Randomized controlled trials (RCTs)the gold standard for evaluating medicationsappear to largely under-report or even neglect to report problems involving drug safety, according to researchers supported by the US Agency for Healthcare Research and Quality (AHRQ).
Their study in today’s Journal of the American Medical Association (JAMA) found that, on average,
medical journal articles about drug trials devote only a third of a page to safety informationsuch as side effects of medications and the frequency and reasons for patient withdrawalsroughly the same space they devote to naming the contributors and their affiliations.
"Adequate reporting of drug safety problems is critical because clinical trials have traditionally been the foundation of high-quality, evidence-based medical practice," said AHRQ Director, John M. Eisenberg, M.D., whose agency is leading federal research efforts to improve patient safety.
According to the study’s authors, John P.A. Ioannidis, M.D., of Tufts University School of Medicine and Greece’s University of Ioannina School of Medicine, and Joseph Lau, M.D., of The New England Medical Center, the quality of safety reporting also appears to be inadequate.
The researchers examined safety reporting in 192 randomized drugs trials, each involving a minimum of 100 patients and at least 50 patients in each study arm. More than 130,000 trial subjects were involved overall. The trials tested medications for seven areas: HIV therapy; antibiotic therapy for acute sinusitis; thrombolysis-clot dissolving drugsfor heart attack; non-steroidal anti-inflammatory drugs for rheumatoid arthritis; high blood pressure treatment in the elderly; antibiotic treatment of Helicobacter pylori, a major cause of stomach ulcers; and selective decontamination of the gastrointestinal tract.
The severity of medication side effects and drug toxicity revealed by abnormal laboratory test results, was adequately explained in only 39 percent and 29 percent, respectively, of drug trial reports, and a further 11 percent and 8 percent, respectively, had partially adequate reporting. Also, while the numbers of patients in each trial study arm who had to be withdrawn because of drug toxicity were cited in 75 percent of trial reports, the specific reasons for these discontinuations were given only 46 percent of the time.
According to the authors, major strides have been made in standardizing the collection, analysis and reporting of efficacy data in clinical trials, but efforts to evaluate and improve the quality of analysis and reporting of safety data lag behind.
Details are in "Reporting of the Safety of Medications in Randomized Trials is Neglected: An Evaluation of Seven Medical areas," published in the January 24, 2001, issue of JAMA.