Daily Epidemiological Dataset of Respiratory Contagious Diseases

Description

Data regarding the variables COVID-19, Severe acute respiratory infections (SARI), Acute respiratory infections (ARIs) , Influenza-like illness (ILI), and Flu types A and B were collected from “Ourworldindata.org”, spanning from early January 2020 to March 17, 2025. COVID-19 data, unlike the other variables, presented a major challenge due to inconsistent reporting frequencies. At the onset of the pandemic, COVID-19 data were reported daily. Following extensive vaccination campaigns and the subsequent decline in daily incidence rates, several countries ceased reporting altogether after late 2023, while others shifted to a weekly framework. The data for SARI, ARIs, ILI, and Flu types A and B were collected with more consistent weekly intervals from the start, as provided on the same source website. This inconsistency in reporting frequency directly affected the quality of the resulting time series. The shift to weekly reporting introduced an artifact into the COVID-19 data: artificial weekly peaks appeared at the start of each week, followed by sharp drops to near-zero values on weekdays. Such behavior is epidemiologically implausible for an infectious disease; once a time-series value reaches a certain threshold, it should propagate, increase, or decrease at a statistically determined rate rather than undergo abrupt, artificial fluctuations. The daily time-series data were generated from the weekly records of the target database using a scientifically grounded methodology, described in a manuscript currently under preparation. This procedure preserves the statistical distribution of the original data throughout the disaggregation process. Thus, applying the same generation technique yielded regular daily time series across variables and reduced the impact of reporting inconsistencies.

Institutions

• Universidad Carlos III de Madrid

  Madrid, Madrid

Categories

Licence