Critical Community Size¶
Study-defined host-population size or range associated with a specified local infection fade-out versus persistence criterion over a stated horizon.
Core Idea¶
Critical community size (CCS) is a pathogen- and setting-specific host-population size or range associated with a declared criterion for local infection fade-out versus persistence over a stated period. It is an operational relation between community size and stochastic transmission outcomes, not a universal count at which infection becomes impossible below and guaranteed above. Disease biology, immunity, demography, vaccination, reporting, and importation assumptions shape an estimate.[1][2]
Bartlett's historical U.S. measles analysis used the size at which post-epidemic fade-out was as likely as not until reintroduction, estimating roughly 250,000–300,000 people under his convention. Wearing and Rohani later analyzed pertussis in England and Wales with different surveillance and model conditions, reporting separate pre-vaccine and vaccine-era ranges. Their numbers illustrate context dependence rather than a disagreement over one fixed biological constant.[1][2]
Structural Signature¶
- Specified infection and host community. Name the pathogen and population unit whose local circulation is studied. A count with neither is not an epidemiological CCS.[1][2]
- Transmission and susceptible-replenishment setting. State the relevant immunity, contact, demographic, and vaccination assumptions that shape successive epidemic troughs. No single SEIR scheme or birth mechanism is required in every study.[2]
- Local fade-out measure. Define what counts as loss of local circulation, whether through a stochastic model or a surveillance proxy such as zero reported cases. A reported zero is not automatically proof of no infections.[1][2]
- Size boundary, risk criterion, and horizon. Relate host population size to the declared fade-out or persistence outcome over time, while distinguishing outside reintroduction from uninterrupted local persistence. A range or even-odds point may be more defensible than a sharp natural cutoff.[1][2]
Without an explicit fade-out outcome and time or probability convention, a population number cannot be interpreted as the study's CCS. Without the host-size relation, the result is simply an epidemic description.[1]
What It Is Not¶
Not a disease-independent population constant. Measles and pertussis analyses use different pathogens and conditions. Not an all-or-none guarantee. Fade-out risk can vary with size without a deterministic cliff. Not the disappearance of a pathogen everywhere. Local chains can end and later be reintroduced. Not a direct readout of every infection. Zero notifications may reflect surveillance as well as transmission.[1][2]
Minimum viable population asks whether a host species can persist; CCS asks about a pathogen's local circulation in a host community. A bare policy cutoff with no disease, outcome, or horizon is a nearby population threshold, but it lacks this epidemiological relation.[1][2]
Scope of Application¶
The term arose in the analysis of recurrent measles epidemics across U.S. urban populations. Bartlett's summary names an as-likely-as-not post-epidemic fade-out criterion, a mean fade-out time of about two years under his convention, and the corresponding historical population estimate. It does not establish a contemporary universal measles number or, by itself, a specific birth-rate mechanism for that estimate.[1]
Wearing and Rohani use England–Wales pertussis data and models to compare fade-out patterns in pre-vaccine and vaccine eras. They also model a background force of infection for outside introductions. The two settings remain in human infection epidemiology; neither licenses importing the seed's wildlife example without separate evidence and admission review.[2]
Clarity¶
Before comparing two CCS claims, ask: which disease, which communities, which fade-out measure, and over what period? Bartlett's “as likely as not” criterion is an explicit probability convention. A surveillance study may use weeks with zero reported cases, which is an observation rule rather than direct access to every infection. Different definitions can yield different numerical boundaries without one being a mathematical error.[1][2]
Also distinguish a locally extinct chain from one sustained by outside arrivals. The same town may report another case later; that fact alone does not show its earlier transmission chain persisted continuously.[1][2]
Manages Complexity¶
Persistence depends on host count, contacts, immunity, vaccination, demographic renewal, stochastic low-case periods, imports, and observation. CCS organizes those variables around one bounded question: at what population size does the declared fade-out outcome have the specified frequency under this setting? The result can summarize a model or data pattern, but its conditions must accompany the number.[1][2]
The summary sacrifices detail if misused. A single threshold value cannot show every epidemic trajectory or separate under-reporting from true transmission loss. Wearing and Rohani's plots of fade-out frequency against size retain the graded relation that a single label can hide.[2]
Abstract Reasoning¶
Define the pathogen, population unit, observation window, local fade-out rule, and importation convention. Estimate or model how fade-out frequency changes with host population size. Then identify the size or range that meets the declared criterion, reporting uncertainty and surveillance limits. If vaccination or immunity assumptions change, reassess the relation rather than carrying the old count forward.[1][2]
This procedure permits a bounded inference about local persistence risk. It does not infer certain extinction below the estimate, certain endurance above it, or a fixed number for another disease.[2]
Knowledge Transfer¶
The size-versus-fade-out question transfers literally from historical measles analysis to pertussis analysis: both relate human community size to loss of local infection under a stated definition. The numerical answer does not transfer, because pathogen, era, immunity, surveillance, and intervention conditions differ. Within the pertussis study itself, pre-vaccine and vaccine-era estimates change under different population-level conditions.[1][2]
The broader intuition that small populations can lose a stochastic process is only an analogy outside the infection setting. This named CCS entry is bound to pathogen persistence and should not be presented as a general population-size Prime.[1][2]
Examples¶
Historical U.S. measles city analysis¶
Bartlett's 1960 summary defines the critical size as the community population at which measles is as likely as not to fade out after a major epidemic until it is reintroduced. He reports roughly 250,000–300,000 persons and a mean time to fade-out of about two years under that analysis. The source is a historical statistical estimate, not a guarantee for each city.[1]
Mapped back: measles in the studied U.S. urban populations provides the specified infection and host community; recurrent epidemic conditions supply the transmission setting, though the accessible summary does not prove one exact replenishment mechanism; as-likely-as-not post-epidemic loss supplies the local fade-out measure; and the stated size, probability convention, and roughly two-year time provide the operational boundary and horizon. Later reintroduction is separated from persistence.[1]
Pertussis across vaccination eras¶
Wearing and Rohani compare pertussis model predictions with England–Wales notifications. Their critical-community-size analysis relates population size to the frequency of weeks with zero reported cases and reports about 150,000–250,000 people before vaccination versus about 800,000–1,000,000 in the vaccination era. They examine an alternative three-consecutive-week definition and include background infection pressure in the model.[2]
Mapped back: pertussis in England–Wales communities is the specified infection and host unit; their modeled immunity, vaccination, and reinfection assumptions give the transmission and replenishment setting; weeks without reported cases provide a local fade-out proxy; and the era-specific ranges under the study's criteria provide the operational size boundary. Notification zeros and imported infections limit any interpretation as an absolute extinction line.[2]
Structural Tensions¶
The two sources establish uncertainty and convention dependence, but do not prove a universal conflict between two objectives that every CCS must balance. A strict zero-case rule and a longer zero-case rule may classify different observations; that is a measurement choice whose consequences must be reported, not an intrinsic biological optimization tradeoff. Likewise, reducing introductions changes the persistence question rather than creating a universal pair of opposed values.[2]
The diagnostic question is whether a numerical claim is being used to predict local transmission under the same fade-out, reporting, and importation convention that generated it. If the convention changes, the estimate must be recalculated or clearly qualified.[1][2]
Structural–Framed Character¶
Evaluative weight: CCS describes a size–fade-out relation; whether persistence or interruption is desirable is a separate public-health judgment. Human-practice dependence: analysts choose a community unit, surveillance definition, model, risk criterion, and time horizon, while infection events constrain the estimate. Institutional origin: epidemic records and reporting systems shape the data, but a particular agency does not create the general relation. Vocabulary travel: “critical size” elsewhere may mean survival or market viability, not pathogen fade-out. Import versus recognition: verify the infection, local outcome, and operational criterion before applying the term.[1][2]
The entry is framed-leaning on the structural–framed spectrum: its stochastic size–outcome relation is analyzable, but the named measurement depends strongly on disease, surveillance, and a declared criterion. No live Prime has been proved a strict direct parent across the admitted cases. Its character: an operational epidemiological measure that keeps its study conditions attached to any reported population range.[1][2]
Structural Core vs. Domain Accent¶
The core is a host-population-size relation to a specified probability or frequency of local pathogen fade-out over a horizon. Measles versus pertussis, U.S. cities versus England–Wales communities, the two reported ranges, and a particular zero-week observation rule are case accents. Remove the infection and local fade-out relation and a remaining “critical” population count is no longer CCS.[1][2]
The named entry does not clear the Prime bar merely because threshold language appears in other fields. The disease and outcome are identity-bearing, and the live Threshold Prime's distinct-transition wording does not strictly cover every operational gradual risk relation here. A broader stochastic population-persistence abstraction is a future-prime question requiring independent cross-domain evidence; it is not asserted as a present parent.[1][2]
Instantiates / Related Primes¶
The approved placement is unparented in the current catalog. Threshold is a nearby word and reasoning aid, but its live distinct or rapid transition identity is stronger than the graded, criterion-dependent fade-out relation shown here. Critical Mass concerns chain growth under a reproduction condition, not a guaranteed CCS genus. Basic Reproduction Number is a transmission ratio, and Minimum Viable Population concerns host species survival. These links are thematic or component relations, not proven strict direct parents for both source cases.[1][2]
An unparented placement does not deny that one can choose an operational cutoff on a smooth risk curve. It says that doing so alone does not make the entire epidemiological measure a strict instance of the current Threshold Prime.[2]
Neighborhood in Abstraction Space¶
Critical Community Size sits in a sparse region of the domain-specific corpus (99th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Survival Analysis & Demographic Rates (16 abstractions)
Nearest neighbors
- Epidemic models on lattices — 0.76
- Infection rate — 0.76
- Clinical Case Definition — 0.75
- Attack rate — 0.73
- Uncertain geographic context problem — 0.73
Computed from structural-signature embeddings · 2026-10-08
Not to Be Confused With¶
A universal herd-immunity number: CCS refers to host population size under local stochastic persistence conditions. The basic reproduction number: a transmission ratio, not population count. Minimum viable host population: survival of the host species, not of a pathogen chain. Zero notifications: a surveillance proxy that may miss cases. Reintroduction after fade-out: a new outside input, not proof of unbroken local persistence. An exact natural cliff: published CCS values use declared study criteria and can be ranges.[1][2]
References¶
[1] M. S. Bartlett, “The Critical Community Size for Measles in the United States,” Journal of the Royal Statistical Society: Series A 123, no. 1 (1960), pp. 37–44, original publisher Summary. https://academic.oup.com/jrsssa/article/123/1/37/7101793 registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x
[2] Helen J. Wearing and Pejman Rohani, “Estimating the Duration of Pertussis Immunity Using Epidemiological Signatures,” PLoS Pathogens 5, no. 10 (2009), e1000647, Results “Critical community size,” Methods, and Figures 3–4. https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1000647 registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x ↩y ↩z ↩27 ↩28