Home  >  CTSA Programs > Translational Science Research Program

Translational Science
Research Program

The focus of Stanford's Clinical, Translational, and Science of Translation Research and Education Program (Element E) is on innovations to develop and test translational science systems for the collection and curation of real-world data with representative, diverse, national populations in coordination with effective and efficient structures for education and training to use these data for impact.

FACULTY LEADS

David Rehkopf, Sc.D., MPH
Faculty Co-Lead

Manisha Desai, PhD
Faculty Co-Lead

Project 1. Develop and evaluate the impact of a hands-on course and data use support structures for using real-world data for clinical and translational research.

Real-world data includes data routinely collected for administrative purposes, such as electronic health record data, medical claims billing data, and data gathered from digital sources. While real-world data users typically come to translational research with clinical experience and some statistical training, a barrier in translational science is a lack of training specific to the use of electronic health records and other real-world data. Our approach in project 1 is distinct and complementary to statistical support from iBERD and courses in statistics and epidemiology in focusing on unique aspects of electronic health records and medical claims data. This Project addresses an important translational science gap, which, if successfully addressed, will accelerate translational research with real-world data.

Project Leads

David Rehkopf, Sc.D., MPH

Manisha Desai, PhD

Ruth O'Hara, PhD

Tina Hernandez-Boussard, MS, PhD

Project 2. Test for systematic bias in social deprivation measures used in translational science infrastructure.

There is an increasing recognition that accounting for social risks in translational research is critical. However, due to a lack of individual data on social determinants of health in most real-world data sets, small area measures are the most widely used metrics for capturing social determinants of health. Through an ongoing project with the U.S. Census Bureau, we will analyze sources of bias in using area-based social deprivation measures and develop recommendations for the use of NIH PhenX-based measures of social determinants of health. We will examine bias in predicting health outcomes by intersectional characteristics of socioeconomic position, gender, race, ethnicity, and rural area of residence. We will work to assess the potential of new area-based measures of structural racism using machine learning approaches to capture real-world sources of area-level data. The findings from this project will be generalizable to guiding the use of area-based social deprivation measures in other EHR data sources including N3C.

Project Leads

David Rehkopf, Sc.D., MPH

Tainayah Thomas, PhD, MPH

Project 3. Modernize the digital research infrastructure of The PRIDE Study to better characterize the health and inequities affecting sexual and gender minority people.

Limited data inhibits our ability to describe the factors influencing health and disease among sexual and gender minority (SGM) communities, to identify inequities, and to develop testable interventions. The Population Research for Identity and Disparities for Equality (PRIDE) Study is a national, online, longitudinal cohort study of SGM adults in the U.S. that examines how the identity and experiences of SGM people influences their physical, mental, and social health. Its current secure, cloud-based infrastructure was built in 2016, and a number of issues limit the full use of this real-world data for translational research. Properly investigating SGM health requires collection of specialized data such as a detailed gender-affirming hormone history (medication, dose, route, frequency, start/stop dates) and timed collection of gender-affirming surgery data to assess expectations, outcomes, complications, and satisfaction. A renovation of The PRIDE Study’s digital infrastructure will curate new types of real-world data for translational research, improve the user experience for retention and data accuracy, and develop a realworld data warehouse for efficient, effective data sharing and analysis to improve our understanding of these underrepresented and vulnerable communities.

Project Lead

Mitchell Lunn, MD, MAS

Spectrum