Assessment of ventilator-associated events using the geometric distribution

Assaly, Ragheb; Barnett, William; Safi, Fadi; Khuder, Sadik A.; Macko, John

doi:10.1016/j.ajic.2016.12.004

Cited by 2 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Rather than relying on traditional measures (events per 1,000 ventilator days) to report changes in the VAE rate, we chose an alternative method to analyze nosocomial infections. More precisely, we used the geometric distribution to calculate a daily probability of a VAE, which is used to determine the likelihood of a rare event occurring over time based on the days between events ( 3 ). The hallmark of the geometric distribution is the ability to model a large number of failures (non-VAEs) before a success (VAE) where the probability of an event is the same regardless of how much time has passed.…”

Section: Methodsmentioning

confidence: 99%

Does a Starting Positive End-Expiratory Pressure of 8 cmH2O Decrease the Probability of a Ventilator-Associated Event?

et al. 2021

Self Cite

View full text Add to dashboard Cite

Introduction: Ventilator-associated events (VAEs) are objective measures as defined by the Centers for Disease Control and Prevention (CDC). To reduce VAEs, some hospitals have started patients on higher baseline positive end-expiratory pressure (PEEP) to avoid triggering VAE criteria due to respiratory fluctuations.Methods: At our institution, VAEs were gathered from January 2014 through December 2019. Using the CDC-defined classifications, VAEs were split into two groups to separate patients with hypoxemia only (VAC) and those with hypoxemia and evidence of inflammation or infection (IVAC-plus). We used the geometric distribution to calculate the daily event probability before and after the protocol implementation. A probability threshold was used to determine if the days between events was exceeded during the post-protocol period.Results: A total of 306 VAEs were collected over the study period. Of those, 155 were VACs and 107 were IVAC-plus events during the pre-protocol period. After implementing the protocol, 24 VACs and 20 IVAC-plus events were reported. There was a non-significant decrease in daily event probabilities in both the VAC and IVAC-plus groups (0.083 vs. 0.068 and 0.057 vs. 0.039, respectively).Conclusion: We concluded a starting PEEP of 8 cmH2O is unlikely to be an effective intervention at reducing the probability of a VAE. Until specific guidelines by the CDC are established, hospitals should consider alternative methods to reduce VAEs.

show abstract

Section: Methodsmentioning

confidence: 99%