880Incidence rates of ventilator-associated events and ventilator-associated pneumonia in the National Healthcare Safety Network, 2012-2013

Background. Surveillance for ventilator-associated events (VAEs) is based on detecting conditions associated with increased oxygen or positive end-expiratory pressure (PEEP) settings. In 2013 the National Healthcare Safety Network (NHSN) replaced traditional ventilator-associated pneumonia (tVAP) with VAE surveillance in adult inpatient units. In August 2013 a minor PEEP criterion change was made. We compared VAE rates before and after the change and tVAP to VAE rates in 9 intensive care unit (ICU) types. Methods. We analyzed data from ICUs with >50 ventilator days that followedNHSN tVAP (2012) and VAE (2013) protocols. We calculated pooled mean (PM) incidence rates/1,000 ventilator days before (January-July, 2013) and after (August-December, 2013) the change, and annual PM (APM) VAE (2013) and tVAP (2012) rates, and compared them using rate ratios (RRs) and 95% confidence intervals (CIs) (OpenEpi 3.01). Results. Of 2,269 ICUs submitting VAE data in 2013, 1766 (78%) reported before and after the change. VAE rates increased after the change in trauma ICUs only (table). Of 1487 ICUs that participated in tVAP and VAE surveillance, VAE rates were highest in trauma and neurology ICUs. VAE rates were significantly higher than tVAP rates in all except burn ICUs (table). VAEs/1,000 vent days, Before and after definition change Events/1,000 vent days, VAE and tVAP