ASSOCIATION BETWEEN PHYSICAL ACTIVITY LEVELS AND CARDIOMETABOLIC MULTIMORBIDITY IN CHINESE MIDDLE-AGED AND OLDER ADULTS

Abstract

Physical activity has been reported to be an effective method to reduce the risk of cardiometabolic multimorbidity (CMM).  In this prospective cohort study, we aimed to determine if there is a significant association between physical activity levels and the incidence of CMM among middle-aged and older adults in China in order to inform efforts to reduce CMM in the study population.  CMM was defined as the concurrent presence of at least two of the following three diseases: heart disease, stroke, and diabetes, based on self-reported physician diagnoses.  This study was based on data from the China Health and Retirement Longitudinal Study (CHARLS) conducted during 2015-2020.  Study subjects were randomly selected using multistage stratified probability proportional sampling from 28 provinces across China.  Inclusion criteria for study subjects were being aged ≥45 years at baseline (2015), having no history of stroke, diabetes or heart disease at baseline, completing all three waves of follow-up, and having complete data regarding physical activity and other study variables.  Exclusion criteria for subjects were having CMM at baseline, incomplete follow-up data and missing key variables.  Physical activity level was assessed using a standardized questionnaire asking about vigorous, moderate, and light activities during the previous week, expressed as metabolic equivalents of task in minutes (MET-minutes) per week.  Study subjects were then equally divided into 4 groups based on their baseline total weekly MET-minutes as follows: Quartile (Q)1 (reference group) (<1,732.5 MET-min/week), Q2 (1,732.5-5,287.5 MET-min/week), Q3 (5,287.5-11,848.5 MET-min/week) and Q4 (≥11,848.5 MET-min/week).  We used time-varying Cox proportional hazards regression analysis and restricted cubic splines to identify significant associations between physical activity levels and CMM incidence.  A total of 4,516 subjects were included in the study (1,129 subjects in each quartile): 52.8% (n = 2,384) females.  The mean (±standard deviation (SD)) age of study subjects was 57.3 (±8.9) years. The mean (±SD) body mass index of subjects was 23.8 (±3.7) kg/m².  During the follow-up period, 151 subjects (3.3%) developed CMM.  The mean (±SD) physical activity levels were 6,190.8 (±7,106.5) MET-min/week for the CMM group and 7,430.9 (±6,779.4) MET-min/week for the non-CMM group (p-value = 0.036).  After adjusting for age, gender, marital status, residence, education, BMI, smoking, drinking, and retirement status, subjects in Q2, Q3, and Q4 had significantly lower hazard ratios (HR) (Q2: HR = 0.537; 95% confidence interval (CI): 0.351-0.821, p-value = 0.004; Q3: HR = 0.490; 95% CI: 0.313-0.767, p-value = 0.002; Q4: HR = 0.479; 95% CI: 0.282-0.814, p-value = 0.007) of developing CMM than subjects in Q1.  Subgroup analyses revealed that the protective effects of physical activity were particularly pronounced among non-smokers (Q2: HR = 0.382; 95% CI: 0.230-0.634, p-value <0.001; Q3: HR = 0.475; 95% CI: 0.293-0.772, p-value = 0.003; Q4: HR = 0.522; 95% CI: 0.297-0.918, p-value = 0.024) with a significant interaction (p-value for interaction = 0.006).  The protective effects of activity were also significantly greater among overweight and obese participants (Q2: HR = 0.568; 95% CI: 0.337-0.958, p-value = 0.034; Q3: HR = 0.319; 95% CI: 0.168-0.605, p-value <0.001; Q4: HR = 0.433; 95% CI: 0.215-0.869, p-value = 0.019).  Restricted cubic spline analysis revealed a significant overall association between total physical activity and CMM risk (p-value for overall = 0.007), although the nonlinear association was not statistically significant (p-value for nonlinear = 0.210).  In summary, subjects with higher levels of physical activity had significantly lower risk of developing CMM than subjects in the lowest activity quartile, and this protective effect was particularly evident among non-smokers and overweight and obese individuals. We conclude that higher physical activity levels are associated with reduced CMM risk among study subjects.  Further studies are needed to determine if interventions to increase physical activity are able to prevent CMM in this population and to elucidate the mechanisms underlying the differential effects across subgroups.