Mentor(s): Kyle Timmerman, Ph.D.

BACKGROUND: Estimating resting metabolic rate (RMR) is important for assessing the dietary needs of hospitalized patients, and for developing caloric recommendations for weight loss/weight maintenance. The Mifflin-St. Jeor (MSJ) and Harris Benedict (HB) equations are commonly used to estimate RMR. The accuracy of these equations compared to measured RMR has been evaluated in previous studies. However, to date, the accuracy of these equations has not been well evaluated in older adults. PURPOSE: To compare the accuracy of the MSJ and HB equations relative to measured RMR in a population of older adults. METHODS: This study included 48 adults between the ages of 58 and 78 (Average age= 64). RMR (kilocalories per day, kcal/d) was measured using indirect calorimetry (ParvoMedics TrueOne 2400) following an 8-10 hour overnight fast. Each test was conducted in lasted approximately 30 minutes. The first 5 minutes of analysis were not used to give time for the subject to get acquainted with the testing method. RMR was additionally estimated using MSJ and HB prediction equations. Bivariate correlation and Bland-Altman plots were used to compare the MSJ and HB estimated RMR to measured RMR. RESULTS: The average measured RMR measured via indirect calorimetry was 1451 ± 251 kcal/day. The average estimated RMR for HB and MSJ were 1625 ± 252 kcal/d and 1607 ± 262 kcal/d. Correlational analysis revealed that both estimation equations were significantly correlated with measured RMR: MSJ (r= 0.39, p= 0.006) and HB (r= 0.445, p= 0.002). CONCLUSION: Both equations slightly over-estimated RMR, and in this population of older adults, HB appeared to have a greater degree of correlation with measured RMR than the Mifflin-St, Jeor equation.

Explore the Project