Margaret McCartney: Getting the abstract in perspectiveBMJ 2018; 361 doi: https://doi.org/10.1136/bmj.k2499 (Published 12 June 2018) Cite this as: BMJ 2018;361:k2499
- Margaret McCartney, general practitioner
Follow Margaret on Twitter at @mgtmccartney
There’s a conference on! Who could have guessed? On 1 June we had the front page headline, “Cancer blood test hailed as ‘holy grail.’”1 On 2 June we had, “NHS cancer patients ‘failing to be correctly monitored.’”2 On 3 June, joy of joys: “GPs handing out too many antibiotics harms cancer survival chances.”3 In case that last one hasn’t treated my colleagues’ hypotension enough, here’s a choice quote: “People see their GP and the GP thinks, ‘Oh my goodness it’s a cancer patient, they need antibiotics.’”3
Let us regroup. This is a meeting of the American Society of Clinical Oncology (ASCO), where research is transmitted to the world through press briefings, abstracts, and talks. We live in a world where completed, peer reviewed research is shared ahead of publication under embargo, to allow journalists time to check facts and produce accurate coverage. Yet works in progress are released to global fanfare, having been presented only in abstract form, without peer review—making it hard to check the researchers’ workings and uncertainties.
Let’s take the cancer blood test “holy grail” article,1 which is a story of several abstracts.4567 A blood test, performed prospectively in people with cancer and people in a control group, can find a DNA marker with a high degree of sensitivity in patients with cancer but has varying, and not very high, specificity (from 56% upwards). This was not a study testing any intervention to see whether deaths could be delayed, and many of the researchers are employed by, and own shares in, the company making the test. As such, one newspaper’s suggestion of the test being available on the NHS “within five years”1 seems rather optimistic.
Works in progress are released to global fanfare, having been presented only in abstract form
Then there’s the story of NHS cancer patients “failing to be correctly monitored.”2 How long should people who have been treated for different types of cancer be followed up by an oncologist? It’s a great question, but we have only an abstract summarising research on a North American database. The research seems to have found wide variation in how long people are followed up for in the United States—in a very different healthcare system from the NHS—and that some cancers are more likely to recur than others. It is clearly not a study showing that changing current practice will decrease the risk of death.8 However, Joyce Robins, of the UK advocacy group Patient Concern, is quoted as saying that it is “terrifying that cancer patients are being abandoned like this.”2
Moving on, we have the story of how “GPs handing out too many antibiotics harms cancer survival chances.” This last one is based on a retrospective data analysis that showed an association between patients with metastatic disease being treated with antibiotics and the earlier progression of disease and death.9 The abstract does not mention who gave the antibiotics or why: perhaps there are other reasons why patients who receive more antibiotics do less well.
All of this research is interesting and has the potential to improve healthcare. A proposal from the Academy of Medical Sciences,10 to “traffic light” press releases to classify them in terms of peer review and how relevant the research is to people, may help; but my question is whether it is useful for the public or professionals to publicise or use press releases of conferences for conference abstracts at all.
Competing interests: See www.bmj.com/about-bmj/freelance-contributors/margaret-mccartney.
Provenance and peer review: Commissioned; not externally peer reviewed.