Site hosted by Build your free website today!
SHN 1662: Anatomy and Physiology


Relationship between maximal oxygen uptake collected by the indirect multistage fitness test and the direct gas analysis method.

By Richard Gardner

Purpose: the purpose of this assignment is to discover whether or not a significant relationship exists between an indirect and direct maximal oxygen uptake test. Results: two experiments took place, an indirect 20-metre multistage fitness test (VO2max 46.7 ± 8.7) (mean ± standard deviation) and a laboratory testing of direct gas analysis using a Monark cycle ergometer (VO2max 52.0 ± 8.8). 11 participants took part in the experiments (8 males and 3 females of mixed sporting population). A Pearson's correlation test was then used to establish that a strong positive correlation existed between the two experiments (r = 0.83, n = 11, one tailed r > 0.05). Final readings were also taken of maxHR (maximum heart rate) following the experiments, (multistage 186.1 ± 13.7, direct gas 184.0 ± 9.7). Conclusion: These results signify that the multistage fitness test can be used as a reliable source for testing VO2max. The tests also concluded that the mean maxHR scores between the two experiments were virtually identical.

To find out if a statistically significant relationship exists between the multistage fitness test and the direct gas analysis method. To assess the accuracy any relationship and to determine if the multistage fitness test can be used as a valid means for measuring VO2max.

Literature review
Definition of terms: Endurance sports - sports whereby a consistent level of aerobic activity is achieved. Indirect VO2max test - test to analyse maximal oxygen uptake without directly testing oxygen. Direct VO2max test - directly testing oxygen to measure maximal oxygen uptake. Multistage fitness test (also called bleep/shuttle test) - an indirect test to evaluate VO2max using a 20-metre track and audiotape. Direct gas analysis method - a direct VO2max test, using Douglas bags and Servomex Gas analyser. Pearson's correlation coefficient test - statistical analysis to determine whether or not a relationship exists between two variables. r value - the correlation coefficient, signifying the strength (if any) of a relationship raging from -0.1 (strong negative relationship) through zero (no relationship) to +0.1 (strong positive relationship). Directional hypothesis - prediction that predicts there will be a relationship between two variables. Null hypothesis - prediction that predicts there will not be a relationship between two variables.

Previous research has been done into the maximal oxygen uptake testing with some interesting results. Paliczka performed an experiment to assess the reliability of the multistage fitness test as a field test of cardio-respiratory endurance (Paliczka et al. 1987). The results on this occasion provided high correlations using Pearsons coefficient, (n = 9, r = 0.93) proving that the multistage fitness test can be used to test maximal oxygen uptake. Grant et al. (1995) also tested the multistage fitness test (along with other indirect tests) against the direct measurement of VO2max, performed on a treadmill. This time the correlation was calculated (n = 22, r = 0.86) and although slightly lower still shows a strong positive correlation between the indirect and direct method. After conducting these tests however, Grant et al. concluded that the cooper walk/run test was the most accurate indirect method of testing VO2max. Clauzon (2002) also discovered a significant correlation testing the multistage fitness test against direct gas analysis using a treadmill, (n = 16, r = 0.931). Clauzon also commented however, that the mean scores between the tests were 23% lower for the multistage fitness test (multistage 36.675, treadmill 47.5975) and went on to signify that participants executing the multistage fitness test may not reach their maximal oxygen uptake before the test is terminated.

Boddington et al. (2001) advocated a study to determine the reliability of a modified 5-metre multistage shuttle test monitor the 'match-related' fitness of female hockey players. The tests took place on 4 occasions within 4 weeks and found that the total distance scores were not significantly different (n = 23, p = 0.99). This concluded that the modified 5-metre multistage shuttle test could be employed to assess fitness changes throughout a season.

Larsen et al. (2002) also investigated a different indirect maximal oxygen uptake test. In this test, participants were asked to travel a distance of 1.5 miles with a 'somewhat hard' exercise intensity (around 13), the VO2max was then calculated using the formula (VO2 max = 65.404 + 7.707 x gender (1 = male, 0 = female) - 0.159 x body mass (kg) - elapsed exercise time (minutes). This test also provided acceptable validity (n = 101, r = 0.86) and the authors concluded the test accurately predicted maximal oxygen uptake. In a similar experiment, George et al. (2000) also got a similar validity level (n = 156, r = 0.88), this time though the validation results were much higher for women (n = 76, r = 0.93) than for men (n = 80, r = 0.74). This could mean that testing VO2max using formulas in this way is not as accurate as first appears.

Following this research it has been decided that the following two hypothesis will be employed:

Experimental hypothesis "There is a statistically significant relationship between the indirect VO2max test (multistage fitness test) and the direct VO2max test (direct gas analysis)."

Null hypothesis "There is no statistically significant relationship between the indirect VO2max test (multistage fitness test) and the direct VO2max test (direct gas analysis)."


The test is principally designed to ascertain whether or not a relationship exists between maximal oxygen uptake readings taken after a multistage fitness test and a direct gas analysis.

11 participants took part in the tests, 3 female 8 male, and of mixed sporting population.


For multistage fitness test: Audiocassette supplied by NCF, cassette player, tape measure, 8 marker cones, flat, non-slippery floor 20 metres in length.

For direct gas analysis method: Weighing scales, Monark cycle ergometer, Douglas bag rack (evacuated), Harvard dry gas meter, breathing tube, mouthpiece, noseclip, Polar heart rate monitor, stopwatch, Servomex gas analyser.


For multistage fitness test: 20 metres were measured using the tape measure and marker cones. The audiocassette was rewound to the start. The tape started with two bleeps which were timed to a one minute interval to make sure that the tape was not stretched. Each participants resting heart rate was taken. The test started with the tape emitting a five-second countdown. Thereafter, the tape emitted a single beep at regular intervals with all participants covering the distance from the start to the marker cone and back, completing a shuttle, before the next beep sounded.

Participants jogged between the start and marker cones at increasingly quicker levels. Starting at level one, the levels gradually became quicker with a single beep marking the end of each shuttle and a triple beep marking the start of a new level. At the end of each shuttle, participants turned round and placed one foot either on or behind the 20-metre mark. If a participant arrived at the mark before the bleep emitted then they stopped jogging and awaited the beep. Each participant ran for as long as possible, until they could no longer keep up with the speed set by the tape. They then withdrew from the test.

Observers made a note of the level and number of shuttles into the level, at which each participant withdrew from the test. An estimate of maximum oxygen uptake was obtained from the booklet supplied with the audiocassette and the heart rate was taken again (HRmax).

For direct gas analysis method: Participants were weighed in kg on the scales. Each Douglas bag was evacuated using the Harvard dry gas meter and then a breathing tube and mouthpiece were attached. The saddle height of the Monark cycle ergometer was adjusted to suit the participant and a noseclip was attached. Following this, the rest condition was taken. This involved the participant sitting on the bike and breathing through the respiratory apparatus for a period of 4 minutes. During the 3rd-4th minute, a valve to the 1st Douglas bag was opened and air collected for 60 seconds.

After 60 seconds, the valve was closed and a chemical analysis was taken. This was done by attaching the Douglas bag to the Servomex Gas analyser. The analyser was allowed to analyse air for 60 seconds and then the O2 (%FeO2) and CO2 (%FeCO2) values were noted. Over the one-minute period, the analyser drew 2 litres of air from the Douglas bag. Following this, the Douglas bag was attached to the Harvard dry gas meter (which had been reset to zero) and the meter drew all of the air from the bag. The volume in litres was then recorded (minus the 2 litres previously consumed by the Servomex Gas analyser).

The exercise condition was performed using the correct resistance (1.5kg for females and 2.0kg for males) and with the participant cycling at a cadence of 60rpm (revs per minute). Gas analysis (chemical and volume) was taken in identically the same way as for the rest condition.

Once all results had been noted in a data collection table, the oxygen consumption and carbon dioxide production during exercise was calculated.

Results given are based on a group mean ± standard deviation. Statistical correlations were calculated using a Pearson's coefficient test using r to express the final level of correlation between -0.1 and 0.1. Table 1 demonstrates the performance results for both experiments, table 2 demonstrates the statistical analysis and figure 1 shows the raw data, including a 'line of best-fit'.

Multistage Fitness Test Direct gas analysis method
Final VO2max ( Final heart rate (bpm) Final VO2max ( Final heart rate (bpm)
46.7 ± 8.7 186.1 ± 13.7 52.0 ± 8.8 184.0 ± 9.7

Table 1: Performance results (n = 11, mean ± SD)

Number (n) Number of tails Degree of accuracy r value
11 1 r < 0.05 0.83

Table 2: Statistical results

Figure 1: Summary of raw data


The findings from this experiment seem to follow similar previous experimental research on indirect and direct maximal oxygen uptake. Paliczka et al. (1987) found that mean VO2max scores were relatively high (at 59 stating that the average for untrained groups was more likely to be around (45 The mean readings taken from the direct gas analysis are also higher than average (52, but the readings taken from the indirect multistage fitness test are much closer to the average (46.7

Although the population of the participants are of mixed sporting origin, all are sports students and as such a good proportion probably take part in at least some moderate intense sporting activity each week. This could account for the higher than average figures using the direct gas analysis. The differences in mean scores could be attributed to participants being better adapted to cycling than to running. An experiment conducted by Gibson et al. (1998) cited in Boddington et al. (2000) found that maximum oxygen consumption during a 20-metre shuttle test was underestimated "in runners but not in squash players". This is as the muscles become stronger in specific areas when performing repeated exercises; over time this lessens the amount of oxygen required to perform those exercises (Brooks 1997).

Mean values show a difference of just over 5.3 between the multistage fitness test and direct gas analysis. Although this does not seem much it is over 10% (46.7 is 89.70% of 52) and suggests on first sight that a relationship may not be strong between the two methods. This percentage though is much higher (13%) than the mean differences that Clauzon (2002) encountered between indirect and direct testing (23% between indirect, 36.675 and direct 47.5975). With the mean differences higher this also suggests that the participants level of fitness was higher than in Clauzon's experiments whom all led a relatively sedentary existence.

The statistical correlation between the multistage fitness test and the direct gas analysis method was calculated using the Pearson correlation coefficient. The correlation was statistically significant (r = 0.83, one tailed, n = 11, r > 0.05). Therefore, the experimental hypothesis (that there is a relationship between direct and indirect methods) has been supported and the null hypothesis (that there is no relationship between direct and indirect methods) has been rejected. Although this correlation is slightly lower than previous research (Paliczk, r = 0.93; Clauzon, r = 0.931), it is statistically significant nonetheless. Since Paliczk and Clauzon both used running to test direct and indirect VO2max, the difference of 0.10 could be attributed to participants' adaptation to cycling as already mentioned.

Grant though, (Grant et al. 1995) obtained results similar from testing the indirect shuttle test against direct VO2max taken from a treadmill (r = 0.86, a difference of just 0.03). Although the participants in this experiment were mainly all involved in endurance based sports, it is also stated that they have "differing levels of ability and training status" (Grant et al. 1995, p148). Given that, as already stated, the participants in this experiment in most likelihood have similar differences in ability and training status, the similarities with the Grant participants are obvious. The reasons therefore for a lower correlation could lie with training status rather than the adaptation to cycling as was previously suggested. Adaptations to training (specifically running) lead to vast increases in VO2max capacity, especially when tested through regularly practiced movements (Wilmore et al. 1999, Howley et al. 1997). Taking this into account, it could be that the higher correlations are obtained by Paliczk and Clauzon through both sets of participants having similar adaptations to running (several in Paliczk's tests were training for competitive endurance events and Clauzon's participants were mostly sedentary). Considering the variety of sports available in higher education it would seem likely that participants in this and the Grant experiment would have clear different adaptations to running. For instance, a person regularly playing basketball would adapt better to the multistage fitness test than a goalkeeper in soccer who does little or no running. The differences in correlation could relate to this and as such it may be more useful to examine athletes with similar adaptations to running if performing this analysis in future.

Being that the correlation is significant it would seem that using the multistage fitness test is a good indicator of maximal oxygen uptake. Some obvious advantages from using this method are: it is relatively cheap (using minimal equipment, compared to the expensive equipment used in direct gas analysis), a large number of participants can be tested simultaneously and so it is less time-consuming that direct methods, and the test is also relatively safe (due to the increase in speed, the heart rate increases steadily). These advantages would mean the test would be ideal when regularly testing a group of individuals fitness (for instance assessing the fitness of an amateur team throughout a season). The test is also ideal for students researching maximal oxygen uptake.

The results from mean final heart rates taken are much closer than the mean scores taken from VO2max with a difference of just over 2 (1.12%). Standard deviations calculated on final heart rates are further apart than for the final VO2max calculations with the multistage fitness test having an approximately 4 higher figure than the direct gas analysis. This again could be suggestive of either adaptations to cycling or running.


From the experiments undertaken, a significant relationship exists between VO2max readings taken from the indirect, multistage fitness test and the direct analysis. The high correlation between results suggests the multistage fitness test is a valid means for measuring VO2max.

Word count 2,607


Boddington, M.K., Lambert, M. I. St Clair Gibson, A. and Noakes, T.D. (2001) Reliability of a 5-m multiple shuttle test Journal of sports sciences 2001, Vol. 19, pp223-228.

Brookes, D. S. (1998) Program design for personal trainers Champaign: Human Kinetics.

Clauzon. (2002). A regression equation to predict true VO2max using a field test (20 meter shuttle run), [Internet site] URL: {accessed 05/02/2002}

George, J. D., Vehrs, P. R., Babcock, G. J., Etchie, M. P., Chinevere, D. T. and Fellingham, G. W. (2000) A modified submaximal cycle ergometer test designed to predict treadmill VO2max Measurement in physical education and exercise science Vol. 4, No. 4, pp229-243.

Grant, S., Corbett, K., Amjad, A.M., Wilson, J., and Aitchison, T. A. (1995) Comparison of methods of predicting maximum oxygen uptake British journal of sports medicine Vol. 29, No. 3, pp147-152.

Howley, E.T. and Franks, B.D. (1997). Health fitness instructors handbook (3rd Edition). Champaign: Human Kinetics.

Larsen, G. E., James, D. G., Alexander, J.l., Felingham, G. W., Aldana, G. S. and Parcell, A. C. (2002) Prediction of maximum oxygen consumption from walking, running and jogging Research quarterly for exercise and sport Vol. 73, No. 1, pp66-72.

Paliczka, V. J., Nichols, A.K. and Boreham, C.A.G. (1987) A multi-stage shuttle run as a predictor of running performance and maximal oxygen uptake in adults British journal of sports medicine Vol. 21, No 4, pp163-165.

Wilmore, J. H. and Costill, D. L. (1999) Physiology of sport and exercise (2nd edition) Champaign: Human Kinetics.

APPENDIX 1 - Raw data

Participant Gender Multistage Fitness Test Method Direct gas analysis method
Final VO2 max ( Final heart rate (bpm) Final VO2 max ( Final heart rate (bpm)
1 F 37.2 191 41 195
2 F 40.1 190 48 167
3 F 37.8 164 43 177
4 M 64 190 69 186
5 M 48 185 44 177
6 M 54.2 194 58 194
7 M 43.9 198 54 177
8 M 45.6 177 48 183
9 M 58 195 62 180
10 M 40.7 203 57 200
11 M 43.6 160 48 188