PDF(Sci-hub)
Abstract：
This randomized controlled trial (RCT) aimed to test the efficacy of enhanced access to Cochrane Handbook (Handbook) recommendations for judging the 2011 Cochrane risk of bias (RoB) domains for improving the adequacy of RoB judgments. Parallel-group RCT with a 1:1 allocation ratio (N = 2271 per group) was conducted. Eligible participants were corresponding authors of all published Cochrane reviews and protocols. After allocation by a random number generator, participants received 20 scenarios for assessing RoB. The intervention group was shown tables from the Handbook with instructions for assessing 2011 RoB tool together with scenarios they were supposed to assess-enhanced access to the Handbook. The control group was shown only a general link to the Handbook. The primary outcome was the proportion of participants that made an adequate judgment of RoB scenarios for analyzed domains. There were 240 responses out of 2020 delivered e-mail invitations in the intervention and 197/2254 in the control group. Only five participants from the intervention group judged RoB adequately in all the 20 scenarios and no one in the control group. The proportion of participants who adequately assessed all the scenarios within a domain was significantly higher in the intervention than in the control group. The frequency of adequate RoB judgments was 7.1% (95% CI: 5.0-9.3%, p < 0.001) higher in the intervention group (76.2%) than in the control group (69.0%). The enhanced access yields more adequate RoB assessments and could be incorporated in software supporting the RoB tool.
摘要：
这项随机对照试验（RCT）旨在测试增强获取Cochrane手册（Handbook）建议的有效性，以判断2011Cochrane偏倚风险（RoB）领域，以改善RoB判断的充分性。进行平行组RCT，分配比例为1：1（每组N=2271）。符合条件的参与者是所有已发表的Cochrane评论和协议的通讯作者。由随机数生成器分配后，参与者接受了20种评估RoB的方案。干预组显示了手册中的表格，其中包含评估2011年RoB工具的说明以及他们应该评估的情景-增强了对手册的访问。对照组仅显示了与手册的一般链接。主要结果是对分析领域的RoB情景做出适当判断的参与者比例。在干预组中，2020年发送的电子邮件邀请中有240个回复，对照组中有197/2254个回复。在所有20种情况下，干预组只有5名参与者对RoB做出了充分的判断，对照组中没有人。在干预措施中，充分评估某个领域内所有方案的参与者比例明显高于对照组。充分的RoB判断的频率为7.1％（95％CI：5.0-9.3％，p<0.001）干预组（76.2％）高于对照组（69.0％）。增强的访问可以产生更充分的RoB评估，并且可以合并到支持RoB工具的软件中。
