Metrics of quality care in veterans: Correlation between primary-care performance measures and inappropriate myocardial perfusion imaging

Background Approximately 10% to 20% of myocardial perfusion imaging (MPI) tests are inappropriate based on professional-society recommendations. The correlation between inappropriate MPI and quality care metrics is not known. Hypothesis Inappropriate MPI will be associated with low achievement of quality care metrics. Methods We conducted a retrospective cross-sectional investigation at a single Veterans Affairs medical center. Myocardial perfusion imaging tests ordered by primary-care clinicians between December 2010 and July 2011 were assessed for appropriateness (by 2009 criteria). Using documentation of the clinical encounter where MPI was ordered, we determined how often quality care metrics were achieved. Results Among 516 MPI patients, 52 (10.1%) were inappropriate and 464 (89.9%) were not inappropriate (either appropriate or uncertain). Hypertension (82.2%), diabetes mellitus (41.3%), and coronary artery disease (41.1%) were common. Glycated hemoglobin levels were lower in the inappropriate MPI cohort (6.6% vs 7.5%; P = 0.04). No difference was observed in the proportion with goal hemoglobin (62.5% vs 46.3% for appropriate/uncertain; P = 0.258). Systolic blood pressure was not different (132 mm Hg vs 135 mm Hg; P = 0.34). Achievement of several other categorical quality metrics was low in both cohorts and no differences were observed. More than 90% of clinicians documented a plan to achieve most metrics. Conclusions Inappropriate MPI is not associated with performance on metrics of quality care. If an association exists, it may be between inappropriate MPI and overly aggressive care. Most clinicians document a plan of care to address failure of quality metrics, suggesting awareness of the problem.