Published by the Students of Johns Hopkins since 1896
September 25, 2020

Some clinical trials improperly report outcomes

By CHRIS REINHARDT | February 21, 2019



Published clinical trials often report on trial outcomes they did not intentionally study.

Clinical trials are essential for testing new medications because they let researchers know whether new medications, from vitamins to vaccines, are safe and effective. However, a new study published in Trials shows that researchers aren’t properly reporting the outcomes of their clinical trials. This may make a new treatment appear to be safer or more effective than the product actually is.

A group called the Centre for Evidence-Based Medicine Outcome Monitoring Project, or COMPare, examines switched outcomes in clinical reports — whether a clinical trial reports the outcomes they set out to measure. For instance, one group may examine blood sugar levels after sixth months of taking a new medication but will instead report on blood pressure. This is important because clinical trials may gather a lot of data, creating more areas for false-positive outcomes to occur. In turn, clinical trials may report on certain outcomes they didn’t specify before the clinical trial or neglect to report upon their original primary outcomes, in turn skewing the safety and efficacy of the treatment being testing.

COMPare analyzed 67 clinical trials published in the top five medical journals: The New England Journal of Medicine, The Lancet, JAMA, BMJ and Annals of Internal Medicine. They compared the trial report with either a protocol or pre-trial registry to determine if outcomes had been switched. When they noticed discrepancies, they sent a letter to the journal to correct the record. COMPare also tracked whether journals published these letters.

The group found that only nine trials in 67 perfectly reported their outcomes. The other 58 trials reported an average of 58.2 percent of their original outcomes. Additionally, each trial added an average of 5.3 outcomes without declaring them. COMPare found that certain journals fared better than others: 96 percent of JAMA trials reported the primary outcomes correctly, while only 44 percent of trials in Annals were correctly reported according to Science. Eighteen of the letters COMPare sent to journals were published, while 32 were rejected by the journals. The remaining eight articles remained unpublished after four weeks.

COMPare is quick to note that adding or changing outcomes in a trial report isn’t wholly unethical or unjustified — rather, novel outcomes should be declared and explained within the trial report. They believe this paints a more realistic picture of the trial’s findings. 

Bella Radant, a senior Biomedical Engineering student with four years of research lab experience, agrees with their findings. She points to a lack of follow up to ensure clinical trials are being reported fairly. 

“Silently adding outcomes is dishonest because you’re only reporting positive outcomes, not negative ones, making your study looking better than it is,” Radant said in an interview with The News-Letter.

While the study points to an important issue within clinical trial reporting, Radant points out that they could have better addressed disparities between journals. 

“I think there are a lot of flaws with the journal system and the way results are processed, published and reported,” she said. 

She noted that the study only looked at the top five medical journals, ignoring pay-to-publish journals altogether. She believes that the reports in those journals may present more switched outcomes, highlighting the importance of peer-reviewed, meticulously edited journals. 

Radant also cites pressure to “publish or perish” as a contributor to data manipulation. The COMPare study, then, only points to a deeper problem: how pressure for successful results leads to unethical reporting and publishing. Addressing this is no easy feat, but something she believes is a necessary shift. However, careful review by journals can help weed out trial reports with skewed outcomes, therefore creating a new pressure to publish a fairer depiction of clinical trials. 

Comments powered by Disqus

Please note All comments are eligible for publication in The News-Letter.

News-Letter Special Editions