Scientists increasingly use group-randomized trials to test interventions when it is impractical or impossible to randomize individual participants. But the quality of these trials has not been examined in the cancer field.
To determine whether group-randomized oncology trials are adequately designed and executed, David M. Murray, Ph.D., of The Ohio State University in Columbus and colleagues identified and examined 75 studies that used group-randomization and were published between 2002 and 2006 in the field of oncology.
The researchers found a substantial number of problems with the trials. For example, only 18 of the studies (24 percent) reported information about how they calculated the necessary sample size. Additionally, just 35 studies (45 percent) used only the analytical methods that Murray and colleagues considered appropriate for a group-randomized trial. Many of the trials relied on analyses that ignored the group randomization altogether.
“Given that 88 percent of these articles also reported statistically significant intervention effects, there is reason to be concerned that many of these reported effects may be type I errors. Continued reporting of spurious effects misleads investigators and policymakers and slows progress toward control and prevention of cancer,” the authors write.
In an accompanying editorial, Timothy R. Church, Ph.D., of the University of Minnesota School of Public Health in Minneapolis writes, “This disjuncture of health research and statistical practice has consequences that go far beyond academic niceties. The initiative to improve the health of all individuals depends heavily on the appropriate interpretation of well-designed studies, an activity that depends in turn heavily on the close marriage of researchers and statisticians.”-Journal of the National Cancer Institute
