Department of Health Research Methods, Evidence, and Impact, McMaster University, Hamilton, ON, Canada.
St. Joseph's Healthcare Hamilton, McMaster University, 501-25 Charlton Avenue East, Hamilton, ON, L8N 1Y2, Canada.
BMC Med Res Methodol. 2017 Dec 29;17(1):181. doi: 10.1186/s12874-017-0459-5.
Evidence shows that research abstracts are commonly inconsistent with their corresponding full reports, and may mislead readers. In this scoping review, which is part of our series on the state of reporting of primary biomedical research, we summarized the evidence from systematic reviews and surveys, to investigate the current state of inconsistent abstract reporting, and to evaluate factors associated with improved reporting by comparing abstracts and their full reports.
We searched EMBASE, Web of Science, MEDLINE, and CINAHL from January 1st 1996 to September 30th 2016 to retrieve eligible systematic reviews and surveys. Our primary outcome was the level of inconsistency between abstracts and corresponding full reports, which was expressed as a percentage (with a lower percentage indicating better reporting) or categorized rating (such as major/minor difference, high/medium/low inconsistency), as reported by the authors. We used medians and interquartile ranges to describe the level of inconsistency across studies. No quantitative syntheses were conducted. Data from the included systematic reviews or surveys was summarized qualitatively.
Seventeen studies that addressed this topic were included. The level of inconsistency was reported to have a median of 39% (interquartile range: 14% - 54%), and to range from 4% to 78%. In some studies that separated major from minor inconsistency, the level of major inconsistency ranged from 5% to 45% (median: 19%, interquartile range: 7% - 31%), which included discrepancies in specifying the study design or sample size, designating a primary outcome measure, presenting main results, and drawing a conclusion. A longer time interval between conference abstracts and the publication of full reports was found to be the only factor which was marginally or significantly associated with increased likelihood of reporting inconsistencies.
This scoping review revealed that abstracts are frequently inconsistent with full reports, and efforts are needed to improve the consistency of abstract reporting in the primary biomedical community.
有证据表明，研究摘要通常与完整报告不一致，并且可能误导读者。在这项范围界定综述中，我们总结了来自系统评价和调查的证据，以调查目前摘要报告不一致的情况，并通过比较摘要和完整报告来评估与改进报告相关的因素。该综述是我们关于基础生物医学研究报告现状系列研究的一部分。
我们检索了 1996 年 1 月 1 日至 2016 年 9 月 30 日的 EMBASE、Web of Science、MEDLINE 和 CINAHL，以检索合格的系统评价和调查。我们的主要结局是摘要与相应完整报告之间的不一致程度，用百分比（百分比越低表示报告越好）或分类评级（如主要/次要差异、高/中/低不一致）表示，由作者报告。我们使用中位数和四分位间距来描述研究之间的不一致程度。没有进行定量综合。总结了纳入的系统评价或调查的数据。
纳入了 17 项研究。报告的不一致程度中位数为 39%（四分位间距：14% - 54%），范围为 4%至 78%。在一些将主要差异与次要差异分开的研究中，主要差异的程度范围为 5%至 45%（中位数：19%，四分位间距：7% - 31%），包括研究设计或样本量指定、指定主要结果测量、呈现主要结果和得出结论方面的差异。会议摘要和完整报告发表之间的时间间隔较长，被发现是唯一与报告不一致的可能性增加有轻微或显著关联的因素。
这项范围界定综述表明，摘要经常与完整报告不一致，需要努力提高基础生物医学界摘要报告的一致性。
