Hospital Comparison in Urology
Are there quality differences between urology departments in Switzerland?
Factsheet
- Schools involved School of Health Professions
- Institute(s) Institute of Health Economics and Health Policy
- Strategic thematic field Thematic field "Caring Society"
- Funding organisation Others
- Duration (planned) 01.09.2025 - 31.05.2026
- Head of project Prof. Dr. Tobias Benjamin Müller
-
Project staff
Mara Julia Hendry
Marie-Lise Sautebin - Keywords Hospital quality, risk adjustment, machine learning
Situation
The Swiss healthcare system builds largely on free hospital choice. But where should I undergo surgery? Without clinically relevant quality indicators, patients cannot make informed provider choices.
Course of action
How can we adjust for systematic differences in patient risk across hospitals so that hospitals treating more complex cases are not unfairly penalized in quality rankings? As part of a model competition, we compared the predictive performance of traditional parametric models (OLS/logit) for adverse clinical events (e.g., complications, infections) with that of machine learning approaches.
Result
Two main findings: (1) Machine learning approaches clearly outperform parametric models and should therefore be used for risk adjustment. (2) Hospitals differ substantially in the quality of care provided for key urological procedures such as cystectomy and prostatectomy in Switzerland.
Looking ahead
The Swiss healthcare system urgently needs more transparency about the quality of care of hospitals so that quality competition finally can unfold and patients can make informed decisions.
