What herd immunity is, and why each disease needs a different threshold

Herd immunity is one of the most used and most misunderstood ideas in public health. The core is a piece of arithmetic, not a slogan. When enough people in a population are immune to a contagious disease — through vaccination or past infection — each infected person, on average, passes it to fewer than one other, and an outbreak shrinks rather than grows. That indirect protection extends to people who are not themselves immune, including newborns too young to be vaccinated and people whose immune systems cannot respond to vaccines [s1]. Reaching that state does not require everyone to be immune, which is the whole point, but the fraction it does require is not a single universal number.

The threshold comes straight from contagiousness

How much of a population must be immune depends on how contagious the disease is, captured by the basic reproduction number, R0 — the average number of people one infected person would infect in a fully susceptible population. A review in Clinical Infectious Diseases lays out the standard relationship: the herd immunity threshold is approximately 1 − 1/R0 [s1]. Feed a disease's R0 into that formula and it returns the share of the population that needs to be immune to halt sustained spread. A disease with an R0 of 2 needs roughly half the population immune; one with an R0 of 5 needs about 80%; one with an R0 of 10 needs about 90% [s1]. The more contagious the disease, the higher and less forgiving the bar.

That is why the required coverage differs so much between diseases. It also explains why the most contagious diseases are the hardest to hold in check: they leave the least room for gaps. The review is careful to note that this simple formula assumes a well-mixed population where everyone is equally likely to contact everyone else, which no real population is — clustering of susceptible people, such as under-vaccinated communities, can let an outbreak run even when average coverage looks high enough [s1].

Measles is the demanding case

Measles is the disease that sets the highest bar, because it is among the most contagious infections known. The World Health Organization treats roughly 95% two-dose vaccination coverage as the target needed to prevent outbreaks and sustain elimination [s3]. That figure follows directly from measles' very high R0: at the values commonly quoted, the 1 − 1/R0 formula lands in the low-to-mid 90s.

But the popular R0 numbers deserve a caution the formula hides. A 2017 systematic review in The Lancet Infectious Diseases examined the widely repeated claim that measles has an R0 of 12 to 18 and found that this range rests on a small number of decades-old studies, while the published estimates in the literature vary far more widely than the tidy figure suggests [s2]. The authors' point is not that measles is less contagious than believed — it is extremely contagious — but that R0 is not a fixed constant of a virus [s2]. It depends on contact patterns, population density and social structure, so the "right" threshold for a disease is genuinely a range that shifts with place and time, not a decimal to be memorised.

Why the concept is easy to misuse

Two misreadings recur. The first is treating herd immunity as a strategy to reach by letting a disease spread through a population. The arithmetic does not care how immunity is acquired, but the cost does: reaching a measles-level threshold through natural infection means most of the population being infected by a disease that can cause pneumonia, brain inflammation and death, whereas vaccination reaches the same threshold without the illness [s1][s3]. The second misreading is treating the threshold as a finish line after which vaccination can relax. Because R0 varies and immunity can wane, coverage that dips below the threshold allows transmission to resume — which is exactly what has driven measles resurgences where vaccination rates have fallen [s3].

The useful way to hold the idea is this: herd immunity is real and protective, it is governed by contagiousness through a simple formula, and the number it produces is a moving target that has to be maintained rather than a one-time achievement [s1]. That the most contagious diseases demand coverage in the mid-90s is not an arbitrary public-health preference; it is what the arithmetic of their spread requires.