Appropriate Use Criteria for Stress Echocardiography: Impact of Updated Criteria on Appropriateness Ratings, Correlation With Pre-Authorization Guidelines, and Effect of Temporal Trends and an Educational Initiative on Utilization
What is the impact of the 2011 appropriate use criteria (AUC) for stress echocardiography (SE) on the distribution of appropriate and inappropriate SE, and furthermore, how well do AUC correlate with the determination of radiology benefits managers (RBM) for precertification of SE?
Medical record review was undertaken to determine the appropriateness of three separate cohorts of patients undergoing SE at a single center. Each SE was then assigned an AUC number as per the 2008 AUC and/or 2011 AUC. Cohort 1 included 209 patients studied in 2008, cohort 2 included 209 patients studied in 2011, and cohort 3 included an additional 111 patients studied later in 2011 after an educational initiative designed to alter ordering patterns.
For the 2008 cohort, 104 studies were considered appropriate, 62 inappropriate, 23 uncertain, and 20 unclassified. These same patients were subject to analysis by the 2011 AUC. Only two patients remained unclassified, 47.8% were appropriate, 30.1% inappropriate, and 21.1% uncertain. For the 2011 cohort analyzed by the 2011 AUC, 39.2% were appropriate, 30.6% inappropriate, 25.8% uncertain, and 4.3% unclassified. Following an educational intervention consisting of grand rounds and other reminders, the distribution of AUC for a final cohort of 111 patients was 43.2% appropriate, 32.4% inappropriate, 20.7% uncertain, and 3.6% unclassified. The most common inappropriate AUC in the entire study group (N = 529) were for AUC 114: evaluation of ischemic equivalent with low pretest probability and an interpretable electrocardiogram. Other common inappropriate use indications included evaluation of asymptomatic patients with low coronary artery disease risk, evaluation of patients with known coronary artery disease <2 years following prior study without symptoms, and perioperative evaluation with intermediate risk, with no clinical risk factors. For the 2011 cohort, which was compared to two different RBM precertification algorithms, 127 SE were in an appropriate or uncertain category, and 62 in an inappropriate category. RBM #1 would have authorized 110 of 127 appropriate or uncertain and declined 17 SE. RBM #1 would have authorized 14 of 62 inappropriate AUC and declined 48. RBM #2 would have authorized only 72 of 127 appropriate or uncertain SE and declined 55, and approved 11 inappropriate AUC and declined 51.
The 2011 AUC improve on the 2008 version with respect to more complete classification of SE indications. There is a stable persistent pattern of inappropriate utilization over time, which remains problematic. The degree of agreement between AUC and determination by RBMs is highly variable.
The 2011 AUC represent an improvement over the 2008 AUC with respect to the number of unclassified studies. For both time frames, there remains a substantial (30.8%) prevalence of inappropriate SE largely due to testing of low-risk individuals with equivocal symptoms or routine follow-up <2 years out from an event in asymptomatic patients. Of note, a modest educational intervention including cardiology grand rounds and other reminders had no impact on the incidence of inappropriate SE. A worrisome finding from this study was the substantial variability of two different RBMs with respect to the degree that their algorithms for determining reimbursement were in line with the AUC. While not specifically developed for determining reimbursement, it was obvious from the beginning that the AUC documents would be potentially utilized for that purpose. The substantial discrepancy between at least one RBM and the AUC is somewhat disturbing with respect to preauthorization trends.
Keywords: Coronary Artery Disease, Echocardiography, Stress, Cardiology, Risk Factors, Medical Records
< Back to Listings