Assessment of spin within diagnostic accuracy meta-analyses published in top pathology journals: a cross-sectional...

"Spin" - the overinterpretation of research findings - has been assessed in experimental and synthesis designs of interventional evidence. Little attention has been given to the assessment of spin in diagnostic evidence bases. Therefore, we conducted a cross-sectional study to assess for spin found within diagnostic accuracy meta-analyses published in top pathology journals. We searched PubMed for diagnostic test accuracy meta-analyses published in top 20 pathology journals from past to present. We applied published methodology to identify the presence of 10 items of actual overinterpretation and 9 items of potential overinterpretation. Authors screened and extracted relevant data from sample studies in a masked duplicate fashion to reduce extraction errors. On September 21st, 2023, we identified 207 articles from PubMed for potential inclusions. After screening, 55 abstracts and full-texts were available for full data extraction. with 80% (44/55) having positive conclusions germane to accuracy or clinical implications. Every meta-analysis contained one item of spin. Most positive conclusions in abstracts (75%; 33/44) and full-texts (79.6%; 35/44) did not adequately reflect pooled estimates while 19 (34.5%) studies employed non-recommended statistical approaches for pooling accuracy measures. Diagnostic test accuracy meta-analyses found within top pathology journals consistently overstate their conclusions. Authors should contextualize diagnostic summary effects within predetermined diagnostic performance. Further, authors should ensure that summary estimates are pooled using bivariate or hierarchical approaches that maintain threshold effects throughout meta-analytic calculations. Readers should cautiously interpret meta-analyses that fail to report sample sizes and confidence intervals of summary estimates.