Technical Efficiency of Kenyatta National Hospital

Abstract

This study uses Data Envelopment Analysis (DEA) and Malmquist Productivity Index (MPI) to estimate the degree of technical efficiency and levels of productivity change at Kenyatta national hospital. The study has also used Tobit regression to identify the factors responsible for efficiencies in the use of resource inputs among the units that constitute the clinical section of KNH. Panel data for eleven years (2000-2010) was used. The resource inputs used were: number of doctors, number of registrars, number of nurses (enrolled/registered and community nurses), number of administrative officers and number of beds, while outputs used were number of outpatients and number of inpatient admissions. The explanatory variables for the Tobit model were: bed occupancy rate (%); average length of stay (days); ratio of the sum of doctors and nurses to number of inpatient admissions number of beds and outpatient visits as a proportion of inpatient admissions. Tested at 95% confidence interval, bed occupancy rate and outpatient visits as a proportion of inpatient admissions were found to be insignificant, while the rest of the explanatory variables were significant. 40% of the decision making units were found to be were technically efficient while 60% were technically inefficient. The mean technical efficiency was 77%. There is need to undertake this kind of study to other hospitals in the country-where it has not been done, whether public or private- as doing so would aid the ongoing health sector reforms.