Physical activity and overweight/obesity among Malaysian adults: findings from the 2015 National Health and morbidity survey (NHMS)

Background: Overweight and obesity are growing health problems both worldwide and in Malaysia due to such lifestyle changes as decreased physical activity (PA), increased sedentary behavior and unhealthy eating habits. This study examined the levels and patterns of PA among normal-weight and overweight/obese adults and to investigate the association between PA level and overweight/obesity in Malaysian adults. Methods: This study used data from the 2015 National Health and Morbidity Survey (NHMS), a nationwide cross-sectional survey that implemented a two-stage stratified random sampling design. Respondents aged 18 years and above (n = 17,261) were included in the analysis. The short version of International Physical Activity Questionnaire (IPAQ) was administered to assess the respondents' PA levels. The respondents' height and weight were objectively measured and body mass index (BMI) was calculated. The respondents were categorized according to BMI as either normal-weight (18.5-24.9 kg/m(2)) or overweight/obese (>= 25 kg/m(2)). Descriptive and complex sample logistic regression analyses were employed as appropriate. Results: Overall, approximately 1 in 2 respondents (51.2%) were overweight/obese, even though the majority (69.0%) reporting at least a moderate level of PA (total PA >= 10 MET-hours/week). In both normal-weight and overweight/obese groups, a significantly higher prevalence of high PA (total PA >= 50 MET-hours/week) was observed among men than women (p < 0.001), but women reported a significantly higher prevalence of low and moderate PA than men (p < 0.001). Men reported significantly higher activity levels (in MET-hours/week) than women with regard to walking, vigorous-intensity PA and total PA (p < 0.001). Overweight/obese men reported a significantly lower level of vigorous-intensity PA and total PA than normal-weight men (p < 0.001). A low level of PA was associated with the risk of overweight/obesity (Adjusted OR = 1.14; 95% CI: 1.01-1.30) compared to a high level of PA among men but not among women. Conclusions: The levels of PA were inversely related to the risk of overweight/obesity in men but not in women. Programs designed to reduce overweight/obesity rates should encourage the practice of moderateto vigorous-intensity PA. Future research should consider using longitudinal and prospective approaches that simultaneously measure dietary intake, PA and BMI among Malaysian adults to investigate the actual relationship between PA and overweight/obesity.