TUM School of Medicine, Institute of General Practice and Health Services Research, Technical University of Munich, Munich, Germany.
TUM School of Medicine, Institute of Medical Informatics, Statistics and Epidemiology, Technical University Munich, Munich, Germany.
NPJ Prim Care Respir Med. 2021 Nov 25;31(1):46. doi: 10.1038/s41533-021-00258-4.
The study aimed to evaluate the diagnostic accuracy of contact history and clinical symptoms and to develop decision rules for ruling-in and ruling-out SARS-CoV-2 infection in family practice. We performed a prospective diagnostic study. Consecutive inclusion of patients coming for COVID-PCR testing to 19 general practices. Contact history and self-reported symptoms served as index test. PCR testing of nasopharyngeal swabs served as reference standard. Complete data were available from 1141 patients, 605 (53.0%) female, average age 42.2 years, 182 (16.0%) COVID-PCR positive. Multivariable logistic regression showed highest odds ratios (ORs) for "contact with infected person" (OR 9.22, 95% CI 5.61-15.41), anosmia/ageusia (8.79, 4.89-15.95), fever (4.25, 2.56-7.09), and "sudden disease onset" (2.52, 1.55-4.14). Patients with "contact with infected person" or "anosmia/ageusia" with or without self-reported "fever" had a high probability of COVID infection up to 84.8%. Negative response to the four items "contact with infected person, anosmia/ageusia, fever, sudden disease onset" showed a negative predictive value (NPV) of 0.98 (95% CI 0.96-0.99). This was present in 446 (39.1%) patients. NPV of "completely asymptomatic," "no contact," "no risk area" was 1.0 (0.96-1.0). This was present in 84 (7.4%) patients. To conclude, the combination of four key items allowed exclusion of SARS-CoV-2 infection with high certainty. With the goal of 100% exclusion of SARS-CoV-2 infection to prevent the spread of SARS-CoV-2 to the population level, COVID-PCR testing could be saved only for patients with negative response in all items. The decision rule might also help for ruling-in SARS-CoV-2 infection in terms of rapid assessment of infection risk.
这项研究旨在评估接触史和临床症状的诊断准确性，并制定用于家庭实践中排除和确诊 SARS-CoV-2 感染的决策规则。我们进行了一项前瞻性诊断研究。连续纳入了 19 家普通诊所前来进行 COVID-PCR 检测的患者。接触史和自我报告的症状作为指标测试。鼻咽拭子的 PCR 检测作为参考标准。共有 1141 名患者的完整数据可用，其中 605 名（53.0%）为女性，平均年龄为 42.2 岁，182 名（16.0%）COVID-PCR 检测呈阳性。多变量逻辑回归显示，“与感染者接触”（优势比 9.22，95%置信区间 5.61-15.41）、嗅觉丧失/味觉丧失（8.79，4.89-15.95）、发热（4.25，2.56-7.09）和“疾病突然发作”（2.52，1.55-4.14）的比值最高。有“与感染者接触”或“嗅觉丧失/味觉丧失”或伴有自我报告“发热”的患者，COVID 感染的可能性高达 84.8%。四项“与感染者接触、嗅觉丧失/味觉丧失、发热、疾病突然发作”的阴性反应显示出 0.98（95%置信区间 0.96-0.99）的阴性预测值（NPV）。446 名（39.1%）患者符合这一条件。完全无症状、无接触、无风险地区的 NPV 为 1.0（0.96-1.0）。84 名（7.4%）患者符合这一条件。总之，四项关键项目的结合可以高度确定地排除 SARS-CoV-2 感染。如果目标是 100%排除 SARS-CoV-2 感染以防止 SARS-CoV-2 在人群水平上传播，那么只有对所有项目均无反应的患者才可以进行 COVID-PCR 检测。该决策规则也可以帮助快速评估感染风险，从而确定 SARS-CoV-2 感染。
