Towards more realistic measures of accessibility to emergency departments in Sweden

Assuring that emergency health care (EHC) is accessible is a key objective for health care planners. Conventional accessibility analysis commonly relies on resident population data. However, the allocation of resources based on stationary population data may lead to erroneous assumptions of population accessibility to EHC. Therefore, in this paper, we calculate population accessibility to emergency departments in Sweden with a geographical information system based network analysis. Utilizing static population data and dynamic population data, we investigate spatiotemporal patterns of how static population data over- or underestimates population sizes derived from temporally dynamic population data. Our findings show that conventional measures of population accessibility tend to underestimate population sizes particularly in rural areas and in smaller ED's catchment areas compared to urban, larger ED's-especially during vacation time in the summer. Planning based on static population data may thus lead to inequitable distributions of resources. This study is motivated in light of the ongoing centralization of ED's in Sweden, which largely depends on population sizes in ED's catchment areas.