Cost and Allocative Efficiency of Public Hospitals in Ireland – A Robust DEA Approach

This study evaluates the cost and allocative efficiencies of 37 Irish public hospitals (IPH) from July 2017 to June 2018, using a unique dataset containing in-depth information on inputs, outputs, and input prices. Conventional DEA and two-stage double-bootstrap (DB) DEA models are applied to derive robust bias-corrected efficiency scores for IPH. The DB DEA results show that Irish hospitals operate with an average bias-corrected cost efficiency score of 0.756. Thus, IPH are considerably cost-inefficient as they could reduce their costs by almost 25% and still produce the same output level. Moreover, the average DB DEA allocative efficiency is derived at 0.965, whilst the mean bias-corrected technical efficiency is 0.783, indicating that technical (productive) inefficiencies are the primary driver of cost inefficiencies in IPH. The DB DEA results also suggest that reducing private practice in public hospitals negatively influences cost efficiency. Furthermore, local frontier results are generated by truncating the data sample into various hospital groups by type and size. These results demonstrate that large Type-4 hospitals are the most cost-efficient hospital units, which deal with the highest volume and the most complex cases. There is also consistency in the results, where the smaller Type-2 and Type-3 hospitals, which provide emergency care in small urban or regional settings, are the least cost-efficient. Altogether, the findings indicate that, whether applying the full-sample DB DEA frontier or the local DB DEA frontier approach, significant cost savings and reductions in input use can be achieved in IPH.