Rating of perceived exertion (RPE) scales are commonly used during exercise to subjectively estimate the metabolic workload or effort. These scales offer a potentially simple and inexpensive method for the estimation and regulation of exercise intensity during exercise testing, for exercise prescription, and for assessing exercise intensity in field studies. In 1991, a simpler RPE scale ranging from 1 to 10 was first proposed for use in children (32), and since this original work, a number of scales have been developed specifically for use by children (19,20,29,31,33). Children's scales were developed because some children lack the cognitive ability or vocabulary to understand the descriptors of exercise intensity associated with adult perceived exertion scales (5,6,8,22). In an effort to improve validity of the instruments, child RPE scales typically include numbers, verbal descriptors, and illustrations that depict various stages of effort (6).
The OMNI scale of perceived exertion (OMNI scale) is a pediatric specific RPE scale with multiple derivations (18-20,22,29). The OMNI scale, designed for rating the intensity of cycling (OMNI bike, Fig. 1) includes illustrations of a child becoming more fatigued while cycling to the top of a hill. The OMNI bike scale has demonstrated concurrent and construct validity when using a simple estimation-of-effort paradigm during aerobic cycle ergometer exercise (22), and it can also be used for producing a prescribed exercise intensity in children (20). To date, only the scale's designers (22) have demonstrated concurrent and construct validity for the children's version of the OMNI bike scale.
The Cart and Load Effort Rating (CALER) scale of perceived exertion (Fig. 2) was also developed for use during cycle ergometer exercise and is an abridged version of the Children's Effort Rating Table (CERT), with identical numbers and wording (6,17). The CALER scale is an effective tool for regulating exercise intensity in children (6,17). However, the CALER scale has not yet been examined for concurrent validity, nor has the CALER scale been compared against a validated RPE scale to determine its construct validity.
While pediatric-specific RPE scales have been validated for their prediction of effort during submaximal exercise (18,21,22,24,33) there are limited data regarding these scales' ability to accurately reflect perceived effort during specific points of submaximal exercise or during maximal exercise (13,14,29). Indeed, some evidence suggests that children, when performing progressive exercise, may underrepresent their effort relative to their physiologic effort during submaximal (24) and maximal (13) exercise intensities during an estimation paradigm.
The purposes of this study were to use a perceptual estimation paradigm to independently test the concurrent validity of the CALER and OMNI bike perceived exertion scales for children while performing a maximal progressive cycle ergometer exercise test, and to test the construct validity of the CALER scale, using the OMNI bike scale as the criterion. The validity of children's ability to indicate maximal RPE during maximal exercise was also determined.
Study participants included 16 boys (N = 15 Caucasian, N = 1 African American) and 16 girls (N = 16 Caucasian). In the present study, the OMNI bike and CALER RPE scales produced correlation coefficients of r = 0.89 and when compared with V˙O2, respectively, and r = 0.93 and 0.92 when compared with heart rate; respectively. Based on the lowest of these correlation coefficients (r = 0.88) and a null correlation coefficient of r = 0.6, 21 subjects were needed to achieve a power of 0.8 at α = 0.05. Therefore, the current sample size (N = 32) was deemed to provide ample statistical power. These children were recruited from the local community through newspaper advertisements, targeted mailings, and from a database of subjects who had previously contacted the laboratory to participate in studies. Clinical disorders, including cardiovascular, neuromotor, or cognitive disorders that would interfere with exercise testing or perceived exertion, were prescreened, using a phone interview with each child's parent. Before the CALER and OMNI bike scales were described, each child was asked whether he or she had ever seen the scales before. All children indicated that they had not seen either scale before the testing session. Written informed parental consent and child assent were obtained for each subject before participation. This study was approved by the University at Buffalo School of Medicine institutional review board.
A perceptual estimation paradigm was used to test the concurrent validity of the CALER and OMNI bike scales. This was accomplished using a progressive exercise test on a magnetically braked cycle ergometer (model EC-1600, Cateye Fitness, Dallas, TX). Children became familiar with the cycle by pedaling at 50 rpm and 0.6-kp resistance for 2 min. The exercise test then consisted of an initial work rate set at 30 W, which increased by 10 W·min−1 until children could not maintain a pedal cadence of 50 rpm. During the final 20 s of each minute, children reported their undifferentiated (total body) perceived exertion using both the CALER and OMNI bike scales. The workloads for the perceptual estimation paradigm were selected based on previous research that was designed to establish ventilatory threshold and V˙O2peak in children (12). The order of presentation of the two scales was counterbalanced across subjects, meaning that half of the subjects were presented with the CALER scale first and then the OMNI scale during each stage of the cycle ergometer test. The RPE scales were presented in the opposite order for each stage of the test for the other half of the subjects. Heart rate was recorded during the final 20 s of each minute during the test, and V˙O2 was averaged across the last 30 s of each minute of exercise.
Height was assessed using a calibrated stadiometer (Seca, Hamburg, Germany) to the nearest 1.0 mm. A digital scale measured weight to the nearest 0.05 kg, which was rounded to the nearest 0.1 kg, with the subjects wearing shorts and a T-shirt, using a digital scale (Tanita, Arlington Heights, IL). All of the anthropometric measures were performed by the same experienced anthropometrist (J.E.B.).
The rate of oxygen consumption and ventilation were measured continuously throughout the cycle ergometer test, using a Vmax metabolic cart (Sensormedics, Yorba Linda, CA). The Vmax system was calibrated before each test with two separate gas mixtures supplied by Sensormedics, one containing 16% O2 with 4% CO2 and a reference mixture containing 21% O2 with balanced N2. Heart rate was recorded using a Polar HR Monitor (Polar Electro, Inc., Woodbury, NY). V˙O2peak was recorded as the average V˙O2 (mL·kg−1·min−1) during the last 30 s of the final stage of the test (12).
Immediately before the cycle ergometer progressive exercise test, children were shown the CALER and OMNI bike scales one at a time, and their use was explained with a standardized set of instructions. Perceived exertion was defined for both scales as, "How tired does your body feel during exercise?" (29). Ratings were reported as an estimate of the perceived exertion of the total body (undifferentiated). The same written instructions were read to each child to describe the use of the scales. The instructions were taken from the original OMNI scale validation paper (29) and read verbatim when explaining the use of the OMNI bike scale. The instructions for the CALER scale were modified from the instructions from the OMNI scale. The scale numbers and verbal descriptors in the script for the CALER scale were changed to correspond with the low (1: very, very easy) and high (10: so hard I'm going to stop) anchors of the CALER scale. Children were told that there were no "right or wrong" answers. Finally, children were instructed to think about an exercise intensity that was the same as the pictorial descriptor of the boy pedaling at the bottom and the top of the hill (OMNI bike) and with an empty or full wagon (CALER). Cognitive anchors were described for the OMNI bike scale and were modified for the CALER scale to include descriptions of the CALER pictorial descriptors (29,33). For the two scales, children were required to point to a number representing their RPE, because the mouthpiece used to collect expired air made it difficult to provide a verbal rating of RPE.
Differences in physical characteristics and V˙O2peak between boys and girls were analyzed with separate, one-factor analyses of variance (ANOVA). Random effects models were used to test increases in V˙O2, heart rate, and the CALER and OMNI RPE scales across all stages of the exercise test. For each model, linear relationships were tested with the time-variant predictor of exercise test stage and a random intercept. Gender was included as a time-invariant fixed effect, but it was not a significant predictor and was therefore not included in the final models. Children required a different number of stages of the cycle ergometer test to reach V˙O2peak, depending on their fitness level. One advantage of random effects models is that children could complete a different number of stages, yet all children could remain in the analysis without deleting data from the later test stages. Random effects models also account for the relationship of successive (interdependent) measurements of the same child by introducing a random effect for "child" into the linear regression model (9).
Concurrent and construct validity of the CALER and OMNI bike RPE scales was assessed using random effects models and Pearson correlations. For the random effects model, linear relationships were tested with the time-variant predictor of exercise test stage, and a random intercept. Gender was not a significant predictor in any of the models and was therefore not included in the final models. For the correlation analysis, correlation coefficients (r) were calculated for each subject by regressing their CALER RPE scores from each stage of the cycle ergometer exercise test against their OMNI RPE scores from each stage. Each subject's CALER and OMNI bike scores were then correlated with their set of heart rate and V˙O2 scores from each exercise test stage. The correlation coefficients (r) from all subjects were then averaged and reported as a single mean r value for each comparison. Concurrent validity was assessed by testing the linear relationships of RPE scale (CALER, OMNI bike) scores with heart rate and V˙O2 across all stages of the cycle ergometer test (9). Construct validity was assessed by testing the association between the CALER scale scores and the OMNI bike scale across all test stages. The OMNI bike scale served as the construct for the CALER scale because the OMNI scale has been previously validated (22,23,28). Separate tests of proportion were performed to determine whether the proportion of maximal RPE scale (CALER, OMNI bike) during the final stage of the progressive cycle ergometer test differed from 94.5 ± 3%. The value of 94.5 ± 3% (189 ± 6 bpm) was the mean proportion of the predicted maximal heart rate of 200 bpm (25) attained during the final stage of the exercise test. While the average maximal respiratory exchange ratio (RER) children achieved (1.03 ± 0.075) was indicative of maximal effort (RER > 1.0) (26), the maximum percentage of RPE scale indicated was compared only to the percentage of predicted peak heart rate achieved (94.5 ± 3%) because previous research (15) has demonstrated that maximal achieved heart rate is typically lower than predicted maximal heart rate during a maximal cycle ergometer test.
Physical characteristics for boys and girls are shown in Table 1. There were no differences in age (P = 0.99), height (P = 0.10), weight (P = 0.74), and BMI percentile (P = 0.48) (11) for the boys and girls. Boys had significantly greater V˙O2peak (P ≤ 0.05) than the girls. Table 2 lists the V˙O2, heart rate, and CALER and OMNI RPE scores for boys and girls across the 10 stages of the cycle ergometer test. V˙O2 (P < 0.001), heart rate (P < 0.001), and CALER (P < 0.02) and OMNI bike (P < 0.001) RPE scores significantly increased from one stage to the next successive stage across all 10 stages of the cycle ergometer test.
Figures 3 and 4 illustrate the results of the random effects models and correlation analysis used to test the concurrent and construct validity of the RPE scales. Both the OMNI bike and CALER scales were linearly related (P < 0.001, random effects model) with V˙O2 (Fig. 3, upper panel) (r = 0.89 and 0.88, respectively) and heart rate (Fig. 3, lower panel) (r = 0.93 and 0.92, respectively) across cycle ergometer test stages. An additional random effects model and correlation analysis (Fig. 4) was used to determine the construct validity of the CALER RPE scale against the validated OMNI bike RPE scale. Scores from the CALER scale were significantly (P < 0.001 random effects model) associated with scores from the OMNI bike scale (r = 0.93) across stages of the cycle ergometer test.
Figure 5 illustrates the percentage maximum of the RPE scales during the final stage of the cycle ergometer test. Both the proportion of the CALER (75 ± 20%, P < 0.001) and OMNI bike (74 ± 19%, P < 0.001) RPE scales were less (P ≤ 0.001) than the proportion of predicted maximal heart rate (94.5 ± 3%) achieved during the final stage of the cycle ergometer test.
A perceptual estimation paradigm was used to test the concurrent validity of the CALER and OMNI bike perceived exertion scales during a maximal progressive cycle ergometer exercise test. A novel aspect of this study is that it is the first to examine the concurrent validity of the CALER scale. The CALER scale contains illustrations of an individual pulling a cart filled with bricks, which differ from the OMNI bike scale's illustrations of a child pedaling a bike up hill. Both the CALER and OMNI bike scales displayed concurrent validity, as demonstrated by their ability to predict V˙O2 (r = 0.88 and 0.89, respectively) and heart rate (r = 0.92 and 0.93, respectively) during the course of a maximal progressive cycle ergometer exercise test, and, despite their different illustrations, each of the scales performed equally well. These results corroborate previous data that demonstrated a strong correlation between various forms of the OMNI scale and physiologic measures of effort during progressive exercise (22,24,29). Whereas the CALER scale has been demonstrated to be an effective tool for regulating exercise intensity in children (6,17), this is the first experiment to test the construct validity of the CALER scale, and it is a valid method for estimating effort during a progressive exercise test. As shown in Figure 3, the scales had very similar relationships with physiologic measures, and therefore it is not surprising that the CALER scale demonstrated excellent construct validity when compared with the OMNI scale (r = 0.93) (Fig. 4). These results were similar in both genders and occurred even though the CALER scale uses a slightly different numeric scale and different visual depicters from the OMNI bike scale.
Previous research (7,10,27) has noted that the theoretical maximal RPE of 20 on the Borg scale is infrequently reported at volitional exhaustion in adults. The present study extends this finding to children. In a previous experiment (24) that used a progressive submaximal treadmill exercise protocol, children indicated only 52% of the maximum RPE scale while exercising at a heart rate and V˙O2 of 82% and 70% of their predicted maximum values, respectively. Therefore, if children were to indicate a maximal score on the RPE scale at the predicted maximal heart rate or V˙O2, it was speculated that they would have needed to increase their RPE scores at a greater rate than had occurred at lower workloads. As hypothesized from previous work (24), at maximal exertion children indicated CALER and OMNI bike RPE scores that were 75 ± 20% and 74 ± 19% of the total scale, which was less than the proportion of predicted maximal heart rate (94.5 ± 3%). Therefore, despite the fact that the RPE scales were compared only against the percentage of predicted heart rate maximum achieved during the final stage of the maximum cycle ergometer test, and not against 100%, the RPE scales, which were specifically designed to assess effort in bicycling exercise in children, did not perform as well at maximal exertion. One explanation for this result is that children are unfamiliar with the physical sensations associated with maximum effort because they may seldom experience it. Thus, they may lack the necessary perceptual memory anchors required to accurately indicate maximal RPE when exercising at maximal effort. During their final stage of exercise, children may have felt that because they were still turning the pedals and exercising, they were not at maximal effort. Children who were repeatedly exposed to a maximal exercise test on a cycle ergometer and who were provided feedback regarding their ability to match their perceived and physiologic effort may learn to assign perceived exertion scores that are more similar to their percentage of physiologic effort at near-maximal and maximal exercise. Previous studies designed to evaluate the reliability of the scales for regulating exercise intensity in children have demonstrated that children's interpretation of these scales may improve with repeated exposure (6,17). However, while this approach may improve a child's ability to recognize his or her own maximal effort, it would improve the functionality of the RPE scales only for children who had received such training. Additionally, the number of exercise sessions necessary to improve a child's ability to accurately and reliably assign an RPE to his or her percentage of physiologic effort has not yet been studied.
Although these scales appear to successfully illustrate effort during submaximal exercise, the scales may benefit from modifications to their upper range. Perhaps modifying verbal descriptors, or changing the linearity of the scales so that the final stages increase at a greater rate or increase in a curvilinear fashion, could improve agreement with physiologic measures at peak exercise intensities (3). A curvilinear scale has been proposed (4) and is being evaluated to determine whether it provides improved validity during intense exercise. It is also possible that different instructions for the scales' use during near-maximal and maximal exercise may improve validity. Future research should consider further exploration of these potential modifications.
Children's reluctance to report pain or discomfort (1,16,30) could also lead to a discrepancy in the proportions of the RPE scale and maximal heart rate. Pain ratings reported by adults increase with RPE during both progressive and interval exercise (2). It has not yet been determined whether children, who are reluctant to report pain or discomfort in clinical situations (1,16,30), are also reluctant to report their levels of exertion, which they may associate with discomfort, during exercise testing.
In conclusion, concurrent validity was demonstrated for both the CALER and OMNI bike scales for assessing effort perception in youth performing a maximal progressive exercise test on a cycle ergometer. The CALER and OMNI bike RPE scales performed similarly well when assessing perceived exertion in children, and neither scale was superior. The present study extends previous research by being the first to demonstrate the construct and concurrent validity of the CALER scale. Construct validity was established, as the CALER scale was significantly associated with the previously validated OMNI bike scale across a progressive exercise test. However, both the construct and concurrent validity demonstrated in the present study are limited to the estimation paradigm used during a progressive exercise test. While progressive exercise tests are commonly used (22-24) to assess the validity of RPE scales, such protocols, which systematically increase in difficulty at predetermined time points, may lend themselves to a certain degree of validity, as it is possible that the progressively increasing intensity of the exercise test encourages subjects to respond with a greater numeric value from one stage to the next. Future research should consider using a series of trials conducted at random percentages of peak resistance or peak capacity, in an effort to avoid the possibility that progressive exercise may encourage increasing RPE scores.
This study also extends previous research by demonstrating that children, when using pediatric-specific RPE scales, do not indicate a proportion of an RPE scale that is equal to the proportion of physiologic effort. The cause for this disparity is unclear. Future research is warranted, to determine whether repeated exposure to maximal exertion would improve children's ability to perceive their effort at or near maximal exertion. The scales' instructions may also require modification, or children may be reluctant to report high perceived efforts during peak exercise. The development of well-validated RPE scales that provide an accurate estimation of effort across a wide range of intensities would be a simple and inexpensive tool for assessing and regulating exercise intensity during exercise tests and, possibly, the assessment of the intensity of free-living physical activity.
This work was supported by Gatorade Sports Science Institute and Mark Diamond Research Foundation grants to Dr. Jacob E. Barkley and RO1 HD42766 to Dr. Roemmich. Dr. Barkley is currently with the School of Exercise Leisure and Sport, Kent State University, Kent, OH.
1. Clarke EB, French B, Bilodeau ML, Capasso VC, Edwards A, Empoliti J. Pain management knowledge, attitudes and clinical practice: the impact of nurses' characteristics and education. J Pain Symptom Manage
2. Cook DB, O'Connor PJ, Eubanks SA, Smith JC, Lee M. Naturally occurring muscle pain during exercise: assessment and experimental evidence. Med Sci Sports Exerc
3. Eston R, Lamb KL. Effort perception
. In: Armstrong N, Van Mechelen W, editors. Pediatric Exercise Science and Medicine
. Oxford (UK): Oxford University Press; 2000. p. 85-91.
4. Eston R, Parfitt G. Effort Perception
. In: Armstrong N, editor. Paediatric Exercise Physiology
. London (UK): Elsevier; 2006. p.275-98.
5. Eston RG, Lamb KL, Bain A, Williams AM, Williams JG. Validity of a perceived exertion scale for children: a pilot study. Percept Mot Skills
6. Eston RG, Parfitt G, Campbell L, Lamb KL. Reliability of effort perception
for regulating exercise intensity in children using the cart and load effort rating (CALER) scale. Pediatr Exerc Sci
7. Faulkner J, Parfitt G, Eston R. Prediction of maximal oxygen uptake from the ratings of perceived exertion and heart rate during a perceptually-regulated sub-maximal exercise test in active and sedentary participants. Eur J Appl Physiol
8. Groslambert A, Mahon AD. Perceived exertion: influence of age and cognitive development. Sports Med
9. Hedeker D, Gibbons RD. MIXOR: a computer program for mixed-effects ordinal regression analysis. Comput Methods Programs Biomed
10. Kay D, Marino FE, Cannon J, St Clair Gibson A, Lambert MI, Noakes TD. Evidence for neuromuscular fatigue during high-intensity cycling in warm, humid conditions. Eur J Appl Physiol
11. Kuczmarski RJ, Ogden CL, Guo SS, et al. 2000 CDC growth charts for the United States: methods and development. Vital Health Stat
12. Mahon AD, Duncan GE, Howe CA, Del Corral P. Blood lactate and perceived exertion relative to ventilatory threshold: boys versus men. Med Sci Sports Exerc
13. Mahon AD, Ray ML. Ratings of perceived exertion at maximal exercise in children performing different graded exercise test. J Sports Med Phys Fitness
14. Marinov B, Mandadjieva S, Kostianev S. Pictorial and verbal category-ratio scales for effort estimation in children. Child Care Health Dev
15. Moreira-da-Costa M, Russo AK, Picarro IC, et al. Maximal oxygen uptake during exercise using trained or untrained muscles. Braz J Med Biol Res
16. O'Brien S, Dalton JA, Konsler G, Carlson J. The knowledge and attitudes of experience oncology nurses regarding the management of cancer-related pain. Oncol Nurs Forum
17. Parfitt G, Shepherd P, Eston RG. Reliability of effort production using the children's CALER and BABE perceived exertion scales. J Exerc Sci Fitness
18. Pfeiffer KA, Pivarnik JM, Womack CJ, Reeves MJ, Malina RM. Reliability and validity of the Borg and OMNI rating of perceived exertion scales in adolescent girls. Med Sci Sports Exerc
19. Robertson RJ, Goss FL, Andreacci JL, et al. Validation of the children's OMNI-resistance exercise scale of perceived exertion. Med Sci Sports Exerc
20. Robertson RJ, Goss FL, Bell JA, et al. Self-regulated cycling using the children's OMNI scale
of perceived exertion. Med Sci Sports Exerc
21. Robertson RJ, Goss FL, Boer N, et al. OMNI scale
perceived exertion at ventilatory breakpoint in children: response normalized. Med Sci Sports Exerc
22. Robertson RJ, Goss FL, Boer NF, et al. Children's OMNI scale
of perceived exertion: mixed gender and race validation. Med Sci Sports Exerc
23. Robertson RJ, Goss FL, Dube J, et al. Validation of the adult OMNI scale
of perceived exertion for cycle ergometer
exercise. Med Sci Sports Exerc
24. Roemmich JN, Barkley JE, Epstein LH, Lobarinas CL, White TM, Foster JH. Validity of PCERT and OMNI walk/run ratings of perceived exertion. Med Sci Sports Exerc
25. Rowland T. Aerobic exercise testing protocols. In: Pediatric Laboratory Exercise Testing: Clinical Guidelines
. Champaign (IL): Human Kinetics Press; 1993. p. 19-41.
26. Rowland T. Exercise testing. In: Rowland T, editor. Development Exercise Physiology
. Champaign (IL): Human Kinetics; 1996. p.27-47.
27. St Clair Gibson A, Lambert MI, Hawley JA, Broomhead SA, Noakes TD. Measurement of maximal oxygen uptake from two different laboratory protocols in runners and squash players. Med Sci Sports Exerc
28. Utter AC, Robertson RJ, Green JM, Suminski RR, McAnulty SR, Nieman DC. Validation of the adult OMNI scale
of perceived exertion for walking/running exercise. Med Sci Sports Exerc
29. Utter AC, Robertson RJ, Nieman DC, Kang J. Children's OMNI scale
of perceived exertion: walking/running evaluation. Med Sci Sports Exerc
30. Van Hulle Vincent C. Nurses' knowledge, attitudes, practices: regarding children's pain. MCN Am J Matern Child Nurs
31. Williams JG, Eston R, Furlong B. CERT: a perceived exertion scale for young children. Percept Mot Skills
32. Williams JG, Eston RG, Stretch C. Use of the rating of perceived exertion to control exercise intensity in children. Pediatr Exerc Sci
33. Yelling Y, Lamb KL, Swaine IL. Validity of a pictorial perceived exertion scale for effort estimation and effort production during stepping exercise in adolescent children. Eur Phys Educ Rev
Keywords:©2008The American College of Sports Medicine
CYCLE ERGOMETER; CHILD; EFFORT PERCEPTION; OMNI SCALE