ing was defined with a value < 1.14 for an individual subject in the ratio (energy intake)/(basal metabolic rate computed according to Schofield). Using this criteria, 55% (n = 229) of the subjects were classified as UR. The proportion of women was increased in the UR group compared to non-UR subjects (91 % vs 67%, P < 0.000,001) and analyses were performed separately in each gender. In women as in men, obese UR compared to non-UR were characterized by an increased protein intake (about 20% vs

more »

... 5% of total energy intake, P < 0.000,001) and an increased proportion of restraint subjects (P < 0.05), whereas age and BMI were comparable between the two groups. In conclusion, in obese patients seeking advice in a hospital setting, dietary underreporting: 1) concerns predominantly female subjects, 2) may be related to certain macronutrients, and 3) appears to be associated with dietary restriction.