Background: Undiagnosed chronic obstructive pulmonary disease (COPD) and asthma remain prevalent health issues. The current global and Canadian prevalence reported for obstructive lung disease do not reflect the true prevalence since undiagnosed cases remain missed and uncounted. Spirometry testing is viewed as the current gold standard for diagnosing obstructive lung disease. However, barriers associated with inaccessibility and underuse have contributed to undiagnosed lung disease. While guidelines advise against spirometry for asymptomatic persons, active case-finding for persons at-risk and those presenting with symptoms has been recommended. Given early treatment and management has the potential to improve health-related quality of life and reduce the progression of lung decline, identifying undiagnosed lung disease is critical to preventing adverse health outcomes. To date, this marks the first study to incorporate both obstructive lung diseases into a single-case finding instrument.
Objective: To develop and validate a case-finding questionnaire to identify undiagnosed COPD and asthma in community-dwelling adults, and to prospectively evaluate reliability and predictive performance.
Methods: This study uses data obtained from the Undiagnosed Chronic Obstructive Pulmonary Disease and Asthma Population (UCAP) study from June 2017 to March 2020. Eligible participants were >18 years, had a history of chronic respiratory symptoms, and had no previous physician diagnosis of obstructive lung disease. Presence of obstructive lung disease was confirmed with spirometry. Multinomial logistic regression and recursive partitioning were used to develop a case-finding questionnaire. Predictors available from six questionnaires completed during spirometry visit. Diagnostic accuracy of the models was used to evaluate performance. Risk score externally validated in a cohort of participants recruited between October 2020 and January 2021 at study sites open during the COVID-19 pandemic.
Results: Derivation cohort included 1615 participants, with 136 ultimately diagnosed with asthma and 195 diagnosed with COPD. A 13-item questionnaire was developed using logistic regression: age, pack-years of cigarette smoking, wheeze, cough, sleep, chest tightness, level of tiredness, physical activity limitation, occupational exposure, primary or second-hand smoke exposure, frequency of chest attacks, and salbutamol medication. Internal validation showed an area under the curve (AUC) of 0.79 (0.70-0.90) for COPD and 0.64 (0.45-0.80) for asthma. At a predicted probability of greater than or equal to 6%, specificity was 17% for no OLD, sensitivity was 91% for asthma, and sensitivity was 96% for COPD. External cohort included 74 subjects, with 8 diagnosed with COPD and 6 diagnosed with asthma. The AUC for COPD was 0.89 (95% CI: 0.62-0.90) and AUC was 0.65 (95% CI: 0.63-0.72) for asthma. Sensitivity was 100% for both asthma and COPD, specificity was 13%, and positive predictive value was 23%.
Conclusion: The 13-item case-finding questionnaire was shown to be reliable and with modest predictive ability in identifying COPD and asthma. Prospective evaluation with the UCAP study is still ongoing to recruit a larger sample to re-evaluate predictive performance.
Identifer | oai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/42683 |
Date | 17 September 2021 |
Creators | Huynh, Chau |
Contributors | Aaron, Shawn David |
Publisher | Université d'Ottawa / University of Ottawa |
Source Sets | Université d’Ottawa |
Language | English |
Detected Language | English |
Type | Thesis |
Format | application/pdf |
Page generated in 0.0105 seconds