Development and Evaluation of Polar OwnCal Feature

The energy expenditure assessment in Polar heart rate monitors is called the OwnCal (in M-series heart rate monitors) or OwnCal_S (in S-series heart rate monitors) and it is based on prediction equations developed and evaluated in a series of research projects. Polar Electro Oy started this research co-operation with the UKK Institute, Tampere, Finland in 1995. Later the Department of Human Biology at the University of Maastricht, The Netherlands, joined the project. The first project was a comparative study about energy expenditure assessment methods (Fogelholm et al. 1998). In this study doubly labeled water measurements were done in obese women and compared with Caltrac accelerometer, Fitty3 pedometer, heart rate monitoring (Polar PE-3000), physical activity log and food diary. Heart rate, with individual HR-EE (energy expenditure) equations, gave an unbiased estimation of daily EE in obese females despite large individual errors in some cases. It was concluded that the critical points in using HR as a measure of total daily EE are (1) the choice of FLEX-point (HR discriminating sedentary and activity EE) and (2) the discrimination of activity vs. emotion related HR above the FLEX-point.

After the study described above, it was considered that the association between exercise energy expenditure and heart rate should be solved. A total of 86 women and men was measured on a bicycle ergometer and during walking in a graded exercise protocol to define the heart rate/energy expenditure -relation. As a result, generalizable gender specific regression equations including age, body weight and heart rate were developed for the energy expenditure assessment during exercise for adults (Hiilloskorpi et al. 1997,1999b). The reference measure for the EE in the laboratory was gas-analysis based on the Weir equation (1949). The prediction yielded r² of 0.70 in walking. The mean deviation between the predicted and measured EE during walking was 0.4 kcal/min and the limits of agreement (mean±2SD) were between -4 and 3 kcal/min. In a validation study the measured EE was underestimated by 2.7% in women and overestimated by 6.5% in men on an average during walking. In cycling the corresponding values were overestimations by 5.4% and 18%, respectively. Despite the rather high overestimation in men in cycling, the obtained EE estimation accuracy was considered satisfactory. The accuracy was more than any other commercially available assessment method for field settings (activity recorders, accelerometers, pedometers) could report, and Polar adopted this for the technical development of the OwnCal feature.

After this basic study the regression equations developed have been cross-validated in a sample of 135 adults at the Cooper Institute for Aerobics Research (unpublished). In this data the mean differences between the predicted and measured EE were 2% in women and 10% in men during incremental walking-jogging.

Further, the SmartEdge heart rate monitor prototype developed was used to study the accuracy of the energy expenditure equations on 50 adults in the field setting during walking-jogging and cycling protocols (Hiilloskorpi et al. 1998). The results showed that the mean difference between the energy expenditure predicted by Polar SmartEdge equations and the measured (Cosmed K4) energy expenditure was in females during cycling and walking -2.5 kcal/min and in males in cycling -1.2 kcal/min and in walking -1.1 kcal/min. In two thirds of the subjects the predicted energy expenditure differed from the measured value by less than 15% in women and in men by less than 14 % in cycling and 12% in walking. The difference between the gas analyzers (Medikro in the laboratory, Cosmed in the field measurements) confused the interpretation of these results and partly explained the underestimation in both genders in this study (Hiilloskorpi et al. 1999a). However, based on the results it was concluded that despite this underestimation, the EE equations built in SmartEdge provide an estimate of exercise EE that is satisfactorily accurate in healthy adults.

The OwnCal equations have been further validated in two studies. Altogether 40 healthy adults were measured on treadmill and on cycle ergometer at the Paavo Nurmi Centre, Turku, Finland (Kapanen et al. 2000). The equations were validated on heart rate levels 120-165 bpm. In men the mean difference between the measured and estimated EE was about 2 kcal/min in cycling and about 1 kcal/min in walking-running. In women the corresponding values were less than 0.5 kcal/min both in cycling and in walking-running. In all other cases except for in women during walking-jogging, the SmartEdge™ prediction values were slight overestimations.

In another study of over 100 adults at the Sports Science Institute of South-Africa, Cape Town (unpublished), the SmartEdge EE equations were shown to predict EE at lower exercise intensities (<10 kcal/min) with a reasonable accuracy (the mean deviation 1.6 kcal/min).

Because the OwnCal development was focused on less fit to moderately fit individuals, the next step was to update the EE estimation calculation to be more accurate in highly fit individuals/athletes. This resulted in the development of OwnCal_S. OwnCal_S estimates EE from exercise heart rate and individual HR_max and VO_2max (measured or predicted values). The development group consisted of 105 20-59 year old men and women (data collected in collaboration with the Cooper Institute, Dallas, Texas). The developed model was further validated on 101 men and women (data collected in collaboration with the UKK Institute, Tampere, Finland). All subjects went through a maximal treadmill test with continuous heart rate measurement and respiratory gas analysis. In the validation data, the error of estimate (mean+/-SD) for OwnCal_S was -0.7+/-1.3 and -0.3+/-0.8 kcal/min and standard error of estimate 1.4 and 0.9 kcal/min in men and women, respectively (Kinnunen et al. 2000). Thus, good accuracy for EE estimation was obtained in OwnCal_S.

In Physical Activity and Health: a Report of the Surgeon General (U.S. Department of Health and Human Services 1996) it is stated that activity leading to an increase in daily energy expenditure of approximately 150 kcal/day, equivalent to about 1000 kcal/week, is associated with substantial health benefits. OwnCal/OwnCal_S provides a useful alternative for assessing exercise dose in kilocalories.

References

Fogelholm M, Hiilloskorpi H, Laukkanen R, Oja P, Van Marken Lichtenbelt W, Westerterp K. Assessment of physical activity and energy expenditure in overweight women. Med Sci Sports Exerc 30(8), 1191-1197, 1998.

Hiilloskorpi H, Fogelholm M, Laukkanen R, Oja P, Natri A, Mänttäri A. Factors affecting the relation between heart rate and energy expenditure. International Congress of Movement and Sport in the Life-Cycle of Woman, Lahti, Finland. Book of Abstracts, nr 31, 1997.

Hiilloskorpi H, Fogelholm M, Laukkanen R, Pasanen M, Oja P. Validation of gender spesific equations for predicting energy expenditure during exercise. Med Sci Sports Exerc 30(5), 330, 1998.

Hiilloskorpi H, Mänttäri A, Fogelholm M, Pasanen M, Laukkanen R. The comparison between three different gas-analysers. Med Sci Sports Exerc 31(5), nr 1787, 1999a.

Hiilloskorpi H, Fogelholm M, Laukkanen R, Pasanen M, Oja P, Mänttäri A, Natri A. Factors affecting the relation between heart rate and energy expenditure during exercise. Int J Sports Med, 20, 438-443, 1999b.

Kapanen J, Laukkanen R, Hiilloskorpi H, Fogelholm M, Heinonen O. Estimation of EE during exercise by equation based on heart rate. Med Sci Sports Exerc 32(5), nr 984, 2000.

Kinnunen H, Nissilä S. Estimation of exercise energy expenditure from heart rate and aerobic capacity. Proceedings of 5th Annual Congress of the ECSS, Jyväskylä, Finland, 19-23 July 2000, p 395.

U.S. Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, 1996, p 147.

Weir J. New methods for calculating metabolic rate with special reference to protein metabolism. J Physiol 109, 1-9, 1949.