Factors associated with the accuracy of reported energy expenditure

Physical inactivity has been associated with an increased risk for several chronic diseases such as colon cancer and cardiovascular disease. Accurate measurements of reported energy expenditure (REE) is vital to understanding the relationship between physical activity and disease. The current study was conducted to determine what factors are associated with the accuracy of self reported physical activity and hence total energy expenditure (TEE), using doubly labeled water as the reference method for TEE. We studied physiologic and psychological factors that we believe would be related to reporting accuracy (RA=REE/TEEx100). A total of 35 women (43.57±9.52 years) with an average body mass index (BMI) of 28.48±8,03 kg/m² participated in the study. The women completed a physical activity questionnaire (PAQ), a series of behavioral scores and participated in a doubly labeled water for a 9 day period to measure TEE. The psychological factors included: Marlow-Crowne Social Desirability Scale (MC_SDS), Weinberger Adjustment Inventory. The Eating Disorder Inventory (EDI), Revised Restrained Eating Scale and their subscales. A series of Stunkard's Silhouettes were used to have subjects determine if their body size was larger (negative) or smaller (positive) than a perceived healthy body size (healthy). The women's REE from the PAQ was 7.56±1.75 MJ. Their mean TEE was 9.98±2.46 MJ. The mean RA was 77.08±13.43%. TEE and RA were negatively correlated (r=-0.4028, p=0.0164). BMI had the strongest association of the physiological measurements with RA and was entered into all regression models using psychological scores. The best regression model for predicting RA was: RA=93.62+0.52(MC_SDS)-0.99(BMI)-4.58(healthy)-0.58(maturity fears from the EDI). Our results indicate that: as a woman's fear of growing old increases RA decreases, as BMI increases RA also decreases, as MC_SDS scores increase RA increases and women who perceive themselves as being less healthy have a greater RA. In conclusion, research should consider the psychological profiles of the participants and their body composition when obtaining REE.