Abstract
We consider statistical methods for benchmarking clinical centers based on a dichotomous outcome indicator. Borrowing ideas from the causal inference literature, we aim to reveal how the entire study population would have fared under the current care level of each center. To this end, we evaluate direct standardization based on fixed versus random center effects outcome models that incorporate patient-specific baseline covariates to adjust for differential case-mix. We explore fixed effects (FE) regression with Firth correction and normal mixed effects (ME) regression to maintain convergence in the presence of very small centers. Moreover, we study doubly robust FE regression to avoid outcome model extrapolation. Simulation studies show that shrinkage following standard ME modeling can result in substantial power loss relative to the considered alternatives, especially for small centers. Results are consistent with findings in the analysis of 30-day mortality risk following acute stroke across 90 centers in the Swedish Stroke Register.
Highlights
In recent years, the interest in profiling hospital performance has grown among different stakeholders including government and health insurers, hospitals and clinicians, and last but not least the patients
We focus on statistical methods to estimate center performance on a binary quality indicator such as 30-day mortality
We aim to overcome this using doubly robust methods (Robins and others, 2007) that build on a fixed center effects model but utilize inverse weighting by the so-called propensity score (PS) (Shahian and Normand, 2008), which is the probability of being treated in the observed center based on patient characteristics
Summary
The interest in profiling hospital performance has grown among different stakeholders including government and health insurers, hospitals and clinicians, and last but not least the patients. Health-care quality deserves careful statistical analysis yielding relevant and relatively simple measures with clear interpretation for hospital evaluation.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
