Economic Impact of Ventilator-Associated Pneumonia in a Large Matched Cohort

OBJECTIVE. To evaluate the economic impact of ventilator-associated pneumonia (VAP) on length of stay and hospital costs. DESIGN. Retrospective matched cohort study. SETTING. Premier database of hospitals in the United States. PATIENTS. Eligible patients were admitted to intensive care units (ICUs), received mechanical ventilation for >= 2 calendar-days, and were discharged between October 1, 2008, and December 31, 2009. METHODS. VAP was defined by International Classification of Diseases, Ninth Revision (ICD-9), code 997.31 and ventilation charges for >= 2 calendar-days. We matched patients with VAP to patients without VAP by propensity score on the basis of demographics, administrative data, and severity of illness. Cost was based on provider perspective and procedural cost accounting methods. RESULTS. Of 88,689 eligible patients, 2,238 (2.5%) had VAP; the incidence rate was 1.27 per 1,000 ventilation-days. In the matched cohort, patients with VAP (n = 2,144) had longer mean durations of mechanical ventilation (21.8 vs 10.3 days), ICU stay (20.5 vs 11.6 days), and hospitalization (32.6 vs 19.5 days; all P < .0001) than patients without VAP (n = 2,144). Mean hospitalization costs were $99,598 for patients with VAP and $59,770 for patients without VAP (P < .0001), resulting in an absolute difference of $39,828. Patients with VAP had a lower in-hospital mortality rate than patients without VAP (482/2,144 [22.5%] vs 630/2,144 [29.4%]; P < .0001). CONCLUSIONS. Our findings suggest that VAP continues to occur as defined by the new specific ICD-9 code and is associated with a statistically significant resource utilization burden, which underscores the need for cost-effective interventions to minimize the occurrence of this complication.