外科元证据中模棱两可情况的频率：对炎症性肠病文献中系统评价的综述

Frequency of equivocation in surgical meta-evidence: a review of systematic reviews within IBD literature.

作者信息

Delaney John D, Holbrook John T, Dewar Robert K, Laws Patrick J, Engel Alexander F

机构信息

Colorectal Surgery, Northern Clinical School, University of Sydney, Sydney, New South Wales, Australia.

Royal Prince Alfred Hospital, Camperdown, New South Wales, Australia.

出版信息

BMJ Open. 2017 Dec 19;7(12):e018715. doi: 10.1136/bmjopen-2017-018715.

DOI:10.1136/bmjopen-2017-018715

PMID:29259063

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC5778281/

Abstract

OBJECTIVE

To assess the level of equivocation among level 1 evidence in ulcerative colitis and Crohn's disease and determine whether any predisposing factors are present.

METHOD

MEDLINE, Embase, CINHAL and Cochrane were searched from 2006 to 2017. Papers were scored using AMSTAR and categorised into surgical (S), medical (M) or medical and surgical (MS) groups. The ability of each paper to make a recommendation and conclusiveness in doing so was recorded.

RESULTS

278 papers were assessed. 82% (n=227) could make a recommendation, 18% (n=51) could not. There was a significant difference in ability to provide a recommendation between S and M (P=0.003) but not MS and M (P=0.022) nor S and MS (P=0.79). Where a recommendation was made, S papers were more likely to be tempered than M papers (P=0.014) but not MS papers (P=0.987).

CONCLUSIONS

Surgical meta-evidence within the inflammatory bowel disease domain is more than twice as likely as medical meta-evidence to be unable to provide a recommendation for clinical practice. Where a recommendation was made, surgical reviews were twice as likely to temper their conclusion.

摘要

目的

评估溃疡性结肠炎和克罗恩病一级证据中的模棱两可程度，并确定是否存在任何诱发因素。

方法

检索2006年至2017年的MEDLINE、Embase、CINHAL和Cochrane数据库。使用AMSTAR对论文进行评分，并分为手术（S）、内科（M）或内科与手术（MS）组。记录每篇论文提出建议的能力及其结论的确定性。

结果

共评估了278篇论文。82%（n = 227）的论文能够提出建议，18%（n = 51）的论文不能。S组和M组在提出建议的能力上存在显著差异（P = 0.003），但MS组和M组（P = 0.022）以及S组和MS组（P = 0.79）之间没有显著差异。在提出建议的情况下，S组论文比M组论文更有可能缓和语气（P = 0.014），但MS组论文则不然（P = 0.987）。