You are not currently logged in.
Access JSTOR through your library or other institution:
If You Use a Screen ReaderThis content is available through Read Online (Free) program, which relies on page scans. Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
The Probability of Causation under a Stochastic Model for Individual Risk
James Robins and Sander Greenland
Vol. 45, No. 4 (Dec., 1989), pp. 1125-1138
Published by: International Biometric Society
Stable URL: http://www.jstor.org/stable/2531765
Page Count: 14
Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
Preview not available
In this paper we offer a mathematical definition for the probability of causation that formalizes the legal and ordinary-language meaning of the term. We show that, under this definition, even the average probability of causation among exposed cases is not identifiable from epidemiologic data. This is because the probability of causation depends both on the unknown mechanisms by which exposure affects disease risk and competing risks, and on the unknown degree of heterogeneity in the background disease risk of the exposed population. We derive the maximum and minimum values for the probability of causation consistent with the observable population quantities. We also derive the relationship of the "assigned share" (excess incidence rate as a proportion of total incidence rate) to the probability of causation.
Biometrics © 1989 International Biometric Society