🏢 Conceptual Overview
Classical epidemic models typically assume that all individuals are equally susceptible to infection. In reality, biological, behavioral, and social differences create substantial heterogeneity in vulnerability. The Distributed Susceptibility (Frailty) SIR Model extends the standard SIR framework by explicitly incorporating individual-level variation in susceptibility, referred to as frailty.
This framework explains a key empirical observation in epidemics: outbreaks often decelerate faster than predicted by homogeneous models. Highly frail individuals are infected early, leaving a residual susceptible population that is, on average, more resilient. As a result, transmission slows even when a large fraction of individuals remains uninfected.
🏗️ Compartmental Structure and Flow
The total population is structured by a continuous frailty variable x, representing relative susceptibility to infection.
- Distributed Susceptibles, S(x, t)
The susceptible population is described as a distribution over frailty levels x. Individuals with larger x values are more likely to acquire infection upon exposure. - Infected, I(t)
Individuals who have become infected. In the basic formulation, disease progression does not depend on the original frailty level. - Recovered, R(t)
Individuals who have cleared the infection and acquired immunity.
Flow Structure
Susceptible individuals transition to infection at rates proportional to their frailty. Over time, the susceptible pool becomes progressively enriched with low-frailty individuals, dynamically reshaping population-level risk.
🧮 Mathematical Formulation
The dynamics of the frailty-structured SIR model are described by the following system.
Evolution of the Susceptible Distribution
dS(x, t) / dt
= − x · λ(t) · S(x, t)
Infected Population Dynamics
dI(t) / dt
= ∫ x · λ(t) · S(x, t) dx − γ · I(t)
Force of Infection
λ(t)
= β · I(t) / N
Recovered Population Dynamics
dR(t) / dt
= γ · I(t)
In this formulation, infection acts as a selective process, preferentially removing individuals with higher frailty from the susceptible pool.
🌤️ Weather-Driven Frailty Scaling
Environmental conditions can modulate effective susceptibility by influencing viral stability and host defenses. A common approach is to model the transmission coefficient β as a function of absolute humidity AH:
β(AH)
= β₀ · exp( − α · AH(t) )
Lower humidity increases airborne viral survival and mucosal vulnerability, effectively amplifying the operational frailty of the population even when intrinsic susceptibility remains unchanged.
📋 Table 1. Parameter Definitions
| Parameter | Definition |
|---|---|
| S(x, t) | Density of susceptible individuals with frailty x at time t |
| I(t) | Number of infected individuals |
| R(t) | Number of recovered individuals |
| x | Individual frailty (relative susceptibility) |
| λ(t) | Force of infection |
| β | Base transmission coefficient |
| β₀ | Baseline transmission under dry conditions |
| γ | Recovery rate |
| α | Environmental sensitivity coefficient |
| AH(t) | Time-varying absolute humidity |
| N | Total population size |
📊 Table 2. Typical Parameter Ranges
| Parameter | Typical Range | Interpretation |
|---|---|---|
| β | 0.2 – 1.0 day⁻¹ | Baseline transmission intensity |
| γ | 0.1 – 0.33 day⁻¹ | Infectious period of 3–10 days |
| Mean(x) | 1.0 (normalized) | Average susceptibility |
| Var(x) | 0.1 – 2.0 | Degree of heterogeneity |
| α | 0.05 – 0.5 | Sensitivity to humidity |
| N | Fixed | Population size |
High variance in frailty indicates strong heterogeneity, where a small fraction of highly susceptible individuals drives early transmission.
🎯 Applicability and Limitations
Applicability
• Explaining rapid epidemic deceleration without full susceptible depletion
• Correcting overestimates of herd immunity thresholds
• Interpreting vaccine efficacy and trial outcomes
• Modeling endemic persistence in heterogeneous populations
Key Assumptions and Weaknesses
• Assumes frailty is constant over time
• Frailty distributions are difficult to infer from incidence data
• Often conflates biological susceptibility with behavioral exposure
• Typically assumes proportionate mixing
Despite these challenges, frailty-based models provide critical insight into real-world epidemic dynamics that homogeneous models cannot capture.
📚 References
- Vaupel, J. W., Manton, K. G., & Stallard, E. (1979). The impact of heterogeneity in individual frailty on the dynamics of mortality. Demography.
- Diekmann, O., & Heesterbeek, J. A. P. (2000). Mathematical Epidemiology of Infectious Diseases: Model Building, Analysis and Interpretation. Wiley.
- Gomes, M. G. M., et al. (2020). Individual variation in susceptibility or exposure to SARS-CoV-2 lowers the herd immunity threshold. medRxiv / Journal of Theoretical Biology.
- Novozhilov, A. S. (2008). On the spread of epidemics in a closed heterogeneous population. Mathematical Biosciences.
🌲 Analogy for Clarity
Imagine a forest in which trees have different levels of dryness. A standard model assumes all trees are equally dry. In the distributed susceptibility model, the fire rapidly consumes the driest trees first. Even if many trees remain, the fire slows or stops because the remaining forest is greener and more resistant. The epidemic behaves the same way: transmission weakens as the most vulnerable individuals are removed from the susceptible pool.