Poisson regression with missing durations of exposure

In this paper, we develop Poisson-type regression methods that; require the durations of exposure be measured only on a possibly nonrandom subset of t he cohort members. These methods can be used to make inferences about the i ncidence density during exposure as well as the ratio of incidence densitie s during exposure versus not during exposure. Numerical studies demonstrate that the proposed methods yield reliable results in practical settings. We describe an application to a population-based case-control study assessing the transient increase in the risk of primary cardiac arrest during leisur e-time physical activity.