Multivariate Bayesian Logistic Regression for Analysis of Clinical Study Safety Issues

TL;DR

This paper introduces multivariate Bayesian logistic regression (MBLR), a novel statistical method for analyzing sparse clinical safety data across multiple studies, enabling better detection of adverse event risks and subgroup vulnerabilities.

Contribution

The paper presents MBLR, a new Bayesian approach that borrows strength across related issues and handles sparse data, improving safety analysis in clinical studies.

Findings

MBLR effectively analyzes sparse safety data.

The method identifies vulnerable subgroups based on covariates.

Simulations demonstrate favorable distributional properties.

Abstract

This paper describes a method for a model-based analysis of clinical safety data called multivariate Bayesian logistic regression (MBLR). Parallel logistic regression models are fit to a set of medically related issues, or response variables, and MBLR allows information from the different issues to "borrow strength" from each other. The method is especially suited to sparse response data, as often occurs when fine-grained adverse events are collected from subjects in studies sized more for efficacy than for safety investigations. A combined analysis of data from multiple studies can be performed and the method enables a search for vulnerable subgroups based on the covariates in the regression model. An example involving 10 medically related issues from a pool of 8 studies is presented, as well as simulations showing distributional properties of the method.