Objectives: To quantify the association of ambient air pollution (particulate matter, PM 2.5) exposure with medically attended acute respiratory illness among infants with bronchopulmonary dysplasia (BPD). Study Design: Single center, retrospective cohort study of preterm infants with BPD in Metropolitan Philadelphia. Multivariable logistic regression quantified associations of annual mean PM 2.5 exposure (per μg/m 3) at the census block group level with medically attended acute respiratory illness, defined as emergency department (ED) visits or hospital readmissions within a year after first hospital discharge adjusting for age at neonatal intensive care unit (NICU) discharge, year, sex, race, insurance, BPD severity, and census tract deprivation. As a secondary analysis, we examined whether BPD severity modified the associations. Results: Of the 378 infants included in the analysis, 189 were non-Hispanic Black and 235 were publicly insured. Census block PM 2.5 level was not significantly associated with medically attended acute respiratory illnesses, ED visits, or hospital readmissions in the full study cohort. We observed significant effect modification by BPD grade; each 1 µg/m 3 higher annual PM 2.5 exposure was medically attended acute respiratory illness (aOR 1.65, 95% CI: 1.06-2.63) among infants with grade 1 BPD but not among infants with grade 3 BPD (aOR 0.83, 95% CI: 0.47-1.48) (interaction p=0.024). Conclusions: Cumulative PM 2.5 exposure in the year after NICU discharge was not significantly associated with medically attended acute respiratory illness among infants with BPD. However, infants with grade 1 BPD had significantly higher odds with higher exposures. If replicated, these findings could inform anticipatory guidance for families of these infants to avoid outdoor activities during high pollution days after NICU discharge.

Objective: Home pulse oximetry is often prescribed to children with chronic disease upon hospital discharge. Children monitored at home may generate >20 alarms every 8 hours, contributing to premature discontinuation of monitoring. We aimed to improve the home oximetry ordering process using clinical decision support (CDS), supporting more liberal oxygen saturation (SpO 2) alarm limits. Methods: Within a large single-center improvement project to increase informativeness of alarms in the hospital and in patients’ homes, we compared home care oximetry orders of discharged children pre-post CDS implementation. Order parameters included low SpO 2 limit, specification of intensity of use, an intervention plan, pulse oximetry probe prescription, and order completeness. We extracted order details 6 months pre-CDS and 6 months post-CDS with a one-month washout period. The CDS intervention used a letter template to include all required home oximeter order elements and provide more liberal age-specific default alarm limits. Results: There were 100 orders in the pre-CDS epoch (7/1/2021-12/31/2021) and 112 orders in the post-CDS epoch (2/1/2022-7/31/2022). The median low SpO 2 alarm limit post-CDS implementation (87%, IQR 87%-90%) was significantly lower than pre-CDS (90%, IQR 90%-90%, p=<0.001). In the post-CDS epoch significantly more orders included an intervention plan (80.4% versus 31%, p<0.001), prescribed pulse oximeter probes (85.7% versus 52.0%, p<0.001), and were complete (68.8% versus 13.0%, p<0.001). Conclusions: CDS implementation resulted in a significant decrease in median low SpO 2 limit and a significant increase in home oximetry order completeness. These changes may decrease home oximetry alarm burden and improve caregiver experiences with home oximetry.