Toward the publication of every clinical trial

In 1980, researchers at the Nottingham University Hospitals ran a clinical trial testing whether lorcainide — a class 1C anti-arrhythmic drug — could be used to treat people after a heart attack. After six weeks on the drug, they found that the drug was effective at restoring the heart to a normal rhythm. However, nine of the 49 people who had received lorcainide died compared to one person in the placebo group. The results were not published, although one of the researchers, John Hampton, later wrote that they tried to publish, but multiple journals rejected their manuscript (1).

Unfortunately, this example of publication bias — the selective publication of certain studies based on their results — endangered the lives of future heart attack patients. Several years later, investigators running the Cardiac Arrhythmia Suppression Trial (CAST) found that giving people similar class 1C anti-arrhythmic agents also resulted in more deaths in the treatment groups (2). It’s possible that publishing the results of the first trial could have saved their lives.

Of course, not every example of publication bias is a matter of life and death, but there are additional reasons that the results of every clinical trial — no matter their results — should either be published or reported in an easily-accessible platform. First and foremost, publishing the results fulfills a reciprocal informal agreement between the investigators and patients who gave their time and undertook risks to participate, and it also ensures that the investigators meet the ethical obligation to honor the patients’ informed consent. Some consent forms directly state that the results will be published or that patients may expect that the research will add to scientific knowledge. Without publishing the results, it’s more likely that the work will be forgotten. This can be upsetting to many trial participants, as studies show that many choose to participate in order to help future patients (3). In a qualitative study led by researchers at the University of British Columbia, one patient said, “If you don’t publish … then how is it to be paid forward to help other people?” (4).

Publishing the results of clinical trials also allows current patients to make informed decisions about their care based on the most up-to-date results. In 2004, the medical biostatistician Alessandro Liberati wrote in the BMJ about his experience as a multiple myeloma patient trying to decide whether to receive a second autologous stem cell transplant to improve his condition (5). He found that there had been four randomized controlled trials that could shed light on this question — but none of them had been published. He wrote, “As a patient I felt even more strongly about what I’ve been fighting for throughout my career. Research results should be easily accessible to people who need to make decisions about their own health. … Why was I forced to make my decision knowing that information was somewhere but not available?”

Four years before Liberati’s piece, in 2000, the FDA mandated that the NIH launch ClinicalTrials.gov with the goal of providing transparent information about clinical trials to the public. In 2007, the FDA added an amendment that requires sponsors to report all results to the site within one year of trial completion. Yet, compliance is still an issue. In January 2025, the latest estimate showed that 76.8 percent of trials do not report results within 12 months (6). In terms of publications in journals, the data show that the rates are lowest for clinical trials with negative results. A study published in January 2025 revealed that the percentage of negative trials published has increased modestly in the last 20 years — from 27.6 percent to 37.4 percent (7).

To make significant strides in reporting and publishing the results of all clinical trials, greater buy in from scientists, industry sponsors, and journal reviewers will be necessary. Perhaps the next 20 years will show a different story.

References