A two-page mailed questionnaire, sent to subscribers of a walking magazine and to participants of walking events, solicited information on demographics (age, race, education), walking history (age when began walking at least 12 miles/week, current average weekly mileage), weight history (greatest and current weight; weight when started walking; least weight as a walker; body circumferences of the chest, waist, and hips; bra cup size), diet (vegetarianism and the current weekly intakes of alcohol, red meat, fish, fruit, vitamin C, vitamin E, and aspirin), current and past cigarette use, history of heart attacks and cancer, and medications for blood pressure, thyroid, cholesterol, or diabetes (20,21,22). Walking distances were reported in miles per week, body circumferences in inches, and body weights in pounds. These values were converted to kilometers per day, centimeters, and kilograms, respectively. The analyses were restricted to nonsmoking, nondiabetic subjects who provided complete data on height, weight, education, and intakes of meat, fruit and alcohol.
Intakes of meat and fruit were based on the questions “During an average week, how many servings of beef, lamb, or pork do you eat”, and “…pieces of fruit do you eat”. Alcohol intake was estimated from the corresponding questions for 4-oz (112 ml) glasses of wine, 12-oz (336 ml) bottles of beer, and mixed drinks and liqueurs. Alcohol was computed as 10.8 g per 4-oz glass of wine, 13.2 g per 12-oz bottle of beer, and 15.1 g per mixed drink. Correlations between these responses and values obtained from 4-day diet records in 110 men were r = 0.46 and r = 0.38 for consumptions of meat and fruit, respectively. These values agree favorably with published correlations between food records and more extensive food frequency questionnaires for red meat (r = 0.50), and somewhat less favorably for fruit intake (r = 0.50) (23).
The walkers' BMIs were calculated as the weight in kilograms divided by height in meters squared. Self-reported body circumferences of the waist, hip, and, chest were in response to the question “Please provide, to the best of your ability, your body circumferences in inches” without further instruction. Self-reported height and weight from the questionnaire have been found previously to correlate strongly with their clinic measurements (r = 0.96 for both) (24). Self-reported waist and hip circumferences are somewhat less precise as indicated by their correlations with self-reported circumferences on a second questionnaire (r = 0.84 and r = 0.79, respectively) and with their clinic measurements (r = 0.68 and r = 0.63, respectively) (24). Self-reported chest circumferences also demonstrate strong test—retest correlations across repeated questionnaires (r = 0.93) and somewhat weaker correlation relative to their clinic measurement (r = 0.77) (24). The study protocol was reviewed by the University of California Berkeley committee for the protection of human subjects, and all subjects provided a signed a statement of informed consent.
Results are presented as mean ± s.e. or slopes ± s.e. except where noted. With the exception of the sample description of Table 1, all analyses are adjusted for age (age and age2), education, and alcohol intake, and all analyses are sex-specific. Multiple linear regression analyses were used to estimate the relationship between the walkers' BMI and body circumferences (dependent variables) and their reported meat or fruit intake (independent variables) when adjusted for age, education, and alcohol consumption (covariates). The sample was divided into walking increments of <1.5 km/day, 1.5–3, 3–4.5, and >4.5 km/day and regression coefficients for meat or fruit were calculated separately within each stratum. To test whether the regression slope differed significantly by distance walked, we combined the data over all distance categories tested whether the coefficient for the interaction between exercise and diet (i.e., “fruit × distance walked” or “meat × distance walked”) differed significantly from zero in a model that also included the separate main effects of exercise (distance walked) and diet (fruit or meat). Thus the formal test for a significant exercise by diet interaction treated walking distance as a continuous variable, whereas the stratified analyses for illustrating the differences in slope treated walking distance as a categorical variable. We also created a composite variable that combined meat and fruit intake into a single dietary index. Specifically, multiple linear regression analyses was used to estimate the best linear combination of meat and fruit intake for predicting BMI or body circumferences within the least-active distance category (<1.5 km/day), i.e., where the walkers' BMI or body circumferences were the dependent variables and their reported meat and fruit intake were the independent variables when adjusted for age, education, and alcohol consumption. Separate dietary indexes were computed for males and females and for BMI and each circumference variable. The same dietary index was applied to all walkers (i.e., irrespective of their reported distance walked) and used to estimate the regression slope for BMI vs. the dietary index within each distance category, and to test for an exercise by diet interaction as described above for meat and fruit. Virtually identical significance levels were obtained for distance by dietary index interactions when the coefficients for the index were calculated on all the data rather than just the least-active walkers. All analyses were performed separately for males and females.
Table 1. Characteristics of subjects by average distance walked per day