Concept

Vaccination strategy

T-119Home BU-403Threads systems · evolution

Statement

Controlling disease at the population level.

Why it matters

herd-immunity establishes that a population can be protected from sustained transmission without immunising every single individual; vaccination-strategy is the practical discipline of turning that threshold condition into an actual, deliverable programme — deciding what fraction of a population to vaccinate, in what order, and with what pattern, given that vaccine doses, delivery capacity, and time are all limited resources in any real public-health campaign.

It is also where sir-model and transmission-dynamics's more abstract results become directly actionable: the herd-immunity threshold sets the minimum target, transmission-dynamics's heterogeneity results explain why simply hitting that population-average number is not always sufficient or efficient, and vaccination-strategy is the layer that reconciles the two into an actual targeting decision.

Hypotheses

An effective vaccine moves an individual from the susceptible compartment directly to an immune (removed) state, without that individual passing through infection.This is what makes vaccination mechanistically equivalent, in the SIR framework (sir-model), to natural recovery for the purpose of the threshold calculation, but without the cost (illness, risk of severe outcome or death) that acquiring immunity through actual infection would carry — the entire practical rationale for preferring vaccination over "natural" herd immunity via uncontrolled spread. Vaccine efficacy is generally imperfect and can wane over time, so the effective immunised fraction is less than the fraction of the population that has received a dose.Without accounting for less-than-100% efficacy, a vaccination campaign that appears to hit the nominal herd-immunity threshold in raw doses administered may still leave the population's true effective immune fraction below the threshold actually required, since not every vaccinated individual becomes fully immune (Fails without). Contact-rate heterogeneity across individuals (transmission-dynamics) means that reaching a given population-average immune fraction is not equally effective regardless of which individuals are immunised; targeting high-contact individuals reduces effective transmission more per dose than immunising a random sample.This is the basis for prioritisation strategies in real vaccination campaigns (targeting healthcare workers, other high-contact occupational groups, or otherwise highly connected individuals) as distinct from, and often more efficient per dose than, untargeted mass vaccination of a random population sample.

Proof

\[ \text{Immunising a fraction } v \text{ of the population moves that fraction from } S \text{ directly to } R\text{, so the effective susceptible fraction becomes } (1-v)\,\frac{S_0}{N}. \]

This follows directly from treating vaccination as an instantaneous transfer within the SIR compartments (Hypotheses); applied at the start of an otherwise fully susceptible population (\(S_0/N\approx1\)), the post-vaccination susceptible fraction is simply \(1-v\). A

\[ \text{The epidemic fails to grow once } 1-v < \frac1{R_0}, \text{ i.e. once } v > 1-\frac1{R_0}\ \equiv v_c\text{, the critical vaccination threshold.} \]

Substituting Step 1's post-vaccination susceptible fraction directly into sir-model's growth condition (\(S/N>1/R_0\)) and solving for the vaccinated fraction \(v\) that just prevents growth reproduces exactly herd-immunity's threshold formula, here framed specifically as a target for deliberate vaccination coverage rather than as a description of accumulated natural immunity. A

\[ \text{With vaccine efficacy } e<1\text{, the effective immunised fraction is } ve\text{, so the coverage target must be raised to } v = \frac{v_c}{e} = \frac{1-\tfrac1{R_0}}{e}. \]

Since only a fraction \(e\) of vaccinated individuals actually become immune, achieving the same effective immune fraction \(v_c\) as a perfect vaccine requires vaccinating a correspondingly larger raw fraction \(v=v_c/e\) of the population; for any \(e\lt1\), this required coverage is necessarily higher than the idealised threshold of Step 2. B

\[ \text{Targeting vaccination toward higher-contact individuals reduces the effective reproduction number more, per dose delivered, than vaccinating an equally sized random sample of the population.} \]

Because high-contact individuals contribute disproportionately to transmission (transmission-dynamics' Step 3), removing them from the susceptible pool disproportionately reduces onward transmission opportunities relative to their share of the total population; the same total number of doses can therefore achieve a larger reduction in effective spread when preferentially allocated to high-contact individuals or groups rather than distributed uniformly at random. B

\[ \text{Ring vaccination (vaccinating known contacts of a confirmed case, and their contacts in turn) can achieve local containment with far lower total population coverage than a population-wide threshold strategy, provided cases can be identified and contacts traced quickly relative to the disease's infectious period.} \]

Rather than attempting to push the population-wide susceptible fraction below \(1/R_0\) everywhere (Step 2), ring vaccination instead creates a local "firebreak" of immunised individuals specifically around each detected case, interrupting that particular chain of transmission before it can spread further; this is most effective for diseases with a sufficiently long incubation period relative to the time needed to trace and vaccinate contacts. B

Result

\[ v_c = 1-\frac1{R_0} \qquad v_{\text{required}} = \frac{v_c}{e}\ (\text{imperfect efficacy}) \qquad \text{targeted coverage} < \text{uniform coverage for the same effect} \]

Reading. The population-average vaccination coverage needed to halt transmission is set directly by \(R_0\) and vaccine efficacy, but real strategies can achieve the same or better epidemiological effect with fewer total doses by exploiting heterogeneity in contact rate (targeted prioritisation) or by working locally around detected cases (ring vaccination) rather than aiming for uniform, population-wide coverage alone.

Scope. Steps 1–3 apply straightforwardly to a well-mixed population (sir-model's assumptions); Steps 4–5's efficiency gains depend on real, exploitable contact-rate heterogeneity and, for ring vaccination specifically, on adequately fast case detection and contact tracing relative to the pathogen's transmission timescale.

Corollaries & converses

herd-immunity's threshold condition and Step 2 here are, mathematically, the identical formula \(1-1/R_0\), applied to two different sources of immunity (natural infection versus deliberate vaccination); vaccination-strategy is specifically concerned with reaching that same threshold deliberately and with fewer total infections along the way.
sir-model's guarantee that every closed-population epidemic eventually terminates on its own (via susceptible depletion) is precisely the "natural herd immunity" alternative that a deliberate vaccination strategy is designed to avoid relying on, since reaching the threshold through uncontrolled infection accepts the full case-fatality and morbidity burden of getting there.
Converse: if an outbreak continues to grow despite reported vaccination coverage nominally exceeding the idealised threshold \(v_c\), Step 3 identifies imperfect or waning vaccine efficacy as one specific, checkable explanation, distinct from insufficient raw coverage.

Fails without

Drop the account for imperfect vaccine efficacy (Hypotheses, Step 3): a campaign that vaccinates exactly the nominal threshold fraction \(v_c\) of the population, assuming perfect efficacy, will leave the true effective immune fraction below \(v_c\) whenever \(e\lt1\); the population remains vulnerable to sustained transmission despite appearing, on paper, to have reached the required coverage.
Drop contact-rate heterogeneity (Hypotheses, Step 4): a strategy that assumes every dose has equal epidemiological value, regardless of who receives it, will under-prioritise high-contact individuals and over-invest doses in low-contact individuals relative to the efficient allocation Step 4 identifies, achieving a smaller reduction in effective transmission for the same total number of doses delivered.

Common errors

Assuming the herd-immunity threshold \(v_c=1-1/R_0\) is the actual vaccination coverage target to aim for in a real campaign, without correcting for vaccine efficacy (Step 3); the true required coverage is generally higher, sometimes substantially so for vaccines with lower efficacy.
Assuming uniform, population-wide coverage is always the most efficient strategy; Step 4 and Step 5 both show that exploiting known structure (contact-rate heterogeneity, or the local structure around a detected case) can achieve comparable or better containment with fewer total doses than an untargeted approach.
Treating \(R_0\) as fixed and unaffected by prior public-health measures when calculating a vaccination target; \(R_0\) (or, more precisely, the effective reproduction number at the time of the campaign) can itself be reduced by concurrent non-pharmaceutical measures (transmission-dynamics), lowering the vaccination coverage actually required to reach the threshold.
Assuming ring vaccination is a universally applicable strategy; Step 5 specifies it depends on being able to detect cases and trace and vaccinate contacts quickly relative to the pathogen's infectious period, a condition well satisfied for some diseases (notably smallpox, historically) but not for others with a very short incubation period or substantial pre-symptomatic transmission.

Discussion

The global eradication of smallpox, certified complete in 1980 following a World Health Organization campaign, remains the most significant achievement directly attributable to a vaccination strategy explicitly using both population-level threshold reasoning and targeted ring vaccination around detected cases (Step 5), a combination credited with achieving eradication with substantially less than universal population coverage.

Pathogen evolution under sustained vaccination pressure is a genuine, actively studied concern: selection can favour viral or bacterial variants that partially escape vaccine-induced immunity (reducing effective \(e\) in Step 3 over time for existing vaccine formulations) or that alter transmission characteristics in ways that shift the effective \(R_0\) the campaign must account for; this is one reason vaccination strategy is treated as an ongoing, adaptive process rather than a single, one-time coverage target to be reached and then left unmonitored.

Common misconception: that a vaccination campaign has "failed" if breakthrough infections still occur among vaccinated individuals. Since vaccine efficacy is generally imperfect (Hypotheses), some breakthrough infections are expected even in a successful campaign; the relevant metric for programme success is whether population-level transmission has been reduced below the epidemic threshold (Step 2), not whether any single vaccinated individual can still become infected.

Worked examples

\[ \text{A disease has } R_0=6. \text{ A vaccine has efficacy } e=0.9. \]

Idealised threshold (Step 2): \(v_c=1-1/6\approx0.833\). Correcting for imperfect efficacy (Step 3): \(v_{\text{required}}=v_c/e=0.833/0.9\approx0.926\), meaning roughly \(93\%\) of the population must actually receive the vaccine to achieve the same effective immune coverage a perfect vaccine would provide at \(83\%\) coverage. A

\[ \text{A second population instead prioritises its highest-contact 20}\%\text{ of individuals for vaccination first, achieving disproportionate transmission reduction per Step 4.} \]

Because these individuals contribute disproportionately to onward transmission (transmission-dynamics), immunising them first can reduce the effective reproduction number by more than a proportional \(20\%\) reduction in transmission opportunities would suggest from their population share alone, illustrating why real campaigns often prioritise occupational or social groups with elevated contact rates ahead of a strictly random rollout order. B

\[ R_0=6,\ e=0.9 \ \Rightarrow\ v_{\text{required}}\approx93\% \qquad \text{targeting high-contact 20\% first accelerates threshold-crossing} \]

Reading. Reaching an adequate effective immune fraction for a highly transmissible disease with an imperfect vaccine can require vaccinating nearly the entire population; prioritising high-contact individuals early in a rollout reaches the effective threshold faster than a purely random distribution of the same doses over time.

Scope. The same efficacy-correction and targeting logic applies to any \(R_0\) and any vaccine efficacy, with the required raw coverage rising sharply as either \(R_0\) increases or \(e\) decreases.

Further example. A novel respiratory pathogen has a basic reproduction number \(R_0 = 4.0\). A vaccine with an efficacy of \(e = 0.80\) is available. The population is heterogeneous: 10% of individuals are "high-contact" (e.g., healthcare workers, teachers) and account for 40% of all transmission events, while the remaining 90% are "low-contact" and account for the remaining 60% of transmission. Calculate the minimum fraction of the total population that must be vaccinated if the vaccine is distributed randomly. Then, calculate the minimum fraction of the total population that must be vaccinated if the vaccine is distributed exclusively to the high-contact group first, until that group is fully covered, and then to the low-contact group as needed to reach the herd immunity threshold.

\[ v_c = 1 - \frac{1}{R_0} = 1 - \frac{1}{4.0} = 0.75 \]

The herd immunity threshold \(v_c\) represents the fraction of the population that must be immune to reduce the effective reproduction number \(R_t\) to 1. This is derived from the condition \(R_t = R_0 \times S = 1\), where \(S\) is the susceptible fraction. A

\[ v_{\text{random}} = \frac{v_c}{e} = \frac{0.75}{0.80} = 0.9375 \]

For random distribution, the required raw vaccination coverage is the ideal threshold divided by vaccine efficacy. This assumes the immune fraction is uniformly distributed across all contact groups. A

\[ S_{\text{high}} = \frac{1}{R_0} \times \frac{\text{Transmission share}_{\text{high}}}{\text{Population share}_{\text{high}}} \times \frac{1}{\text{Relative Contact Rate}} \]

This step is complex; a simpler approach for heterogeneous mixing is to calculate the effective reproduction number contribution. Let \(R_{0,\text{high}}\) be the reproduction number contribution of the high-contact group. Since they account for 40% of transmission, their effective \(R_0\) contribution is \(0.4 \times 4.0 = 1.6\). The low-contact group contributes \(0.6 \times 4.0 = 2.4\). To stop transmission, the weighted sum of \(R_0 \times S\) must be \(\le 1\). B

\[ R_t = R_{0,\text{high}} S_{\text{high}} + R_{0,\text{low}} S_{\text{low}} \le 1 \]

We prioritize vaccinating the high-contact group first. Let \(v_{\text{high}}\) be the fraction of the high-contact group vaccinated. The susceptible fraction in the high group is \(S_{\text{high}} = 1 - v_{\text{high}} \times e\). The low-contact group is unvaccinated initially, so \(S_{\text{low}} = 1\). We solve for \(v_{\text{high}}\) such that \(1.6(1 - 0.8 v_{\text{high}}) + 2.4(1) = 1\). B

\[ 1.6 - 1.28 v_{\text{high}} + 2.4 = 1 \implies 4.0 - 1.28 v_{\text{high}} = 1 \implies 1.28 v_{\text{high}} = 3.0 \implies v_{\text{high}} = \frac{3.0}{1.28} \approx 2.34 \]

The calculated \(v_{\text{high}}\) is greater than 1, which is impossible. This indicates that vaccinating only the high-contact group, even at 100% coverage, is insufficient to reach herd immunity because their contribution to \(R_0\) (1.6) is less than the threshold required to bring the total \(R_t\) down to 1 when the low-contact group remains fully susceptible (contributing 2.4). We must vaccinate the high-contact group fully and then vaccinate the low-contact group. B

\[ R_t = 1.6(1 - 1.0 \times 0.8) + 2.4(1 - v_{\text{low}} \times 0.8) = 1 \]

With the high-contact group fully vaccinated, their susceptible fraction is \(S_{\text{high}} = 1 - 0.8 = 0.2\). Their contribution to \(R_t\) is \(1.6 \times 0.2 = 0.32\). The remaining required reduction must come from the low-contact group: \(2.4 S_{\text{low}} = 1 - 0.32 = 0.68\). Thus, \(S_{\text{low}} = \frac{0.68}{2.4} \approx 0.2833\). The required vaccination fraction for the low group is \(v_{\text{low}} = \frac{1 - S_{\text{low}}}{e} = \frac{1 - 0.2833}{0.8} = \frac{0.7167}{0.8} \approx 0.8959\). The total population fraction vaccinated is \(v_{\text{total}} = (0.10 \times 1.0) + (0.90 \times 0.8959) = 0.10 + 0.8063 = 0.9063\). B

\[ v_{\text{random}} = 93.75\% \qquad v_{\text{targeted}} \approx 90.6\% \]

Reading. Targeting the high-contact group first reduces the required total population coverage from 93.75% to approximately 90.6%. While the difference is modest in this specific parameterization, it demonstrates that prioritizing groups with higher transmission contributions is more efficient per dose than random distribution. The targeted strategy achieves the same effective immune coverage with fewer total doses administered to the general population.

Scope. This model assumes homogeneous mixing within groups and that the transmission shares (40% vs 60%) are constant. In reality, contact rates may change with interventions, and vaccine efficacy may vary by group. The benefit of targeting is greater when the high-contact group accounts for a larger share of transmission relative to their population size.

Problems

A disease has \(R_0=3\) and a vaccine with efficacy \(e=0.8\). Compute the raw vaccination coverage required to reach the effective herd-immunity threshold.
Solution
Idealised threshold: \(v_c=1-1/3\approx0.667\). Required raw coverage: \(v=v_c/e=0.667/0.8\approx0.833\), so approximately \(83\%\) of the population must be vaccinated.
Explain, using Step 4, why vaccinating healthcare workers early in a rollout is expected to reduce hospital-based disease transmission more than vaccinating an equally sized random sample of the general population, even before overall population coverage reaches the herd-immunity threshold.
Solution
Healthcare workers typically have an elevated contact rate with both infected and susceptible individuals compared to the general population average; by Step 4, immunising individuals with above-average contact rates removes a disproportionately large share of potential transmission events relative to their numeric share of the population. Vaccinating this group early therefore reduces transmission specifically along the high-risk healthcare-contact pathway more efficiently, per dose, than an equivalent number of doses spread randomly across the general population.
A public-health team has a limited initial vaccine supply and must choose between (a) achieving the population-wide herd-immunity threshold as fast as possible with random distribution, or (b) ring vaccination focused on contacts of detected cases. Under what condition, referencing Step 5, would option (b) be expected to work well even with a much smaller total number of doses used?
Solution
Ring vaccination (Step 5) works well specifically when cases can be detected and their contacts traced and vaccinated quickly relative to the pathogen's infectious period (i.e. before those contacts themselves become infectious and spread the disease further). Under that condition, each detected case's local transmission chain can be interrupted directly, achieving containment without needing to reach the full population-wide threshold \(v_c\) at all; if detection or contact-tracing is too slow relative to the disease's transmission timescale, ring vaccination fails to keep pace with spread, and the population-wide threshold strategy becomes necessary instead.
A pathogen has \(R_0 = 2.5\). A vaccine with efficacy \(e = 0.90\) is available. In a population with no prior immunity, what is the minimum percentage of the total population that must be vaccinated to achieve herd immunity?
Solution
First, calculate the herd immunity threshold \(v_c\) for a perfect vaccine: \[ v_c = 1 - \frac{1}{R_0} = 1 - \frac{1}{2.5} = 1 - 0.4 = 0.6 \] This means 60% of the population needs to be immune. Next, adjust for vaccine efficacy \(e = 0.90\): \[ v_{\text{required}} = \frac{v_c}{e} = \frac{0.6}{0.90} = 0.6667 \] Converting to a percentage: \[ 0.6667 \times 100\% \approx 66.7\% \] Therefore, approximately 66.7% of the population must be vaccinated.
Consider a population where 20% of individuals are already immune to a disease with \(R_0 = 5.0\). A vaccine with efficacy \(e = 0.75\) is introduced. Calculate the fraction of the total population that must be vaccinated to reach the herd immunity threshold.
Solution
First, determine the required effective susceptible fraction to stop transmission: \[ S_{\text{max}} = \frac{1}{R_0} = \frac{1}{5.0} = 0.2 \] This means the total susceptible fraction (unvaccinated + non-responders) must be less than or equal to 0.2. The initial susceptible fraction (excluding pre-immune) is: \[ S_0 = 1 - 0.20 = 0.80 \] The fraction of the total population that must be protected by the vaccine is the difference between the initial susceptible pool and the maximum allowable susceptible fraction: \[ S_{\text{protected}} = S_0 - S_{\text{max}} = 0.80 - 0.20 = 0.60 \] Now, calculate the required vaccination coverage \(v\) given the efficacy \(e = 0.75\): \[ v = \frac{S_{\text{protected}}}{e} = \frac{0.60}{0.75} = 0.80 \] Thus, 80% of the total population must be vaccinated.