Award
NIH Reporter #5F31DK141237-02
Adjusting for selection bias due to missing eligibility data in electronic health records-based observational studies
Recipient
Harvard University D/B/A Harvard School of Public Health
Award Amount
$36,238.00
Ceiling
$36,238.00
Awarded
August 13, 2025
Identifier
5F31DK141237-02
The award funds research to develop statistical methods to correct selection bias from missing eligibility data in EHR-based studies, focusing on bariatric surgery outcomes.
Description
This project focuses on developing statistical methods to address selection bias caused by missing data in electronic health records (EHR) research. It aims to create inverse probability weighting (IPW) frameworks tailored for sequential target trial emulations examining time-to-event endpoints, and to develop robust estimators that are resistant to model misspecification. The research will analyze long-term health outcomes among individuals who underwent bariatric surgery and non-surgical patients eligible for such surgery, to evaluate the efficacy and safety of bariatric procedures in treating obesity and type 2 diabetes.