To evaluate the association between blood pressure variability (BPV) and mortality in the elderly, all blood pressure measurements recorded in a cohort of individuals 65 years and older were collected and the association between BPV coefficient of variation (BPV divided by mean arterial pressure) was calculated. Mortality during a 10-year period was compared between BPV coefficient of variation quartiles. Overall, 39 502 individuals 65 years and older were included in the analysis, of which 31 737 (80.3%) were hypertensive; 12 817 (32.4%) individuals died during the study period. Mortality was lower in the second and third blood pressure quartiles compared with the first quartile in both the normotensive and hypertensive groups. In both normo-tensive and hypertensive individuals, mortality was higher in the fourth quartile, but it was more pronounced in normotensive individuals (odds ratio, 1.18; 95% confidence interval, 1.06-1.31 in hypertensive individuals vs odds ratio, 1.27; 95% confidence interval, 1.17-1.37 in normotensive individuals). High and low BPV are associated with mortality in both hypertensive and normotensive elders. 1 | INTRODUCTION Although mean arterial pressure is probably the single most important factor associated with cardiovascular morbidity and mortality, systolic blood pressure (SBP) variability (BPV), defined as variation in SBP levels over a period, has been proposed to play a pivotal role in the association between blood pressure (BP) control and cardiovascular morbidity and mortality. Systolic BPV has been found to be associated with mortality and adverse cardiovascular outcomes in several studies 1-3 but not in others. 4 Data regarding the association of long-term systolic BPV with mortality in the elderly are lacking. In this paper, the association between long-term systolic BPV and mortality was evaluated in a large cohort of hypertensive and normotensive community-dwelling individuals 65 years and older. 2 | PATIENTS AND METHODS 2.1 | Setting and patients Clalit Health Services (CHS) is the largest health management organization in Israel, with more than 4 million members. Its comprehensive computerized data warehouse aggregates and stores continuous real-time input from physicians and health service providers, including demographic parameters, medical diagnoses, findings from in-hospital and outpatient laboratory tests, and medications dispensed, for each of its members. The cohort for the present retrospective study consisted of individuals insured by CHS-Tel Aviv District branch, which insures a mainly urban and Jewish population. All individuals who had at least one available BP measurement in 2002 were identified, but only those in whom BP was recorded at least five times in different years were included in the analysis. Systolic BPV was calculated as the standard deviation of SBP from the mean of all SBP measurements (normally distributed) recorded during the follow-up period (2002-2012). For the purpose of analysis, we used the BPV coefficient of variation (CV). This val...