Estimation of energy expenditure using heart rate and triaxial accelerometery in overweight and obese adults

Christopher C. Howe, Chris Easton

Research output: Contribution to journalMeeting Abstractpeer-review


To date, no study has examined the validity of the Actitrainer tri-axial accelerometer (AT) (Actigraph, Florida, USA) in combination with heart rate (HR) for prediction of energy expenditure (EE) during walking in overweight and obese adults.PURPOSE: To examine the validity of existing regression equations (Freedson and vector magnitude equations (VM)) based on accelerometer counts (AC) from the AT for predicting EE in overweight and obese adults during treadmill and outdoor walking. Further, to examine whether novel gender-specific equations combining AC, HR and body mass (BM) can enhance the accuracy of the EE prediction.METHODS: Twenty overweight or obese adults (eight males and twelve females, age 43 ± 11 years, BMI 30.5 ± 4.9 kg/m2) completed a 30 min incremental walking protocol on a treadmill in a laboratory setting (4-6.5 km·h-1). VO2 and VCO2 were measured throughout via breath-by-breath indirect calorimetry (IC) and EE for each stage was calculated using the Weir equation. AC and HR were continuously measured using the AT and Polar HR monitor. Novel gender-specific EE prediction equations were generated by multiple linear regression using AC, HR and BM. In a separate study, ten overweight and obese adults (five males and five females, age 44 ± 13 years, BMI 30.4 ± 4.0 kg/m2) conducted a self-paced three kilometre external walk during which EE, AC and HR were measured as before. EE was estimated using existing and novel prediction equations for both treadmill and external walks and compared to IC.RESULTS: The Freedson equation overestimated EE (7.9 ± 1.3 kcal.min-1 vs. IC: 5.5 ± 1.1 kcal.min-1, P<0.01) during treadmill walking as did the VM equation (7.1 ± 1.1 kcal·min-1, P<0.01). There was no difference between EE estimated using the novel gender specific prediction equations and EE measured by IC during either treadmill (P=0.993, standard error of estimate (SEE) 0.17 kcal.min-1) or outdoor walking (P=0.114, SEE 0.77 kcal.min-1).CONCLUSIONS: The Freedson and VM equation significantly overestimated EE in overweight and obese adults during walking exercise. However, novel gender specific prediction equations based on VM, HR and BM appear to provide valid estimations of EE in this population during both treadmill and self-paced external walking.
Original languageEnglish
Pages (from-to)S467
JournalMedicine & Science in Sports & Exercise
Issue number5S
Publication statusPublished - May 2012
Externally publishedYes


Dive into the research topics of 'Estimation of energy expenditure using heart rate and triaxial accelerometery in overweight and obese adults'. Together they form a unique fingerprint.

Cite this