A threshold analysis assessed the credibility of conclusions from network meta-analysis.
A threshold analysis assessed the credibility of conclusions from network meta-analysis.
复制标题
DOI：
10.1016/j.jclinepi.2016.07.003
复制
发表时间：
2016-12
影响因子：
7.2
通讯作者：
Welton, Nicky J.
中科院分区：
文献类型：
--
作者：
Caldwell, Deborah M.;Ades, A. B.;Dias, Sofia;Watkins, Sarah;Li, Tianjing;Taske, Nichole;Naidoo, Bhash;Welton, Nicky J.
关键词：
基于网络荟萃分析（NMA）评估治疗建议的可靠性。我们认为NMA中的证据可能存在偏见。我们依次对每一个两两对比，使用一系列结构化的阈值分析来问：(1)“在改变我们的决定之前，这个证据基础中的偏差会有多大？”(2)“如果决定改变，新的建议是什么？”我们通过两个nma来说明该方法，其中对nma进行了分级建议评估，发展和评估（GRADE）评估：体重减轻和骨质疏松症。根据GRADE， 4个减肥NMA评估为“低”质量，6个为“中等”质量；对于骨质疏松症，6个是“低”，9个是“中度”，1个是“高”。阈值分析表明，减肥网络中10个估计中有3个的合理偏差可能改变了治疗建议。对于骨质疏松症，16项估计中有6项的合理偏差可能改变建议。可能的偏倚改变治疗建议和最初的GRADE评估之间没有关系。对个体NMA对比的可靠性判断并不能帮助决策者理解治疗建议是否可靠。阈值分析揭示了最终的建议是否对数据中可能的偏差程度具有稳健性。
To assess the reliability of treatment recommendations based on network meta-analysis (NMA). We consider evidence in an NMA to be potentially biased. Taking each pairwise contrast in turn, we use a structured series of threshold analyses to ask: (1) “How large would the bias in this evidence base have to be before it changed our decision?” and (2) “If the decision changed, what is the new recommendation?” We illustrate the method via two NMAs in which a Grading of Recommendations Assessment, Development and Evaluation (GRADE) assessment for NMAs has been implemented: weight loss and osteoporosis. Four of the weight-loss NMA estimates were assessed as “low” and six as “moderate” quality by GRADE; for osteoporosis, six were “low,” nine were “moderate,” and 1 was “high.” The threshold analysis suggests plausible bias in 3 of 10 estimates in the weight-loss network could have changed the treatment recommendation. For osteoporosis, plausible bias in 6 of 16 estimates could change the recommendation. There was no relation between plausible bias changing a treatment recommendation and the original GRADE assessments. Reliability judgments on individual NMA contrasts do not help decision makers understand whether a treatment recommendation is reliable. Threshold analysis reveals whether the final recommendation is robust against plausible degrees of bias in the data.
登录
查看更多内容
影响因子：
9.8
作者：
Lu, Guobing;Welton, Nicky J.;Ades, A. E.
通讯作者：
Ades, A. E.
DOI：
10.1111/j.1467-985x.2004.00349.x
发表时间：
2005-01-01
影响因子：
2
作者：
Greenland, S
通讯作者：
Greenland, S
DOI：
10.1111/j.1467-985x.2010.00639.x
发表时间：
2010-01-01
影响因子：
2
作者：
Dias, S.;Welton, N. J.;Ades, A. E.
通讯作者：
Ades, A. E.
DOI：
10.1016/0197-2456(86)90046-2
发表时间：
1986-09-01
期刊：
CONTROLLED CLINICAL TRIALS
影响因子：
--
作者：
DERSIMONIAN, R;LAIRD, N
通讯作者：
LAIRD, N
影响因子：
2
作者：
Koenig, Jochem;Krahn, Ulrike;Binder, Harald
通讯作者：
Binder, Harald