Times are displayed in (UTC-04:00) Eastern Time (US & Canada) Change
3/10/2025 |
8:30 AM – 12:00 PM |
Conference A
W04: Conducting Retrospective Analyses Using Aggregated Real World Data in Cosmos: Lessons, Challenges, and Best Practices
Presentation Type: Workshop
Session Credits: 3
Conducting Retrospective Analyses Using Aggregated Real World Data in Cosmos: Lessons, Challenges, and Best Practices
Presentation Time: 08:30 AM - 12:00 PM
Abstract Keywords: Outcomes Research, Clinical Epidemiology, Population Health, Data-Driven Research and Discovery, Data Mining and Knowledge Discovery, Secondary Use of EHR Data, Public Health Informatics, Data Quality, Clinical and Research Data Collection, Curation, Preservation, or Sharing, Machine Learning, Generative AI, and Predictive Modeling
Primary Track: Clinical Research Informatics
Programmatic Theme: Health Data Science and Artificial Intelligence Innovation: From Single-Center to Multi-Site
Cosmos is an aggregated EHR platform containing information on nearly 300 million patient records from over 290 participating health systems for the purpose of accelerating medical research and creating tools that improve patient care. This instructional introductory workshop will introduce attendees to best practices for assessing feasibility and conducting research using Cosmos. The workshop will introduce the different toolsets that can be used in the platform, covering benefits and limitations of each tool and comparison between them. Presenters will share their experiences from real-world analyses, including discussion of obstacles encountered in the real-world data and strategies they employed to overcome those challenges. Discussion will include methodology, code review, and best practices for conducting studies using Cosmos. After learning about these topics, attendees will be invited to break into groups and discuss ideas for retrospective analyses of EHR data, then work with presenters who will apply Cosmos tools to assess feasibility of attendee ideas. During this phase, presenters and attendees will collaborate to design initial analyses and begin to undertake those analyses. While this session will be specific to Cosmos, the lessons learned regarding the inherent biases in aggregated EHR-derived RWD and how to account for those biases are broadly applicable in the medical informatics community. No prior programming experience is required from attendees, nor is access to Cosmos required – however, Cosmos users will especially benefit from this session and are encouraged to independently leverage Cosmos themselves in collaboration with presenters during the interactive component of this workshop.
Speaker(s):
Harry Freedman, BS
Epic
Yasir Tarabichi, MD, MSCR
MetroHealth
Kevin Dysart, MD, MBI
Nemours Alfred I. duPont Hospital for Children
Laila Rasmy, PhD, MSc, MBA, RPh.
UTHealth MSBMI
Gabriel Najarro, PA-C
Emory Healthcare
Author(s):
Harry Freedman, -- - Epic; Yasir Tarabichi, MD, MSCR - MetroHealth; Kevin Dysart, MD, MBI - Nemours Alfred I. duPont Hospital for Children; Gabriel Najarro, PA-C - Emory Healthcare; Laila Rasmy, PhD, MSc, MBA, RPh. - UTHealth MSBMI;
Presentation Time: 08:30 AM - 12:00 PM
Abstract Keywords: Outcomes Research, Clinical Epidemiology, Population Health, Data-Driven Research and Discovery, Data Mining and Knowledge Discovery, Secondary Use of EHR Data, Public Health Informatics, Data Quality, Clinical and Research Data Collection, Curation, Preservation, or Sharing, Machine Learning, Generative AI, and Predictive Modeling
Primary Track: Clinical Research Informatics
Programmatic Theme: Health Data Science and Artificial Intelligence Innovation: From Single-Center to Multi-Site
Cosmos is an aggregated EHR platform containing information on nearly 300 million patient records from over 290 participating health systems for the purpose of accelerating medical research and creating tools that improve patient care. This instructional introductory workshop will introduce attendees to best practices for assessing feasibility and conducting research using Cosmos. The workshop will introduce the different toolsets that can be used in the platform, covering benefits and limitations of each tool and comparison between them. Presenters will share their experiences from real-world analyses, including discussion of obstacles encountered in the real-world data and strategies they employed to overcome those challenges. Discussion will include methodology, code review, and best practices for conducting studies using Cosmos. After learning about these topics, attendees will be invited to break into groups and discuss ideas for retrospective analyses of EHR data, then work with presenters who will apply Cosmos tools to assess feasibility of attendee ideas. During this phase, presenters and attendees will collaborate to design initial analyses and begin to undertake those analyses. While this session will be specific to Cosmos, the lessons learned regarding the inherent biases in aggregated EHR-derived RWD and how to account for those biases are broadly applicable in the medical informatics community. No prior programming experience is required from attendees, nor is access to Cosmos required – however, Cosmos users will especially benefit from this session and are encouraged to independently leverage Cosmos themselves in collaboration with presenters during the interactive component of this workshop.
Speaker(s):
Harry Freedman, BS
Epic
Yasir Tarabichi, MD, MSCR
MetroHealth
Kevin Dysart, MD, MBI
Nemours Alfred I. duPont Hospital for Children
Laila Rasmy, PhD, MSc, MBA, RPh.
UTHealth MSBMI
Gabriel Najarro, PA-C
Emory Healthcare
Author(s):
Harry Freedman, -- - Epic; Yasir Tarabichi, MD, MSCR - MetroHealth; Kevin Dysart, MD, MBI - Nemours Alfred I. duPont Hospital for Children; Gabriel Najarro, PA-C - Emory Healthcare; Laila Rasmy, PhD, MSc, MBA, RPh. - UTHealth MSBMI;