Time Course Analysis of Age-Gender Effect on Severity of COVID-19 Outbreak in Spain and Italy

Marta Niedostatek, Maria Cristina Valsecchi, Carlo Vittorio CannistraciResearch 2020 · DOI

Abstract

Several data science studies have analysed the gender effect on COVID-19 with at least one of these limitations: 1) missing comparison across countries; 2) data not age-stratified; 3) analysis based on single date of epidemic period instead of time-course; 4) few variables analysed; 5) gender bias not adjusted by country’s population in that strata. Here, we address these limitations. A wide range of variables on the severity of COVID-19 in relation to gender and age, are analysed over an extended time course from March 2020 to when data are publicly available. Spanish and Italian data only are considered because they are the unique open access data to be comprehensive and harmonized according to a comparable format. Altogether our findings offer two key evidence-driven recommendations. First, since data collection is disharmonic across Europe, the creation of a European institute for standards in biomedical data collection could play a crucial role for fast open-source dissemination and analysis of harmonized data, which in turn could foster rapid and coordinated decision making in emergency periods. Second, since COVID-19 severity particularly impacts 60+ males, containing interventions might be more age/gender-adaptive and, to increase effectiveness and efficiency, focus to contrast contagion of these categories at risk.

Introduction

Early in the COVID-19 pandemic, it became clear that men and women were affected differently. However, many early data science studies suffered from "snapshot bias"—they looked at data from a single day or a single country without accounting for age or the underlying population distribution. For example, if a country has significantly more women than men in the 90+ age bracket, a raw count showing more female deaths in that bracket might lead to the false conclusion that the virus is more lethal to women. This paper addresses five specific gaps: lack of cross-country comparisons, lack of age-stratification, reliance on single-date snapshots, limited variables, and failure to adjust for population imbalances within age strata.

Methods

The researchers conducted a longitudinal analysis of COVID-19 data from Spain and Italy, covering the period from March 2020 through October 2020. These countries were selected because they provided the most comprehensive, open-access, and harmonized datasets in Europe at the time.

Population Adjustment

A critical technical approach was the normalization of all variables (confirmed cases, hospitalizations, ICU admissions, and deaths) by dividing them by the specific country’s population for that gender and 10-year age strata. This ensures that the "gender gap" reflects actual risk rather than demographic distribution.

Formula for population-adjusted difference in confirmed cases

Statistical Analysis

Rather than looking at one day, the team analyzed the evolution of the gender gap over time, using the area under the curve to quantify the total difference in impact. They employed Pearson Chi-Square tests to compare proportions and applied the Benjamini-Hochberg procedure to control the false discovery rate during multiple hypothesis testing. When reporting errors were identified in Spanish ICU data, cubic spline interpolation was used to recreate missing functional data without biasing the results.

Key Insights and Results

The Severity Gap

While the number of infections was highly dependent on age and social role (with women aged 20–59 showing higher infection rates, likely due to "professional exposure" in healthcare and retail), the severity of the disease was consistently gendered. In both countries, the difference in male vs. female deaths and ICU admissions grew substantially over time, particularly for those aged 60+.

Formula for population-adjusted difference in death rates

The 90+ Trend Inversion

One of the most surprising findings was in the 90+ age group. Initially, more males were infected, but by mid-April, the trend inverted to significantly more females. The authors speculate this might be due to early testing biases where limited tests were prioritized for the "more severe" (male) cases.

Consistency Across Borders

Despite different national lockdown strategies, the age-gender trends in Italy and Spain were remarkably similar. Hospitalized males were significantly more likely to require ICU treatment than hospitalized females, especially in the 60–79 age range. The "naïve case fatality rate" was also significantly higher for males in the 60+ strata.

Conclusion and Policy Recommendations

The study highlights that COVID-19 severity is a robust biological and sociological pattern that transcends national borders. However, the research was limited by the lack of harmonized data across the rest of Europe.

The authors propose two major shifts in pandemic management:

  1. Data Standardization: The creation of a European Institute for Standards in Biomedical Data Collection to ensure data is immediately comparable across borders during emergencies.
  2. Adaptive Interventions: Future social interventions, such as lockdowns or contact tracing, should be "age/gender-adaptive" to focus resources on the most at-risk categories, specifically males over 60.

Summary updated

Abstract

Abstract Several data science studies have analysed the gender effect on COVID19 with at least one of these limitations: 1) missing comparison across countries; 2) data not age-stratified; 3) analysis based on single date of epidemic period instead of time-course; 4) few variables analysed; 5) gender bias not adjusted by country’s population in that strata. Here, we address these limitations. A wide range of variables on the severity of COVID-19 in relation to gender and age, are analysed over an extended time course from March 2020 to when data are publicly available. Spanish and Italian data only are considered because they are the unique open access data to be comprehensive and harmonized according to a comparable format. Altogether our findings offer two key evidence-driven recommendations. First, since data collection is disharmonic across Europe, the creation of a European institute for standards in biomedical data collection could play a crucial role for fast open-source dissemination and analysis of harmonized data, which in turn could foster rapid and coordinated decision making in emergency periods. Second, since COVID-19 severity particularly impacts 60+ males, containing interventions might be more age/gender-adaptive and, to increase effectiveness and efficiency, focus to contrast contagion of these categories at risk.
BibTeX
@inproceedings{niedostatek2020time,
  author = {Marta Niedostatek and Maria Cristina Valsecchi and Carlo Vittorio Cannistraci},
  title = {{Time Course Analysis of Age-Gender Effect on Severity of COVID-19 Outbreak in Spain and Italy}},
  booktitle = {Research},
  year = {2020},
  doi = {10.21203/rs.3.rs-131287/v1},
}
Figure