The finalists
Ten universities. Ten projects. Ten academics.
Shahram Rezapour
Tuberculosis transmission in China is the focus of this mathematical case study, which builds a fractal-fractional model of the disease. The work develops existence and stability theories for the model and supports them with numerical simulations, contributing tools for understanding how such an epidemic evolves over time.
Tanja Stadler
Quantification of the spread of SARS-CoV-2 variant B.1.1.7 in Switzerland
Full Professor of Computational Evolution at ETH Zurich and president of the Swiss COVID-19 Science Advisory Panel, Tanja Stadler develops phylodynamic methods. Her cross-institution genomic surveillance study quantified the transmission advantage of the Alpha variant in Switzerland, sequencing a large share of national cases to guide public health responses.
Charles C. Y. Wang
How Much Should We Trust Staggered Difference-In-Differences Estimates?
Difference-in-differences is among the most used tools for drawing causal conclusions from observational data, and staggered adoption is the common real-world case. The work asks how much confidence such staggered estimates deserve, a question of method rather than subject matter, and therefore one that bears on findings reached across many fields.
Majid Ezzati
Association of Tropical Cyclones With County-Level Mortality in the US
Tropical cyclones have a devastating effect on society, yet a comprehensive picture of their association with deaths from specific causes over many years has been lacking. Analysing 33.6 million deaths across 1,206 US counties over 31 years using a Bayesian statistical model, the study evaluates how county-level cyclone exposure relates to monthly death rates from various causes.
James R. Carpenter
Missing data: A statistical framework for practice
Missing data are everywhere in medical research, and it is often unclear when simply dropping incomplete records is acceptable. This article offers both a practical framework and a more formal account, relating complete-case analysis, maximum likelihood, multiple imputation and Bayesian methods to one another, and showing with examples and code how multiple imputation supports the sensitivity analyses that remain rare in practice.
Gianluca Baio
Regional excess mortality during the 2020 COVID-19 pandemic in five European countries
Professor of Statistics and Health Economics and Head of Statistical Science at UCL, Gianluca Baio develops Bayesian methods for health evaluation. In a five country collaboration with Imperial College and European partners he modelled regional excess mortality during the 2020 COVID-19 pandemic, quantifying the true toll of the outbreak beyond reported deaths.
Mihaela van der Schaar
Causal machine learning for predicting treatment outcomes
Choosing the right treatment for an individual patient depends on anticipating how that particular patient, not the average patient, will respond. The study addresses causal machine learning, methods designed to estimate the effects of interventions rather than simple associations, as a route to predicting treatment outcomes in medicine.
Tadahiro Oh
Using ideas from paracontrolled calculus, the work proves local well-posedness for a renormalised version of the three-dimensional stochastic nonlinear wave equation with quadratic nonlinearity, forced by space-time white noise. Two new ingredients distinguish it from the parabolic setting: careful exploitation of dispersion at a multilinear level, and novel random operators whose regularity overcomes the lack of smoothing.
Christl Donnelly
Resurgence of SARS-CoV-2: Detection by community viral surveillance
Professor of Applied Statistics at Oxford, Christl Donnelly applies statistical epidemiology to infectious disease outbreaks. As part of the multi-institution REACT-1 collaboration with Imperial College London she analysed community swab surveillance across England, detecting the resurgence of SARS-CoV-2 in real time and informing national decisions on pandemic interventions.
Anna Heath
Economic Evaluation of Cost and Time Required for a Platform Trial vs Conventional Trials
Platform trials let new treatments join a study already under way, but little is known about what they cost to run. Surveying 146 international experts and drawing on the real entry dates of ten interventions into a long-running prostate cancer trial, this evaluation compares the money and time required for a platform trial against a series of conventional ones.
Exclusive to partners
The Subject Awards are open to partner universities.
Partners verify their data, feature their academics, and are eligible for the Subject Awards. Finalists are identified from open data; winners are decided by our editorial team and announced on 15 September 2026.
Explore partnership →The wider slate
Fifteen more candidate partnerships behind the finalists.
COVID-19 in Ethiopia: a geospatial analysis of vulnerability to infection, case severity and death
GRADE guidance 36: updates to GRADE's approach to addressing inconsistency
Matching Methods for Causal Inference with Time‐Series Cross‐Sectional Data
Mathematical modeling of COVID-19 epidemic with effect of awareness programs
Viral load and contact heterogeneity predict SARS-CoV-2 transmission and super-spreading events
Simulating Survival Data Using the <b>simsurv</b> <i>R</i> Package
A tool to assess risk of bias in non-randomized follow-up studies of exposure effects (ROBINS-E)