Search

This secondary analysis was designed to evaluate the independent effect of physical activity (PA) on hemoglobin A1c (HbA1c) level in Chinese patients with type 2 diabetes mellitus (T2DM). A total of 799 T2DM patients from eight communities of Shanghai, China, were randomized into one control arm and three intervention arms receiving 1-year interventions of health literacy, exercise, or both. PA was measured using the International Physical Activity Questionnaire at baseline, 12 months, and 24 months and quantified as metabolic equivalents (Mets). A multiple level mixed regression model was applied to evaluate the associations between PA and HbA1c level. After adjusting for potential confounders including interaction of PA level with initial PA or HbA1c, a significant improved HbA1c was observed for the patients in the medium versus the lowest tertile groups of overall PA at 12 months (β: -3.47, 95%CI: -5.33, -1.60) and for those in the highest versus the lowest tertile group at 24 months (β: -0.50, 95%CI: -1.00, -0.01), resulting in a β (95%CI) of -3.49 (95%CI: -5.87, -1.11) during the whole two-year period of follow-up. The negative association was also observed when the subjects were classified according to their exercise levels using the World Health Organization (WHO) recommendation as a cut-off point. The beneficial effect of higher PA level was only observed among patients having a lower level of baseline HbA1c or PA or both (all p values for interaction <0.05). Our results provide evidence for the beneficial effect of PA and suggest that the exercise intervention should be addressed to the physically inactive patients to improve their PA level to a physiological threshold.