This prospective cohort study consisted of 544,749 individuals who participated in a self-funded comprehensive health surveillance program offered by Taiwan MJ Health Management Institution between 1998 and 2016. We included 236,216 adults who had at least two repeated MetS measures 5.9 (4.6) years apart and were followed up for mortality over 18.8 (5.2) years. Participants were classified according to the change in their MetS status as follows: MetS-free at both time points (n=173,116), MetS-developed (n=22,607), MetS-recovered (n=13,616), and MetS-persistent (n=26,877). Multivariable Cox proportional hazards model was used to determine the association between change in MetS status and risk of all-cause and CVD mortality.

Over the 4,436,842 person-years follow-up period, 14,226 participants died, including 2,671 (19%) of CVD-related causes. The crude CVD mortality rate per 1,000 person-years in the study groups were MetS-free, 0.32; MetS-developed, 0.75; MetS-recovered, 1.22; and MetS-persistent, 2.00 (P < 0.001). Compared to the persistent MetS group, participants in the MetS-recovered group had a lower risk of all-cause (adjusted hazard ratio [aHR], 0.87; 95%CI, 0.82-0.92) and CVD mortality (aHR, 0.81; 95% confidence interval [CI], 0.71-0.93). Development of MetS increased the risk for all-cause (aHR, 1.11; 95%CI, 1.05-1.17) and CVD mortality (aHR, 1.22; 95%CI, 1.07-1.39), compared to the MetS-free group.