337 - Impact of a Clinical Pathway for the Emergency Management of Bronchiolitis by Preferred Language

Poster Number: 337
Publication Number: 337.204

Robert H. Rosen, Boston Combined Residency Program, Boston, MA, United States; Michael C. Monuteaux, Boston Children's Hospital, Boston, MA, United States; kenneth michelson, Boston Children's Hospital, Boston, MA, United States; Anne Stack, Boston Children's Hospital, Boston, MA, United States; Andrew Fine, Boston Children's Hospital, Boston, MA, United States

Background: Clinical pathways designed to standardize care have proliferated in recent years, but their impact on healthcare equity is poorly understood.

Objective: We examined the impact of a clinical pathway for the management of bronchiolitis on the care of children of families with English versus non-English preferred language presenting to the emergency department (ED) of a quaternary care children’s hospital.

Design/Methods: We included all encounters for patients aged 1-12 months with an ED discharge diagnosis of acute bronchiolitis by ICD-9 or -10 code from 1/1/2010-10/31/2020. The bronchiolitis pathway was introduced on 11/4/2011, so the pre-pathway period was 1/1/2010-10/31/2011 and the post-pathway period was 1/1/2012-10/31/2020 (after a two-month washout). We performed interrupted time series analyses with a three-way interaction term (time x period x language) to assess the differential impact of the clinical pathway by language (i.e., children of families with English versus non-English preferred language) on the following outcomes: chest x-ray (CXR) acquisition, albuterol administration, 7-day ED return visit, 72-hour ED return to admission, antibiotic treatment, and steroid treatment. Analyses were adjusted for complex care condition at index visit.

Results: There were 1485 encounters in the pre-period (77% English, 14% non-English, 8% missing) and 7840 encounters in the post-period (79% English, 15% non-English, 6% missing). The pre-post effect on albuterol administration significantly differed by language; among non-English preferred language, there was an upward trend in rates in the pre-period followed by a downward trend in the post-period, whereas there was no pre-period trend and a slight downward trend in the post-period among English preferred language (three-way interaction p = 0.022). The pre-post effect for all other outcomes did not differ by language. There were significant decreases in albuterol administration, CXR acquisition, and antibiotic treatment. There were no significant pre-post effects for steroid treatment, 7-day ED return visit, or 72-hour ED return to admission.Conclusion(s): A clinical pathway was associated with improvements in care regardless of patient language. The impact on albuterol administration was slightly greater among families with non-English versus English preferred language, but this was not the case with other outcomes. Further study is needed to understand the impact of common quality improvement interventions on equity.
Curriculum VitaeRosen.Robert CV_Harvard Format 1.3.22.pdf
Table 2. Interrupted Time Series Model for Key Outcomes Among All Patients without Stratification by LanguageValues in table represent odds ratios (95% confidence interval), adjusted for presence or absence of a complex care condition (CCC) at time of presentation to the ED and primary language. See explanation in table 1 for guidance on interpreting the reported odds ratios for each slope comparison.