General characteristics at baseline.

Background and aims
There is a demand for longitudinal studies that use both objective and subjective measures of physical activity to investigate the association of physical activity with the change in carotid intima-media thickness (CIMT). In order to investigate such association, we conducted an 8-year follow-up study that used both objective and subjective measures of physical activity.
Methods
This cohort study used subsamples of the ongoing Korean Genome and Epidemiology Study (KoGES). Included participants were between 49 to 79 years of age at baseline. Exclusion criteria included incomplete assessments of pedometer/accelerometer, international physical activity questionnaire (IPAQ), and baseline CIMT. Participants with a history of cardiovascular diseases were further excluded. Linear regression models were used for the main analysis. Age differences were assessed by stratifying the participants into < 60 years and ≥ 60 years.
Results
After removing excluded participants, 835 participants were included in the final analysis (age, 59.84 ± 6.53 years; 326 (39.04%) males). 453 participants were < 60 years and 382 participants were ≥ 60 years. The daily total step count was inversely associated with the percent change in overall CIMT over 8-years (β = -0.015, standard error = 0.007, P = 0.034). This association was present among participants in the < 60-year-old group (β = -0.026, standard error = 0.010, P = 0.006), but not among participants in the ≥ 60-year-old group (β = -0.010, standard error = 0.011, P = 0.38).
Conclusions
The findings suggest that taking preemptive actions of increasing physical activity may prevent the incidence of atherosclerosis.