implementation of a "local" learning health system, clinical quality can
be improved while simultaneously reducing health care costs, demonstrated a recent pilot study by a team from Nationwide Children's Hospital
and The Ohio State University.
was published in a recent issue of Developmental Medicine and Child
Neurology with an editorial published online in the Journal of the American Medical Association
‘In a learning health system, the electronic health record is utilized to drive research and personalized treatments based on data from patients with similar conditions and risk factors.’
Providing high-quality patient care while reducing costs is a
significant goal in the current health care reform environment. The
Institute of Medicine has specifically called for the establishment of
"learning health systems" to address this challenge.
In a learning
health system, the electronic health record is utilized to drive
research and personalized treatments based on data from patients with
similar conditions and risk factors.
Current efforts to establish learning health systems have been
designed primarily around the integration of electronic health records
among multiple hospital systems. While substantial progress has been
made with this approach, the operational and technical barriers of
bringing together such diverse hospital systems led the Nationwide
Children's/Ohio State team to consider an alternative based on the
concept of a "local" learning health system.
Local systems start with the integration of research, clinical care
and quality improvement within a specific health service and then use
the knowledge gained to systematically deliver quality improvement and
cost savings. Those clinical improvements can be expanded to other
programs within the hospital or across the health care network.
The team from Nationwide Children's and Ohio State developed and
implemented a local learning health system model of care with
optimization of the electronic health record and a robust care
coordination system at its core. The project, named the "Learn From
Every Patient" (LFEP) Program, was piloted with a group of 131 children
treated through the Nationwide Children's Cerebral Palsy (CP) Program.
Children treated in the LFEP Program during the 12-month study
period experienced a 43% reduction in total inpatient days; a 27% reduction in inpatient admissions; a 30% reduction in
emergency department visits; and a 29% reduction in urgent care
visits. LFEP Program implementation resulted in reductions in health
care costs of $1.36 million.
"Introducing electronic health record-supported care that integrated
clinical care, quality improvement and distinct, clinician-driven
research resulted in large reductions in health care utilization,
greatly reduced healthcare charges and improved care coordination,"
explains William E. Smoyer, vice president, Clinical and
Translational Research and director, Center for Clinical and
Translational Research at Nationwide Children's Hospital, senior author
on the study paper and lead author of the editorial.
The cost of implementing the pilot program, including care
coordination expenses, totaled approximately $225,000 over the first
year. However, these costs were only 16% of the reduced total
health care expenditures realized during the first year of the program,
representing a savings of approximately $6 for each $1 invested. In
addition, the clinicians working in the CP clinic reported that this
approach simplified their documentation, enabling more patients to be
seen during each session.
Consistent with the goals of a learning health system, the data
collected in the electronic health record included five
clinician-inspired "learning projects" conducted during routine clinical
care for the pilot LFEP group. One of these studies has been published
in a peer-reviewed journal and one is in press.
"The use of the LFEP Program in our clinic has resulted in robust
research data that can be used for publication and for implementing
evidence-based improvements in clinical care," says author Garey Noritz, medical director of the Cerebral Palsy Program at Nationwide
Children's Hospital and associate professor at The Ohio State
University. "For example, children with cerebral palsy are prone to hip
dislocation, so our practice had been to perform x-rays on every child
every year. An examination of our data revealed that this was excessive
for children with mild CP. We have since changed our protocol for
these patients, reducing their exposure to radiation, as well as costs."
"These results demonstrate that a learning health system can be
developed and implemented in a cost-effective manner, and can integrate
clinical care and research to systematically drive simultaneous clinical
quality improvement and reduced health care costs," notes co-author
Peter Embi, associate professor and interim chair of Biomedical
Informatics and chief research information officer at The Ohio State
As described in the journal article, the early experience of this
local learning health system (i.e. a bottom-up approach) suggests that
this method has the potential to be an effective complementary or
alternative strategy to the top-down approach of learning health systems
to achieve the overall goal of clinical transformation at the national
level. The LFEP approach is being used as the foundation for the
recently created Cerebral Palsy Research Network, which will connect multiple CP programs across the U.S. and Canada.
"The adage that our local environment will affect our global health
efforts has never been truer than now. With all of the externalities
impacting health care, we need to be innovative, progressive and
integrative relative to research and operations at every opportunity,"
notes senior author, Susan Moffatt-Bruce, professor of
Surgery and Biomedical Informatics and chief quality officer at The Ohio
State University Wexner Medical Center.
"There are numerous future challenges related to demonstrating the
potential value of using local learning health systems to develop a
national adaptable learning health system," says Dr. Smoyer. "We still
need to investigate whether local learning health systems can be
successfully scaled across the wide variety of clinical programs that
comprise the nation's health care systems. However, these preliminary
findings from implementing this disruptive innovation at a single center
demonstrate that learning health systems are indeed able to be
implemented and that their feasibility and effectiveness can be tested."