Loading...

Data Science in Clinical Care (2024-2025)

The Data Science in Clinical Care team examined how clinical decision support (CDS) tools can be more effectively implemented and sustained in real-world healthcare settings. Building on prior work, students, faculty and clinicians investigated the complex systems that shape the adoption and use of CDS tools at Duke University Hospital.

This year, the team studied two use cases, the first being a Readmission Risk Score designed to identify patients at high risk of hospital readmission. Using participatory methods such as Group Model Building, the team engaged case managers, physicians and therapists in workshops to map the dynamic factors influencing adoption. Their analysis revealed that external pressures, such as institutional mandates, could spur initial use, but sustained adoption depended on whether clinicians viewed the tools as useful in their daily workflows. These findings were translated into causal loop diagrams and simulation models that demonstrated how perceptions of value directly shaped long-term CDS integration.

The second use case focused on groin surgical site infections to assess clinicians’ views on modifiable risk factors and inform the development of a new CDS tool tailored to this issue. Results from the project were shared at the 2025 Bass Connections Showcase and the System Dynamics Society Conference. The team is preparing a manuscript to further disseminate findings while continuing to expand training opportunities for students in implementation science, qualitative research and health informatics. Their work highlights how participatory system dynamics can strengthen CDS adoption by aligning technology design with the realities of clinical practice.

Timing

Fall 2024 – Spring 2025

Team Outputs

System Dynamics Society Conference (Presentation, Boston, MA, August 2025)

Readmission Risk Score project (Fortin Foundation Bass Connections Showcase, April 16, 2025)
 

See earlier related team, Data Science in Clinical Care (2023-2024).

 

Image: Sara Schick and Leah Kranja assist a patient in infusion at Duke Women’s Cancer Care Raleigh (Patient is staff posing, Ivey Aronowitz), by Erin Hull/Duke Health

Team Leaders

  • Adam Johnson, School of Medicine: Surgery
  • Nina Sperber, School of Medicine, School of Medicine: Population Health Sciences

Graduate Team Members

  • Sarah Haas, Population Health Sciences-MS
  • Hana Shafique, Medical Student
  • Eden Singh, Medical Student
  • Shiling Zheng, Biomedical Engineering-MS

Undergraduate Team Members

  • Samantha Hamelsky, Statistical Science (BS)
  • Theresa Kiki-Teboum, Chemistry (AB)
  • Afraaz Malick, Computer Science (AB)
  • Rishab Pulugurta, Interdepartmental
  • Jacqueline Rodriguez, Public Policy (AB); Computer Science (AB2)
  • Kriti Vasudevan, Computer Science (BS); Public Policy (AB2)

Team Contributors

  • Benjamin Goldstein, School of Medicine: Biostatistics and Bioinformatics
  • Scott Rockart, Fuqua School of Business