OBJECTIVE: Medical treatment of healthy individuals with high normal blood pressure (BP) is recommended if there is subclinical organ damage (SOD). We examined which markers of SOD to use based on their supplementary prognostic value. METHODS: In a population sample of 1968 individuals, aged 41, 51, 61 and 71 years, without diabetes, prior stroke or myocardial infarction, not receiving any cardiovascular, antidiabetic or lipid-lowering medications, we measured urine albumin/creatinine ratio, carotid atherosclerotic plaques, carotid/femoral pulse wave velocity and left ventricular mass index. RESULTS: During a median follow-up of 12.8 years, the composite endpoint (CEP) of cardiovascular death, nonfatal myocardial infarction and stroke occurred in 153 individuals, of whom 32 had high normal BP. Presence of high normal BP was associated with increased risk of CEP [hazard ratio, 1.8 (95% confidence interval, 1.0-3.1; P = 0.046), optimal BP as reference group, adjusted for age and sex]. In the 337 individuals with high normal BP, using all four markers of SOD increased the sensitivity (number of CEPs in the group in which antihypertensive treatment was indicated divided by total number of CEPs) of the European Society of Hypertension risk classification chart significantly from 47 to 88% (P = 0.001) and the proportion of individuals in whom antihypertensive drug treatment was indicated from 22 to 57% (P < 0.001). Using two of pulse wave velocities of more than 12 m/s, atherosclerotic plaques or urine albumin/creatinine ratio of at least the 90th percentile did not produce significantly worse results. Seventy-five percent of individuals with three or more traditional risk factors had SOD. CONCLUSION: In healthy individuals with high normal BP, measuring two of pulse wave velocities, atherosclerotic plaques or urine albumin/creatinine ratio was sufficient to significantly improve risk prediction.