Increased stroke incidence in Lund-Orup, Sweden, between 1983 to 1985 and 1993 to 1995.

BACKGROUND AND PURPOSE Some studies suggest that the incidence of stroke may continue to change. We examined recent temporal trends in a defined geographical area of southern Sweden. METHODS Medical records at the University Hospital of Lund (hospital district population 224 126 in 1993) were retrospectively screened for possible first-ever strokes during 1993 to 1995. Included patients were classified into pathological subtypes (cerebral infarction, intracerebral hemorrhage, subarachnoid hemorrhage, and undetermined pathological type) and according to the Oxfordshire Community Stroke Project (OCSP) classification system. Stroke patients from a previous study from 1983 to 1985 in the same area were reevaluated with the same criteria. Epidemiological data for the 2 time periods were compared. RESULTS There were 998 patients with first-ever stroke in 1983 to 1985 and 1318 in 1993 to 1995. The total incidence rate per 100 000 person-years (age-adjusted to the European population) increased from 134 (95% confidence limits [CL] 126 to 143) to 158 (95% CL 149 to 168). The incidence rate for patients <75 years of age increased from 94 (95% CL 85 to 103) in 1983 to 1985 to 117 (95% CL 108 to 127) in 1993 to 1995, whereas the incidence rate for patients >/=75 years was stable. The age-adjusted incidence rates for the OCSP subtypes lacunar and posterior circulation syndromes increased significantly, by 30% and 55%, respectively. CONCLUSIONS A marked increase in the incidence of first-ever stroke was observed, surprisingly mainly confined to people <75 years of age. The underlying causes of this increase must be explored in future studies.