188宝金博页面版

1 Scientific Data | (2025) 12:355 | https://doi.org/10.1038/s41597-025-04629-1www.nature.com/scientificdataSPIRIt-CONSORt-tM: a corpus for assessing transparency of clinical trial protocol and results publicationsLan Jiang 1,6  ? , Colby J Vorland 2,6 , Xiangji Ying 3,6 , Andrew W Brown 4,5 , Joe D Menke 1 , Gibong Hong 1 , Mengfei Lan 1 , Evan Mayo-Wilson 3 & Halil Kilicoglu 1  ?Randomized controlled trials (RCTs) can produce valid estimates of the benef i ts and harms of therapeutic interventions. However, incomplete reporting can undermine the validity of their conclusions. Reporting guidelines, such as SPIRIT for protocols and CONSORT for results, have been developed to improve transparency in RCT publications. In this study, we report a corpus of 200 RCT publications, named SPIRIT-CONSORT-TM, annotated for transparency. We used a comprehensive data model that includes 83 items from SPIRIT and CONSORT checklists for annotation. Inter-annotator agreement was calculated for 30 pairs. The dataset includes 26,613 sentences annotated with checklist items and 4,231 terms. We also trained natural language processing (NLP) models that automatically identify these items in publications. The sentence classif i cation model achieved 0.742 micro-F1 score (0.865 at the article level). The term extraction model yielded 0.545 and 0.663 micro-F1 score in strict and lenient evaluation, respectively. The corpus serves as a benchmark to train models that assist stakeholders of clinical research in maintaining high reporting standards and synthesizing information on study rigor and conduct.Background & SummaryRandomized controlled trials (RCTs) are foundational to evidence-based medicine 1 . When well-designed and rigorously conducted, RCTs can provide valid estimates of ef f ects of therapeutic interventions 2 . For RCTs to benef i t clinical practice and health policy, they must be reported thoroughly and transparently 2,3 . Complete reporting facilitates the assessment of RCT validity and applicability 2 . Given the high cost and time investment of RCTs, transparent reporting also helps avoid unnecessary duplication and research waste 2,4 . Unfortunately, even well-conducted RCTs of t en suf f er from inadequate reporting 2,5,6 .T h e SPIRIT 2013 Statement 7,8 and CONSORT 2010 Statement 2,9 are reporting guidelines that aim to enhance the reporting quality of RCT protocols and results, respectively. CONSORT 2010 (referred to as CONSORT for brevity, henceforth) consists of a checklist and a participant fl owchart. Th e checklist includes 25 items essen-tial for understanding the design, implementation, analysis, and results of parallel RCTs. CONSORT has been widely endorsed by journals, publishers, and editorial organizations, and its adoption has been found to be posi-tively correlated with completeness of reporting 10,11 . However, studies have also repeatedly shown that key meth-odological details like allocation concealment remain poorly reported even in articles published in endorsing journals 5,6 . An overview of systematic reviews found that CONSORT adherence was reported to be inadequate in 88% of the reviews 6 . Th e complementary SPIRIT 2013 guidelines 7,8 (referred to as SPIRIT, henceforth) consist of recommended items and a fi gure to be included in trial protocols. SPIRIT includes many applicable items from CONSORT, especially items related to methodology, and of t en encourages authors to report more infor-mation than would typically be included in a results report. Protocols are widely used to appraise trial conduct 1 University of Illinois Urbana-Champaign, School of Information Sciences, Champaign, IL, 61820, USA. 2 indiana University, School of Public Health, Bloomington, IN, 47405, USA. 3 University of North Carolina Chapel Hill, Gillings School of Global Public Health, Chapel Hill, NC, 27599, USA. 4 University of Arkansas for Medical Sciences, Little Rock, AR, 72205, USA. 5 Arkansas Children’s Research Institute, Little Rock, AR, 72202, USA. 6 These authors contributed equally: Lan Jiang, Colby J Vorland, Xiangji Ying. ? e-mail: lanj3@illinois.edu; halil@illinois.eduDATA DESCRIPTOROPEN