This was briefly the background, when I decided to test the "J"-curve hypothesis in the ATBC Study cohort with 30,000 male participants

First, I analyzed the relation between physical activity and common incidence and I found no support to the "J"-curve. This work was published in MSSE 2003;35:1815-20.

Matthews et al: Moderate to vigorous physical activity and risk of upper-respiratory tract infection.

MSSE. 2002;34(8):1242-8.

http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=12165677&query_hl=8&itool=pubmed_docsum
Matthews had 547 participants in contrast to the 14401 participants in my ATBC cohort study. In any case, such a small difference, 20%, is very easily explained by unadjusted confounders. To prove that such a small difference (20% effect) is real would require randomized trials. Furthermore, even if the 20% difference would actually be caused by the greater level of exercise, we can calculate that some 4 people should exercise for one year in order to prevent one episode of the common cold, which is quite a banal infection (e.g. in comparison with pneumonia). Thus the cost-benefit ratio is very low even if there was an effect.
After the common cold work, I analyzed the relationship between physical activity and the incidence of pneumonia.

Pneumonia is the most common severe infection. Therefore it is a good outcome to test whether variation in physical activity has actual effects on infection risk.

I found no evidence that greater physical activity at work or at leisure would be associated with lower risk of pneumonia.

The participants of the ATBC Study had variation in their work and leisure time physical activity, but it was not possible to pick persons who had very heavy physical stress. Therefore the work does not test the right hand part of the "J"-curve as I explicitly state in the paper, but the results are highly relevant to the valley part of the "J"-hypothesis, i.e. the speculated benefit of moderate exercise on the immune function.

I submitted my manuscript to MSSE and 2 reviewer read it and both suggested rejection. I received only one reviewer comments (#1) which I found to be such a poor scientific argumentation so that I wrote counterarguments to the editorial office and therefore my manuscript was sent to a third reviewer, which also suggested rejection.

Essentially all of the arguments of the three reviewers were unsound and I am presenting my counterarguments to the three reviewers below.

I was surprised that the editors of MSSE were unable to see that those reviewer comments of my pneumonia manuscript were of such a poor scientific quality that they should not be used as a basis to consider whether my manuscript is good or not.