Department of Emergency Medicine, Odense University Hospital, Sdr. Boulevard 29, DK-5000 Odense C, Denmark; Department of Emergency Medicine, Hospital of South West Jutland, Finsensgade 35, DK-6700 Esbjerg, Denmark.
Department of Anaesthesiology, Juliane Marie Centret, Rigshospitalet, Blegdamsvej 9, 2100 København Ø, Denmark.
J Crit Care. 2018 Apr;44:404-406. doi: 10.1016/j.jcrc.2017.12.020. Epub 2017 Dec 28.
To evaluate the inter-observer reliability of nurses assessing respiratory rate.
We presented seven minimum 60-seconds long videos of thoraces of non-identifiable patients breathing to experienced nurses from several Danish emergency departments. Two videos were assessed by 50 raters while five were reviewed by eight. The videos were shown using an online system that also recorded the counted respiratory rate.
A total of 140 nurses participated with a median of 15years' experience. The range of counted respiratory rate was minimum 10 on each video. For videos evaluated by eight nurses, average Inter Class Coefficient (ICC) was 0.662 (0.000-0.960) and individual ICC 0.197 (0.000-0.750). For the two case-videos analyzed by 50 nurses, average ICC was 0.0 (0.000-0.991) and individual ICC 0.0 (0.000-0.677).
We found a wide variation in measurements of RR with both few and many observers assessing exactly the same patients.
评估护士评估呼吸频率的观察者间可靠性。
我们向来自丹麦多个急诊部门的有经验护士展示了七段时长至少为 60 秒的非特定患者胸部的视频。两段视频由 50 名评估者评估，五段视频由 8 名评估者评估。视频使用在线系统展示，该系统还记录了计数的呼吸频率。
共有 140 名护士参与，中位数为 15 年的工作经验。每个视频的计数呼吸率范围最小值为 10。对于由 8 名护士评估的视频，平均组内相关系数（ICC）为 0.662（0.000-0.960），个体 ICC 为 0.197（0.000-0.750）。对于由 50 名护士分析的两个病例视频，平均 ICC 为 0.0（0.000-0.991），个体 ICC 为 0.0（0.000-0.677）。
我们发现，即使评估的是完全相同的患者，使用少量或大量观察者进行 RR 测量时，其结果也存在很大差异。
