When medicine & media sugarcoat the true impact of patient harms

Gary Schwitzer is the founder and publisher of HealthNewsReview.org.  He tweets as @garyschwitzer.

Click here to read more stories about real people harmed by misleading messages, or share your own story.

An important paper, “Reporting harms more transparently in trials of cancer drugs,” was published in The BMJ last week. Some excerpts follow, but also additional thoughts about how journalism and public relations (which sometimes are indistinguishable) make the matter worse.

The paper’s lead author is Bishal Gyawali, MD, PhD an oncologist from Nepal currently working as a Research Fellow at Harvard and Brigham & Women’s Medical Center in Boston.

The research examined how cancer drug harms were described in clinical trial results published in five major medical journals in 2016.  They looked at 122 such trials in journals, and found that 53 of them – 43% – used terms that downplayed harms.

The paper points to vague descriptions of harms regarding three cancer drugs:  “Acceptable adverse-event profile….a manageable and mostly reversible safety profile….the tolerability was good overall.” With subjective terms like these, it’s anyone’s guess what “acceptable” levels of nausea or “manageable fatigue” means.

Additionally, Gyawali and his team found these vague descriptors at odds with what they saw in the actual data, where side effects were more common than what the phrases above implied.

The following excerpt is a good summary of what the researchers saw in their broader analysis (parenthetical notes are ours):

Terms used to downplay the harms of cancer drugs, and reasons not to use them

  • Acceptable—Acceptable to whom? Were the patients asked if the toxicities were “acceptable” to them? (Toxicity is defined as the extent to which something is poisonous or harmful.)

  • Manageable—Serious events and deaths can never be considered manageable. Even manageable toxicities incur burden and decrease patients’ quality of life. (In other words, even side effects perceived as manageable by researchers can add up to a miserable day–or week, or month, or year–for a cancer patient.)

  • Feasible—What is the threshold for feasibility of a treatment? Will the mention of “the treatment is feasible” be enough to obtain patient’s consent to a treatment? (In other words, what does it mean when researchers report that it is possible or reasonable to do the treatment?)

  • Favourable toxicity profile—Favourable compared with what? The threshold of enduring toxicities and thus favourability is different from patient to patient.

  • Tolerable or well tolerated—Only the patient can decide whether any side effect is tolerable

  • Safe—Any cancer treatment that has resulted in a treatment related death cannot be considered safe

And here is an excerpt of the paper’s key messages, as laid out by the journal:

  • Vague and subjective terms can lessen the perception of harm and influence decisions about treatment

  • All cancer trials should fully report adverse events and avoid subjective terms

An oncologist who saw a disconnect between what he read and what he saw in practice

Bishal Gyawali, MD

Gyawali emailed me some background on why he pursued this topic.

“I felt it was some sort of deception. To the family of a patient who died due to treatment toxicities, I imagined what would they feel if they read this report that said ‘toxicities were acceptable’ or ‘treatment was safe.’ Further, I felt it was inappropriate for us clinicians to label a toxicity our patients experience as acceptable. My view is that only patients who experience those side effects can say whether it was acceptable or not. This downplaying of toxicities is very harmful because it makes us clinicians believe that a drug is safer than it actually is.”

He felt so strongly about what he found that he posted a Tweetorial to further explain the research he and his colleagues had done.

News & PR messages that make matters worse

Now think about what happens when journalists or marketing people try to translate these kinds of clinical trial results that are described in these ways in medical journals. The vague, subjective terminology serves to minimize the sense of potential harms, while the discussion of potential benefits is often emphasized and exaggerated.

This is another example of the polluted stream of health care information that we’ve been discussing so often on this site, and another example of patient harm from misleading media when you include journal articles under the umbrella of media, as we do. Journals are daily and weekly purveyors of news, courting and thriving on the attention.

Whether a news story or a PR news release adequately explained and quantified the potential harms of an intervention is one of our 10 review criteria. In our explanation of why that harms criterion matters, we state:

Many stories emphasize or exaggerate potential benefits while minimizing or completely ignoring potential harms. They may:

• Fail to mention potential harms.

• Fail to quantify potential harms.

• Fail to describe the severity of potential harms.

• Fail to account for “minor” side effects that could have a significant impact on a patient’s life.

• Rely too heavily on a patient anecdote about safety.

• Rely too heavily on a researcher’s comment that an approach “appears to be safe” – without supporting data.

Step back and look at how many of the journal articles on cancer drug trials soft-pedaled harms. In  Gyawali’s paper, 43% of cancer drug trial reports contained one or more of the vague euphemistic terms for drug safety mentioned above.  In our data, when journalists report on stories with claims about new interventions, 63% of more than 2,600 news stories we’ve reviewed over the past 12 years were graded unsatisfactory on this criterion. That is a terrible reflection on that huge sample of daily news.  Public relations news releases are even worse: 78% of the nearly 600 PR news releases we’ve reviewed were graded unsatisfactory on the harms criterion. This helps you to see the combined amplified effect of the polluted stream of health care research news as it reaches an uninformed public.


We see it not only in media messages about cancer drugs, but in stories about all sorts of health care interventions. Some examples:

Gyawali agrees that journalists become complicit in spreading the misinformation on harms.

Media usually pick up the abstracts/conclusions (from the journal article) to make generalized statements or headlines such as “a new cancer drug is found to be effective in XXX cancer and is totally safe.” No cancer drug is safe, it’s always a trade-off. This understanding is crucial because most cancer drugs provide only modest benefits, very few are truly game-changers. Thus, statements such as “toxicities were manageable” gets propagated to media from scientific papers and then to patients who will have false beliefs of exaggerated benefits and diminished harms.

Readers of these trial reports–physicians, patients, media people–will all trust these generalized statements on toxicities that paint the new drug as having better risk-benefit profile than what actually exists. This will ultimately harm the patients.

I think it’s the responsibility of media to dig deeper and report the true incidence of toxicities rather than copy these generalized phrases.

You might also like

Comments (4)

Please note, comments are no longer published through this website. All previously made comments are still archived and available for viewing through select posts.

Andrew DePristo

November 6, 2018 at 4:09 am

This article makes a big deal out of a commonly understood issue: all drugs have side effects. Aside from the obvious situations like liver damage from acetaminophin, antibiotics are generally considered “safe.” Yet antibiotics can cause severe GI distress (e.g., uncontrollable diarrhea) in some patients and not in others. If the occurrence of severe distress is low, the drug would have to be described as generally safe and well-tolerated. Unless one wants to provide a list of every patient and side effect or a bar chart of side effect vs number of patients, the description is informative. It is up to the physician to interpret the data properly for his/her patients. Bishal Gyawali, MD, PhD needs to do his job better.

    Gary Schwitzer

    November 6, 2018 at 8:47 am


    The paper in The BMJ analyzed descriptions of cancer drugs – not descriptions of acetaminophen or antibiotics.

    And it didn’t make “a big deal” about all drugs having side effects, but, rather, explored the disconnect between how side effects from the cancer drugs in the journal articles in question were described in the papers and what occurred in the trials.

    Finally, I would remind you of our comments policy, which clearly states:

    Comments should primarily discuss the quality (or lack thereof) in journalism or other media messages about health and medicine. This is not intended to be a forum for definitive discussions about medicine or science.

    In the future, please abide by that policy. And your final comment on Dr. Gyawali is bordering on the kind of ad hominem attack that we discourage on this website. What do you know about how he does his job?

    Gary Schwitzer

Bishal Gyawali

November 6, 2018 at 9:16 am

Dear Andrew DePristo,
Thanks for suggesting that I need to do my job better…that’s actually my life motto, to do my job better today than I did yesterday. My job as a doctor is to take care of cancer patients and cancer drugs (unlike acetaminophen or antibiotics) are not safe; it’s always a trade-off because even the benefits from these drugs, despite the side effects, are modest. If you read our paper, you’ll see that the CONSORT statement also cautions against using these generalized terms and that our main point was the disconnect between the side effects profile of the drugs and the use of downplaying terms, which is harmful to patients who need to make delicate decisions. I can email you the PDF if you’d like to take a look. Because we not only raise the problem, we have also suggested some solutions in our paper that doesn’t involve making bar chart as you suggest. I don’t know what your job is but I pray that you’re doing it to the best of your abilities. PS: I have no conflicts of interest to declare. Warm regards, Bishal

Sally Schott

November 6, 2018 at 9:31 am

Somebody doesn’t know what they’re talking about.

And it isn’t Dr Gyawali.