Date of Award


Level of Access Assigned by Author

Campus-Only Thesis

Degree Name

Master of Science (MS)


Food Science and Human Nutrition


Susan Sullivan

Second Committee Member

Adrienne White

Third Committee Member

Shihfen Tu


The provision of too few or too many calories via nutrition support therapy to a critically ill patient can result in adverse health outcomes and hinder recovery. Indirect calorimetry (IC), which is a measurement of gas exchange, is the most accurate method of assessing calorie needs in hospitalized patients; however, its use is not always possible. When IC is not an option, clinicians may choose from a number of predictive equations, which utilize patient variables such as height, weight and age, to estimate energy needs. It is unclear which predictive equation is the most accurate for use in most critically ill patients. The primary objective of this study was to evaluate the performance of several predictive equations compared to measured resting energy expenditure as determined by indirect calorimetry. This study involved a retrospective evaluation of medical records from Maine Medical Center between 2008 and 2011. Data was collected for patients who had their calorie needs measured via indirect calorimetry during this time period. For each of these patients, data was collected in order to estimate calorie needs using eleven predictive equations. The best performing equation was the Penn State 2010 equation; however, this equation was designed for older, obese patients and the sample size for this subgroup was small. Of equations applicable to all patients, the Penn State 2003b equation was the most accurate. The least accurate equation was the ACCP formula, which utilizes only patient weight data. The metabolic rate of critically ill patients in this study was highly variable. More research is needed to identify variables that can better predict energy needs in order to generate more accurate predictive equations for clinical use. Since even the most accurate predictive equation cannot accurately predict energy needs in all patients, indirect calorimetry should be used when possible.