Apps

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

Immortal Time Bias

Also Known As: Survival Bias in Cohort Studies Time-Dependent Confounding
Discourse Mechanics ID: immortal_time_bias

Definition

A bias in observational studies where a period of follow-up during which the outcome cannot occur (because the exposure has not yet happened) is misclassified as exposed person-time. This artificially inflates the exposed group's survival time and makes the exposure appear protective.

Examples

A study of Oscar winners' longevity counts the years before winning the Oscar as 'winner' person-time, during which the person was 'immortal' (they had to survive to win). This artificially increases the winners' apparent life expectancy.

A study claims that patients who complete a full 12-week cardiac rehabilitation program have 40% lower mortality than those who don't. But patients who died in the first 8 weeks were automatically classified as 'non-completers' — they couldn't have completed the program because they died. The completers were immortal during those early weeks by definition, inflating their apparent survival advantage.

Researchers find that employees who receive a promotion live longer on average than those who don't. However, the years worked before receiving the promotion — during which death would have prevented the promotion from ever occurring — are counted as 'promoted' person-time, making promoted employees appear healthier than they actually are relative to non-promoted peers.

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

    Is a cohort study comparing exposed and unexposed groups over time?

    Type: binary
  2. 2

    Is there a period during which exposed individuals could not have experienced the outcome by definition?

    Type: binary
  3. 3

    Is this 'immortal' time being counted in the exposed group's person-time, artificially lowering their event rate?

    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.

Hierarchical Context