The mortality related to heat during Summer of 2022 in Europe could have exceeded 70,000 deaths, according to a study by the Barcelona Institute for Global Health (ISGlobal), a center supported by the “la Caixa” foundation. The results, published in The Lancet Regional Health – Europe, Uprate initial estimates of associated mortality Record temperatures on the European continent.
In one previous studypublished in the magazine Natural medicineThe same authors used epidemiological models applied to weekly temperature and mortality data in 823 regions in 35 European countries to estimate a premature mortality of 62,862 people in 2022. The authors themselves recognized that using weekly data led to an underestimation of heat-related mortality, as daily data series are required to accurately estimate the influence of warm temperatures on mortality.
The aim of the new study was to develop a theoretical framework that enables quantification Errors arising from the use of aggregated dataB. weekly or monthly temperature and mortality series.
This aggregated datas are particularly useful because they are available in real time through institutions such as Eurostat, allowing health risks to be quantified within a few days. To develop this theoretical framework, the research team added the daily temperature and mortality series of 147 regions from 16 European countries between 1998 and 2004. Estimates of heat- and cold-related mortality were then analyzed and compared at different levels of aggregation: daily, weekly, biweekly, and monthly.
The results showed differences in epidemiological estimates according to the temporal aggregation scale. In particular, it was found that the weekly, biweekly and monthly models generally underestimate the effects of heat and cold compared to the daily model and that this underestimation increases the larger the temporal aggregation scale.
Specifically, the daily data model for the period 1998–2004 estimated annual cold and heat mortality at 290,104 and 39,434 premature deaths, respectively, while the weekly model underestimated these numbers by 8.56% and 21.56%, respectively.
The underestimation increases as the temporal aggregation scale increases.
“It is important to highlight that during periods of extreme cold and heat, such as the summer of 2003, these differences were very small, with an underestimate of only 4.62% in the weekly data model,” he points out. Joan Ballester ClaramuntISGlobal researcher leading the project CUSTOMIZE EARLY of the European Research Council.
Recalculate mortality using weekly data
The research team applied this theoretical framework to re-estimate mortality caused by 2022’s record temperatures. According to the calculations now carried out using this new methodological approach, mortality in this study would have been underestimated by 10.28%, which would mean that the actual heat-related mortality in 2022 would be 70,066 deaths according to the daily data model, rather than the original estimates 62,862 deaths.
“In general, we have found that using monthly aggregated data does not allow us to estimate the short-term impact of ambient temperatures. On the other hand, the use of weekly data provides sufficient precision in mortality estimates to be used in real-time practice epidemiological surveillance and in the Generation of public policysuch as activating emergency plans to prevent the effects of heat and cold waves,” says the ISGlobal researcher.
This is an advantage because research teams often encounter bureaucratic hurdles that make it difficult or impossible to design large-scale epidemiological studies with daily data. In cases where it is not possible to have daily data, according to Ballester, using weekly time series, easily accessible in real time at European level, is a good solution as it offers “good value”. Approach of the estimates obtained in the model daily data“.