Abstract:Nick Watts and colleagues1Watts N Amann M Arnell N et al.The 2020 report of the Lancet Countdown on health and climate change: responding to converging crises.Lancet. 2021; 397: 129-170Summary Full Text Full Text PDF PubMed Scopus (477) Google Scholar did wide-reaching analyses to track the emerging health effect of climate change. Among other indicators they estimated “Indicator 1.1.3: heat-related mortality” and “Indicator 3.3: premature mortality from ambient air pollution” and concluded: “average heat-related mortality per year…a total of 296 000 deaths in 2018” and “premature deaths…from ambient PM2·5 have increased…from 2·95 million deaths in 2015 to 3·01 million deaths in 2018”. The Article1Watts N Amann M Arnell N et al.The 2020 report of the Lancet Countdown on health and climate change: responding to converging crises.Lancet. 2021; 397: 129-170Summary Full Text Full Text PDF PubMed Scopus (477) Google Scholar estimated numbers of deaths by attributable fractions, which disallows identification of the number of premature deaths due to exposure.2Hammitt 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: 674-695Crossref Scopus (21) Google Scholar The resulting bias is not quantified by typical confidence or uncertainty intervals and it could be substantial (in fact, the numbers of premature deaths due to exposure could be much larger than calculated by the authors).2Hammitt 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: 674-695Crossref Scopus (21) Google Scholar, 3Morfeld P Erren TC Uncertainties in the GBD 2017 estimates on diet and health.Lancet. 2019; 3941802Summary Full Text Full Text PDF Scopus (4) Google Scholar Greenland and Robins4Greenland S Robins JM Conceptual problems in the definition and interpretation of attributable fractions.Am J Epidemiol. 1988; 128: 1185-1197Crossref PubMed Scopus (3) Google Scholar, 5Robins JM Greenland S Estimability and estimation of excess and etiologic fractions.Stat Med. 1989; 8: 845-859Crossref PubMed Scopus (100) Google Scholar identified this methodological error decades ago. Research into how health and climate change are linked is too important to ignore, rather than acknowledge and factor in, well known limitations of epidemiological data. We declare no competing interests. Countdown on health and climate change: too important for methodological errors – Authors' replyWe thank Peter Morfeld and Thomas Erren for their interest in the 2020 report of the Lancet Countdown.1 They argue that the use of attributable fractions is inadequate to estimate premature mortality from ambient air pollution or exposure to heat, and base their argument on the work by Hammitt and colleagues,2 which states that mortality does not allow one to “distinguish between situations in which a few people die much earlier than they would have if unexposed and situations in which many people die a little earlier than they would have if unexposed”. Full-Text PDF The 2020 report of The Lancet Countdown on health and climate change: responding to converging crisesFor the Chinese, French, German, and Spanish translations of the abstract see Supplementary Materials section. Full-Text PDF

Countdown on Health and Climate Change: Too Important for Methodological Errors – Authors' Reply