Compliance with STARD Checklist among Studies of Coronary CT Angiography: Systematic Review

Abstract

To determine the compliance of prospective diagnostic accuracy studies investigating computed tomographic (CT) coronary angiography with Standards for Reporting of Diagnostic Accuracy (STARD) guidelines. Relevant studies were identified by means of a systematic literature search that included the MEDLINE, EMBASE, and Institute for Scientific Information Web of Science databases. Prospective studies that compared CT with conventional coronary angiography for the evaluation of the coronary arteries were included. STARD compliance was assessed by three independent investigators using 21 of the original 25 STARD checklist items. Items with the qualifier "if done" (items 13, 23, and 24) were excluded because they were not applicable to all studies. Owing to the inclusion criteria, all studies fulfilled item 9; therefore, this item was excluded as well. The correlation between the total score and multiple variables was tested with a linear regression model. One hundred thirty studies published in 44 scientific journals were included in the analysis. There was a significant correlation between the year of publication and STARD-adopting versus non-STARD-adopting journals, with the total STARD score based on a linear regression model. Studies published in STARD-adopting journals showed a significantly higher total STARD score than those published in nonadopting journals (15.4 ± 2.7 vs 14.1 ± 2.7; P = .018). Linear regression analysis yielded an increase in the total STARD score of 0.30 points (95% confidence interval: 0.03, 0.57; P = .031) per year. Adequate reporting of individual items varied between 17% (item 20b) and 97% (item 1). The overall compliance with reporting guidelines of prospective diagnostic accuracy studies of CT coronary angiography is moderate to good, and STARD-adopting journals have greater STARD compliance than nonadopting journals. Online supplemental material is available for this article.