Physical activity and obesity: findings from cross-sectional and prospective studies

Obesity results from an excessive energy intake or a low level of physical activity or both. Cross-sectional studies indicate that obese individuals are less active than their lean counterparts. However, whether a reduced physical activity is the cause or the consequence of obesity is unclear. We measured total energy expenditure (TEE, doubly labeled water) over 7 days in free-living conditions and sleeping metabolic rate over a few hours (SMR, indirect calorimetry) in a respiratory chamber in 91 mostly sedentary, adult, non-diabetic Pima Indians (63M/28F, 36 plus or minus 12 y, 35 plus or minus 10% body fat; mean plus or minus SD). Energy expenditure due to physical activity (EEACT) was calculated as (TEE - (SMR + 0.1* TEE)). Physical activity level (PAL) was calculated as TEE/SMR. Fat-free mass (FFM, kg) and fat mass (FM, kg) were measured at baseline by hydrodensitometry or DEXA. Baseline and follow-up body weights were available in 58 (36M/22F) of the 91 subjects. Cross-sectionally, both EEACT and PAL were negatively correlated with percent body fat (r = -0.22; p = 0.04 and r = -0.34; p < 0.01, respectively). In a multiple regression model, EEACT (993 plus or minus 316 kcal/d) was best predicted by the following equation (564 + 14 FFM - 6.4 FM - 5.1 age, R super(2) = 0.22; p < 0.01). PAL (1.80 plus or minus 0.23) was best predicted by the following equation (2.16-0.01 FM-0.003 age; R super(2) = 0.14, p < 0.01). In a prospective analysis, baseline EEACT and PAL were not correlated with body weight changes (which ranged from -6 to +20 kg over a follow-up of 3.4 plus or minus 2.6 y). This was still true after accounting for differences in baseline body composition and follow-up duration. Our cross-sectional data in an adult population at high risk for obesity confirm that increased adiposity is associated with reduced physical activity. Our prospective data, however, do not lend support to the hypothesis that a low level of physical activity is a major risk factor for the development of obesity in Pima Indians.