Economic Burden of Ventilator-Associated Pneumonia Based on Total Resource Utilization

Abstract

Objectives. To characterize the current economic burden of ventilator-associated pneumonia (VAP) and to determine which services increase the cost of VAP in North American hospitals. Design and Setting. We performed a retrospective, matched cohort analysis of mechanically ventilated patients enrolled in the North American Silver-Coated Endotracheal Tube (NASCENT) study, a prospective, randomized study conducted from 2002 to 2006 in 54 medical centers, including 45 teaching institutions (83.3%). Methods. Case patients with microbiologically confirmed VAP (n = 30) were identified from 542 study participants with claims data and were matched by use of a primary diagnostic code, and subsequently by the Acute Physiology and Chronic Health Evaluation II score, to control patients without VAP (n = 90). Costs were estimated by applying hospital-specific cost-to-charge ratios based on all-payer inpatient costs associated with VAP diagnosis-related groups. Results. Median total charges per patient were $198,200 for case patients and $96,540 for matched control patients (P <.001); corresponding median hospital costs were $76,730 for case patients and $41,250 for control patients (P = .001). After adjusting for diagnosis-related group payments, median losses to hospitals were $32,140 for case patients and $19,360 for control patients (P = .151). The median duration of intubation was longer for case patients than for control patients (10.1 days vs 4.7 days; P < .001), as were the median duration of intensive care unit stay (18.5 days vs 8.0 days; P < .001) and the median duration of hospitalization (26.5 days vs 14.0 days; P < .001). Examples of services likely to be directly related to VAP and having higher median costs for case patients were hospital care (P < .05) and respiratory therapy (P < .05). Conclusions. VAP was associated with increased hospital costs, longer duration of hospital stay, and a higher number of hospital services being affected, which underscores the need for bundled measures to prevent VAP. Trial Registration. NASCENT study ClinicalTrials.gov Identifier: NCT00148642.