Benchmarking mechanical ventilation services in teaching hospitals

The purpose of this investigation was to examine the technical efficiency of mechanical ventilation nonsurgery (DRG 475) among University Hospital Consortium (UHC) hospitals that consists of volunteer, teaching hospitals across the nation. The data for this study was retrieved from the 1997 UHC database that includes charge and discharge information for 69 hospitals. Data on 7961 patients classified with mechanical ventilation were aggregated to the hospital level. We retained data from a total of 62 hospitals, the other seven hospitals had missing data. The research questions were (1) Do UHC hospitals differ significantly in their efficiencies in the treatment of mechanically ventilated patients? (2) What inputs and outputs contribute most to the inefficiencies associated with mechanical ventilation? Of the 62 hospitals analyzed using data envelopment analysis technique, 10 were considered efficient and 52 were inefficient as compared to their benchmark peers. Efficient and inefficient hospitals did significantly differ between the transferred output variable and between the respiratory, laboratory, and radiology input variables. All inputs demonstrated excessive resource utilization among inefficient hospitals as compared to efficient hospitals. A total reduction of about $19 million dollars in ancillary services would need to occur for inefficient hospitals to approach the frontier of efficient hospitals. This study demonstrates that mechanical ventilation is costly, yet the specified ancillary services are capable of being reduced yielding technical efficiency as demonstrated by 10 efficient hospitals.