Translation of Exercise Testing to Exercise Prescription Using the Talk Test : The Journal of Strength & Conditioning Research

Secondary Logo

Journal Logo

Original Research

Translation of Exercise Testing to Exercise Prescription Using the Talk Test

Jeans, Elizabeth A; Foster, Carl; Porcari, John P; Gibson, Mark; Doberstein, Scott

Author Information
Journal of Strength and Conditioning Research: March 2011 - Volume 25 - Issue 3 - p 590-596
doi: 10.1519/JSC.0b013e318207ed53
  • Free
  • Press Release



Exercise intensity for endurance training is often defined in terms of a target heart rate (HR), expressed relative to maximal HR (%HRmax or %HR reserve) or relative exercise capacity (%maximal metabolic equivalents [METs] or %maximal MET reserve) (2). However, the range of recommended intensities is very wide, ranging from very easy to very hard exercise. Additionally, the ‘relative percent’ concept of exercise prescription may lack individual precision and has been the subject of challenge for many years (22). A considerable body of recent evidence in athletes suggests that training prescriptions are better defined in terms of the exercise intensity relative to physiologic thresholds defined by the ventilatory threshold or blood lactate accumulation (1,3,4,9-11,19-21,28) than in terms of %O2max or %HRmax. Additionally, both participants and trainers are often as interested in the absolute dimensions of the exercise prescription (e.g., running 7:00 per mile pace may be appropriate as background training intensity for 1 person, whereas 7:30 per mile pace might be appropriate for another) as in the target HR value itself. This is particularly true for those who guide the conditioning program of athletes, where the goals of training may be either to construct a primary training program (e.g., as in runners) or to construct a supplemental conditioning program (e.g., as in team sport athletes to supplement their sport specific training). Prescription of intensity may be a particularly sensitive issue, because athletes apparently have a maximal amount of high-intensity (e.g., greater than the ventilatory threshold) training that they can tolerate (27,28). Additionally, there is evidence that athletes may not be particularly adept at executing the training plans designed by coaches (12), highlighting the need for simple and practical approaches to guiding exercise training intensity. Early work from our laboratory demonstrated that maximal exercise test-derived parameters could be ‘‘translated’’ into absolute exercise intensities that could provoke a given HR response during training (13,16,17), at least for ordinary fitness prescriptions. However, this approach requires a maximal exercise test, which is not always logistically possible in fitness programs or desirable in athletes (who may want to limit the amount of ‘race intensity’ or ‘game intensity’ training they do). Thus, there is a need for a systematic method of translating submaximal exercise test responses into training prescriptions.

Laboratory techniques for guiding training based on submaximal exercise responses using either ventilatory or blood lactate responses are well accepted and of established value. However, the time and resources to provide this kind of technologically intensive guidance is not widely available for the less than elite athlete. Over the last decade, the Talk Test (TT) has emerged as a technique applicable to submaximal exercise that is a reasonable surrogate of the ventilatory threshold (VT) and respiratory compensation threshold (RCT) in a variety of populations, including athletes (25), healthy nonathletes (7,24), patients with chronic disease (30) and of the ischemic threshold in patients who develop exertional ischemia (6). The Talk Test technique has been shown to be sensitive enough to reflect changes in VT attributable to interventions such as blood donation and training (14). However, we have also shown that the HR response during steady-state or interval exercise training may be systematically higher (e.g., cardiovascular drift) than predicted from the Talk Test during incremental exercise (15). This is the same problem as when absolute training intensities are predicted from measurements of VT/RCT or from blood lactate measures obtained during incremental exercise. Because of the lag in responses of the detection method (talking or breathing pattern/blood lactate accumulation), the absolute exercise intensity that produces desired responses during training is often lower than that intensity during incremental exercise. In a recent report from our laboratory (15), we described a strategy for translating the Talk Test responses (e.g., accounting for cardiovascular drift) observed during an incremental exercise test into training intensities during 20-minute steady-state training sessions in sedentary young adults. These results suggested a strategy whereby the results of a submaximal exercise evaluation could be used to define an absolute training intensity below the VT, but without the technical requirements of measuring VT in the laboratory. If such a strategy could be validated in well-trained populations, then the process of defining training intensities for the conditioning of athletes would be meaningfully simplified and perhaps lead to better training prescriptions (1,2,28). Accordingly, we sought to repeat the basic experimental protocol of our earlier study (15), in a group of well-trained individuals to determine how to translate incremental exercise test results into appropriate exercise training advice based on Talk Test responses. Additionally, we sought to evaluate the reproducibility of Talk Test responses during incremental exercise testing.


Experimental Approach to the Problem

To test the hypothesis that exercise intensity would have to be downregulated by ∼10% to yield comparable results during a 40-minute training session vs. those observed during an incremental test, we evaluated the responses of healthy, well-trained subjects during 40-minute steady-state training bouts at different intensities. The intensities were defined based on Talk Test responses during an incremental exercise test.


The subjects were 14 healthy volunteers (Table 1). All subjects provided written informed consent, and the protocol was approved by the local ethics committee. All subjects were regular exercisers (minimum of 120 minutes running weekly). Although none of the subjects were intercollegiate level athletes, most routinely entered local road races. Typically, their performances for 10-km road races were in the range of 40-55 minutes. Thus, although we acknowledge that these subjects were not competitive athletes, they all trained regularly and systematically, approached their training in a goal-oriented manner (participating in occasional local road races), and all understood training within the context of threshold-based training zones (10,19,20,27,28), which we believe is the dominant approach to training in athletes, rather than in terms of %O2max or %HRmax that is more typical of the American College of Sports Medicine-based model widely used for fitness prescription (2).

Table 1:
Demographic characteristics of the subjects.*†


Each subject was studied during treadmill running in the laboratory on 6 different occasions, with at least 72 hours between tests. The subjects were asked to refrain from heavy exercise for 48 hours before testing and abstain from alcohol consumption for 24 hours. None of the subjects used tobacco. The first test in each subject (designed to define subject characteristics and to habituate the subjects to the laboratory) was an incremental exercise test to fatigue with measurement of respiratory gas exchange via open circuit spirometry (AEI Inc, Pittsburgh, PA, USA). The analyzers and volume turbine were calibrated with reference gases and room air, and a 3-L syringe, respectively. Gas analysis data were integrated over 30 seconds. No O2max verification procedures were used, and the highest 30-second value observed during the test was accepted as O2peak. Ventilatory and respiratory compensation thresholds were defined based on conventional criteria (7,14,15,24,25,30). Heart rate was measured using radiotelemetry (Polar Electo Oy, New York, NY, USA), and integrated over 5 seconds. The test began with a 3-minute walking warm-up. At this point, the treadmill velocity was set to 2.2 m·s−1 (5 miles·h−1) for 2 minutes, and incremented by 0.22 m·s−1 (0.5 miles·h−1) every 2 minutes until the subject indicated that they could no longer continue. The grade of the treadmill belt was constant at 1% throughout the test.

The rating of perceived exertion (RPE) was measured using the category ratio (0-10) scale (5). The second test was identical to the first, except that respiratory gas exchange was not measured. Instead, the subject was asked to recite the “Pledge of Allegiance” during the last 30 seconds of each exercise stage. Immediately after reciting this standard paragraph, the subject was asked ‘can you speak comfortably?’ Over the course of several studies (7,14,15,24,25,30) we have noted that at intensities below the VT, subjects quickly and definitely respond to the question with ‘yes.’ At an intensity that approximates the VT, the subject will typically equivocate their answer (e.g., ‘yes, but it' getting harder'). It is almost always clear to both subject and test administrator that the point of comfortable speech has passed. We have referred to this as the equivocal (EQ) stage in our previous studies. By definition, the stage before this is the last positive (LP) stage of the Talk Test. Later during the test, at approximately the intensity of the RCT, the subject will often fail to complete the paragraph, or respond to the question about comfortable speech with a definite ‘no,’ which we have referred to as the Negative stage of the Talk Test (NEG) (25). The speed at the last time the subjects reported that they could speak comfortably (LP), the speed at the first time they rated speech comfort as ‘EQ’ and the speed at the first time the subjects definitely could not speak comfortably (NEG) were recorded.

Tests 3-5 were 40-minute runs designed to be at an intensity that might be used in steady-state training, at intensities ranging from recovery to race tempo. After a warm-up period, which followed the incremental protocol until the Talk Test stage consistent with that stage, the speed remained constant for the remainder of the 40-minute exercise bout. Comments by the subjects after completion of the tests indicated that the NEG stage was just slower than their normal race pace for 10-km events. Indeed, many subjects had significant difficulty finishing the 40-minute duration of this stage. The ‘stage before the LP stage’ (LP-1), the ‘last positive’ (LP) and ‘EQ’ stages were selected based on recent results from our laboratory (15) and on evidence that Talk Test responses have both a delay and a drift over time (14). Relevant measures (HR, RPE, Talk Test) were made at 5-minute intervals throughout each exercise bout. To allow quantitative assessment during the 40-minute exercise bouts, Talk Test responses were coded as 1 = yes, speech is comfortable; 2 = EQ, speech is possible but not entirely comfortable; and 3 = no, speech is definitely not comfortable, as in our previous study (15). The order of presentation of the 3 exercise intensity bouts was random. The last laboratory visit was a repeat of the second session, an incremental exercise test to fatigue, without gas exchange, but with the Talk Test performed at the end of each 2-minute stage.

Statistical Analyses

Mean values for HR, RPE, and the Talk Test Score during the 40-minute exercise bouts were compared using repeated-measures ANOVA to test the hypothesis that the absolute intensity during an exercise training bout would have to be reduced by ∼10% (e.g., 1 exercise stage) to allow appropriate training intensities to be achieved. When justified by ANOVA, post hoc analyses were made using the Tukey honestly significant difference (HSD) procedure. Responses during the 2 incremental exercise tests were compared using the Pearson correlation coefficient. Statistical significance was accepted when p ≤ 0.05.


Mean values for the running speed, HR and RPE at the LP-1, LP, and EQ stages of the Talk Test and during the maximal incremental exercise are presented in Table 2. Responses of HR and %maximal HR during the 40-minute exercise bouts are presented in Figure 1. There were significant main effects for both exercise time and Talk Test classification. There was no significant interaction for exercise time by Talk Test classification. After 20 minutes of the exercise bout, the HR and %maximal HR were significantly greater during the EQ bout than during the LP-1 and LP bouts, and remained significantly greater for the remainder of the bout. With the exception of the EQ bout, the %maximal HR remained within the broad ACSM guidelines for exercise training (2). During all 3 exercise bouts, there was a significant increase in HR and %HRmax between 20 and 40 minutes.

Table 2:
Responses of the subjects during submaximal and maximal portions of the incremental exercise test.*†
Figure 1:
Mean responses of the heart rate (HR) (upper panel) and %HRmax (lower panel) during steady-state running bouts of 40-minute duration. The shaded zone in the lower panel is the %maximal heart rate (%HRmax) recommended by ACSM for exercise training of healthy individuals (2). The %HRmax during the last 20 minutes of the runs was significantly different during the equivocal (EQ) bout vs. the last positive stage of the talk test (LP-1) and last positive stage (LP) bouts.

Responses of RPE and the Talk Test score during the 40-minute exercise bouts are presented in Figure 2. For both RPE and the Talk Test score, there were significant main effects for exercise time and Talk Test classification as well as a significant interaction for exercise time by Talk Test classification. After 20 minutes, the RPE during the EQ stage was significantly greater than during the LP-1 or LP bouts and remained greater for the reminder of the bout. With the exception of the EQ exercise bout, the RPE remained within the broad ACSM guidelines for exercise training (2). After 20 minutes, the Talk Test score was significantly greater during the EQ bout than during the LP-1 or LP exercise bouts. Although there was a drift in the Talk Test score during all bouts, such that not all subjects were unequivocally comfortable speaking during either the LP-1 or LP stages, the magnitude of drift was small, and it would be fair to say that the subjects were generally speaking comfortably during the duration of both LP-1 and LP exercise bouts.

Figure 2:
Mean responses of rating of perceived exertion (RPE) (upper panel) and the Talk Test Score (lower panel) during steady-state running bouts of 40-minute duration. The shaded zone in the upper panel is the RPE recommended by ACSM for exercise training of healthy individuals (2). Both the Talk Test score and the RPE were significantly greater during the last 20 minutes of the equivocal (EQ) bout than during the last positive stage of the Talk Test (LP-1) and the last positive stage (LP) bouts.

The reproducibility of the exercise responses from the 2 incremental exercise tests are presented in Table 3 and Figure 3. The speed (r = 0.84), HR (r = 0.81) and RPE (r = 0.71) at stages matched for Talk Test responses during the 2 incremental exercise tests were generally well correlated. There were no significant differences between tests for the relevant outcome measures (Table 3). The slope of the regression line between tests for all 3 outcome measures approximated 1.0. Together, these responses support the concept that responses during the Talk Test are adequately reproducible.

Table 3:
Exercise responses during the 2 incremental exercise tests.*†
Figure 3:
Comparisons of the speed (upper panel), heart rate (middle panel) and rating of perceived exertion (lower panel) during repeated incremental exercise bouts matched for Talk Test responses (last positive stage [LP] and equivocal stage [EQ]).


The primary finding of this study was that the amount of downregulation of the absolute intensity compared to incremental exercise was less in this group of well-trained individuals than in our earlier report of a similar protocol with sedentary individuals (15). In that sense, our hypothesis regarding the magnitude of downregulation of exercise intensity is not supported. Exercise performed at the LP intensity observed during incremental exercise produced responses during a 40-minute steady-state exercise bout consistent with exercise intensity at or below the VT. Whether this in attributable to the well-trained subjects making a more accurate estimate of their responses during incremental exercise, or to a fundamental difference in the response to steady-state exercise is not evident from the present data, although a case can be made for either explanation.

Although fitness training has traditionally been organized in terms of %maximal HR or %maximal METs (2), recent evidence has suggested that the overall organization of a training program can be better understood, and prescribed, in terms of the percent of time above and below physiologic markers defined by the VT and LT (1,3,4,9-11,19-21,26-28). Within this context, there is evidence that the majority of training by many high level athletes is performed below the VT or the LT, which would approximate the LP stage of the Talk Test. Although it is clear that for competitive fitness to be developed, a certain amount of training must be performed at higher intensities (8,18,21,23,29) (>RCT or blood lactate concentration of 4 mmol·L−1), it seems that the upper limit of this high-intensity training approximates 10-20% of total weekly training time. Further, this intensity represents the reference value for training intensities that are likely to approximate O2max if continued for long enough. This intensity would be consistent with the NEG stage of the Talk Test. In a serious athlete, training 15+ h·wk−1, this would represent 1.5-3.0 h·wk−1 of high-intensity training, which is in basic agreement with the volume of training performed in studies focused on high-intensity training (21). Recent evidence in team sport athletes, suggests that, in contrast to endurance athletes, the requirement for sport-specific training requires that somewhat >20% of training is performed at a higher intensity (1). However, case evidence suggests that even this type of athlete profits from performing relatively more low intensity ‘background training’ (28). This finding suggests that either the base of endurance athletes' training or conditioning training in team sport athletes should be performed at relatively low intensity (e.g., <VT or <LT or ≥LP stage of the Talk Test.

Because extensive evidence suggests that the Talk Test appear to be strongly related to physiologic thresholds (7,14,24,25,30), the Talk Test may be a simple and reasonably accurate surrogate of directly measured values for either the VT/LT or RCT/4 mmol·L−1 when it is not possible or feasible to make direct threshold measurements. Our earlier results, using HR results to define absolute exercise training intensities relative to target HR in healthy individuals suggested that workloads needed to be downregulated to account for the physiologic ‘drift’ that often occurs during sustained submaximal exercise (13,16,17). The most recent evidence from our laboratory (15), performed in sedentary individuals, suggests that the magnitude of downregulation is ∼10%, or at least one stage based on exercise test responses. If the EQ stage is assumed to more or less approximate the VT (7,25,30), then the downregulation required to make sure that the exercise training intensity is below the VT was 2 stages (e.g., LP-1). In the present data, the subjects were able to stay within reasonable levels of steady-state exercise intensity even while running at the LP stage, probably secondary to the well-trained nature of the subjects. Nevertheless, if the goal of background training is to be definitely below the VT/LT, then it still seems prudent to recommend training at the LP-1 stage, to provide a margin of error. On the other hand, other studies from our laboratory have shown that during interval exercise at intensities above and below the VT, predicted using the Talk Test, that training exercise responses often lagged behind those predicted because it apparently takes ∼4 minutes to accommodate Talk Test responses (14). However, in the setting of steady-state submaximal exercise, such as that used in the present study, the tendency of exercise responses to ‘drift’ over time was clearly evident.

Practical Applications

If 1 of the goals of the coach or conditioning specialist is to suggest appropriate intensities for the background training of athletes, the present results suggest that a simple submaximal evaluation, based on Talk Test responses during incremental exercise can define the absolute speed necessary to provoke appropriate training intensities. The results suggest that the intensity associated with the LP stage of the Talk Test may be appropriate for steady-state training at an intensity below the VT/LT. However, given the substantial body of evidence suggesting that background training should be at intensities below the VT, it seems more prudent to suggest that the LP-1 stage from the incremental test would be more appropriate. This study extends our previous work with sedentary individuals into the realm of well-trained individuals. If this tool, provided by Talk Test responses, can be extended into the realm of serious endurance athletes, then it may provide a valuable tool even for the coaches of high-level athletes.


This study was supported by a grant from the Graduate Council of the University of Wisconsin-La Crosse.


1. Algroy, A, Hetelid, KJ, Seiler, S, and Stray Pederson, JI. Quantifying training intensity distribution in a group of Norwegian professional soccer players. Int J Sports Physiol Perf In press.
2. American College of Sports Medicine. ACSM's Guidelines for Exercise Testing and Prescription (8th ed.) Philadelphia, PA: Lippincott, Williams and Wilkins, 2010.
3. Billat, VL. Interval training for performance: A scientific and empirical practice. Special recommendations for middle and long distance running. Sports Med 31: 13-31, 2001.
4. Billat, VL, Lepretre, PM, Heugas, AM, Laurence, MH, Salim, D, and Koralsztein, JP. Training and bioenergetic characteristics in elite and female Kenyan runners. Med Sci Sports Exerc 35: 297-304, 2003.
5. Borg, GA. Borg's Perceived Exertion and Pain Scales. Champaign, IL: Human Kinetics, 1998.
6. Cannon, C, Foster, C, Porcari, JP, Skemp-Arlt, KM, Fater, DCW, and Backes, R. The Talk Test as a measure of exertional ischemia. Am J Med Sports 6: 52-56, 2004.
7. Dehart-Beverley, M, Foster, C, Porcari, J, Fater, DCW, and Mikat, RP. Relationship between the Talk Test and the ventilatory threshold. Clin Exerc Physiol 2: 34-38, 2000.
8. Driller, MW, Fell, JW, Gregory, JR, Shing, CM, and Williams, AD. The effects of high intensity interval training in well-trained rowers. Int J Sports Physiol Perf 4: 110-121, 2009.
9. Esteve-Lanao, J, Foster, C, Seiler, S, and Lucia, A. Impact of training intensity distribution on performance in elite athletes. J Strength Cond Res 21: 943-949, 2007.
10. Esteve-Lanao, J, San Juan, AF, Earnest, CP, Foster, C, and Lucia, A. How do endurance runners actually train? Relationship with competition performance. Med Sci Sports Exerc 37: 496-504, 2005.
11. Fiskerstrand, A and Seiler, KS. Training and performance characteristics among Norwegian international rowers 1970-2001. Scand J Med Sci Sports 14: 303-310, 2004.
12. Foster, C, Heimann, KM, Esten, PL, Brice, G, and Porcari, JP. Differences in perceptions of training by coaches and athletes. S Afr J Sports Med 8: 3-7, 2001.
13. Foster, C, Lemberger, K, Thompson, NN, Sennett, SM, Hare, J, Pollock, M, Pels, AE, and Schmidt, DH. Functional translation of exercise responses from graded exercise testing to exercise training. Am Heart J 112: 1309-1316, 1986.
14. Foster, C, Porcari, JP, Anderson, J, Paulson, M, Smaczny, D, Webber, H, Doberstein, S, and Udermann, B. The Talk Test as a marker of exercise training intensity. J Cardiopulm Rehabil 28: 24-30, 2008.
15. Foster, C, Porcari, JP, Gibson, M, Greany, J, Talati, N, and Recalde, P. Translation of submaximal exercise test responses to exercise prescription using the Talk Test. J Strength Cond Res 23: 2425-2429, 2009.
16. Foster, C and Thompson, NN. Functional translation of exercise responses to recreational activities. J Cardiopulm Rehabil 11: 373-377, 1991.
17. Foster, C, Thompson, NN, and Bales, S. Functional translation of exercise responses using combined arm-leg ergometry. Cardiology 78: 150-155, 1991.
18. Gorostigia, EM, Walter, CB, Foster, C, and Hickson, RC. Uniqueness of interval and continuous training at the same maintained exercise intensity. Eur J Appl Physiol 63: 101-107, 1991.
19. Guellich, A and Seiler, S. Lactate profile changes in relation to training characteristics in junior elite cyclists. Int J Sports Physiol Perf 5: 316-327, 2010.
20. Guellich, A, Seiler, S, and Emrich, E. Training methods and intensity distribution of young world-class rowers. Int J Sports Physiol Perf 4: 448-460, 2009.
21. Helgerud, J, Engen, LC, Wislof, U, and Hoff, J. Aerobic endurance training improves soccer performance. Med Sci Sports Exerc 33: 1925-1931, 2001.
22. Katch, V, Weltman, A, Sady, S, and Freedson, P. Validity of the relative percent concept for equating exercise intensity. Eur J Appl Physiol 39: 219-227, 1978.
23. Laursen, PB, Shung, CM, Peake, JM, Coombes, JS, and Jenkins, DG. Interval training program optimization in highly trained endurance cyclists. Med Sci Sports Exerc 34: 1801-1807, 2002.
24. Persinger, R, Foster, C, Gibson, M, Fater, DCW, and Porcari, JP. Consistency of the Talk Test for exercise prescription. Med Sci Sports Exerc 36: 34-38, 2004.
25. Recalde, PT, Foster, C, Skemp-Arlt, KM, and Porcari, JP. The Talk Test as a simple marker of ventilatory threshold. S Afr J Sports Med 8: 5-8, 2003.
26. Schumacher, YO and Muller, P. The 4000-m team pursuit cycling world record: Theoretical and practical aspects. Med Sci Sports Exerc 34: 1029-1036, 2002.
27. Seiler, S and Kjerland, GO. Quantifying training intensity distribution in elite endurance athletes: Is there evidence for an optimal distribution. Scand J Med Sci Sports 16: 49-56, 2006.
28. Seiler, S and Tonnessen, E. Intervals, thresholds and long slow distance: The role of intensity and duration in endurance training. Sports Sci 13: 32-54, 2009.
29. Stepto, NK, Hawley, JA, Dennis, SC, and Hopkins, WG. Effects of different interval training programs on cycling time trial performance. Med Sci Sports Exerc 31: 736-741, 1998.
30. Voelker, SA, Foster, C, Porcari, JP, Skemp-Arlt, KM, Brice, G, and Backes, R. Relationship between the Talk Test and ventilatory threshold in cardiac patients. Clin Exerc Physiol 4: 120-123, 2002.

exercise therapy; training intensity; %HRmax

© 2011 National Strength and Conditioning Association