Charan Jaykaran, Saxena Deepak, Yadav Preeti, Kantharia N D
Department of Pharmacology, Govt. Medical College, Surat, Gujarat, India.
Indian J Psychol Med. 2011 Jul;33(2):141-4. doi: 10.4103/0253-7176.92062.
Quality of methodological and statistical parameters published in various Indian Medical Journals are usually debated in terms of appropriateness. Aim of this manuscript is to analyze and validate efficacy studies on antidepressant drugs published in Indian Medical Journals regarding quality of various methodological and statistical parameters used.
An in-depth review of Efficacy studies on Antidepressant published in Indian Medical Journals was conducted. Articles were retrieved from all possible sources like Published Journals, downloaded through Pubmed, MedInd, Opengate, Medknow, and Cochrane. These studies were analyzed and validated for various methodological and statistical parameters. Descriptive statistics for various parameters with confidence interval (CI) is also reported herewith.
Of 32 articles reviewed, primary and secondary endpoints and sample size calculation was reported in only one article (3.1%; 95% CI, 0.5 to 15.7%). Power and CI was not reported in any of the reviewed articles. Inclusion and exclusion criteria were reported in 9 (28.3%; 95% CI, 15.5 to 45.3%) articles. Information regarding randomization was mentioned in 12 (37.5%; 95% CI, 22.9 to 54.7%) articles, whereas reporting of blinding or open label status of the study was mentioned in 21 (65.6%; 95% CI, 48.3 to 79.5%) articles. Assumptions of statistical tests were not reported in any of the reviewed articles. Statistical tests were used in only 14 (43.7%; 95% CI, 28.1 to 60.6%) articles of which 70% of were inappropriate. All studies were underpowered for small and medium effect size.
Quality of reporting of methodological and statistical aspects of antidepressant efficacy studies published in Indian medical journals are poor and hence makes difficult to conclude on issues related to validity of these studies.
各类印度医学期刊上发表的方法学和统计学参数的质量通常在适当性方面存在争议。本手稿的目的是分析和验证印度医学期刊上发表的关于抗抑郁药物疗效研究中所使用的各种方法学和统计学参数的质量。
对印度医学期刊上发表的抗抑郁药物疗效研究进行了深入回顾。文章从所有可能的来源检索,如已发表的期刊,通过PubMed、MedInd、OpenGate、Medknow和Cochrane下载。对这些研究进行了各种方法学和统计学参数的分析与验证。同时还报告了各种参数的描述性统计及置信区间(CI)。
在32篇被审查的文章中,仅有1篇文章(3.1%;95%CI,0.5至15.7%)报告了主要和次要终点以及样本量计算。在任何一篇被审查的文章中均未报告效能和CI。9篇文章(28.3%;95%CI,15.5至45.3%)报告了纳入和排除标准。12篇文章(37.5%;95%CI,22.9至54.7%)提到了随机化信息,而21篇文章(65.6%;95%CI,48.3至79.5%)提到了研究的盲法或开放标签状态报告。在任何一篇被审查的文章中均未报告统计检验的假设。仅14篇文章(43.7%;95%CI,28.1至60.6%)使用了统计检验,其中70%不恰当。所有研究对于中小效应量的检验效能均不足。
印度医学期刊上发表的抗抑郁药物疗效研究在方法学和统计学方面的报告质量较差,因此难以就这些研究的有效性相关问题得出结论。