Champions for Children’s Health
NACHRI 2009 Annual Meeting Session
Comparative Effectiveness: Embracing Bold Actions to Improve Outcomes for Children
Ramesh Sachdeva, MD, PhD, DBA, JD, Vice President, Quality
and Outcomes
Tom Rice, MD, FAAP, FCCP, Medical Director, Pediatric Intensive Care
Unit
Children’s Hospital of Wisconsin, Milwaukee, WI
Randall Wetzel, MB, BS, FAAP, FCCM, Chairman, Department of
Anesthesiology Critical Care Medicine, Professor of Critical Care
Medicine
Childrens Hospital Los Angeles, Los Angeles, CA
Mary Gorman, MPH, MBA, Vice President, Education, Member Services and
Management Information Services
NACHRI, Alexandria, VA
Learning objectives: After viewing this session attendees should be able to:
-
Understand the need for risk-adjustment for conducting comparative effectiveness analysis
-
Lead to a better understanding of the process of severity risk-adjustment
-
Provide participants the opportunity to conduct hands on severity risk-adjustment simulation
Background/context:
The U.S. Congress is implementing the American Recovery and Reinvestment
Act of 2009 as part of President Obama's economic recovery plan. A
key strategy within this plan relates to health care reform.
Comparative effectiveness is a key component in this health care reform
strategy. Comparative effectiveness relates to the ability to
compare outcomes of medical care and identify best practices that add
the greatest value. However, a crucial step for successful
comparative effectiveness analysis requires the ability to compare
patient groups that have differences in underlying severity or
risk. This comparison can be accomplished by the scientific
process of severity risk-adjustment. In order to leverage the
available funding for comparative effectiveness for children, pediatric
institutions will need to understand and implement state-of-the-art
severity risk-adjustment methods.
Project description:
The Virtual Pediatric Intensive Care Unit (VPS) system, a joint endeavor
between the National Association of Children’s Hospitals and
Related Institutions (NACHRI), Childrens Hospital Los Angeles (CHLA),
and the National Outcomes Center (NOC), has successfully developed and
implemented severity risk-adjustment methods for comparative
effectiveness of 81 pediatric institutions in the U.S. comprising of
over 250,000 critically ill children. Risk adjustment in the VPS
system utilizes severity of illness systems such as the Pediatric Risk
of Mortality (PRISM) and Pediatric Index of Mortality (PIM). These
systems include clinical and laboratory factors that have been validated
in children for risk-adjustment. This process involves a formal
statistical approach (using unconditional logistic regression) to
severity risk-adjust outcomes in children, i.e. the quantitative
severity risk for each child is computed and severity risk groups are
created. These severity risk groups can then be used to
statistically adjust for differences in case-mix when conducting
comparative effectiveness. This process of severity
risk-adjustment has been well-described and established in the outcomes
research field. However, its recent application for comparative
effectiveness to perform quality improvement is a relatively new and
bold paradigm shift. Embracing this paradigm shift will allow the
meaningful implementation of comparative effectiveness analyses to
determine the true value of health care strategies which will facilitate
the successful health care reform for children.
Results/outcomes:
The VPS experience for performing severity risk-adjustment to conduct
comparative effectiveness analysis is used as a case study.
Pediatric institutions participating in the VPS collect information
related to PRISM or PIM systems for severity risk-adjustment,
facilitating meaningful comparisons of outcomes across pediatric
intensive care units (PICUs). Results use standard levels of
statistical significance (p<0.05). For example, individual PICU
mortality rates can be compared to the national experience meaningfully
after performing severity risk-adjustment which adjusts for differences
in patient-mix. Similarly, length of stay (LOS) in PICUs can be
compared meaningfully by first performing severity risk-adjustment which
adjusts for LOS differences due to patient-mix. This process has
led to a high level of buy-in by clinicians and administrators related
to comparative analysis results for their PICUs.
Lessons learned:
Severity risk-adjustment is a critical step for conducting comparative
effectiveness in a reliable manner for clinicians and
administrators. The case study using the VPS illustrates that
severity risk-adjustment can be successfully performed across multiple
institutions using robust statistical principles. The experience
from the VPS in the PICU setting has direct implications for other
clinical settings for performing comparative effectiveness analysis
related to the care for children. Understanding of the severity
risk-adjustment process provides the necessary foundation for conducting
comparative effectiveness analyses to determine the value (relation
between quality and cost) of health care interventions which will be a
major area of focus under the national economic stimulus package
focusing on health care reform.
Future of project/next steps:
Severity risk-adjustment methods utilized in the VPS continue to be
refined further. Diagnosis based severity risk-adjustment models
are being developed. This provides greater precision and accuracy
in performing severity risk-adjustment when performing comparative
effectiveness as it relates to specific categories or disease
entities. The large VPS dataset also provides a unique opportunity
to conduct more frequent and rapid recalibration of severity
risk-adjustment systems for even greater calibration of this
process. This is important because it leads to the ability to
perform highly accurate severity risk-adjustment that can account for
recent changes in clinical practice. Current efforts are also
being focused to expand the experience for severity risk-adjustment from
the PICU to other areas in pediatric institutions such as cardiac
critical care and critically ill neonates.