Fig. 3.

Associations between accelerometer-based physical activity (aPA) measures and incident coronary heart disease (CHD) risk in the Cox regression model for incident CHD on the aPA measures of total volume, total number of events, mean duration, and peak intensity, assuming linear relationships, and mean intensity via a cubic penalized spline function (non-linear function to account for significant non-linearity), without and with adjustment for the baseline covariates of age, sex, ethnicity, Townsend deprivation index, smoking status, alcohol intake frequency, and assessment center