How does staff expertise impact the technical efficiency of radiology departments? Evidence from a two-stage DEA analysis with metafrontiers and non-discretionary inputs

Abstract In many countries, rising demand for diagnostic services coupled with decreasing healthcare budgets makes it necessary to increase the technical efficiency of radiology departments. This study examines how the expertise of radiology technologists—specifically their education, training, and specialization—affects the technical efficiency of 168 radiology sub-units within a leading university hospital in Italy which performed over 700,000 tests between 2021 and 2022. In our study, each Decision-Making Unit (DMU) represents an organizational unit operating only one type of imaging technology (X-ray, Magnetic Resonance Imaging, or Computerized Tomography). Each piece of equipment can be allocated to only one DMU during the time horizon, while radiology technologists overseeing these examinations can be assigned to multiple DMUs without restrictions. The bootstrapped two-stage Data Envelopment Analysis (DEA) approach was taken as a reference since it provides a robust way to draw statistical inferences. To account for the specificity of the case, in the first stage, we enhanced the DEA model by incorporating non-discretionary inputs and applying the metafrontier approach. The integration of non-discretionary inputs reflects the fact that equipment allocation is beyond managerial control in the short term due to structural or external constraints. The metafrontier approach was selected because it enables valid efficiency comparisons across DMUs using different technologies. The second stage involves a truncated regression analysis through which we identify efficiency determinants. Our findings show that while technologists’ experience can negatively affect efficiency, their specialization has a positive effect. Staff training is crucial for maximizing the positive impact of new technologies on efficiency. Additionally, the presence of elderly patients and emergency referrals further reduces efficiency. This study offers healthcare managers insight into how to optimize resource allocation and enhance efficiency and provides a robust method for evaluating efficiency in settings with units employing different technologies and non-discretionary inputs.