Multilevel attributable fraction scenario
Combine exposure-category prevalence and causal risk-ratio assumptions.
Your entries stay in this page’s memory and are not submitted to us. Downloads are generated locally. Reloading or leaving the page clears your entries. Do not enter patient information.
Fractions must be nonnegative and sum to one within 1e−9; risk ratios positive. Include the reference category with RR=1. Categories must be exhaustive and exclusive.
The starting values are hypothetical. Results update as you edit. Use aggregate or synthetic inputs only.
Calculation results
- Population relative risk
- 1.35
- Attributable fraction scenario
- 0.25925926
- Interpretation
- Requires causal and intervention assumptions, not association alone.
How to interpret the output
PAF=(sum(p×RR)−1)/sum(p×RR) compares observed exposure mix with everyone at reference risk. A causal interpretation requires valid causal risk ratios, transportability and a feasible intervention without other changes. It can be negative for protective exposure mixes. This is not a case-based adjusted PAF estimator.
Method referenceSource checked 2026-11-06. Read the reproducible calculator validation report.