Questions: SEIR Models Incorporating Latent Periods

Question 1

Compared to a SIR model with the same transmission rate β and recovery rate γ, what does adding an Exposed compartment (SEIR) change?

• A. R₀ increases because the latent period extends total time an individual affects transmission
• B. R₀ decreases because exposed individuals are not yet infectious
• C. R₀ is unchanged, but the epidemic grows more slowly and peaks later
• D. R₀ is unchanged and the epidemic curve is identical to the SIR model

Answer: (C)

R₀ = β/γ depends only on the transmission rate and recovery rate — not on the latent period. Adding the E compartment does not change how many people one infectious person ultimately infects; it only delays when new infections become infectious. The epidemic curve is stretched out: initial growth is slower because new infections flow through E before becoming I, and the peak occurs later and slightly lower. Options A and B are common misconceptions — the latent period is a delay, not a multiplier or divider of transmission potential.

Question 2

A public health team detects the first 10 confirmed infectious cases of a novel respiratory disease with an estimated latent period of 6 days. About how far ahead of this observation is the true epidemic?

• A. The epidemic is approximately at the same point — confirmed cases track actual infections closely
• B. The epidemic is approximately 6 days ahead, because cases are not observable until they become infectious
• C. The epidemic is approximately 6 days behind, because exposed individuals will become cases in the future
• D. The latent period tells us nothing about the gap between the epidemic and observed cases

Answer: (B)

In SEIR, individuals in the E compartment are infected but not yet detectable as cases (they are not yet infectious and typically not yet symptomatic). The lag between true infection and observable cases is approximately 1/σ — the mean latent period. With σ such that 1/σ = 6 days, the epidemic is roughly 6 days ahead of what the current case count reflects. This is why knowing the latent period is critical for early warning systems: it tells you how far the epidemic has already progressed beyond what you can see.

Question 3

In a SEIR model, the epidemic peak occurs later and is slightly lower than in a SIR model with identical R₀.

• T. True
• F. False

Answer: True

This is correct. The E compartment introduces a delay in the feedback loop: newly infected individuals must pass through the Exposed phase before becoming Infectious and transmitting further. This slows the initial exponential growth, delays the buildup of the Infectious compartment, and consequently delays and slightly flattens the epidemic peak relative to a SIR model with the same R₀ and parameters. The final epidemic size (total fraction infected) is similar, but the trajectory differs.

Question 4

Adding an Exposed (E) compartment to the SIR model increases the basic reproduction number R₀ because infected individuals now spend more total time in the system before recovering.

• T. True
• F. False

Answer: False

R₀ = β/γ is unchanged by the E compartment. R₀ depends on the transmission rate β (how fast S→I contact occurs) and the recovery rate γ (how fast I→R occurs). The latent period 1/σ adds a delay before an exposed person becomes infectious, but it does not affect how many secondary infections one infectious person causes during the infectious period. The E compartment changes the timing of the epidemic, not its reproductive potential.

Question 5

Why does adding an Exposed compartment not change R₀ but does change the initial epidemic growth rate? What is the relationship between these two quantities?

Model answer: R₀ measures the average number of secondary cases produced by one infectious individual in a fully susceptible population — it depends only on β (transmission rate) and γ (recovery rate). The initial growth rate r of the epidemic also depends on σ (the rate of leaving the E compartment): r is smaller in SEIR than in SIR with the same R₀ because new infections must pass through E before contributing to transmission. R₀ and r are related but distinct: a given R₀ can correspond to different growth rates depending on the generation interval structure.

This distinction matters practically. Two diseases with the same R₀ but different latent periods will grow at different rates and have different intervention windows. A disease with a long latent period (like tuberculosis) grows slowly even with a high R₀, giving more time for intervention. A disease with a short latent period grows quickly even with a modest R₀, leaving little time. This is why epidemiologists need both R₀ and the generation interval — not just R₀ alone — to characterize an outbreak's dynamics.