文献速递流行病学与生物统计医学教育与科研方法

疼痛系统评价会议摘要选择性发表、结果不可靠且报告质量差

Conference abstracts describing systematic reviews on pain were selectively published, not reliable, and poorly reported.
PubMed 文献 · J Clin Epidemiol · 2019-09-15

该研究筛查2008至2016年五届世界疼痛大会的会议摘要,找出描述系统评价(SR)的摘要,并于2018年4月检索其对应全文发表。共纳入143篇SR摘要,其中90篇(63%)已发表全文;在79对可评估摘要-全文配对中,40%存在某种形式的不一致,16%存在定性不一致;摘要对PRISMA-A各条目的中位依从率为33%。结论认为疼痛SR会议摘要存在选择性发表、结果不可靠且报告质量差。

要点

  • 143篇疼痛SR会议摘要中,90篇(63%)至2018年4月已发表全文。
  • 79对可评估配对中40%存在不一致,13对(16%)效应方向不同。
  • 摘要对PRISMA-A各条目依从率中位数为33%(IQR 29%–42%)。

核心结论

  • 疼痛系统评价的会议摘要存在选择性发表、结果不可靠且报告质量差的问题。

研究设计与样本

  • 研究筛查2008至2016年五届世界疼痛大会(WCPs)的摘要以找出描述系统评价(SR)的摘要,并由两名作者于2018年4月通过PubMed和Google Scholar检索对应全文发表,将摘要中提取的方法和结局与全文比较,并依据PRISMA for Abstracts(PRISMA-A)清单评价摘要报告质量。

主要发现

  • 共识别143篇描述SR的会议摘要,其中90篇(63%)至2018年4月已发表为同行评议期刊全文。
  • 从会议展示到发表的中位时间为5个月(四分位距:-0.25至14个月)。
  • 在79对可评估不一致性的摘要-全文配对中,40%的配对存在某种形式的不一致。
  • 13对(16%)分析配对存在定性不一致(效应方向不同)。
  • 摘要对每个PRISMA-A清单条目的依从率中位数为33%(四分位距:29%至42%)。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

OBJECTIVE: The objective of the study was to determine the reporting quality of systematic review (SR) abstracts presented at World Congresses on Pain (WCPs) and to quantify agreement in results presented in those abstracts with their corresponding full-length publications. STUDY DESIGN AND SETTING: We screened abstracts of five WCPs held from 2008 to 2016 to find abstracts describing SRs. Two authors searched for corresponding full publications using PubMed and Google Scholar in April 2018. Methods and outcomes extracted from abstracts were compared with their corresponding full publications. The reporting quality of abstracts was evaluated against the PRISMA for Abstracts (PRISMA-A) checklist. RESULTS: We identified 143 conference abstracts describing SRs. Of these, 90 (63%) were published as full-length articles in peer-reviewed journals by April 2018, with a median time from conference presentation to publication of 5 months (interquartile range: -0.25 to 14 months). Among 79 abstract-publication pairs evaluable for discordance, there was some form of discordance in 40% of pairs. Qualitative discordance (different direction of the effect) was found in 13 analyzed pairs (16%). The median adherence by abstracts to each PRISMA-A checklist item was 33% (interquartile range: 29% to 42%). CONCLUSION: Conference abstracts of pain SRs are selectively published, not reliable, and poorly reported.

来源:J Clin Epidemiol,查看原文。版权归原作者所有。

引用

  1. Saric L, Dosenovic S, Saldanha IJ, et al. Conference abstracts describing systematic reviews on pain were selectively published, not reliable, and poorly reported. J Clin Epidemiol. 2019. PMID: 31533073. doi:10.1016/j.jclinepi.2019.09.011

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。