Changes in the prevalence of respiratory symptoms over time in a multinational cohort study

Abstract

Background: In ageing adults, various factors may affect the prevalence of respiratory symptoms associated with chronic lung diseases, including tobacco smoking. Whether the prevalence of respiratory symptoms has changed in populations of adults across the world as they aged is unknown. Aim: This study aimed to describe the changes in the prevalence of respiratory symptoms over follow-up of an adult cohort across several world regions. An additional study aim was to evaluate the influence of smoking on those changes. Methods: A prospective (cohort) study conducted in 18 sites of the multinational population-based Burden of Obstructive Lung Disease (BOLD) study collected data on the respiratory symptoms of 3,954 participants seen on 2 occasions. Respiratory symptoms included in this study were chronic cough, chronic phlegm, dyspnoea, and wheezing. Descriptive statistics were calculated to obtain the absolute change in the prevalence of respiratory symptoms. Logistic generalised estimating equations (GEE), with exchangeable correlation adjusting for age, sex and BMI, were used to determine whether the absolute change in the prevalence of respiratory symptoms was related to smoking status. Results: After a median of 9 years of follow-up, chronic cough prevalence exhibited a decrease of 1.4%. The prevalence of wheezing and dyspnoea increased by 1.8% and 1.6%, respectively. While the prevalence of chronic phlegm showed minimal change (+0.1%), increases were mainly found in European sites, including Reykjavik (Iceland; +0.39%), Bergen (Norway; +0.39%), and Uppsala (Sweden; +0.11%). The prevalence changes in chronic cough across sites was variable but small. In contrast, changes in dyspnoea and wheezing prevalence displayed considerable variation across Asian sites, such as large reductions in Nampicuan (Philippines; -1.3%) coupled with increases in most of the Indian sites, especially Pune (+2.4%), in terms of dyspnoea prevalence. Current smokers were 1.3 times more likely to report dyspnoea, 3 times more likely to report chronic cough or chronic phlegm, and 4 times more likely to report wheezing than never-smokers. Ex-smokers were 2 times more likely to report chronic cough, chronic phlegm, and wheezing compared to never-smokers. There were no differences in the change in prevalence of dyspnoea among ex-smokers. Conclusion: The prevalence of chronic cough decreased overall across different sites, while the prevalence of wheezing and dyspnoea increased. The net change in chronic cough prevalence was attributable to smoking cessation. This study suggests that the increase of dyspnoea and wheezing prevalence is unlikely to be explained by changes in smoking behaviour.