International cohorts show differing cardiometabolic pathways to stroke

Researchers analyzed harmonized longitudinal data from 247,863 participants aged 45–100 in CHARLS, ELSA, HRS, KLoSA, MHAS, and SHARE, covering China, England, the USA, South Korea, Mexico, and Europe. The observational analysis modeled 17 transitions among six states: healthy, hypertension alone, diabetes alone, both conditions, stroke, and death. Cause-specific hazard models assessed associations with age, sex, education, and smoking, accounting for within-person correlation and including death as a competing state.

There were 599,161 observed transitions, including 40,911 to death. The baseline proportion classified as healthy ranged from 45.5% in Mexico to 74.5% in China. Among participants leaving the healthy state, hypertension alone was the most common destination across countries (42.2–66.9%), followed by death (9.0–33.7%). Each additional year of age was associated with higher stroke hazards from the healthy state (hazard ratio range, 1.032–1.084), hypertension alone (1.022–1.050), and combined hypertension and diabetes (1.013–1.032).

Country-by-age interactions were statistically significant for all major transitions (Wald p<0.05). Smoking was associated with progression to combined hypertension and diabetes, while education showed protective associations in several cohorts. The findings support attention to local cardiometabolic prevention needs, but do not establish causal effects. Differences in cohort design and measurement remain competing explanations for the cross-national variation, and the transition percentages are conditional on leaving a health state rather than absolute population risks.