Bias against negative studies in newspaper reports of medical research

--To assess if the reporting of controversial medical journal articles by newspapers reflects the existence of a bias against negative studies (those showing no effect), we compared the rates of newspaper reporting of two studies, one negative and one positive, published back-to-back in the March 20, 1991, issue of JAMA. Both studies analyzed an area of public health concern, radiation as a risk for cancer.

--Seven computerized on-line databases were screened for daily newspapers published in North America during the week following JAMA's publication of the two studies. These databases had full-text access to 168 daily newspapers. Newspapers identified with reports of the two studies were analyzed for length and quality of the reports.

--Seventeen newspapers, publishing 19 reports on the two studies, were identified. Nine reports were dedicated solely to the positive study and 10 reports covered both studies. None of the reports were dedicated to the negative study only. In reports covering both studies, the mean length of the positive reports was significantly longer than the mean length of the negative reports (354 +/- 181 words vs 192 +/- 178 words; P = .04). The mean quality score of the positive reports was significantly higher than that of the negative reports (10.1 +/- 3.4 vs 5.9 +/- 4.9; P = .02).

--The number, length, and quality of newspaper reports on the positive study were greater than news reports on the negative study, which suggests a bias against news reports of studies showing no effects or no adverse effects.