Overview

Hill's criteria


Show Summary Details

Quick Reference

The British medical statistician Austin Bradford Hill (1897–1991) identified eight criteria that can be used to help distinguish statistically significant associations that indicate causation, i.e., an association in which a cause produces an effect. The criteria are as follows:1. Strength of the association, i.e., the size of the risk, as measured by appropriate statistical tests.2. Consistency, i.e., the association is replicated in different settings and using different methods.3. Specificity, i.e., a single specific putative cause produces a specific effect.4. Dose-response relationship, i.e., an increasing amount of exposure to an agent, either in amount or duration, increases the risk of disease.5. Temporal relationship, i.e., exposure to the cause always precedes the effect.6. Biological plausibility, i.e., the relationship accords with currently accepted understanding of pathophysiological processes.7. Coherence, i.e., the association is compatible with current understanding of theory and practice.8. Experiment, i.e., the condition can be prevented or relieved by the regimen of an experiment, e.g., a randomized controlled trial.

1. Strength of the association, i.e., the size of the risk, as measured by appropriate statistical tests.

2. Consistency, i.e., the association is replicated in different settings and using different methods.

3. Specificity, i.e., a single specific putative cause produces a specific effect.

4. Dose-response relationship, i.e., an increasing amount of exposure to an agent, either in amount or duration, increases the risk of disease.

5. Temporal relationship, i.e., exposure to the cause always precedes the effect.

6. Biological plausibility, i.e., the relationship accords with currently accepted understanding of pathophysiological processes.

7. Coherence, i.e., the association is compatible with current understanding of theory and practice.

8. Experiment, i.e., the condition can be prevented or relieved by the regimen of an experiment, e.g., a randomized controlled trial.

Subjects: Public Health and Epidemiology.


Reference entries

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.