The clinical significance of isolated systolic hypertension (systolic blood pressure greater than or equal to 160 mmHg and diastolic blood pressure less than 90 mmHg) has long been recognized, but its prevalence and correlates have not been well characterized. A community-based study was carried out by the Yang-Ming Crusade in 1987-1988 in Pu-Li Town, Taiwan. Of the 2573 registered residents over 30 years old, 1738 were interviewed, and their fasting blood samples were drawn and tested. The prevalence of isolated systolic hypertension was 2.1%. Age-specific prevalence increased with age. No significant difference was found between men and women. No trend was found at the urbanizational level. To study the significant correlates of isolated systolic hypertension, univariate analyses were applied first. Stratified analyses by age and by sex were used for interaction assessment. Based on the above findings as well as from the clinical point of view, logistic regression was used for multivariate analyses. Logistic regression analysis showed that after controlling the covariates simultaneously, four variables were significantly correlated with isolated systolic hypertension: age (greater than or equal to 50 vs. less than 50 years, OR = 3.4, 95% CI = 1.6-7.2); diabetes (yes vs. no, OR = 2.4, 95% CI = 1.2-4.7); blood urea nitrogen (greater than or equal to 25 vs. less than 25 mg/dl, OR = 2.1, 95% CI = 1.2-3.9); and physical activity (frequent vs. infrequent, OR = 1.8, 95% CI = 1.0-3.1). In comparison with definite (greater than or equal than 160/95 mmHg) and borderline (140/90-160/95 mmHg) hypertension as defined by WHO, the different sets of predictors and the possible adverse effect of frequent physical activity on isolated systolic hypertension were found and discussed.