Publication Bias Found in Ischemic Stroke Trials

Trials showing harmful results tend to be published more slowly

THURSDAY, Sept. 28 (HealthDay News) -- Clinical trials in ischemic stroke that indicate harmful outcomes are substantially underreported, with a trend toward lags in time to publication compared with trials showing beneficial results, according to a report in the Sept. 26 issue of Neurology.

David S. Liebeskind, M.D., from the University of California Los Angeles Stroke Center, and colleagues identified 178 acute ischemic stroke trials conducted between 1955 and 1999 that included 73,949 subjects and evaluated 75 pharmacologic or non-pharmacologic interventions.

The researchers said non-publication bias was evident because a greater proportion of harmful outcomes were in unpublished studies than in published trials (75 percent versus 6 percent) and because of underreporting of smaller, non-beneficial studies in acute stroke. Moreover, they found that non-beneficial studies were slower to proceed to publication, and that there was a longer delay to publication for non-beneficial corporate pharmaceutical-sponsored trials.

"Publication bias is evident in the acute stroke research literature, supporting the need for prospective trial registration," Liebeskind and colleagues conclude.

Several of the study authors have taken lecture fees from, or been advisors for, drug companies.

Abstract
Full Text (subscription or payment may be required)

Physician's Briefing