BACKGROUND Despite an association between visit-to-visit blood pressure (BP) variability (VV-BPV) and cardiovascular (CV) outcomes, many studies performed during the past 4 years have shown conflicting results. This study investigated the impact of VV-BPV on CV outcomes in the Korean National Health Insurance Service (NHIS) database-National Sample Cohort. METHODS From the 2002 Korean NHIS database (n = 47,851,928), sample subjects with 3 or more BP measurements (n = 51,811) were divided into 2 groups according to a 10 mm Hg cutoff in the SD of systolic BP (SD-SBP). The CV outcomes of these groups were compared by sensitivity analyses using various sampling methods. RESULTS Irrespective of sampling method, subjects with SD-SBPs ≥10 mm Hg had higher rates of CV events or death, nonfatal myocardial infarction (MI) or stroke, and total mortality, but were not associated with CV mortality. The hazard ratios for CV events or death, nonfatal MI or stroke, CV mortality, and total mortality were 1.43 (95% confidence interval [CI], 1.25-1.63, P < 0.01), 1.45 (95% CI, 1.27-1.65, P < 0.01), 1.32 (95% CI, 0.89-1.94, P = 0.17), and 1.18 (95% CI, 1.01-1.38, P = 0.04), respectively. CONCLUSIONS Increased VV-BPV was an independent risk factor for future CV outcomes, independent of mean BP status, even in normotensive subjects and in all subgroups, except females. Similar VV-BPV values in the sensitivity analyses suggest VV-BPV is a reproducible phenomenon, reflecting the various types of intrinsic physiologic properties.