Research
My research sits at the intersection of population health, demography, epidemiology, and statistical computing. I am especially interested in how better methods and better data systems can improve measurement, surveillance, and decision-making in global health.
The themes below summarize the main threads of my work. Each section includes a short overview together with selected publications and direct links to papers and downloads.
AI in Global Health
I study how language models and multimodal machine learning can strengthen health measurement in low-resource settings. This work focuses on verbal autopsy analysis, where narrative text and structured symptom data can be combined to improve cause-of-death assignment and make health estimates more timely and scalable.
Key publications
Leveraging Language Models and Machine Learning in Verbal Autopsy Analysis
This paper examines how modern language models can learn from verbal autopsy narratives and how multimodal pipelines can combine text with structured questionnaire responses for more reliable cause-of-death classification.
Disease Surveillance and Global Epidemiology
I use surveillance, survey, and modeled mortality data to study how health risks and causes of death change across place and time. The goal is to produce population-level evidence that is locally grounded enough to guide policy, while still allowing meaningful cross-country comparison.
Key publications
Temporal changes in cause of death among adolescents and adults in six countries in eastern and southern Africa in 1995-2019
A multi-country surveillance study using verbal autopsy data to trace how cause-specific mortality changed across eastern and southern Africa, with a clear picture of shifting burdens by age, sex, site, and period.
National and sub-national levels and causes of mortality among 5-19-year-olds in China in 2004-2019
This analysis maps mortality patterns among older children and adolescents in China, highlighting regional disparities, changing injury profiles, and the importance of sustainable surveillance systems.
Estimating seroprevalence of SARS-CoV-2 in Ohio: A Bayesian multilevel poststratification approach with multiple diagnostic tests
This work develops a Bayesian multilevel poststratification framework for estimating SARS-CoV-2 seroprevalence under low positivity, imperfect diagnostics, and nonresponse, showing how rigorous statistical modeling can strengthen public health surveillance.
Global, regional, and national causes of under-5 mortality in 2000-15: an updated systematic analysis with implications for the Sustainable Development Goals
An updated global cause-of-death analysis for children under five, designed to support SDG-era priority setting by showing where preventable child deaths remain concentrated.
Demographic Methodology
I build demographic estimation methods for settings where data are incomplete, sparse, or uneven across age groups. Much of this work uses Bayesian modeling to recover age-specific mortality and fertility schedules with transparent uncertainty, so estimates remain useful for planning and evaluation.
Key publications
A flexible Bayesian framework to estimate age- and cause-specific child mortality over time from sample registration data
This paper proposes a flexible Bayesian framework for estimating child mortality jointly across age, cause, and time, offering a principled alternative to fragmented multistage workflows.
Adapting and validating the log quadratic model to derive under-five age- and cause-specific mortality (U5ACSM): a preliminary analysis
A methodological extension of the log-quadratic model that improves estimation of under-five mortality by both age and cause, using China sample registration data as an empirical test bed.
SVD-Bayes: A Singular Value Decomposition-based Approach Under Bayesian Framework for Indirect Estimation of Age-specific Fertility and Mortality
My master's thesis introduces an SVD-based Bayesian approach for recovering detailed age schedules of fertility and child mortality from summary birth history data, with uncertainty estimates included from the start.
Data Quality in Population Surveys
I examine how measurement error enters demographic and health data, especially when survey instruments ask respondents to reconstruct difficult or sensitive life events. This research looks closely at reporting processes, classification error, and validation against surveillance data to improve the reliability of population estimates.
Key publications
Provider-mother interactions are associated with birth outcome misclassifications in household surveys: A case-control study in Guinea-Bissau
This study shows how provider-mother interactions can shape whether births are later reported as stillbirths or neonatal deaths in household surveys, with implications for the validity of mortality measurement.
Biases in survey estimates of neonatal mortality: results from a validation study in urban areas of Guinea-Bissau
A validation study comparing survey reports against HDSS records to show how reporting error can systematically bias neonatal mortality estimates in low-resource settings.
Health Outcome Research
Alongside demographic and population-health methods, I have contributed to applied health outcomes research that synthesizes evidence for clinical decision-making. This work reflects a broader interest in connecting rigorous analysis with questions that matter directly for patient care and health systems.
Key publications
Diabetes medications as monotherapy or metformin-based combination therapy for type 2 diabetes: a systematic review and meta-analysis
This systematic review and meta-analysis compares common medication strategies for type 2 diabetes, helping clarify the evidence base behind treatment choice and escalation.