📉 Conceptual Overview
Infectious disease transmission rarely occurs in a homogeneous population. Real epidemics are shaped by age-specific biological susceptibility, social behavior, and contact patterns. The Age-Structured SIR Model explicitly incorporates these heterogeneities by partitioning the population into discrete age cohorts and coupling them through age-dependent contact rates. This framework is fundamental for designing targeted interventions such as age-specific vaccination or school-based control measures.
🏗️ Compartmental Structure and Flow
The total population is divided into n distinct age groups (for example, infants, school-age children, adults, and elderly). Within each age group i, individuals occupy one of three epidemiological states:
- Susceptible (Sᵢ)
Individuals in age group i who are at risk of infection. - Infected (Iᵢ)
Individuals in age group i who are currently infectious. - Recovered (Rᵢ)
Individuals in age group i who have acquired immunity.
Within each age group, individuals progress through the classical S → I → R pathway. Age groups are dynamically linked through a contact matrix, which governs transmission between groups. In addition, demographic aging moves individuals from age group i to i+1 over time, coupling the compartments across age classes.
🧮 Mathematical Formulation
The Age-Structured SIR model consists of a system of 3n ordinary differential equations. For each age group i = 1, … , n, the dynamics are given by:
Susceptible Population
dSᵢ(t) / dt
= Bᵢ − ∑ⱼ₌₁ⁿ βᵢⱼ Sᵢ(t) Iⱼ(t) / Nⱼ(t) − ( μᵢ + aᵢ ) Sᵢ(t) + aᵢ₋₁ Sᵢ₋₁(t)
Infected Population
dIᵢ(t) / dt
= ∑ⱼ₌₁ⁿ βᵢⱼ Sᵢ(t) Iⱼ(t) / Nⱼ(t) − ( γᵢ + μᵢ + aᵢ ) Iᵢ(t) + aᵢ₋₁ Iᵢ₋₁(t)
Recovered Population
dRᵢ(t) / dt
= γᵢ Iᵢ(t) − ( μᵢ + aᵢ ) Rᵢ(t) + aᵢ₋₁ Rᵢ₋₁(t)
These equations describe infection, recovery, natural mortality, and demographic aging simultaneously. The coupling term βᵢⱼ links susceptibility in age group i to infectious individuals in age group j through structured contact patterns.
🌤️ Integration of Climatic Variables
Transmission intensity often varies with environmental conditions such as temperature or humidity. This effect can be incorporated by allowing the transmission coefficient to depend on a climatic variable W. A commonly used formulation is a Gaussian-shaped thermal performance curve:
β(W) = βₘₐₓ · exp [ − 0.5 · ( ( W − Wₒₚₜ ) / σ )² ]
In this representation, transmission peaks at an optimal climatic condition and declines smoothly as conditions deviate from that optimum. As a result, the model becomes non-autonomous, with time-varying parameters driven by environmental dynamics.
📋 Table 1. Parameter Definitions
| Parameter | Definition |
|---|---|
| Sᵢ | Number of susceptible individuals in age group i |
| Iᵢ | Number of infectious individuals in age group i |
| Rᵢ | Number of recovered individuals in age group i |
| Nᵢ | Total population size of age group i |
| Bᵢ | Entry or birth rate into age group i |
| βᵢⱼ | Transmission coefficient between age groups i and j |
| γᵢ | Recovery rate in age group i |
| μᵢ | Natural mortality rate in age group i |
| aᵢ | Aging rate from age group i to i + 1 |
| W | Climatic variable (e.g., temperature or humidity) |
| Wₒₚₜ | Optimal climatic value for transmission |
| σ | Climatic sensitivity (breadth) parameter |
| βₘₐₓ | Maximum transmission rate |
📊 Table 2. Typical Parameter Ranges
| Parameter | Typical Range | Interpretation |
|---|---|---|
| βᵢⱼ | 0.01 – 1.0 | Depends strongly on contact intensity; often higher among children |
| γᵢ | 0.07 – 0.5 day⁻¹ | Corresponds to infectious periods of 2–14 days |
| aᵢ | 1 / (duration of age class) | For a 10-year age class, approximately 1 / 3650 day⁻¹ |
| μᵢ | 10⁻⁵ – 10⁻³ day⁻¹ | Strongly age-dependent, higher in infants and elderly |
| σ | 1 – 10 (units of W) | Determines sensitivity to climatic variation |
🎯 Applicability and Limitations
Applicability
• Pediatric and childhood infections where school-age contacts dominate transmission
• Evaluation of age-targeted vaccination strategies and indirect herd effects
• Diseases with strong age-specific morbidity and mortality patterns
Key Assumptions and Weaknesses
• Assumes fixed or proportionate mixing within and between age groups
• Assumes permanent immunity unless explicitly extended to SIRS dynamics
• Requires detailed contact-matrix data, which are difficult to measure accurately
• Parameter dimensionality increases quadratically with the number of age groups, raising risks of over-fitting
📚 References
1. Anderson, R. M., & May, R. M. (1991). Infectious Diseases of Humans: Dynamics and Control. Oxford University Press.
2. Schenzle, D. (1984). An age-structured model of pre- and post-vaccination measles transmission. IMAMedical Genetics.
3. Vynnycky, E., & White, R. G. (2010). An Introduction to Infectious Disease Modelling. Oxford University Press.
4. Iannelli, M. (1995). Mathematical Theory of Age-Structured Population Dynamics. Giardini Editori e Stampatori.
5. Keeling, M. J., & Rohani, P. (2008). Modeling Infectious Diseases in Human and Animals. Princeton University Press.
🔍 Intuitive Analogy
The population can be visualized as a system of connected reservoirs, each representing an age group. The pipes between reservoirs correspond to contact patterns, with varying widths reflecting interaction intensity. Environmental conditions act like a valve controlling flow through the system, amplifying or dampening transmission as climate changes.