Abstract: Body Mass Index (BMI) is an important source of clinical knowledge about a patient and is increasingly used in public health research to understand health outcomes. In electronic health records (EHR), patient height and weight information are messy. For BMI measurements to be used in clinical studies, efforts to account for this heterogeneity of clinical data are necessary to make BMI measurements robust and accurate. I posit a framework that handles this heterogeneity of data.

Learning Objective 1: Implement state of the art informatics and statistical methodology to improve height and weight data for BMI calculation from clinical EHR data to improve data quality.


Nnenna Ibeanusi (Presenter)

John Borsi, IBM
Wei Yao, IBM

Presentation Materials: