Pisegna Jessica M, Kaneoka Asako, Leonard Rebecca, Langmore Susan E
Boston Medical Center, FGH Building 820 Harrison Ave, Boston, MA, 02118, USA.
Boston University, Sargent College, 635 Commonwealth Ave., Boston, MA, 02215, USA.
Dysphagia. 2018 Feb;33(1):100-108. doi: 10.1007/s00455-017-9838-7. Epub 2017 Aug 30.
The goal of this work was to better understand perceptual judgments of pharyngeal residue on flexible endoscopic evaluation of swallowing (FEES) and the influence of a visual analog scale (VAS) versus an ordinal scale on clinician ratings. The intent was to determine if perceptual judgments of residue were more accurately described by equal or unequal intervals. Thirty-three speech language pathologists rated pharyngeal residue from 75 FEES videos representing a wide range of residue severities for thin liquid, applesauce, and cracker boluses. Clinicians rated their impression of the overall residue amount in each video on a VAS and, in a different session, on a five-point ordinal scale. Residue ratings were made in two separate sessions separated by several weeks. Statistical correlations of the two rating methods were carried out and best-fit models were determined for each bolus type. A total of 2475 VAS ratings and 2473 ordinal ratings were collected. Residue ratings from both methods (VAS and ordinal) were strongly correlated for all bolus types. The best fit for the data was a quadratic model representing unequal intervals, which significantly improved the r values for each bolus type (cracker r = 0.98, applesauce r = 0.99, thin liquid r = 0.98, all p < 0.0001). Perceptual ratings of pharyngeal residue demonstrated a statistical relationship consistent with unequal intervals. The present findings support the use of a VAS to rate residue on FEES, allowing for greater precision as compared to traditional ordinal rating scales. Perceptual judgments of pharyngeal residue reflected unequal intervals, an important concept that should be considered in future rating scales.
这项工作的目标是更好地理解在吞咽功能的软性内镜评估（FEES）中对咽部残留物的感知判断，以及视觉模拟量表（VAS）与顺序量表对临床医生评级的影响。目的是确定残留物的感知判断是否通过等距或不等距能得到更准确的描述。33名言语语言病理学家对75段FEES视频中的咽部残留物进行评级，这些视频代表了稀薄液体、苹果酱和饼干团块等广泛的残留物严重程度。临床医生在一次评估中使用VAS对每个视频中总体残留物量的印象进行评级，并在另一次评估中使用五点顺序量表进行评级。残留物评级在相隔数周的两个独立环节中进行。对两种评级方法进行了统计相关性分析，并为每种团块类型确定了最佳拟合模型。总共收集了2475个VAS评级和2473个顺序评级。两种方法（VAS和顺序量表）对所有团块类型的残留物评级都高度相关。数据的最佳拟合是一个代表不等距的二次模型，这显著提高了每种团块类型的r值（饼干r = 0.98，苹果酱r = 0.99，稀薄液体r = 0.98，所有p < 0.0001）。咽部残留物的感知评级显示出与不等距一致的统计关系。目前的研究结果支持使用VAS对FEES中的残留物进行评级，与传统的顺序评级量表相比，能实现更高的精度。咽部残留物的感知判断反映了不等距，这是未来评级量表中应考虑的一个重要概念。
