Objectives To examine the extent to which contextual factors explain emergency department (ED) visits and ED revisits, additional to that explained by individual factors.

Design A register-based prospective cohort study.

Setting Swedish region of Dalarna.

Participants Participants were 16 543 community-living adults aged 80 or older who were residents of the Dalarna region of Sweden, excluding older adults who moved out of Dalarna or into residential care during the study period.

Outcome measures Dependent variables were initial ED visit, and at least one ED revisit within 30 days of an initial ED visit.

Results Approximately 36% of the participants visited the ED during the study period with 18.9% returning to the ED within 30 days. For both initial ED visits and ED revisits, the addition of contextual factors to models containing individual factors significantly improved model fit (p<0.001; p<0.022) and the amount of variance explained in the outcome. In the final models, initial ED visit was significantly associated with older age, number of chronic diseases, receipt of home help, number of primary care visits, proportion of 80+ in the population and shorter distance to the ED; while an ED revisit was significantly associated with greater use of social care, number of hospital admissions and disposition (discharged; admitted to hospital) at initial ED visit.

Conclusion Contextual factors explain variance in initial ED visit, additional to that explained by individual factors alone, which indicates inequitable access to ED care. These findings suggest considering local variations in contextual factors in order to improve health-related outcomes among older adults.