Development of Community Healthcare, Hirosaki University Graduate School of Medicine, Hirosaki-shi, Aomori, Japan.
General Medicine, Hirosaki University Graduate School of Medicine, Hirosaki-shi, Aomori, Japan.
PLoS One. 2018 May 10;13(5):e0197163. doi: 10.1371/journal.pone.0197163. eCollection 2018.
To elucidate the accuracy and optimal cut-off point of self-diagnosis and clinical symptoms of seasonal influenza compared with rapid influenza diagnostic tests as the reference standard, we conducted a cross sectional observational study at a rural clinic in Japan.
Data during three influenza seasons (December 2013 to April 2016) were retrospectively collected from the medical records and pre-examination sheets of 111 patients aged >11 years (mean age 48.1 years, men 53.2%) who were suspected of influenza infection and underwent rapid influenza diagnostic testing. Patients' characteristics (age, sex, and past medical history of influenza infection), clinical signs (axillary temperature, pulse rate, cough, joint and muscle pain, and history of fever [acute or sudden, gradual, and absence of fever]), duration from the onset of symptoms, severity of feeling sick compared with a common cold (severe, similar, and mild), self-reported likelihood of influenza (%), and results of rapid influenza diagnostic tests.
At the optimal cut-off point (30%) for estimation of self-diagnosis of seasonal influenza, the positive likelihood ratio (LR+) was 1.46 (95% confidence interval 1.07 to 2.00) and negative likelihood ratio (LR-) was 0.57 (0.35 to 0.93). At a 10% cut-off point, LR-was 0.33 (0.12 to 0.96). At an 80% cut-off point, LR+ was 2.75 (0.75 to 10.07). As for clinical signs, the combination of acute or sudden onset fever and cough had LR+ of 3.27 (1.68 to 6.35). Absence of cough showed LR-of 0.15 (0.04 to 0.61).
Self-diagnosis of influenza using the optimal cut-off point (30%) was not found useful for ruling in or ruling out an influenza diagnosis. However, it could be useful when patients self-report extremely high (80%) or low (10%) probability of having influenza. Clinically useful signs were the combination of history of fever and cough, and absence of cough.
为了阐明自我诊断和季节性流感的临床症状与快速流感诊断检测相比的准确性和最佳截断点，我们在日本的一家农村诊所进行了一项横断面观察性研究。
我们回顾性地收集了三个流感季节（2013 年 12 月至 2016 年 4 月）期间的 111 名年龄大于 11 岁（平均年龄 48.1 岁，男性占 53.2%）的疑似流感感染患者的病历和预诊表中的数据，这些患者接受了快速流感诊断检测。患者的特征（年龄、性别和流感感染的既往病史）、临床症状（腋窝温度、脉搏率、咳嗽、关节和肌肉疼痛以及发热史[急性或突然、逐渐和无发热]）、从症状发作到就诊的时间、与普通感冒相比的感觉严重程度（严重、相似和轻度）、自我报告的流感可能性（%）以及快速流感诊断检测的结果。
在季节性流感自我诊断的最佳截断点（30%），阳性似然比（LR+）为 1.46（95%置信区间为 1.07 至 2.00），阴性似然比（LR-）为 0.57（0.35 至 0.93）。在 10%截断点时，LR-为 0.33（0.12 至 0.96）。在 80%截断点时，LR+为 2.75（0.75 至 10.07）。就临床症状而言，发热和咳嗽的急性或突然发作组合的 LR+为 3.27（1.68 至 6.35）。无咳嗽的 LR-为 0.15（0.04 至 0.61）。
使用最佳截断点（30%）进行流感的自我诊断对于诊断或排除流感的诊断并不有用。但是，当患者自我报告极高（80%）或极低（10%）的流感可能性时，这可能会有用。有用的临床症状是发热和咳嗽的病史以及无咳嗽的组合。
