Abstract
We describe a spreadsheet method for implementing Bayesian causality assessment of suspected adverse drug reactions (ADRs). The method can be implemented on paper or using a computerised spreadsheet program. Both approaches ensure coherence in the combination of component estimates to arrive at an overall assessment. The components themselves are obtained from the literature or the clinical judgment of the user. We demonstrate use of the method using a case of pseudomembranous colitis suspected to be due to antibiotics. The approach has the following advantages: 1. It allows the simultaneous assessment of multiple causes (no limit); 2. It allows the user to use relative likelihoods which are often easier to estimate than absolute probabilities; and 3. The computer program provides instant calculation and graphical display of the posterior probabilities allowing the impact of any factor to be instantly appreciated
We envisage that the method will be useful for the assessment of groups of similar cases in a regulatory or pharmaceutical setting.
Get full access to this article
View all access options for this article.
