The Åstrand-Ryhming test/method under the magnifying glass A review of research articles
Dan Andersson
IDROTTSHÖGSKOLAN I STOCKHOLM D-uppsats i Idrott 2004
D-UPPSATS I IDROTT 61-80 POÄNG (10 POÄNG) VID MAGISTERUTBILDNING I IDROTT 2004 PÅ IDROTTSHÖGSKOLAN I STOCKHOLM
The Åstrand-Ryhming test/method under the magnifying glass A review of research articles
Dan Andersson
Handledare: Peter G. Schantz
Abstract
Aim: Validity studies of the Åstrand-Ryhming test/method are somewhat contradictory and the need of a review of the accumulated results is urgent. The aim of this study was to interpret and evaluate scientific results about the internal and external validity as well as the reproducibility of the Åstrand-Ryhming test (1954) and the Åstrand-Ryhming test (1960) and, if motivated, suggest presumptive modifications.
In accordance with the aim the following questions were formulated: (1) Is the test valid? (2) For whom is it valid? (3) Is it reproducible? (4) For whom is it reproducible? (5) Is there a scientific base for suggesting modifications of the test?
Method: Initially the characteristics and bases of the two tests/methods were sorted out. Subsequent to that a search for appropriate articles was conducted. The search resulted in 14 articles that had a title including Åstrand or Åstrand-Ryhming. Each article was evaluated and scrutinized. Uncertainty or lack of key information reduced the credibility of the article.
Results and conclusion: Based on the limited number of articles that were a part of this review it is not possible to decide whether or not the Åstrand-Ryhming test/method is valid. Consequently it is not possible to decide for which population the test is valid. It is additionally not possible to decide if the Åstrand-Ryhming test/method is reproducible or for whom the test is reproducible.
It is most likely possible to improve the accuracy of the estimations by means of modifying the Åstrand-Ryhming test/method. Suggestions of modifications include: (1) An elimination of the maximal treadmill test data (which is integrated in the 1960 nomogram), (2) A recommendation of new guidelines for the selection of workload, (3) That each test subject, prior to the initial test, conducts a pre-test, in order to get acquainted with the laboratory environment, the test equipment, and the testing procedures. (4) The introduction of a correction factor for obese individuals and (5) The construction of (a) new age correction factors and (b) separate age correction factors for the two sexes.
2
Contents
Abstract
Page 2
Record of figures and tables
Page 4
1. Introduction
Page 5
2. Aim and questions
Page 7
3. Method
Page 8
4. Background
Page 13
4.1. Characteristics of the Åstrand-Ryhming test (1954)
Page 13
4.1.1. Construction of the nomogram
Page 14
4.1.2. Material and methods
Page 16
4.1.3. Validity
Page 18
4.2. Characteristics of the Åstrand-Ryhming test (1960)
Page 21
4.2.1. Material and methods
Page 21
4.2.2. Adjustment of the original nomogram
Page 24
4.2.3. Correction factor for age
Page 25
4.2.4. Accuracy of prediction
Page 25
5. Results
Page 29
5.1.The articles
Page 29
5.1.1. An overview of the articles
Page 29
5.1.2. Group 1, with focus on validity
Page 32
5.1.3. Group 2, with focus on modification
Page 46
6. Discussion
Page 54
6.1. Question 1: Is the test valid?
Page 56
6.2. Question 2: For whom is it valid?
Page 62
6.3. Question 3: Is it reproducible?
Page 62
6.4. Question 4: For whom is it reproducible?
Page 64
6.5. Question 5: Is there a scientific base for suggesting modifications of the test? Page 64 7. Conclusion
Page 70
References
Page 71
3
Record of figures and tables Record of figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17
The 1954 nomogram…(Åstrand & Ryhming, 1954) Page 13 The relationship between pulse rate… (Åstrand & Ryhming, 1954) Page 14 Mechanical efficiency when cycling… (Åstrand & Ryhming, 1954) Page 15 Maximal oxygen intake calculated…(Åstrand & Ryhming, 1954) Page 17 Heart rates in relation to oxygen uptake…(Åstrand, P.-O., 1952) Page 17 The 1960 nomogram…(Åstrand, I., 1960) Page 22 Correction factor for the nomogram…(Åstrand, I., 1960) Page 26 Deviation in % between predicted and…(Åstrand, I., 1960) Page 28 Individual predictions of maximal oxygen…(Jessup et al, 1974) Page 36 Comparison between direct determined… (Teräslinna et al, 1966) Page 41 A plot of the differences of the estimated…(Wisén et al, 1995) Page 45 Maximal oxygen intake estimated with the…(Wisén et al, 1995) Page 45 The Legge-Banister nomogram…(Legge and Banister, 1986) Page 48 Maximal oxygen intake calculated… (Legge and Banister, 1986) Page 50 Nomogram for workload selection…(Terry et al, 1977) Page 53 Figure for ranking research articles about the Åstrand-Ryhming… Page 57 Figure for ranking the Åstrand-Ryhming test/method. Page 62
Record of tables Table 1 Table 2 Table 3 Table 4A Table 4B Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 11 Table 12 Table 13 Table 14
Error of methods when maximal… (Åstrand & Ryhming, 1954) Page 18 Average values of maximal heart rate…(Åstrand, P.-O., 1952) Page 20 Number of subjects in the different…(Åstrand, I., 1960) Page 23 An overview of the articles, questions 1-9 Page 30 An overview of the articles, questions 10-18 Page 31 Correlation coefficients, standard error…(Cink et al, 1981) Page 33 Comparison of predicted and measured…(Jessup et al, 1977) Page 34 Comparison among the predicted maximal…(Jetté et al, 1979) Page 37 Åstrand, Sjöstrand and directly measured maximal…(Kasch, 1984) Page 38 Raw values for the observed maximal oxygen…(Macsween, 2001) Page 40 Data from the 31 test subjects…(Teräslinna et al, 1966) Page 41 Means and standard deviations for estimated…(Williams, 1975) Page 43 Estimated reliabilities of additional test days…(Williams, 1975) Page 43 Correlation coefficients and standard error…(Siconolfi et al, 1982) Page 51 A comparison between the two nomograms Page 55
4
1. Introduction
Knowledge about an individual’s capacity for cardiovascular work is of interest, e.g., when selecting people for special tasks in the military service or in heavy industrial work. It is also of great interest for coaches, athletes and exercise scientists. An individual’s capacity for physical work is dependent mainly upon the supply of oxygen to the working muscles. In order to determine the aerobic capacity you will have to conduct a test, maximally or submaximally.
Measuring VO2 max directly demands sophisticated equipment, highly skilled technicians, time and money. It is also, for a large number of subjects, associated with discomfort and/or pain. “For these reasons it is desirable to have a valid substitute for assessing VO2 max that could be used by exercise and fitness programs in industry, hospitals, universities, schools, YMCAs, clubs, and recreation departments.”1
The Åstrand-Ryhming test/method is an example of such a substitute. The procedure is simple and no sophisticated laboratory equipment is needed. The test is widely used and several researchers have evaluated and scrutinized the test. The test does not require an all out effort, and is for the majority of individuals not associated with discomfort or physical pain.
However, validity studies are somewhat contradictory and the need of a review, with an interpretation of the accumulated results, is urgent. Such a review will shed new light over the test and have potential impact on both medical care and the physiological testing of the highperformance athlete. For this reason I have chosen to position the Åstrand-Ryhming test/method under the magnifying glass.
Initially it is important to sort out the characteristics and bases of the two tests; the Åstrand Ryhming test (1954) and the Åstrand-Ryhming test (1960). For a general discussion about the test and further information about advantages, disadvantages and validity see Åstrand & Rodahl (1986) or Legge and Banister (1986). 2,3 1
Kasch, F.W. (1984) “The validity of the Åstrand and Sjöstrand submaximal tests.” The Physician and Sports
medicine, 54, page 47. 2
Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd
Edition. New York: McGraw-Hill Book Company, page 369-370.
5
When approaching scientific literature about “The test” one quickly realizes that the tests from 1954 and from 1960 have a wide variety of names such as the Åstrand test, the ÅstrandRyhming test and the Å-R test. In order to be distinct and to differentiate between them the following definitions concerning the tests/methods will be used in this essay.4 •
The Åstrand-Ryhming test (1954) = the test with the accompanying nomogram described by P.-O. Åstrand, and I. Ryhming in the article from 1954. 5
•
The Åstrand-Ryhming test (1960) = the test with the accompanying nomogram described by I. Åstrand in her thesis from 1960.6
•
The Åstrand-Ryhming test/method = if it is uncertain which of the test/nomogram that has been investigated or when writing about the test in general.
3
Legge B.J., Banister E.W. (1986) “The Åstrand-Ryhming nomogram revisited.” J Appl Physiol, 61 (3), page
1203-1209. 4
P.-O. Åstrand and I. Ryhming married each other in the 1950´s. As a consequence I. Ryhming became I.
Åstrand. 5
Åstrand, P.-O. & Ryhming, I. (1954) “A nomogram for calculation of aerobic capacity (physical fitness) from
pulse rate during submaximal work.” J Appl Physiol, 7, page 218-221. 6
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta Physiologica
Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960.
6
2. Aim and questions
Aim
To interpret and evaluate scientific results about the internal and external validity as well as the reproducibility of the Åstrand-Ryhming test (1954) and the Åstrand-Ryhming test (1960) and, if motivated, suggest presumptive modifications.
Questions
1. Is the test valid? 2. For whom is it valid? 3. Is it reproducible? 4. For whom is it reproducible? 5. Is there a scientific base for suggesting modifications of the test?
7
3. Method
This essay is a review of research articles. To find relevant articles one has to conduct a search. It is important to be aware of that another selection method is likely to have resulted in the sorting out of different research articles and possibly, dissimilar or conflicting answers. Due to a large number of hits a restriction was made, see below. The selection method was conducted in the following manner:
Step 1
A search was made on PubMed (http://www.ncbi.nlm.nih.gov/PubMed/) on the following words:
astrand ergometer cycle test = 11 hits. astrand ergometer bicycle test = 10 hits. astrand test = 90 hits. astrand ryhming nomogram = 10 hits. astrand rhyming nomogram = 4 hits.
In total: 125 hits.
The following results were excluded due to the large amount of hits:
ergometer bicycle test = 1475 hits. ergometer cycle test = 1558 hits. bicycle ergometer = 2480 hits. cycle ergometry = 2185 hits.
In total: 7698 hits.
Step 2
Was a closer examination together with Dr Schantz and resulted in a sorting out of 45 relevant articles (out of the abovementioned 125 hits). 8
Step 3
Which of these 45 articles has a title including Åstrand or Åstrand-Ryhming? The answer is 15. However one of them, Pollock M.L., Linnerud A.C. (1971) “Observations of the ÅstrandRhyming nomogram as related to the evaluation of training.” Am Correct Ther J, 25 (6), page162-165 was excluded due to the aim. Consequently 14 articles remained.
1. Cink R.E., Thomas T.R. (1981) “Validity of the Åstrand-Ryhming nomogram for predicting maximal oxygen intake.” Br J Sports Med, 15 (3), page 182-185.
2. Cullinane E.M., Siconolfi S., Carleton R.A., Thompson P.D. (1988) “Modification of the Åstrand-Ryhming submaximal bicycle test for estimating VO2 max of inactive men and women.” Med Sci Sports Exerc, 20 (3), page 317-318.
3. Jessup G.T., Riggs C.E., Lambert J., Miller W.D. (1977) “The effect of pedalling speed on the validity of the Åstrand-Ryhming aerobic work capacity test.” J Sports Med, 17 (4), page 367-371.
4. Jessup G.T., Terry J.W., Landiss C.W. (1975) “Prediction of workload for the Åstrand-Ryhming test using stepwise multiple linear regression.” J Sports Med, 15 (1), page 37-42.
5. Jessup G.T., Tolson H, Terry J.W. (1974) “Prediction of maximal oxygen intake from Åstrand-Ryhming test, 12-minute run, and anthropometric variables using stepwise multiple regression. Am J Phys Med, 53(4), page 200-207.
6. Jette M. (1979) “A comparison between predicted VO2 max from the Åstrand procedure and the Canadian Home Fitness Test.” Can J Appl Sport Sci, 4 (3), page 214-218.
7. Kasch, F.W. (1984) “The validity of the Åstrand and Sjöstrand submaximal tests.” The Physician and Sports medicine, page 47-51, 54.
9
8. Legge B.J., Banister E.W. (1986) “The Åstrand-Ryhming nomogram revisited.” J Appl Physiol, 61(3), page 1203-1209.
9. Macsween A. (2001) “The reliability and validity of the Åstrand nomogram and linear extrapolation for deriving VO2 max from submaximal exercise data.” J Sports Med, 41 (3), page 312-317.
10. Siconolfi S.F., Cullinane E.M., Carleton R.A., Thompson P.D. (1982) ”Assessing VO2 max in epidemiological studies: a modification of the Åstrand-Ryhming test.” Med Sci Sports Exerc, 14 (5), page 335-338.
11. Teräslinna P., Ismail A.H., MacLeod D.F. (1966) ”Nomogram by Åstrand and Ryhming as a predictor of maximum oxygen intake.” J Appl Physiol, 21(2), page 513515.
12. Terry J.W., Tolson H., Johnson D.J., Jessup G.T. (1977) “A workload selection procedure for the Åstrand-Ryhming test.” J Sports Med, 17 (4), page 361-36.
13. Williams, L. (1975) “Reliability of predicting maximal oxygen intake using the Åstrand-Ryhming nomogram.” Res Quart, 46, page 12-16.
14. Wisén A.G., Wohlfart B. (1995) “A comparison between two exercise tests on cycle; a computerized test versus the Åstrand test.” Clin Physiol, 15 (1), page 91-102.
Each article will be scrutinized and the credibility will be evaluated. The credibility depends on the entirety of the study. The entirety of the study depends on if it have been conducted and presented in a systematic and logical manner. Uncertainty and/or lack of key information will reduce the credibility of the article.
When dissecting the articles the following questions were used as tools:
1. Which nomogram did the author/authors examine?
2. Was the first test excluded? 10
3. Which cycle ergometer was used?
4. Have they stated that they have checked the calibration of the ergometer?
5. Which were the workloads during submaximal work?
6. Pedalling speed?
7. How many submaximal tests with each person were conducted?
8. How many subjects were used?
9. How old were the subjects?
10. How many were men and how many were women?
11. Did they calculate BMI on the test subjects?
12. If BMI wasn’t calculated - which was the BMI using their reported values for mean weight and height?
13. How did they correct the values - age correction or heart rate correction?
14. What was the subjects’ health status?
15. Were they trained or untrained?
16. Which activity was used when determining submaximal VO2. 17. Which activity was used when determining maximal VO2? 18. Did they report the criteria of reaching maximal VO2?
11
Validity & reproducibility (reliability) A valid measure is one, which is measuring what it is supposed to measure.7 Validity consists of internal and external validity. The internal validity of a study refers to the accuracy of the research in determining the relationship between the dependent and independent variables (the integrity of the experimental design). The external validity is to what extent the research results can apply to a wider range of situations. If a result of a test/measure repeatedly is identical or closely similar the reproducibility (reliability) is high.
Envision that a scientist is interested in quantifying obesity. By systematically and carefully measuring the size of the feet (with a well calibrated instrument) data is collected. This is probably not going to make him/her any wiser. Determining BMI is probably a more valid measure. However, this must be done systematically; a mere glance in order to estimate BMI will lead to low reliability and low validity. Validity implies reliability; a valid measure has to be reliable, but a reliable measure does not have to be valid. 8
In a diagram, comparing for example determined max VO2 with estimated max VO2, it is essential that a line of identity, an ideal line, is drawn (e.g. figure 4). The line of identity represents perfect correlation (r = 1.0). With a line of identity included you can compare your regression line with the ideal line. If a line of identity is not included it is easy to be misled and assume that the presented values are better than they are (e.g. figure 9). Consequently you can have a high r-value but an incorrect (with respect to the ideal line) slope of the line (i.e. distant from the ideal line), implying over- or underestimation.
Parameters such as standard deviation (SD) and correlation coefficients (r) are interesting but blunt; the SD is based on average values and is merely correct with respect to the average values. The r-value is also an insufficient measure of a method’s validity.
7
Gunnarson R. Validitet och reliabilitet [Dept of Prim Health Care Göteborg University - Research methodology
web site]. March 13, 2002. Available at: http://infovoice.se/fou. Accessed June 26, 2002. 8
Ibid.
12
4. Background
4.1. Characteristics of the Åstrand-Ryhming test (1954)
The Åstrand-Ryhming test (1954) with its accompanying nomogram for calculation of aerobic capacity submaximally was presented in 1954 (figure 1). 9 It is based on the relationship between pulse rate during work and actual oxygen intake in % of a subject’s maximal aerobic capacity = a certain amount of oxygen is required for each workload (figure 2).
FIGURE 1
9
A nomogram is a graphic representation that consists of several lines marked off to scale and arranged in such a
way that by using a straightedge to connect known values on two lines an unknown value can be read at the point of intersection with another line.
13
FIGURE 2
4.1.1. Construction of the nomogram Research from Åstrand10 and Ryhming11 together with the investigations of Wahlund 12 concerning mechanical efficiency constitutes cornerstones in the attempts at creating a nomogram. Wahlund investigated men of various physical fitness and detected a pretty constant mechanical efficiency when cycling on a cycle ergometer. 469 men were tested; the investigation covered patients with various diagnoses or symptoms of heart or lung diseases (n = 376) and ordinary healthy subjects (n = 40) as well as moderately trained healthy subjects (n = 26) and athletes (n = 27). 10
Åstrand, P.-O. (1952) ”Experimental Studies of Physical Working Capacity in Relation to Sex and Age.”
Thesis. Munksgaard, Copenhagen. 11
Ryhming, I. (1953) “A modified Harvard Step test for the evaluation of physical fitness.” Arbeitsphysiologie,
15, page 235-250. 12
Wahlund, H. (1948) “Determination of the physical working capacity.” Acta Med. Scandinav. Suppl. 215.
Thesis. Stockholm.
14
The possibility to estimate oxygen uptake from workload within a range of + 8% in two-thirds of the cases was reported.13 Wahlund used a Krogh cycle ergometer with a pedalling rate of 60 rpm. To control the velocity a speedometer was used. The men being tested were instructed to keep a pace corresponding to 5 m/s.
Åstrand and Ryhming has shown that the oxygen uptake can be indicated from work level within a range of + 6% in two thirds of the subjects. “Identical values for mechanical efficiency were obtained for men and women (figure 3).” 14 A wide scattering of the subject’s mechanical efficiency, i.e. from 15–30 %, will not change the shape of the nomogram but simply result in more errors when predicting maximal aerobic capacity.
FIGURE 3
13
Wahlund, H. (1948) “Determination of the physical working capacity.” Acta Med. Scandinav. Suppl. 215.
Thesis. Stockholm, page 32. 14
Åstrand, P.-O. & Ryhming, I. (1954) “A nomogram for calculation of aerobic capacity (physical fitness) from
pulse rate during submaximal work.” J Appl Physiol, 7, page 218-219.
15
4.1.2. Material and methods
Åstrand and Ryhming used the Krogh cycle ergometer. According to Åstrand, P.-O. (personal communication) the ergometer was calibrated with controlled weights.15 When constructing the nomogram 27 male and 31 female well-trained subjects in the age of 20–30 years were used as a basis (figure 4).16 All of them were students at the Gymnastiska Centralinstitutet, Stockholm (G.C.I., College of physical education). The maximal oxygen intake was determined in maximal tests on the treadmill or on the cycle ergometer. In addition to this between three and five submaximal tests were conducted on several days. Subjects not reaching a steady state between 125 and 170 bpm were excluded. 17 “Within these limits there is normally an almost linear increase in metabolism with heart rate.” 18 The submaximal tests were a cycle test (900 kg m/min for women and 1200 kg m/min for men) where the O2 intake was determined. The pedalling frequency when testing the subjects submaximally was 50 rpm.19 Heart rates from 140 to 220 bpm were recorded at an oxygen uptake of 3.0 litres/min. A heart rate of 180 bpm represented an oxygen uptake of 2.0 litres/min for some women and as much as 5.0 litres/min for some men. The scattering results probably depend on differences in the subject’s maximal aerobic power.20
When the males were exercising at 50 % of their maximal oxygen uptake, their heart rates were on the average 128 bpm and when exercising on 70 % of their maximal aerobic power, the average heart rates were 154. For women the results were 138 and 164 bpm respectively (figure 5). The SD was 8-9 bpm. The findings in the above mentioned experiments made it possible to include scales in a nomogram with work levels (cycle test) and body weights (step test). When reading horizontally from the scale work level to the scale with O2 intake the 15
Personal communication with Åstrand, P.-O., 1/2-2004.
16
Åstrand, P.-O. & Ryhming, I. (1954) “A nomogram for calculation of aerobic capacity (physical fitness) from
pulse rate during submaximal work.” J Appl Physiol, 7, page 219. 17
Eriksson, M. & Larsson, B., (2001) ”Ergometri – konditionstest på cykel – en undersökning av hur
Åstrandstestet utvecklades och används idag”. C-essay from Idrottshögskolan, Stockholm, page 12. 18
Åstrand, P.-O. & Ryhming, I. (1954) “A nomogram for calculation of aerobic capacity (physical fitness) from
pulse rate during submaximal work.” J Appl Physiol, 7, page 221. 19
Ibid.
20
Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd
Edition. New York: McGraw-Hill Book Company, page 375-376.
16
actual energy output can be obtained. Together with information about the subject’s heart rate it is possible to estimate the maximal oxygen uptake (figure 1). Maximal heart rate was not a parameter in the 1954 study.21
FIGURE 4
FIGURE 5
21
Personal communication with Åstrand, P.-O., 1/2-2004.
17
4.1.3. Validity
Åstrand & Ryhming plotted the aerobic capacity calculated from the nomogram against the actual determined aerobic capacity. “For two-thirds of the cases the difference (standard deviation) will be less than 6.7 % for men and 9.4 % for women. With a lower rate of work, 600 and 900 kg m/min for women and men respectively, the standard deviation was higher, 14.4% for women and 10.4% for men” (see table 1). 22
TABLE 1
22
Åstrand, P.-O. & Ryhming, I. (1954) “A nomogram for calculation of aerobic capacity (physical fitness) from
pulse rate during submaximal work.” J Appl Physiol, 7, page 220.
18
The validity of the nomogram was controlled by comparing maximal oxygen intake calculated from the nomogram with the direct determined maximum using 18 well-trained male subjects, 18-19 years of age. The standard deviation was less than 7 %.23 For 31 females and 28 males between the ages of 20-30 years, the maximal oxygen uptake was calculated from both heart rate and O2 intake when doing a cycle test (600 and 900 kg/min), and from heart rate and O2 intake when doing a step test. These two values were compared and the standard deviation was 9.5 % and 7.3 % respectively.
The prediction of a subjects maximal VO2, from submaximal data can, when using the nomogram correctly, be made with a standard deviation of less than + 10 %.24 According to the authors the nomogram is based on results from experiments with healthy men and women between the ages of 20-30 years. The validity when testing younger or older people or patients with diseases is not known.
When determining maximal oxygen uptake Åstrand and Ryhming uses different testing equipment such as the cycle ergometer and the treadmill.25 This can be due to the fact that in his thesis from 1952, Experimental Studies of Physical Working Capacity in Relation to Sex and Age, P.-O. Åstrand detected no difference between the maximal values obtained during cycling and running (table 2).26 However, this might cause problems considering that the direct determined maximal oxygen uptake attained in various types of exercise such as cycling and running is likely to result in different values depending on whether or not you are an ordinary subject or a specially trained subject.27 In addition to this, running on a treadmill produces higher maximal oxygen uptake than cycling on an ergometer (on the average some 4-8 %).28 23
Åstrand, P.-O. & Ryhming, I. (1954) “A nomogram for calculation of aerobic capacity (physical fitness) from
pulse rate during submaximal work.” J Appl Physiol, 7, page 220 24
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.”. Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 45. 25
Åstrand, P.-O. & Ryhming, I. (1954) “A nomogram for calculation of aerobic capacity (physical fitness) from
pulse rate during submaximal work.” J Appl Physiol, 7, page 219. 26
Åstrand, P.-O. (1952) ”Experimental Studies of Physical Working Capacity in Relation to Sex and Age.”
Thesis. Munksgaard, Copenhagen, page 37-38. 27
Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd
Edition. New York: McGraw-Hill Book Company, page 356. 28
Ibid.
19
P.-O. Åstrand states in an interview from 2000 that when testing the subjects submaximally a higher pulse rate was detected in the first test compared to subsequent tests.29 This was presumed to be a result of anxiety.30 According to P.-O. Åstrand the subjects first had to get acquainted with the experimental procedures. The first determinations with each subject were therefore excluded and not used for the calculations of values.31
The exclusion is not to be found in the article from Journal of Applied Physiology, where the nomogram originally was presented, but merely in P.-O. Åstrand´s dissertation Experimental studies of physical working capacity in relation to sex and age from 1952 and in the c-essay by Eriksson & Larsson Ergometri – konditionstest på cykel - en undersökning av hur Åstrandstestet utvecklades och används idag. 32, 33 Therefore, anyone who reproduces P.-O. Åstrand´s and I. Ryhming´s procedure from the article in Journal of Applied Physiology might be doing it inadequately.
TABLE 2
29
Eriksson, M. & Larsson, B., (2001) ”Ergometri – konditionstest på cykel – en undersökning av hur
Åstrandstestet utvecklades och används idag”. C-essay from Idrottshögskolan, Stockholm, page 16-17. 30
Ibid.
31
Åstrand, P.-O. (1952) ”Experimental Studies of Physical Working Capacity in Relation to Sex and Age.”
Thesis. Munksgaard, Copenhagen, page 20. 32
Ibid.
33
Eriksson, M. & Larsson, B., (2001) ”Ergometri – konditionstest på cykel – en undersökning av hur
Åstrandstestet utvecklades och används idag”. C-essay from Idrottshögskolan, Stockholm, page 16-17.
20
Furthermore, it is difficult to find out whether or not a test subject has participated in one or several studies. This implies that the same person might have been (1) a part of the construction of the nomogram and (2) a part of Åstrand´s and Ryhming´s own validity study of the same nomogram. If this is the case Åstrand and Ryhming might have circle-validated themselves.34 In the interview from 2000, which is to be found in the above named c-essay, P.-O. Åstrand states that he neither has any notes left nor the memory of how the division was made.35
4.2. Characteristics of the Åstrand-Ryhming test (1960)
I. Åstrand adjusted and improved the original nomogram from 1954 and expanded the possibility to use “her” nomogram as a predictor of aerobic capacity with special reference to age and sex, see figure 6.36
4.2.1. Material and methods
Determinations of oxygen uptake and heart rate at submaximal and maximal work for physically active female and male subjects were used as the basis for the calculations. When constructing the nomogram I. Åstrand used 44 female and 100 male subjects as test subjects. The subjects were: housewives, 20-65 years of age, draymen37 50-64 years of age, males 5668 years of age who were members of a health club in Philadelphia, USA, males 27-45 years of age who participated in a physical training program (moderate training for 5-6 weeks).
She also used data from the Åstrand-Ryhming test from 1954 (29 male and 32 female students), making it a total of 205 subjects. The subjects from 1954 were all students at Gymnastiska Centralinstitutet, Stockholm (G.C.I., College of physical education). For number of subjects in the different materials and in various age groups see table 3.
34
Eriksson, M. & Larsson, B., (2001) ”Ergometri – konditionstest på cykel – en undersökning av hur
Åstrandstestet utvecklades och används idag”. C-essay from Idrottshögskolan, Stockholm, page 17. 35
Ibid.
36
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 45-60. 37
Drayman = somebody who drives a vehicle for a brewery
21
FIGURE 6
22
TABLE 3
In her investigations I. Åstrand used the Krogh and the von Döbeln cycle ergometer. The ergometers were calibrated.38 According to Eriksson & Larsson the submaximal tests, with one exception, were executed in the same way as in the construction of the first nomogram.39 The only exception was that I. Åstrand used both the Krogh and the von Döbeln cycle ergometer. In the nomogram from 1954 only the Krogh ergometer was used. According to P.O. Åstrand (personal communication) these two ergometers result in identical results, i.e. a given work load demands the same oxygen intake.40
“The work was performed on at least 2 days with light and heavy loads each day. On the last experimental day the subjects attempted to reach their maximal levels with regard to oxygen uptake.” 41 The work period was usually 5-10 min. An exception to this was when the subjects
38
Personal communication with Åstrand, P.-O., 1/2-2004.
39
Eriksson, M. & Larsson, B., (2001) ”Ergometri – konditionstest på cykel - en undersökning av hur
Åstrandstestet utvecklades och används idag.” C-essay from Idrottshögskolan, Stockholm, page 19. 40
Personal communication with Åstrand, P.-O. 23/6-2003.
41
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 49.
23
were working on the maximal load - when doing this the work period was sometimes limited to 2-3 min.
”The values used for oxygen uptake and heart rate refer to the last minutes of work on each load and are usually mean values obtained either from determinations on two or more occasions, or from two or more successive determinations on the same occasion. An exception to this are the determinations of oxygen uptake made at maximal work, whereby the highest value was used if it differed from the other determinations by more than 0,08 l/min.”42 To establish the degree of strain the blood lactate concentration was measured. The values indicate that the subjects worked at or close to their maximum.
4.2.2. Adjustment of the original nomogram
The nomogram from 1954 was adjusted by means of:
(1) A correction of the mechanical efficiency, since the mechanical efficiency is considerably lower at lower loads, e.g. at 300 kpm/min than at 900 kpm/min.43 The error becomes relatively large when using the lower loads if this is not being considered. The use of lower loads is necessary for certain populations, e.g. older individuals. This was put into practice by changing the relationship between the workload scale and the VO2 scale. (2) The construction of separate workload scales for the two sexes, since there is a difference in oxygen uptake at a certain workload between males and females. “However, the net mechanical efficiency is the same.”44 There was also a small difference in oxygen uptake between younger and older subjects at a specific workload. It did not however justify different scales for the two groups; instead a mean value for all age groups was used.45
A finding when constructing the nomogram was that a better agreement between measured and predicted maximal oxygen uptake was obtained if the subjects used relatively high 42
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 49. 43
Ibid.
44
Ibid.
45
Ibid.
24
workloads.46 “Thus, there is a small systematic difference in the maximal oxygen uptake calculated from low and high work loads.”47 It was recommended that only pulse rates between 125 and 170 bpm should be used for calculations.
4.2.3. Correction factor for age
Older people do not reach maximal pulses as high as younger people, thus an age correction factor was introduced (figure 7). However, I. Åstrand states that it is possible that the proposed age correction factors are too low causing underestimation when using the presented values.48, 49 The correction factor is applicable on both males and females.50 The material used for determining the correction factor is relatively small.51
Irma Åstrand expresses that the original idea with the nomogram was that it should be easy to handle and that too many correction factors might - in fact - limit the use of the nomogram.52
4.2.4. Accuracy of prediction
The accuracy of the nomogram is somewhat higher when using relatively higher loads and when applying the nomogram on well-trained, younger persons than to the whole material (SD = + 10% in comparison with the whole material, + 15%), see figure 8.53 “The reason for this difference is probably that the physical education students represent a selected group of subjects with a comparatively large oxygen uptake capacity. Moreover, they have had more severe physical training than the other subjects. Physical training does not only increase the 46
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 49. 47
Ibid, page 50.
48
Ibid, page 54.
49
The Åstrand-Ryhming test/method has often been accused of suffering from underprediction, see chapter 7,
Results and discussion, for a discussion about underestimation. 50
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 54-55. 51
Ibid.
52
Personal communication with I. Åstrand, 23/6-2003.
53
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 55-56.
25
maximal oxygen capacity, but also changes the individual slopes of the curves for heart rate and oxygen uptake. This means that the nomogram might have had a slightly different appearance if another original material had been used.”54
FIGURE 7
Can higher accuracy be gained when performing repeated tests with the same subjects? In a study from 1958 by I. Åstrand on inexperienced persons (draymen, n = 81) some of them (n =11) had, at a certain workload (900 kpm/min), lower values for several physical parameters on the second experimental day. 55 “Usually the values for heart rate, oxygen intake, pulmonary ventilation and blood lactate concentration for a given work load were lower on the second experimental day than on the first; mechanical efficiency was accordingly higher.”56 Åstrand continues: “In many cases the difference in oxygen intake and in heart rate from day one to day two was 0.2 litres/min. and 15 or more beats/min. respectively.”57 However, the average difference was not statistically significant.58
54
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 56-57. 55
Åstrand, I. (1958) “The physical work capacity of workers 50-64 years old.” Acta Physiologica Scandinavica
42, page 73-86. 56
Ibid, page 77.
57
Ibid.
58
Ibid.
26
When the original nomogram was constructed the first determinations with each subject were excluded and not used for the calculations of values. In the study on draymen (1958), that was a part of designing the second nomogram, the first determinations with each subject were also excluded consequently. Thus, it is likely that this was done in all of the different studies that were a part of constructing the modified nomogram.59 This is confirmed by I. Åstrand and P.O. Åstrand (personal communication).60 According to I. Åstrand the higher mechanical efficiency on the second test day, in comparison with on the first test day, can be explained with lower tension/nervousness among the test subjects.61
In the original nomogram the sole testing advice used when testing the subjects submaximally was the cycle ergometer. When testing the subjects maximally both the cycle ergometer and the treadmill were used. In the modified nomogram from 1960 I. Åstrand choose only to use the cycle ergometer. Data from the original nomogram was however included in the shaping of the modified nomogram. This means that some of the testing results, of which the modified nomogram is based upon, also derive from tests conducted on the treadmill.
I. Åstrand emphasizes that the method of measuring work load and heart rate submaximally in order to determine a persons aerobic capacity only gives a hint of a subjects aerobic capacity. For exact information one has to measure the aerobic work capacity directly.62
59
However, this is not stated in all of the articles.
60
Confirmed by I. Åstrand (23/6-2003) and P.-O. Åstrand (1/2-2004) in personal communication.
61
Åstrand, I. (1958) “The physical work capacity of workers 50-64 years old.” Acta Physiologica Scandinavica
42, page 77. 62
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 59.
27
FIGURE 8
28
5. Results 5.1. The articles
This chapter aims to introduce each article, mainly with respect to the questions presented in chapter 3. The 14 articles about the Åstrand-Ryhming test/method can more or less be divided into one of the following two groups depending on the aim: group 1, with focus on validity and group 2 with focus on modification.
Group 1, focus on validity 1. 2. 3. 4. 5. 6. 7. 8. 9.
Cink R.E., Thomas T.R. (1981), Jessup G.T., Riggs C.E., Lambert J., Miller W.D. (1977) Jessup G.T., Tolson H., Terry J.W. (1974) Jetté, M. (1979) Kasch, F.W. (1984) Macsween, A. (2001) Teräslinna P., Ismail A.H., MacLeod D.F. (1966) Williams, L. (1975) Wisén, A.G., Wohlfart B. (1995)
Group 2, focus on modification 1. 2. 3. 4. 5.
Cullinane E.M., Siconolfi S., Carleton R.A., Thompson P.D. (1988) Jessup G.T., Terry J.W., Landiss C.W. (1975) Legge B.J., Banister E.W. (1986) Siconolfi S.F., Cullinane E.M., Carleton R.A., Thompson P.D. (1982) Terry J.W., Tolson H., Johnson D.J., Jessup G.T. (1977)
5.1.1. An overview of the articles If the authors haven’t stated which nomogram they have examined and not mentioned any correction factors for age it was assumed that they investigated the original nomogram, i.e. the Åstrand-Ryhming nomogram (1954). A hyphen (-) in the following two tables (4A & 4B) implies that the answer is not to be found in the article or that it is difficult to interpret what the authors suggest.
29
TABLE 4A, An overview of the articles, questions 1-9 Questions Which nomogram?
Exclusion of the first test?
Which cycle ergometer?
Was the ergometer calibrated?
Workload, submax work?
Pedalling speed?
Number of submax tests?
How many subjects were used?
Age of the subjects?
Cink et al (1981)
1960
No
Monark
-
75/100/150 W
50 rpm
1
51 (40)
18-33 years
Jessup et al (1977)
1954
No
Schwinn Ergometric
Yes
-
50 & 80 rpm
2
30
18-24 years
Jessup et al (1974)
1954
No
Monark
Yes
-
-
1
40
18-23 years
Jetté (1979)
1960
No
Monark
-
-
-
1
64
20-54 years
Kasch (1984)
1960
No
Monark
-
-
50 rpm
1
83
30-66 years
Teräslinna et al (1966)
1960
No
Monark
-
900 kpm/min
50 rpm
1
31
23-49 years
Williams (1975)
1954
No
-
-
450 kpm (starting point)
-
4
31
College-age
Monark, yes Cat Eye, not possible
-
50 rpm
4
20
31-49 years
Articles
Wisén et al (1995)
1954
No
Monark and Cat Eye Ergociser
Jessup et al (1975)
1954
No
Monark
Yes
600 kpm (starting point)
-
1
60
18-23 years
Legge and Banister (1986)
1954
No
Quinton (QI-854)
Yes
Various work loads64
50 rpm
2/3 (1Å-R-test)
39
20-29 years
Siconolfi et al (1982)
1960
No
Philbin
Yes
♀=300 kpm ♂=600 kpm (starting point)
50 rpm
1
113
20-70 years
Terry et al (1977)
1954
No
Monark
Yes
600 kpm
-
1 (if the desired HR was reached)
60
18-23 years
50 rpm
3-5
64
20-30 years
50 rpm
>2
205
20-68 years
Åstrand & Ryhming 1954
1954
Yes
Krogh
Yes
♀=9001500 kpm ♂=600-900 kpm
I. Åstrand 1960
1960
Yes
Krogh & von Döbeln
Yes
Various work loads
63
The article by Cullinane et al (1988) and by Macsween (2001) is not included in table 4A and table 4B due to
the fact that they did not conduct any trials. 64
Various work loads, 50 W incremental steps, each lasting 3 minutes at a pedal rate of 90 rpm to ∼80% of a
subjects age-predicted VO2 max.
30
TABLE 4B, An overview of the articles, questions 10-18 Questions
Sex of the test subjects?
Did they calculate BMI?
BMI, using reported figures?
Correction of values?
Health status?
Training status?
Activity, submax VO2?
Activity, max VO2?
Criteria of reaching max VO2?
Cink et al (1981)
♂
No
23.6
Age correction
-
Various physical fitness
Cycle ergometer
Cycle ergometer
Yes
Jessup et al (1977)
♂
No
23.7
-
-
-
Cycle ergometer
Cycle ergometer
-
Jessup et al (1974)
♂
No
22.3
-
-
Students of phys.ed.
Cycle ergometer
Treadmill
-
Jetté (1979)
♂ 35 ♀ 29
No
♂ 25.6 ♀ 21.2
Age correction
-
Sedentary
Cycle ergometer
Treadmill
Yes
No
24.5
Age correction
Healthy
-
Cycle ergometer
Cycle ergometer
Yes
No
No figures were reported
Age correction
Healthy
Members of a fitness program
Cycle ergometer
Cycle ergometer
Yes
Articles
Kasch (1984)
♂
Teräslinna et al (1966)
♂
Williams (1975)
♀
No
No figures were reported
-
-
Students of phys.ed.
Cycle ergometer
-
-
Wisén et al (1995)
♀
No
22.25
Age correction
Healthy
Various physical fitness
Cycle ergometer
-
-
No
22.4
-
-
Students of phys.ed.
Cycle ergometer
Treadmill
-
Jessup et al (1975)
♂
Legge and Banister (1986)
♂
No
No figures were reported
-
-65
Various physical fitness
Cycle ergometer
Cycle ergometer
Yes
Siconolfi et al (1982)
At least 10 ♂ & ♀/age decade
No
No figures were reported
-
Healthy
Mainly inactive
Cycle ergometer
Cycle ergometer
Yes
No
22.4
-
-
Students of phys.ed.
Cycle ergometer
Treadmill
-
♂ 35 ♀ 29
-
♂ 22.6 ♀ 21.9
-
Healthy
Well trained
Cycle ergometer
Cycle ergometer & Treadmill
Yes
♂ 129 ♀ 76
-
♂ 24.2 ♀ 22.4
-
Healthy
Physically active
Cycle ergometer
Cycle ergometer66
Yes
Terry et al (1977) Åstrand & Ryhming 1954 I. Åstrand 1960
♂
65
Legge and Banister merely states that the subjects were medically examined prior to beginning the study.
66
Data from the original nomogram was however included in the 1960 nomogram. Consequently some of the
testing results derive from tests conducted on the treadmill.
31
5.1.2. Group 1, with focus on validity Cink et al (1981) evaluated the predicted VO2 max from the Åstrand-Ryhming test (1960) using the cycle ergometer as the sole exercise mode.67 They followed the recommended submaximal test protocol with a pedalling speed of 50 rpm using two sets of correction factors (1) Age correction by I. Åstrand (1960) and (2) Age correction by von Döbeln et al. 68 Cink et al used a Monark ergometer (although they have not stated that they have checked the calibration, but they have stated that the seat height was adjusted so that the subjects knee was slightly flexed when the ball of the foot rested on the pedal at the lowest point in a revolution). BMI was not calculated in the study; using the authors reported figures for mean weight and height resulted in a BMI of 23.6.69 40 men between the ages of 18-33 years conducted two tests, one submaximal and one maximal. The subjects were of various physical fitness. It is not stated whether or not the subjects were healthy. The maximal test was conducted after the submaximal test. A pedalling speed of 60 rpm was used and the workload was 25 W for the first two minutes. The workload was increased by 25W every minute until the subject was exhausted. The criteria of reaching maximal VO2 were: exhaustion, ECG changes, marked dyspnoea, confusion, pallor, or pain in the chest, arms or jaw. The result indicate roughly that 2/3 of the VO2 max measurements elicited during a maximal cycle ergometer test will be within + 0.42 L min-1 (r = 0.76) or 5.7 ml kg-1min-1 (r = 0.83) of the VO2 predicted from the Åstrand-Ryhming test (1960), see table 5.70 It was concluded that there was no significant difference between the measured and the predicted means. A diagram showing the correlation between the directly determined maximum oxygen uptake and the prediction of maximum oxygen uptake is missing. 67
Cink R.E., Thomas T.R. (1981). “Validity of the Åstrand-Ryhming nomogram for predicting maximal oxygen
intake.” Br J Sports Med, 15 (3), page 182-185. 68
von Döbeln, W., Åstrand, I., Bergström, A. (1967). ”An analysis of age and other factors related to maximal
oxygen uptake.” J Appl Physiol, 22, page 934-938. 69
When converting weight in pounds to weight in kg 0.4536 multiplied the pounds and when converting height
in inches to height in cm 0.2540 multiplied the inches. Figures for conversion were found on www.onlineconversion.com. 70
The usual measure of linear association is the Pearson correlation coefficient, symbolized by an r.
32
TABLE 5
Jessup et al (1977) investigated the effect of pedalling speed on the validity of the ÅstrandRyhming test (1954).71 Research had suggested that perceived exertion was reduced at higher pedalling speeds for equivalent power outputs. The purpose of the study was to determine the validity of the Åstrand-Ryhming test (1954) administered at 50 and 80 rpm. The cycle ergometer was a calibrated Schwinn Ergometric (an electronic cycle ergometer) and was used in all tests, submaximal as well as maximal. BMI was not calculated in the study; using the authors reported figures for mean weight and height resulted in a BMI of 23.7. 30 male volunteer subjects between the ages of 18-24 years performed on consecutive days (a) the Åstrand-Ryhming test (1954) administered at 50 rpm; (b) the Åstrand-Ryhming test (1954) administered at 80 rpm; and (c) a maximum oxygen intake determined at the end of a maximal graded exercise test. The graded exercise test consisted of performing consecutive six-minute work bouts at 300, 600, 900 and 1200 kpm, or until the exercise heart rate reached 180 bpm, at which point the subject was pushed to maximum aerobic power by increasing the pedal speed and resistance. The criterion of reaching max VO2 is lacking as well as information about the subjects’ health- and training status.
71
Jessup G. T., Riggs C. E., Lambert J., Miller W. D. (1977) “The effect of pedalling speed on the validity of the
Åstrand-Ryhming aerobic work capacity test.” J Sports Med, 17 (4), page 367-371.
33
TABLE 6
The results revealed a correlation coefficient of r = 0.64 and r = 0.63 for the ÅstrandRyhming test (1954) administered at 50 and 80 rpm, respectively. This can be compared with the Åstrand-Ryhming test (1954) with an r = 0.71 and with the Åstrand-Ryhming test (1960) with an r = 0.78. The correlation between the test conducted at 50 rpm compared with the test conducted on 80 rpm was r = 0.85. 72 See table 6 for a comparison of predicted and measured scores by different investigators. Jessup et al (1974) compared the prediction of maximal oxygen intake with simple regression models, the Åstrand-Ryhming test (1954) and a 12-Minute Run with multiple regression
72
According to Åstrand & Rodahl the oxygen uptake varies with pedal frequency, see Åstrand, P.-O. & Rodahl,
K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd Edition. New York: McGrawHill Book Company, page 365. The discrepancy probably depends on that Jessup et al (1977) and Åstrand & Rodahl (1986) have examined different types of cycle ergometers, i.e. the mechanically braked cycle ergometer and the electromechanically braked cycle ergometer.
34
models in young volunteer subjects.73 They used 40 volunteer male college students enrolled in physical education activity courses. The men were between the ages of 18-23 years with a mean of 19.5 years. Simple regression techniques such as the Åstrand-Ryhming test/method and the 12-minute Run were administered approximately one week apart and compared with multiple regression models. When testing the students submaximally a calibrated Monark and standard procedures were used. The maximal test was the Balke treadmill test.74 BMI was not calculated; using the authors reported figures for mean weight and height resulted in a BMI of 22.3. The data was analysed with a computer program that performed a stepwise multiple linear regression (UCLA BIMED 02R). The individual correlations between max VO2 and prediction from the Åstrand-Ryhming test/method was r = 0.64 (figure 9). Information about the subjects’ health- and training status and criteria of reaching max VO2 are lacking. Jette (1979) compared the predicted maximal oxygen consumption derived from the Canadian Home Fitness Test (CHFT)75 and the Åstrand-Ryhming test (1960) to the observed VO2 max determined from a progressive multi-stage treadmill test.76 64 sedentary subjects between the ages of 20-54 years old (35 men and 29 women) were used as test subjects. The tests were conducted on two different occasions. On occasion 1: CHFT and the Åstrand-Ryhming test (1960) with a minimum of 30 minutes in between the two tests, occasion 2: maximal VO2 test using a multi-stage treadmill test. 73
Jessup, G.T., Tolson H, Terry J.W., (1974) “Prediction of maximal oxygen intake from Åstrand-Ryhming test,
12-minute run and anthropometric variables using stepwise multiple regression.” Am J Phys Med, 53 (4), page 200-207. 74
For additional information about the Balke treadmill test see Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of
work physiology. Physiological Bases of exercise.” 3rd Edition. New York: McGraw-Hill Book Company, page 362. 75
The Canadian Home Fitness test is a safe, simple, self-administered fitness test. It is a steptest - the steps are
20.3 cm high. For further information about the CHFT see Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd Edition. New York: McGraw-Hill Book Company, page 380. 76
Jette M. (1979) “A comparison between predicted VO2 max from the Åstrand procedure and the Canadian
Home Fitness Test.” Can J Appl Sport Sci, 4 (3), page 214-218.
35
FIGURE 977
Jette used a Monark cycle ergometer (no statement of calibration). The criteria of reaching maximal VO2 was if the test subject had achieved a heart rate equal to or in excess of his predicted maximum heart rate established for his age and indicated that he could no longer continue walking. Information about the subjects’ health status is lacking. According to Jetté the Åstrand-Ryhming test (1960) underpredicts the observed VO2 max particularly in males, while this discrepancy is not to be found in the female population.78 The Åstrand-Ryhming test (1960) provides reasonably accurate predictions among young active individuals but tends to underestimate when it is employed in sedentary and older subjects.79 The predictive correlation coefficients for the Åstrand-Ryhming test (1960) were for men and women combined r = 0.47, for men r = 0.54 and for women r = 0.59. The mean VO2 max measured on the treadmill for males and females combined was 34.6 + 6.0 ml kg-1min-1 while the Åstrand-Ryhming test/method predicted a mean VO2 max of 29.6 + 6.5 ml kg-1min-1 and 77
78
An obvious drawback with presenting values in this way is that a line of identity (an ideal line) is missing. Jette M. (1979) “A comparison between predicted VO2 max from the Åstrand procedure and the Canadian
Home Fitness Test.” Can J Appl Sport Sci, 4 (3), page 216-217. 79
Ibid.
36
the CHFT predicted a mean VO2 max of 34.8 + 5.0 ml kg-1min-1 (see table 7 for further information).
TABLE 7
Kasch (1984) investigated the validity of the Åstrand-Ryhming test (1960).80 He also included two other tests in the same session, the Sjöstrand test and a direct VO2 measurement. According to the author conducting all three tests at one testing session minimizes any diurnal or other differences that might occur in protocols using separate tests on the same or alternate days.81 The test was continuous, using the three Sjöstrand stages of six minutes each and finishing after the 18th minute with a directly measured VO2 max. The 3rd six-minute stage of the Sjöstrand portion was used for determining the Åstrand-Ryhming VO2 max prediction.82 The five- and six-minute heart rates were then averaged and the VO2 max prediction was 80
Kasch, F. W. (1984) “The validity of the Åstrand and Sjöstrand submaximal tests.” The Physician and
Sportsmedicine, page 47-51, 54. 81
Ibid, page 48.
82
The recommended test protocol was thus neglected. Each test subject had pedalled for at least 12 minutes
when beginning the Åstrand-Ryhming test/method.
37
determined from a table in relation to workload, sex and age. The criterion of reaching max VO2 was the levelling off the VO2 with increasing workloads. Kasch choose to elicit pulse rates between 120-170 instead of pulse rates between 125-170 bpm.83,84 83 healthy men between the ages of 30-66 years were used as test subjects. They had to be free of infection, free of heart disease, normotensive85 and cleared by a physician. Kasch used a Monark cycle ergometer, but no statement of checked calibration is to be found in the article. BMI was not calculated in the study; using the reported figures for mean weight and height resulted in a BMI of 24.5. It is not stated whether or not the subjects were trained or untrained.
TABLE 8
The predictive correlation coefficient for the Åstrand-Ryhming test (1960) was r = 0.58 (age corrected) and r = 0.46 uncorrected (table 8). The Åstrand-Ryhming test (1960) underpredicted the directly measured VO2 max by 21%. According to Kasch the ÅstrandRyhming test (1960) is an inappropriate substitute for direct measurement of maximal oxygen intake.
83
The average heart rate was 142 + 9.6 bpm.
84
To get reliable estimates of VO2, max, it is recommended to elicit heart rates of at least 125 bpm, see Åstrand,
I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 49. 85
Blood pressure below 140/90 mmHg.
38
Macsween (2001) investigated the reliability and validity of the Åstrand-Ryhming test/method and the linear extrapolation for deriving VO2 max from submaximal exercise data.86 25 normal subjects (12 men, 13 women) participated in the study (mean age 28.6 years, range 18-50). They completed 52 maximal trials (Bruce treadmill protocol).87 No submaximal tests were conducted. All subjects were healthy.88 Data was analysed to compare the reliability and validity of linear extrapolation to three estimates of heart rate maximum using a computer simulation of Åstrand-Ryhming test/method. BMI was not calculated in the study; using the reported figures for mean weight and height resulted in a BMI of 23.3. The criterion of reaching maximal VO2 was exhaustion. Since a computer model of the Åstrand-Ryhming test/method was utilized no submaximal or maximal trials were conducted on the cycle ergometer.89 Age predicted maximal heart rate was determined by three commonly used methods; 220-age (referred to as the Emax) and from regression equations of Londeree and Moeschbeger (referred to as the EAmax and EBmax respectively).90 The Åstrand-Ryhming prediction produced the greatest bias at 3.48 ml kg-1min-1 overestimate and the extrapolation Emax (220 – age) the lowest at 0.46 ml kg-1min-1. See table 9 for raw values of each derived measure (Omax = observed VO2 max, Pmax = Åstrand-Ryhming nomogram prediction, Emax = age predicted maximal heart rate – 220 – age, EAmax = regression equations of Londeree and EBmax = regression equations of Moeschbeger). 91
86
Macsween A. (2001) “The reliability and validity of the Åstrand nomogram and linear extrapolation for
deriving VO2 max from submaximal exercise data.” J Sports Med, 41 (3), page 312-317. 87
For additional information about the protocol developed by Bruce (1971) see Åstrand, P.-O. & Rodahl, K.
(1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd Edition. New York: McGraw-Hill Book Company, page 362-363. 88
Macsween A. (2001). “The reliability and validity of the Åstrand nomogram and linear extrapolation for
deriving VO2 max from submaximal exercise data.” J Sports Med, 41 (3), page 313. 89
Ibid, page 314.
90
Ibid.
91
Londeree and Moeschbeger developed equations for estimations of maximal heart rate. Londeree, B.R.,
Moeschbeger, B.L. (1982) ”Effect of age and other factors on maximal heart rate.” Res Q, 53, page 297-304.
39
TABLE 9
Teräslinna et al (1966) investigated the Åstrand-Ryhming test (1960) as a predictor of maximum oxygen intake.92 The purpose of the study was to obtain a validity coefficient for the final form of the nomogram developed by Åstrand & Ryhming. 31 healthy men (23-49 years of age, mean 35.8) engaged in sedentary occupations were used as test subjects. The cycle ergometer was a Monark. Neither BMI nor any figures for mean weight and height were reported. The test subjects were engaged in an adult fitness program (however nothing is mentioned about their training status). See table 10 and figure 10 for additional data including a comparison between determined and predicted maximum oxygen uptake. A submaximal workload of 900 kpm/min and a pedalling speed of 50 rpm were used. The maximal test was a modified Balke test (on a cycle ergometer). The subjects rode until exhaustion or until two consecutive heart rate determinations were constant. The correlation between maximum oxygen uptake and the prediction from the nomogram resulted in an r = 0.69 (uncorrected for age). Using age correction factors resulted in an r = 0.92. Thus the correlation increases when using the correction factors for age. Teräslinna et al concluded that the nomogram from 1960 with accompanying age correcting factors is a satisfactory predictor of maximal oxygen uptake.
92
Teräslinna P., Ismail A.H., MacLeod D.F. (1966) “Nomogram by Åstrand and Ryhming as a predictor of
maximum oxygen intake.” J Appl Physiol, 21 (2), page 513-515.
40
TABLE 10
FIGURE 10
Predicted maximum oxygen uptake (l/min)
Comparison between direct determined maximum oxygen uptake and the prediction of maximum oxygen uptake. A line of identity is drawn (values from Teräslinna et al , 1966).
4,50 4,00 3,50 3,00 2,50 2,00 2,00
2,50
3,00
3,50
4,00
Direct determined maximum oxygen uptake (l/min)
41
4,50
Williams (1975) investigated the reliability of predicting VO2 max using the ÅstrandRyhming test/method.93 31 female physical education students were used to study the testretest reliability of oxygen uptake estimated from the Åstrand-Ryhming test/method. Williams did not calculate BMI on the subjects and no figures of mean weight and height are to be found in the article. The name of the cycle ergometer and possible calibration is also lacking. “Adjustments” were done with the cycle ergometer.94 Information about the subjects health status and if they were trained or untrained is also missing. The subjects conducted four tests, two tests on the same day, which was repeated exactly one week later (no maximal test was conducted). Thus, each woman’s oxygen uptake was estimated on four different occasions. After a 10-minute rest, the procedure was repeated. Prior to the beginning of the second test the heart rate was checked to be sure that it had returned to within 10 beats of the original resting heart rate. There was no significant day effect or day by trial effect. However, there was a significant trial effect. Inspection of the means revealed that the estimated oxygen uptake for the second trial was lower (for means and standard deviations see table 11). It was concluded that oxygen uptake estimated from the Åstrand-Ryhming test/method tends to be low when a single trial is administered. ANOVA95 findings reveal that additional trials over days are preferable to additional trials administered within days (see table 12 for estimated reliabilities of additional test days). Thus, reliability increases with the number of times the test has been administered.
93
Williams, L. (1975) “Reliability of predicting maximal oxygen intake using the Åstrand-Ryhming
nomogram.” Res Quart, 46, page 12-16. The study/article was conducted in partial fulfilment of the requirement for a Master’s of Education degree. 94
It is unclear what Williams means with adjustments.
95
ANOVA = Analysis of variance. ANOVA is a technique for quantifying and partitioning sample variance in
experimental data. It is also called F test after Fischer, the originator.
42
TABLE 11
TABLE 12
Wisén et al (1995) compared two exercise tests on cycle - a computerized test versus the Åstrand-Ryhming test/method.96 The aim of the study was to assess agreement and reproducibility between the VO2 max obtained from the Åstrand-Ryhming test/method and with the one of the computerized submaximal extrapolation tests (Cat Eye Ergociser EC 1500). When evaluating the Åstrand-Ryhming test/method the original nomogram from 1954 with age-correction factors from 1960 were used. A calibrated Monark ergometer model 90810 and the recommended submaximal test protocol including a pedalling speed of 50 rpm were used. The Cat Eye Ergociser EC 1500 was not possible to calibrate. 20 healthy women participated in the study.97
96
Wisén A.G., Wohlfart B. (1995) “A comparison between two exercise tests on cycle; a computerized test
versus the Åstrand test.” Clin Physiol, 15 (1), page 91-102. 97
Four subjects were on medication: oestrogen 25 mg/daily (1 person), nasal budesonide (2 persons),
amitryptilin 25 mg/daily (1 person).
43
BMI was not calculated; using the reported figures for mean weight and height resulted in a BMI of 22.25. The women were divided into two groups of ten according to their age (mean 35.3 years and 46.9 years, range 31-49). Prior to each test the subjects were shown the equipment and informed about the testing procedures. Each subject participated in four separate test sessions (test-retest with each method). Tests were performed over two consecutive days; two tests - one with each method - were performed on each day with 2 h in between, on approximately the same time of the day. The computerized submaximal test was performed for 10 minutes with a pedalling rate of 60 rpm. The workloads were automatically set at three different levels. A line was fitted to the data collected at the end of the 7th and 10th minute (work rate and heart rate). The regression line was extended until it reached the predicted maximum heart rate (204 – (0.69 × age)).98 The VO2 max was calculated assuming that 1 l of oxygen corresponds to 5 kcal and that the human efficiency for cycle exercise is 23%.99 The VO2 max estimated from the Cat Eye Ergociser was lower compared with the VO2 max estimated from the Åstrand-Ryhming test/method. For a plot of the differences of the estimated VO2 max against the mean VO2 max see figure 11. A statistically significant linear relationship was found between the two methods, r = 0.85 (figure 12). Some of the estimates derived from the computerized test had large errors reducing the agreement between the two tests. The variation (the coefficient of reproducibility) expressed in % of mean max VO2 was19 % for the Åstrand test and 34 % for the computerized test. Wisén et al concludes that the computerized method gives VO2 max estimates that are lower and more uncertain than the Åstrand-Ryhming test/method. Therefore, it is not recommendable to use the two methods interchangeably.
98
Current research proposes that the most correct equation for estimating maximal heart rate is that of Inbar
(1994): 205.8 - (0.685 x age). For a manuscript written to provide insight into the history of maximal heart rate see Robergs, A.R., Landwehr, R. (2002) “The surprising history of the 220 – age equation.“ Journal of Exercise Physiology online. Volume 5, May 2002, page 1-10. 99
Wisén A.G., Wohlfart B. (1995) “A comparison between two exercise tests on cycle; a computerized test
versus the Åstrand test.” Clin Physiol, 15 (1), page 93.
44
FIGURE 11
FIGURE 12
45
5.1.3. Group 2, with focus on modification The article by Cullinane et al (1988), “Modification of the Åstrand-Ryhming submaximal bicycle test for estimating VO2 max of inactive men and women”, is really not an article, it is a letter to the editor-in-chief of Medicine and Science in Sports and Exercise. 100 The letter describes and clarifies the methodology of an earlier article by the same authors, Siconolfi et al (1982).101 Furthermore, the letter let us know that the original paper contained an error, inadequate increments had been recommended for men under 35 years of age. The letter and the article by Cullinane et al focused on the Åstrand-Ryhming test from 1960. Jessup et al (1975) developed a multiple regression equation to predict individual workloads for the Åstrand-Ryhming test (1954) in order to guide the novice experimenter in workload selection.102 A calibrated Monark cycle ergometer was used when testing the subjects submaximally. 60 male subjects between the ages of 18-23 years were test subjects. They were all students at the department of physical education at Texas A & M University. BMI was not calculated; using the reported figures for mean weight and height resulted in a BMI of 22.4. When testing the subjects maximally the Balke treadmill test of maximal work capacity was conducted. The test was ended at exhaustion. Based upon these scores the subjects were assigned to two equal groups (experimental group and cross-validation group). Information about the test subjects health- and training status is not to be found. The data from the experimental group were analysed with a computer program that performed a stepwise multiple linear regression (UCLA BIMED 02R). In order to establish the dependent variable of workload the subjects rode the cycle ergometer at 600 kpm using standard ÅstrandRyhming protocol. The independent variables were age, height, weight, resting heart rate,
100
Cullinane E.M., Siconolfi S., Carleton R.A., Thompson P.D. (1988) “Modification of the Åstrand-Ryhming
submaximal bicycle test for estimating VO2 max of inactive men and women.” Med Sci Sports Exerc, 20 (3), page 317-318. 101
Siconolfi S.F., Cullinane E.M., Carleton R.A., Thompson P.D. (1982) ”Assessing VO2 max in
epidemiological studies: a modification of the Åstrand-Ryhming test.” Med Sci Sports Exerc, 14 (5), page 335338. 102
Jessup G.T., Terry J.W., Landiss C.W. (1975) “Prediction of workload for the Åstrand-Ryhming test using
stepwise multiple linear regression.” J Sports Med, 15 (1), page 37-42.
46
resting blood pressure, one-minute exercise heart rate (at 600 kpm), leg length standing, and static leg strength (at 115° extension). Workload for the Åstrand-Ryhming test/method was calculated for the cross-validation group using the established multiple regression equation. In order to test the goodness of fit between the criterion maximal oxygen intake values and those predicted from the Åstrand-Ryhming test/method the Kolmogorov-Smirnov one-sample test was applied.103 Application of stepwise multiple linear regression analysis for the prediction of workload required to attain a sixth-minute heart rate near 165 bpm yielded a multiple correlation of r = 0.763. The multiple regression equation was: Workload = 1123.452 + Age (1.746) + Weight (2.201) −Minute heart rate (7.386). The correlation between predicted and criterion maximal oxygen intake values was r = 0.65. The authors conclude that the novice experimenter, when administering the Åstrand-Ryhming test/method on a similar population, might improve the validity of workload selection by using the equation. It was additionally concluded that the Åstrand-Ryhming test/method only should be used in general fitness-screening situations where more sophisticated techniques are unavailable.104 Legge and Banister (1986)105 aimed to improve the accuracy of the Åstrand-Ryhming test (1954) by developing a new nomogram, based on the linear relationship between VO2 and ∆ HR, the later being defined as the elevation of exercise heart rate above that reached during zero-load pedalling at 90 rpm (see figure 13 for the Legge-Banister nomogram).106 Three groups of males (ages from 20-29 years of age) took part in the study. Two were core groups of trained individuals (n = 15) and untrained (n = 10). The third group (n = 14) was a
103
The Kolmogorov-Smirnov test tries to determine if two datasets differ significantly. The KS-test has the
advantage of making no assumption about the distribution of data. 104
Jessup G.T., Terry J.W., Landiss C.W. (1975) “Prediction of workload for the Åstrand-Ryhming test using
stepwise multiple linear regression.” J Sports Med, 15 (1), page 40. 105
Legge B.J., Banister E.W. (1986) “The Åstrand-Ryhming nomogram revisited.” J Appl Physiol, 61 (3), page
1203-1209. 106
The nomogram was developed in the same manner as the original nomogram.
47
mixed group of trained (n = 5), untrained (n = 5) and moderately trained (n = 4) individuals used to validate the predictive nomogram for VO2 max developed from the test result of the core groups. Legge and Banister states that the test subjects were medically examined prior to the beginning of the study but nothing is mentioned about the results of the investigation. In order to reduce learning effects and nervous tension the test subjects familiarized themselves with the laboratory environment, the test equipment and the testing procedures on two occasions prior to actual data collection. Submaximal and maximal tests were conducted on each of two occasions for untrained individuals to establish the test-retest reliability. Submaximal tests were conducted prior to maximum tests. Trained individuals were measured on three occasions since they were in training and any variations in their ability throughout the test period needed evaluation.
FIGURE 13
48
Each type of test was conducted on a Quinton, QI-854, an electromechanically braked ergometer, which was calibrated prior to each study.107 A ramp test was used to determine VO2 max. Subjects completed ramp-slope increments of 16.6 W / min for the first 3 min of the test from a base line of 4 min zero load pedalling at 90 rpm, equivalent to 33 W, which was increased to 16.6 W/30 s until exhaustion. BMI was not calculated and no figures for height were reported. The correlation between VO2 max and those predicted from the Legge and Banister nomogram was significantly higher (r = 0.98) than those made from the original ÅstrandRyhming test (r = 0.80), an underestimation. See figure 14 for a plot of maximal oxygen intake calculated from the 1954 nomogram in relation to the direct determined maximum. The Åstrand-Ryhming test (1954) was completed during one of the test sessions, with a work rate sufficient to stimulate a steady state HR of ~ 150 bpm. Siconolfi et al (1982) modified the Åstrand-Ryhming test (1960) in order to make it more suitable for older, sedentary subjects.108 Initial exercise rates were changed and allowed older men and women to be tested at a lower initial work rate.109 An initial exercise rate of 24.5 W (150 kpm ⋅ min-1) was proposed for men 35 year and older and for women of all ages. For men under the age of 35 the initial exercise rate was 49.0 W (300 kpm ⋅ min-1). A calibrated
107
In the mechanically braked cycle ergometer the resistance to the wheel is based upon the tightening (friction)
of the belt that surrounds the flywheel. Mechanically braked ergometers provide a constant force, but a nonconstant power at varying pedal revolutions. Any change in pedal revolutions per minute will alter the work (w) and power (P). The electromechanically braked cycle ergometers do not require strict attention to pedal frequency. If the distance (D) changes because the pedal revolutions increase or decrease on a constant-power ergometer, the force (F) factor compensates electronically to maintain a constant power level. For further information about similarities and dissimilarities between mechanically braked cycle ergometers and electromechanically braked cycle ergometers see Adams, G. (2002) Exercise Physiology Laboratory manual, 4th edition. McGrawHill, New York, page 170-171. Legge and Banister assumed 25 % efficiency in cycle ergometry. 108
Siconolfi S.F., Cullinane E.M., Carleton R.A., Thompson P.D. (1982) “Assessing VO2 max in
epidemiological studies: a modification of the Åstrand-Ryhming test.” Med Sci Sports Exerc, 14 (5), page 335338. 109
Åstrand and Rodahl recommends a work rate of 50 watts for older persons or persons who are completely
untrained, see Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd Edition. New York: McGraw-Hill Book Company, page 370.
49
Philbin cycle ergometer was used. 113 healthy volunteers were used in the study; there were at least 10 men and women per age decade from 20-70 yr. FIGURE 14
The subjects were randomly assigned to one of two groups: a test group of 50 subjects and a validity group of 63 subjects. Volunteers on medication known to affect heart rate were excluded, as were individuals with heart diseases. Data from the test group was used to derive equations for predicting VO2 max by using multiple-regression analysis. Data from the validity group was used to test the validity of the derived equations by using the Pearson product-moment correlation, standard error of estimate, and t-ratios.110 The test subjects performed a submaximal cycle test followed by a 10-min rest period (agecorrection factors developed by I. Åstrand were not used). In order to test the subjects maximally they pedalled for 1 min with zero resistance. This was followed by an elevation of exercise rate by 24.5 W (150 kpm ⋅ min-1) each minute until the subject could not maintain a pedal rate of 50 rpm. BMI was not calculated and no relevant figures of weight/height were reported.
110
The t-test deals with problems associated with inference based on small samples: the calculated mean and
standard deviation may by chance deviate from the "real" mean and standard deviation.
50
The criterion of reaching max VO2 was when the final two VO2 measurements were less than 250 ml ⋅ min-1 and the respiratory quotient for the last measurement was greater than 1.0. The multiple regression equations derived from data of the test group were: •
For males:
y = 0.348 (X1) – 0.035 (X2) + 3.011 (r = 0.86; SEE = 0.359 liter ⋅ min-1)
•
For females:
y = 0.302 (X1) – 0.019 (X2) + 1.593 (r = 0.97; SEE = 0.199 liter ⋅ min-1)
where y is the VO2 (litre ⋅ min-1), X1, is the VO2 (litre ⋅ min-1) from the nomogram, not corrected for age), and X2 is the age in years. No significant differences were found between the directly measured VO2 max and the VO2 max estimated by Siconolfi et al. The correlation coefficients between measured and estimated VO2 max ranged from 0.92-0.93 for the age groups (see table 13 for correlation coefficients and standard error of estimates).
TABLE 13
51
Terry et al (1977) established a procedure for eliciting correct, i.e. relatively high, pulse rates for the Åstrand-Ryhming test (1954). 111 A heart rate of 160-165 bpm is often mentioned as optimal.112 Terry presented a nomogram for workload selection aiming to elicit such pulse rates (see figure 15). A calibrated Monark cycle ergometer was used. 60 male college freshmen113 enrolled in the physical education program at Texas A & M University participated in the study. The mean age was 19.7 years with an average height of 177.8 cm and an average weight of 70.9 kg resulting in a BMI of 22.4. Each subject performed the Balke Treadmill test of optimal work capacity in order to measure maximal oxygen intake. The test was ended at exhaustion. The values were used to establish two equal groups. One group of 30 students was randomly designated as the experimental group and the other as the cross validation group. Information about health and training status is lacking. The experimental group was tested to collect data on the independent variables of age, height, weight, resting heart rate, resting blood pressure, one minute exercise heart rate, leg length standing and on the cycle ergometer. By subjecting the experimental group to workload tests on the cycle ergometer the dependent variable of workload was obtained. A multiple regression equation utilizing age, weight and one minute exercise heart rate to predict workload was developed by subjecting the data to a stepwise regression analysis. A nomogram was developed from the regression equation. The application of the nomogram is relatively simple; after obtaining weight and the 600 kpm/min exercise heart rate one proceeds with entering these two quantities in the nomogram. The intersection determines the workload to be used when administering the Åstrand-Ryhming test/method. The correlation between predicted and criterion maximal oxygen intake values was r = 0.65. No comparison
111
Terry J.W., Tolson H., Johnson D.J., Jessup G.T. (1977) “A workload selection procedure for the Åstrand-
Ryhming test.” J Sports Med, 17 (4), page 361-363. 112
Ibid. Page 362. See also Jessup et al (1975) page 36-37 for a discussion about optimal workload intensity.
113
A freshman is a student in the first year at a college or university.
52
between direct determined maximum oxygen uptake and prediction of maximum oxygen uptake was presented. FIGURE 15
53
6. Discussion The aim of this study was: To interpret and evaluate scientific results about the internal and external validity as well as the reproducibility of the Åstrand-Ryhming test (1954) and the Åstrand-Ryhming test (1960) and, if motivated, suggest presumptive modifications. In accordance with the aim the following questions were formulated: 1. Is the test valid? 2. For whom is it valid? 3. Is it reproducible? 4. For whom is it reproducible? 5. Is there a scientific base for suggesting modifications of the test? According to P.-O. Åstrand the most important use for the Åstrand-Ryhming test/method is in testing the same individual on several separate occasions, for instance during a period of training or after an injury. By doing so it is possible to determine whether or not the circulatory training for NN has been effective.114 Åstrand and Ryhming and I. Åstrand have repeatedly declared that the Åstrand-Ryhming test/method is a screening, selecting or evaluating tool and that the validity - when testing individuals with characteristics different from those in the studies - is unknown. 115, 116 P.-O. Åstrand expresses in an interview from 2000 that the Åstrand-Ryhming test from 1960 is a more sophisticated tool than the test from 1954.117 With the test from 1960 I. Åstrand
114
Åstrand, P.-O. (1979) ”Work tests with the bicycle ergometer.” Varberg, page 19.
115
Åstrand, P.-O. & Ryhming, I. (1954) “A nomogram for calculation of aerobic capacity (physical fitness) from
pulse rate during submaximal work” J Appl Physiol, 7, page 218-221. 116
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 45-60. 117
Eriksson, M. & Larsson, B., (2001) ”Ergometri – konditionstest på cykel – en undersökning av hur
Åstrandstestet utvecklades och används idag”. C-essay from Idrottshögskolan, Stockholm. Appendix 1, question 11.
54
expanded the possibility to use the Åstrand-Ryhming test/method as a predictor of aerobic capacity with special reference to age and sex. Consequently, it is surprising that quite a few of the scrutinized articles deal with the Åstrand-Ryhming test from 1954, instead of with the modified nomogram from 1960. The difference between the two nomograms is larger with respect to lower loads. The use of lower loads is necessary, especially for sedentary or older persons (see table 14 for a comparison of the two nomograms).
TABLE 14, A comparison of the two nomograms
Nomogram
1954
1960
♂ (1200 kpm/min & 166 bpm)
3.60 l/min
3.60 l/min
♂ (900 kpm/min & 130 bpm)
4.15 l/min
4.10 l/min
♂ (600 kpm/min & 150 bpm)
2.20 l/min
2.25 l/min
♀ (900 kpm/min & 160 bpm)
3.20 l/min
3.10 l/min
♀ (600 kpm/min & 160 bpm)
2.20 l/min
2.25 l/min
♀ (450 kpm/min & 132 bpm)
2.40 l/min
2.75 l/min
Sex, workload, HR
The major problems with how the different studies have been conducted were: •
A possible use of inappropriate equipment or uncalibrated equipment.
•
That the first test with the each subject wasn’t excluded.118
•
An incorrect test protocol.
•
An insufficient arrangement/presentation of the derived values.
•
An unsatisfactory measure of reaching max VO2. 119
118
No one has excluded the results of the first test.
119
According to Åstrand and Rodahl (1986) there are two main criteria of reaching a subject’s maximal aerobic
power: (1) There is no further increase in oxygen uptake despite further increase in the rate of exercise (levelling off) and (2) the blood lactate concentration exceeds 8-9 mM. See Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd Edition. New York: McGraw-Hill Book Company, page 301.
55
6.1. Question 3: Is the test valid? Nine (out of 14) articles have primarily dealt with the validity of the Åstrand-Ryhming test/method.120 Of these nine four has focused on the original nomogram and four on the modified nomogram, the ninth article was conducted by Macsween (2001), who used a computer simulation of the Åstrand-Ryhming test/method and information about which nomogram the computer model focused on is lacking. It is important to be aware of that another selection method is likely to have resulted in the sorting out of different research articles and possibly, dissimilar or conflicting answers. The credibility of an article depends on the entirety of the study. The entirety of the study depends on if it have been conducted in a systematic and logical manner. Uncertainty and/or lack of key information will reduce the credibility of the article. In a diagram, comparing for example determined max VO2 with estimated max VO2, it is key that a line of identity, an ideal line, is drawn (e.g. figure 4). The line of identity represents perfect correlation (r = 1.0). With a line of identity included you can compare your regression line with the ideal line. If a line of identity is not included it is easy to be misled and assume that the presented values are better than they are (e.g. figure 9). Consequently you can have a high r-value but an incorrect (with respect to the ideal line) slope of the line (i.e. distant from the ideal line), implying over- or underestimation. Several authors have presented parameters such as standard deviation (SD) and correlation coefficients (r). These parameters are interesting, but blunt; the SD is based on average values and is merely correct with respect to the average values. The r-value, as mentioned earlier, is also an insufficient measure of a method’s validity. With this in respect merely three articles can be used when answering question 1: Teräslinna et al (1966), Jessup et al (1974) and Legge and Banister (1986). These articles have high credibility but have come to different conclusions about the Åstrand-Ryhming test/method (figure 16).
120
Legge and Banister (1986) also validated the Åstrand-Ryhming test (1954). However, the primary purpose
with their article was to develop a new nomogram.
56
FIGURE 16
A number of investigators have reported that the Åstrand-Ryhming test/method suffers from consistent underestimation, for references see the introduction in Legge and Banister (1986).121 Underestimation can depend on a disregard of the age correction factors or that the age correction factors are too low, something that I. Åstrand already stated in her thesis from 1960.122 Another possible explanation as to why some investigators have concluded that the ÅstrandRyhming test/method suffers from underestimation can possibly be found in the fact that, when constructing the original nomogram (and most likely the modified nomogram), the first 121
Legge B.J., Banister E.W. (1986) “The Åstrand-Ryhming nomogram revisited.” J Appl Physiol, 61 (3), page
1203-1209. 122
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 54.
57
determinations with each subject were excluded and not used for the calculations of values.123 A lower pulse rate, on the second test occasion (with the same workload) will result in a higher estimation of max VO2. Consequently, conducting only one test is likely to result in an underestimation. Åstrand & Rodahl examines in Textbook of work physiology the reliability of predicting maximal aerobic power from submaximal heart rate response to a given rate of exercise. The following factors might influence the prediction of maximal oxygen uptake from data recorded during submaximal tests: 124 •
The fact that there are many exceptions to the linear increase in heart rate with increase in oxygen uptake.
•
That maximal heart rate declines with age. If old and young people are included in the same study, the older subjects will be consistently overestimated compared with the younger. However, the correct use of the age correction factors will combat this problem.
•
The mechanical efficiency may vary + 6% (cycle ergometer). This means that subjects with a low mechanical efficiency will be underpredicted when measuring maximal oxygen uptake or vice versa.
Fear, enthusiasm and/or emotional stress may also affect the heart rate response at submaximal work. However, the heavier the work rate the less pronounced is the nervous effect on the heart rate. Therefore it is advisable that the test load should result in a heart rate of at least 150 bpm for younger subjects.125 Untrained persons are often underestimated; extremely well trained persons are often overestimated.126
123
In the study on draymen (1958), that was a part of designing the second nomogram, the first determinations
with each subject were excluded consequently. Thus, it is likely that this was done in all of the different studies that were a part of constructing the modified nomogram, something that is confirmed by I. Åstrand (23/6-2003) and P.-O. Åstrand (1/2-2004) in personal communication. 124
Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd
Edition. New York: McGraw-Hill Book Company, page 373-374. 125
Ibid, page 374-375.
126
Ibid, page 376.
58
The heart rate and VO2 responses to tests on the treadmill are different than those from the responses to tests with the cycle ergometer.127 The use of the treadmill interchangeably with the cycle ergometer is likely to reduce the accuracy of the predicted VO2 max.128 On the cycle ergometer the energy output can be predicted with greater accuracy than for any other type of exercise.129, 130 According to I. Åstrand the cycle ergometer is preferable to the treadmill, especially on older persons.131 In the original nomogram from 1954 the sole testing advice used when testing the subjects submaximally was the cycle ergometer. When testing the subjects maximally both the cycle ergometer and the treadmill were used. Thus, when evaluating the Åstrand-Ryhming test (1954), and determining the max VO2, it is of minor importance whether you choose to work on the cycle ergometer or on the treadmill. When constructing the modified nomogram from 1960 I. Åstrand decided only to use the cycle ergometer. Some data, from the original nomogram, was however added/included in the shaping of the modified nomogram. This means that some of the results, of which the modified nomogram is based upon, also derive from tests conducted on the treadmill. Nevertheless, the majority of the results are based upon results made on the cycle ergometer. Therefore, when evaluating the Åstrand-Ryhming test (1960), and determining max VO2, the use of the cycle ergometer is preferable.
127
Cink R.E., Thomas T.R. (1981). “Validity of the Åstrand-Ryhming nomogram for predicting maximal oxygen
intake.” Br J Sports Med, 15 (3), page 184. 128
Legge B.J., Banister E.W. (1986). “The Åstrand-Ryhming nomogram revisited.” J Appl Physiol, 61(3), page
1203. 129
Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd
Edition. New York: McGraw-Hill Book Company, page 363. 130
Neder, J.A., Nery, L.E., Andreoni, S., Sachs, A., Whipp, B.J. (2000). “Oxygen cost for cycling related to leg
mass in males and females, aged 20 to 80.” Int J Sports Med, 2000, 21, page 263. 131
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 59.
59
Teräslinna et al (1966) studied the Åstrand-Ryhming test (1960) using the recommended submaximal test protocol, which is of great importance.132 The cycle ergometer (Monark) was the sole testing equipment, however it is uncertain whether or not it was calibrated. For a comparison between direct determined maximum oxygen uptake and the prediction of maximum oxygen uptake see figure 10. Teräslinna et al (1966) stresses the use of the age correction factors. Jessup et al (1974) compared the Åstrand-Ryhming test (1954) with the determined maximal oxygen intake in young volunteer subjects.133 When testing the students submaximally a calibrated Monark and standard procedures were used. The maximal test was the Balke treadmill test of physical working capacity. For a comparison between determined maximal oxygen uptake and the prediction of maximum oxygen uptake see figure 9. Jessup’s study indicates that the Åstrand-Ryhming test (1954) results in underestimation. Legge and Banister (1986) aimed to improve the accuracy of the Åstrand-Ryhming test (1954) by developing a new nomogram (see figure 13 for the Legge-Banister nomogram). Males between the ages of 20-29 years took part in the study. Each type of test was conducted on a Quinton, QI-854, an electromechanically braked ergometer, which was calibrated prior to each study. The use of the Quinton ergometer reduces the credibility of the article. During one test session the Åstrand-Ryhming test (1954) was completed. See figure 14 for maximal oxygen intake calculated from the nomogram in relation to the direct determined maximum. Legge and Banister’s study (1986) indicates that the Åstrand-Ryhming test (1954) results in underestimation. The quality among cycle ergometers are varying, hence it is imperative to use an ergometer of equal quality as was done when the test/method was constructed. There are mainly two types of ergometers: The mechanically braked ones where the resistance to the flywheel is based upon the tightening (friction) of the belt that surrounds the flywheel. The mechanically braked 132
For further information about how the “normal” test is conducted see Åstrand, P.-O. & Rodahl, K. (1986)
“Textbook of work physiology. Physiological Bases of exercise.” 3rd Edition. New York: McGraw-Hill Book Company, page 370-371 or Åstrand, P.-O. (1979) ”Work tests with the bicycle ergometer.” Varberg. 133
Jessup G.T., Tolson H., Terry J.W. (1974) “Prediction of maximal oxygen intake from Åstrand-Ryhming test,
12-minute run, and anthropometric variables using stepwise multiple regression.” Am J Phys Med, 53 (4), page 200-207.
60
ergometers that provide a constant force, but a nonconstant power at varying pedal revolutions. Any change in pedal revolutions per minute will alter the work (W) and power (P). The other type of ergometer is the electromechanically braked cycle ergometer that doesn’t require strict attention to pedal frequency. If the distance (D) changes because the pedal revolutions increase or decrease on a constant-power ergometer, the force (F) factor compensates electronically to maintain a constant power level.134 The Monark ergometer is a mechanically braked ergometer. Teräslinna et al (1966) and Jessup et al (1974) used a Monark cycle ergometer. Legge and Banister (1986) used a Quinton cycle ergometer. Figure 17 was used when trying to determine whether or not the Åstrand-Ryhming test/method is a good predictor of a person’s aerobic capacity. The results from Teräslinna et al (1966) indicate that the probability is high that the Åstrand-Ryhming test (1960) is valid. Jessup et al (1974) and Legge and Banister (1986) focused on the Åstrand-Ryhming test (1954). According to them the test/method underestimates the aerobic capacity. Does the underestimation depend on that Jessup et al (1974) and Legge and Banister (1986) focused on the original nomogram instead of the modified nomogram? The difference between the two nomograms is not very large (table 14). Most likely Jessup et al (1974) and Legge and Banister (1986) would have reached approximately the same results, i.e. underestimation, regardless of which nomogram they had evaluated. Since the test subjects were young (they were more or less the same age as the test subjects who were used when designing the original nomogram), the influence of the age correction factors is expected to be negligible. It is more likely to believe that the underestimation detected by Jessup et al (1974) and by Legge and Banister (1986) depends on something else, e.g., whether or not the first determination was excluded.
134
For further information about similarities and dissimilarities between mechanically braked cycle ergometers
and electromechanically braked cycle ergometers see Adams, G. (2002) “Exercise Physiology Laboratory manual”, 4th edition. McGraw-Hill, New York, page 170-171.
61
The answer to question 1, based on the articles by Teräslinna et al (1966), Jessup et al (1974) and by Legge and Banister (1986), is that it is not possible to decide whether or not the Åstrand-Ryhming test/method is valid.
FIGURE 17
6.2. Question 4: For whom is it valid? The answer to question 2 is that it is not possible to decide for which population the ÅstrandRyhming test/method is valid.
6.3. Question 5: Is it reproducible? Two articles have dealt with the reproducibility of the Åstrand-Ryhming test/method. The article by Wisén et al (1995) has high credibility, the article by Williams (1975) has a few drawbacks, which reduces the trustworthiness of the article.135 As a consequence merely one article will be used when answering question 3, the article by Wisén et al (1995). Wisén et al compared two exercise tests on cycle - a computerized test versus the ÅstrandRyhming test/method. When evaluating the Åstrand-Ryhming test/method the original 135
No figures of mean weight and height were presented. The name of the cycle ergometer and possible
calibration is also lacking. Information about the subjects health status and if they were trained or untrained is also missing.
62
nomogram with age correction factors from 1960 were used. The reason for not using the modified nomogram is unclear. The aim of the study was to assess agreement and reproducibility between the VO2 max obtained from the Åstrand-Ryhming test/method and with the one of the computerized submaximal extrapolation tests (Cat Eye Ergociser EC 1500). A calibrated Monark ergometer and the recommended submaximal test protocol for the Åstrand-Ryhming test/method were used. The Cat Eye Ergociser EC 1500 was not possible to calibrate. 20 healthy women participated in the study.136 The women were divided into two groups of ten according to their age (mean age 35.3 and mean age 46.9 resulting in a total mean age of 41.1 years). Prior to each test the subjects were shown the equipment and informed about the testing procedures. Tests were performed over two consecutive days. Two tests, one with each method, were performed on each day with 2 h in between. Thus, each subject participated in four separate test sessions (test-retest with each method). The VO2 max estimated from the computerized test was lower compared with the VO2 max estimated from the Åstrand-Ryhming test/method. A statistically significant linear relationship was found, r = 0.85 (figure 12). Some of the estimates derived from the computerized test had large errors reducing the agreement between the two tests. The coefficient of reproducibility was 7.5 ml kg-1min-1 for the Åstrand-Ryhming test/method and 11.4 ml kg-1 min-1 for the computerized test (the average maximal VO2 intake was approximately 40 ml kg-1min-1 for the Åstrand-Ryhming test/method). Wisén et al concluded that the computerized method gave VO2 max estimates that were lower and more uncertain than the Åstrand-Ryhming test/method. Since intra-individual variations are higher at lower intensities reproducibility is likely to be higher at higher intensities.137 By replacing “low validity” and “high validity” with “low reproducibility” and “high reproducibility” figure 16 can be applied when answering question 3. 136
Four subjects were on medication: oestrogen 25 mg/daily (1 person), nasal budesonide (2 persons),
amitryptilin 25 mg/daily (1 person). 137
Terry J.W., Tolson H., Johnson D.J., Jessup G.T. (1977). “A workload selection procedure for the Åstrand-
Ryhming test.” J Sports Med, 17 (4), page 361-362.
63
The work by Wisén et al indicates that there is a possibility that the Åstrand-Ryhming test/method might be reproducible. It is however not, on the basis of merely one article possible to make a decision about whether or not the Åstrand-Ryhming test/method is reproducible.
6.4. Question 6: For whom is it reproducible? The answer to question 4 is that it is not possible to decide for whom the Åstrand-Ryhming test/method is reproducible.
6.5. Question 7: Is there a scientific base for suggesting modifications of the test? Before proceeding with answering question 5 it is necessary to take a quick look in the rear view mirror. Åstrand and Ryhming and I. Åstrand have declared that the Åstrand-Ryhming test/method is a screening, selecting or evaluating tool. It is possible that a number of the investigators might have misinterpreted the original aim, i.e. Kasch (1984) who investigated the validity of the Åstrand-Ryhming test/method (1960) and concluded (about the ÅstrandRyhming test/method and the Sjöstrand test): “…neither test is an adequate substitute for direct measurement of VO2 max and that great caution should be used in interpreting the results of such tests.” 138 None of the originators have stated that the Åstrand-Ryhming test/method is a substitute for a direct measurement of a persons max VO2. It is evident that a discrepancy exists between those who view the test as a substitute for a direct measurement of the aerobic capacity and those who view the test/method as a rough estimation of the same. One possible explanation to why there have been misunderstandings about the aim can possibly be found in the fact that the scientific interest in public health among researchers started quite early in Sweden. In order to be able to investigate large numbers of individuals a more blunt tool (than a direct measurement of max VO2) had to be accepted. This was different from the situation in many other countries where tests were conducted more scarcely and (always) had to be a 100 % accurate. When the Åstrand
138
Kasch, F.W. (1984). ” The validity of the Åstrand and Sjöstrand submaximal tests.” The Physician and
Sportsmedicine, page 47.
64
Ryhming test/method was “exported” a number of scientists might have misunderstood the original aim. In personal communication with I. Åstrand she expressed that one of the original thoughts with the nomogram was that it should be easy to handle and that too many correction factors might limit the use of the test/method.139 If you, erroneously, view the test as a substitute for a direct measurement of an individuals max VO2 you abandon the original idea. It can however be interesting to reflect about whether or not a scientific base for suggesting modifications of the test exists. As mentioned earlier, the heart rate and VO2 responses to tests on the treadmill are different than those from the responses to tests with the cycle ergometer.140 The use of the treadmill interchangeably with the cycle ergometer is likely to reduce the accuracy of the predicted VO2 max.141 When constructing the modified nomogram from 1960 I. Åstrand decided only to use the cycle ergometer. Some data, from the original nomogram, was however added/included in the shaping of the modified nomogram. This means that some of the results, of which the modified nomogram is based upon, also derive from tests conducted on the treadmill. By eliminating the maximal treadmill test data it is possible to improve estimations from the Åstrand-Ryhming test/method.142 A relatively high workload is required to generate the best estimate of the aerobic capacity. A heart rate up to, or above, 150 bpm is often being mentioned as optimal.143, 144 Intra-individual 139
Personal communication with I. Åstrand 23/6-2003.
140
Cink R.E., Thomas T.R. (1981). “Validity of the Åstrand-Ryhming nomogram for predicting maximal oxygen
intake.” Br J Sports Med, 15 (3), page 184. 141
Legge B.J., Banister E.W. (1986). “The Åstrand-Ryhming nomogram revisited.” J Appl Physiol, 61(3), page
1203. 142
Cink R.E., Thomas T.R. (1981). “Validity of the Åstrand-Ryhming nomogram for predicting maximal oxygen
intake.” Br J Sports Med, 15 (3), page 184. 143
Terry J.W., Tolson H., Johnson D.J., Jessup G.T. (1977). “A workload selection procedure for the Åstrand-
Ryhming test.” J Sports Med, 17 (4), page 362. 144
Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd
Edition. New York: McGraw-Hill Book Company, page 374-375.
65
variations are higher at lower intensities compared to at higher intensities, consequently, it is vital to try to establish a procedure that elicits a correct, i.e. a relatively high, heart rate.145 Terry et al (1977) developed a nomogram for workload selection (figure 15). The nomogram by Terry et al (1977) was developed from a multiple regression equation. The statistical figures obtained when using multiple regression equations/models are impressive; correlation coefficients of or above 0.90 are not unusual. However the applicability and practicability of these models are doubtful, especially when relating them to large groups. Using multiple regression equations/models can include several more or less sophisticated respiratory and cardiovascular variables. The method is not new, von Döbeln et al146 involved skeletal weight and Hermiston and Faulkner147 involved fat-free weight as variables. A disadvantage with the method is that it requires sophisticated laboratory equipment, highly skilled technicians and plenty of time. When the original nomogram was constructed the first test with each subject was excluded. 148 It is likely that this also was done whilst constructing the modified nomogram.149 P.-O. Åstrand says in an interview that when testing the subjects submaximally a higher pulse rate was detected in the first test compared to subsequent tests, probably a result of anxiety.150 A higher mechanical efficiency on the second test day is, according to I. Åstrand, due to lower tension among the test subjects.151
145
Terry J.W., Tolson H., Johnson D.J., Jessup G.T. (1977). “A workload selection procedure for the Åstrand-
Ryhming test.” J Sports Med, 17 (4), page 361-362. 146
von Döbeln W., Åstrand, I., Bergström, A. (1967) ”An analysis of age and other factors related to maximal
oxygen uptake” J Appl Physiol, 22, page 934-938. 147
Hermiston, R.T., Faulkner, J.A. (1971) ”Prediction of maximal oxygen intake by a stepwise regression
technique.” J Appl Physiol, 30, page 833-837. 148
Åstrand, P.-O. (1952) ”Experimental Studies of Physical Working Capacity in Relation to Sex and Age.”
Thesis. Munksgaard, Copenhagen, page 20. 149
This is confirmed by I. Åstrand (personal communication, 23/6-2003) and by P.-O. Åstrand (personal
communication 1/2-2004). In the study on draymen, that was a part of I. Åstrand´s thesis, (Åstrand, I., 1958), the results from the second work test were used. 150
Eriksson, M. & Larsson, B., (2001) ”Ergometri – konditionstest på cykel – en undersökning av hur
Åstrandstestet utvecklades och används idag”. C-essay from Idrottshögskolan, Stockholm, page 16-17. 151
Åstrand, I. (1958) “The physical work capacity of workers 50-64 years old.” Acta Physiologica
Scandinavica, 42, page 77.
66
This information is hard to find. As a consequence - anyone who reproduces the procedure described by Åstrand and Ryhming in Journal of Applied Physiology might be doing it inadequately. At any rate, it is clever that prior to the first real test advise each subject to conduct a ”pre-test” in order to become familiarized with the laboratory environment, the test equipment, and the testing procedures. By doing so you improve the accuracy of the estimations with respect to the determined aerobic capacity. A lower pulse rate, on the second test occasion (with the same workload) will result in a higher estimation of max VO2. Conducting merely one test is likely to result in an underestimation. When the Åstrand-Ryhming test/method was developed obesity was probably not as widespread as it is today. It is plausible that a relatively high BMI (obeseness) affects the correctness of the estimations of the aerobic capacity negatively. 152 For that reason an introduction of some kind of correction factor for obese individuals might be appropriate. However BMI as a single parameter cannot explain the underestimation observed by, for instance, Jessup et al (1974), since the test subjects in that investigation had almost identical BMI with the test subjects involved in designing the original nomogram. Neder et al (2000) studied the relationship between VO2 and leg mass (LM) and total body mass (TBM) in 71 randomly selected sedentary men and women aged 20–80 years. 153 Research had suggested that the O2 cost was linearly and positively related to LM during cycle ergometry. DEXA technique (Dual Energy X-ray Absorptiometry), allowed the LM to be accurately measured and apportioned into fat and bone free lean components. Neder et al found that gross VO2 and gross efficiency were more strongly related to LM than TBM. It was concluded that LM provides the preferred frame of reference for predicting oxygen cost for cycle ergometry at 60 rpm in sedentary subjects, independent of age or gender.
152
Kasch (1984) reported a correlation coefficient (r) of 0.58; his population had a BMI of 24.5. Jetté (1979)
reported a correlation coefficient (r) of 0.54; her population had a BMI of 25.6. See table 4B for information about BMI of the test subjects in the different studies including those involved in designing the nomogram from 1954 and from 1960. 153
Neder, J.A., Nery, L.E., Andreoni, S., Sachs., Whipp, B.J. “Oxygen cost for cycling related to leg mass in
males and females, aged 20 to 80.” Int J Sports Med, 2000; 21, page 263-269.
67
Since HRmax declines with age, I. Åstrand introduced age correction factors 1960 (figure 7). It is possible that those are too low, causing underestimation.154 The material used for determining the age correction factor is relatively small, thus a common factor for both sexes was recommended.155 It would be interesting to investigate a larger number of subjects in order to find out whether or not separate correction factors for men and women would result in more exact estimations. In some of his investigations P.-O. Åstrand has observed that females were able to reach maximal pulses higher than males, however the difference is not significant.156 It is nevertheless, according to P.-O. Åstrand, a fact that a small heart can beat faster than a large heart. In general, the size of the female heart is smaller than the size of the male heart. Considering the arguments above there seems to be a base for suggesting modifications of the test/method in order to improve the accuracy of the estimations: 1. An elimination of the maximal treadmill test data (which is integrated in the 1960 nomogram). 2. A recommendation of new guidelines for the selection of workload (in order to elicit relatively high pulse rates). 3. The introduction of a pre-test prior to the initial true test in order for the test subjects to get acquainted with the laboratory environment, the test equipment, and the testing procedures. 4. To introduce a correction factor for obese individuals. 5. The construction of (a) new age correction factors and (b) separate age correction factors for the two sexes.157 Legge and Banister (1986) aimed to improve the accuracy of the Åstrand-Ryhming test (1954) by developing a new nomogram, based on the linear relationship between VO2 and ∆ HR, the later being defined as the elevation of exercise heart rate above that reached during
154
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age.” Acta
Physiologica Scandinavica, vol. 49, supplementum 169. Thesis. Stockholm 1960, page 49. 155
Ibid, page 54-55.
156
Personal communication with Åstrand, P.-O., 1/2-2004.
157
This modification (nr 5) does not derive from the research articles. It derives from I. Åstrand´s thesis.
68
zero-load pedalling at 90 rpm (see figure 13 for the Legge and Banister nomogram).158 The nomogram developed by Legge and Banister is truly different from the Åstrand-Ryhming test (1960) and it is difficult to view it as a modification. It is fairer to consider the L-B nomogram as an alternate nomogram/test and not as a basis for suggesting modifications of the ÅstrandRyhming test/method. In order to estimate VO2 max several tests uses the heart rate response to exercise based on the 220-age equation. Current research questions the estimation of HRmax using the abovenamed equation (the SD for HRmax within an age group is + 10 bpm).159 The equation is often found in textbooks without explanations or references. Furthermore, the formula is not developed from original research; it is developed from observations based on data from approximately 11 references consisting of published research or unpublished scientific compilations. “Consequently, the formula HRmax =220-age has no scientific merit for use in exercise physiology and related fields.”160 If HRmax needs to be estimated, the most accurate general equation is that of Inbar (205.8-(0.685 x age)).161
158
Legge B.J., Banister E.W. (1986) “The Åstrand-Ryhming nomogram revisited.” J Appl Physiol, 61 (3), page
1203-1209. 159
Robergs, A.R., Landwehr, R. (2002) “The surprising history of the 220-age equation.“ Journal of Exercise
Physiology online. Volume 5, May 2002, page 1-10. 160
Ibid, page 1.
161
Ibid, page 7.
69
7. Conclusion The aim of this study was to interpret and evaluate scientific results about the internal and external validity as well as the reproducibility of the Åstrand-Ryhming test (1954) and the Åstrand-Ryhming test (1960) and, if motivated, suggest presumptive modifications. In accordance with the aim the following questions were formulated: (1) Is the test valid? (2) For whom is it valid? (3) Is it reproducible? (4) For whom is it reproducible? (5) Is there a scientific base for suggesting modifications of the test?
Based on the limited number of articles that were a part of this review it is not possible to decide whether or not the Åstrand-Ryhming test/method is valid. Consequently it is not possible to decide for which population the test is valid. It is additionally not possible to decide if the Åstrand-Ryhming test/method is reproducible or for whom the test is reproducible.
It is most likely possible to improve the accuracy of the estimations by means of modifying the Åstrand-Ryhming test/method. Suggestions of modifications include: (1) An elimination of the maximal treadmill test data (which is integrated in the 1960 nomogram), (2) A recommendation of new guidelines for the selection of workload, (3) That each test subject, prior to the initial test, conducts a pre-test, in order to get acquainted with the laboratory environment, the test equipment, and the testing procedures. (4) The introduction of a correction factor for obese individuals and (5) The construction of (a) new age correction factors and (b) separate age correction factors for the two sexes.
70
References Adams, G. (2002) “Exercise Physiology Laboratory manual”, 4th edition, New York: McGraw-Hill Book Company.
Cink R.E., Thomas T.R. (1981) “Validity of the Åstrand-Ryhming nomogram for predicting maximal oxygen intake.” Br J Sports Med, 15 (3), page 182-185.
Cullinane E.M., Siconolfi S., Carleton R.A., Thompson P.D. (1988) “Modification of the Åstrand-Ryhming submaximal bicycle test for estimating VO2 max of inactive men and women.” Med Sci Sports Exerc, 20 (3), page 317-318.
Eriksson, M. & Larsson, B. (2001) ”Ergometri – konditionstest på cykel. En undersökning av hur Åstrandstestet utvecklades och används idag.” C-essay from Idrottshögskolan, Stockholm.
Hermiston, R.T., Faulkner, J.A. (1971) ”Prediction of maximal oxygen intake by a stepwise regression technique.” J Appl Physiol, 30, page 833-837.
Jessup G.T., Riggs C.E., Lambert J., Miller W.D. (1977) “The effect of pedalling speed on the validity of the Åstrand-Ryhming aerobic work capacity test.” J Sports Med, 17 (4), page 367371.
Jessup G.T., Terry J.W., Landiss C.W. (1975) “Prediction of workload for the ÅstrandRyhming test using stepwise multiple linear regression.” J Sports Med, 15 (1), page 37-42.
Jessup G.T., Tolson H, Terry J.W. (1974) “Prediction of maximal oxygen intake from Åstrand-Ryhming test, 12-minute run, and anthropometric variables using stepwise multiple regression. Am J Phys Med, 53(4), page 200-207.
Jette M. (1979) “A comparison between predicted VO2 max from the Åstrand procedure and the Canadian Home Fitness Test.” Can J Appl Sport Sci, 4 (3), page 214-218.
Kasch, F.W. (1984) “The validity of the Åstrand and Sjöstrand submaximal tests.” The Physician and Sports medicine, page 47-51, 54. 71
Legge B.J., Banister E.W. (1986) “The Åstrand-Ryhming nomogram revisited.” J Appl Physiol, 61(3), page 1203-1209.
Londeree, B.R., Moeschbeger, B.L. (1982) ”Effect of age and other factors on maximal heart rate.” Res Q, 53, page 297-304.
Macsween A. (2001) “The reliability and validity of the Åstrand nomogram and linear extrapolation for deriving VO2 max from submaximal exercise data.” J Sports Med, 41 (3), page 312-317.
Neder, J.A., Nery, L.E., Andreoni, S., Sachs, A., Whipp, B.J. (2000) “Oxygen cost for cycling related to leg mass in males and females, aged 20 to 80.” Int J Sports Med, 21, page 263-269.
Robergs, A.R., Landwehr, R. (2002) “The surprising history of the 220 – age equation.“ Journal of Exercise Physiology online. Volume 5, May 2002, page 1-10.
Ryhming, I. (1953) “A modified Harvard Step test for the evaluation of physical fitness.” Arbeitsphysiologie, 15, page 235-250.
Schantz, P.G. (2004) Unpublished data.
Siconolfi S.F., Cullinane E.M., Carleton R.A., Thompson P.D. (1982) ”Assessing VO2 max in epidemiological studies: a modification of the Åstrand-Ryhming test.” Med Sci Sports Exerc, 14 (5), page 335-338.
Teräslinna P., Ismail A.H., MacLeod D.F. (1966) ”Nomogram by Åstrand and Ryhming as a predictor of maximum oxygen intake.” J Appl Physiol, 21(2), page 513-515.
Terry J.W., Tolson H., Johnson D.J., Jessup G.T. (1977) “A workload selection procedure for the Åstrand-Ryhming test.” J Sports Med, 17 (4), page 361-36.
von Döbeln, W., Åstrand, I., Bergström, A. (1967) ”An analysis of age and other factors related to maximal oxygen uptake.” J Appl Physiol, 22, page 934-938.
72
Wahlund, H. (1948) “Determination of the physical working capacity.” Acta Med Scandinavica, suppl 215. Thesis, Stockholm.
Williams, L. (1975) “Reliability of predicting maximal oxygen intake using the ÅstrandRyhming nomogram.” Res Quart, 46, page 12-16.
Wisén A.G., Wohlfart B. (1995) “A comparison between two exercise tests on cycle; a computerized test versus the Åstrand test.” Clin Physiol, 15 (1), page 91-102.
Åstrand, I. (1960) “Aerobic work capacity in men and women with special reference to age”. Acta Physiologica Scandinavica, vol. 49, suppl 169. Thesis, Stockholm.
Åstrand, I. (1958) “The physical work capacity of workers 50-64 years old.” Acta Physiologica Scandinavica 42, page 73-86.
Åstrand, P.-O. (1952) ”Experimental Studies of Physical Working Capacity in Relation to Sex and Age.” Thesis, Copenhagen.
Åstrand, P.-O. (1979) ”Work tests with the bicycle ergometer.” Varberg.
Åstrand, P.-O. & Rodahl, K. (1986) “Textbook of work physiology. Physiological Bases of exercise.” 3rd Edition, New York: McGraw-Hill Book Company.
Åstrand, P.-O. & Ryhming, I. (1954) “A nomogram for calculation of aerobic capacity (physical fitness) from pulse rate during submaximal work.” J Appl Physiol, 7, page 218-221.
Electronic references
Gunnarson R. Validitet och reliabilitet [Dept of Prim Health Care Göteborg University Research methodology web site]. March 13, 2002. Available at: http://infovoice.se/fou. Accessed June 26, 2004. Journal of Exercise Physiology online: http://www.css.edu/users/tboone2/asep/fldr/fldr.htm. Accessed Oct 13, 2003.
73
Online Conversion: http://www.onlineconversion.com. Accessed July 1, 2003. PubMed: http://www.ncbi.nlm.nih.gov/PubMed/. Accessed Jan 22, 2003 – Jan 30, 2003.
Personal communication
Personal communication with Åstrand, I. (23/6-2003). Personal communication with Åstrand, P.-O. (23/6-2003). Personal communication with Åstrand, P.-O. (1/2-2004).
74
![[DIPT. : SYRPHIDAE] FOR THE CONTROL OF APHIDS UNDER GLASS](https://kipdf.com/img/300x300/dipt-syrphidae-for-the-control-of-aphids-under-gla_5ac214291723dd0be6669122.jpg)
