Statistics and Actuarial Science, University of Waterloo –
Addressing Informative Presence Bias in Analyses of Electronic Health Records
Medical researchers frequently investigate relationships between complex diseases like autism spectrum disorder and other potentially related conditions.
He will develop statistical tools to correct the bias that exists in examinations of autism patients due to the fact that research often relies on electronic health records. The use of these records can lead to spurious associations given that patients with autism often visit the doctor more often than others. These proposed methods have the potential to improve the way medical research is conducted whenever patients’ conditions lead them to have higher than average interaction with the medical system.
Recipient of a Banting-CANSSI Ontario Discovery Award in Data Science.