LANGUAGE BIAS IN RANDOMIZED CONTROLLED TRIALS PUBLISHED IN ENGLISH AND GERMAN

Background Some randomised controlled trials (RCTs) done in German-spe aking Europe are published in international English-language journals and others in national German-language journals. We assessed whether a uthors are more likely to report trials with statistically significant results in English than in German. Methods We studied pairs of RCT re ports, matched for first author and time of publication, with one repo rt published in German and the other in English. Pairs were identified from reports fround in a manual search of five leading German-languag e journals and from reports published by the same authors in English f ound on Medline. Quality of methods and reporting were assessed with t wo different scales by two investigators who were unaware of authors' identities, affiliations, and other characteristics of trial reports. Main study endpoints were selected by two investigators who were unawa re of trial results. Our main outcome was the number of pairs of studi es in which the levels of significance (shown by p values) were discor dant. Findings 62 eligible pairs of reports were identified but 19 (31 %) were excluded because they were duplicate publications. A further t hree pairs (5%) were excluded because no p values were given. The rema ining 40 pairs were analysed. Design characteristics and quality featu res were similar for reports in both languages. Only 35% of German-lan guage articles, compared with 62% of English-language articles, report ed significant (p<0.05) differences in the main endpoint between study and control groups (p=0.002 by McNemar's test). Logistic regression s howed that the only characteristic that predicted publication in an En glish-language journal was a significant result. The odds ratio for pu blication of trials with significant results in English was 3.75 (95% CI 1.25-11.3). Interpretation Authors were more likely to publish RCTs in an English-language journal if the results were statistically sign ificant. English language bias may, therefore, be introduced in review s and meta-analyses if they include only trials reported in English. T he effort of the Cochrane Collaboration to identify as many controlled trials as possible, through the manual search of many medical journal s published in different languages will help to reduce such bias.