Implementing Respiratory Illness Care Pathways in General Hospitals: A Cluster Randomized Trial.

BACKGROUND AND

OBJECTIVES

A prior study at a freestanding children's hospital tested simultaneous implementation of multiple clinical pathways and demonstrated improved guideline adherence and patient health outcomes. This high-efficiency approach has not been studied in general hospitals, which face unique implementation barriers. Our objective was to determine the effectiveness of a multi-condition pathway intervention in general hospitals.

METHODS

A pragmatic, cluster-randomized trial (1:1 allocation to intervention versus waitlist control) conducted from July 2022 to June 2024 in general hospitals across the United States. Admissions of children with a primary diagnosis of asthma, pneumonia, or bronchiolitis were analyzed. The intervention included inpatient pathways for asthma, pneumonia, and bronchiolitis implemented using audit and feedback, electronic order sets, iterative improvement cycles, and mentoring/facilitation. The primary outcome was mean composite adherence (percentage) across 7 evidence-based practices (2-3 analyzed for each diagnosis). Secondary outcomes included mean adherence to each practice, length of stay, transfer to intensive care, and 30-day hospital readmission or emergency revisit.

RESULTS

Forty-three hospitals were randomized, and 11 760 hospital admissions were analyzed. The intervention was associated with significant increases in mean composite adherence to evidence-based practices relative to control (absolute increase 9.9 percentage points [95% CI: 7.4-12.4]) and increases in 4 of 7 practices. It was also associated with a significant decrease in intensive care transfers (adjusted odds ratio 0.69 [95% CI 0.54-0.89]) but no significant effects on length of stay or 30-day readmission/revisit.

CONCLUSIONS

The multicondition pathway intervention was effective in general hospitals, demonstrating significant improvements in adherence to evidence-based care over a sustained period.