Table 4 Summary of results of relevant studies dealing with the reporting quality in diagnostic accuracy studies using the STARD Checklist
First author | Study year | Article year | Medical field | Included studies | Mean total STARD Score a | Important findings |
|---|---|---|---|---|---|---|
Hong et al.* | 2018 | 2016 | Imaging/Magnetic Resonance | 142 | 16.6/30 (55) | Articles published in journals with higher IFs (17.2 vs. 16; p = 0.001) and STARD-adopting journals (17.5 vs. 16.4; p = 0.01) achieved higher total STARD scores. No evidence of a difference in the total STARD score was found for mode of data collection and imaging modality |
Zarei et al.*, b | 2018 | 2015 | Radiology | 151 | (69.45) | Several items, such as providing a registration number (1.1%), full study protocol (10.7%), reporting adverse events (14.9%), a prespecified sample size (16.11%), analyses prespecified from exploratory (28.19%), and the distribution of alternative diagnoses (26.17%) were infrequently reported |
Choi et al.* | 2016 | 2011–2015 | Radiology | 63 | 20/27 (74) | With the effect of exposure time partialled out, the STARD score did not significantly correlate with citation numbers (partial correlation coefficient = 0.15, p = 0.23) |
Hogan et al.* | 2020 | 2018 | Pathology | 171 | 15.4/34 (45) | Articles that were published in STARD-adopting journals (16.1 vs. 14.8, p = 0.018) reported significantly more items compared to STARD-nonadopting journals. No evidence of a difference in the total STARD score was found for IF, citation number, and pathology (sub-)specialty |
Michelessi et al.* | 2017 | 2003–2014 | Ophthalmology/glaucoma | 106 | 16.8/31 (54.1) | An increase in the total STARD score was found for publication year (OR: 1.03 per year, p = 0.03) and for journals with IF > 3.5 vs. < 2 (OR: 1.22, p = 0.03) |
Korevaar et al | 2014 | 2012 | General Medicine | 112 | 15.3/25 (61) | Articles published in 2012 reported on 1.7 items (95% Cl 0.9–2.5) more than in 2004 c. Significantly more items were reported in studies published in general journals than in discipline-specific journals (17.7 vs 14.8, p = 0.002), for single gate studies vs. multiple gate studies (16.8 vs. 12.1, p < 0.001), and for studies assessing imaging tests compared with laboratory tests and other types of tests (17 vs. 14 vs. 14.5; p < 0.001) |
Walther et al | 2014 | 2003–2011 | Imaging/CT angiography | 130 | 14.4/21 (69) | Articles published in STARD-adopting journals had a significantly higher total STARD score (15.4 vs. 14.1; p = 0.018 than STARD-nonadopting journals. From 2003 to 2011, the total STARD score increased by an average of 0.30 points (p = 0.03) per year |