Assessing COVID-19-Related Excess Mortality Using Multiple Approaches—Italy, 2020–2021
Authors:
- Emiliano CeccarelliECEmiliano Ceccarelli
- Maria DorrucciMDMaria Dorrucci
- Giada MinelliGiada Minelli
- Giovanna Jona LasinioGLGiovanna Jona Lasinio
- Sabrina PratiSPSabrina Prati
- Marco BattagliniMBMarco Battaglini
- Gianni CorsettiGCGianni Corsetti
- Antonino BellaABAntonino Bella
- Stefano BorosSBStefano Boros
- Daniele PetroneDPDaniele Petrone
- Flavia RiccardoFRFlavia Riccardo
- Antonello MaruottiAMAntonello Maruotti
Open Access
- 17 December 2022
- journal article
- research article
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 19 (24) , 16998
Abstract
Introduction: Excess mortality (EM) is a valid indicator of COVID-19’s impact on public health. Several studies regarding the estimation of EM have been conducted in Italy, and some of them have shown conflicting values. We focused on three estimation models and compared their results with respect to the same target population, which allowed us to highlight their strengths and limitations. Methods: We selected three estimation models: model 1 (Maruotti et al.) is a Negative-Binomial GLMM with seasonal patterns; model 2 (Dorrucci et al.) is a Negative Binomial GLM epidemiological approach; and model 3 (Scortichini et al.) is a quasi-Poisson GLM time-series approach with temperature distributions. We extended the time windows of the original models until December 2021, computing various EM estimates to allow for comparisons. Results: We compared the results with our benchmark, the ISS-ISTAT official estimates. Model 1 was the most consistent, model 2 was almost identical, and model 3 differed from the two. Model 1 was the most stable towards changes in the baseline years, while model 2 had a lower cross-validation RMSE. Discussion: Presently, an unambiguous explanation of EM in Italy is not possible. We provide a range that we consider sound, given the high variability associated with the use of different models. However, all three models accurately represented the spatiotemporal trends of the pandemic waves in Italy.Keywords
This publication has 38 references indexed in Scilit:
- COVID-19 and Excess All-Cause Mortality in the US and 18 Comparison CountriesJAMA, 2020
- Excess mortality during the COVID-19 outbreak in Italy: a two-stage interrupted time-series analysisInternational Journal of Epidemiology, 2020
- Estimating weekly excess mortality at sub-national level in Italy during the COVID-19 pandemicPLOS ONE, 2020
- Temporal dynamics in total excess mortality and COVID-19 deaths in Italian citiesBMC Public Health, 2020
- Excess all-cause mortality during the COVID-19 pandemic in Europe – preliminary pooled estimates from the EuroMOMO network, March to April 2020Eurosurveillance, 2020
- Estimating weekly excess mortality at sub-national level in Italy during the COVID-19 pandemicmedRxiv, 2020
- How Large Was the Mortality Increase Directly and Indirectly Caused by the COVID-19 Epidemic? An Analysis on All-Causes Mortality Data in ItalyInternational Journal of Environmental Research and Public Health, 2020
- Correlates, facilitators and barriers of physical activity among primary care patients with prediabetes in Singapore – a mixed methods approachBMC Public Health, 2020
- Root mean square error (RMSE) or mean absolute error (MAE)? – Arguments against avoiding RMSE in the literatureGeoscientific Model Development, 2014
- Residual Analysis for Linear Mixed ModelsBiometrical Journal, 2007