Instructor
Department of Biomedical Data Science
Dr. Tamang is the Assistant Faculty Director, Data Science, at the Stanford Center for Population Health Science and an Instructor at the Department of Biomedical Data Science. As a computer scientist with training in biology, health services research and biomedical informatics, Dr. Tamang works with interdisciplinary teams of experts on population health problems of public interest. Integral to her work, is the analysis of large and complex population-based datasets, using techniques from natural language processing, machine learning and deep learning. She brings extensive experience with US and Danish population-based registries, Electronic Medical Records from various vendors, administrative claims and other types of observational health data sources in the US and internationally; also, constructing, populating and applying knowledge-bases for automated reasoning. Dr. Tamang has developed open source tools for the extraction of health information from unstructured free-text patient notes and licensed machine learning prediction models to health analytics startups. Dr. Tamang affiliated with the Department of Veterans Affairs, the National Bureau of Economic Review, the Department of Rheumatology, University of California San Francisco and in Denmark, the Department of Clinical Epidemiology, Arhaus University. In addition to her more traditional research activities, she also functions a mentor for the Stanford working group Stats for Social Good. As a SAGE researcher, Dr. Tamang’s work will focus on opioid safety.
https://profiles.stanford.edu/suzanne-tamang
SAGE Project: Safer Opioid Prescribing for Vulnerable Populations and Developing the US Opioid Atlas
We are using Veterans Health Administration data to examine trends among veterans with a diagnoses opioid use disorder that experienced a Serious Adverse Event (e.g., non-fatal overdose and suicide attempts) with a focus on under represented minority older adults. Our analysis is based on the approximately eight million veterans that sought care at the VHA from 2016 through the end of 2018. Although our FDA collaborators have conducted similar work in other large population-based US datasets such as Optum and Truvan, their ability to characterize trends among under-represented minorities that are at high risk of opioids is limited. Our findings have the potential to improve mental healthcare management at the VA and patient-centered care initiatives at the FDA.
To help opioid investigators study these questions, we will develop a data visualization tool for exploring data on legal opioid ordering, by country and time, collected by Electronic Medical Record systems across 42 US States. Our design borrows ideas from the data visualization and multivariate statistics communities, especially the principles of linking and dimensionality reduction. Our work is relevant to policymakers and pain researchers who wish to systematically assess country-level factors that contribute to differences in opioid access for patients with cancer and surgery-related pain. Our tools. and the code behind it, will be freely available with an open source license.
