Technical efficiency analysis of provincial hospitals under two different contracting modalities in Afghanistan

Abstract

The study aimed to measure the Technical Efficiency (TE) of delivering health services in provincial hospitals under the contracted-in and contracted-out modalities in Afghanistan. Input oriented Data Envelopment Analysis (DEA) was employed to measure the TE in the ten provincial hospitals, five from each contracting modality. Using the HOSPICAL cost allocation tool cost analysis of these hospitals was carried out, in order to support the TE score, with other KPIs. The results showed that in the five contracted-in hospitals the average cost per bed day was US$20.3 and each OPD costed on average US$3.1/visti. The average BOR found 81% with the average length of 3.1 days ALOS. On the other hand the averge cost of each bed day was US$23.8 in the five cotracted-out hospitals, the average OPD cost was found US$1.9/visit in the five contracted-out hospitals. The overall average BOR and ALOS for these hospitals were 68% and 2.9 days. DEA analysis revealed that out of 10 hospitals in the study 6 of them were efficient with the TE of 100%; 3 from each contracting modality. The average TE score for the five contracted-in hospitals was 0.94, while average TE score for the five contracted-out hospitals was 0.96, which shows relatively higher efficiency in the contracted-out modality hospitals. Results of the Tobit regression analyses indicated that from the five independent variables ( BOR, ALOS, Cost/OPD, Cost/Bed day and dummy vairable for contracting type) that were regressed against the technical efficiency scores of the hospitals, only BOR and cost/OPD visit had positive relationship. But none of them was quite significant, which could be due to small sample size.