Contact
Joseph Mihaljevic, PhD
SHERC Research Project
SSCIMA: Integrating analysis of socio-economic sub-population dynamics to improve spatial models of infectious disease
By the summer of 2020, serious health disparities associated with COVID-19 were becoming apparent. Nearly 50 percent of detected COVID-19 infections in Arizona were associated with the Latinx population and 13 percent were associated with the American Indian population, even though these populations make up only 31.7 percent and 3.9 percent of the overall population, respectively.
Through their SHERC project, Joseph Mihaljevic and his co-investigators are dedicated to addressing these health disparities by developing effective prediction models for infectious diseases, like COVID-19, that accurately examine small spatial scales that affect transmission rates and disease outcomes in local communities.
Mihaljevic is an assistant professor in the School of Informatics, Computing, and Cyber Systems, and his co-investigators include Eck Doerry, professor, School of Informatics, Computing and Cyber Systems and Ye Chen, assistant professor, Department of Mathematics and Statistics.
Project aims
- Develop a geostatistical analysis method to discover “socio-spatially similar” sub-populations and quantify transmission linkages between them.
- Design algorithms that enable accurate and automated parameter estimation for complex epidemiological models.
- Engage community participation by developing outreach and educational elements.
“Our overall objective is to create better disease forecasts at spatial scales that are most relevant for decision-making [about the spread of COVID-19 or other pandemics]. We want models to be as accurate as possible so that public health stakeholders are empowered to use these model predictions in a useful way,” Mihaljevic said. “Moreover, we want these models to reflect the fact that populations differ in meaningful ways that can lead to health disparities. By bringing together experts in infectious disease, computer science, and mathematics, we hope to create usable and actionable technologies that will prepare us for future disease outbreaks.”
Their goal is to improve their spatial models to fit local disease data while also improving the accuracy of local forecasts. They also plan to eventually create software tools and outreach programs that will train new users to implement their disease modeling methods to respond to a variety of diseases.
Giving back to science and the community
The team hopes to turn their technical developments into freely available software and to create an online testing bench for modelers.
To assist communities, they plan to develop a 90-minute, freely available webinar to build on the success of SHERC’s earlier Infectious Disease Epidemiology Bootcamp, which hosted a series of web-based modules that assisted with immediate public health workforce development during the COVID-19 pandemic.
The course will offer an introduction to epidemiological modeling, how to interpret common model outputs, and how to critically evaluate whether a model may or may not be relevant to some specific local context.
They will also design an in-person half-day workshop to hold at academic conferences that train modelers to use our software package and online testing bench. It will include scenario-based training that demonstrates how to use our software to respond during a novel epidemic.
Funding: The study is funded by NIMHD/NIH U54MD012388
About the investigators
