Managers’ Experience and Efficiency of Regional & Selected Ghanaian Hospitals in Maternal & Newborn Health Service Provision

ABSTRACT

Despite the numerous studies on technical efficiency in the Ghanaian health sector, none has included regional hospitals, which are secondary referral hospitals.  Researching the efficiency of this group of hospitals is important given their critical role and the fact that they employ substantial amounts of public resources. This thesis provides evidence on the relative technical efficiency of all regional hospitals, plus some selected large district/municipal/metropolitan hospitals, in Ghana. It further examines the effect of managers’ experience on hospital technical inefficiency, while controlling for age, gender, and other institutional factors such as hospital size, the potential demand for maternal & newborn health service, and hospital ownership. Using panel data from the District Health Information Management System 2 (DHIMS 2) platform of the Ghana Health Service and primary data from the sampled hospitals, the study adopted a two-step estimation approach. In the first stage, relative technical efficiency scores were calculated using the Data Envelopment Analysis approach. Specifically, the Malmquist Total Factor Productivity index using a DEA frontier was employed due to the panel nature of the data. The second stage estimation employed a Tobit model to examine the variation in technical inefficiency among the sampled hospitals. Mean technical efficiency scores, under variable returns to scale assumption, were 0.868, 0.954, 0.860, 0.855, and 0.814 for the respective years between 2013 and 2017. There was generally a decline in mean technical efficiency scores over the study period. The percent of hospitals that were 100% technically efficient, under the assumption of variable returns to scale, decreased over the successive years - 53.33%, 51.11%, 46.67%, 44.44%, and 37.78%. More hospitals became less efficient over time. The key policy variable in the Tobit model estimation, hospital managers’ experience, was found to be a statistically insignificant determinant of hospital technical inefficiency; likewise, age. Hospital size is positively and significantly related to the level of hospital inefficiency, though with a small effect. Potential demand for maternal & newborn health services of a hospital was statistically significant with an expected negative sign. Mission hospital ownership was statistically insignificant though, a positive sign of its coefficient points to Government hospitals as comparatively more efficient. 

The study thus recommends: an expanded scope of measuring management experience/practices to re-examine its effect on efficiency, and intra-region redistribution of hospital beds across the country; integration of health facilities efficiency measurement into the routine data and health sector performance monitoring system - as opposed to setting up a parallel system; a qualitative study on the management systems of fully efficient hospitals and the least efficient ones, to provide qualitative insights on factors driving success in efficiency improvement and root-causes of poor performance; the creation of a platform for peer to peer learning and sharing of best practices on efficiency improvement for health facilities; and hospitals should design innovative strategies for deeper penetration into communities to increase output coverages.