Obesity Prevalence In Physically Active And Inactive Adults By Sex, Race, Education, And Age

Abstract

More than 65% of US adults are overweight or obese. The relations between physical activity (PA) and weight loss and weight maintenance have been documented, but little information is available about population estimates of obesity prevalence by level of PA. PURPOSE Describe the prevalence of obesity by PA level for a representative sample of US adults, and examine differences by sex, race, education, and age. METHODS 9,449 adults (4,216 men 5,233 women) responded to the National PA and Weight Loss Survey, a random digit dial survey of US households conducted Sept to Dec 2002. As part of a larger survey, respondents reported their age, race/ethnicity, education, height, and weight. Body mass index was calculated as wt (kg)/ht (m)2 and categorized as healthy weight (< 25.0), overweight (25.0–29.9), or obese (≥ 30.0). Respondents reported days/wk and min/day of non-occupational moderate and vigorous PA. PA was categorized as meeting public health recommendations (≥ 5 days ≥ 30 min moderate intensity PA or ≥ 3 days ≥ 20 min vigorous intensity PA), insufficiently active (not enough PA to meet recommendations), or inactive (no reported PA). Prevalence estimates were calculated for BMI category by PA level. Prevalences of obesity by PA level were stratified by sex, race/ethnicity, education, and age. RESULTS The age-adjusted prevalence of healthy weight was directly associated with PA level (46% meet PA recommendations, 41% insufficiently active, 38% inactive). The prevalence of obesity was inversely associated with PA level (17% meet PA recommendations, 23% insufficiently active, 27% inactive). Among adults who met PA recommendations, there was little difference in obesity prevalence between adults engaging in > 2,010 MET-min/wk of PA (16%, 95% CI = 15,18) and adults engaging in ≤2,010 MET-min/wk of PA (17%, 95% CI = 16,19). Inactive women had higher obesity prevalence than inactive men. Non-Hispanic blacks and Hispanics had higher obesity prevalence than non-Hispanic whites for all PA levels. Persons with at least some college had lower obesity prevalence than those with high school education or less for all PA levels. Obesity prevalence was higher with older age up to the highest age group (65+ years) for all PA levels. Obesity prevalence was higher with lower PA for all demographic groups. CONCLUSION Obesity prevalence is inversely associated with PA level, and follows a dose response pattern that plateaus at high levels of PA. Demographic differences in obesity prevalence were observed, but the relationship between obesity and PA was consistent within sex, race/ethnicity, education, and age groups. Supported by CDC Cooperative Agreement, SIP 20–01, U48/CCU409664.