New-onset atrial fibrillation can be falsely associated with increased length of stay in ICU due to immortal time bias

Lu, Y.; Chen, Tenggao

doi:10.1186/s13054-020-2757-2

Cited by 1 publication

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several previous studies have suggested an independent association between NOAF during critical illness and hospital mortality using logistic regression analysis [14,[16][17][18]. However, logistic regression may underestimate the association between NOAF and ICU mortality due to immortal time bias [37,39,50]. We treated NOAF as a time-varying covariate in our primary analysis to mitigate this issue and reveal a stronger association than has been demonstrated in studies using logistic regression.…”

Section: Discussionmentioning

confidence: 99%

Hospital outcomes associated with new-onset atrial fibrillation during ICU admission: A multicentre competing risks analysis

Bedford

Gerry

Hatch

et al. 2020

Journal of Critical Care

View full text Add to dashboard Cite

New onset atrial fibrillation (NOAF) in critically ill patients has been associated with increased shortterm mortality. Analyses that do not take into account the time-varying nature of NOAF can underestimate its association with hospital outcomes.We investigated the prognostic association of NOAF with hospital outcomes using competing risks methods. Materials and MethodsWe undertook a retrospective cohort study in three general adult intensive care units (ICUs) in the UK from June 2008 to December 2015. We excluded patients with known prior atrial fibrillation or an arrhythmia within four hours of ICU admission.To account for the effect of NOAF on the rate of death per unit time and the rate of discharge alive per unit time we calculated subdistribution hazard ratios (SDHRs). ResultsOf 7541 patients that fulfilled our inclusion criteria, 831 (11.0%) developed NOAF during their ICU admission. NOAF was associated with an increased duration of hospital stay (CSHR 0.68 (95% CI 0.63-0.73)) and an increased rate of in-hospital death per unit time ). This resulted in a strong prognostic association with dying in hospital (adjusted SDHR 2.04 (1.79 -2.32)). NOAF lasting over 30 minutes was associated with increased hospital mortality. ConclusionsUsing robust methods we demonstrate a stronger prognostic association between NOAF and hospital outcomes than previously reported.

show abstract