Geographic distribution of trials in Cochrane system­atic reviews

Tags: Poster
Vist G, Walleser S, McLellan F, Ghersi D

Background: When the aim is to provide the evidence base for global recommendations, all the evidence should be sought, without restrictions on language or geography. It is hoped that there are no such restrictions in Cochrane systematic reviews. Objectives: To study whether Cochrane systematic reviews place language restrictions on searches for evidence, and to study the geographical distribution of the trials included in Cochrane systematic reviews. Methods: All new reviews (n=112) published in the Cochrane Database of Systematic Reviews (CDSR) Issue 1, 2009 were included. The search strategy was examined to determine whether any language restrictions existed. Data were extracted from the Table of Included Studies for each review on country for each included trial. Results: One of the 112 new reviews in CDSR had imposed language restrictions to the search. Twenty-four reviews had no included studies, and in 27 reviews the countries of the included trials had not been reported. Seven reviews had included trials from four different continents; 18 reviews included trials from three different continents; and 13 reviews included trials from two different continents. Several reviews included studies from only one continent: nine reviews included studies only from Europe; seven from only North America; three from only Africa; one from only Asia; and three from only Australia. Conclusions: There appear to be few Cochrane systematic reviews that impose language restrictions on searches. We acknowledge the different implications from excluding trials published in different languages and from actively searching for trials in different languages. However, there is room for improvement in the reporting of the location where the included studies were conducted.