OBJECTIVE: To measure activity related energy expenditure in Prader-Willi syndrome (PWS) corrected for body size. SUBJECTS: 17 PWS subjects (10 females, seven males, age 7.5-19.8 y) and 17 obese controls, matched for gender and bone age. MEASUREMENTS: Basal metabolic rate (BMR) was measured by ventilated hood and average daily metabolic rate (ADMR) was measured with doubly labelled water. Activity induced energy expenditure (AEE) was calculated as 0.9ADMR - BMR. Activity related energy expenditure was corrected for body size using the following measures: AEE per kg body weight (AEE/kg), ADMR/BMR (PAL), and the residual of the regression of ADMR on BMR (rADMR). Group differences were analyzed by analysis of covariance adjusting for bone age, fat mass (FM) and gender. RESULTS: ADMR, AEE and PAL were lower (P <0.01) in the PWS group compared with the control group (7.14 +/- 1.72, 1.07 +/- 0.69 and 1.33 +/- 0.15 MJ/day compared with 9.94 +/- 2.64, 2.56 +/- 1.03 and 1.55 +/- 0.12 MJ/day respectively). The variance of AEE/kg and PAL was significantly explained by gender and PWS, while AEE was additionally explained by FM. The variance of rADMR was explained by PWS and not by FM or gender CONCLUSION: Activity related energy expenditure is decreased in PWS compared with controls adjusted for bone age, FM and gender.