儿童针灸领域系统评价的方法学质量和偏倚风险评估

The evaluation of methodology quality and risk of bias of the systematic reviews of acupuncture in children

目的本文采用AMSTAR-2和ROBIS工具评价纳入儿童针灸领域系统评价/meta分析的方法学质量和偏倚风险,探讨两种评价工具的特点与不同,为合理选择AMSTAR-2和ROBIS工具提供参考依据。方法计算机检索PubMed、EMBASE、Cochrane Library、CBM、CNKI、维普和万方数据库,收集儿童针灸领域的系统评价/Meta分析,检索时限均为建库至2018年1月。两名评价员分别独立采用AMSTAR-2和ROBIS工具评价纳入研究的方法学质量和偏倚风险。结果共纳入66篇系统评价/Meta分析,CDSR 7篇、国际期刊34篇、中文期刊25篇。①AMSTAR-2评价结果显示,有7个条目报告率<50%,其中报告率最低的3个条目分别是“是否报告系统评价纳入研究的基金资助信息(7.58%,5/66)”、“研究设计的选择依据是否给予解释(10.61%,7/66)”和“是否提供排除研究的清单以及排除理由(10.61%,7/66)”。②ROBIS评价结果显示,92.42%(61/66)的文献总体偏倚风险较高,其中四个领域(领域1研究的纳入排除标准、领域2研究的检索和筛选、领域3数据提取和质量评价和领域4数据合成和结果呈现)的偏倚风险高的文献分别占总数的78.79%(52/66)、81.82%(54/66)、40.91%(27/66)和66.67%(44/66)。③有15.15%(10/66)的文献AMSTAR-2显示方法学质量较高,ROBIS却显示其具有较高的偏倚风险。结论当前儿童针灸领域的系统评价质量仍有待提高。质量评价和偏倚风险评估结果大多数情况较为一致,但高方法学质量的文献也可能存在高风险偏倚。建议根据具体需求选择AMSTAR-2和ROBIS工具。

Objectives To evaluate the methodology quality and risk of bias of the systematic reviews/metaanalyses(SRs/MAs)of acupuncture in the field of children by the AMSTAR-2 and ROBIS tools.Methods We searched three international databases(PubMed,Cochrane Library,EMBASE),four Chinese databases(CBM,CJFD,CSJD,and WANFANG)from inception to January 2018.For each SR,two reviewers independently conducted AMSTAR-2(16 items)and ROBIS assessments with consensus and discussed challenges encountered.Results A total of 66 SRs/MAs were included,of which 7 were from CDSR,34 were from international journals and 25 were from Chinese journals.The results of AMSTAR-2 showed that there were 7 items with reporting rate<50%,among which 3 items with the least reported rate were the sources of funding for the studies included in the review(7.58%,5/66),selection of the study designs for inclusion in the review?(10.61%,7/66)and a list of excluded studies and justify the exclusions(10.61%,7/66).The results of ROBIS showed that the overall risk of biases was“high”in sixty-one SRs 92.42%(61/66).The reporting rates of four field(Domain 1:study eligibility criteria;Domain 2:identification and selection of studies;Domain 3:data collection and study appraisal;Domain 4:synthesis and findings)were 78.79%(52/66),81.82%(54/66),40.91%(27/66)and 66.67%(44/66),respectively.There are 15.15%(10/66)of the literature AMSTAR-2 showing a high methodological quality,but ROBIS shows a higher risk of bias.Conclusions The quality of systematic reviews in the field of acupuncture for children still needs to be improved.The results of quality evaluation and risk assessment of bias are mostly consistent,but the literature with high methodological quality may also have high risk bias.It is recommended to choose AMSTAR-2 and ROBIS tools according to specific needs.