Bradford Hill Criteria

Description

A structured checklist of viewpoints (strength, consistency, temporality, plausibility and others) for building confidence that an observed association reflects causation when experiments are impossible.

Hill proposed nine viewpoints for judging whether an association is causal: strength, consistency, specificity, temporality, biological gradient, plausibility, coherence, experiment and analogy. They are not proof but a disciplined way to weigh observational evidence. Vivek Gopinath's playbook applies them as a narrative check in UX when randomisation is not possible: temporality (the change preceded the effect), strength (a 19-point lift), consistency (replicated across segments) and plausibility (aligns with mental-model evidence). The value is in forcing explicit reasoning about alternative explanations rather than declaring causality from correlation.

Attributes

Loss type
Unmigrated
Stage
Post-launch
Effort
Hours

Steps

  1. Write the causal claim and the observational evidence for it.
  2. Check temporality: did the design change precede the outcome shift?
  3. Assess strength and dose-response (e.g. size of lift, more exposure giving more effect).
  4. Check consistency across segments, platforms and time periods.
  5. Assess plausibility and coherence with qualitative mechanism evidence.
  6. Summarise which viewpoints are met and state the residual uncertainty.

Sources

  • Vivek Gopinath (2025), Forward-Thinking UX Research Playbook: A Mixed-Methods, Decision-Ready Approach from Prediction to Validation
  • Lost Methods in UX — A Complete Treatise (Gopinath)
  • Hill AB (1965) The environment and disease: association or causation? Proc R Soc Med 58:295-300
  • Rothman KJ & Greenland S (2005) Causation and causal inference in epidemiology. AJPH
  • Hill, The environment and disease: association or causation? (1965), Proceedings of the Royal Society of Medicine
  • Rothman, Greenland & Lash, Modern Epidemiology (3rd ed., 2008)

Also known as

  • Hill's criteria for causation
  • Hill's viewpoints
  • Hill's criteria
  • Bradford Hill considerations
  • Bradford Hill Criteria for UX Causation
  • Bradford Hill viewpoints
  • Hill's considerations
  • Bradford Hill Criteria for Causal Assessment

Categories

  • Epidemiology & Population Health