Abstract
This talk reviews the problem of causality assessment of adverse drug reactions from the point of view of probability theory and decision analysis. Causality assessment is a special case of the general problem of probability appraisal, the measurement of uncertainty. The purpose of assessing causality is to help make correct decisions. Thus, probability theory and decision analysis can offer useful lessons for workers in the causality assessment enterprise. In particular: (a) There seems to be some confusion about just what the causality problem is and how individual causality assessments can be incorporated with other information to guide clinical, industrial, regulatory, and epidemiological decision making. Rational decision analysis of the situations to which causality assessment is applied requires that this confusion be resolved. (b) Arguments from descriptive and prescriptive probability theory favor the use of algorithms to improve the accuracy, reliability and validity of causality assessment. (c) Probability theory can suggest criteria for good causality assessment algorithms that can be used to compare and improve existing algorithms.
Get full access to this article
View all access options for this article.