Schizophrenia Daily News Blog

Reporting of safety for medications

Safety reporting in randomized trials of mental health interventions.

Papanikolaou PN, Churchill R, Wahlbeck K, Ioannidis JP.
Am J Psychiatry. 2004 Sep;161(9):1692-7.

Prior to a medication being offered to the public there are scientific procedures known as randomized controlled trials (RCT) that are used to test the safety and effectiveness of a medication (Click here for more information about different types of clinical trials). In psychiatry, it is unclear of how adequate the reporting of safety information is in publications of randomized trials. As such, the authors for this study explored whether the reporting of safety information, including withdrawals due to toxicity, clinical adverse events, and laboratory-determined toxicity, is neglected in randomized controlled trials of therapeutic and preventive interventions.

The authors randomly selected 200 entries from a European registry known as PsiTri which holds information about controlled trials across the field of mental health. In 2002 this registry contained 16,504 controlled trials. For this study, of the 200 randomly picked entries, 142 were randomized trials of medications or other types of interventions. They analyzed these entries for adequacy and relative emphasis of their content on safety issues.

They found that majority of the eligible trials were for medications. Among these medication trials, only 21% had adequate (as defined by their qualitative and quantitative measures) reporting of clinical adverse events and only 16% had adequate reporting of laboratory-determined toxicity, while 32.0% reported both the numbers and the reasons for withdrawals due to toxicity. They also found that medication trials devoted 1/10 of a page in their results sections to safety, and 58% devoted more space to the names and affiliations of authors than to safety. According to the authors, none of non-medication trials had adequate reporting of clinical adverse events or laboratory-determined toxicity. They also found a neglect of safety information in long-term trials and less space had been given to safety in trials conducted in the United States. On a more encouraging note, schizophrenia trials were found to devote more space to safety than trials in other mental health areas.

Overall, the authors suggest that safety reporting is largely neglected across trials of mental-health-related interventions. They suggest that this neglect of safety in mental health-related trials is similar to what has been reported for other medical fields including HIV therapy, hypertension in the elderly and nonsteroidal anti-inflammatory drugs for rheumatoid arthritis. The results from this study are discouraging since many patients can be unaware of adverse effects of medications even if it is rare might want to know the potential toxicity of various medications. Yet, the study highlights the need for further development and reporting of evaluation tools that can appraise safety reporting in specific categories of trials.

However, there are several limitations to keep in mind about this study. Firstly, from these results it is difficult to decide whether the lack of safety information in a published article was because such information was never collected or whether it was collected but not reported. Secondly, qualitative and quantitative factors used to determine adequacy can be different based on the investigators. Even though this study used previously validated tools, other criteria for adequacy may have found different results. Another limitation is that it is not easy to make attribution and causal claims in behavioral, psychological, or social interventions, which might account for the underreporting in these cases. Fourthly, the registry that was used does not contain all trials and most samples were small (which limits power for safety). Finally, mental health outcomes by their very nature are an important part of benefits and toxicity; and as such these effects may be difficult to tease apart.

This study was supported by a concerted action grant from the European Commission Quality of Life Programme.

Click here to find this article on PubMed

Author: Farzin Irani