三位不同职称喉科医师对咽喉反流体征评分量表的应用研究

Study on the consistency of reflux score evaluated by three different level of throat physicians

目的不同职称不同医院3位喉科医师应用反流体征评分量表（refluxfindingscore，RFS）评分的一致性研究。方法任意抽取门诊因咽喉疾病就诊患者的110张喉镜图片（每人次1张）。由2位高级职称医师（A、B）和1位初级职称医师（c）对110张喉镜图片进行两次（间隔1周）双盲RFS评分。分析整理数据，对RFS进行信度和效度的评估。结果3位医师初次评估平均（x±s，下同）RFS总分：A为（9．05±2．54）分；B为（8．80±2．20）分；C为（8．98±2．21）分。再次重测平均RFS总分：A为（9．20±2．47）分，B为（9．03±2．14）分，C为（8．91±2．30）分。3位平均RFS总分重测信度分别为0．860、0．800、0．837，P值均〈0．01；各项条目重测信度：A为0．662～1．000，B为0．486～1．000，C为0．613～1．000。3位医师两次平均RFS总分比较差异无统计学意义（χ2=1．553，P=0．907）。在110例图片中，把RFS总分〉7分判定为咽喉反流患者，3位喉科医师的组间一致性分别为83．6％（κ=0．617）、85．5％（κ=0．644）、89．1％（κ＝0．720），P值均〈0．01；3位医师两次评分的一致性分别为91．8％（κ＝0．807）、81．8％（κ=0．534）、90．9％（κ=0．741），P值均〈0．01。结论不同职称、不同医院喉科医师对咽喉反流体征评分量表的应用结果一致，不受教育背景和临床经验影响，在国内可以推广使用。

Objective The consistency of reflux finding score （RFS） was studied by three ditterent level of throat physicians. Methods One hundred and ten laryngeal photos were chosen to assess the RFS test-retest reliability on two separate occasions at least more than one week. Results The mean total RFS scores for doctor A were 9. 05 ±2. 54, doctor B were 8.80 ±2. 20, doctor C were 8.98 ±2. 21 at the initial screening, and 9.20±2. 47, 9.03 ±2. 14 and 8.91 ±2. 30 respectively at the repeat evaluation. The test- retest reliability of total RFS scores of three doctors were 0. 860, 0. 800 and 0. 837 respectively, P all 〈 0. 001. The test-retest reliability of each item scores for doctor A were from 0. 662 to 1. 000, doctor B were from 0. 486 to 1. 000 and doctor C were from 0. 613 to 1. 000. There were no differences in the total RFS scores comparing among each evaluation of the three doctors （ χ2 = 1. 553, P = 0. 907）. Total RFS scores more than 7 were considered as abnormal. The interobsever consistency was as follows ： A with B was 83.6% （κ=0.617, P=0.000）,A with C was 85.5% （κ=0.644, P =0.000）,B with C was 89.1% （κ= 0. 720, P = 0. 000）. The intraobserver consistency was 91.8% （ κ= 0. 807, P = 0. 000）, 81.8% （ κ = 0. 534, P = 0. 000）, 90. 9% （ κ = 0. 741, P = 0. 000 ） respectively. Conclusions The result of this study shows that the assessment of RFS is not influenced by different educational backgrounds and clinical experience. RFS can be applied widely in China.