The number of papers published in Indian medical journals has increased rapidly during the last decade. Hassan et al. wished to assess the quality of the published papers’ study design, statistical analysis, results presentation, and the interpretation of the results. All original research papers published during 2003 and 2013 in the ten Indian journals with the highest impact factors were reviewed. The main questions were about the type and frequency of study design, errors in study design, statistical analyses, and implementation of CONSORT checklist in randomized clinical trials.
From 2003 to 2013 the proportion of erroneous statistical analyses was about the same 25% (80/320) in 2003 and 22.6% (111/490) in 2013. Significant improvement was seen in the proportion of papers using statistical tests, from 42.5% (250/588) to 56.7 % (439/774). The proportion of errors in study design decreased from 41.3% (243/588) to 30.6% (237/774). The number of randomized clinical trials remained low, 7.3% (43/588), most having statistical errors (41 papers, 95.3%). The majority of the published studies were retrospective both in 2003 79.1% (465/588) and in 2013 78.2% (605/774). Major improvements were seen in results presentation, 82.2% (263/320) as compared with 66.3% (325/490), and interpretation, 32.5% (104/320) as compared to 17.1% (84/490).
The authors conclude that Indian medical research has made progress regarding defects in study design but that there remains ample room for improvement regarding the statistical analysis.
Hassan S, Yellur R, Subramani P, Adiga P, Gokhale M, Iyer MS, Mayya SS. Research Design and Statistical Methods in Indian Medical Journals: A Retrospective Survey. PLoS One 2015;10:e0121268