Effect Modifiers of the Association of High-Flow Nasal Cannula and Bronchiolitis Length of Stay.

BACKGROUND AND OBJECTIVES: High-flow nasal cannula (HFNC) therapy for hospitalized children with bronchiolitis is associated with a longer length of stay (LOS) when used outside of the ICU. We sought to explore the association between HFNC and LOS to identify if demographic and clinical factors may modify the effect of HFNC usage on LOS. METHODS: In this multicenter retrospective cohort study, we used a combination of hospital records and the Pediatric Health Information System. We included encounters from September 1, 2018 to March 31, 2020 for patients <2 years old diagnosed with bronchiolitis. Multivariable Poisson regression was performed for the association of LOS with measured covariates, including fixed main effects and interaction terms between HFNC and other factors. RESULTS: Of 8060 included patients, 2179 (27.0%) received HFNC during admission. Age group, weight, complex chronic condition, initial tachypnea, initial desaturation, and ICU services were significantly associated with LOS. The effect of HFNC on LOS differed among hospitals (P < .001), with the estimated increase in LOS ranging from 32% to 139%. The effect of HFNC on LOS was modified by age group, initial desaturation, and ICU services, with 1- to 6-month-old infants, patients without initial desaturation, and patients without ICU services having the highest association between HFNC and LOS, respectively. CONCLUSIONS: We identified multiple potential effect modifiers for the relationship between HFNC and LOS. The authors of future prospective studies should investigate the effect of HFNC usage on LOS in non-ICU patients without documented desaturation.

Reducing the Frequency of Pulse Oximetry Alarms at a Children's Hospital.

BACKGROUND AND OBJECTIVES: Alarm fatigue is exacerbated by frequent, nonactionable physiologic monitor alarms. Overutilization of pulse oximetry (SpO2) compounds this alarm burden. Narrow default alarm limits and overutilization of continuous (CSpO2) rather than intermittent monitoring contribute to nonactionable alarms. There were 1.12 million SpO2 alarms on included units during the baseline period, of which 41.0% were for SpO2 ≥ 88%. We aimed to decrease SpO2 alarms per patient day by 20% within 12 months. METHODS: This quality improvement study included patients admitted January 2019 to June 2022. Intensive care and cardiology units were excluded. Interventions included (1) changing default alarm SpO2 limits on monitors from <90% to <88%, (2) changing SpO2 order default from continuous to intermittent, and (3) adding indication requirements for CSpO2. Outcome measures were total SpO2 alarms and alarms for SpO2 ≥ 88% per patient day. Balancing measures were high acuity transfers and code blues without CSpO2 ordered. Control charts were used for each. RESULTS: Our study included 120 408 patient days with 2.98 million SpO2 alarms. Total SpO2 alarms and alarms for SpO2 ≥ 88% per patient day decreased by 5.48 (30.57 to 25.09; 17.9%) and 4.48 (12.50 to 8.02; 35.8%), respectively. Special cause improvement was associated with changing default monitor alarm parameters. Balancing measures remained stable. CONCLUSIONS: SpO2 monitors alarm frequently at our children's hospital. Widening default alarm limits was associated with decreased SpO2 alarms, particularly nonactionable alarms (≥88%). This high-reliability intervention may be applied, when appropriate, to other monitor alarm parameters to further mitigate alarm burden.

Impact of High Flow Nasal Cannula on Resource Utilization in Bronchiolitis.

OBJECTIVES: High flow nasal cannula (HFNC) is increasingly used for children hospitalized with bronchiolitis. We aimed to validate identification of HFNC use in a national database, then compare resource utilization among children treated with and without HFNC. METHODS: In this cross-sectional, multicenter study, we obtained clinical and resource utilization data from the Pediatric Health Information System (PHIS) database for healthy children aged 1 to 24 months admitted for bronchiolitis. We assessed HFNC use based on a combination of billing codes and reviewed charts at 2 hospitals to determine their accuracy. We compared costs, length of stay, and readmissions between the HFNC and no HFNC groups at hospitals utilizing the HFNC codes. RESULTS: The PHIS codes demonstrated 90.4% sensitivity and 99.3% specificity to detect HFNC use as verified by chart review at 2 hospitals. However, only 24 of 51 PHIS hospitals used these codes for ≥1% of patients with bronchiolitis. Within those hospitals, children treated with HFNC had greater total costs ($7054 vs $4544; P < .001), greater daily costs ($2922 vs $2613; P < .001), and longer length of stay (57.6 vs 41.6 hours; P < .001). Those treated with HFNC were less likely to be readmitted at 3 and 7 days (P < .001), but by 14 days, readmissions were similar in the 2 groups. CONCLUSIONS: Billing codes for HFNC are inconsistently applied across PHIS hospitals; however, among those hospitals that routinely apply these codes, HFNC was associated with more intense resource utilization. Standardization of billing practices for HFNC would allow future study to more broadly describe the value of HFNC.