Eur J Appl Physiol (2007) 100:645–651 DOI 10.



Cross-validation of the 20- versus 30-s Wingate anaerobic test
C. Matthew Laurent Jr. · Michael C. Meyers · Clay A. Robinson · J. Matt Green

Accepted: 19 March 2007 / Published online: 12 April 2007 © Springer-Verlag 2007

Abstract The 30-s Wingate anaerobic test (30-WAT) is the most widely accepted protocol for measuring anaerobic response, despite documented physical side eVects. Abbreviation of the 30-WAT without loss of data could enhance subject compliance while maintaining test applicability. The intent of this study was to quantify the validity of the 20-s Wingate anaerobic test (20-WAT) versus the traditional 30-WAT. Fifty males (mean § SEM; age = 20.5 § 0.3 years; Ht = 1.6 § 0.01 m; Wt = 75.5 § 2.6 kg) were randomly selected to either a validation (N = 35) or crossvalidation group (N = 15) and completed a 20-WAT and 30-WAT in double blind, random order on separate days to determine peak power (PP; W kg¡1), mean power (MP; W kg¡1), and fatigue index (FI; %). Utilizing power outputs (relative to body mass) recorded during each second of both protocols, a non-linear regression equation (Y20WAT+10 = 31.4697 e¡0.5[ln(Xsecond/1174.3961)/2.63692]; r2 = 0.97; SEE = 0.56 W kg¡1) successfully predicted (error »10%) the Wnal 10 s of power outputs in the crossvalidation population. There were no signiWcant diVerences between MP and FI between the 20-WAT that included the predicted 10 s of power outputs (20-WAT+10)

and the 30-WAT. When derived data were subjected to Bland–Altman analyses, the majority of plots (93%) fell within the limits of agreement (§2SD). Therefore, when compared to the 30-WAT, the 20-WAT may be considered a valid alternative when used with the predictive non-linear regression equation to derive the Wnal power output values. Keywords Leg power · Work capacity · Cycle ergometry · Sprint test

Introduction The most commonly employed protocol for the measurement of anaerobic response is the 30-s Wingate anaerobic test (30-WAT). Developed almost 30 years ago, the 30WAT involves a maximal exertion bout on a cycle ergometer to evaluate peak power (PP), mean power (MP), and fatigue index (FI; Bar-Or 1987; Bar-Or et al. 1977). When performing a 30-WAT, a subject typically exhibits a sharp rise in power output, reaching peak power within the Wrst few seconds. Typically, subjects are unable to maintain this output, leading to an exponential decline in power throughout the remaining duration of the test (Bar-Or et al. 1977; Marquardt et al. 1993). During this prolonged period of maximal eVort, the accumulation of [H+] and lactate as byproducts of anaerobic glycolysis results in a drop in blood pH (lactic acidosis). The increased acidity impairs enzyme activity involved in energy metabolism and reduces maximal muscle Wbre recruitment (Allen et al. 1992; Davis 1985). In addition, the acute increase of blood glucose as a substrate for glycolysis during maximal exercise can result in hypoglycemia (Vincent et al. 2004). Together these responses may result in unwelcome physical side eVects, including fatigue, headache, dizziness, and

C. Matthew Laurent Jr. (&) · J. Matt Green Department of Kinesiology, The University of Alabama, Box 83012, Moore Hall, Tuscaloosa, AL, 35487-0312, USA e-mail: M. C. Meyers Human Performance Research Laboratory, Department of Sports and Exercise Sciences, West Texas A&M University, Canyon, TX 79016, USA C. A. Robinson Department of Agriculture, West Texas A&M University, Canyon, TX 79016, USA



Eur J Appl Physiol (2007) 100:645–651

nausea (Jacobs et al. 1982). Subject awareness of these side eVects during anaerobic testing may result in submaximal eVort, high attrition rates, or unsuccessful subject completion of subsequent sport performance testing (Jacobs et al. 1982; Marquardt et al. 1993). In 1987, Vandewalle et al. reported high correlation coeYcients between data collected at the 20th and 30th seconds of a WAT. Consequently, they proposed to curtail the duration of the WAT in order to diminish the stress of the test and the involvement of the aerobic metabolism. However, this previous study did not compare the 20-WAT and 30-WAT but the data collected at the 20th and 30th second of the same test. Hence this previous study failed to prove that results of a 20-WAT would have been the same as results collected during the Wrst 20 s of a 30-WAT because of a possible submaximal eVort strategy during a stressful 30-WAT. More recently, Marquardt et al. (1993) reported high intraclass correlations (r > 0.95) between results of a 20 and 30 Wingate tests in 14 subjects and concluded “that a 20-s Wingate test may be a valid and less strenuous alternative to the 30-s Wingate test.” Since its Wrst publication, the 30-WAT has been used in hundreds of studies published in the international literature and it would be interesting to be able to compare the results of a future 20-s anaerobic test with all these experimental data. Moreover, it is likely that many athletes have performed several traditional 30-WATs and, subsequently, could be reluctant to perform new tests whose results cannot be compared with previous scores. The prediction of 30-WAT indices from 20-WAT performances is not individual but statistical in the study by Vandewalle et al. (1987a). In that study, the statistical prediction of the amount of work performed at the 30th seconds (an equivalent of MP) from 20-s data was probably accurate (r = 0.989), however, the regression coeYcient between pedal rate at 20th and 30th was lower (r = 0.882). In the present study we validate a 20-WAT by comparing the data collected during 20 and 30 s all-out exercises in 50 subjects. Moreover, we propose a method which enables an individual and accurate prediction of 30-WAT indices from the data of a 20-WAT (20-WAT+10).

randomly selected students who volunteered from university level physical education classes. Subjects were encouraged not to participate in any strenuous exercise during the 24 h prior to testing, and initially reported to the laboratory 4 h postabsorptive. Subjects were fully informed of the nature of the study and provided written, informed consent in accordance with the accepted guidelines of the Institutional Review Board of the university and the American College of Sports Medicine (1997). Procedures All subjects initially reported to the laboratory whereupon each individual had their height (m) and body mass (kg) determined using calibrated physician scales. Anaerobic response was quantiWed utilizing both a 20-WAT and a 30WAT, performed in a double blind, randomly determined order to measure peak power (W kg¡1), mean power (W kg¡1), and fatigue index (%). The maximum power achieved during the Wrst 5 s of the test is deWned as PP, while MP is deWned as the average power achieved throughout the trial, and the FI reXects the percent power decline during the trial (Bar-Or 1987; Bar-Or et al. 1977). Both WAT protocols were conducted while pedaling a calibrated Monark model 824E cycle ergometer (Monark AB, Varberg, Sweden) with integrated laser-based sensor and computer software (Sports Medicine Industries, Inc., St. Cloud, MN). Prior to both tests, each subject began a warm-up phase consisting of alternating three 30-s intervals of active rest (pedaling against no resistance at 60 rpm) with three 30-s intervals pedaling against increasing resistance of 25, 50, and 75% of test resistance (Vanderford et al. 2004). Prior research has indicated that this form of pre-test loading elicits optimal power production during supramaximal exercise (Burnley et al. 2005). Test resistance was calculated by multiplying the subject’s body mass (kg) by 0.075. Following completion of the warm-up, the subject continued to pedal at 60 rpm with no resistance for another 2 min until initiation of the WAT. Following a 10-s countdown, the resistance was immediately added and the subject was verbally encouraged to pedal as fast as possible for 20 or 30 s. Relative power outputs (W kg¡1) were measured during each second of the two trials, and PP, MP, and FI were calculated according to accepted procedures (Bar-Or 1987; Bar-Or et al. 1977). Within each subject, the 20-WAT and 30-WAT were completed at the same time of day, no less than 48 h and no more than 7 days apart in order to ensure optimal recovery and minimize any confounding anthropometric changes, respectively. In accordance with previously published research regarding test–retest reliability of the WAT (Bar-Or 1987; Granier et al. 1995; Kaczkowski et al. 1982; Vandewalle et al. 1987a), a PP increase ¸8% was considered

Methods Subjects The participants for this study included 50 male, collegeaged students (mean § SEM; age = 20.5 § 0.3 years; Ht = 175.8 § 1.2 cm; Wt = 75.5 § 2.6 kg) with no known cardiovascular/pulmonary disease, metabolic disorders, or medical contraindications to exercise as determined by selfresponse medical history form and interview. Subjects were


Eur J Appl Physiol (2007) 100:645–651


to be a distorted response, presumably due to a learning eVect. In these instances, the initial protocol was repeated 48 h later and used for comparison. In 13 instances, the diVerence between the two protocols still exceeded the stated criteria and, consequently, these subjects were omitted from the study. To assess the presence of negative side eVects following testing, a basic approach was taken. Subjects were asked to provide a yes/no answer with respect to presence of nausea, light-headedness, leg fatigue, and/or any other physical side eVects. This approach was advantageous in that it was not diYcult to administer and permitted the evaluation of side eVects immediately proximal to the test. This was important as side eVects may often be acute and transient and in such case a written survey completed by subjects after recovering from testing would provide less valid information. Statistical analyses Data were subjected to multivariate analyses of variance (MANOVAs), where appropriate, utilizing the Statistical Package for Social Sciences (SPSS, Inc., Chicago, IL) to determine a signiWcant main eVect between protocols. When univariate post hoc procedures (AVOVA) indicated signiWcant diVerences between relative MP outputs and FI, the sample of 50 subjects were randomly assigned to either a validation group (N = 35) or cross-validation group (N = 15). Both linear and non-linear regression analyses were employed to develop an equation in order to determine the best model to predict the Wnal 10 s of relative power outputs. These values were derived from power outputs obtained during seconds 11–20 of the 20-WAT relative to those obtained during seconds 21–30 of the 30-WAT from the validation group. The non-linear formula employed to Wt the data was a three parameter, peak, lognormal equation of the form Y = a e¡0.5[(ln (X/X0)/b)2], where a, b, and X0 are the parameters Wtted, e¡0.5 is an exponential function, ln is the command to perform a normal logarithmic procedure, and X is the second for which the relative power output is being determined. Following regression analyses, the predicted Wnal 10 s were combined with the observed 20-WAT data (20-WAT+10) recorded in the cross-validation group to demonstrate the validity and applicability of the derived equation to a target population. ANOVAs were performed to identify signiWcant diVer-

ences among the PP, MP, and FI of the 20-WAT+10 and the 30-WAT relative power output. To further assess validity and reliability, the Bland–Altman method of comparison was employed to determine the limits of agreement (mean § 2SD) between the two protocols (Altman and Bland 1983; Bland and Altman 1986) within the cross-validation group. Statistical signiWcance was determined a priori at the P = 0.05 level. All data are presented as mean § SEM unless otherwise noted.

Results Peak power, mean power, and fatigue index The mean relative power outputs recorded during the 20-WAT as well as the 30-WAT along with their respective P values and percent diVerences are presented in Table 1. As expected, a signiWcant MANOVA was found for the Wilks’ Lambda rank variable (F3,80 = 0.723; P < 0.001; n ¡ = 0.998) between protocols. Subsequent ANOVAs indicated no signiWcant diVerences between relative PP outputs recorded during the 30-WAT and the 20-WAT; however, signiWcant diVerences were found with regard to relative MP and FI. Mean relative power outputs were similar throughout the Wrst 20 s of both trials, however, a substantial power decline was observed during the Wnal 10 s of the 30-WAT (Fig. 1). Linear versus non-linear regression prediction model Results from linear and non-linear regression analyses revealed that the non-linear model yield similar coeYcients of determination between models. There was an additional two percent of variation observed was explained when using a non-linear versus a linear prediction model (r2 = 0.97 vs. r2 = 0.95, respectively). Additionally, a noticeable discrepancy was revealed in the standard error of the estimate between models, with the non-linear model reporting an error of »10% (SEE = 0.56 W kg¡1) when compared to an error of »15% (SEE = 0.75 W kg¡1) of the linear model observed within the measured population. Consequently, the non-linear regression model was employed to validate and cross-validate the 20-WAT versus 30-WAT.

Table 1 Mean anaerobic power and capacity for 20-WAT and 30-WAT (N = 50)

Variable Peak power (W kg¡1) Mean power (W kg¡1) Fatigue index (%)

30-WAT Mean § SEM 6.8 § 0.2 4.5 § 0.1 58.3 § 1.7

20-WAT Mean § SEM 6.7 § 0.2 5.0 § 0.2 46.1 § 1.7


DiVerence (%) 1.5 10.0 20.9

0.54 0.01* <0.01*

* SigniWcant at P < 0.05 level


648 Fig. 1 Average relative power outputs of all subjects (N = 50) between 20-WAT and 30-WAT

Eur J Appl Physiol (2007) 100:645–651

(N =50)

6.0 5.0


4.0 3.0

2.0 1.0 20-WAT 0.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 30-WAT


Predicted versus observed MP and FI The relative power outputs of the 20-WAT+10 compared to the observed relative power outputs of the 30-WAT for all 35 subjects in the validation group (Fig. 2) was modeled and curve Wtted for all individuals in a three parameter, peak, log-normal equation (Y20WAT+10 = 31.4697 e¡0.5 [ln (Xsecond/1174.3961)/2.63692]). The model was signiWcant
Fig. 2 Average relative power outputs of the validation group (N = 35) between 20-WAT+10 and 30-WAT
(N = 35)
7.0 6.5 6.0 5.5 W·kg-1 5.0 4.5 4.0 3.5 3.0 2.5 30-WAT 2.0

and the slope and intercept in the equation were signiWcantly diVerent from zero (P < 0.001). The mean relative power outputs recorded during the 20WAT+10 as well as the 30-WAT, their respective P values, and percent diVerences are presented in Table 2. MANOVA indicated no signiWcant main eVect for the Wilks’ Lambda rank variable (F3,80 = 0.182; P = 0.909; n ¡ = 0.082) between a 20-WAT+10 and 30-WAT.


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30


Table 2 Mean anaerobic power and capacity for the 20-WAT+10 and the 30-WAT in the validation group (N = 35) Variable Peak power (W kg¡1) Mean power (W kg¡1) Fatigue index (%) 30-WAT Mean § SEM 6.6 § 0.2 4.3 § 0.1 59.4 § 1.7 20-WAT+10 Mean § SEM 6.6 § 0.2 4.4 § 0.1 57.5 § 1.2 P 0.75 0.71 0.20 DiVerence (%) 0.0 2.3 3.3

No values were signiWcantly diVerent at the P < 0.05 level


Eur J Appl Physiol (2007) 100:645–651 Table 3 Mean anaerobic power and capacity for the 20-WAT+10 and the 30-WAT in the cross-validation group (N = 15) Variable Peak power (W kg¡1) Mean power (W kg ) Fatigue index (%)


30-WAT Mean § SEM 6.9 § 0.2 4.6 § 0.2 55.4 § 2.4

20-WAT+10 Mean § SEM 7.1 § 0.3 4.6 § 0.2 58.2 § 1.7

P 0.06 0.58 0.08

DiVerence (%) 2.9 0.0 4.9

No values were signiWcantly diVerent at the P < 0.05 level

The mean observed and predicted relative power outputs for the 20-WAT+10 as well as the 30-WAT in the crossvalidation group, their respective P values, and percent diVerences are presented in Table 3. Furthermore, ANOVAs indicated no signiWcant diVerences between relative PP outputs, relative MP outputs, or FI between the two protocols in the cross-validation group (Fig. 3). Figures 4, 5 illustrate the limits of agreement (§2SD) as determined by the Bland–Altman method of comparison. Predicted values of MP and FI demonstrated acceptable agreement when compared to the actual values of MP and
Fig. 3 Average relative power outputs of the cross-validation group (N = 15) between 20-WAT+10 and 30-WAT

FI observed during the 30-WAT trials, with 93% of all observations falling within the §2SD limits of agreement.

Discussion The 30-WAT has been shown to be highly reliable and applicable in a wide variety of settings, having test–retest reliability of 0.90–0.97 (Bar-Or 1987; Bar-Or et al. 1977; Kaczkowski et al. 1982; Vandewalle et al. 1987b). The 30-WAT has been used to evaluate anaerobic performance

(N = 15)
7.0 6.5 6.0 5.5 W·kg-1 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30




Observed Difference Between Protocols

Fig. 4 Bland–Altman plots of relative mean power outputs illustrating the upper and lower levels of agreement (dashed lines) between 30-WAT and 20-WAT+10

(N = 15)
2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8 -1.0 -1.2 -1.4 -1.6 -1.8 -2.0 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Mean Power



650 Fig. 5 Bland–Altman plots of fatigue index illustrating the upper and lower levels of agreement (dashed lines) between 30-WAT and 20-WAT+10
(N = 15)
24.0 22.0 20.0 18.0 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 -2.0 -4.0 -6.0 -8.0 -10.0 -12.0 45.0

Eur J Appl Physiol (2007) 100:645–651

Observed Difference Between Protocols






Fatigue Index (%)

among children, adolescents, and adults with activity levels ranging from sedentary to athletic (Bar-Or 1987; Groussard et al. 2003; Mastrangelo et al. 2004; Van Someren and Palmer 2003; Vincent et al. 2004). However, the limitations and applicability of the WAT to a variety of populations, including high-level athletes, has been well documented in other studies (Beneke et al. 2002; Medbø and Tabata, 1993). Additionally, detrimental physical side eVects have been documented, and may constitute a threat to optimal subject compliance (Maud and Schultz 1989; Ulmer 1996). Abbreviation of the 30-WAT without loss of data could potentially ensure subject compliance while maintaining test applicability (Marquardt et al. 1993; Smith and Hill 1991). Therefore, the purpose of this study was to quantify the validity of and, subsequently, cross-validate the 20WAT versus 30-WAT. As expected, results from this study indicate an abbreviated version of the 30-WAT produced signiWcantly diVerent relative MP outputs and FI values. In the study by Marquardt et al. (1993), average power output (PO) from seconds 15–20 during 20-WAT was apparently correlated with PO from seconds 15–20 during 30 WAT instead of PO from seconds 25–30, which did not enable the computation of FI during 30-WAT. In a study by Vandewalle et al. (1987a), the correlation coeYcient between velocity at 20 and 30 s was lower than the correlation coeYcient between MP (0.882 vs. 0.989) although these velocities were measured during the same exercise. In the present study, the fatigue indices (FI) corresponding to 20-WAT+10 and 30WAT were calculated for exercises performed on separate days. Nevertheless, the diVerences between FI were only 3.3 and 4.9% (Tables 2, 3 for the validation and cross validation studies, respectively) although the test–retest correlation coeYcients in the literature are low for the fatigue index. Consequently, it is possible to predict the Wnal 10 s

of relative power outputs and attain similar relative MP outputs and FI values observed during a traditional 30-WAT with the use of the derived non-linear regression equation. Comparison of the 20-WAT versus the 30-WAT As seen in Figs. 1, 2, 3, each subject achieved maximal power output within the initial 5 s of the anaerobic bouts. This was followed by a gradual decline in relative power, as expected, throughout the remainder of the test, resulting in a minimal power output during the Wnal 5 s, as observed in earlier studies (Ansley et al. 2004; Bar-Or 1987; Calbet et al. 2003; Gastin 2001; Granier et al. 1995; Murphy et al. 1986; Smith and Hill 1991; Vandewalle et al. 1987a). The lack of signiWcant diVerence in PP reported in this study indicates that similar eVort was demonstrated over both trials. Despite similar relative power outputs observed during the Wrst 20 s of both trials, signiWcant diVerences were observed between the overall relative MP outputs and FI values between the two protocols, as would be expected. While lactate and acidity were not directly assessed in the current study, it is plausible that these by-products potentially associated with fatigue may have diVered between trials and consequently aVected performance. However, it is strongly emphasized that without measures, any conclusions as such would be speculative at best. Validity of the 20-WAT+10 versus the 30-WAT Following the derivation of the non-linear regression equation, predicted power output and FI values resulted in similar relative MP outputs and FI values between the 20WAT+10 and the 30-WAT in the validation and cross validation groups (Figs. 1, 2). Both PP and MP values from the 20-WAT+10 and 30-WAT met or exceeded the test–retest


Eur J Appl Physiol (2007) 100:645–651

651 Bar-Or O (1987) The Wingate anaerobic test: an update on methodology, reliability and validity. Sports Med 4:381–394 Bar-Or O, Dotan R, Inbar O (1977) A 30 s all-out ergometric test: its reliability and validity for anaerobic capacity. Isr J Med Sci 13:126 Beneke R, Pollmann C, Bleif I, Leithauser RM, Hutler M (2002) How anaerobic is the Wingate anaerobic test for humans? Eur J Appl Physiol 87:388–392 Bland JM, Altman DG (1986) Statistical methods for assessing agreement between two methods of clinical measurements. Lancet 1:307–310 Burnley M, Doust JH, Jones AM (2005) EVects of prior warm-up regime on severe-intensity cycling performance. Med Sci Sports Exerc 37:838–845 Calbet JA, De Paz JA, Garatachea N, Cabeza de Vaca S, Chavarren J (2003) Anaerobic energy provision does not limit Wingate exercise performance in endurance-trained cyclists. J Appl Physiol 94:668–676 Davis JA (1985) Anaerobic threshold: review of the concept and directions for future research. Med Sci Sports Exerc 17:6–21 Gastin PB (2001) Energy system interaction and relative contribution during maximal exercise. Sports Med 31:725–741 Granier P, Mercier B, Mercier J, Anselme F, Prefaut C (1995) Aerobic and anaerobic contribution to Wingate test performance in sprint and middle-distance runners. Eur J Appl Physiol 70:58–65 Groussard C, Machefer G, Rannou F (2003) Physical Wtness and plasma non-enzymatic antioxidant status at rest and after a Wingate test. Can J Appl Physiol 28:79–92 Jacobs I, Bar-Or O, Karlsson J, Dotan R, Tesch P, Kaiser P, Inbar O (1982) Changes in muscle metabolites in females with 30-s exhaustive exercise. Med Sci Sports Exerc 14:457–460 Kaczkowksi W, Montgomery DL, Taylor AW, Klissouras V (1982) The relationship between muscle Wber composition and maximal anaerobic power and capacity. J Sports Med Phys Fitness 22:407–413 Marquardt JA, Bacharach DA, Kelly JM (1993) Comparison of power outputs generated during 20 and 30 s Wingate tests. Res Q Exerc Sport 64:A33–A34 Mastrangelo MA, Chaloupka EC, Kang J, Lacke CJ, Angelucci JA, Martz WP, Biren GB (2004) Predicting anaerobic capacity in 11– 13 year-old boys. J Strength Cond Res 18:72–76 Maud PJ, Schultz BB (1989) Norms for the Wingate anaerobic test with comparison to another similar test. Res Q Exerc Sport 60:144–151 Medbø JI, Tabata I (1993) Anaerobic energy release in working muscle during 30 s to 3 min of exhausting bicycling. J Appl Physiol 75:1654–1660 Murphy MM, Patton JF, Frederick FA (1986) Comparative anaerobic power of men and women. Aviat Space Environ Med 57:636–641 Smith JC, Hill DW (1991) Contribution of energy systems during a Wingate power test. Br J Sports Med 25:196–199 Ulmer HV (1996) Concept of an extracellular regulation of muscular metabolic rate during heavy exercise in humans by psychophysiological feedback. Experientia 52:416–420 Van Someren KA, Palmer GS (2003) Prediction of 200-m sprint kayaking performance. Can J Appl Physiol 28:505–517 Vanderford ML, Meyers MC, Skelly WA, Stewart CC, Hamilton KL (2004) Physiological and sport-speciWc skill response of Olympic youth soccer athletes. J Strength Cond Res 18:334–342 Vandewalle H, Heller J, Pérès G, Raveneau S, Monod H (1987a) Etude comparative entre le Wingate test et un test force-vitesse sur egocycle. Sci Sports 2:279–284 Vandewalle H, Pérès G, Monod H (1987b) Standard anaerobic exercise tests. Sports Med 4:268–289 Vincent S, Berthon P, Zouhal H, Moussa E, Catheline M, Betue-Ferrer D, Gratas-Delamarche A (2004) Plasma glucose, insulin and catecholamine responses to a Wingate test in physically active women and men. Eur J Appl Physiol 91:15–21

reliability values previously reported with respect to power outputs observed between two separate 30-WATs (Bar-Or 1987; Kaczkowski et al. 1982). These Wndings suggest that administering a 20-WAT is still advantageous if (a) the derived regression equation is employed to determine relative power outputs of the Wnal 10 s of a WAT to maintain the validity typically associated with the original protocol or (b) by acknowledging the percent diVerences that exist between the protocols reported in this study. Physical response Detrimental physical responses and subsequent subject apprehension have been reported to occur both during and after the 30-WAT, including nausea, dizziness, headaches, and vomiting resulting in less than optimal compliance with the 30-WAT (Jacobs et al. 1982; Maud and Schultz 1989). Following the 30-WAT in this study physical discomfort was observed as three subjects vomited, and approximately 25% subjects reported nausea, light-headedness, headaches, and/ or leg fatigue, despite a 4-min cool-down and provision of a high-carbohydrate drink upon request. While these eVects can only be reported subjectively, the only detrimental side eVect reported following the 20-WAT was leg fatigue. The minimal side eVects (aside from leg exhaustion) seem to indicate a reduced level of discomfort among subjects following the 20-WAT, which could increase subject compliance as well as the test–retest reliability of the protocol.

Conclusion The major Wnding from this study is that a 20-WAT can be considered a valid alternative to 30-WAT when used concomitantly with the prediction of the Wnal 10 s utilizing the regression equation derived in this study. The use of this abbreviated 20-WAT+10 protocol may reduce subject discomfort both during and following the test, thereby maximizing subject compliance and enhancing the applicability and repeatability of the WAT across a variety of populations.

Allen DG, Westerblad JA, Lee JA, Lannergren J (1992) Role of excitation–contraction coupling in muscle fatigue. Sports Med 13:116–126 Altman DG, Bland JM (1983) Measurement in medicine: the analysis of method comparison studies. Statistician 32:307–317 American College of Sports Medicine (1997) Policy statement regarding the use of human subjects and informed consent. Med Sci Sports Exerc 29:5 Ansley L, Robson PJ, Gibson A, Noakes TD (2004) Anticipatory pacing strategies during supramaximal exercise lasting longer than 30 s. Med Sci Sports Exerc 36:309–314


Sign up to vote on this title
UsefulNot useful