Apps

🧪 This platform is in early beta. Features may change and you might encounter bugs. We appreciate your patience!

Incidence-Prevalence Bias

Also Known As: Neyman bias Prevalence-incidence bias Selective survival bias
Aspect ID: incidence_prevalence_bias

Definition

Incidence-prevalence bias (also called Neyman bias) occurs when studying existing cases of a disease gives a distorted picture of risk factors because rapidly fatal or quickly remitting cases have already disappeared from the pool of observable cases. The sample of surviving or persistent cases is not representative of all cases that occurred.

Examples

A study examining risk factors for a severe infection recruits hospitalized patients. However, the most severely ill patients died before reaching the hospital, and mildly ill patients recovered at home without hospitalization. The hospital sample represents only intermediate-severity cases, distorting risk factor estimates.

Researchers studying risk factors for long COVID recruit participants from a post-COVID clinic. Because patients who recovered quickly never sought follow-up care, the sample overrepresents people with prolonged illness, making certain demographic and clinical factors appear far more strongly associated with long COVID than they actually are in the full population of COVID-19 survivors.

A study on risk factors for stroke recruits patients currently hospitalized with stroke. Because strokes with certain risk factor profiles (e.g., very severe hypertension) are rapidly fatal before hospitalization, and minor strokes are managed outpatient, the hospitalized sample misrepresents the true distribution of risk factors across all stroke events in the population.

Verification Steps
Verification Steps
Binary yes/no questions that an AI must answer to detect a reasoning pattern in a text.
Each of the 452 aspects has verification steps — simple yes/no questions designed to systematically detect whether a pattern appears in a text. For ad hominem: "Does the argument attack a person rather than their claim?" For false dichotomy: "Are only two options presented when more exist?" This ensures consistent, reproducible analysis.

Binary (yes/no) questions an LLM must answer to identify this aspect:

  1. 1

    Does the study sample existing cases (prevalent cases) rather than newly diagnosed cases (incident cases)?

    Type: binary
  2. 2

    Could rapidly fatal or quickly resolved cases have been missed because they were no longer present at study entry?

    Type: binary
  3. 3

    Are risk factor estimates from prevalent cases used to draw conclusions about disease etiology?

    Type: binary
  4. 4

    Could the apparent risk factors reflect duration of disease rather than its onset?

    Type: binary
Deep Dive
The expandable detail section on each aspect page with examples, psychology, and counter-strategies.
The Deep Dive section provides in-depth information about each aspect: a real-world example showing the pattern in action, an explanation of why it works psychologically, practical advice on how to counter it, alternative names, and links to related aspects.