News editors should have taken a hint from the editors of the Journal of the American College of Cardiology who gave this research space for only a brief report, not thinking it worthy of a full-length article. The story exaggerates the importance and mischaracterizes the results of the study.
It is interesting that our breath can contain clues to the condition of our hearts. And the idea of a breath test complementing blood and other tests for heart failure is fascinating. But by highlighting “100% accuracy,” this story goes beyond simply describing a tantalizing concept and misleads readers about the status of this line of research. It muddles the real matter, proper diagnosis of patients with suspected heart failure, with the separate question asked in this research: are trace components in the breath of patients with known heart failure distinct from those with other known heart problems? In the end, the story merely reaches an obvious conclusion that better tests would be better… without really assessing whether this sort of breath test might really be better.
This study shares the flaw of many diagnostic accuracy studies – spectrum bias; when you enroll groups of “known to have” and “known not to have” the test performs better.
The value of testing depends on the context, but this research is too preliminary to have any such context. There is not yet any data that could help predict whether this sort of breath test would help or harm patients. So the story is not quite so “simple” as the headline suggests.
This potential test is in the early stages of development. Almost everything about it, from how well it actually works, or how it compares to alternatives, has yet to be studied. The story does quote a researcher saying the test is “theoretically cheap.”
The story merely speculates about potential advantages without describing what they would be. The potential advantage would be more accurate diagnosis, but that was not studied here (no comparison). It is too early to say where such a test might be used, in an ER or in a primary care clinic, or how it would influence treatment. Would it help speed up treatment of some patients who would benefit from rapid intervention or would it avoid hospitalizing people who don’t need aggressive care? The story implies benefits without specifying what they might be.
The story misses the mark here. It quotes a researcher saying “there is no risk involved.” Even if there is no direct physical harm from breathing into a testing device, there could be substantial harm caused by actions based on the test results… including unnecessary hospitalizations and treatments or, conversely, delays in potentially beneficial treatments.
The story does tell readers that this line of research is at an early stage, but then it reports the test had “100 percent accuracy,” without making clear that the study did not test the main question doctors and patients want answered. The headline and lead highlight the need to correctly identify patients with heart failure, when their primary symptoms could be caused by a variety of conditions. However, this small study compared breath samples from patients known to have heart failure with patients known to have other heart problems. There was nothing in this very preliminary study to address the real question: can a breath test given to people suspected of possibly having heart failure tell the difference between those who do and those who don’t?
As the researchers noted, this study was meant to help prove that the concept is worth studying further, but the story suggests that the study showed the test was accurate at diagnosing suspected heart failure patients, which is not what was actually tested.
There’s no disease mongering of heart failure. But in the way the story addresses early research about this test, it smacks of a subset of disease-mongering. The story cites the study authors’ reference to suspicion of heart failure as the most common reason American patients are admitted to a hospital. But it is far too early to say whether this test might ever be relevant to most, or even any, of those patients. Is it disease mongering to say that most heart failure patients confront diagnostic challenges of the type raised by these researchers? Or are most cases pretty clear – meaning that most heart failure patients aren’t really threatened by uncertain diagnoses? These are questions that could have been addressed, weren’t, and therefore, the story made the problem seem bigger than it might really be.
There are quotes from an independent expert. The journal that published the research reported that it was funded by state and federal grants, and that the researchers did not report any disclosures relevant to this study.
The story does list some of the standard tests for heart failure. It also notes that the research is too preliminary to tell whether this sort of breath test would be any better. However, it is likely that readers will get the impression that the new test is an advance, even though there have not been any experiments done to compare different methods.
The story could have compared with the B-type Natriuretic Peptide (BNP) blood test (an inexpensive blood test).
The story reports that this test is experimental and quotes a researcher saying, “There is much more work that needs to be done to get it to the point where it would become widely available.”
This sort of breath test would be a new addition to heart failure diagnosis, if it works.
The story does not appear to rely on a news release.