Background and Purpose Stroke has been the second leading cause of dea
th for all ages in Taiwan since 1983. However, despite the severity of
the threat, stroke prevalence in Taiwan has not yet been addressed in
a nationwide survey. In this study, the stroke prevalence in Taiwan w
as investigated using data from the 1994 National Health Interview Sur
vey. Methods This nationwide survey sought to obtain a nationally repr
esentative sample of households in Taiwan by using three-stage stratif
ied random sampling with a probability proportional to size. In the fi
rst stage, 58 townships were selected, from the 359 townships in Taiwa
n, according to their administrative structure and level of socioecono
mic development. In the second stage, 149 basic administrative regions
(tsun or ii) were selected from the selected 58 townships. Finally, 3
814 households were selected from the 149 selected tsuns or lis. Field
interviews were carried out between October 1994 and December 1994. F
ollow-up interviews with families of stroke patients were made 2 years
later.Results Of the selected households 3119, or 81.8%, responded. A
total of 11 925 persons were interviewed, and 71 of them were stroke
patients, with a crude point prevalence rate of 5.95 per 1000. The str
oke prevalence increased steadily with age, from 0.51 per 1000 in pers
ons aged 35 to 44 years to 113.6 per 1000 in persons aged 85 years or
over. There was a weak association with higher stroke prevalence for p
ersons living in eastern Taiwan or those of lower educational level. T
he overall male/female prevalence ratio was 1.17. The results of follo
w-up interviews showed a cumulative mortality rate of 25.4% within the
2-year period and an ischemic/hemorrhagic stroke ratio of 1.33 in 35
patients whose stroke type could be validated. Conclusions Age was the
most important factor correlating to stroke prevalence. On the basis
of this result, planning policies and programs for stroke prevention i
n Taiwan should give a higher priority to (1) aggressive primary preve
ntion for aged people who are apparently at higher risk of stroke and
(2) early reduction of stroke risk factors in younger aged people whos
e immediate risk of stroke is lower but would increase significantly w
ith age.