Framework reveals how neglecting income, education and ethnicity affects predictions of disease spread based on COVID-19 data

An international team of researchers has developed an innovative approach to modeling epidemics that could transform the way scientists and policymakers predict the spread of infectious diseases. Led by Dr Nicola Perra, Reader in Applied Mathematics, the study published in Scientific advances introduces a new framework that incorporates socioeconomic status (SES) factors – such as income, education and ethnicity – into epidemic models.

“Epidemic models typically focus on age-stratified contact patterns, but that’s only part of the picture,” Dr. Perra said.

“Our new framework recognizes that other factors, such as income and education, play an important role in how people interact and respond to public health measures. By including these SES variables, we are able to create more realistic models that better reflect the real world. epidemic consequences. »

Dr. Perra and colleagues approached this critical surveillance with a framework that uses “generalized contact matrices” to stratify contacts across multiple dimensions, including SES. This allows for a more detailed and realistic depiction of how diseases spread across different population groups, particularly those facing socio-economic disadvantages.

The study demonstrates how failing to account for these variables can lead to significant distortions of epidemic forecasts, compromising both public health strategies and policy decisions.

The team’s approach draws on both formal mathematical derivations and empirical data. Their study establishes that ignoring dimensions of SES can lead to underestimates of key parameters, such as the basic reproduction number (R₀), which measures the average number of secondary infections caused by a single infected person.

Using synthetic and real-world data from Hungary collected during the COVID-19 pandemic, researchers show how including SES indicators provides more accurate estimates of disease burden and reveals crucial disparities in outcomes between different socio-economic groups.

“The COVID-19 pandemic has been a stark reminder that the burden of infectious diseases is not borne equally across the population,” said Dr. Perra.

“Socioeconomic factors have played a decisive role in how different groups have been affected, and yet most of the epidemic models we rely on today still fail to explicitly incorporate these critical dimensions. Our framework brings these variables to the forefront, enabling more comprehensive and actionable insights. “.

The researchers demonstrated how their framework could quantify variations in adherence to non-pharmaceutical interventions (NPIs) such as social distancing and mask wearing across different SES groups. They found that neglecting these factors in models not only misrepresents the spread of disease, but also obscures the effectiveness of public health measures.

Their analysis of Hungarian data further highlighted how heterogeneities in contact patterns related to socioeconomic status can lead to substantial differences in disease outcomes between groups, highlighting the need for more targeted interventions.

“Our results suggest that future contact investigations should expand beyond traditional variables like age and include more nuanced socioeconomic data,” added Dr. Perra. “Including these factors could significantly improve the accuracy of epidemic models and, by extension, the effectiveness of health policies.”

The study highlights the urgent need for more comprehensive epidemic modeling frameworks as societies continue to combat the lingering impacts of COVID-19 and prepare for future pandemics. By moving beyond the conventional focus on age and context, this new approach opens the door to a more detailed understanding of disease transmission and offers a powerful tool to address health inequities.

This work was carried out in collaboration with Adriana Manna (Central European University), Dr. Lorenzo D’Amico (ISI Foundation), Dr. Michele Tizzoni (University of Trento) and Dr. Márton Karsai (Central and European University). Rényi Mathematics Institute). ).