OBJECTIVE:To establish validity of the bicarbonate-urea (BU) method against direct measurements of gaseous exchange (GE) in a whole-body indirect calorimeter and to compare BU and doubly labelled water (DLW) measurements in free-living conditions in the same group of grossly obese women. DESIGN: Energy expenditure (EE) was estimated by the BU method over 24 h concurrently with whole-body indirect calorimetry and subsequently over 5 consecutive days at home concurrently with 14 day DLW. Six women, body mass index (BMI) 52.4710.4 kg/m 2 (s.d.), were studied. RESULTS: Total energy expenditure (TEE) measurements by BU and GE within the metabolic chamber were not significantly different (BU ¼ 11.7971.89 MJ/day and GE ¼ 11.6471.86 MJ/day; mean difference, 0.2570.49 MJ/day, P40.05). Free-living TEE derived from BU and DLW was also similar (13.2871.86 and 13.8672.25 MJ/day, respectively; mean difference 0.1771.33 MJ/day, Po0.05). The measured physical activity level (PAL) in these very obese subjects was within the range reported in other free-living studies in less obese individuals (1.6270.14 using DLW and 1.5670.20 using BU). The BU method was well tolerated by the subjects. CONCLUSIONS: This study in grossly obese subjects, heavier than those participating in previous studies involving tracer methods, demonstrates validity of the BU against GE under controlled metabolic conditions, and the equivalence between BU and DLW under free-living conditions. The results suggest that both tracer methods are valid in this population group. This study also demonstrates the practicalities of using the BU method over 5 days, the longest application of the method so far.