Comparative prevalence of asthma in different groups of athletes : A survey

Centre de Recherche de l’Hôpital Laval, Institut Universitaire de Cardiologie et de Pneumologie, Université Laval, Quebec City, Quebec Correspondence: Dr Louis-Philippe Boulet, Hôpital Laval, 2725 Chemin Sainte-Foy, Quebec City, Quebec G1V 4G5. Telephone 418-656-4747, fax 418-656-4762, e-mail lpboulet@med.ulaval.ca J-B Langdeau, H Turcotte, G Thibault, L-P Boulet. Comparative prevalence of asthma in different groups of athletes: A survey. Can Respir J 2004;11(6):402-406.

A sthma and airway hyperresponsiveness (AHR) are more prevalent in athletes than in the general population (1,2).Studies (3)(4)(5) using methacholine or histamine challenges to evaluate AHR identified a high prevalence of AHR in various sports disciplines; however, studies (6-9) using surveys to evaluate asthma among the athletic population reported a significantly lower prevalence of physician-diagnosed asthma than that reported for AHR.This phenomenon had already been observed among the general population, but in athletes, the reason for the discrepancy between the prevalence of AHR and the diagnosis of asthma remains to be determined.It might be attributed to poor perception of respiratory symptoms or to an acquired tolerance to nociceptive symptoms.Another explanation might be that a high percentage of athletes have developed asymptomatic AHR, with no clinical manifestations.
In a previous study (4), we looked at the prevalence of AHR in four different groups of athletes categorized according to the type of air predominantly inhaled when training: cold air (CA), dry air (DA), humid air (HA) and a mixture of dry and humid air (MA).The prevalence of AHR was significantly higher (49%) in all athletes than in sedentary subjects (28%), but surprisingly, very few athletes reported having a diagnosis of asthma (16%).When the prevalence of AHR among the groups of athletes was compared, we found significant differences.Athletes of the CA group (cross-country skiers, biathletes and speed skaters) and the HA group (swimmers) had significantly higher prevalences of AHR than the two other groups.The prevalence of AHR was not significantly different between athletes of the DA and MA groups and the sedentary group.The main objective of the present study was to evaluate and compare the prevalence of asthma among four groups of athletes categorized according to the type of air predominantly inhaled during training.In addition, the possible influences of a family history of asthma and of living environment on the development of asthma in these athletes were examined.
©2004 Pulsus Group Inc.All rights reserved

Study design and subjects
From the 33 sports disciplines selected for the study, a list of 1469 athletes from the province of Quebec was produced by the sports and physical activity branch of the government of Quebec.A letter explaining the purpose of the survey was sent to each athlete, offering the opportunity to take part in the survey by filling out an enclosed questionnaire (in French).All athletes received a second letter as a reminder four weeks after the initial mailing.The study protocol was approved by the Laval Hospital (Quebec City, Quebec) institutional ethics committee.Seven hundred questionnaires were returned; two questionnaires that were not properly filled out were eliminated.An analysis was performed on the remaining 698 anonymous questionnaires.

Questionnaire
Questions included personal history of asthma (self-reported and physician-diagnosed); severity of asthma; health care system use and medication for asthma; previous allergy testing; past or present history of hay fever; respiratory infections; episodes of cold in the past 12 months; and a family history of asthma, allergies and chronic obstructive pulmonary disease (COPD).Environmental conditions, such as exposure to domestic animals, bedroom carpeting and smoking history, were also recorded.The questionnaire structure is illustrated in Table 1.Questions used in the present survey included some from the validated questionnaire of the European Community Respiratory Health Survey (10) to which specific questions on exercise-related symptoms were added.
In the analysis, athletes were considered to have a positive history of asthma if they self-reported having asthma and/or had had physician-diagnosed asthma at any time.To evaluate the impact of different training environments, according to the type of air inhaled during training, on the respiratory health and allergic conditions in athletes, the four groups of athletes, as described in the introduction, were compared.

Statistical analysis
The results are expressed as mean ± SD and per cent for categorical variables.The analyses of continuous variables were performed using a one-way ANOVA.Subgroups were defined as CA, DA, HA and MA.Comparisons among groups were performed using Scheffe's multiple comparison technique.The normality assumption was verified with the Shapiro-Wilk test, and Bartlett's statistic was used to verify the homogeneity of variances.Categorical variables were analysed using Fisher's exact test.Results were considered significant if P≤0.05.Data were analysed using the statistical package SAS (SAS Institute Inc, USA).

Subject characteristics
Among the participating athletes, 45% were males and 55% were females (Table 2).The majority were white (95%) and most were young (19.0±4.0 years).The number of years and time spent training weekly are summarized in Table 2.The distribution of athletes for the 33 sports disciplines is illustrated in Table 3.The number of sports disciplines by category of athlete were eight in the CA group, 19 in the DA group, three in the HA group and three in the MA group (Table 3).

Prevalence of asthma
According to the criteria (athletes self-reporting asthma and/or having had physician-diagnosed asthma), 107 of 698 subjects (15.3%) had a past or present history of asthma; among these subjects, 92 stated that a physician had made the diagnosis of asthma.Table 3 illustrates the prevalence of asthma within each sports discipline.

Prevalence of asthma by group analysis
The prevalence of asthma varied from 12.6% to 18.6% among the four groups but no significant differences were observed (Table 4).

Severity of asthma and health care use
Of the 92 athletes with physician-diagnosed asthma, 17 described the severity of their asthma as very mild, 37 as mild, 22 as moderate and 14 as severe.Two athletes did not rate the severity of their asthma.Of the 92 athletes, 42 (45.7%)reported having visited the emergency department at least once for asthma and 13 (14.1%)reported having been hospitalized for asthma.

Asthma medication
Of the 92 athletes with physician-diagnosed asthma, 68 (73.9%) reported current or past use of asthma medication.Among these athletes, 38 used inhaled short-acting beta 2agonists alone, five used inhaled corticosteroids alone, 23 used inhaled corticosteroids in combination with short-acting beta 2agonists, one used long-acting beta 2 -agonists alone and one used inhaled corticosteroids in combination with a leukotrienereceptor antagonist.There was no report of asthma medication use among nonasthmatic athletes.

Allergic conditions
Of 159 athletes who reported having been tested for allergies, 143 (89.9%) reported having had at least one positive response (Table 5).Atopic athletes reported positive tests for the following specific allergens: house dust (56.6%), dust mites (35.0%), grasses (19.6%), ragweed (33.6%), trees (24.5%), animal dander (51.0%) and other (52.4%).The presence of allergy (defined as at least one positive test) was highly correlated with asthma (P<0.0001).Atopy was significantly higher in asthmatic athletes (56 of 107, 52.3%) than in nonasthmatic athletes (87 of 591, 14.7%).Hay fever, which was reported by 145 athletes, was also highly correlated with asthma (P<0.0001).By group analysis (CA, DA, HA and MA), there were no significant differences in the prevalence of atopy among the four groups; however, athletes from the DA group had a lower prevalence of hay fever than observed in the other three groups (P=0.04,Table 4).

Respiratory health
The majority of athletes (79.1%) reported at least one episode of cold/influenza in the last 12 months, but this parameter was not correlated with asthma.A smaller percentage of athletes reported having suffered from respiratory infection (12.9%), this parameter was highly correlated with asthma (P<0.0001).
Cold/influenza and respiratory infection prevalences were not significantly different among the four groups of athletes (Table 4).

Family history
A family history of asthma and allergy was not significantly correlated with asthma in athletes, although a family history of COPD was significantly correlated to asthma in athletes (P=0.002)(Table 5).

Living environment and smoking
The prevalence of asthma in athletes exposed to cats, dogs or had bedroom carpeting in their living environment was not significantly different from that in those not exposed to these conditions (Table 5).History of smoking was reported by 91 (13%) athletes.Of these, 44 were current smokers (eight athletes

DISCUSSION
In the present survey, we found a prevalence of self-reported or physician-diagnosed asthma (15.3%) similar to what has been previously reported among athletes.This prevalence is identical to the one found by Weiler et al (8) in summer Olympic Games athletes (15.3%), although their criterion for asthma consisted of a previous diagnosis of asthma or exercise-induced asthma.Later, using the same criterion, the same author found a prevalence of 21.9% in winter Olympic Games athletes (7).Similar figures were also found by Heir and Oseid (11) and Voy (12) who reported a prevalence of asthma and exercise-induced asthma in athletes of 14.0% and 11.2%, respectively, when using questionnaires.
In a previous study (4), we compared the prevalence of AHR in the same four groups of athletes (categorized by type of air inhaled during training) and found it to be remarkably high in swimmers (79%) and in athletes training in CA (52%).With the present study, we were expecting to find a similar tendency, but no significant differences were observed for the prevalence of asthma among the four groups.Moreover, the lowest prevalence of asthma was observed among athletes of the HA group, which had presented the highest prevalence of AHR in our previous study.This suggests a dissociation between what is measured by objective means for airway responsiveness and reported/diagnosed asthma evaluated by a survey questionnaire, especially for certain groups of athletes.Many factors might be responsible for the dissociation between prevalence of AHR and asthma, particularly for athletes of the HA group (swimming, synchronized swimming and water polo).For these athletes, this could indicate an under-recognition of asthma, possibly due to under-reporting or under-perception of respiratory discomfort in these athletes.In this regard, we previously showed that swimmers had a significantly lower perception of exercise-induced breathlessness when compared with the three other groups (13).This could also indicate that there is an overexpression of AHR in athletes training in indoor pools, possibly due to the longterm airway exposure to chlorine derivatives, leading to increased airway sensitivity to methacholine without inducing symptomatic asthma.For athletes training in cold air, similar factors could be involved, especially with regard to a possible overexpression of AHR that could result from long-term exposure to the cold air environment during training.
Perhaps certain athletes -for instance, those of the HA and CA groups -develop a temporal adaptation to nociceptive stimuli, such as respiratory discomfort during high-intensity training.It is also possible that the type of environment in which athletes of the HA or CA group train on a regular basis induces airway inflammation, sufficient to produce AHR, possibly from airway remodelling, but insufficient to generate perceivable symptoms.A possible reason to explain the discrepancy between studies using bronchial challenges (either with exercise or methacholine) and surveys using questionnaires could be the possible long-term temporal adaptation to respiratory symptoms.We have previously reported this feature following allergen challenge-induced late responses (14).In this regard, we recently extensively reviewed the problem of the discrepancy between physiological and clinical features observed in athletes (15).
In the present study, asthma did not correlate with any of the subject characteristics illustrated in Table 2 (all P>0.05).
The strong correlation of atopy and asthma in athletes is similar to the observations of Helenius et al (3).In the present study, we cannot exclude the possibility that patients with respiratory symptoms were more likely to have had allergy tests.It  is also possible that subjects may not have remembered the results of their allergy tests correctly.Nevertheless, the prevalence of atopy was not significantly different among the four groups.This trend is similar to what we observed in our previous study (4).The similar prevalence of atopy observed in the four groups of athletes may suggest that the allergen content in inhaled air is not a major factor influencing the development of sensitization to common allergens in these athletes or that other factors are involved.
A family history of asthma and allergies was not correlated with a personal history of asthma in these athletes.The lack of correlation between a family history of asthma and asthma in athletes indicates that, contrary to what has been demonstrated for the general population, this genetic determinant may not play a major role in the development of asthma in the athletic population.Helenius et al (3) evaluated the family history of asthma in first-degree relatives of athletes and found no association between this parameter and asthma in athletes.Nevertheless, while a family history of asthma has been identified as a risk factor for the development of asthma in the general population (16,17), more extensive studies should further investigate this aspect among athletes.Although the relationship between family history of COPD and personal history of asthma was statistically significant in our athletes, the significance of this finding is unclear and may be related to factors such as childhood exposure to second-hand smoke or an increased awareness toward respiratory conditions among their family.Finally, asthma was not more frequently reported in athletes exposed to animals, with bedroom carpeting or with a history of smoking.
Regarding limitations of the present study, a possible bias may have been introduced by the use of voluntary subjective reports.It is possible that athletes already concerned about their respiratory condition were more likely to fill out the questionnaire.It has been reported that lower response rates are observed when surveys are anonymous (18).The fact that this survey was made on an anonymous basis could perhaps have negatively influenced the response rate; however, the use of written reminders could have enhanced the response rate, as it has been demonstrated that surveys using written reminders obtain higher responses rates (18).The validity of physician-diagnosed asthma, and especially, self-reported asthma, could be one of the areas of uncertainty.Nevertheless, this method is currently used and is considered a valuable means of acquiring data on possible manifestations of diseases such as asthma in large populations.Data on the nonresponders would have been useful; however, this information was not available.It could also have been worthwhile to include normal controls in the present study, but our main goal was to compare the prevalence of asthma among groups of athletes rather than between athletes and nonathletes.

CONCLUSIONS
We previously reported that there was a high prevalence of AHR in high-level athletes and that among groups of athletes classified according to the type of air predominantly inhaled during training, this prevalence differed significantly.This led to the examination of the frequency of asthma in such athletes, but more precisely, to compare the prevalence of asthma among the four groups of athletes that had been previously defined.The present study showed a prevalence of asthma consistent with what has been reported by other surveys in athletes.The major observation was that, contrary to the significant variability in prevalence that was previously found for AHR in the four groups of athletes, in the present study, no significant differences were observed among the groups.In addition, athletes with asthma reported having confirmed allergies, hay fever and respiratory infection more frequently than those without asthma.There was a certain homogeneity among the four groups of athletes, regardless of the type of air they regularly inhaled during training.Lastly, a family history of asthma (with the exception of COPD) and environmental factors such as animal, carpet and smoke exposure did not correlate with the diagnosis of asthma in these athletes.
Prevalence of asthma in different groups of athletes Can Respir J Vol 11 No 6 September 2004 403 Prevalence of asthma in different groups of athletes Can Respir J Vol 11 No 6 September 2004 405

TABLE 1
1. Have you ever suffered from asthma? 2. Have you ever been diagnosed as having asthma by a physician?If yes, was your asthma defined as very light, light, moderate or 12.Is there one member of your family (father, mother, brother or sister) that has ever had one of the following conditions: asthma, allergies and/or emphysema?13.Which of the following are present where you live: cat, dog and/or car pet in your bedroom?14.Are you a current smoker?If yes, do you smoke less than 5 cigarettes/day, or 5 or more cigarettes/day?15.Are you an exsmoker?If yes, were you smoking less than 5 cigarettes/day, or 5 or more cigarettes/day?*Translated from French

TABLE 3
Prevalence of asthma* in various sports disciplines *Self-reported and/or physician-diagnosed asthma

TABLE 4
reported smoking five or more cigarettes/week and 36 reported smoking less than five cigarettes/week) and 47 were exsmokers (seven athletes estimated smoking five or more cigarettes/week and 40 estimated smoking fewer than five cigarettes/week).Past or present history of smoking was not associated with asthma.

TABLE 5
Prevalence of various parameters and their correlation with history of asthma in athletes