Cross-sectional examination of physical and social contexts of episodes of eating and drinking in a national sample of US adults

摘要

Abstract Objective The current study characterizes associations between physical and social contexts of self-reported primary episodes of eating/drinking and sociodemographic and obesity-related variables in US adults. Design Multinomial logistic regression was used to analyse a nationally representative sample of adults from the 2006a€“2008 American Time Use Survey. Models identifying physical (where) and social (whom) contexts of primary eating/drinking episodes at the population level, controlling for demographic characteristics, weight status and time of eating, were conducted. Setting USA. Subjects A nationally representative sample of US adults (n 21 315). Results Eating/drinking with immediate family was positively associated with age (OR = 1?·15 (95 % CI 1?·04, 1?·27) to 1?·23 (95 % CI 1?·09, 1?·39)), education level (OR = 1?·16 (95 % CI 1?·03, 1?·30) to 1?·36 (95 % CI 1?·21, 1?·54)), obesity (OR = 1?·13 (95 % CI 1?·04, 1?·22)), children in the household (OR = 3?·39 (95 % CI 3?·14, 3?·66)) and time of day (OR = 1?·70 (95 % CI 1?·39, 2?·07) to 5?·73 (95 % CI 4?·70, 6?·99)). Eating in the workplace was negatively associated with female gender (OR = 0?·65 (95 % CI 0?·60, 0?·70)) and children in the household (OR = 0?·90 (95 % CI 0?·83, 0?·98)), while positively associated with non-white status (OR = 1?·14 (95 % CI 1?·01, 1?·29) to 1?·47 (95 % CI 1?·32, 1?·65)) and time of day (OR = 0?·25 (95 % CI 0?·28, 0?·30) to 5?·65 (95 % CI 4?·66, 6?·85)). Women (OR = 0?·80 (95 % CI 0?·74, 0?·86)), those aged >34 years (OR = 0?·48 (95 % CI 0?·43, 0?·54) to 0?·83 (95 % CI 0?·74, 0?·93)) and respondents with children (OR = 0?·69 (95 % CI 0?·63, 0?·75)) were less likely to eat in a restaurant/bar/retail than at home. Overweight and obese respondents had a greater odds of reporting an episode of eating in social situations v. alone (e.g. immediate family and extended family; OR = 1?·13 (95 % CI 1?·04, 1?·22)) and episodes occurring in restaurant/bar/retail locations (OR = 1?·12 (95 % CI 1?·03, 1?·23) to 1?·14 (95 % CI 1?·05, 1?·24)). Conclusions Findings underscore the multidimensional nature of describing eating/drinking episodes. Social and physical contexts for eating/drinking and their demographic correlates suggest opportunities for tailoring interventions related to diet and may inform intervention targeting and scope.