cancer when cancer was present – means that nearly half of patients with cancer who would have the test done would not detect their cancer this way. In patients with early-stage cancer – the kind one would want to catch with an effective screening test – the sensitivity was much lower. The test had a sensitivity rate of less than 17% for stage 1 cancer and a sensitivity rate of 40% for stage 2 cancer. in the journal article in which the study results were published, tried to downplay the relevance of the sensitivity statistics. But the coverage I saw went around the entire sensitivity issue.
There was also no truly independent perspective in the story. It read like a PR message. Finding healthy skeptics wasn’t difficult on social media either.
Doctor Susan Bewley busted The Guardian on Twitter:
Another British doctor took just one word to respond to the news:
In the US, cancer specialist Dr. Kevin button:
The guard was not alone; You can find dozens of unquestioning stories online.
Let me make that clear: as Dr. Bewley suggests there are potential benefits but inevitable harm to having mass screening done. Any journalist or news organization who writes about screening tests must discuss these tradeoffs.
Since we’ve written so much on the subject, we have an introduction that journalists and the general public should know –
Understand medical tests: sensitivity, specificity, and positive predictive value
Just take a look back at our archive of bad media news about cancer blood tests … universal blood tests … “simple” cancer blood tests. You will see the damage that wrong communication with the public can do.
When I looked up The Guardian’s story, I saw that this week this story was also published:
The Guardian allowed one study author to claim that this “confirms … that drinking coffee protects against severe liver disease”.
Later the story turned around. It took 11 paragraphs – or 365 words – of the Guardian to reveal:
However, the study has limitations, including the fact that it cannot prove that coffee itself reduces the risk of chronic liver disease, while participants were only asked about their coffee drinking habits at one point in time.
Confirmed Coffee Is Protective? Or can’t you prove that coffee reduces the risk? You can’t have both.
Thanks To Source