We answer questions about the biological, behavioural, and environmental (health systems and structural) mechanisms that underpin infectious disease epidemics. We develop and integrate causal inference and mathematical models with the best available data to test hypotheses and to help inform public health and policy decisions under a Program Science framework. Our primary focus is on HIV and other sexually transmitted infections (STI), and more recently, on COVID-19 and Ebola virus disease.

Program Science is a scholarly field that emerged from what public health programs were doing, observing, and asking. It is a unified scope of scientific inquiry embedded in public health programs. A unique feature of Program Science is its focus on emergent phenomena in complex adaptive systems — systems which can be studied with mathematical models.

**Our objectives are to:**
1. Appraise Epidemics (“know your epidemic”): to understand why epidemics establish and persist where and when they do, and what leads to differences in their trajectories, size, and characteristics across regions.
2. Maximize Program Impact (“plan the response”): to optimize the design and delivery of public health programs by health-system and epidemic context.
3. Forecast Data Priorities (“monitor and adapt the response”): to systematically assess the influence of data uncertainty on model projections to prioritize future data collection.

Overview: Mathematical Modeling and Program Science

We are working to understand the upstream causes and epidemic consequences of heterogeneity in the risks of HIV/STI acquisition and transmission. Heterogeneity may reflect variability in biology, behavior, partnership-level features, and structural pathways that interact and modify each of the former. We are studying between-person [heterogeneity](https://pubmed.ncbi.nlm.nih.gov/25014543/), within-person heterogeneity, and network-level [heterogeneity](https://pubmed.ncbi.nlm.nih.gov/32589341/), focusing on “places” as proxies for socio-sexual networks. The focus of these projects has been on epidemics of HIV, syphilis, [mpox](https://pubmed.ncbi.nlm.nih.gov/39260880/), and more recently on [COVID-19](https://pubmed.ncbi.nlm.nih.gov/40069958/).

**We seek to:**
- Creatively explore and disentangle the role of – and interactions between – social and structural inequalities in establishing and sustaining epidemics
- Identify the conditions under which fixed resources could be harnessed to reduce health inequalities
- Develop, test, and validate new ways of estimating causes and consequences of heterogeneity

Examples include our body of work examining approaches to quantify the relative “contribution” of unmet needs (prevention and treatment gaps) to onward HIV/STI transmission (transmission population attributable fraction, [tPAF](https://pubmed.ncbi.nlm.nih.gov/34189863/)). Projects are conducted in close partnership with academic colleagues, public health decision-makers, and community partners in Ukraine, Kenya, South Africa, Eswatini, and Sierra Leone; and with the UNAIDS Reference Group.

*Supported by: Canadian Institutes of Health Research, National Institutes of Health Research, Ontario Ministry of Science Early Career Researcher Award, Ontario HIV Treatment Network.*

Causes and Consequences of Heterogeneity in Epidemics

This body of work focuses on how we draw inferences from “what if” scenarios in mathematical models and in observational epidemiological studies. We test the mechanisms that could lead to systematic biases, identify the conditions under which these biases could “matter”, and then try to develop methodological solutions to reduce these biases. To do this, we use simulation studies alongside causal inference methods. 

**We seek to:**
- Creatively explore the biological and network-level properties that could lead to biases in model-based inference or answers
- Identify the conditions under which these biases matter (e.g., change decisions)
- Develop, test, and validate approaches to reduce biases

Examples include our body of work [examining](https://pubmed.ncbi.nlm.nih.gov/37715459/) mixing [patterns](https://pubmed.ncbi.nlm.nih.gov/35004190/) (“who contacts whom”), [causes](https://pubmed.ncbi.nlm.nih.gov/40413178/) of false negative vaccine effectiveness, [turnover](https://pubmed.ncbi.nlm.nih.gov/32913937/) or periodic risks, and new ways of looking at herd immunity using empirical data. 

*Supported by: Canadian Institutes of Health Research, National Institutes of Health, Natural Sciences and Engineering Research Council Emerging Infectious Disease Modeling Networks, University of Toronto Data Sciences Institute and Temerty Centre for AI Research and Education in Medicine, and the Canadian Immunization Research Network, Ontario Ministry of Science Early Career Researcher Award, St. Michael’s Hospital Foundation Research Innovation Council.*

Reducing Biases in Mathematical and Observational Epidemiological Studies

We work with communities and community scholars with lived experience through a range of approaches, from engagement with community advisory boards to community-led mathematical modeling with participatory methods. We are trying to ground our modeling in lived experience, to democratize knowledge, and to re-think theories, knowledge systems, and the power structures that govern a model’s architecture and its use in public health decision-making. 

**We seek to:**
- Re-imagine the “how” of developing and analyzing mathematical models with communities affected by the health conditions under study and affected by the decisions that stem from models
- Co-developing and applying approaches and frameworks for community-led mathematical modeling
- Co-develop new theories and model architectures that draw on various disciplines, various knowledge systems, and are situated in the local context 

Under the leadership of community partners in Kenya, led by Jeffrey Walimbwa at the Community Technical and Research Support Hub, we have embarked on the [co-development](https://journals.ub.umu.se/index.php/jcsh/article/view/1151) of a framework for community participatory mathematical modeling of HIV/STI epidemics. We are working with a community expert group on modeling COVID-19 in the shelter system in Toronto, Canada; with a community advisory board on modeling of mpox in Ontario, Canada; and partnering with the Center for Community Based Research and a focused community advisory board on modeling syphilis epidemics in British Columbia, Canada. We have also been working with Chiefs of Ontario to co-develop a COVID-19 [scenario model](https://nccid.ca/webcast/voices-of-sovereignty/). In partnership with Well Living House, we are collaborating on efforts led by Indigenous scholars and elders on decolonize mathematical modeling. 

*Supported by: New Frontiers in Research Exploration Fund, Canadian Institutes of Health Research, Temerty Faculty of Medicine Knowledge Translation Grant.*

Research Focus Areas