Force of Infection
The force of infection is the instantaneous per-capita hazard rate at which susceptible individuals in a population acquire an infectious disease, linking pathogen prevalence, contact patterns, and age-specific seroprevalence dynamics.
Force of infection is the instantaneous per-capita hazard rate at which susceptible individuals in a population acquire an infectious pathogen per unit of time or chronological age. Commonly denoted by the Greek letter lambda, λ(a, t), this epidemiological parameter represents the conditional probability per unit time that an uninfected person of age a becomes infected at calendar time t. Unlike the incidence rate, which divides new infections by the total population or total person-time at risk, the force of infection conditions specifically on the susceptible compartment and scales directly with the prevalence of infectious hosts or environmental reservoirs in compartmental models in epidemiology.
Mathematical Formulation and Transmission Mechanics
Survival-Analytic Hazard Definition
In mathematical epidemiology and survival analysis, acquisition of an infection is treated as a time-to-event process in which birth or entry into a risk cohort marks time zero and seroconversion or clinical infection marks the event. If T denotes the random age or time at first infection, the force of infection λ(a) for a stationary endemic pathogen is defined as the hazard function of T:
λ(a) = limΔa → 0 [P(a ≤ T < a + Δa | T ≥ a) / Δa]
Because λ(a) is a rate rather than a dimensionless probability, its physical dimension is inverse time (such as year−1, month−1, or day−1). The probability that an individual remains uninfected up to age a—the susceptible survival fraction S(a)—is obtained by exponentiating the negative cumulative force of infection from birth to age a:
S(a) = exp(−∫0a λ(u) du)
When λ is constant across all age groups and over calendar time, the duration spent in the susceptible state follows an exponential distribution with mean age at first infection equal to A = 1 / λ. In an endemic setting with lifespans characterized by expectation of life L, the basic reproduction number R0 for a lifelong immunizing infection in a homogeneously mixing population is approximated by R0 ≈ 1 + L / A = 1 + λL.
Frequency-Dependent versus Density-Dependent Transmission
In dynamic transmission models such as the Susceptible-Infectious-Recovered (SIR) framework, the force of infection is not a fixed constant like a mortality hazard; instead, it is an endogenous function of the infectious population I(t). For directly transmitted human respiratory and sexually transmitted infections, contact rates normally saturate with population size, yielding frequency-dependent (mass-action) transmission:
λ(t) = β · [I(t) / N(t)]
Here, N(t) is total population size and β is the effective transmission rate, which decomposes into the average per-capita contact rate c multiplied by the conditional transmission probability per infectious contact p (β = cp). Conversely, for wildlife zoonoses or airborne livestock pathogens where contact frequency increases linearly with spatial crowding, density-dependent transmission defines the hazard as λ(t) = βI(t). For vector-borne infections such as malaria or dengue transmitted by Culicidae, the human force of infection is governed by the entomological inoculation rate (EIR) multiplied by the sporozoite or virion transmission efficiency per bite: λH = ma b z, where m is mosquito density per human, a is the human-biting frequency, b is vector-to-human susceptibility, and z is the infectious proportion of the vector population.
| Epidemiological Metric | Mathematical Expression | Denominator / Conditioning Set | Physical Units | Primary Epidemiological Role |
|---|---|---|---|---|
| Force of Infection (λ) | βI(t) / N(t) | Susceptible person-time at risk | Time−1 (e.g., year−1) | Measures individual infection hazard and transmission intensity |
| Cumulative Force of Infection (Λ) | ∫0a λ(u) du | Susceptible cohort across age span | Dimensionless (cumulative hazard) | Determines cumulative seroprevalence P(a) = 1 − exp(−Λ(a)) |
| Incidence Rate | New cases / Total person-time | Total population person-time | Cases · person-time−1 | Tracks population-level disease burden and healthcare demand |
| Attack Rate (Cumulative Incidence) | 1 − exp(−∫ λ(t) dt) | Initially susceptible cohort | Dimensionless proportion (0 to 1) | Quantifies cumulative fraction infected over an outbreak |
| Basic Reproduction Number (R0) | β / γ (or ≈ 1 + λL) | Index case in naive population | Secondary cases per index case | Establishes threshold for epidemic growth and herd immunity |
Catalytic Models and Seroepidemiological Estimation
Muench's Catalytic Framework
Because many viral and parasitic infections produce subclinical or asymptomatic cases, direct case notification underestimates the true force of infection. In 1934 and 1959, Hugo Muench introduced the catalytic model, drawing an analogy between chemical reaction kinetics and the conversion of seronegative individuals into seropositive individuals with age. In a cross-sectional serosurvey where lifelong IgG antibodies mark prior exposure, the age-specific seroprevalence π(a) = 1 − S(a) obeys the ordinary differential equation:
dπ(a) / da = λ(a) · [1 − π(a)]
Solving this equation allows epidemiologists to estimate age-dependent transmission rates directly from cross-sectional blood samples:
λ(a) = − [d ln(1 − π(a)) / da] = [dπ(a) / da] / [1 − π(a)]
Reversible Catalytic Models and Waning Immunity
For pathogens where antibody titers decay over time or where reinfection occurs (such as Plasmodium falciparum, pertussis, or seasonal coronaviruses), a reversible catalytic model incorporates a seroreversion rate ρ (with units of time−1):
dπ(a) / da = λ · [1 − π(a)] − ρ · π(a)
Under constant transmission and reversion rates, seroprevalence approaches an asymptotic plateau below 100 percent at older ages:
π(a) = [λ / (λ + ρ)] · [1 − exp(−(λ + ρ)a)]
WAIFW Matrices and Age-Structured Mixing
Childhood infections such as measles, rubella, varicella, and mumps exhibit pronounced age variation in λ(a), peaking sharply between ages 5 and 12 due to assortative mixing in primary schools. Roy Anderson and Robert May formalized age-structured transmission using the Who-Acquires-Infection-From-Whom (WAIFW) matrix β(a, a′), which represents the effective contact rate between susceptible individuals of age a and infectious individuals of age a′:
λ(a, t) = ∫0∞ β(a, a′) I(a′, t) da′
Empirical social contact diaries (such as the European POLYMOD survey) combine with age-stratified serological profiles via generalized additive models, fractional polynomials, or non-parametric Nelson–Aalen estimators to reconstruct λ(a) without imposing rigid parametric assumptions.
Public Health Applications and Vaccination Paradoxes
Age-Shifting Under Mass Immunization
Mass vaccination campaigns reduce the prevalence of infectious individuals I(t), which proportionally lowers the force of infection λ experienced by remaining unvaccinated individuals—the mechanistic basis of herd immunity. However, reducing λ directly increases the average age at first infection (A ≈ 1 / λ) among those who remain susceptible.
For diseases where clinical severity increases with host age, an incomplete vaccination campaign that lowers λ without achieving elimination can paradoxically increase the absolute number of severe outcomes. Congenital rubella syndrome (CRS) provides the classic illustration: if routine infant rubella vaccination achieves moderate coverage (for example, 40 to 60 percent), the reduced force of infection shifts the mean age of rubella acquisition from early childhood into the reproductive years (ages 15 to 35). Unless vaccine uptake exceeds a threshold (typically 80 percent or higher, combined with catch-up campaigns in adolescent and adult females), the incidence of intrauterine rubella infection during pregnancy can exceed pre-vaccination levels.
Frequently Asked Questions
How does the force of infection differ from the incidence rate?
The force of infection divides new infections solely by the susceptible person-time at risk, measuring the per-capita hazard faced by an uninfected individual. The population incidence rate divides new infections by the entire population at risk (frequently including immune or previously infected individuals who cannot contract the disease again), measuring overall community burden rather than individual exposure hazard.
Why can the force of infection exceed 1.0 per year?
Because the force of infection is an instantaneous hazard rate rather than a probability, its value is bounded below by zero but has no upper bound of one. A force of infection of λ = 2.0 year−1 implies that the average waiting time to infection is 0.5 years (6 months), corresponding to an annual infection probability of 1 − exp(−2.0) ≈ 86.5 percent for a susceptible person exposed for a full year.
How is the force of infection estimated when disease reporting is incomplete?
Epidemiologists estimate the force of infection from cross-sectional seroprevalence surveys by measuring the rate at which the seronegative fraction declines with age. Using catalytic models, generalized linear models with a complementary log-log link, or survival-analytic estimators for interval-censored data, researchers reconstruct historical transmission intensity without relying on clinical case notifications.
References
- Muench, H. (1959). Catalytic Models in Epidemiology. Harvard University Press. https://doi.org/10.4159/harvard.9780674428928
- Anderson, R. M., & May, R. M. (1991). Infectious Diseases of Humans: Dynamics and Control. Oxford University Press. https://doi.org/10.1093/oso/9780198545996.001.0001
- Hens, N., Aerts, M., Faes, C., Shkedy, Z., Lejeune, O., Van Damme, P., & Beutels, P. (2010). Seventy-five years of estimating the force of infection from current status data. Epidemiology and Infection, 138(6), 802–812. https://doi.org/10.1017/S0950268809990781
- Mossong, J., Hens, N., Jit, M., Beutels, P., Auranen, K., Mikolajczyk, R., ... & Edmunds, W. J. (2008). Social contacts and mixing patterns relevant to the spread of infectious diseases. PLOS Medicine, 5(3), e74. https://doi.org/10.1371/journal.pmed.0050074