Validation and reproducibility of the lung function questionnaire (LFQ) for the diagnosis of COPD in Colombia
Article Category: Original Papers
Published Online: Jul 27, 2020
Page range: 37 - 46
DOI: https://doi.org/10.2478/pneum-2020-0010
Keywords
© 2020 Alirio Rodrigo Bastidas Goyes et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Introduction
The development of clinical prediction scales and their use can reduce under-diagnosis and increase early detection of chronic obstructive pulmonary disease (COPD). The performance of clinical prediction scales in Colombia is unknown. The objective of this study is to evaluate the validity and reproducibility of the lung function questionnaire (LFQ) in Colombia.
Method
A cross-sectional study was performed, with analysis of diagnostic validity and reliability in people over 40 years of age who underwent a spirometry test. The LFQ questionnaire was applied. To assess reproducibility, the test was carried out at two time points: first at the initial consultation; and then 1 day to 1 week after the previous application. Spirometry was performed immediately after the initial questionnaire, meeting the American Thoracic Society criteria.
Results
Among the 1996 subjects included in the analysis, the average age was 65 years (SD: 11.97 years), prevalence of COPD was 21.3%, the intra-class correlation coefficient between the two time points was 0.844 (95% CI: 0.863–0.901) (p < 0.001), and kappa was 0.797 for the dichotomous outcome ≤18 COPD risk points (p < 0.001), validity analysis using the area under the receiver operating characteristic curve for the population evaluated was 0.715 (95% CI: 0.685–0.745); the dichotomous outcome of the questionnaire ≤18 points was as follows: sensitivity – 91.18% (95% CI: 88.0–94.3); specificity – 32.41% (95% CI: 29.8–35.0); positive predictive value – 26.7% (95% CI: 24.1–29.3); negative predictive value – 93.15% (95% CI: 90.7–95.6); likelihood ratio (LR) +: 1.34 (95% CI: 1.28–1.42), LR– 0.27 (95% CI: 0.19–0.39); number needed to diagnose: 4; number needed to misdiagnose: 2 (p < 0.001).
Conclusion
The LFQ questionnaire has good performance for the diagnosis of COPD, especially in populations without previous respiratory symptoms or usual risk factors, optimising the use of spirometry to increase its detection.