Reporting data analysis methods in high impact respiratory journals

Abstract

Background: Data analysis methods play an important role in respiratory research. However, there has been no comprehensive study contrasting the current respiratory journals with other prominent medical journals in their use of statistics. Objective: To evaluate the complexity of data analysis methods in high impact respiratory journals and to compare the statistical reporting in these respiratory articles with reports published in other eminent medical journals. Methods: We analyzed a total of 160 papers published in 2015 in European Respiratory Journal, American Journal of Respiratory and Critical Care Medicine, Chest and Thorax; and 680 papers published between 2007 - 2015 in other medical journals including Lancet and New England Journal of Medicine (NEJM). We classified the articles according to their study design and reviewed the main strategy in the analysis of the primary research question. We also assessed the frequency with which the articles reported various statistical methods. Subsequently, we evaluated differences between the journals. Results: The main findings of the study are: 1) multivariable methods were more often used in respiratory journals in comparison to other medical sub-fields, 2) traditional ways of testing statistical significance were widely used in respiratory articles, 3) statistical procedures were usually extensively described, and 4) the respiratory journals did not display different profiles in their statistical content. Conclusions: Readers of the high impact respiratory journals need to possess a substantial level of statistical expertise if they wish to critically evaluate the design, methodology, data analysis, and interpretation of the findings published in these journals.