Reproducibility of the STARD checklist for diagnostic accuracy studies
Author Information
Author(s): Nynke Smidt, Anne WS Rutjes, Daniëlle AWM van der Windt, Raymond WJG Ostelo, Patrick M Bossuyt, Johannes B Reitsma, Lex M Bouter, Henrica CW de Vet
Primary Institution: Institute for Research in Extramural Medicine, VU University Medical Center
Hypothesis
The study investigates the inter-assessment reproducibility of the STARD checklist items in diagnostic accuracy studies.
Conclusion
The overall reproducibility of the quality of reporting on diagnostic accuracy studies using the STARD statement was found to be good, but substantial disagreements were noted for specific items.
Supporting Evidence
- The overall inter-assessment agreement for all items of the STARD statement was 85%.
- Cohen's kappa for inter-assessment was 0.70, indicating substantial agreement.
- The inter-assessment reliability of the STARD checklist was satisfactory with an ICC of 0.79.
Takeaway
This study looked at how consistently reviewers can evaluate the quality of reports on medical tests, finding that while they mostly agree, there are some tricky parts that cause confusion.
Methodology
Two reviewers independently evaluated 32 diagnostic accuracy studies using the STARD checklist, with a consensus evaluation for disagreements.
Potential Biases
The reviewers were not evaluating the likelihood of bias, only the quality of reporting.
Limitations
The study's findings may be affected by subjective judgment in assessing certain items of the STARD checklist.
Participant Demographics
The studies included 25 cohort and 7 case control studies, with a variety of diseases and tests.
Statistical Information
P-Value
0.70
Confidence Interval
[95% CI: 0.62 to 0.89]
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website