Operational Efficiency Analysis of Nursing Homes and Institutions in Taiwan: Data Envelopment Analysis

Hence, this shows that the government plays an important role in facilitating financial supports of charted nursing homes and institutions. Thus, whether or not nursing homes and institutions are operated efficiently is an important issue worth emphasizing and discussing. Efficiency is defined as either the institution in question is operated at the minimum cost point of input combination given fixed output, or the maximum output combination given fixed cost.

According to the 70 charted nursing homes and institutions in 1996, questionnaire survey is conducted to the manager of each nursing and institution. After considering the DEA input and output specifications, 55 valid sample are obtained. The input variables include: (1) number of doctors and nurses; (2) number of social workers; (3) number of administrators;(4) number of monitoring employees; (5) number of ambulances and transportation vehicles; (6) number of beds.

The output variables include: (1) actual number of inpatients; (2) full, half, and none self-caring inpatients; (3) occupancy rates; (4) average number of servicing; (5) frequency of accidents. In order to perform DEA sensitivity analysis, the 6 inputs and 5 outputs are recombined to form 12 different input-output combinations. Empirical results shows that there are 18 homes and institutions evaluated efficient among 55 valid sample using actual number of inpatients as the output measure.

However, there are 27 homes and institutions evaluated efficient among 52 valid sample using full, half, and none self-caring inpatients as the output measure. Hence, DEA model is sensitive to the input and output specifications. Nonparametric analysis shows that the scale and type of service of nursing home and institution significantly affect relative efficient score of each home and institution with the degree of influence of the former is greater.

Finally, the Censored Tobit analysis shows that, at the 1% level of significance, occupancy rate is positively related to each home and institution\'s operational efficiency among all input-output specifications. In addition, at the 5% level of significance, those institutions with province-supervised, publicly owned, and medium scale are evaluated to have lower relative efficient scores. Further direction of research may consider different quality variables and use panel data to objectively evaluate operational efficiency and productivity changes of nursing homes and institutions.