从循证医学角度分析中华眼科杂志已发表临床疗效研究论文的质量

目的探讨我国眼科领域的临床疗效研究水平,为循证医学(EBM)研究提供可靠证据。方法按照EBM标准,对中华眼科杂志1983—2002年间,发表的全部涉及临床疗效的论文进行检索和分析。结果临床疗效研究论文共152篇;其中属于随机对照试验35篇,临床对照试验8篇,共计43篇,占临床疗效研究论文的28．1％,且在随机分组、对照设置、盲法、基线资料、随访、失访、并发症报告及统计分析等方面均存在较多问题。在属于对照实验的论文中,有10篇(23．3％)论文描述了随机或半随机分组的具体方法;采用安慰剂对照和开放对照的文章分别是3篇(7．0％)和9篇(20．9％);盲法研究性论文7篇(16．3％);有随访结果,并提供较为详细随访资料的论文25篇(58．1％);未见分析混杂、偏倚、干扰和沾染等因素影响临床疗效研究质量的论文;未见在项目设计时进行了样本含量估算的论文;15篇(34．9％)论文中的部分或全部数据未标明显著性检验方法,仅有P值。结论我国眼科领域的部分临床科研设计方法不够严谨,从而影响研究结果的真实性和可靠性。重视对眼科医师进行临床科研设计方法的培训,严格控制研究质量,对提高我国的临床疗效研究水平至关重要。

Analysis of the quality of clinical trials about therapeutic research using the standard of evidence-based medicine

OBJECTIVE: To investigate the quality of clinical trials of ocular therapeutic efficacy in China in order to provide the facts to support the Evidence-Based Medicine (EBM) and improve the clinical research trials. METHODS: All the articles published in Chinese Journal of Ophthalmology from 1983 to 2002 emphasizing random control trial and clinical control trial were evaluated as per EBM standard. RESULTS: The total number of articles for clinical therapeutic efficacy published in the past 20 years were 152, of which 35 articles (23.0%) were random control trials, and 8 (5.3%) articles were clinical control trials. The disparity of quality of research was found as follows. The method of random/half random was described in 10 articles (23.3%). Placebo-control and open-control were reported in 3 (3/43) and 9 articles (20.9%), respectively. Blinded, double-blinded, and single-blinded research was adopted in 7 (16.3%), 3 articles (7.0%), and 4 articles (9.3%), respectively. Ten articles (23.3%) reported strict diagnostic standard including inclusion and exclusion standards while 30 articles (69.8%) didn't use exclusion criteria. Follow-up visits from a few months to years was found in 30 out of 43 articles, of 25 articles (58.1%) provided detailed follow-up visit data; 19 articles (44.2%) didn't supply any information of the cases lost follow-up, only 7 articles (16.3%) reported the cause and process of the cases lost follow-up. Adverse events were not consistently reported in the articles. Not a single article estimated sample size when the study was designed. Most articles included small or medium size samples, while lacking of multi-center, large sample, randomized, double-blinded and placebo-controlled clinical trials. Statistical significance for either part or all data except P value was not indicated in 15 articles (34.9%). The indexes assessing the clinical significance of clinical therapeutic efficacy include relative risk reduction, absolute risk reduction, number needed to treat, however, these indexes were not used in all the articles. CONCLUSIONS: In summary, the clinical trials related to ocular therapeutic efficacy published between 1983 and 2002 were not very well designed and the accuracy and the reliability of the results were greatly affected. Therefore, it is imperative to improve the quality of the research by enhancing the design and analysis of the research in future.