Some comments about the background of: Hemilä et al. (2006) Int j sports Medicine 27: 336-341 Physical activity and the risk of pneumonia

with marathon runners has been frequently cited as an evidence that heavy exercise increases risk of common cold

13 colds in 44 runners

6 colds in 80 non-runners

>> total 19 common cold episodes in the study
Peters et al. Am J Clin Nutr 1993;57:170-4

with marathon runners has also been frequently cited as an evidence that heavy exercise increases risk of the common cold

28 colds in 41 placebo-runners

18 colds in 39 placebo-nonrunners

>> total 46 common cold episodes in the study on physical activity (excluding the vitamin C arms)
These 4 papers have been cited, for example, by Nieman in his review on physical activity and the risk of common cold in MSSE 1994;26:128-39 (his Table 1)

Thus, for example, according to Nieman, these 4 studies with 11 (or 12) to 46 cold episodes are large enough to allow meaningful conclusions.

In our cohort study we have 71 pneumonia cases to analyze the association between physical activity and the risk of pneumonia, and accordingly the statistical power is substantially higher than in those 4 common cold studies that were acceptable for Nieman in his 1994 review.

Thus, the reviewer’s comment that our pneumonia cases are so few that they do not allow meaningful epidemiological analysis, is not valid if we compare e.g. with Nieman’s analyses of common cold studies.

Furthermore, pneumonia is much more severe and much more rare infection compared with the common cold so that we should not expect analyses of pneumonia to have more cases than analyses of the common cold; rather the contrary.

8) The reviewer states: “DISCUSSION: Once again, I feel the literature review is slanted towards nonsupport of the physical activity-infection link. There are numerous animal and human studies showing that the link between infection and physical activity is important, depending on the pathogen studied, the severity of the exercise workload, and other factors.

In any case, our presentation of previously published literature is not at all relevant to the validity of our new analyses.

Also, our manuscript is not a review and we had the upper limit of 30 references.

Also, in our introduction we were referring to three reviews by different authors and two cohort studies.

As regards the validity of our study, the outcome of “pneumonia” used in our manuscript was acceptable for professional pulmonologists as the original ATBC Study pneumonia analysis was published in Chest.