Asthma and Swimming: A Meta-Analysis
Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA 30322, USA. Journal of Asthma
(Impact Factor: 1.8).
11/2008; 45(8):639-47. DOI: 10.1080/02770900802165980
In this meta-analysis, studies on swimming and asthma were divided into four groups: Group I compared frequency of asthma among elite swimmers to that of other athletes; Group II examined the association between asthma and swimming during childhood; Group III evaluated effects of swimming programs on asthma severity and pulmonary function; and Group IV compared immediate respiratory effects of swimming to those of other types of exercise. The summary results were expressed as meta-odds ratios (ORs) for binary endpoints such as presence of asthma, and meta-differences for continuous endpoints such as changes in post-exercise pulmonary function tests (PFTs). All summary measures of effect were calculated using random effects models accompanied by a corresponding 95% confidence interval (CI) and a test for heterogeneity. In the analysis comparing frequency of asthma among elite swimmers to that among other athletes (Group I), meta-ORs ranged from 2.3 to 2.6 with all 95% CIs excluding 1.0. The corresponding meta-ORs reflecting the association between asthma and swimming pool use during childhood (Group II) were in the 0.63-0.82 range and were not statistically significant. In comparison to swimming, running and/or cycling was associated with a statistically significant four-to six-fold increase in exercise induced bronchospasm. Although asthma is more commonly found among elite swimmers than among other high-level athletes, it is premature to draw conclusions about the causal link between swimming and asthma because most studies available to date used cross-sectional design, because the association is not confirmed among non-competitive swimmers, and because asthmatics may be more likely to select swimming as the activity of choice because of their condition.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.