Background
Although statistical procedures for pooling of several epidemiological metrics are generally available in statistical packages, those for meta-analysis of diagnostic test accuracy studies including options for multivariate regression are lacking. Fitting regression models and the processing of the estimates often entails lengthy and tedious calculations. Therefore, packaging appropriate statistical procedures in a robust and user-friendly program is of great interest to the scientific community.
Methods
metadta is a statistical program for pooling of diagnostic accuracy test data in Stata. It implements both the bivariate random-effects and fixed-effects model, allows for meta-regression, and presents the results in tables, a forest plot and/or summary receiver operating characteristic (SROC) plot. For a model without covariates, it also quantifies heterogeneity using an I2 statistic that accounts for the mean-variance relationship, and correlation between sensitivity and specificity, a typical characteristic of diagnostic data. To demonstrate metadta, we applied the program on two published meta-analyses on: 1) the sensitivity and specificity of cytology and other markers including telomerase for primary diagnosis of bladder cancer; and 2) the accuracy of human papillomavirus testing on self-collected versus clinician-collected samples to detect cervical precancer.
Results
Without requiring a continuity correction, metadta generated a pooled sensitivity and specificity of 0.77 [95% CI: 0.70, 0.82] and 0.91 [95% CI: 0.75, 0.97] respectively of telomerase for the diagnosis of primary bladder cancer. metadta allowed to assess the relative accuracy of human Papilloma virus (HPV) testing on self- versus clinician-taken specimens in matched studies taking into account two covariates. Under the condition of using assays based on target-amplification, HPV tests were similarly sensitive to detect cervical pre-cancer, irrespective of clinical setting.
Conclusion
The metadta program implements state of art statistical procedures in an attempt to close the gap between methodological statisticians and systematic reviewers. With metadta, we hope to popularize even further, the use of appropriate statistical methods for diagnostic meta-analysis.