运动干预阿尔茨海默病患者认知功能元分析文献的方法学与证据质量评价

Quality Evaluation and Improvement of Meta-Analysis Literatures on Cognitive Function of Exercise Intervention in Alzheimer's Disease Patients

目的评估运动对阿尔茨海默病(Alzheimer's Disease,AD)患者认知功能干预效果元分析文献的方法学质量及其结局指标的证据质量,并提出相应建议。方法通过PubMed、Scopus和中国知网等数据库,搜集运动干预AD患者认知功能效果相关的元分析文献,检索时限均为各数据库建库至2020年12月31日。采用AMSTAR2量表评价所纳入文献的方法学质量,并依据GRADE系统对所纳入文献的结局指标进行证据质量分级。结果共纳入9篇元分析文献。AMSTAR2量表评价结果显示,4篇文献方法学质量为低,5篇文献为极低;GRADE评价结果显示,在纳入的17项结局指标中,中级证据3项,低级证据9项,极低级证据5项。方法学缺陷包括:未撰写计划书和提供注册号,文献筛选、数据提取和文献排除理由未详细描述,系统评价实施过程不严谨,未说明纳入研究的资助来源和相关利益冲突。证据质量降级因素包括:设计和实施的局限性,结果不一致,精确度不够,可能存在发表偏倚。结论元分析证据显示,运动对AD患者的认知功能具有一定的改善效果,但元分析文献的方法学质量整体偏低,证据质量相对较低。建议本领域元分析研究者遵循严谨的元分析报告规范,期刊编辑部严格把关元分析报告研究,高校及循证医学中心在政策支持下联合建设注册方案平台,开设元分析理论与实践课程,共同为运动防治AD领域提供更加科学、准确、高质量的临床试验证据。

Objective To provide an overview of meta-analysis on the effect of exercise on the cognitive function of patients with Alzheimer’s disease(AD), and to put forward improvement suggestions. Methods PubMed,Scopus, CNKI, and other databases were electronically retrieved. The retrieval time was from the date of database construction to December 31, 2020. According to the inclusion criteria and exclusion criteria, final articles were selected and relevant literature information was extracted by screening the abstract and full text. The methodological quality of the included meta-analysis was evaluated using AMSTAR2 tool, and the quality of the outcome indicators of the included meta-analysis was further graded according to the GRADE system. Results A total of 9 meta-analyses were included. The AMSTAR2 evaluation results showed that 4 studies were in low methodological quality, and 5 studies were in extremely low quality. The GRADE evaluation results showed that among the 17 outcome indicators included, 3 were intermediate quality indicators, 9 were low ones and 5 were extremely low level ones. The main deficiencies of methodology quality were that no protocol was provided or registered, no details of literature screening, data extraction, and reasons for exclusion, no source of funding and related conflicts of interest for each included literature. In addition, the implementation process of systematic reviews was not rigorous. The main flaws of evidence quality included designable and implementation limitations,inconsistent results, insufficient precision, and possible publication bias. Conclusions The meta-analysis indicates that exercise can improve the cognitive function of AD patients to some extent, but the methodological quality of relevant meta-analysis is relatively low and the quality of the evidence is generally low. It is suggested that the related authors should follow rigorous meta-analysis report specifications. Journal editorial departments should strictly control meta-analysis report. Universities and evidence-based medicine centers build registration scheme platforms with policy support, and offer courses on meta-analysis theory and practice, which will provide a more scientific, accurate and high-quality clinical trials evidence for the field of exercise prevention and treatment of AD.