Herd immunity

Concept

Herd immunity

T-117Home BU-403Threads systems · evolution

Statement

The immune fraction that halts transmission.

Why it matters

basic-reproduction-number establishes \(R_0\), the expected number of secondary infections produced by one case in a fully susceptible population; herd immunity shows how reducing the susceptible fraction of that population — via prior infection or vaccination — can push the effective reproduction number below 1, halting sustained spread even without every single individual being immune. It is the direct, quantitative link between an individual-level immunological fact (who is protected) and a population-level epidemiological outcome (whether an outbreak can sustain itself at all).

The threshold derived here is also what makes population-level vaccination strategy (vaccination-strategy) a coverage-target problem rather than an all-or-nothing one: public health goals are generally set at or above a calculated threshold coverage, not at 100%, and understanding why that threshold is sufficient (and what can undermine it) is essential to interpreting real-world vaccination policy correctly.

Hypotheses

Population mixing is roughly homogeneous: any infected individual is equally likely to contact any other individual.This is the specific assumption behind the simple threshold formula below; real populations have structured, clustered contact patterns (households, workplaces, social networks), and a population that meets the overall average threshold can still contain locally under-immunised clusters that sustain transmission internally despite the population-wide average being above threshold. Immunity (from infection or vaccination) substantially protects against onward transmission, not merely against the immune individual's own symptoms.If immunity blocked disease severity in the immune individual but not their ability to transmit the pathogen to others, the simple threshold calculation below would overstate the true population-level protection a given immune fraction actually provides.

Proof

R_{\text{eff}} = R_0(1-x)

In a population where a fraction \(x\) is immune and mixing is homogeneous (Hypotheses), only the susceptible fraction \((1-x)\) can actually become infected upon contact with an infectious case, so the effective reproduction number is scaled down from the fully-susceptible value \(R_0\) by exactly this factor. A

x_c = 1-\frac{1}{R_0}

An epidemic can only continue growing while \(R_{\text{eff}}>1\); setting \(R_{\text{eff}}=1\) in Step 1 and solving for \(x\) gives the herd immunity threshold \(x_c\), the minimum immune fraction needed to push \(R_{\text{eff}}\) below 1 and cause new-case counts to decline on average rather than grow. A

\text{Because } x_c \text{ depends only on } R_0\text{, more transmissible diseases (higher } R_0\text{) require a higher immune fraction to reach threshold.}

This is a direct, testable consequence of Step 2 relating a pathogen's inherent transmissibility to the vaccination coverage needed to control it: a highly transmissible disease demands correspondingly higher population coverage before the same halting effect is achieved. A

\text{Once } R_{\text{eff}}<1\text{, transmission chains tend to die out faster than they replace themselves, indirectly protecting even the remaining susceptible minority.}

The immune majority reduces the effective contact rate between an infectious case and any remaining susceptible individual, lowering that susceptible individual's own risk of exposure even though they are not themselves immune — the specific indirect protection effect that gives "herd immunity" its name, and which particularly matters for individuals who cannot be vaccinated for a medical reason. A

\text{Public-health vaccination coverage targets are generally set at or above } x_c\text{, with some margin.}

This directly underlies vaccination-strategy's population-level coverage goals: because real mixing is not perfectly homogeneous (t3 Hypothesis) and vaccine-induced immunity is not always 100% effective per dose, a margin above the simple calculated \(x_c\) is generally built into practical coverage targets rather than aiming for exactly the theoretical minimum. B

Result

x_c = 1-\frac{1}{R_0}

Reading. The minimum immune population fraction needed to halt sustained transmission rises with a pathogen's inherent transmissibility \(R_0\), and once reached, the immune majority indirectly protects the remaining susceptible minority by reducing their effective exposure risk.

Scope. A population-average threshold assuming homogeneous mixing and transmission-blocking immunity; real, structured contact patterns and imperfect (leaky) vaccine protection against transmission both require the practical coverage target to sit above the simple calculated \(x_c\) (Hypotheses, Step 5).

Corollaries & converses

sir-model's compartmental framework (Susceptible-Infected-Recovered) is exactly the underlying dynamical model from which the \(R_{\text{eff}}=R_0(1-x)\) relation is derived, with \(x\) here corresponding to the Recovered/immune compartment's fraction of the total population.
transmission-dynamics (contact rate, per-contact infectiousness) is what actually determines a given pathogen's \(R_0\) in the first place, and hence, via Step 2, its herd immunity threshold.
vaccination-strategy's population coverage targets are a direct practical application of \(x_c\), generally set with an added margin above the bare calculated threshold to account for imperfect vaccine effectiveness and non-homogeneous mixing.

Fails without

Immunity protects only against symptomatic disease, not onward transmission (Hypotheses): for an imperfect, "leaky" vaccine or infection-derived immunity that still permits some transmission from an immune individual, the simple \(x_c=1-1/R_0\) formula overestimates the true protective effect of a given immune fraction, since immune individuals could still contribute to sustaining transmission chains despite counting toward \(x\).
Population mixing is strongly non-homogeneous (t3 Hypothesis): a population could sit above the population-average threshold \(x_c\) overall while a locally under-immunised, tightly connected cluster remains well below threshold internally, allowing outbreaks to persist or recur within that cluster even though the overall population-wide average appears well protected.

Common errors

Assuming herd immunity requires 100% of the population to be immune, rather than the \(R_0\)-dependent threshold \(x_c\) (Step 2), which is generally well below 100% for most pathogens.
Treating the herd immunity threshold as a single, fixed, universal number rather than a value specific to each pathogen's own \(R_0\) (Step 3).
Assuming that reaching \(x_c\) guarantees zero further cases; it only guarantees, on average, that case counts decline rather than grow (Step 2) — stochastic transmission chains and cases imported from elsewhere can still cause localised outbreaks even above the calculated threshold.
Ignoring waning immunity, which can erode a previously reached immune fraction back below \(x_c\) over time, generally requiring booster vaccination or renewed exposure to maintain population-level protection.

Discussion

The herd immunity concept became broadly familiar through large-scale vaccination and disease-control campaigns over the 20th century, most notably underlying the global smallpox eradication effort, which the World Health Organization declared successful in 1980 — a case in which sufficiently high global immunity, achieved specifically through vaccination rather than natural infection, eliminated sustained human-to-human transmission of the disease entirely.

Because \(x_c\) depends only on \(R_0\) under the idealised homogeneous-mixing assumption, real-world control efforts for highly transmissible diseases (with correspondingly very high calculated \(x_c\)) can be especially sensitive to any gap between the theoretical threshold and lower, achievable real-world coverage, precisely because the margin for error shrinks as \(x_c\) itself approaches 1.

Common misconception: that herd immunity achieved primarily through widespread natural infection, rather than vaccination, is an equally acceptable or equivalent public-health strategy for reaching the same threshold. Both routes can, in principle, reach the same calculated \(x_c\), but relying on natural infection generally incurs the full disease burden (including severe disease and death) in the substantial fraction of the population that must be infected to reach that threshold, a very different cost from reaching the identical threshold primarily through vaccination.

Worked examples

R_0 \approx 15\ \text{(a highly transmissible disease, within the commonly cited range for measles)}

Applying the Result: \(x_c = 1-\dfrac{1}{15} \approx 0.933\), meaning roughly 93% of the population must be immune to reliably push \(R_{\text{eff}}\) below 1 — a very high bar, directly reflecting this pathogen's high inherent transmissibility (Step 3 of the Proof). A

R_0 \approx 3\ \text{(a moderately transmissible disease)}

By contrast, \(x_c = 1-\dfrac{1}{3} \approx 0.667\), roughly 67% — a substantially lower coverage target than the first case, purely because of the lower \(R_0\), illustrating directly how sensitive the required threshold is to a pathogen's transmissibility. A

R_0=15 \Rightarrow x_c\approx93\%; \qquad R_0=3 \Rightarrow x_c\approx67\%

Reading. A roughly five-fold difference in transmissibility between the two pathogens produces a substantial difference in the vaccination coverage required to achieve population-level control, directly demonstrating why highly transmissible diseases are correspondingly much harder to eliminate through vaccination alone.

Scope. The identical calculation, using the pathogen's own estimated \(R_0\), gives the theoretical minimum coverage target for any disease under the homogeneous-mixing idealisation.

Problems

A pathogen has \(R_0=4\). Compute the herd immunity threshold, and state, using Step 2 of the Proof, what is predicted to happen to case counts on average once this threshold is reached.
Solution
\(x_c = 1-1/4 = 0.75\), so 75% of the population must be immune. Once \(x\) reaches this threshold, \(R_{\text{eff}}=R_0(1-x)=4\times0.25=1\); above this threshold, \(R_{\text{eff}}<1\) and, on average, each existing case produces fewer than one new case, so case counts are predicted to decline over time rather than grow.
A country reaches 85% vaccination coverage against a pathogen with \(R_0=6\), assuming the vaccine is perfectly transmission-blocking. Determine whether this coverage is sufficient to reach herd immunity, and compute \(R_{\text{eff}}\) at this coverage level.
Solution
\(x_c=1-1/6\approx0.833\), so 85% coverage exceeds the calculated threshold. \(R_{\text{eff}}=R_0(1-x)=6\times(1-0.85)=6\times0.15=0.90\), below 1, consistent with case counts declining on average at this coverage level.
Explain why a population with 90% vaccination coverage against a pathogen with \(x_c=0.85\) could still experience a sustained local outbreak, using the t3 Hypothesis.
Solution
The 90% figure is a population-wide average; if mixing is not homogeneous (t3 Hypothesis) and vaccination coverage is unevenly distributed — for instance, concentrated in some communities and much lower in a specific, tightly connected under-vaccinated cluster — that local cluster's own effective immune fraction can sit well below the 85% threshold even while the population-wide average exceeds it, allowing sustained local transmission within the cluster despite the population as a whole appearing well above threshold on average.