Influenza spreads in a crowded barracks or prison. Ebola proliferates when per-capita hospital beds plummet below the typical levels in developed countries. Bedside discrimination leads some patients to die and others to survive.
For decades, researchers have rigorously documented the way social inequality shapes the spread of disease. In parallel, mathematical models have emerged for forecasting epidemics and weighing different public health interventions. But rarely have the two subfields of epidemiology joined forces.
Now, a study links both bodies of work. A paper in Biology Letters introduces a metric for use in the most canonical disease-transmission models, quantifying the degree to which social determinants of health — environmental factors like neighborhood, access to healthcare, or economic status — cause outbreaks.
Modeling done with the new metric reveals that even when a population’s overall transmission risk is low, an epidemic can still form due to disease spread in a disadvantaged group.
“The study helps to affirm the notion that we really are in this together. The wisest public health practice is one that helps everyone prevent the spread of infection,” says SFI Resident Professor Brandon Ogbunu, the paper’s senior author, also an associate professor of ecology and evolutionary biology at Yale University. “This research implores us to distribute resources in infectious disease control equitably across a population, because it ends up benefiting everyone.”
Ogbunu and collaborators, with expertise in epidemiology, statistics, and data science, developed the “structural causal influence” metric to quantify how social factors like crowded living conditions, limited vaccine access, and financial barriers directly shape the way disease spreads — and whether community interventions will actually work.
Rather than integrating their metric into the AI neural networks currently in vogue, the researchers chose to use popular mathematical models like SIR (Susceptible-Infected-Removed) so that humans could more easily read and interpret the results. Adding complexity to these simple models is crucial: ignoring social inequality could lead to grave errors when handling health emergencies.
“A lot of people think of interdisciplinary research as dilettante or big-picture science. Here, we’re showing that different kinds of thinkers can come together to make a deep intellectual contribution that wouldn’t otherwise be possible, with great potential to generalize. This framework could help unify modeling of public health threats beyond epidemics, like chronic disease, pollution, or climate change,” says SFI External Professor Sam Scarpino (Northeastern University), a public health scientist and co-author on the paper.
Already, the new metric can assist decision-makers in weighing tradeoffs, and help advocates show that allocating more public-health resources to their communities benefits the whole population.
“We think social determinants should be central in conversations about how epidemics happen and what amplifies them,” says Ogbunu. “Our study offers an easily computable formula for that, using little more than the sorts of data we’ve already been collecting to develop classical epidemiological models.”
Read the paper "Structural causal influence (SCI) captures the forces of social inequality in models of infectious disease" in Biology Letters (August 19, 2026). DOI: 10.1098/rsbl.2025.0655