Response to the letter to the editor from Morfeld

overseen that after blocking pneumoconiosis as a cause of death the probability of a certain cause can then be described by the ratio between the observed number of this cause and the number of all causes of death except for pneumoconiosis. Therefore, this approach yields SMR = (7 + 6.7)/10.14 = 1.35 for the worked example, which even exceeds the originally proposed adjusted risk estimate. Following Dr. Morfeld’s recommendation, 67*67/137 = 32.8 miners suffering from pneumoconiosis would survive forever or at least up to the end of followup. However, this is exactly what he was actually trying to avoid by his recommendation. In his criticism, Dr. Morfeld refers to a time-dependent method for a counterfactual or potential outcome model (Robins 1998) and his own calculations based on a cohort of German coal miners (Morfeld and Lampert 2004; Morfeld et al. 2005) as a kind of gold standard. The counterfactual concept refers to what would have happened if, contrary to fact, the exposure had been something other than what it actually was. Applying this concept to our exemplary research question, we may analyze what would have happened if a worker had worked as a construction worker on surface, for example, instead of working underground as a coal miner as he did in reality. Of course, in that case he would have not been exposed to coal dust, but his smoking habits could have been quite different also. Smoking is strictly prohibited in underground coal mines, and all miners, purely for reasons of self-preservation, respect this prohibition strictly during the whole underground shift. Therefore, it is plausible that a coal miner smokes noticeably less than his counterfactual sibling, and hence, the potential increase in lung cancer risk due to the exposure to coal dust might be even more than compensated by his reduced tobacco consumption. In addition, breathing difficulties due to a reduced lung function or receiving advice from the I would like to thank the Journal for giving me the opportunity to respond to comments from Dr. Morfeld on my suggested approach to adjust standardized mortality ratios (SMRs) for competing causes. The aim of my suggested approach (Möhner 2015) was to develop a method, which can even be applied to published results from cohort studies like the US coal miner mortality study (Attfield and Kuempel 2008), which I used as an example. Contrary to Dr. Morfeld’s assumption, neither the number of observed cases nor the overall number of expected cases has been changed by my proposed approach. This is reflected in the basic assumption