Stroke Incidence and Case Fatality According to Rural or Urban Residence: Results From the French Brest Stroke Registry
Grimaud, O.; Lachkhem, Y.; Gao, F.; Padilla, C.; Bertin, Mélanie.; Nowak, E.; Timsit, S.
Stroke 50(10): 2661-2667
Background and Purpose- Recent findings suggest that in the United States, stroke incidence is higher in rural than in urban areas. Similar analyses in other high-income countries are scarce with conflicting results. In 2008, the Brest Stroke Registry was started in western France, an area that includes about 366 000 individuals living in various urban and rural settings. Methods- All new patients with stroke included in the Brest Stroke Registry from 2008 to 2013 were classified as residing in town centers, suburbs, isolated towns, or rural areas. Poisson regression was used to analyze stroke incidence and 30-day case fatality variations in the 4 different residence categories. Models with case fatality as outcome were adjusted for age, stroke type, and stroke severity. Results- In total, 3854 incident stroke cases (n=2039 women, 53%) were identified during the study period. Demographic and socio-economic characteristics and primary healthcare access indicators were significantly different among the 4 residence categories. Patterns of risk factors, stroke type, and severity were comparable among residence categories in both sexes. Age-standardized stroke rates varied from 2.90 per thousand (95% CI, 2.59-3.21) in suburbs to 3.35 (95% CI, 2.98-3.73) in rural areas for men, and from 2.14 (95% CI, 2.00-2.28) in town centers to 2.34 (95% CI, 2.12-2.57) in suburbs for women. Regression models suggested that among men, stroke incidence was significantly lower in suburbs than in town centers (incidence rate ratio =0.87; 95% CI, 0.77-0.99). Case fatality risk was comparable across urban categories but lower in rural patients (relative risk versus town centers: 0.76; 95% CI, 0.60-0.96). Conclusions- Stroke incidence was comparable, and the 30-day case fatality only slightly varied in the 4 residence categories despite widely different socio-demographic features covered by the Brest Stroke Registry.