Analysis of ROX Index Pre- and Post-Interventions: A Pilot Perspective

Abstract

Background: The ROX index is a validated tool to assist clinicians in the decision of whether to escalate care from utilizing a high-flow nasal cannula (HFNC) to mechanical ventilation in patients diagnosed with acute respiratory failure due to pneumonia. In 2020, an in-house quality improvement pilot analysis found statistical significance when comparing expired subjects to survived subjects with the intervention of the ROX index suggesting potential for improvement in patient care with the use of the ROX index. This pilot serves to compare the 2020 pilot analysis to the post-intervention subject population.

Methods: The UC Davis Health Respiratory Care Department RTs were educated on the use of the ROX index in April-November 2022 with a high-fidelity hands-on simulation skills lab attended by 116 UC Davis Health RTs. The ROX index was included in the adult HFNC hospital policy published in December 2022 and added to the Epic EMR flowsheet to auto-populate since July 2022. This is a single-center retrospective observational analysis of consecutive subjects from December 2022-April 2023 if ≥18 y old, diagnosed with acute respiratory failure due to pneumonia, ARDS, and/or COVID-19, utilized HFNC, and subsequently intubated. A subject met inclusion criteria at the second failed ROX index (SFR) score using Epic EMR automated ROX index of <4.88. Excel t-Test: Two-Sample Assuming Unequal Variances used for statistical analyses.

Results: Pre-intervention, authors included n = 76 for variable comparison. Statistical significance was found in (median, IQR) HFNC start to SFR in hours, P = .029 (E) 11.3 (38.14-2.25) vs (S) 2.58 (12-0.5) and SFR to mechanical ventilation start in hours, P = .01 (E) 33.68 (87.22-8.42) vs (S) 9.95 (25.42-3.07). Post-intervention, authors included n = 7 for variable comparison. No statistical significance was found in (median, IQR) HFNC start to SFR in hours, P = .71 (E) 1.88 (0.93-3.23) vs (S) 0.4 (0.29-5.63) or SFR to mechanical ventilation start in hours, P = .15 (E) 12.03 (4.42-25.03) vs (S) 0.68 (0.47-1.09).

Conclusions: The change from statistical significance pre-intervention to no statistical significance post-intervention may suggest an improvement in patient care for this population as it relates to the use of the ROX index. Continued monitoring of the post-intervention subject population is needed. Direct statistical analyses of pre- and post-intervention subject population is required as time elapses.