A Multiple Stage Approach for Performance Improvement of Primary Healthcare Practice

Abstract

Chilean primary healthcare practice is analyzed using a Data Envelopment Analysis (DEA) multiple stage approach. We estimate the efficiency level of 259 municipalities nationwide. Since the efficiency score by itself is of limited value for decision making, we use a multivariate tool to help explain the effect of relevant factors. First, we use a cluster analysis to homogenize the units under study. Second, we use DEA to estimate the efficiency levels, which varies from 61% to 71% for urban municipalities, and from 51% to 56% in rural ones. Third, we use bootstrap to estimate confidence intervals for the efficiency scores, and a Biplot method to identify adequate variables to include in the Tobit Model, which is our last stage. We identify six factors associated with rural municipalities' operational efficiency, and two with urban ones. Knowing the efficiency level of municipalities can help determine ways to improve their efficiency.

Factors affecting primary health care centers’ economic and production efficiency

“…Finally, and for the economic production model case a rather interesting result worth mentioning is the negative and significant effect that the population density (POPDEN) variable exerts leading to lower technical efficiency scores in contrast with (POPGR) and the production model. The specific finding is in accordance with similar studies (Ramírez-Valdivia et al 2011;Pelone et al 2013) and can be explained in terms of increased demand for health services in these areas and greater pressure that this demand creates (Cordero-Ferrera et al 2011).…”

“…Therefore in the last few years several studies have focused in a two-stage approach to compare efficiencies (Cordero-Ferrera et al 2011;Linna et al 2003;Kontodimopoulos et al 2007;Ramírez-Valdivia et al 2011). However, all of these studies employ other conventional procedures of regression (Tobit estimator, OLS, etc) and did not make inferences based upon DEA results with the two stage bootstrapping procedure that are used in this paper.…”

“…The argument behind the inclusion of socio-demographic factors was bounded to the idea of higher pressure on demand and the consequent impact on HC's technical and scale efficiency. In this study, we analyze differences in HCs technical and scale efficiency by considering population density (POPDEN) (Pelone et al 2013;Cordero-Ferrera et al 2011), population growth (POPGR) (Ramírez-Valdivia et al 2011;Linna et al 2003) and the proportion of inhabitants working on agricultural sector (AGROPOP) (Cordero-Ferrera et al 2011).…”

“…Rurality (Puenpatom and Rosenman 2008;Ramírez-Valdivia et al 2011;Marathe et al 2007) is one of the factors that can impede or facilitate HC's efficiency, while catchment population can be considered as the second one (Kontodimopoulos et al 2007;Zavras et al 2002). Relative to the physical establishment of HCs that located in cities with 50,000 or more inhabitants or in remote areas we included the variable LOCATION that classifies the HCs as rural or urban.…”

Primary Care Efficiency Measurement Using Data Envelopment Analysis: A Systematic Review

“…Largely, DEA has been used to measure the relative efficiency of a wide range of activities provided by different PC centers or districts within national/regional PC systems [21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36]. In a large number of studies DEA has also been used to compare PC providers in delivering specific activities (e.g.…”