ABSTRACT
Risk adjustment is often necessary for outcome quality indicators (QIs) to provide fair and accurate feedback to healthcare professionals. However, traditional risk adjustment models are generally oversimplified and not equipped to disentangle complex factors influencing outcomes that are out of a healthcare professional's control. We present VIRGO, a novel variational Bayes model trained on routinely collected, large administrative datasets to risk-adjust outcome QIs. VIRGO uses detailed demographics, diagnosis, and procedure codes to provide individualized risk adjustment and explanations on patient factors affecting outcomes. VIRGO achieves state-of-the-art on external datasets and features capabilities of uncertainty expression, explainable features, and counterfactual analysis capabilities. VIRGO facilitates risk adjustment by explaining how patient factors led to adverse outcomes and expresses the uncertainty of each prediction, allowing healthcare professionals to not only explore patient factors with unexplained variance that are associated with worse outcomes but also reflect on the quality of their clinical practice.
ABSTRACT
CONTEXT: In health care, monitoring of quality indicators (QIs) in general urology remains underdeveloped in comparison to other clinical specialties. OBJECTIVE: To identify, synthesise, and appraise QIs that monitor in-hospital care for urology patients. EVIDENCE ACQUISITION: This systematic review included peer-reviewed articles identified via Embase, MEDLINE, Web of Science, CINAHL, Global Health, Google Scholar, and grey literature from 2000 to February 19, 2021. The review was carried out under the Preferred Reporting Items of Systematic Reviews and Meta-Analyses (PRISMA) guidelines and used the Appraisal of Indicators through Research and Evaluation (AIRE) tool for quality assessment. EVIDENCE SYNTHESIS: A total of 5111 articles and 62 government agencies were screened for QI sets. There were a total of 57 QI sets included for analysis. Most QIs focused on uro-oncology, with prostate, bladder, and testicular cancers the most represented. The most common QIs were surgical QIs in uro-oncology (positive surgical margin, surgical volume), whereas in non-oncology the QIs most frequently reported were for treatment and diagnosis. Out of 61 articles, only four scored a total of ≥50% on the AIRE tool across four domains. Aside from QIs developed in uro-oncology, general urological QIs are underdeveloped and of poor methodological quality and most lack testing for both content validity and reliability. CONCLUSIONS: There is an urgent need for the development of methodologically robust QIs in the clinical specialty of general urology for patients to enable standardised quality of care monitoring and to improve patient outcomes. PATIENT SUMMARY: We investigated a range of quality indicators (QIs) that provide health care professionals with feedback on the quality of their care for patients with general urological diseases. We found that aside from urological cancers, there is a lack of QIs for general urology. Hence, there is an urgent need for the development of robust and disease-specific QIs in general urology.