Objective To investigate the relationship of brachial-ankle pulse wave velocity (baPWV) with cardiocerebral vascular events in the elderly. Methods A total of 10 312 employees of Kailuan Group who aged ≥60 years and completed physical examinations and baPWV test in 2006,2008, 2010,2012, 2014 and 2016 were included in the study. After the subjects with previous histories of cardiocerebral vascular disease (CVD) or atrial fibrillation or with incomplete blood pressure data were excluded, and 9 586 subjects were finally included in the analysis. According to the tertile of baPWV in each group, the subjects were divided into Q1 tertile, Q2 tertile and Q3 tertile groups. SAS 9.4 software was used for data analysis. According to the data type, one-way ANOVA, Kruskal-Wallis rank sum test andχ² test were used to compare the data between groups. Kaplan-Meier analysis was used to calculate the cumulative incidence of cardiocerebral vascular events, and Log-rank test was adopted to compare the cumulative incidence of cardiocerebral vascular events in different tertile groups. Cox proportional hazard model was applied to analyze the effect of baPWV on cardial cerebrovascular and stroke events. Results (1) During the average follow-up of(5.68±2.68) years, there were 538 cases having cardio-cerebrovascular events and 434 cases having stroke. The cumulative incidence of cardio-cerebrovascular events was 5.17%, 8.72% and 11.53%, respectively, in the Q1, Q2 and Q3 tertile groups. Log-rank test showed that there were significant differences in the cumulative incidence of terminal events among the 3 group (P<0.001). The cumulative incidence of stroke was 4.16%, 7.37% and 9.27%, respectively in the 3 groups. Log-rank test indicated that there were significant differences in the cumulative incidence of end events among the 3 groups (P<0.001).(2)Cox regression analysis suggested that after adjusting for other confounding factors, the HR (95%CI) of cardiocerebral vascular events and stroke events were 1.60 (1.26-2.03) and 1.72 (1.31-2.25) in Q2 tertile group, and 2.21 (1.74-2.80) and 2.39 (1.8-3.13) in Q3 tertile group when compared with those in Q1 group. For each additional standard deviation of baPWV, the HR (95%CI) of cardiocerebral vascular events and stroke events were 1.21 (1.12-1.32) and 1.22 (1.13-1.34), respectively. Conclusion baPWV is an independent risk factor for cardiocerebral vascular events in the elderly.