随机对照试验摘要的报告质量：对主要牙科综合期刊的调查

Reporting quality of randomized controlled trial abstracts: survey of leading general dental journals.

作者信息

Hua Fang, Deng Lijia, Kau Chung How, Jiang Han, He Hong, Walsh Tanya

出版信息

J Am Dent Assoc. 2015 Sep;146(9):669-678.e1. doi: 10.1016/j.adaj.2015.03.020.

DOI:10.1016/j.adaj.2015.03.020

PMID:26314976

Abstract

BACKGROUND

The authors conducted a study to assess the reporting quality of randomized controlled trial (RCT) abstracts published in leading general dental journals, investigate any improvement after the release of the Consolidated Standards of Reporting Trials (CONSORT) for Abstracts guidelines, and identify factors associated with better reporting quality.

METHODS

The authors searched PubMed for RCTs published in 10 leading general dental journals during the periods from 2005 to 2007 (pre-CONSORT period) and 2010 to 2012 (post-CONSORT period). The authors evaluated and scored the reporting quality of included abstracts by using the original 16-item CONSORT for Abstracts checklist. The authors used risk ratios and the t test to compare the adequate reporting rate of each item and the overall quality in the 2 periods. The authors used univariate and multivariate regressions to identify predictors of better reporting quality.

RESULTS

The authors included and evaluated 276 RCT abstracts. Investigators reported significantly more checklist items during the post-CONSORT period (mean [standard deviation {SD}], 4.53 [1.69]) than during the pre-CONSORT period (mean [SD], 3.87 [1.10]; mean difference, -0.66 [95% confidence interval, -0.99 to -0.33]; P < .001). Investigators reported 3 items-interventions, objective, and conclusions-adequately in most of the abstracts (> 80%). In contrast, the authors saw sufficient reporting of randomization, recruitment, outcome in the results section, and funding in none of the pre-CONSORT abstracts and less than 2% of the post-CONSORT abstracts. On the basis of the multivariate analysis, a higher impact factor (P < .001) and a publication date in the post-CONSORT period (P = .003) were associated significantly with higher reporting quality.

CONCLUSIONS

The reporting quality of RCT abstracts from leading general dental journals has improved significantly, but there is still room for improvement.

PRACTICAL IMPLICATIONS

Joint efforts by authors, reviewers, journal editors, and other stakeholders to improve the reporting of dental RCT abstracts are needed.

摘要

背景

作者开展了一项研究，以评估发表于主要牙科综合期刊上的随机对照试验（RCT）摘要的报告质量，调查《试验报告统一标准》（CONSORT）摘要指南发布后报告质量是否有所改善，并确定与更好报告质量相关的因素。

方法

作者在PubMed中检索了2005年至2007年（CONSORT之前时期）和2010年至2012年（CONSORT之后时期）发表于10种主要牙科综合期刊上的RCT。作者使用原始的16项CONSORT摘要清单对纳入的摘要的报告质量进行评估和评分。作者使用风险比和t检验比较两个时期各项目的充分报告率和整体质量。作者使用单变量和多变量回归来确定更好报告质量的预测因素。

结果

作者纳入并评估了276篇RCT摘要。在CONSORT之后时期，研究者报告的清单项目（均值[标准差{SD}]，4.53[1.69]）明显多于CONSORT之前时期（均值[SD]，3.87[1.10]；均值差异，-0.66[95%置信区间，-0.99至-0.33]；P<.001）。研究者在大多数摘要（>80%）中充分报告了3个项目——干预措施、目的和结论。相比之下，作者发现CONSORT之前的摘要中没有一篇充分报告随机化、招募、结果部分的结果和资金情况，CONSORT之后的摘要中这一比例不到2%。基于多变量分析，较高的影响因子（P<.001）和CONSORT之后时期的出版日期（P=.003）与较高的报告质量显著相关。