Abstract
Dietary patterns represent a broader picture of food consumption, and are better correlated with a variety of health outcomes. However, few studies have been conducted to explore the associations between dietary patterns and obesity in Southwest China. Data from the 2010-2012 National Nutrition Survey in the province of Yunnan, Southwest China, were analyzed (n = 1604, aged 18-80 years). Dietary data were collected using the 24 h dietary recall over three consecutive days. Height, weight, and waist circumference were measured following standard methods. Exploratory factor analysis was used to identify dietary patterns. Logistic regression was used to explore the association between dietary patterns and obesity. Three distinct dietary patterns were identified, which were labeled as traditional, modern, and tuber according to their key components. With potential confounders adjusted, adults in the highest quartile of the modern pattern were at higher risk of general and central obesity (odds ratio (OR) 1.95, 95% confidence interval (CI) 1.15-3.48; OR 2.01, 95% CI 1.37-2.93). In contrast, adults in the highest quartile of the tuber pattern were at lower risk of general and central obesity (OR 0.34, 95% CI 0.15-0.61; OR 0.64, 95% CI 0.43-0.95) but at higher risk of underweight (OR 2.57, 95% CI 1.20-6.45). No significant association was found between the traditional pattern and obesity. Moreover, dietary pattern differences occurred due to the differences in socio-demographic characteristics. In conclusion, the modern dietary pattern was positively, and the tuber pattern negatively, associated with general and central obesity among adults in Southwest China.