Associations of Alcoholic Beverage Consumption with Dietary Intake, Waist Circumference, and Body Mass Index in US Adults: National Health and Nutrition Examination Survey 2003-2012

Background Findings from studies of alcohol and obesity measures (eg, waist circumference [WC] and body mass index [BMI; calculated as kg/m(2)]) are conflicting. Residual confounding by dietary intake, inconsistent definitions of alcohol consumption across studies, and the inclusion of former drinkers in the nondrinking comparison group can contribute to the mixed literature. Objective This study examines associations of alcoholic beverage consumption with dietary intake, WC, and BMI. Design Cross-sectional data from the 2003-2012 National Health and Nutrition Examination Survey were analyzed. Participants/setting Adults 20 to 79 years of age (n = 7,436 men; n = 6,939 women) were studied. Main outcome measures Associations of alcoholic beverage consumption with energy (kcal), macronutrient and sugar intakes (% kcal), WC, and BMI were determined. Statistical analyses performed Multivariable linear regression models were used to determine associations of average daily volume and drinking quantity (ie, drinks per drinking day) with dietary intake and obesity measures. Former and never drinkers were analyzed as distinct categories; associations of drinking with WC and BMI were examined with and without adjustment for dietary intake variables. Results Heavier-drinking men (>= 3 drinks/day) and women (>= 2 drinks/day) consumed less nonalcoholic energy (beta - 252 kcal/day, 95% CI -346 to -159 kcal/day and beta - 159 kcal/day, 95% CI -245 to -73 kcal/day, respectively) than moderate drinkers (1 to 2 drinks/day in men and 1 drink/day in women). By average daily drinking volume, differences in WC and BMI between former and moderate drinkers were +1.78 cm (95% CI 0.51 to 3.05 cm) and +0.65 (95% CI 0.12 to 1.18) in men and +4.67 cm (95% CI 2.95 to 6.39 cm) and +2.49 (95% CI 1.64 to 3.34) in women. Compared with moderate drinking, heavier drinking volume was not associated with WC or BMI among men or women. In men, drinking >= 5 drinks/drinking day was associated with higher WC (beta 3.48 cm, 95% CI 1.97 to 5.00 cm) and BMI (beta 1.39, 95% CI 0.79 to 2.00) compared with men who consumed 1 to 2 drinks/drinking day. In women, WC and BMI were not significantly different for women drinking >= 4 drinks/drinking day compared with 1 drink/drinking day. Conclusions Differences in dietary intake across drinking subgroups and separation of former drinkers from nondrinkers should be considered in studies of alcohol intake in relation to WC and BMI.