Secondary Logo

Share this article on:

Self-reported symptoms and exercise-induced asthma in the elite athlete


Medicine and Science in Sports and Exercise: February 2001 - Volume 33 - Issue 2 - p 208-213
CLINICAL SCIENCES: Clinical Investigations

RUNDELL, K. W., J. IM, L. B. MAYERS, R. L. WILBER, L. SZMEDRA, and H. R. SCHMITZ. Self-reported symptoms and exercise-induced asthma in the elite athlete. Med. Sci. Sports Exerc., Vol. 33, No. 2, 2001, pp. 208–213.

Purpose: The purpose of this study was to compare self-reported symptoms for exercise-induced asthma (EIA) to postexercise challenge pulmonary function test results in elite athletes.

Methods: Elite athletes (N = 158; 83 men and 75 women; age: 22 ± 4.4 yr) performed pre- and post-exercise spirometry and were grouped according to postexercise pulmonary function decrements (PFT-positive, PFT-borderline, and PFT-normal for EIA). Before the sport/environment specific exercise challenge, subjects completed an EIA symptoms-specific questionnaire.

Results: Resting FEV1 values were above predicted values (114–121%) and not different between groups. Twenty-six percent of the study population demonstrated >10% postexercise drop in FEV1 and 29% reported two or more symptoms. However, the proportion of PFT-positive and PFT-normal athletes reporting two or more symptoms was not different (39% vs. 41%). Postrace cough was the most reported symptom, reported significantly more frequently for PFT-positive athletes (P < 0.05). Sensitivity/specificity analysis demonstrated a lack of effectiveness of self-reported symptoms to identify PFT-positive or exclude PFT-normal athletes. Postexercise lower limit reference ranges (MN-2SDs) were determined from normal athletes for FEV1, FEF25–75% and PEF to be −7%, −12.5%, and −18%, respectively.

Conclusion: Although questionnaires provide reasonable estimates of EIA prevalence among elite cold-weather athletes, the use of self-reported symptoms for EIA diagnosis in this population will likely yield high frequencies of both false positive and false negative results. Diagnosis should include spirometry using an exercise/environment specific challenge in combination with the athlete’s history of asthma symptoms.

Sports Science and Technology Division, United States Olympic Committee, Lake Placid, NY 12946

March 2000

May 2000

The prevalence of exercise-induced asthma (EIA) has been estimated to be 4–20% in the general population and 11–50% in specific athlete populations (2,9,10,16,30,32). This variability may be due to specific environmental demands on the airways of athletes and/or the criteria used for diagnosis. EIA in athletes is typically gauged by self-reported symptoms (12,16) and/or airway responsiveness assessed using pharmacological (3,16,26) or exercise (18,22,23,26,32) challenges. Several investigators have relied on the methacholine challenge (3,16) to evaluate EIA, although it has been shown that between 17% and 73% who respond as positive to an inhalation challenge do not test positive to an exercise challenge (26). Others have recommended a laboratory exercise challenge of 60–80% of maximal heart rate (4,17,26,27). Still others (18,22,23) have used a sport-specific high-intensity exercise challenge for EIA assessment.

Few studies (12,16,21,23) have evaluated whether or not symptoms always occur with EIA among the elite athlete population, yet self-reported symptoms are often the basis for a diagnosis of EIA (6,25). Currently, there are no standardized guidelines for EIA pulmonary function testing of elite athletes. Even among the nonathletic population, there is no universally accepted postexercise lower limit of pulmonary function change between EIA positive and normal individuals. Cut-off criteria for exercise-induced change in FEV1 or PEF has ranged from 10% to 20%(1,2,7,8,20,24,27,28). However, some investigators have used more objective criteria by determining the cut-off as the mean plus 2 SDs obtained from nonasthmatic subjects (13,14,23). By using this method, a fall of 6–7% in FEV1 is considered abnormal in the elite athlete population (13,14,23).

We have recently demonstrated that exercise intensity in cold dry air and not exercise duration is critical to effective diagnosis of EIA (23). In that study, the postexercise fall in FEV1 between sport-specific “field” exercise challenges of less than 2 min and of over 1 h were not different. Moreover, exercise at a similar intensity during an 8-min laboratory exercise challenge at 21°C, 50% RH resulted in 78% of “field” test positive athletes testing normal for EIA. We believe that the sport-specific exercise challenge is most sensitive and specific for broncho-provocation in the elite athlete population. Not only is it the most meaningful evaluation because it simulates the training and competition environment, it requires a near-maximal effort that may be necessary to trigger EIA in elite athletes. Therefore, the purpose of this study was to evaluate questionnaire data and pulmonary function test results in elite athletes (N = 158) who performed a sport-specific exercise challenge, to determine the efficacy of using self-reported symptoms for EIA diagnosis.

Back to Top | Article Outline



Elite athletes (N = 158; 83 men and 75 women; age: 22 ± 4.4 yr) provided written informed consent to participate in this study. The Institutional Review Board of the Sport Science and Technology Division of the United States Olympic Committee (USOC) approved experimental procedures. Seventy-eight (49.4%) subjects were Olympians, 49 others were members of World Championship Teams, and the remaining 31 were top developmental athletes in their respective sports. Subject sports were biathlon (N = 43), canoe/kayak (N = 35), cross-country skiing (N = 21), women’s ice hockey (N = 29), Nordic combined (N = 7), and speed skating (N = 23). The athletes were evaluated for the presence of EIA using pre- and post- pulmonary function tests (PFT) with a sport/environment specific “field” exercise challenge (race or race simulation). Before testing, each subject completed and signed a questionnaire that queried the presence or absence of four common symptoms of EIA in the athlete population; coughing, wheezing, chest tightness/trouble breathing (dyspnea), and excessive mucus formation.

Back to Top | Article Outline

Pulmonary function test procedures.

Actual competition or simulated competition was the sport-specific exercise challenge for EIA provocation. Competitions included Olympic Trials, World Team Trials, World Cup Competition, and U.S. National Championships. Time duration for the exercise challenge was dependent upon event duration and ranged from approximately 1 min 20 s for speed skaters to over 1 h for cross-country skiers. Temperature (−20 to +4°C) and humidity (<40% RH) were not controlled during the exercise challenge, but conditions were cold and dry, and generally representative of the ordinary environmental challenge faced by these athletes. Exercise intensity for all trials was competition specific and at competition pace.

Standard spirometry evaluations were performed pre- and post-exercise challenge using a calibrated computerized 10-L rolling dry-seal spirometer (Model 2130, SensorMedics, Yorba Linda, CA). Before each exercise challenge and warm-up, baseline spirometry was performed by obtaining three consistent trials that involved the following procedure: 1) three normal tidal volume breaths; 2) maximal inhalation; 3) forced maximal exhalation; and 4) maximal inhalation. The best-of-three trials were used for analysis. After baseline spirometry was obtained, the athletes followed their usual warm-up schedule and completed the exercise challenge. Subsequently, they performed pulmonary function tests (PFT), at 5, 10, and 15 min postexercise (23). Pulmonary function decrements were determined by subtracting each postexercise value of FVC, FEV1, FEF25–75%, and PEF from best-of-three preexercise baseline values, dividing by baseline values, and multiplying by 100. Decrements of greater than 10% in FEV1 were considered as positive indications of EIA (grouped as PFT+). Reductions from baseline values of >7% but less than 10% in FEV1 and/or 15% in FEF25–75%, and/or 10% in PEF were considered borderline for EIA (PFT-B). Previous studies (13,14,23) have suggested that a lower limit reference (mean plus 2 SDs from asymptomatic normal athletes) for FEV1 for elite athletes should be >6.5% decrement in FEV1. Subjects who did not meet these criteria were grouped as normal (PFT-N). Prediction equations for resting lung functions (FVC, FEV1, FEF25%-75%, and PEF) used were incorporated in the SensorMedics computer software (15).

Back to Top | Article Outline

Statistical analysis.

Statistical comparisons between groups (PFT+, PFT-B, PFT-N, with [s] or without [ns] self-reported symptoms) for PFT criterion variables were made by ANOVA. Differences in reported symptoms between groups were determined using the chi-square test. Sensitivity and specificity tests were done to determine the predictive value of self-reported symptoms for identifying true EIA positives (PFT+) and excluding true EIA negatives (PFT-N). For all statistical comparisons, the level of significance was set as P < 0.05.

Back to Top | Article Outline


Preexercise pulmonary functions for all groups (PFT+, PFT-B, PFT-N, s, and ns) were above predicted values (Table 1). No differences were noted for percent predicted values between or within groups.

Table 1

Table 1

Figure 1 depicts postexercise pulmonary function values (expressed as percent change from baseline spirometry values) for PFT+, PFT-B, and PFT-N athletes. Data are presented as the greatest postexercise fall from preexercise baseline spirometry values (expressed as percent difference). No differences in postexercise reduction for any pulmonary functions were noted within groups between subjects who reported symptoms and those who did not report symptoms (Table 2). For all pulmonary functions, significant differences were found between groups (P < 0.05).

Figure 1

Figure 1

Table 2

Table 2

Table 3 (Panel A) lists self-reported symptoms for each group. The proportion “postrace” coughing was significantly higher for the PFT+ group than for PFT-B and PFT-N groups (P < 0.05). All PFT+ athletes who reported symptoms had a postrace cough. No other differences for self-reported symptoms between groups were noted. For all groups, postrace cough was reported more frequently than the other symptoms (P < 0.05). Panel B (Table 3) lists the number of symptoms reported by each group. Two symptoms were reported proportionally more among PFT+ athletes than among PFT-N athletes (P < 0.05). No other differences in the number of reported symptoms were noted between groups.

Table 3

Table 3

The percent of athletes within groups who reported at least one symptom was not different between groups [61% (25/41), 57% (17/30), and 45% (39/87) for PFT+, PFT-B, and PFT-N, respectively]. However, of all athletes who reported symptoms, one or more symptoms were reported significantly more for PFT-N than for PFT+ and PFT-B, and PFT+ athletes reported more often than PFT-B (Table 4, P < 0.05). No difference was evident between PFT+ and PFT-N who reported two or more symptoms, but both reported more frequently than PFT-B (P < 0.05).

Table 4

Table 4

Table 5 provides an index of the failure of self-reported symptoms by elite athletes for EIA screening. Table 5 shows data from PFT+ and PFT-N athletes, but results do not differ when PFT-B athletes were used in the analysis (data not shown). The proportion of true diagnosis (a match between symptoms and PFT test results), the proportion of symptom positives that are PFT+ (sensitivity), or the proportion of symptom negatives that are PFT-N (specificity) demonstrate that self-reported symptoms do not provide accurate EIA diagnoses.

Table 5

Table 5

Spirometry values of PFT-N subjects were used to determine lower limit PFT reference ranges for the elite athlete population (Table 6). The mean maximal exercise-induced change in pulmonary functions plus 2 SDs was established as lower limit reference ranges for FVC, FEV1, FEF25–75%, and PEF. Variability within groups for FEF25–75% and PEF (except for PFT-N (s), PEF) was significantly higher than that of FVC and FEV1 (P < 0.05).

Table 6

Table 6

Back to Top | Article Outline


A standardized questionnaire is typically used to estimate the prevalence of asthma (6,25) and has been applied extensively with elite athletes (1,16,21,30,31). Additionally, symptoms are often the basis for the diagnosis and treatment of EIA. Our data suggest that questionnaires may be useful for epidemiological studies when two or more symptoms are reported but are not specific enough for diagnosis. Twenty-six percent of our study population demonstrated a >10% postexercise reduction in FEV1 and 29% reported two or more symptoms. However, of the 26% PFT+ athletes, only 44% (18 of 41) reported two or more symptoms. Although this is significantly greater than the 22% (19 of 87) PFT-N athletes who reported two or more symptoms (P < 0.05), the proportion among the symptom reporting population is not different (39% vs. 41%; for PFT+ and PFT-N, respectively). This suggests that symptom-based diagnosis of EIA is not more accurate than a coin toss. The lack of high sensitivity or specificity demonstrates that self-reported symptoms are poor indices of pulmonary dysfunction. Diagnosis should include spirometry with an exercise/environment-specific challenge in combination with the athlete’s history of asthma symptoms.

We grouped athletes into three categories, based on PFT results. Those athletes who met the accepted cut-off criteria of >10% postexercise reduction in FEV1 were defined as PFT+(1,2,4,7,8,19,24,27,28). Athletes who demonstrated decrements of >15% for FEF25–75%(1,4,17), and/or >10% for PEF (2,8,19,27,28), and/or >7% (but less than 10%) for FEV1 (13,14,23), were defined as PFT-B. Athletes with normal postexercise spirometry were defined as PFT-N. Formation of the PFT-B group was based on established criteria for FEF25–75% and PEF and recently published work defining a cut-off for FEV1 as a reduction greater than the mean postexercise reduction + 2 SDs (13,14,23). Studies (13,14) using elite Finnish runners defined FEV1 lower limit cut-off criteria for probable EIA at 6.5% (MN+2SDs from a nonatopic athlete population). This value is similar to the 7% value we defined from our PFT-N group and not different than we have recently found in a smaller population of elite cold weather athletes (23). The athletes in our PFT-B group clearly demonstrate pulmonary flow dysfunction that could affect athletic performance (5,29). Whether or not seasonal variation in bronchial responsiveness would shift subjects in this group to PFT+ or PFT-N remains to be determined. Heir (11) suggested that fluctuations in bronchial responsiveness are related to seasonal variations in cross-country skiers.

The prevalence of 26% PFT+ athletes in our study population is consistent with other studies using elite cold weather athletes as subjects (18,22). By combining the PFT+ group with the PFT-B group, the prevalence of EIA for our study population would be remarkably high at 45%. However, this is not different than what we have recently reported for specific elite athlete populations using a sport/environment specific exercise challenge (32). We found that 43% of U.S. Olympic short-track speed skaters and 50% of U.S. Olympic cross-country skiers had a greater than 10% postexercise reduction in FEV1 (32). The use of the sport/environment specific “field” exercise challenge to provoke broncho-responsiveness is not novel to our work. Mannix et al. (18) identified a 35% and Provost-Craig et al. (22) a 30% prevalence of EIA among elite figure skaters based on a >10% reduction in FEV1 after a sport/environment specific exercise challenge. We (23) have recently shown that within the elite athlete population, 78% of “field” challenge PFT+ athletes were PFT-N after a similar intensity treadmill run challenge in the laboratory environment (21°C, 50% RH).

Baseline spirometry in this population was 15% to >20% above predicted values for the general population. Resting spirometry did not provide insight to airway abnormalities. This is not surprising because even in the frank asthmatic, resting spirometry is often normal (>90% predicted) (5). The proportion of athletes who demonstrated baseline FEV1 values below 100% predicted was not different between PFT+ and PFT-N groups (4.9% vs. 5.8%). FEV1 baseline values below 100% predicted for PFT+ and PFT-N were not different (93 ± 1.4% vs. 92 ± 5%) but did approach 2 SDs away from mean values and as such would be considered abnormal.

A survey by Heir and Oseid (12) found that the prevalence of asthma “as diagnosed by a physician” was 14% among 153 cross-country skiers versus 5% among 306 matched controls. They also reported that 86% of the skiers and 35% of the control subjects reported at least one symptom (chest tightness, shortness of breath, cough, wheezing, or excess mucus production) (12). In our study, approximately 51% of this elite athlete population (N = 158) and 62% of the population of cross-country skiers (N = 21) reported at least one symptom. The frequency of two or more, three or more, and four symptoms reported by the skiers in the Heir and Oseid study (11) was 58%, 34%, and 18%, respectively. We found respective frequencies of only 29%, 13%, and 3% for our study population, and 48%, 19%, and 5% among our cross-country skiers. Heir and Oseid (11) also found that coughing was the most reported symptom among skiers and control subjects. However, unlike our study, where coughing was the most common symptom among EIA+ athletes, every symptom but coughing was more frequent in their subjects with asthma (11). It is interesting to note that among the elite athletes in our study, the PFT-N athletes reported one or more symptoms significantly more than the PFT+ athletes and reported two or more symptoms as frequently (cf. Table 3). Even if the PFT-B group was combined with the PFT+ group, no significant difference was noted from PFT-N for frequency of two or more reported symptoms (59% vs. 41%).

Although we only reported sensitivity and specificity between PFT+ and PFT-N, all combinations using the PFT-B group were preformed with similar results. Our data show that a positive symptom is insensitive to detecting PFT defined EIA and a negative symptom is not specific. The highest sensitivity found was 61% and the highest specificity observed was 85%. To be of diagnostic value, sensitivities above 85% and specificities greater than 92% are necessary. In this light, self-reported symptoms by our study population neither identify true positives nor exclude true negatives.

In conclusion, our data presents an interesting clinical dilemma in that the physician typically relies heavily on self-reported symptoms for diagnosis, yet we have shown that this is not appropriate with elite athletes. The high prevalence of EIA among elite cold weather athletes indicates a need for appropriate diagnostic procedures, which may not be readily available to the general practitioner. In this population, reasonable estimates of the prevalence of EIA can be made using self-reported symptoms, but diagnosis without pulmonary function testing using a sport/environment specific exercise challenge will yield high frequencies of both false positive and false negative results. Future studies with elite athletes should focus on the efficacy of using a more detailed patient history to improve EIA diagnosis.

This study was supported by the United States Olympic Committee.

The views, opinions, and findings contained in this report are those of the authors and should not be construed as an official United States Olympic Committee position.

The authors would like to express gratitude to the athletes who participated in this study.

Address for correspondence: Kenneth W. Rundell, Ph.D., Sports Science & Technology Division, United States Olympic Committee at Lake Placid, 421 Old Military Rd., Lake Placid, NY 12946; E-mail:

Back to Top | Article Outline


1. American Thoracic Society. Lung function testing: selection of reference values and interpretive strategies. Am. Rev. Respir. Dis. 144: 1202–1218, 1991.
2. Anderson, S. D., N. M. Connolly, and S. Godfrey. Comparison of bronchoconstriction induced by cycling and running. Thorax 26: 396–401, 1971.
3. Avital A., C. Springer, E. Bar-Yishay, and S. Godfrey. Adenosine, methacholine, and exercise challenges in children with asthma or pediatric chronic obstructive pulmonary disease. Thorax 50: 511–516, 1995.
4. Bar-Or, O. Pediatric sports medicine for the practitioner. In:Physiologic Principles to Clinical Applications. New York: Springer-Verlag, 1983, pp. 88–125.
5. Beck, K. C., K. P. Offord, and P. D. Scanlon. Bronchoconstriction occurring during exercise in asthmatic subjects. Am. J. Respir. Crit. Care Med. 149: 352–375, 1994.
6. Charpin, D., D. Vercloet, and J. Charpin. Epidemiology of asthma in western Europe. Allergy 43: 481–492, 1988.
7. Deal, E. C., E. R. Mcfadden, Jr., R. H. Ingram, F. J. Breslin, and J. J. Jaeger. Airway responsiveness to cold air and hyperpnea in normal subjects and in those with hay fever and asthma. Am. Rev. Resp. Dis. 121: 621–628, 1980.
8. Eggleston, P. A., R. R. Rosenthal, S. A. Anderson, et al. Guidelines for the methodology of exercise challenge testing of asthmatics. J. Allergy Clin. Immunol. 64: 642–645, 1979.
9. Haby, M. M., S. D. Anderson, J. K. Peat, C. M. Mellis, B. G. Toelle, and A. J. Woolcock. An exercise challenge protocol for epidemiological studies of asthma in children: comparison with histamine challenges. Eur. Respir. J. 7: 431–449, 1994.
10. Haby, M. M., J. K. Peat, C. M. Mellis, S. D. Anderson, and A. J. Woolcock. An exercise challenge for epidemiological studies of asthma in children: validity and repeatability. Eur. Respir. J. 8: 729–736, 1995.
11. Heir, T. Longitudinal variations in bronchial responsiveness in cross-country skiers and control subjects. Scand. J. Med. Sci. Sports. 4: 134–139, 1994.
12. Heir, T., and S. Oseid. Self-reported asthma and exercise-induced asthma symptoms in high-level competitive cross-country skiers. Scand. J. Med. Sci. Sports 4: 128–133, 1994.
13. Helenius, I. J., H. O. Tikkanen, and T. Haahtela. Exercise-induced bronchospasm at low temperature in elite runners. Thorax 51: 628–629, 1996.
14. Helenius, I. J., H. O. Tikkanen, and T. Haahtela. Occurrence of exercise-induced bronchospasm in elite runners: dependence on atopy and exposure to cold air and pollen. Br. J. Sports Med. 32: 125–129, 1998.
15. Knudson R. J., R. C. Slatin, M. D. Lebowitz, and B. Burrows. The maximal expiratory flow volume curve. Am. Rev. Respir Dis. 113: 587–600, 1976.
16. Larsson, L., P. Hemmingsson, and G. Boethius. Self-reported obstructive airway symptoms are common in young cross-country skiers. Scand. J. Med. Sci. Sports. 4: 124–127, 1994.
17. Mahler, D. A. Exercise-induced asthma. Med. Sci. Sports Exerc. 25: 554–561, 1993.
18. Mannix, E. T., M. O. Farber, P. Palange, P. Galassetti, and F. Manfredi. Exercise-induced asthma in figure skaters. Chest 109: 312–315, 1996.
19. Mcfadden E. R. Jr., I. A. Gilbert. Exercise-induced asthma. N. Engl. J. Med. 330: 1362–1367, 1994.
20. Morris, J. F. Spirometry in the evaluation of pulmonary function. West J. Med. 125: 110–111, 1976.
21. Nystad, W., J. Harris, and J. S. Borgen. Asthma and wheezing among Norwegian elite athletes. Med. Sci. Sports Exerc. 32: 266–270, 2000.
22. Provost-Craig, M. A., K. S. Arbour, D. C. Sestilli, J. J. Chabalko, and E. Ekinol. The incidence of exercise-induced bronchospasm in competitive figure skaters. J. Asthma 33: 67–71, 1996.
23. Rundell, K. W., R. L. Wilber, L. Szmedra, D. M. Jenkinson, L. B. Mayers, and J. Im. Exercise-induced asthma screening of elite athletes: field versus laboratory exercise challenge. Med. Sci. Sports Exerc. 32: 309–316, 1999.
24. Rupp, N. T., M. F. Guill, and D. S. Brudno. Unrecognized exercise-induced bronchospasm in adolescent athletes. Am. J. Dis. Child. 146: 941–944, 1992.
25. Sears M. R. Epidemiological trends in bronchial asthma. In:Asthma, Its Pathology and Treatment, M. A. Kaliner, P. J. Barnes, C. G. A. Persson (Eds). New York: Dekker, 1991, 49:1–49.
26. Smith, L. Exercise testing in adults and children. In:Exercise-Induced Asthma, E. R. McFadden, Jr. (Ed.). New York: Dekker, 1999, 130:235–259.
27. Sterk, R. H., L. M. Fabbri, P. H. Quanjer, Airway responsiveness: et al. : standardized challenge testing with pharmacological, physical and sensitizing stimuli in adults. Eur. Respir. J. 6 (Suppl. 16): 53–83, 1993.
28. Tan, R. A., and S. L. Spector. Exercise-induced asthma. Sports Med. 25: 1–6. 1998.
29. Tikkanen, H. O., and J. E. Peltonen. Asthma: cross-country skiing. Med. Sci. Sports Exerc. 31: (Suppl.) S99, 1999.
30. Voy, R. O. The U.S. Olympic Committee experience with exercise-induced bronchospasm, 1984. Med. Sci. Sports Exerc. 18: 328–330, 1984.
31. Weiler, J. M., T. Layton, and M. Hunt. Asthma in United States Olympic athletes who participated in the 1996 summer games. J. Allergy Clin. Immunol. 102: 722–726, 1998.
32. Wilber, R. L., K. W. Rundell, L. Szmedra, D. M. Jenkinson, J. Im, and S. D. Drake. Incidence of exercise-induced bronchospasm in Olympic winter sport athletes. Med. Sci. Sports Exerc., in press.


© 2001 Lippincott Williams & Wilkins, Inc.