Correction: Unravelling Myths: A Difference-In-Differences Analysis of Post-Vaccination Mortality in Italy During COVID-19

In this article, an unintentional aggregation error was identified in the mortality data for the 0–9 age group. This issue arose due to a misalignment in the classification process, where the age classes 0, 1–4, and 5–9 were incorrectly grouped during data preprocessing. Specifically, an error occurred in the software routine that prematurely assigned a common category to these age groups. This led to an underrepresentation of mortality figures for this category. However, this issue did not affect the results for other age groups (analysed separately) or alter the study’s conclusions. To ensure complete accuracy and methodological transparency, the 0–9 age group has been removed from the analysis, as it was not central to the vaccination policy and was entirely outside the scope of the hypotheses and arguments examined in this study. Indeed, vaccination for children aged 0–4 (from 6 months) was approved at the end of 2022, while for ages 5–9 was introduced only in December 2021 (source: AIFA, AIFA). Furthermore, vaccination coverage in this age group remained minimal, with only 0.03% of children aged 0–4 (from 6 months) and 38.6% of those aged 5–11 receiving at least one dose (source: Ministry of Health, 01/11/2024). Given these circumstances, this age group was not the focus of the study and was never part of the discussion or conclusions. This adjustment does not impact the findings for the other age groups or the overall robustness of the study. During this review process, the following minor revisions were made to correct typographical inconsistencies and improve readability, particularly for a non-econometric audience: Throughout the paper, we consistently referred to age class groups (e.g., 10–19, 20–29, etc.), removing previous adjunct classifications (e.g., “age group 1,” “age group 2”). Table 1 was reformulated to better distinguish the descriptive data of the two datasets before merging them for the core analysis. The “municipalities” column was removed from Table 1, as this information is already detailed in Table 8. The captions for Tables 3 and 4 were modified to specify the “average of municipal mortality rates” and clarify that the scale is per 1,000 inhabitants. The labelling for the first two classes (females) in Fig. 4; Table 2 has been relabelled. Values on the bars in Figs. 3 and 4 are uniform for a more explicit visual representation in the accompanying graphs. The captions for Tables 6 and 7 were revised to specify the effects of positive and negative coefficients related to the increase or decrease of mortality. Table 8 was reorganised for improved clarity. Additionally, to enhance transparency and reproducibility, the syntax files of these post-publication changes have been made publicly available on ZENODO [https://zenodo.org/records/14849952]. These materials also include replicating the results for the 0–9 age group, conducted purely as an econometric test. The new estimates confirm the previous findings, showing no significant variation in mortality trends within this age group.