Health Policy and Health Services Research
Keith Kocher, MD, MPH
University of Michigan
Disclosure Relationship(s): Agency for Healthcare Research and Quality (List your financial relationship role.): Grants/Research Support Recipient
Blue Cross Blue Shield of Michigan/Blue Care Network (List your financial relationship role.): Grants/Research Support Recipient
Background: As the US shifts to value-based care, there are opportunities to safely reduce over-testing in emergency departments (EDs). The Michigan Emergency Department Improvement Collaborative (MEDIC), a statewide quality network connecting a diverse set of unaffiliated EDs, provides a unique opportunity to benchmark appropriate use of imaging and estimate the burden of low-value studies
Methods: MEDIC is a physician-led partnership supported by a major third party payer with 15 member hospitals (25,000-125,000 ED visits annually) contributing data to a central registry. Data were obtained from electronic health records and chart review by trained abstractors. Baseline performance data were collected for 4 quality measures: appropriateness of computed tomography (CT) for minor head injury for (1) adults (Canadian Rule) and (2) children (PECARN Rule); (3) chest x-ray (CXR) utilization for children with asthma, bronchiolitis, or croup; and (4) diagnostic yield of CTs for suspected pulmonary embolism (PE). Estimates of excess imaging studies in 2017 were calculated based on Achievable Benchmark of Care (ABCTM), a method for determining quality improvement targets across a population of performance. Direct standardization was then used to project estimates for 10,000 annual ED visit increments with 20% pediatrics. Spending estimates were calculated for excess studies based on fair market price quotes from the Healthcare Bluebook using the 15 site locations in Michigan and reported as ranges.
Results: From 6/1/16 – 10/31/17, the MEDIC registry contained data on 1,124,227 ED visits, representing ~22% of all ED visits in Michigan, including 261,157 (23.2%) for children. Visits reflected typical demographics and acuity found in the general ED population. Across the collaborative, we found substantial variation, suggesting the potential to avoid 1,519 head CTs; 3,308 CXRs; and 4,254 PE CTs in 2017 if the ABCTM targets were achieved. Estimated spending on these excess studies ranged from $3.59 – $5.02 million. This represents an excess of 14 head CTs, 122 CXRs, 24 PE CTs, and $34,760 – $48,189 for each additional 10,000 ED visit volume increment per year
Conclusion: There is substantial opportunity for avoiding low value imaging studies in EDs. If ABCTM targets can be safely achieved, the result would be significant health care savings.