Measuring efficiency in primary health care: the effect of exogenous variables on results
- SourceAvailable from: José Manuel Cordero Ferrera
[Show abstract] [Hide abstract]
- "Actually, most of the studies that have attempted to incorporate this information have been limited to performing a secondstage analysis in order to identify potential explanatory factors of inefficient behavior, but they have not incorporated the effect of these variables into the efficiency scores (Zavras, Tsakos, Economou, & Kyriopoulos, 2002; Kontodimopoulos, Moschovakis, Aletras, & Niakas, 2007; Ramirez-Valdivia, Maturana, & Salvo- Garrido, 2010). More recently, Kontodimopoulos, Papathanasiou, Tountas, and Niakas (2010) and Cordero, Crespo, and Murillo (2010) estimated corrected efficiency scores including information about the characteristics of the population covered by each primary care center using two traditional approaches such as the three-stage model (Muñiz, 2002) and four-stage model (Fried, Schmidt, & Yaisawarng, 1999), which adjust original input and output values to obtain a measure of managerial inefficiency that controls for the effect of exogenous factors. However, as we mentioned in the previous section, there have not been previous empirical studies incorporating this information through a conditional nonparametric model. "
ABSTRACT: This paper uses a fully nonparametric approach to estimate efficiency measures for primary care units incorporating the effect of (exogenous) environmental factors. This methodology allows us to account for different types of variables (continuous and discrete) describing the main characteristics of patients served by those providers. In addition, we use an extension of this nonparametric approach to deal with the presence of undesirable outputs in data, represented by the rates of hospitalization for ambulatory care sensitive condition (ACSC). The empirical results show that all the exogenous variables considered have a significant and negative effect on efficiency estimates.European Journal of Operational Research 01/2015; 240(1):235–244. DOI:10.1016/j.ejor.2014.06.040 · 1.84 Impact Factor