Isiguzo Godsent C, Zunza Moleen, Chirehwa Maxwell, Mayosi Bongani M, Thabane Lehana
1Department of Medicine, Groote Schuur Hospital and University of Cape Town, Cape Town, South Africa.
2Cardiology Unit, Department of Medicine, Federal Teaching Hospital, Abakaliki, Nigeria.
Pilot Feasibility Stud. 2018 May 31;4:107. doi: 10.1186/s40814-018-0302-8. eCollection 2018.
Pilot trials are miniature researches carried out with the sole aim of acting as the precursor for larger more definitive studies. Abstracts are used to summarize and introduce the findings to the reading audience. There is substantive empirical evidence showing that abstracts, despite their important roles, are not informative enough, lacking the necessary details. This systematic survey was designed to assess the quality of reporting of heart failure pilot trial abstracts. The quality of reporting was defined as the completeness of reporting based on adherence to the CONSORT extension for reporting of pilot trial abstracts. We also identified factors associated with reporting quality.
We searched MEDLINE (PubMed), Cochrane Controlled Trials Register, Scopus, and African-wide information databases for abstracts from heart failure pilot trials in humans published from 1 January 1990 to 30 November 2016. These were assessed to determine the extent of adherence to CONSORT extension checklist for reporting of abstracts of pilot trials. We screened identified studies for inclusion based on title and abstract. Data were independently extracted by two reviewers using the checklist. We used regression analysis to assess the association between completeness of reporting (measured as the number of items in the CONSORT extension checklist for reporting of abstracts in pilot trials contained in each abstract) and factors influencing the quality of the reports.
Two hundred and twenty-eight (228) articles were retrieved, of which 92 met the inclusion criteria. The mean CONSORT extension score was 8.3/16 (standard deviation 1.7); the least reported items were the source of funding (1% [1/92]), trial registration (13% [12/92]), randomization sequence (13% [12/92]), number randomized to each arm (16% [15/92]), and number analyzed in each arm (16% [15/92]). Multivariable regression analysis showed that pharmacological intervention pilot trials [incidence rate ratio (IRR) = 0.88; 95% confidence interval (CI), 0.81-0.97] were significantly associated with better reporting. Other factors such as structured abstract (IRR = 1.10; 95% CI, 0.99-1.23) and CONSORT endorsement (IRR = 1.10; 95% CI, 0.99-1.23) only showed minimal relationship with better reporting quality.
The quality of reporting of abstracts of heart failure pilot trials was suboptimal. Pharmacological intervention was significantly associated with better reporting. These findings are consistent with previous research on reporting of trials.
预试验是开展的小型研究，其唯一目的是作为更大规模、更具确定性研究的先导。摘要用于向读者总结和介绍研究结果。有大量实证证据表明，摘要尽管发挥着重要作用，但信息不够丰富，缺乏必要细节。本系统综述旨在评估心力衰竭预试验摘要的报告质量。报告质量定义为基于遵循预试验摘要报告的CONSORT扩展版的报告完整性。我们还确定了与报告质量相关的因素。
我们检索了MEDLINE（PubMed）、Cochrane对照试验注册库、Scopus和全非洲信息数据库，以获取1990年1月1日至2016年11月30日发表的关于人类心力衰竭预试验的摘要。对这些摘要进行评估，以确定其遵循预试验摘要报告的CONSORT扩展版清单的程度。我们根据标题和摘要筛选确定的研究以纳入分析。两名审阅者使用清单独立提取数据。我们使用回归分析来评估报告完整性（以每个摘要中包含的预试验摘要报告的CONSORT扩展版清单中的项目数量衡量）与影响报告质量的因素之间的关联。
共检索到228篇文章，其中92篇符合纳入标准。CONSORT扩展版平均得分是8.3/16（标准差1.7）；报告最少的项目是资金来源（1%[1/92]）、试验注册（13%[12/92]）、随机化序列（13%[12/92]）、每组随机分配的人数（16%[15/92]）以及每组分析的人数（16%[15/92]）。多变量回归分析表明，药物干预预试验[发病率比（IRR）=0.88；95%置信区间（CI），0.81 - 0.97]与更好的报告显著相关。其他因素，如结构化摘要（IRR = 1.10；95%CI，0.99 - 1.23）和CONSORT认可（IRR = 1.10；95%CI，0.99 - 1.23）与更好的报告质量仅显示出极小的关系。
心力衰竭预试验摘要的报告质量欠佳。药物干预与更好的报告显著相关。这些发现与先前关于试验报告的研究一致。
