Case study
Developing a Prognostic Information System for Personalized Care in Real Time
Authors:
Tracy A. Lieu ,
The Permanente Medical Group, Kaiser Permanente Northern California; Division of Research, Kaiser Permanente Northern California, US
About Tracy A.
Tracy Lieu, MD, MPH, is Director of the Division of Research, Kaiser Permanente Northern California. She leads a group of 600 people who conduct studies in clinical effectiveness, delivery science, and epidemiology to enhance the health of Kaiser Permanente members and society at large. Her national roles have included membership on the U.S. Preventive Services Task Force and the Advisory Committee on Immunization Practices and the chair of the Health Services Organization and Delivery study section of the National Institutes of Health. From 1999 to 2012, Dr. Lieu was a professor and the founding director of the Center for Healthcare Research in Pediatrics in the Department of Population Medicine, Harvard Pilgrim Health Care Institute and Harvard Medical School. Dr. Lieu is a practicing pediatrician who has led nationally recognized work in vaccine safety and policy and childhood asthma. She was elected to the National Academy of Medicine for her research in decision sciences and cost-effectiveness analysis in health policy.
Lisa J. Herrinton,
The Permanente Medical Group, Kaiser Permanente Northern California; Division of Research, Kaiser Permanente Northern California, US
Dimitri E. Buzkov,
The Permanente Medical Group, Kaiser Permanente Northern California; Division of Research, Kaiser Permanente Northern California, US
Liyan Liu,
The Permanente Medical Group, Kaiser Permanente Northern California; Division of Research, Kaiser Permanente Northern California, US
Deborah Lyons,
The Permanente Medical Group, Kaiser Permanente Northern California; TPMG Technology Group, Kaiser Permanente Northern California, US
Romain Neugebauer,
The Permanente Medical Group, Kaiser Permanente Northern California; Division of Research, Kaiser Permanente Northern California, US
Tami Needham,
The Permanente Medical Group, Kaiser Permanente Northern California; TPMG Technology Group, Kaiser Permanente Northern California, US
Daniel Ng,
The Permanente Medical Group, Kaiser Permanente Northern California; Division of Research, Kaiser Permanente Northern California, US
Stephanie Prausnitz,
The Permanente Medical Group, Kaiser Permanente Northern California; Division of Research, Kaiser Permanente Northern California, US
Kam Stewart,
The Permanente Medical Group, Kaiser Permanente Northern California; TPMG Technology Group, Kaiser Permanente Northern California, US
Stephen K. Van Den Eeden,
The Permanente Medical Group, Kaiser Permanente Northern California; Division of Research, Kaiser Permanente Northern California, US
David M. Baer
The Permanente Medical Group, Kaiser Permanente Northern California; Department of Oncology, Kaiser Permanente Oakland Medical Center, US
Abstract
Context: Electronic medical records hold promise to transform clinical practice. However, technological and other barriers may preclude using them to guide care in real time. We used the Virtual Data Warehouse (VDW) to develop a tool that enables physicians to generate real-time, personalized prognostic information about survival after cancer.
Case description: Patients with cancer often ask their oncologists, “Have you ever seen a patient like me?” To help oncologists answer this question, we developed a prototype Prognostic Information System (PRISM), a web-based tool that gathers data about the index patient from Kaiser Permanente’s clinical information systems, selects a historical cohort of similar patients, and displays the survival curve of the similar patients relative to key points in their treatment course.
Findings and major themes: The prototype was developed by a multidisciplinary team with expertise in oncology, research, and technology. We have completed two rounds of user testing and refinement. Successful development rested on: (1) executive support and a clinical champion; (2) collaboration among experts from multiple disciplines; (3) starting with simple cases rather than ambitious ones; (4) extensive research experience with the Virtual Data Warehouse, related databases, and an existing query tool; and (5) following agile software development principles, especially iterative user testing.
Conclusion: Clinical data stored in health care systems’ electronic medical records can be used to personalize clinical care in real time. Development of prognostic information systems can be accelerated by collaborations among researchers, technology specialists, and clinicians and by use of existing technology like the Virtual Data Warehouse.
How to Cite:
Lieu TA, Herrinton LJ, Buzkov DE, Liu L, Lyons D, Neugebauer R, et al.. Developing a Prognostic Information System for Personalized Care in Real Time. eGEMs (Generating Evidence & Methods to improve patient outcomes). 2019;7(1):2. DOI: http://doi.org/10.5334/egems.266
