Authors: (including presenting author): :
Mak MY(1), Lam HKC(2), Chau JPC(3), Lo SHS(3), Choi KC(3), Tong DWK(2)
Affiliation: :
(1)Community Nursing Services, Pamela Youde Nethersole Eastern Hospital, (2) Nursing Services Department, Head Office, Hospital Authority, (3)The Nethersole School of Nursing, Faculty of Medicine, The Chinese University of Hong Kong
Keyword 1: :
Point-of-Care Ultrasound
Keyword 2: :
Nasogastric Tube Placement
Keyword 3: :
Clinical Decision-Making
Keyword 4: :
Interrater Reliability
Introduction: :
Evidence suggests Point-of-Care Ultrasound (POCUS) is useful for confirming nasogastric tube (NGT) placement due to its real time versatile imaging modality. Its clinical utility depends on operator competence and standardised protocols to ensure interpretation accuracy and methodological rigour.
Objectives: :
This study aimed to assess the interrater reliability of a standardised POCUS protocol for NGT placement verification and subsequent feeding safety decisions.
Methodology: :
This is a multicenter observational study included adults (≥18 years) undergoing NGT placement confirmation via POCUS from December 2024 to February 2025. Following NGT insertion by a non-study nurse, a trained rater performed a standardised examination comprising interpretation of neck and subxiphoid views, supplemented by a dynamic fogging test. An experienced nurse certified in emergency ultrasound, serving as gold standard, independently blindly reviewed all video recordings and data forms with images detailing placement confirmation and feeding safety recommendations. Interrater reliability for scan interpretation and decision-making was assessed using Krippendorff's alpha, a robust coefficient suitable for small samples, multiple raters, and categorical nominal data.
Result & Outcome: :
The sample included 67 adult patients (mean age: 82.21, SD: 11.21) from six community sites. The raters are nine nurses who had completed a comprehensive 3-month POCUS training programme and accrued 4 months of post-training experience. There was substantial interrater agreement: α=0.85 (95%CI: 0.48–1.00) for neck scanning, α=0.86 (95%CI: 0.58–1.00) for subxiphoid scanning, and α=0.88 (95%CI: 0.65–1.00) for feeding safety decisions. The dynamic fogging test demonstrated perfect agreement (α=1.00; 95%CI: 1.00–1.00). A high level of interrater reliability supports the validity of the training and examination protocol developed for this study. The results demonstrate methodological robustness, justifying the employment of this standardised approach in a large cohort, phase 2 main study. However, the wide confidence intervals for the neck and subxiphoid scans, compared to the perfect agreement for the dynamic fogging test, highlight these sonographic views involve greater skill complexity than the binary fogging test. This suggests a need for future training refinement, with enhanced emphasis on sonographic image acquisition and interpretation, to ensure consistent high-fidelity performance across all POCUS operators.
