Global mortality burden attributable to non-optimal temperatures
Katrin Burkart and colleagues1 present the results of an ambitious study on the global mortality burden attributable to non-optimal temperatures by the Global Burden of Diseases, Injuries,and Risk Factors Collaborators. They report that 2·98% of deaths globally could be attributed to non-optimal temperatures in 2019; 2·37% of deaths from low temperatures and 0·63% of deaths from high temperatures. Although estimates of the heat mortality burden are broadly consistent with existing literature, the contribution of cold temperatures to this burden differs substantially to assessments at global and regional scales.2, 3, 4 We believe that these differences are the result of crucial methodological limitations of the study,1 mainly the failure to adequately address the complexities of temperature–mortality relationships, probably resulting in an underestimation of the effects. Burkart and colleagues only accounted for the effects on the same day (ie, lag 0), whereas substantial epidemiological data shows the presence of lagged effects of temperature (up to 3 weeks for cold temperatures) or mortality displacement, or both.5 Additionally, the applied method does not account for seasonality or long-term trends—strong confounders in this analysis.5 In the appendix, we illustrate how markedly different the results of the two approaches are using data from Greater London, UK. A critical lens needs to be applied to any analytical framework to ensure its suitability and to increase confidence in the results. Burkart and colleagues’ analyses1 would have benefited from method developments in climate epidemiology from the past 20 years. Robust and reliable estimates of the burden of non-optimal temperatures are increasingly important in a changing climate.
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | artículo biblioteca |
| Language: | English |
| Published: |
Elsevier
2022-03-19
|
| Subjects: | Global mortality, Temperatures, |
| Online Access: | http://hdl.handle.net/10261/266912 https://api.elsevier.com/content/abstract/scopus_id/85126606685 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Katrin Burkart and colleagues1 present the results of an ambitious study on the global mortality burden attributable to non-optimal temperatures by the Global Burden of Diseases, Injuries,and Risk Factors Collaborators. They report that 2·98% of deaths globally could be attributed to non-optimal temperatures in 2019; 2·37% of deaths from low temperatures and 0·63% of deaths from high temperatures.
Although estimates of the heat mortality burden are broadly consistent with existing literature, the contribution of cold temperatures to this burden differs substantially to assessments at global and regional scales.2, 3, 4 We believe that these differences are the result of crucial methodological limitations of the study,1 mainly the failure to adequately address the complexities of temperature–mortality relationships, probably resulting in an underestimation of the effects.
Burkart and colleagues only accounted for the effects on the same day (ie, lag 0), whereas substantial epidemiological data shows the presence of lagged effects of temperature (up to 3 weeks for cold temperatures) or mortality displacement, or both.5 Additionally, the applied method does not account for seasonality or long-term trends—strong confounders in this analysis.5 In the appendix, we illustrate how markedly different the results of the two approaches are using data from Greater London, UK.
A critical lens needs to be applied to any analytical framework to ensure its suitability and to increase confidence in the results. Burkart and colleagues’ analyses1 would have benefited from method developments in climate epidemiology from the past 20 years. Robust and reliable estimates of the burden of non-optimal temperatures are increasingly important in a changing climate. |
|---|