We initiated a quality improvement work group to focus on Ventilator Associated Pneumonia (VAP) in our intensive care unit (ICU). Our project incorporated multiples strategies through the quality improvement process to decrease the incidence of VAP.
Multiple strategies included: (1) head of bed elevation, (2) daily wake up from sedation, (3) hand hygiene, (4) oral care, (5) enteral feeding protocol to prevent aspiration, (6) insulin protocol. Collaborative meetings of stake holders were held on a monthly basis. Control charts were distributed via e-mail along with a monthly lecture series and data displays in the ICU. The CDC definition of VAP was used; data was benchmarked against the National Nosocomial Infection Surveillance (NNIS) mean. Incidence of VAP was collected per 1,000 ventilator days. A single infection control officer was responsible for data tracking, display and dissemination.
Baseline data included 919 ventilator days (August, September 2002) with a mean VAP of 10.9%. After protocol (February 2004), VAP decreased to 4 VAP/487 ventilator days, 8.2%; and (March 2004) 3 VAP/432 ventilator days, 6.9%. Device utilization for mechanical ventilation (February and March) was 919 ventilator days/1,219 patient days (75%).
Multiple quality improvement strategies can be incorporated under a single project to evaluate impact on ventilator associated pneumonia. A coordinated approach with an identified team leader and frequent feedback to interested parties using multiple media is effective in reducing the incidence of ventilator associated pneumonia.
Decreasing the incidence of ventilator associated pneumonia through a quality improvement paradigm will save both lives and money. A further benefit should be more bed availability within the ICU. Many quality improvement strategies are able to decrease costs and improve quality while simultaneously providing more resources. This constellation of benefits is rarely achieved with new medical technology.
Figure Legend: VAP decreased from 11.6% (August 2002) to 6.9% (March 2004) with the protocol. NNIS pooled mean equals 5.8%.
W. McGee, None.