The Fred Hutch Biostatistics Program hosts seminars featuring presentations by Fred Hutch and outside scientists to share their latest developments and recent research. Each seminar includes an hour-long presentation and discussion during which speakers showcase their work and findings.
This seminar will be via Zoom.
In this big data era, we can easily observe substantial amounts of clinical data in large observational studies or electronic health records (EHR). Data accuracy can vary according to measurement methods. For example, self-reported medical history can include bias, such as recall bias or response bias. In contrast, gold standard diagnostic tests are less likely to be biased but may not be available on all individuals in a large prospective study due to cost or participant burden. We are motivated to study what benefit we can gain by augmenting analyses of the gold standard disease outcome with error-prone self-reported disease diagnoses in regression for time-to-disease onset. The proposed model addresses left-truncation and interval-censoring in time-to-disease onset outcomes while correcting errors in self-reported disease diagnosis in a joint likelihood for the gold standard and error-prone outcomes. The proposed model is applied to the Hispanic Community Health Study/ Study of Latino data to quantify risk factors associated with diabetes onset.