Abstract P233: Changes in Diet, Lifestyle Behaviors, and Body Weight Over Two Years in Mexican Women

Abstract

Background: Some prospective observational studies have evaluated baseline diet and lifestyle behaviors, rather than changes in diet and lifestyle in relation to weight gain. Evaluating changes in diet and lifestyle on changes in adiposity may more appropriately identify effective strategies aimed at preventing weight gain. Methods: In a prospective cohort of Mexican female teachers (n=11,296) aged 35-65 y, we evaluated how two-year changes in diet and lifestyle behaviors relate to body weight change. We excluded women who were pregnant or who had prevalent heart disease, diabetes, or cancer at baseline, and those with implausible energy intake or missing baseline dietary, lifestyle, or weight data. We collected information on habitual consumption of foods and beverages with a validated food frequency questionnaire. Weight was self-reported and previously validated. We obtained data on physical activity, smoking, alcohol use, and TV watching through a self-administered questionnaire. We calculated the change in weight and diet/lifestyle behaviors by subtracting the information obtained in 2008 from that in 2006. We used multivariable linear regression models adjusting for age, state of residence, menopausal status, hormone replacement therapy, and all diet and lifestyle behaviors, simultaneously. Results: Increasing weekly servings of individual food components was positively associated with two-year body weight for intake of potatoes (0.06 kg; 95%CI 0.006, 0.12), corn (0.22 kg; 95%CI 0.006, 0.43), corn tortilla (0.03 kg; 95%CI 0.02, 0.04), sweets and desserts (0.01 kg; 95%CI 0.001, 0.03), Mexican dishes (0.13 kg; 95%CI 0.08, 0.17), and sodas (0.07 kg; 95%CI 0.05, 0.10). We observed an inverse association for intake of fruits (-0.006 kg; 95%CI -0.011, -0.001), vegetables (-0.013 kg; 95%CI -0.019, -0.008), and whole grains (-0.025 kg; 95%CI -0.045, -0.005). Lifestyle behaviors were also independently associated with weight change, including smoking (ever smokers=0.22 kg, starters=-0.24 kg, quitters=0.18 kg; p=0.04), physical activity level (decrease in category=0.14 kg, increase in category=-0.04 kg p=0.03), and TV watching (0.04 kg per hour/d; p=0.04). Changes in consumption of nuts, refined grains, low-fat and whole-fat dairy products, red and processed meats, beans, rice, diet sodas, and alcohol use were not associated with two-year weight change. Conclusions: Two-year changes in specific dietary and lifestyle behaviors were associated with modest changes in body weight. These results may be useful for targeted weight management messages.