|
Session Title: The Next Generation Rapid-Learning Health System: Combining Decision Support Tools, Organizational Decision Making, and Dissemination and Implementation Science to Improve Healthcare
|
|
Panel Session 231 to be held in the Granite Room Section B on Thursday, Nov 6, 9:15 AM to 10:45 AM
|
|
Sponsored by the Health Evaluation TIG
|
| Chair(s): |
| Arne Beck,
Kaiser Permanente,
arne.beck@kp.org
|
| Abstract:
In this series of presentations, we provide the conceptual framework for a next generation rapid learning health system, along with examples of specific applications in healthcare program evaluation. Decision support tools are foundational to rapid-learning health systems. The first presentation will describe the use of these tools to conduct healthcare program evaluations in Kaiser Permanente Colorado. The second presentation will address the organizational context which can facilitate or impede the effectiveness of a rapid-learning health system. The third presentation will focus on the application of dissemination and implementation science to spread best practices identified through the rapid-learning evaluation process. The fourth presentation will be a case study of the development, evaluation, and dissemination of a palliative care program in Kaiser Permanente nationally that illustrates the aforementioned next generation rapid-learning health system concepts.
|
|
Rapid Learning Health Systems: Decision Support Tools
|
| Arne Beck,
Kaiser Permanente,
arne.beck@kp.org
|
|
Kaiser Permanente Colorado is developing a rapid-learning health system where electronic health records (EHRs) provide real-time opportunities to answer such questions as:
- What is the evidence base for procedures?
- What explains variations and increases in health care spending and use?
- How can the health of minorities and special populations be improved?
KP's HealthConnect is the largest civilian EHR system. It links all KP facilities and provides members, physicians and authorized providers with online access to clinical information 24/7. Additional tools include disease registries, technologies for collecting patient-centered and/or patient-health plan information and supporting patient-provider communication, and cost capture systems. This presentation describes rapid-learning system tools employed at KP Colorado that enable us to evaluate the impact of planned and natural health services experiments, and provide information for decision making with increasing rapidity. Examples illustrating the use of these tools in several healthcare program evaluations will be discussed.
|
|
|
Building a Rapid Learning Health System Culture
|
| R Sam Larson,
Kaiser Permanente,
r.sam.larson@kp.org
|
|
Rapid learning tools need organizational systems and processes that bring this rich information into the decision making realm of the organization. Like many nonprofits and healthcare organizations, KP Colorado lacks a policy on how and who makes decisions to stop, continue, select or scale-up healthcare programs and initiatives. Yet, KPCO is actively asking questions about the relative improvement in quality, service and affordability of most of our healthcare programs and seeking out emergent promising practices. This presentation explores how KPCO is intentionally moving from an implicit to an explicit culture of evaluation by identifying and coming alongside several strategic initiatives to conduct formative and summative evaluations and to incorporate findings into a rapid-learning cycle. We also discuss how we document those programs identified as 'best practices'. We conclude with a broad discussion on building an organizational culture supportive of rapid-learning and evidence-based decision making.
| |
|
Spreading Best Practices in a Rapid Learning Health System
|
| James Dearing,
Kaiser Permanente,
james.w.dearing@kp.org
|
|
The original rapid-learning health system concept focused primarily on the use of electronic health records (EHRs) to evaluate planned and natural health services experiments in real-time fashion. Little attention was given to the actual dissemination and implementation of healthcare programs if the results from these experiments were deemed successful. A critical component of a next generation rapid-learning system should be the application of dissemination and implementation science to understand barriers to adoption of best practices and to facilitate rapid, efficient, and effective spread of identified such practices. This presentation will describe the concepts from dissemination and implementation science, illustrate their application in healthcare, and demonstrate how, when paired with rapid-learning system decision support tools, they can be integrated into a more comprehensive and robust rapid-learning system model.
| |
|
Developing, Evaluating and Disseminating Best Practices in a Rapid Learning Health System: Palliative Care in Kaiser Permanente
|
| Doug Conner,
Kaiser Permanente,
douglas.a.connor@kp.org
|
|
Kaiser Permanente recently implemented an inpatient palliative care team to all of its hospitals in the US. This implementation was based on evaluation of a pilot study and randomized control trial (RCT) of the team in 3 hospitals. The pilot study highlighted the resistance of hospital staff and physicians to both the concept of a multi-disciplinary palliative care team and to the use of a RCT. These findings were used to develop educational strategies to improve acceptance of the larger RCT by hospital staff and physicians. During the RCT, interim analyses were done to document patient benefits and cost savings providing evidence to health plan leaders, engaged in budget planning, for the subsequent implementation of the team as a continuing consult service. Final evaluation provided evidence that the model followed by one hospital team was most effective and this model was implemented in the national roll-out to all hospitals.
| |