When a new infectious disease emerges, one of the most critical questions is: how fast will it spread? Mathematical epidemiology provides the tools to answer this question, using equations and models to predict the trajectory of outbreaks, evaluate the impact of interventions, and guide public health decision-making. From the early work of Daniel Bernoulli on smallpox vaccination in the 18th century to the sophisticated computational models that informed COVID-19 policy, mathematics has been an indispensable weapon against epidemics.
The SIR Model: A Foundation
The cornerstone of mathematical epidemiology is the SIR model, developed by William Ogilvy Kermack and Anderson Gray McKendrick in 1927. This model divides a population into three compartments: Susceptible (S) individuals who can be infected, Infectious (I) individuals who can transmit the disease, and Recovered (R) individuals who have gained immunity. Differential equations describe how individuals move between compartments over time.
The SIR model captures a fundamental insight: an epidemic grows when each infected person transmits the disease to more than one susceptible person, and it declines when this transmission rate drops below one. This threshold behaviour is governed by the basic reproduction number, R0 — the average number of secondary infections produced by a single infected individual in a fully susceptible population.
The Basic Reproduction Number R0
R0 is perhaps the most important parameter in epidemiology. Measles has an R0 of 12-18, making it one of the most contagious diseases known. Seasonal influenza has an R0 of approximately 1.3, while the original strain of SARS-CoV-2 had an R0 estimated at 2.5-3.5. When R0 exceeds 1, an epidemic grows exponentially in its early stages; when it falls below 1, the outbreak declines. Public health interventions — vaccination, social distancing, quarantine, mask-wearing — all work by reducing the effective reproduction number below the critical threshold of 1.
The concept of herd immunity follows directly from R0. When a sufficient proportion of the population is immune (through vaccination or prior infection), the effective reproduction number drops below 1 even without individual protective measures. The herd immunity threshold is calculated as 1 – 1/R0 — for measles (R0 = 15), approximately 93 percent of the population must be immune to prevent sustained transmission.
Beyond Simple Models
Modern epidemic modelling has moved far beyond the basic SIR framework. Age-structured models account for differences in contact patterns and susceptibility across age groups — critical for diseases like influenza that disproportionately affect the elderly. Network models represent the actual social contact patterns through which diseases spread, capturing the outsized role of highly connected individuals (“superspreaders”) in driving transmission.
Spatial models incorporate geographic information, modelling how diseases spread across cities, regions, and countries via transportation networks. During the COVID-19 pandemic, mobility data from smartphones enabled real-time modelling of how changes in movement patterns affected transmission — an unprecedented integration of digital data with epidemiological modelling.
Agent-based models simulate the behaviour of millions of individual “agents” — digital representations of people — each with distinct characteristics, daily routines, and contact patterns. These models can evaluate complex, targeted interventions that simpler models cannot capture, such as the impact of school closures, workplace policies, or targeted vaccination campaigns.
Lessons from COVID-19
The COVID-19 pandemic was a defining moment for mathematical epidemiology. Models informed decisions on lockdowns, travel restrictions, hospital capacity planning, and vaccine prioritisation in virtually every country. Imperial College London’s model, projecting millions of deaths without intervention, directly influenced policy in the United Kingdom and United States.
The pandemic also exposed limitations. Models are only as good as their input data — early COVID-19 models were hampered by uncertain estimates of transmission rates, infection fatality rates, and the prevalence of asymptomatic infection. Communicating uncertainty to policymakers and the public proved challenging, with models sometimes interpreted as precise predictions rather than scenario-based projections.
Despite these challenges, mathematical epidemiology remains essential for evidence-based public health preparedness, providing the quantitative framework through which society understands, anticipates, and responds to infectious disease threats.
