ENRICH-US Trial Links Fidelity To Pediatric GI Surgery Outcomes

Key Takeaways

• Overall postoperative length of stay did not differ significantly across study phases.
• Compared with baseline, the sustainability phase was associated with faster return to a regular diet and lower in-hospital opioid exposure.
• In sensitivity analyses, receipt of at least 13 protocol elements was associated with shorter stays and fewer complications, and no clinically significant complications were attributed to the intervention.

Across 18 US pediatric surgery sites, the ENRICH-US trial found no significant phase-based change in postoperative length of stay among 597 patients aged 10 to 18 years. The cohort included older children and adolescents undergoing elective gastrointestinal procedures. That phase comparison reflected rollout timing rather than the number of protocol elements received by individual patients. Implementation fidelity still increased as the 21-element protocol spread across participating hospitals, and sensitivity analyses showed outcome signals that aligned more closely with higher patient-level fidelity than with rollout phase alone.

The trial was a prospective type 2 hybrid implementation-effectiveness study at 18 US pediatric surgery sites. This stepped-wedge cluster-randomized clinical trial included at least 9 months of control care, 12 months of implementation, and 1 to 2 years of sustainability. Usual care served as the control comparator, and the intervention was a 21-element enhanced recovery protocol for elective procedures such as bowel resection, ostomy closure, stricturoplasty, and colectomy. Implementation support included a structured toolkit, a 1-year group learning collaborative, a repository of tools, quarterly feedback reports, and site report cards. Among 597 patients, median age was 15 years, and primary and secondary outcomes included postoperative hospital length of stay, opioid use, diet timing, complications, readmission, and patient-reported HRQOL.

Postoperative length of stay remained similar across phases, and most secondary outcomes also showed no significant differences. Time to regular diet decreased by 36.1% between control and sustainability phases, with a 95% CI of 5.91% to 56.61%. Average morphine milligram equivalents per hospital day fell by 55.9%, with a 95% CI of 11.4% to 78.0% and P=.02. Postoperative complications and readmission did not differ significantly by phase, while subgroup signals favored bowel resection or ileocecectomy and lower opioid use in selected procedures. Interaction testing was limited by sample size, and later-phase recovery measures improved without an overall phase-based difference in length of stay.

Median patient-level fidelity increased from 11 protocol elements at baseline to 14 during implementation and remained 14 during sustainability, with P<.001. In a post hoc sensitivity analysis, patients receiving at least 13 elements had a 1.14-day shorter median length of stay, with a 95% CI of 0.27 to 2.01. That high-fidelity group also had fewer complications, with an adjusted odds ratio of 0.48 and a 95% CI of 0.28 to 0.82. Site-level fidelity correlated with EHR order set integration, Kendall 0.46, 95% CI 0.15 to 0.69, P=.02, and with site culture or attitudes, 0.41, 95% CI 0.09 to 0.65, P=.03. No clinically significant complications were attributed to the intervention, and in this older pediatric GI cohort the outcome signal aligned more closely with fidelity than with rollout phase alone during follow-up.