Biostatistics Support

The Office of Clinical Research provides biostatistical support for non-grant and pre- and post-award clinical research projects. We specialize in: 

  • clinical research design
  • null hypothesis significance testing
  • linear modeling
  • data mining, tidying, and visualization
  • manuscript, grant, abstract, and poster/podium presentation preparation
  • Cerner Learning Health Network powered by Oracle - “a nationwide network of diverse health systems that share de-identified, real-world data to help advance clinical research and trials”
  • Cosmos Data Model powered by Epic - “the largest integrated database of clinical information in the US”

To request biostatistical support, please complete this form and we will respond to you shortly.

Acknowledgement and Authorship

We expect acknowledgement or co-authorship for our analytical products based on good faith assessment of our contributions.

Open Science

We conduct analyses primarily in the R statistical computing language. To promote transparent and reproducible clinical research, we encourage investigators to agree to publishing R code and de-identified data (in the form of GitHub repositories) for each manuscript that reaches publication.

Additional Resources

The Biostatistics and Epidemiology Consulting Service (BECS): provides biostatistical and epidemiologic support to improve the quality of health-related research conducted at the George Washington University and by affiliated research groups.

Biomedical Informatics Center: The Biomedical Informatics Center has access to the Cerner Health Facts database and uses statistical and machine learning methods for the development of the desired dynamic risk assessment.

SMHS and University Research Core Facilities


Cosmos combines billions of clinical data points in a way that forms a high quality, representative, and integrated data set that can be used to change the health and lives of people everywhere. Cosmos includes information from 200 Epic organizations, with the dataset containing more than 184 million patients and 7 billion encounters.

There are two types of Cosmos access: 

  1. Cosmos SlicerDicer access is intended for users who have a research project use case that would benefit from access to summary-level national data. 
  2. Cosmos Data Science access is intended for researchers and data scientists familiar with R, Python, and/or SQL who have a research or project use case that would benefit from access to deidentified national data. This access requires additional training at Epic headquarters in Verona, WI or online with successful completion of a project and exam. Only users with an articulable project or use case and prior R/Python/SQL experience should receive this level of access.

For more information, checkout the Cosmos website.

To request Cosmos access, please complete the Cosmos Access Request Form. Send completed forms to

Cerner Learning Health Network

Learning Health Network (LHN) comprises 113 health systems representing over 100 million patient records across 44 states and Washington, DC. These heterogeneous, de-identified data are merged into Oracle Health’s Real-World Data (RWD) core model and housed in Cerner’s HealtheDataLab cloud-based data science ecosystem, built and deployed on Amazon Web Services.

GWU has been a contributing member of LHN since November 2020 and has access to RWD and HealtheDataLab (pronounced "healthy datalab"). MFA researchers may work with Office of Clinical Research personnel to develop research proposals that leverage these tools and to also request Letters of Support from LHN. MFA researchers have submitted ten LHN research proposals as of April 2023. For more information, please contact