RESUMO
BACKGROUND: Heart failure and pneumonia are among the most measured and expensive conditions to treat in the United States across all payer types and are top of mind for value-driven hospital organizations and payers seeking to not only improve the quality of care for patients but also reduce unnecessary spending. Care standardization potentially leads to better patient outcomes and reduced excess costs but is a difficult objective to achieve. METHODS: A pre-post analysis of clinical practice, patient outcomes, and cost was designed to determine if serial measurement and feedback using simulated patients improves patient care quality and reduces costs for two common conditions cared for by hospitalists: pneumonia and heart failure. Care decisions measured using the simulations were compared to patient-level data collected by the system. RESULTS: Intrafacility care variation seen among Novant Health's 11 facilities employing hospitalists decreased from 14.9% to 8.5%, and overall quality-of-care scores by individual providers improved by 14.6 percentage points from study start to end. Overall, care changes (for example, troponin usage, palliative care consults, beta blocker orders) documented in the simulated patients matched the available patient-level data. Care standardization around evidence-based practices, as measured by the simulations, was associated with appreciable decreases in patient length of stay and readmissions, amounting to nearly $1.1 million in savings for Novant Health. CONCLUSION: An approach using simulated patients that includes serial measurement and feedback may help significantly reduce practice variation between different facilities in a health system and reduce costs substantially without negatively affecting outcomes.