Correspondence
Conceptual Misunderstanding
;
We read the article (1) with great interest. The main basis for the calculation consists of weekly data from the German Federal Statistical Office regarding overall mortality. We wish to point out a conceptual misunderstanding: numbers of heat-related deaths—that is, the numbers of people who would have lived longer without the occurrence of “heat weeks/waves” (1)—cannot be adequately determined from these data.
As we explained in (2), the number of affected persons cannot be concluded from differences in death rates because the differences determined in (1) between modeled mortality rates and hypothetical mortality trend that would result if the mean weekly temperature did not rise above 20°C are consistent with many different constellations from heat related deaths. The actual number of heat related deaths could be much higher than determined by the authors. Neither the point estimates in the Table (1) nor the associated confidence intervals for the numbers of heat related deaths are therefore scientifically robust (proof in [3]).
Instead of the problematic “number of heat-related deaths,” the metric “heat-related years of life lost per exposed person” should be used.. This metric estimates the expected value of the lifetime lost because of heat stress and can in principle be determined and usefully interpreted from the data of the Federal Statistical Office (proof in [3]).
Finally we wish to emphasize that the facts we named were mathematically explained and published by the leading epidemiologist methodologists Greenland and Robin as early as in the 1980s (proof in [2, 3]).
DOI: 10.3238/arztebl.m2022.0332
PD Dr. Peter Morfeld
Prof. Dr. med. Thomas C. Erren, M.P.H.
Institut und Poliklinik für Arbeitsmedizin,
Umweltmedizin und Präventionsforschung, Uniklinik Köln,
Universität zu Köln
peter.morfeld@ruhr-uni-bochum.de
| 1. | Winklmayr C, Muthers S, Niemann H, Mücke HG, an der Heiden M: Heat-related mortality in Germany from 1992 to 2021. Dtsch Arztebl Int 2022; 119: 451–7 VOLLTEXT |
| 2. | Morfeld P, Erren TC: Warum ist die „Anzahl vorzeitiger Sterbefälle durch Umweltexpositionen“ nicht angemessen quantifizierbar? Gesundheitswesen 2019; 81: 144–9 CrossRef MEDLINE |
| 3. | Hammitt JK, Morfeld P, Tuomisto JT, Erren TC: Premature deaths, statistical lives, and years of life lost: identification, quantification, and valuation of mortality risks. Risk Anal 2020; 40: 674–95 CrossRef MEDLINE PubMed Central |
