Relationships of physical activity with dietary behaviors among adults

Background. Physical activity and diet are important influences on health, but few data are available about the relationship between these two factors . The purpose of this study was to examine relationships between physical a ctivity and dietary quality and to identify determinants of the combination of sedentary behavior and suboptimal diet. Methods. The design of this study was cross-sectional. The setting was a la rge managed-care organization and the participants were 1,322 racially dive rse men and women ages 25-91 years. We categorized subjects' physical activ ity into vigorous, moderate, and sedentary based on answers to two validate d interviewer-administered questions about intensity and duration of specif ied activities. Dietary assessment was by means of a validated short food f requency questionnaire. We defined suboptimal diet as consuming unhealthful quantities of at least two of the following five food groups: fruits, vege tables, whole grain foods, whole-fat dairy foods, and red and processed mea ts. Results. Seven hundred fifty-four (57%) subjects were sedentary and 617 (47 %) consumed a suboptimal diet. Using multiple linear regression, we found t hat sedentary individuals consumed smaller amounts of foods and nutrients c onsidered to be healthful, such as fruits and vegetables, fiber, calcium, f olate, and vitamins A, C, and E, than more active participants. For nutrien ts considered to be harmful, such as saturated fat, trans fat, and dietary cholesterol, the association with physical activity was inverse. In multipl e logistic regression analyses, the strongest sociodemographic correlates o f the joint presence of inactivity and poor diet were less education [odds ratio for 1-year decrease 1.14 (95% confidence interval 1.06, 1.22)], nonwh ite race [1.48 (1.05, 2.07)], and nonmarried status [1.49 (1.06, 2.10)]. Conclusions. Physical activity and diet quality are correlated behaviors. S uboptimal diet and sedentary behavior tend to cluster in individuals who ar e less educated, not married, and of nonwhite race. Programs that target di et and activity together, informed by their joint determinants, may attain enhanced outcomes. (C) 2001 American Health Foundation and Academic Press.