Geographical patterns of cardiac arrests: An exploratory model

Abstract

The geographical distribution of out-of-hospital cardiac arrest has not been studied but is of importance both epidemiologically and programmatically, for the planning of pre-hospital emergency care. In this study, 525 cardiac arrests in Seattle are sampled and the census tract of their occupance noted. A predictive model is developed to explain the geographical distribution of the cardiac arrest cases. The regression model indicates a high degree of statistical explanation (R² = 0.94), based upon 5 independent variables. Using population alone as an independent variable, the model is only marginally less powerful (R² = 0.91). The study concludes that such a prediction model is of use in the geographical allocation of emergency units based upon response time minimization.