In-Person and Virtual Options

In-Person: Parkvale 305 A/B (lunch provided)

Zoom Meeting: Register Here

In public health and biomedical association studies, binary outcome variables such as diagnoses may be subject to misclassification, resulting in substantial bias in effect estimates. In this talk, new strategies are presented to recover unbiased parameter estimates in association studies with covariate-dependent outcome misclassification, without requiring gold standard measures. A study of gender-related misdiagnosis rates in myocardial infarction is presented to demonstrate the utility of this analysis technique in cases where misclassification not only occurs but is more prevalent for certain groups.