Dietary patterns as predictors of body fat and BMI in women: a factor analytic study

To identify independent patterns of diet using factor analysis to determine the extent to which dietary patterns account for differences in body fat percentage (BF%) and body mass index (BMI). Also, to ascertain the extent to which the associations are influenced by age, education, menopause, energy intake, and physical activity.

Study design was cross-sectional.

Study setting was approximately 20 cities in the Mountain West.

The study included 281 apparently healthy female nonsmokers.

Diet was assessed using 7-day weighed food records, and foods were categorized using the American Diabetes and American Dietetic Associations Exchange Lists and expressed as servings per 1000 kcal. BF% was measured using the Bod Pod, and physical activity was estimated using accelerometers worn for 1 week.

We used factor analysis, general linear models, and partial correlations.

Three dietary patterns were identified: (1) Prudent Pattern, (2) Low-fat Milk, and (3) Meat. Higher consumption of the Prudent Pattern corresponded with significantly lower BF% (F = 8.5, p = .0038) and BMI (F = 4.4, p = .0363). The Low-fat Milk pattern was inversely related to BF% (F = 5.4, p = .0207) and BMI (F = 9.5, p = .0023). Higher intake of the Meat pattern was related to higher levels of BF% (F = 4.5, p = .0346) and BMI (F = 4.2, p = .0418).

These findings support an association between dietary patterns and body composition. Dietary patterns reflect the complex interrelationships inherent in day-to-day eating and are strongly related to differences in BF% and BMI in women.