Abstract
Introduction: We aimed to quantitatively evaluate the educational effectiveness of an on-the-job training (OJT) program, termed "Constipation Ultrasound Rounds," designed to teach constipation assessment skills to nurses without prior ultrasound experience.
Methods: This prospective observational study enrolled 36 nurses at a single center. Following a one-hour lecture, participants attended weekly OJT rounds led by an expert gastroenterologist. Nurses were stratified into Group A (1-3 rounds of experience, Novices) and Group B (≥ 4 rounds, Experienced). Agreement rates for assessing bladder filling, rectal stool presence, and stool consistency between nurses and the expert were evaluated using Cohen's kappa coefficient.
Results: Group A demonstrated high agreement with the expert: bladder (κ = 0.876), rectal stool (κ = 0.828), and stool consistency (κ = 1.0). This suggests that the novice group accurately replicated the procedure immediately after observing the expert. In contrast, Group B, who performed scans independently before receiving feedback, showed only fair to moderate agreement: bladder (κ = 0.512), rectal stool (κ = 0.366), and stool consistency (κ = 0.569). Discrepancies in Group B were more frequent in challenging cases such as patients with minimal bladder filling or anatomical variations.
Conclusion: Initial skill acquisition using a "See One, Do One" approach was highly effective for training nurses in constipation ultrasound assessment; however, skill mastery requires repeated practice and structured feedback. This two-phase model consisting of initial demonstration-based instruction followed by independent evaluation and feedback may serve as a practical and scalable framework for POCUS-based nursing education in real-world clinical settings.
