Stroke incidence and survival in the beginning of the 21st century in Southern Sweden - Comparisons with the late 20th century and projections into the future

Background and Purpose-We report trends of stroke incidence and survival up to year 2001/2002 in Lund-Orup, Sweden, and projections of future stroke incidence in Sweden. Methods-Lund Stroke Register, a prospective population-based study, included all first-ever stroke patients, between March 1, 2001 and February 28, 2002, in the Lund-Orup health care district. Institution-based studies for 1983 to 1985 and 1993 to 1995 were used for comparison. We calculated age-standardized incidence and Cox proportional hazards analysis of survival (stroke subtype, sex, age group, and study period in the analysis). Minimum follow-up was 46 months. Based on our register's stroke incidence and the official Swedish population projection, a projection for future stroke incidence on a national basis was calculated. Results-We included 456 patients with first-ever stroke in 2001/2002. The age-standardized incidence (to the European population) was 144 per 100 000 person-years (95% CI 130 to 158) in 2001/2002, 158 (95% CI 149 to 168) in 1993 to 1995, and 134 (95% CI 126 to 143) in 1983 to 1985. Cox proportional hazard analysis indicated decreased risk of death after stroke in 2001/2002 (hazard ratio 0.80; 95% CI 0.67 to 0.94) compared with 1993 to 1995. Up to year 2050, the annual number of new stroke patients in Sweden may increase by 59% based solely on demographic changes. Conclusions-Despite possible underestimation of stroke incidence during the previous institution-based studies, the increased stroke incidence between 1983 to 1985 and 1993 to 1995 did not continue in 2001/2002. The long-term survival after stroke continues to improve. As the elderly population is growing in Sweden, stable incidence and increasing survival will result in a rapidly increasing prevalence of stroke patients in Sweden.