Introduction: The ?door-to-mask? time is a UK-wide quality metric for NIV services. We examined the utility of this metric in predicting in-hospital death, as the COVID pandemic resulted in a significant prolongation of this metric due to stringent infection control measures.
Methods: Data was extracted for all acute NIV applications for hypercapnic respiratory failure pre and post 2020 (01 Apr-31 Dec, 2019 and 2021, respectively). Categorical and continuous variables were compared using Chi-squared and Mann-Whitney U tests, respectively. Logistic regression was used to predict in-hospital death adjusted odds ratios (OR) across both cohorts combined.
Results: Total patients receiving NIV decreased by 26.2%, 2019 to 2021 (83 vs 45 patients). Age (69 vs 69, p=0.93) was not significantly different. NIV indication was not significantly different, the majority, 87 (66.9%) vs 63 (65.6%), received NIV for COPD, 2019 vs 2021, respectively. Door-to-mask-times increased, 327 to 651 mins (p<0.001), 2019 vs 2021. However, in-hospital death was not significantly different, 27 (20.8%) vs 28 (29.2%) (p=0.15). Age and pH did affect in-hospital mortality, OR ratio (95% CI), 1.037 (1.01, 1.07) (p=0.02) and 0.009 (0.00, 0.70) (p=0.03), respectively. However, door to mask time did not significantly affect OR of in-hospital mortality, 1.004 (1.00, 1.01), (p=0.07).
Conclusions: Within our cohort, 'door-to-mask' time was not a strong predictor of mortality, hence we need to explore new metrics such as ?decision-to-mask time? and/or a surrogate of the physical quality of NIV application and synchronisation (e.g. Maximum pressures achieved) for our future quality improvement journey.