How to Reduce Unnecessary Postoperative Labs After Elective General Surgery
Eleah D. Porter1, *Julia L. Kelly1, *Lye-Yeng Wong2, *Allison R. Wilcox1, Christina V. Angeles1,2
1Dartmouth-Hitchcock Medical Center, Lebanon, NH;2Geisel School of Medicine, Hanover, NH
Objective: Laboratory tests are often obtained empirically post-surgery without specific indications. Our goal was to 1) assess utilization and clinical impact of routine postoperative labs (POLs) and 2) evaluate provider perceptions of their necessity and how to reduce occurrence.
Design:
1. Retrospective cohort study of patients undergoing routine (no documented clinical indication or concern) POLs.
2. 15-item validated questionnaire administered to providers.
Setting:
1. Academic medical center.
2. Two teaching hospitals: academic-center and community-affiliated.
Patients:
1. Adult elective general surgery patients admitted for ≤48 hours, January-June 2018.
2. Surgical trainees (ST) and independent practitioners (IP).
Main Outcome Measures:
1. Lab-driven change in clinical care.
2. Provider survey responses.
Results:
1. 345 patients met inclusion. Routine POLs were obtained on 87% (301/345) of patients, of which 89% (269/301) had an abnormality by institutional lab thresholds. Lab results prompted a change in care in 11% (33/301) of tested patients (e.g., electrolyte repletion).
2. A convenience sample of 78 providers were administered survey (32 STs, 46 IPs). Overall, 86% agreed with the statement that unnecessary routine POLs are ordered and 85% attributed this practice to “habit.” However, STs and IPs did not universally agree, as STs were significantly more likely to attribute “concern a senior member will ask” in comparison to IPs (63% vs. 48%, p=0.014). There was no consensus regarding a single intervention to reduce practice, but the most common answers were cost transparency (33%) and faculty role modeling (27%).
Conclusions: Routine POL testing after elective general surgery has limited clinical impact. Surgical providers agree that unnecessary POLs are ordered and varied survey responses suggest that a multimodal approach will be needed to reduce empiric POL testing.
Back to 2019 Posters
