Why standard regression can fail to capture the true speed of an epidemic

Exponential growth and epidemics · 8:57 · Video: 3Blue1Brown · Watch at the source

In short: Estimating how quickly a disease spreads is vital for public health, yet common statistical methods often rely on flawed assumptions. This video explores the mechanics of exponential and logistic growth and the mathematical nuances of tracking outbreaks.

In the early stages of an outbreak, infections typically follow an exponential growth pattern. While simple linear regression on the logarithms of case counts is a common approach, it can be misleading. For instance, looking at the R² of linear regressions on cumulative data can be problematic because daily changes are not independent, even if they appear random.

Researchers are now turning to more robust frameworks, such as Hidden Markov Models (HMM), to better account for unobserved infectious populations. Unlike standard negative binomial regression, which may underestimate uncertainty, HMMs can provide more reliable estimates. This approach can even be extended into a logistic model to account for how growth slows as the susceptible population depletes.

Opens the Intellect+ feed with this card first. Swipe up for the next one.

Sources

To cite this card, link to https://intellectplus.co.uk/app/science-exponential-growth-and-epidemics.

How we source, check and correct cards: Editorial standards.

General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.

Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?

Related

Finding something worth knowing…