Evidences from the literature search indicate that there have been limited studies in the area of measuring efficiency in health care delivery in the developing nations of Africa. This is somewhat not encouraging given the scarcity of health resources in the continent and the fact that inefficient utilisation of these scarce resources exacts higher penalty in terms of forgone health benefits. Plausible explanations for this glaring neglect of research into efficient mode of care delivery in the continent may be found, partly, in the lack of appropriate data for such studies and poor appreciation of statistical data by managers of the health system of most African countries.
The few existing studies in hospital efficiency in Africa principally used data envelopment analysis as their major analytical tool. The appropriateness of data envelopment analysis for these studies hinged on its capacity to handle multiple inputs and outputs, non-specification of functional production form relating inputs to outputs and ability to produce accurate result with small samples are some of the reasons that endeared data envelopment analysis (DEA) to health care researchers in Africa. Indeed, Norman and Stoker (1991) have indicated that in many cases particularly public sector organizations there are no known functional form.
Since most of these studies were largely focused on public health facilitie,s the preference for DEA is somewhat justified.
However, Wrouters (1990) employed econometric approach to study the costs and efficiency of a sample of 42 private and public health facilities in Ogun State. The sample for the study included a heterogeneous range of facilities which comprised of comprehensive health centers, primary health care clinics, maternities, health clinics, and dispensaries. Wrouters analyzed costs and efficiency, estimating a production and cost function, and deriving
associated measures of efficiency. Technical efficiency was assessed by estimating a production function and deriving measures of marginal product of health workers.
So far, data envelopment analysis approach has been applied to health facilities in only few countries in Africa. The concentrations of the studies are more in the southern African region than elsewhere in the continent. Kirigia, et al (2001) studied 155 primary health care clinics in Kwazulu-Natal province in South Africa. The study found 70 percent of the clinics studied to be technically inefficient. Similarly, in 2002 Kirigia assessed the technical efficiency of 54 public hospitals using DEA methodology in Kenya and found that 26 percent of the hospitals were technically inefficient. The study singled out inefficient hospitals and provided the magnitudes of specific inputs reduction or output needed to attain technical efficiency.
Zere (2000) investigated hospital efficiency in South Africa using DEA and DEA based malmquist productivity index. In 2006, Zere leading other health researchers assessed the technical efficiency of 30 district hospitals in Namibia using DEA. Recurrent expenditures, beds and nursing staff were used as inputs in the DEA model while outpatients visit and inpatient days were used as the model’s output. Findings from the study suggested the presence of substantial degree of pure technical and scale efficiency with increasing returns to scale being the predominant form of inefficiency observed.
Another study in Angola assessed technical efficiency and changes in productivity in the nation’s public municipal hospitals. The study based on a three-year panel data from 28 public municipal hospitals found an increase in productivity by 4.5 percent over the period 2000-2002. The increased productivity was attributable to efficiency rather than innovation (Kirigia, 2008). Indeed, in a resource poor countries where deployment of additional resources to any sector of the economy, in the face of competition from other sectors, could be problematic, increased efficiency should be a natural response to raising outputs.
Further in the Southern African axis, Masiye, et al (2006) used data envelopment analysis to estimate the degree of technical, allocative and cost efficiency in private and public health centres in Zambia. The authors’ interest was to research the efficient management of human
resources in the health centres in Zambia. And, of the few studies in Africa, this work appeared to be the only one that included private-owned facilities in the sample studied. The study found private facilities to be more efficient than the public facilities. Indeed, about 88 percent of these facilities were found to be both cost and allocatively efficient; 83 percent of the 40 health centres in the study were technically efficient.
In addition, Masiye (2007) investigated the Zambian health system performance using the DEA methodology. Data gathered from 30 hospitals on institutional expended resources and output profiles indicated that Zambia hospitals were operating at 67 percent level of efficiency: which implied that significant resources were being wasted in the Zambian health system. Forty percent of the hospitals investigated were found to be efficient. However, input congestion and size of the health facilities were found to be a major source of the inefficiency observed in the health system. It seems worrisome that size could be a problem or a major cause of resource wastage in any health system in Africa when viewed against the need to expand health service provision to a significant proportion of the population. Size, however, may remain a problem if political considerations are given priority above the overall interest of the nation with regards to locating health facilities in places that the best health interest of the populace could be best served.
Research evidence exists of hospital efficiency study in Botswana. Thekke, et al (2003) presented relative efficiency indices for the services rendered by health districts and specific hospitals in Botswana. The study which covered 22 health districts and gathered data on 13 hospitals combined stochastic frontiers analysis and data envelopment analysis in analysing the efficiencies of the facilities studied. Indeed, this study stands out as the only one to have used the two methodologies, even though data envelopment analysis was considered superior. Result of the analysis indicated that three districts have efficiency score of less than one, that is, inefficient. Trends in these reviewed studies are that most of the studies were conducted by researchers outside the academics and/or are based outside the shore of Africa with few members of the research team being African based.
It could be said that there have been studies, though scanty, on efficiency in health care delivery in the southern part of Africa. However, interests in health care efficiency and studies in this direction haves been quite limited elsewhere in Africa. Ghana and Sierra Leone furnished a ready example of countries outside the southern African sub-region that have cases of studies on health care efficiency. Kwakey (2004) effort in Ghana is more of a pioneering study on health or hospital efficiency in the West African sub-region. He employed DEA to measure the relative efficiency of 20 selected hospitals in Ghana in 2004 which suggests that the history of DEA application in West Africa is relatively recent. His study was followed by Osei,et al (2005) which was a pilot study based on data from public health centres and 17 public hospitals in Ghana. The study indicated that 47 percent of the hospitals were technically inefficient and ten (10) or 59 percent of these were scale inefficient.
Furthermore, of the 17 health centres studied, 18 percent were found to be technically inefficient with 8 health centres been scale inefficient. The sample size of the health facilities on both sides of hospital and health centres was deemed too small to permit generalisation of the result from the study for the whole country. Another study was conducted based on a larger sample size. Akazili, et al (2008) using DEA focussed on the efficiency of public health centres in Ghana with the objective of determining the degree of efficiency of these centre and recommending performance targets for the inefficient ones. The study based on a sample size of 89 health centres showed that as much as 65 percent of these facilities were technically inefficient, that is , using resources that they did not actually need.
Similarly, another study in Sierra Leone equally applied data envelopment analysis to measure both the technical and scale efficiency of a sample of public peripheral units in Sierra Leone (Renner, et al, 2005). In the tradition of revealing poor resource usage in most health systems of African countries, the study revealed that 59 percent of the 37 peripheral health units were technically inefficient and 65 percent been scale inefficient. The implication of these inefficiencies in the health systems of African countries lies in the limitations it imposes on government in extending care accessibility to the population. It sounds credible that we should be questioning the issue of scarcity of health resource in our
care system. These studies in Ghana and Sierra Leone appeared, to the best of our knowledge, to be the few cases of health care efficiency studies outside the South Africa sub- region.