Technical, scale, and size efficiency in nursing home care: a nonparametric analysis of Connecticut homes

Maindiratta (1990) questioned the usefulness of the concept of scale efficiency in production as the most productive scale size (MPSS) usually requires altering the scale of output produced. If the decision making unit (DMU) is required to deliver a specific output bundle, then altering output along with the input scale to reach the MPSS is not a valid recommendation. He proposed a measure of the size efficiency of a DMU. In this paper, we apply Data Envelopment Analysis (DEA) to examine the levels of technical, scale, and size efficiency of individual nursing homes providing health care to the elderly. The data used relate to the operations of 140 nursing homes from Connecticut, USA during the year 1982-83. Maindiratta's model is input-oriented. By contrast, our study is output-oriented and we appropriately reformulate Maindiratta's model. The findings show that in several cases size efficiency is less than unity. This suggests that the most efficient production of output would require restructuring of the nursing home under investigation as more than one small unit. We also compare the efficiency levels of 'for-profit' homes with those of 'not-for-profit' homes.