Proper Reporting of Statistical Parameters in Clinical Trials Published in Indian Medical Journals. Is Inclusion of Statistician Play any Significant Role?

Sir, It is observed that the clinical trials published in various medical journals are poor in reporting of various methodological aspects, including sample size calculation and power. It is difficult to generalize the results of clinical trials having poor reporting of statistics to the normal patient population and conducting these clinical trials raises ethical issues. It is always advisable to take the help of a statistician both during the analysis of data and design of clinical trials. Not associating a statistician while designing clinical trials is associated with the poor reporting of statistical parameters in clinical trials. [3] This study is designed to check the function of statisticians in clinical trials published in five Indian medical journals and see the impact of employing these statisticians on reporting of some important statistical parameters. Clinical trials published in four Indian medical journals were selected for analysis. Statistical parameters selected for comparisons were the following: calculation of sample size, sample size, calculation of power, appropriate statistical tests, significant P value after Bonferroni correction, number of primary endpoints, and calculation of post hoc power of the study. All clinical trials published in four Indian medical journals in 9 years (January 2000 to December 2008) were downloaded. Each author critically appraised these clinical trials for the various statistical parameters on the basis of a predesigned proforma. Information regarding the role of a statistician was seen at authors' section, acknowledgement, and methods section of clinical trials. These trials were also appraised for information on calculation of sample size, even partial calculation of sample size were taken into account. Information regarding the exact sample size was noted in the proforma. Authors also surveyed all clinical trials for information regarding calculation of power before starting of study (design phase). Appropriate statistical tests were evaluated based on the aim of the study, type of data, and distribution of data. These clinical trials were also appraised for methods of adjustment of multiple endpoints and whether the P value is still significant after adjustment of multiple endpoints on the basis of Bonferroni correction. Endpoints are considered as events or outcomes that can be measured objectively to determine whether the intervention being studied is beneficial. We did not include adverse effects as endpoints of trials. Post hoc power of trials was also calculated at 50% difference between the outcomes with the help of G Power software. [4] Discrepancies observed between the authors …