Resistance exercise is a popular form of exercise for individuals whose goal is to increase daily caloric expenditure and alter body composition (11). The National Strength and Conditioning Association and the American College of Sports Medicine (ACSM) recommend that resistance exercise be performed regularly by healthy individuals as a part of a balanced fitness program (9,29). The ACSM recommends progressive resistance training 2–3 days per week by healthy individuals, which should include at least 1 set of 8–10 exercises that focus on the major muscle groups, with multiple sets providing even greater benefits (29). The benefits of resistance training can include an increase in muscular strength, power, and endurance; increased high-density lipoprotein cholesterol; decreased low-density lipoprotein and total cholesterol; decreased blood pressure; increased glucose tolerance and insulin sensitivity; and increased bone mass and connective tissue strength (reviewed in (9,19)). An added benefit of resistance training is that it contributes to total daily caloric expenditure (18,24); therefore, it may be a useful modality for losing weight and improving body composition. Previous studies examining the energy cost of resistance training have shown that an acute bout of resistance exercise can result in an energy expenditure of ∼70–289 kcal, depending on the type of resistance exercise protocol (e.g., single set vs. multiple set, upper vs. lower body) and the individual (gender, fat-free mass, etc.) (2,3,13,21,25–27,31). Increases in caloric expenditure as a result of resistance training can also be attributed to excess postexercise oxygen consumption (33), increases in fat-free mass (29), and increases in resting metabolic rate (24).
An accurate measure of energy expenditure during resistance exercise is useful when designing resistance training programs that are focused on altering body composition. Although it is possible to accurately measure energy expenditure during resistance exercise, many of the methods of energy assessment are difficult to use because of the unique nature of resistance training. For example, direct calorimetry may not be suitable because of limitations in space for resistance exercise equipment in a metabolic chamber. Portable indirect calorimetry can be used to accurately calculate energy expenditure, but this method of assessment is expensive and not practical for most resistance training programs. Recently, it has been suggested that accelerometers can be used to accurately predict energy expenditure during resistance exercise (31). This method of assessing energy expenditure is convenient when measuring the energy cost of resistance exercise because accelerometers are relatively inexpensive and portable, and their small size does not interfere with movements undertaken during resistance training.
Recently, Rawson and Walsh (31) developed a prediction equation to estimate resistance exercise energy expenditure based on uniaxial (vertical) waist accelerometer counts, fat-free mass, and sex (R2 = 0.90). It is possible that the use of a triaxial accelerometer would be able to predict energy expenditure with greater accuracy because resistance exercise takes place in multiple planes (frontal, sagittal, transverse). For example, during the biceps curl exercise, movement occurs in several planes, not solely the sagittal (vertical) plane. In this situation, a uniaxial accelerometer would only be able to measure acceleration along the sagittal plane, whereas a triaxial accelerometer would be able to measure acceleration in all 3 planes, thus providing a more accurate depiction of acceleration during this exercise. In the study performed by Rawson and Walsh (31), the uniaxial accelerometer counts were measured at the wrist and ankle and also at the waist, and although the wrist and ankle counts correlated to energy expenditure (r = 0.31, p = 0.10; r = 0.50, p < 0.01, respectively), these did not improve the ability to predict energy expenditure. Because upper-body and lower-body movements take place in several planes during resistance exercise, the use of triaxial accelerometers at the wrist and ankle may improve the correlation between accelerometer counts and energy expenditure as the counts will be measured in all 3 planes.
Accelerometers are portable and easy to use, and are a very practical tool for estimating energy expenditure during exercise. If appropriate regression equations can be created, it may be possible to use accelerometry to estimate energy expenditure during resistance exercise. This would be useful to those resistance training individuals who need to closely monitor energy balance (e.g., weight-class athletes). Therefore, the purpose of this study was to estimate resistance exercise energy expenditure using triaxial accelerometry. The primary hypothesis of this study was that triaxial accelerometer counts would be correlated with energy expenditure during resistance exercise, thereby demonstrating that triaxial accelerometer counts can be used to predict resistance exercise energy expenditure. Secondary hypotheses included the following: (a) total waist accelerometer counts would have a higher correlation with net energy expenditure compared with wrist and ankle counts, (b) wrist accelerometer counts would be greater than ankle and waist counts, and (c) exercise heart rate plus waist accelerometer counts would improve the estimate of resistance exercise energy expenditure.
Experimental Approach to the Problem
To determine whether triaxial accelerometers could be used to estimate resistance exercise energy expenditure, 30 healthy men and women performed a resistance exercise protocol while wearing triaxial accelerometers and a portable metabolic system. A regression analysis using the accelerometer data was performed to develop an equation to predict energy expenditure during the resistance exercise protocol. All subjects reported to the laboratory in the morning after an overnight fast on 2 separate occasions. The first visit determined body composition characteristics and strength (10RM), and the second assessed energy expenditure during the resistance exercise protocol while wearing triaxial accelerometers. During visit 1, age, height, weight, skinfold thickness, circumference measurements, limb lengths, and 10RM strength were assessed. During visit 2 (3–7 days after visit 1), energy expenditure during a resistance exercise protocol was estimated using a Cosmed K4b2 (Cosmed, Rome, Italy) portable metabolic system and a portable blood lactate analyzer (Lactate Plus; Nova Biomedical Corporation, Waltham, MA, USA). In addition, triaxial accelerometers were worn at the wrist, waist, and ankle during the resistance exercise protocol. Subjects were instructed to maintain their regular diet the day before testing and were allowed to drink water ad libitum until the time of testing. Additionally, subjects were asked to refrain from caffeine ingestion for 12 hours before and alcohol ingestion for 24 hours before both testing sessions.
Thirty healthy, college-aged subjects (15 men, 15 women) who engaged in recreational resistance training were recruited for this study. All subjects were currently engaged in resistance training during the time of the study and had been resistance trained for a minimum of 6 months before enrolling in the study. Although subjects were not highly trained, the majority were already familiar with all the exercises performed in the study and were familiarized with the equipment during the first visit. Subject characteristics are reported in Tables 1–3 (mean ± SD). Before participation, all subjects were informed of the risks of the study, and read and signed an institutionally approved informed consent form. In addition, a Physical Activity Readiness Questionnaire was completed by all subjects to determine eligibility for the study. Subjects were excluded from the study if they had known cardiovascular or musculoskeletal problems. This study was approved by the Institutional Review Board of Bloomsburg University for the use of human subjects.
Anthropometry and Body Composition
During visit 1, height was measured using a wall-mounted stadiometer and body mass assessed using a calibrated balance scale. Skinfold thickness was assessed with Lange skinfold calipers at 3 sites (men: chest, abdomen, and thigh; women: triceps, suprailiac, and thigh). Skinfolds for each site were measured 3 times, with the average of the 3 trials used to determine body density (14,15). Body density was then converted to body composition using the Siri equation (12). Upper-body and lower-body limb lengths were assessed with a spring-loaded Gulick II anthropometric tape measure (Country Technologies, Inc., Gays Mills, WI, USA) according to the procedures described by Lohman et al. (22). During the limb length measurements, subjects stood in the anatomical position, and arm length was measured from the acromion process to the dactylion and leg length measured from the greater trochanter to the base of the calcaneus. Circumference measurements at the neck, shoulder, chest, waist, abdomen, hip, thigh, calf, ankle, upper arm, wrist, and forearm were taken using a Gulick II anthropometric tape measure as described by Heyward and Wagner (12).
The 10RM strength test was conducted using standardized procedures as recommended by the National Strength and Conditioning Association (1). Ten–repetition maximum strength tests were performed for the Smith machine bench press, Smith machine shoulder press, Smith machine squat, leg extension, leg curl, lat pull-down, triceps push-down, and barbell biceps curl. Subjects were verbally instructed on the proper technique for each exercise and then completed 10 repetitions with a very light weight. Three to 5 additional sets of 10 repetitions were performed with progressive resistance (∼5% increase in each set for upper-body and 10% for lower-body exercises), with 2 minutes rest in between each set. The 10RM for each exercise was determined as the heaviest load that the subject could lift for 10 repetitions.
Resistance Exercise Protocol
The resistance exercise protocol was based on the ACSM recommendations for resistance exercise to achieve good health (29). The protocol was designed to work all major muscle groups using a moderate load (10RM) with relatively short rest intervals because this combination of moderate intensity and short rest intervals has been shown to elicit maximal energy expenditure (16,30). Subjects began by pedaling on a cycle ergometer for 5 minutes at 60% of their age-predicted maximum heart rate. The subjects then performed a resistance exercise protocol that consisted of 8 exercises: Smith machine bench press, Smith machine shoulder press, Smith machine squat, leg extension, leg curl, lat pull-down, triceps push-down, and barbell biceps curl. Subjects performed 2 sets of 8–10 repetitions for each exercise at a 10RM intensity because this type of protocol is beneficial for increasing muscular strength/endurance, lean body mass, and bone density, and also for improving many health-related factors such as blood pressure, lipid profiles, and insulin sensitivity (9,19). There was a seated 1-minute rest in between sets and a seated 2-minute rest in between exercises. During the rest intervals, subjects were instructed to remain as still as possible to minimize accelerometer counts. Cadence for each exercise was controlled with a metronome set at 60 b·min−1, with the concentric phase lasting 1 second and the eccentric phase lasting 1 second. Total load lifted during the resistance exercise session was calculated using the following formula: lifting volume = ([load × reps]set1 + [load × reps]set2). This formula was used for each exercise, and the lifting volume for all exercises was summed to calculate the total lifting volume for the entire session.
Energy expenditure during the resistance exercise protocol was assessed using a Cosmed K4b2 portable indirect calorimeter. The Cosmed K4b2 calorimeter was calibrated before each use in accordance with the manufacturer's guidelines. Before the resistance exercise protocol, the unit was attached to the subject's chest with a harness, and a Hans Rudolph face mask (Hans Rudolph, Kansas City, MO, USA) covered the nose and mouth via the use of a head strap. The subject was then seated in a chair for 20 minutes to determine resting metabolic rate. At the end of the 20-minute rest period, and 2 minutes after completion of the resistance exercise protocol, a blood lactate measurement was taken from the subject via a 25-μL fingerstick. Blood lactate was converted to oxygen equivalent values, with an increase of 1 mmol·L−1 of lactate representing 3 ml O2 per kg body weight (10). The oxygen equivalent was then converted to kilocalories as 5.0 kcal per liter of O2 (34).
Before the resistance exercise protocol, 4 triaxial accelerometers (3 ActiGraph GT3X and 1 ActiTrainer) were secured to the subject's right wrist (1 GT3X between the ulnar and radial styloid processes), right side of the waist (1 GT3X on the hip), left side of the waist (1 ActiTrainer on the hip), and right ankle (1 GT3X superior to the lateral malleolus). The ActiGraph accelerometers used in this study are depicted in Figure 1. In a typical accelerometer, a seismic mass within the accelerometer deforms a piezoelectric material during acceleration, thereby generating an electric potential, which is converted into acceleration counts. These accelerometers are able to detect acceleration in 3 planes of movement, distinguish between periods of exercise and rest, and recognize the postural orientation of the subject. The acceleration measured by the triaxial accelerometers were recorded as activity counts, and these counts were used to estimate movement intensity and energy expenditure (38). In addition, the ActiTrainer accelerometer was synchronized with a heart rate monitor (Polar Electro, Kempele, Finland) to record heart rate data throughout the entire workout. All accelerometer and heart rate data were recorded in 1-second epochs, and data from the accelerometers were downloaded to a computer via the ActiLife computer software program.
Independent t-tests were used to assess differences in subject characteristics between sexes. A 1-way analysis of variance (ANOVA) was used to compare wrist, waist (ActiGraph GT3X and ActiTrainer), and ankle accelerometer counts in all 3 axes, and Tukey's post hoc analysis was used to examine significant interaction effects. Pearson product moment correlation coefficients were used to examine the relationship between accelerometer counts and measured energy expenditure during the resistance exercise protocol. To test our hypothesis, that triaxial accelerometers can be used to estimate resistance exercise energy expenditure, regression analysis was used to develop an equation to estimate resistance exercise energy expenditure from accelerometer counts. Anthropometric and training load variables were also included in the regression equation because these have been shown to be correlated to resistance exercise energy expenditure (6,7,20,31,32). Average exercise heart rate was included in the analysis to determine if heart rate data could be used to further improve the estimate of resistance exercise energy expenditure. Statistical significance for all analyses was set at p ≤ 0.05.
Men were significantly taller and had a higher body mass and fat-free mass, whereas the women had a significantly greater body fat percentage and fat mass (all p < 0.01). Men also had significantly greater neck, shoulder, chest, waist, abdomen, hip, arm, wrist, and forearm circumferences, and longer upper-body and lower-body limb lengths (all p < 0.05; Table 2). Age did not differ between men and women. Total lifting volume during the resistance exercise protocol and 10RM strength for all exercises were significantly greater for men than for women (all p < 0.001; Table 3). Resting metabolic rate, gross energy expenditure during exercise, and net energy expenditure during exercise were significantly higher for men than for women, and men had a significantly greater blood lactate response to the exercise protocol (all p ≤ 0.001; Table 4). The duration of the resistance exercise protocol was 28.5 minutes, with 6.0 minutes of exercise time. The average heart rate during the resistance exercise protocol was 121.9 b·min−1.
Accelerometry and Energy Expenditure
There were no significant differences between ActiTrainer waist counts and ActiGraph GT3X waist counts (p > 0.05 for all 3 axes and sum of counts); therefore, the ActiGraph GT3X waist counts were used for all analyses. One-way ANOVA revealed a significant difference (p < 0.001) between the sum of accelerometer counts by location, with the sum of the counts of activity from all 3 axes being greatest at the wrist (134,249 ± 20,567), followed by the ankle (50,722 ± 15,252) and waist (26,730 ± 7,437; Figure 2). Net energy expenditure was significantly correlated with vertical (r = 0.67, p < 0.001; Figure 3), horizontal (r = 0.43, p = 0.02), third axis (r = 0.36, p = 0.048), and sum of counts at the waist (r = 0.50, p = 0.005; Figure 4), and horizontal counts at the wrist (r = –0.40, p = 0.03; Table 5). The sum of the counts of activity at the wrist and ankle did not significantly correlate to net energy expenditure (p > 0.05; Figures 5 and 6).
A linear regression analysis was conducted to develop an equation to predict energy expenditure, with fat-free mass explaining the greatest percentage of the variance in the estimate of resistance exercise energy expenditure (R2 = 0.68). The addition of sex and the sum of accelerometer counts in all 3 axes at the waist increased the final R2 of the regression equation to estimate resistance exercise energy expenditure to 0.73. Although anthropometric and training load variables were correlated with net energy expenditure, these did not have a meaningful effect on enhancing the ability to predict energy expenditure. In addition, average exercise heart rate did not correlate to net energy expenditure (r = 0.21, p = 0.27).
The following equation was developed to estimate resistance exercise energy expenditure using sex (1 = female, 2 = male), fat-free mass in kilograms (FFMkg), and the sum of accelerometer counts in all 3 axes at the waist (counts): net kilocalories = 36.175 (sex) + 0.916 (FFMkg) + 0.001 (counts) – 31.125.
The current study demonstrates that a triaxial accelerometer worn at the waist can estimate resistance exercise energy expenditure. However, based on previous work by Rawson and Walsh (31), a triaxial accelerometer does not offer any significant benefit compared with a uniaxial accelerometer in estimating resistance exercise energy expenditure. A linear regression equation using fat-free mass, sex, and the sum of waist accelerometer counts in all 3 axes was developed and accounted for 73% of the variance in energy expenditure during a resistance exercise protocol. In accordance with our hypothesis, total waist accelerometer counts had a higher correlation (r = 0.50, p = 0.005) with net energy expenditure than total wrist (r = −0.25, p = 0.18) and total ankle (r = −0.07, p = 0.72) counts, and wrist counts (134,249) were significantly greater than ankle (50,722) and waist (26,730) counts (p < 0.001). Contrary to our hypothesis, the addition of exercise heart rate had no meaningful effect on enhancing the ability to predict net energy expenditure.
Resistance training increases daily energy expenditure and is typically prescribed for individuals who are trying to lose weight and improve body composition. Although resistance exercise is often recommended for inclusion in exercise routines for weight loss, there are few data on the amount of energy expended during resistance exercise workouts, possibly because resistance exercise energy expenditure is difficult to assess. In an effort to accurately quantify energy expenditure during resistance exercise, several groups have attempted to estimate energy expenditure using regression equations that take into account anthropometric variables, total weight lifted during a resistance exercise session (20), and distance of weight lifted (6,7,32). Although the linear regression equations generated from these studies may provide simple convenient estimates of energy expenditure, some of the methods used may be impractical (e.g., measuring distance of weight lifted during exercise), and the equations may not be applicable toward all resistance exercise protocols.
This study used a triaxial accelerometer worn at the waist to improve the estimate of resistance exercise energy expenditure using linear regression. However, although a triaxial accelerometer was used, it did not provide any additional benefit to the estimation of energy expenditure over a uniaxial accelerometer. In the previous study by Rawson and Walsh (31), fat-free mass explained the greatest percentage of variance in the estimate of resistance exercise energy expenditure (R2 = 0.85), and the addition of sex and uniaxial accelerometer counts at the waist were able to improve the R2 of the regression equation from 0.85 to 0.90. Similarly, using the same resistance exercise protocol, we found that fat-free mass explained the greatest percentage of variance (R2 = 0.68), and the addition of sex and the sum of triaxial accelerometer counts from all 3 axes at the waist were able to improve the R2 of the regression equation from 0.68 to 0.73. Another linear regression equation was generated using fat-free mass, sex, and vertical accelerometer counts at the waist, which resulted in the same R2 (0.73) as the equation using the sum of waist counts as a variable. This further supports the notion that using the sum of accelerometer counts at the waist offers no benefit over using only vertical counts at the waist when estimating resistance exercise energy expenditure.
While triaxial accelerometers appear to provide no additional benefit in the estimate of resistance exercise energy expenditure over uniaxial accelerometers, a possible reason may be the way in which accelerometer data were interpreted in this study. Although linear regression equations have been used to accurately estimate energy expenditure during several types of physical activity, such as walking, running, and activities of daily living, these activities typically consist of regular rhythmic accelerations, therefore permitting the use of simple linear regression (38). In contrast, both acceleration and energy expenditure during resistance exercise may vary depending on volume and intensity of exercise (16,35), contraction velocity (13,23), and the length of work-rest intervals (30). In addition, during resistance exercise, some limbs are moving while some may be inactive, making accelerometer location crucial when describing movement during resistance exercise. Because of the variable nature of resistance exercise, linear regression analysis of accelerometer data may not be ideal to estimate resistance exercise energy expenditure.
A possible new method of estimating energy expenditure during physical activity may use triaxial accelerometry and real-time classification approaches (e.g., quadratic discriminant analysis, neural network approach, hidden Markov models) for analyzing accelerometer data rather than linear regression analyses (28). Recently, Karantonis et al. (17) demonstrated the ability of a triaxial accelerometer worn at the hip and a real-time classification system to detect 12 different types of daily physical activities with an overall accuracy of 90.8%. Similarly, Bonomi et al. (4) used decision tree models to analyze data from a triaxial accelerometer worn at the lower back and were able to classify 20 different types of activities with an overall accuracy of 93%. A study by Chang et al. (8) that is more practical to the field of strength and conditioning, used naive Bayes classifiers and hidden Markov models to analyze data from 2 triaxial accelerometers: one located within a lifting glove and one worn at the hip. The accelerometers identified 9 different resistance exercises with 90% accuracy and counted repetitions with an overall miscount rate of >5%. Triaxial accelerometers may be more useful in identifying activity types during resistance exercise than uniaxial accelerometers (Figure 7), which may improve the ability to estimate energy expenditure for specific exercises in a resistance exercise protocol. From a practical aspect, this may allow triaxial accelerometers to be used to estimate energy expenditure during a variety of resistance training protocols with different exercise types (e.g., free-weight exercises, machine exercises, Olympic lifts, etc.).
We hypothesized that average exercise heart rate could be used in combination with accelerometer counts to improve the estimate of resistance exercise energy expenditure. However, average exercise heart rate was not significantly correlated with net energy expenditure in this study and therefore did not have a meaningful effect when estimating energy expenditure. Heart rate can be used as a predictor of energy expenditure during certain exercises (e.g., walking, running) because of the linear relationship between heart rate and energy expenditure. However, the linear relationship between energy expenditure and heart rate diminishes at rest and during very high-intensity exercise. Resistance exercise consists mainly of brief periods of high-intensity exercise separated by rest intervals; therefore, a strong relationship between heart rate and energy expenditure may not be seen during this type of exercise. Previously, Strath et al. (36) demonstrated that analyzing continuous, minute-by-minute heart rate and accelerometry data was accurate in estimating free-living energy expenditure. Additionally, Brage et al. (5) were able to use branched-chain modeling of simultaneous heart rate and accelerometry data to accurately estimate energy expenditure during free living and exercise. The methods used by Strath et al. (36) and Brage et al. (5) to combine real-time heart rate and accelerometer data were able to estimate energy expenditure more accurately than accelerometry alone (37). It appears that simultaneously analyzing physiological variables (e.g., heart rate) and accelerometer data may be more effective than linear regression in estimating energy expenditure during resistance exercise. Because movement and heart rate during resistance exercise are quite variable depending on the type of resistance exercise program, it would make sense physiologically and biomechanically to use nonlinear statistical methods to interpret these data. If a nonlinear method of interpreting simultaneous heart rate and accelerometer data were developed, it would be extremely useful for estimating resistance exercise energy expenditure and would make for a practical measurement because the ActiTrainer accelerometer records simultaneous accelerometer and heart rate data in 1-second epochs.
The findings of this study demonstrate that although a triaxial accelerometer worn at the waist can be used to estimate energy expenditure during resistance exercise, it does not provide any added benefit over a uniaxial accelerometer. In addition, average heart rate during the resistance exercise protocol was unable to improve the accuracy of the regression equation. It is possible that triaxial accelerometers and heart rate monitors may be more useful in predicting energy expenditure during resistance exercise when accelerometer and heart rate data are analyzed using more advanced real-time statistical analyses rather than linear regression. Future research should examine the benefit of using real-time accelerometer and heart rate data along with other real-time physiological data to improve the estimate of resistance exercise energy expenditure.
Determining the energy cost of a resistance exercise session is important for athletes and individuals who are focused on energy balance and altering body composition. An accurate measure of resistance exercise energy expenditure is especially important for weight-class athletes who engage in resistance training because their energy balance must be tightly regulated to maintain a proper body weight. Currently, accelerometers appear to be the only practical tool for estimating energy expenditure during resistance exercise because they are portable and do not interfere with resistance exercise movements. In addition, triaxial accelerometers can be useful in tracking the number of repetitions and sets performed during a resistance exercise workout. This can be particularly useful for coaches because they would be able to record each of their athletes' sets and repetitions during a workout simply by downloading data from the athletes' accelerometers. By combining triaxial accelerometry with physiological measures (i.e., heart rate) and using advanced statistical analyses, it may be possible to improve the estimate of resistance exercise energy expenditure and ultimately develop an accelerometer program that can be used to estimate energy expenditure during different types of resistance exercise protocols.
This research was supported by a National Strength and Conditioning Association Graduate Research Grant (to M.J.S.). The results of this study do not constitute endorsement of the product by the authors or the National Strength and Conditioning Association.
1. Baechle TB, Earle RW. Essentials of Strength and Conditioning (3rd ed.). Champaign, IL: Human Kinetics, 2008.
2. Binzen CA, Swan PD, Manore MM. Postexercise oxygen consumption and substrate use after resistance exercise in women. Med Sci Sports Exerc 33: 932–938, 2001.
3. Bloomer RJ. Energy cost of moderate-duration resistance and aerobic exercise. J Strength Cond Res 19: 878–882, 2005.
4. Bonomi AG, Goris AH, Yin B, Westerterp KR. Detection of type, duration, and intensity of physical activity using an accelerometer
. Med Sci Sports Exerc 41: 1770–1777, 2009.
5. Brage S, Brage N, Franks PW, Ekelund U, Wong MY, Andersen LB, Froberg K, Wareham NJ. Branched equation modeling of simultaneous accelerometry and heart rate
monitoring improves estimate of directly measured physical activity energy expenditure. J Appl Physiol 96: 343–351, 2004.
6. Byrd R, Hopkins-Price P, Boatwright J, Kinley K. Prediction of the caloric cost of the bench press. J Appl Sport Sci Res 2: 7–8, 1988.
7. Byrd R, Pierce K, Gentry R, Swisher M. Predicting the caloric cost of the parallel back squat in women. J Strength Cond Res 10: 184–185, 1996.
8. Chang K, Chen MY, Canny J. Tracking free-weight exercises. In: Proceedings of the UbiComp 2007: Ubiquitous Computers. 2007, Berlin, Springer, p. 19–37.
9. . National Strength and Conditioning Association Position Statement: Health aspects of resistance exercise and training. Strength Cond J 23: 9–23, 2001.
10. di Prampero PE, Ferretti G. The energetics of anaerobic muscle metabolism: A reappraisal of older and recent concepts. Respir Physiol 118: 103–115, 1999.
11. Donnelly JE, Blair SN, Jakicic JM, Manore MM, Rankin JW, Smith BK. American College of Sports Medicine Position Stand. Appropriate physical activity intervention strategies for weight loss and prevention of weight regain for adults. Med Sci Sports Exerc 41: 459–471, 2009.
12. Heyward VH, Wagner DR. Applied Body Composition Assessment (2nd ed.). Champaign, IL: Human Kinetics, 2004.
13. Hunter GR, Seelhorst D, Snyder S. Comparison of metabolic and heart rate
responses to super slow vs. traditional resistance training
. J Strength Cond Res 17: 76–81, 2003.
14. Jackson AS, Pollock ML. Generalized equations for predicting body density of men. Br J Nutr 40: 497–504, 1978.
15. Jackson AS, Pollock ML, Ward A. Generalized equations for predicting body density of women. Med Sci Sports Exerc 12: 175–181, 1980.
16. Kang J, Hoffman JR, Im J, Spiering BA, Ratamess NA, Rundell KW, Nioka S, Cooper J, Chance B. Evaluation of physiological responses during recovery following three resistance exercise programs. J Strength Cond Res 19: 305–309, 2005.
17. Karantonis DM, Narayanan MR, Mathie M, Lovell NH, Celler BG. Implementation of a real-time human movement classifier using a triaxial accelerometer
for ambulatory monitoring. IEEE Trans Inf Technol Biomed 10: 156–167, 2006.
18. Kirk EP, Donnelly JE, Smith BK, Honas J, Lecheminant JD, Bailey BW, Jacobsen DJ, Washburn RA. Minimal resistance training
improves daily energy expenditure and fat oxidation. Med Sci Sports Exerc 41: 1122–1129, 2009.
19. Kraemer WJ, Ratamess NA, French DN. Resistance training
for health and performance. Curr Sports Med Rep 1: 165–171, 2002.
20. Kuehl K, Elliot D, Goldberg L. Predicting caloric expenditure during multi-station resistance exercise. J Appl Sport Sci Res 4: 63–67, 1990.
21. Lagally KM, Cordero J, Good J, Brown DD, McCaw ST. Physiologic and metabolic responses to a continuous functional resistance exercise workout. J Strength Cond Res 23: 373–379, 2009.
22. Lohman T, Roche AF, Martorell R. Anthropometric Standardization Reference Manual. Champaign, IL: Human Kinetics, 1988.
23. Mazzetti S, Douglass M, Yocum A, Harber M. Effect of explosive versus slow contractions and exercise intensity on energy expenditure. Med Sci Sports Exerc 39: 1291–1301, 2007.
24. Melanson EL, Sharp TA, Seagle HM, Donahoo WT, Grunwald GK, Peters JC, Hamilton JT, Hill JO. Twenty-four-hour metabolic responses to resistance exercise in women. J Strength Cond Res 19: 61–66, 2005.
25. Ortego AR, Dantzler DK, Zaloudek A, Tanner J, Khan T, Panwar R, Hollander DB, Kraemer RR. Effects of gender on physiological responses to strenuous circuit resistance exercise and recovery. J Strength Cond Res 23: 932–938, 2009.
26. Phillips WT, Ziuraitis JR. Energy cost of the ACSM single-set resistance training
protocol. J Strength Cond Res 17: 350–355, 2003.
27. Phillips WT, Ziuraitis JR. Energy cost of single-set resistance training
in older adults. J Strength Cond Res 18: 606–609, 2004.
28. Pober DM, Staudenmayer J, Raphael C, Freedson PS. Development of novel techniques to classify physical activity mode using accelerometers. Med Sci Sports Exerc 38: 1626–1634, 2006.
29. Pollock ML, Gaesser GA, Butcher JD, Després JP, Dishman RK, Franklin BA, Garber CE. American College of Sports Medicine Position Stand. The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy adults. Med Sci Sports Exerc 30: 975–991, 1998.
30. Ratamess NA, Falvo MJ, Mangine GT, Hoffman JR, Faigenbaum AD, Kang J. The effect of rest interval length on metabolic responses to the bench press exercise. Eur J Appl Physiol 100: 1–17, 2007.
31. Rawson ES, Walsh TM. Estimation of resistance exercise energy expenditure using accelerometry. Med Sci Sports Exerc 42: 622–628, 2010.
32. Robergs RA, Gordon T, Reynolds J, Walker TB. Energy expenditure during bench press and squat exercises. J Strength Cond Res 21: 123–130, 2007.
33. Schuenke MD, Mikat RP, McBride JM. Effect of an acute period of resistance exercise on excess post-exercise oxygen consumption: Implications for body mass management. Eur J Appl Physiol 86: 411–417, 2002.
34. Scott CB. Contribution of blood lactate to the energy expenditure of weight training. J Strength Cond Res 20: 404–411, 2006.
35. Scott CB, Leighton BH, Ahearn KJ, McManus JJ. Aerobic, anaerobic, and excess postexercise oxygen consumption energy expenditure of muscular endurance and strength: 1-set of bench press to muscular fatigue. J Strength Cond Res 25: 903–908, 2011.
36. Strath SJ, Bassett DR Jr, Thompson DL, Swartz AM. Validity of the simultaneous heart rate
-motion sensor technique for measuring energy expenditure. Med Sci Sports Exerc 34: 888–894, 2002.
37. Strath SJ, Brage S, Ekelund U. Integration of physiological and accelerometer
data to improve physical activity assessment. Med Sci Sports Exerc 37: S563–S571, 2005.
38. Troiano RP. Translating accelerometer
counts into energy expenditure: Advancing the quest. J Appl Physiol 100: 1107–1108, 2006.