Adherence to Enhanced Recovery Protocols in NSQIP and Association With Colectomy Outcomes. Issue 3 (March 2019)

• Record Type: Journal Article
• Authors: Berian, Julia R.; Ban, Kristen A.; Liu, Jason B.; Ko, Clifford Y.; Feldman, Liane S.; Thacker, Julie K.
• Is Part Of: Annals of surgery. Volume 269:Issue 3(2019)
• Publication Date: 2019-03
• Subjects: colectomy outcomes -- enhanced recovery -- implementation science -- protocol adherence -- protocol compliance -- quality improvement; Surgery -- Periodicals; 617.005
• DOI: 10.1097/SLA.0000000000002566
• Languages: English
• ISSNs: 0003-4932

Abstract

Abstract : Objective: To evaluate the effect of protocol adherence on length of stay (LOS) and recovery-specific outcomes after colectomy. Background: Enhanced recovery protocols (ERPs) may decrease postoperative morbidity and LOS; however, the effect of overall protocol adherence remains unclear. Methods: Using American College of Surgeons' National Surgical Quality Improvement Program colectomy data (July 2014–December 2015) and 13 novel ERP variables, propensity scores were constructed for low (0–5), moderate (6–9), and high adherence (10–13 components). Prolonged LOS (>75th percentile, uncomplicated cases) was modeled with multivariable logistic regression with robust standard errors, adjusted for hospital-level clustering and propensity score. Secondary recovery-specific outcomes were modeled with negative binomial regression. Subgroup analysis was conducted on uncomplicated cases. Results: Among 8139 elective colectomies at 113 hospitals, LOS increased with decreasing adherence (4.3 days [SD 3.3] high adherence vs 7.8 [SD 6.8] low adherence; P < 0.0001). High adherence was associated with fewer complications, including postoperative ileus, compared with moderate ( P < 0.0001) and low adherence ( P < 0.0001). High-adherence patients achieved recovery milestones earlier (vs low adherence), with return of bowel function at 1.9 (vs 3.7) days, tolerance of diet at 2.4 (vs 5.4) days, and oral pain control at 2.7 (vs 5.0) days ( P < 0.0001). Risk-adjusted odds of prolonged LOS were significantly increased for low (odds ratio 2.7, 95% confidence interval 2.0–3.6) and moderate-adherence (odds ratio 1.7, 95% confidence interval 1.4–2.1) groups. In a negative binomial regression, time to recovery was 60% to 95% longer for low versus high adherence ( P < 0.0001). Conclusions: In this large, multi-institutional North American data registry, high adherence to ERPs was associated with earlier recovery, decreased complications, and shorter LOS. ERPs can improve outcomes; however, benefits correlate with adherence. Abstract : Supplemental Digital Content is available in the text