LAUSR

sufficiently explained. We elaborated on this in our first response (Heroux et al. 2016), arguing that the criticized calculation of ‘premature deaths’ produces a reasonable albeit ambiguous estimate, for which reason calculation of years of life lost is a more preferable approach. We would like to point out that the HRAPIE report really is about identification of concentration–response functions to be further used in health impact assessments, and therefore did not pretend to provide a discussion of estimating etiologic fractions. Morfeld and Erren single out the one numerical example of an impact assessment given in our paper, We thank Morfeld and Erren for their continued interest in the WHO Health risks of air pollution in Europe (HRAPIE) report (WHO Regional Office for Europe 2013). The key point of contention seems to be the interpretation of the numbers of ‘premature deaths’ associated with air pollution (or any other) exposure. In the IJPH article that is at the basis of the two letters written by Morfeld and Erren (Heroux et al. 2015), the limitations of calculating and using numbers of ‘premature deaths’ were perhaps not