Aims: Primary outcome of this observational study was to compare weight changes in two groups of overweight and obese individuals: subjects who had a diet prescribed on the base of resting energy expenditure (REE) measured by indirect calorimetry and subjects whose REE was estimated by a predictive equation. In addition, we analyzed differences in weight and metabolic parameter variation in subjects with and without an adequate to predicted REE. Methods: We retrospectively analyzed data of 355 overweight and obese patients: 215 on a diet based on REE measured by indirect calorimetry and 140 following a diet based on REE estimated by the Harris–Benedict equation. Anthropometric and metabolic parameters were evaluated for 18 months from baseline. Propensity score adjustment was used to adjust for known differences between the groups being compared. Results: A significant greater decrease in body weight was observed in the group that underwent indirect calorimetry compared to the group that did not undergo it (p < 0.001). No significant differences were observed between patients with not adequate to predicted REE compared to patients with adequate to predicted REE. Conclusions: A weight reduction program based on REE measurement appears more effective than a dietary program based on predictive formulas. This study suggests the routine use of indirect calorimetry in all weight reduction procedures. © 2017 Springer-Verlag Italia S.r.l.