Characteristics of Long-Term Care Hospital Ventilator-Associated Events, National Healthcare Safety Network, 2016–2018

Abstract

Background: Ventilator-associated event (VAE) reporting to the CDC NHSN began in 2013. VAE reporting from long-term care hospitals (LTCHs) to the NHSN was required from January 2016 through September 2018 as part of the CMS LTCH Quality Reporting Program (QRP). We describe the incidence and characteristics of LTCH VAEs during the required reporting period. Methods: We analyzed VAE data reported to the NHSN from January 2016 through December 2018, from the LTCH ward and critical care locations participating in surveillance according to the NHSN protocol. We have described characteristics of VAE, and we determined the distribution of VAE types: ventilator-associated conditions (VAC), infection-related ventilator-associated complications (IVAC), and possible ventilator-associated pneumonia (PVAP). Furthermore, we calculated pooled mean VAE rates per 1,000 ventilator days, and we determined the rate distributions for locations with ≥20 units reporting >50 ventilator days per year. Results: Overall, 493 LTCHs reported 22,359 location months of VAE data from ward and critical care locations. In total, 5,290 VAEs were reported, of which 3,871 (73%) were VAC, 961 (18%) were IVAC, and 458 (9%) were PVAP. Also, 42% (2,241) of VAEs occurred in female patients, and 1,305 (25%) occurred in patients who died during their hospitalization. The median time from LTCH admission to VAE onset was 18 days (IQR, 9–37), and from initiation of mechanical ventilation to VAE onset was 22 days (IQR, 10–43). Pathogens were identified from 454 PVAPs, with Pseudomonas aeruginosa (43% of PVAPs) and Staphylococcus aureus (26%) being the most common organisms. Annual pooled mean incidence rates in critical care locations ranged from 2.11 to 2.62 VAEs per 1,000 ventilator days, whereas rates in ward locations ranged from 1.36 to 1.67 VAEs per 1,000 ventilator days (Table 1). Conclusions: During a period of required reporting, pooled mean LTCH VAE rates remained low. Most VAEs in LTCHs were reported as VACs. Additional work is needed to understand the clinical events associated with LTCH VAE, including whether most VAEs truly represent non–infection-related events or reflect limited evaluation to identify infection-related complications. This distinction might influence the identification of appropriate interventions to reduce LTCH VAE rates.

Funding: None

Disclosures: None