Investigating health system performance: An application of data envelopment analysis to Zambian hospitals

BACKGROUND: Zambia has recently articulated an ambitious national health program designed to meeting health-related MDGs. Public expectations are high and Zambia continues to receive significant resources from global and bilateral donors to support its health agenda. Although the lack of adequate resources presents the most important constraint, the efficiency with which available resources are being utilised is another challenge that cannot be overlooked. Inefficiency in producing health care undermines the service coverage potential of the health system. This paper estimates the technical efficiency of a sample of hospitals in Zambia. METHODS: Efficiency is measured using a DEA model. Vectors of hospital inputs and outputs, representing hospital expended resources and output profiles respectively, were specified and measured. The data were gathered from a sample of 30 hospitals throughout Zambia. The model estimates an efficiency score for each hospital. A decomposition of technical efficiency into scale and congestion is also provided. RESULTS: Results show that overall Zambian hospitals are operating at 67% level of efficiency, implying that significant resources are being wasted. Only 40% of hospitals were efficient in relative terms. The study further reveals that the size of hospitals is a major source of inefficiency. Input congestion is also found to be a source of hospital inefficiency. CONCLUSION: This study has demonstrated that inefficiency of resource use in hospitals is significant. Policy attention is drawn to unsuitable hospital scale of operation and low productivity of some inputs as factors that reinforce each other to make Zambian hospitals technically inefficient at producing and delivering services. It is argued that such evidence of substantial inefficiency would undermine Zambia's prospects of achieving its health goals.