By Clinical Informatics News Staff

October 6, 2016 | A group of stakeholders recently published a list of recommendations in BMJ to improve adverse event (AE) reporting in clinical trial publications (DOI: 10.1136/bmj.i5078). The authors, members of Medical Publishing Insights & Practices (MPIP), a partnership among pharmaceutical companies and the International Society for Medical Publication Professionals, hoped “to provide recommendations intended to supplement existing guidelines and address key challenges when reporting adverse event data in clinical trial publications.” The list of five recommendations for improvement include Identifying and communicating the most clinically relevant drug AE data as part of a comprehensive safety profile; reporting timing, frequency, duration, and other potentially relevant descriptors; and more.

The first recommendation on the collective’s list is to identify and communicate the most clinically relevant AE data as part of a comprehensive safety profile. The intent of the clinical relevance recommendation is to broaden AE reporting beyond what is mandated by regulators and to leverage the clinical experience and expertise of physician investigators to judge which AEs should be highlighted. Three prominent factors that stand out are: death, serious AEs as defined by the US FDA, and AEs that led to the discontinuation of a trial agent. The authors also state that the “clinical relevance” filter should be applied against a background of comprehensive reporting of AEs in the body, data tables, and supplemental section of the published paper.

A second recommendation made by the writers is that clinical trial publications should report timing, frequency, duration, and other potentially relevant descriptors when clinically appropriate. Time and frequency are especially important, according to the paper, “given that aggregated metrics often fail to state when an adverse event was observed, how many times it was reported in individual patients, or how long it lasted.” Some AEs are not considered serious, but are nonetheless important to report. The article gives an example of a headache being an AE. Noting one headache that lasts an hour or so is a vital distinction when compared to perpetual headaches over a week. Some clinical trial publications might write up both instances under the nonspecific category of “headache.” A focus on solidifying the reports of such factors as time, frequency, and duration should enable clinical trial publications to provide accurate statistics when listing AEs. In addition to this, the writers also clarify that “where feasible, it might be more informative to provide relative risks of relevant adverse events, rather than simply reporting incidence rates, as a means of better conveying the likelihood of a patient experiencing these events.”

Using statistical analysis for clinically relevant AEs where appropriate is the third recommendation on the list. The authors say, “While formal statistical analyses of adverse event data can be useful, most clinical trials (with the exception of those measuring a specific adverse event as a primary or co-primary endpoint) are not designed with sufficient power to make definitive conclusions about adverse events.” The writers go on to offer four principles to guide statistical analysis when AEs were not included in the primary endpoints. Trial reports should specify numerators and denominators; report confidence intervals around absolute risk differences; share the underlying rationale and methodology for any analyses in the statistical analysis plan and in the publication; and report all formal statistical tests, even if not statistically significant, to avoid selective reporting and because the results may still be clinically informative.

The final two recommendations are to avoid the use of overly general text descriptions for AEs, including in abstracts; and discuss AE findings in the broader context of available evidence. The writers argue that clinical trial publications too often use broad generalizations such as “safe and well tolerated” or “no clinically significant or unexpected adverse events” to summarize the overall safety profile. They then list a few recommendations for alternative text descriptions. The reason for discussing AEs findings in a broader context is because individual trial results rarely provide a complete representation of a drug’s likely AE profile.

The conclusion of the article stresses that there is room for improvement in the realm of clinical trial publications. The writers of the article wish for nothing more than to improve the transparency, clinical relevance, and credibility of adverse event reporting.