Association of baseline level of physical activity and its temporal changes with incident hypertension and diabetes mellitus.

Background The association between baseline and temporal changes in physical activity and incident hypertension or diabetes mellitus in initially non-hypertensive or non-diabetic subjects is rarely known. Methods Among individuals who underwent consecutive comprehensive health screenings, their physical activity level was measured using a self-reported international physical activity questionnaire. First, subjects were classified into four categories: no regular physical activity with a sedentary lifestyle; minimal physical activity (<75 min/week); insufficient physical activity (≥75 min but <150 min/week); and sufficient physical activity (≥150 min/week). Second, subjects were sub-grouped, based on temporal changes in physical activity level between baseline and consecutive follow-up: increase, no change, and decrease. Results Finally, among 174,314 subjects (mean age 36.7 ± 6.9 years), 5544 (3.18%) and 21,276 (12.2%) developed incident diabetes mellitus and arterial hypertension, respectively. After a multivariate adjustment, sufficient baseline physical activity was associated with significantly lower risk for incident hypertension (hazard ratio 0.89; 95% confidence interval (CI) 0.81 to 0.97), but the difference was not significant, and showed a lower trend in diabetes mellitus incidence (hazard ratio 0.87; 95% CI 0.69 to 1.04) in reference to no regular physical activity group. Regardless of the baseline physical activity level, subjects with a temporal increase in physical activity showed significantly decreased risk for incident hypertension (hazard ratio 0.93; 95% CI 0.87 to 0.99) and diabetes mellitus (hazard ratio 0.83; 95% CI 0.74 to 0.92) compared with those with a temporal decrease in their physical activity level. Conclusion Both sufficient baseline physical activity level and its temporal increase were associated with a lower risk of incident hypertension and diabetes mellitus in a large, relatively healthy, cohort.