Department of Epidemiology and Public Health, University Hospital, Rennes, France.
Univ Rennes, CHU Rennes, Inserm, EHESP, Irset (Institut de recherche en santé, environnement et travail) - UMR_S 1085, Rennes, France.
Clin Respir J. 2020 Sep;14(9):813-821. doi: 10.1111/crj.13201. Epub 2020 May 28.
People at risk of chronic obstructive pulmonary disease (COPD) can benefit from appropriate medical management before severe symptoms appear. This study assesses the value of the COPD Assessment Test (CAT) questionnaire for screening dairy farmers, who tend to be slow or reluctant to seek health care.
During the time period 2012-2017, 2089 randomly selected dairy farmers in Brittany (France) were invited to complete self-administered questionnaires (including the CAT) and to undergo an occupational health check-up using an electronic mini-spirometer and conventional spirometry. Those showing symptoms suggestive of COPD and/or a ratio FEV /FEV < 80% were sent to a pulmonologist for a further check-up, including spirometry with a reversibility test. Multivariate logistic models based on CAT scores and socio-demographic or work-related factors were developed to predict COPD.
The 1231 farmers who underwent the occupational health check-up included 1203 who met the inclusion/exclusion criteria. Pulmonologist identified 16 (1.3%) cases of COPD. A multivariate logistic regression model (covariates: CAT sum score, on-farm time, BMI, smoking status, free-stall mulching) provided an area under the receiver-operating characteristic curve (AUC) of 0.87 (95% CI: 0.75-0.98). Using a cut-off of 0.007 gave a sensitivity of 93.8% and a specificity of 62.4%. Another model that included CAT breathlessness and the same covariates performed marginally better (AUC = 0.88, 95% CI: 0.77-0.98).
Our predictive models can both benefit dairy farmers by providing early diagnosis and management of their COPD and avoid unnecessary, costly spirometry during the screening process.
有发生慢性阻塞性肺疾病（COPD）风险的人群，在出现严重症状之前，可以从适当的医疗管理中获益。本研究评估 COPD 评估测试（CAT）问卷在筛选奶牛场工人中的价值，他们往往行动缓慢或不愿寻求医疗保健。
在 2012 年至 2017 年期间，布列塔尼（法国）的 2089 名随机选择的奶牛场工人受邀填写自我管理问卷（包括 CAT），并使用电子迷你肺活量计和常规肺活量计进行职业健康检查。那些表现出 COPD 症状和/或 FEV/FVC < 80%的人被送往肺病专家进行进一步检查，包括带有可逆性测试的肺活量计检查。基于 CAT 评分和社会人口统计学或工作相关因素的多变量逻辑模型用于预测 COPD。
在接受职业健康检查的 1231 名农民中，有 1203 名符合纳入/排除标准。肺病专家发现 16 例（1.3%）COPD 病例。多变量逻辑回归模型（协变量：CAT 总分、在农场时间、BMI、吸烟状况、自由卧床覆盖）提供了 0.87（95%CI：0.75-0.98）的接收者操作特征曲线（ROC）下面积。使用 0.007 的截断值可获得 93.8%的敏感性和 62.4%的特异性。包含 CAT 呼吸困难和相同协变量的另一个模型表现稍好（AUC=0.88，95%CI：0.77-0.98）。
我们的预测模型可以通过为奶牛场工人提供 COPD 的早期诊断和管理来使他们受益，并避免在筛选过程中进行不必要的、昂贵的肺活量计检查。
