The exposure-adjusted incidence rate (EAIR) is an estimator proposed for assessing safety events in clinical trials and is sometimes used on efficacy endpoints when the follow-up time varies over the subjects, regardless of whether the follow-up time refers to duration of exposure to treatment or time in the study. The estimator relies on a strong assumption that the time to event is exponentially distributed with the same rate parameter for all subjects within a treatment group. Despite its apparent simplicity, the interpretation of EAIR might not be straightforward, especially in the presence of heterogeneity in the individual rates of event occurrence in the population under study. This article discusses potential issues with the EAIR estimator, as well as the interpretation of the resulting estimate, both within a given study and when considering a combination of multiple studies (aggregation, integration, pooling, meta-analysis). Considerations for confidence interval estimation are also discussed.
Kordzakhia et al. (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: