FIGURE 4.

Follow-up data: the effect of inflammation, CV diseases, smoking, and RA disease duration on the arterial aging. The arterial aging rate was calculated by dividing the difference of the two consecutive arterial age estimates with the time interval between the two measurements. The left panel shows least-square estimates of five risk factors on the rate: smoking, disease duration ≤10 years (disdur < 10), history of cardiovascular disease (cvevent), CRP 5 > 5 mg/L, and concomitant antihypertensive drug taking (bp_high). When the 95% confidence intervals do not cross zero the effects are significant, which is true only for crp5 (p = 0.024) and disdur < 10 (p = 0.05). In the absence of any risk factors (which includes that the disease started more 10 years ago), the aging rate is close to zero (base). The combined effect of two risk factors such as CRP and smoking shown in the figure is an estimated marginal effect. The two boxplots illustrate the effect of crp5 and disdur < 10.