Eight studies (n=13,856 participants, range 80 to 6,270) were included in the review. In all studies the appropriateness of the indication for OGD was pre-endoscopically assessed by endoscopists; endoscopic findings served as the reference standard. Six studies used ASGE guidelines to assess appropriateness and two studies used EPAGE guidelines.
According the guidelines, 10,643 OGD indications were appropriate and 3,010 (22%) were inappropriate; 203 participants were excluded from the analysis because their indications were not included in the current guidelines.
Accuracy of appropriateness guidelines for relevant endoscopic findings: Pooled estimate of sensitivity was 85% (95% CI 84% to 86%). Pooled estimate of specificity was 28% (95% CI 27% to 29%). Pooled positive likelihood ratio was 1.18 (95% CI 1.1 to 1.3). Pooled negative likelihood ratio was 0.6 (95% CI 0.5 to 0.7). There was evidence of significant between-study heterogeneity, which was not explained by any of the variables considered. Egger's test for publication bias was not significant.
Accuracy of appropriateness guidelines for cancer: All studies reported data on cancer detection; 267 carcinomas were detected in the included studies (prevalence=2%). Pooled estimate of sensitivity was 97% (95% CI 94% to 98%). Pooled estimate of specificity was 22% (95% CI 22% to 23%). Pooled positive likelihood ratio was 1.2 (95% CI 1.1 to 1.4). Pooled negative likelihood ratio was 0.2 (95% CI 0.05–0.9). There was evidence of significant between-study heterogeneity, which was not explained by any of the variables considered. Egger's test for publication bias was significant.