Rurality and avoidable hospitalization in a Spanish region with high population dispersion.

BACKGROUND This study analyses the association between rurality and local rate of avoidable hospitalizations in a Spanish region with high population dispersion. METHODS Ecological study using a municipality in the region of Castile and Leon (Spain) as the spatial unit of analysis. The variables used to operationalize rurality included the following: distance to hospital, population density, mean socio-economic level and percentage of the population aged >65 years. We calculated relative risk (RR) and 95% confidence intervals (CI) using the conditional autoregressive spatial model proposed by Besag, York and Mollié, with explanatory variables. RESULTS The number of avoidable hospitalizations was 9923 or 4.5% of all admissions. The age- and gender-adjusted avoidable hospitalization rate was 4.06 per 1000 persons. Spatial analysis showed that two variables, distance from municipality of residence to reference hospital and percentage of population aged >65 years were inversely associated with risk for avoidable hospitalization [RR=0.996 (95% CI 0.993-0.999) and RR=0.989 (95% CI 0.982-0.996), respectively]. CONCLUSIONS It is important to determine whether these lower avoidable hospitalization rates reflect an adequate level of accessibility and quality of primary care health services for rural populations or, in the contrary, they reveal access barriers to hospital care.