Reply to Harald Walach

All the methodological limitations of our study concerning baseline differences in variables (‘passive smoking’, ‘childcare’, etc.) are discussed in the article published by Homeopathy and explain why the two groups were considered as globally comparable. We are aware that performing a regression analysis might have corrected these bias but as it was a post hoc analysis, we did not plan to perform it in our protocol study. So the results of our post hoc study should be interpreted with caution, this is why we wrote in our article that methodologically speaking ‘this post hoc analysis is limited’ and ‘double-blind randomised clinical trials are required to confirm these results’ in terms of efficacy. Moreover, our conclusion is very cautious because we conclude by a supposition that ‘homeopathy may be or could constitute a cost-effective alternative to antibiotics’.