Aim To explore the relationship between change of ideal cardiovascular health score (△ICHS) and the ankle-brachial index (ABI) in the elderly population. Methods Through stratified random sampling, a total of 5852 cases were enrolled from 101510 Kailuan Group employees who participated in the health examination in 2006-2007, and among which 5440 individuals met the inclusion criteria. 1 physical examination every 2 years, and ABI was measured by ultrosoud in 2010-2011 examination. Excluding the incomplete data, 4524 cases were included for the final analysis. Ideal cardiovascular health score (ICHS) was performed in the population, according to the 7 cardiovascular health indexes proposed by the American Heart Association. According to the △ICHS of 2006-7,0-2011 examination, the relationship between △ICHS and ABI was analyzed by linear regression and Logistic regression. Results With the △ICHS changed from low to high, △SBP, △DBP, △BMI, △TC and △FBG all decreased (P＜0.05). The proportions of ideal salt habit and physical exercise population significantly increased by about 9.3% and 17.2% respectively, and the proportion of ideal smoking population decreased by about 4.6%. SBP, DBP, BMI, TC and FBG were slightly higher than that in baseline. With the increase of △ICHS, ABI gradually increased (P＜0.05). Linear regression analysis showed that △ICHS was positively related with ABI, and after adjusting for various confounding factors, the △ICHS increased by 1 point, ABI increased by 0.004 (B＝0.004, P＝0.004). Logistic regression analysis showed that after adjusting for various confounding factors, compared with the ICHS reducing group, the ABI abnormality of ICHS invariant group and ICHS elevating group revealed a downward trend. The △ICHS increased by 1 point, abnormal ABI risk reduced by about 21% (OR:0.9,5%CI:0.68-0.91). Conclusions △ICHS is positively correlated with ABI. △ICHS is an independent predictor of atherosclerosis in the elderly population.