PUMA screening tool to detect COPD in high-risk patients in Chinese primary care–A validation study

Abstract

The early stage of chronic obstructive pulmonary disease (COPD) is not easily recognized. Screening tools can help to identify high-risk patients in primary care settings for spirometry and may be helpful in the early detection in COPD and management. This study aims to validate the PUMA questionnaire for use in Chinese primary care settings. This cross-sectional study recruited participants (≥40 years old, current or former smoker with ≥10 packs of cigarette per year) in primary health care clinics in Hong Kong. The Chinese version of the PUMA questionnaire was administered by trained research staff to participants awaiting consultation. COPD diagnosis was confirmed by spirometry (post-bronchodilator FEV1/FVC <0.70). A total 377 patients were recruited of which 373 completed the spirometry. The percentage of participants diagnosed with COPD (post-bronchodilator FEV1/FVC <0.70) was 27.1%. A higher PUMA score was more likely to have an advanced stage of GOLD classification (P = 0.013). The area under the ROC curve of the PUMA score was 0.753 (95%CI 0.698–0.807). The best cut-point according to Youden’s index for PUMA score was ≥6 with sensitivity 76.5%, specificity 63.3% and negative predictive value (NPV) 63.3%. A cut-off point of PUMA score ≥5 was selected due to higher sensitivity of 91.2%, specificity of 42.6% and high NPV of 92.7%. PUMA score performed better than CDQ and COPD-PS in the area under the ROC curve (0.753 versus 0.658 and 0.612 respectively), had higher sensitivity than COPD-PS (91.2% versus 61%) and had higher specificity than CDQ (42.6% versus 13.1%). The use of PUMA as a screening tool was feasible in Chinese primary care and can be conducted by trained staff and health professionals. The validation results showed high sensitivity and high NPV to identify high risk patient with COPD at cut-off point of ≥5. It can be useful for early detection and management of COPD.

Comparative analysis between PUMA and CAPTURE questionnaires for chronic obstructive pulmonary disease (COPD) screening in smokers

“…established a cut-off value similar to the value proposed by a previous study [12], which had a sensitivity of 76.5% and specificity of 63.3%. The previous study showed that PUMA scores of ≥5 offered higher sensitivity (91.2%) but lower specificity (42.6%), influenced by factors such as age and smoking prevalence [12].…”

“…established a cut-off value similar to the value proposed by a previous study [12], which had a sensitivity of 76.5% and specificity of 63.3%. The previous study showed that PUMA scores of ≥5 offered higher sensitivity (91.2%) but lower specificity (42.6%), influenced by factors such as age and smoking prevalence [12]. The AUC of 0.783 highlights the PUMA questionnaire's effectiveness in distinguishing COPD from non-COPD cases among smokers, showing better predictive accuracy than the CAPTURE questionnaire.…”

“…The moderate positive correlation between the PUMA and CAPTURE questionnaires (Spearman's rho 0.636) indicated that while both tools are aligned in their assessment of COPD risk, they might be targeting slightly different aspects of COPD risk factors and symptoms [12,24]. We found that The CAPTURE questionnaire's optimal cut-off was identified as 4, with a sensitivity of 48.2% and specificity of 88.6% for detecting clinically significant, undiagnosed COPD.…”

“…Smokers' demographic information, including age, gender, smoking history, and age of smoking initiation, was collected through the initial interview. Smoking intensity was calculated based on the amount of cigarettes consumed: light smoker (<20 pack-years), moderate smoker (20−30 pack-years), and heavy smoker (>30 pack-years) [12]. Smokers were screened for COPD by completing questionnaires with the PUMA and CAPTURE screening tools through face-to-face interviews.…”

“…The subjective symptoms included dyspnea, sputum, and cough, and each score ranged from 0 to 1 point. The maximum possible score is 9, and those scoring 5 or higher were considered to be at risk of COPD [12].…”