2018 Section 6 - Laryngology, Voice Disorders, and Bronchoesophalogy

chronic cough risk factors at the individual and community level in the general population.

Cough is one of the most common symptoms that patients seek medical attention for in general practice. 1 Most of these consultations are because of acute cough caused by benign upper airway infections that resolve on its own. However, a substantial fraction of the patients suffers from chronic cough, de fi ned as a cough lasting > 8 weeks. 2,3 The prevalence of chronic cough in populations is up to 12%. 4-6 Clinic-based studies have suggested asthma and gastroesophageal re fl ux disease as common causes of chronic cough, 7-11 but currently we know little about the The Copenhagen General Population Study is an epidemiologic population-based cohort study of the general population of Denmark and was initiated in 2003 with ongoing enrolment. 12,13 All individuals in Denmark are assigned a unique identi fi cation number at birth or immigration and recorded in the national Danish Civil Registration System. By using this number, individuals 20 to 100 years of age were randomly selected from the national Danish Civil Registration System to re fl ect the adult white Danish population. Since the initiation of the study, > 100,000 individuals have been examined; however, only 14,669 were asked consecutively from 2013 to 2014 about chronic cough. All participants completed a comprehensive questionnaire and underwent a physical health examination. Questionnaires were reviewed at the day of attendance by an investigator together with the participant. The study was conducted according to the Declaration of Helsinki and was approved by Herlev and Gentofte Hospital and the regional ethics committee (identi fi cation No. H-KF-01-144/01). Written informed consent was obtained from all participants. Additional information on the applied methods can be found in e-Appendix 1 and e-Tables 1-4 . De fi nition of Chronic Cough Chronic cough was de fi ned as an af fi rmative response to the question “ Do you have cough lasting > 8 weeks? ” in accordance with the American College of Chest Physicians, the British Thoracic Society, and the European Respiratory Society guidelines. 8-10 The question was incorporated in the ongoing Copenhagen General Population Study from January 1, 2013. Individuals identi fi ed with chronic cough completed the Leicester Cough Questionnaire, a validated health-related quality of life questionnaire speci fi cally developed for patients with chronic cough, to determine the severity and impact of chronic cough on different aspects of health. 14 Potential Risk Factors The potential risk factors for chronic cough were chosen based on previous fi ndings and reports from specialized clinics and in line with the American College of Chest Physicians, the British Thoracic Society, and the European Respiratory Society guidelines. 8-10 Also, because dietary intake of food and beverages may in fl uence gastroesophageal re fl ux disease, believed to be a major cause of Results Among the 14,669 adults in the Copenhagen General Population Study, 554 (4%) had chronic cough. Prevalence of chronic cough in the general population Materials and Methods Study Design and Participants

We identi fi ed and ranked risk factors for chronic cough in never smokers former smokers, and current smokers, at the level of the individual and community using data from 14,669 individuals from the Copenhagen General Population Study. Because tobacco smoking is a major risk factor for chronic cough, we strati fi ed the analyses by smoking status. chronic cough, different dietary factors were also included as potential risk factors. Information on potential risk factors was collected from the Copenhagen General Population Study and the national Danish Patient Registry. By using the unique identi fi cation number of participants recorded in the national Danish Civil Registration System, it is possible to track records in the national Danish Patient Registry to obtain inpatient and outpatient hospital contacts. The national Danish Patient Registry has an overall high validity 15-18 and covers all public and private Danish hospitals from 1977 until November 2014; therefore, we did not lose track of a single person during this period. All diagnoses recorded in the national Danish Patient Registry are made by a medical doctor according to national law. Statistical Analyses Wilcoxon nonparametric rank-sum test was used for continuous data, and Pearson c 2 test was used for categorical data. All the 26 potential risk factors were dichotomized prior to analyses. Identi fi cation of potential risk factors for chronic cough was done using a stepwise estimation process in a logistic regression model with chronic cough as the dependent variable and age as a fi xed covariate on a continuous scale. Risk factors were chosen according to level of signi fi cance and entered into the logistic regression model one by one until no more predictors had P < .05. The stepwise estimation process identi fi ed risk factors for chronic cough in all individuals. Subsequently, the identi fi ed risk factors were next used with forced entry in analyses strati fi ed by smoking status. Ranking of risk factors for chronic cough at the level of the individual was according to the magnitude of the OR adjusted for age, overall and strati fi ed for smoking status. Ranking of risk factors for chronic cough at the level of the community was according to the magnitude of the population attributable risk (PAR), calculated as ( f [OR ‒ 1]) / (1 þ f [OR ‒ 1]), where f is the frequency of the risk factor in the population and OR is the OR for chronic cough in the population, overall and strati fi ed for smoking status. Information on potential risk factors was 99.9% complete; therefore, we performed multivariate imputation using chained equations to fi ll out the missing values 19 ; however, results were similar without the use of imputation. Statistical analyses were performed using STATA/SE 13.1 for Windows (StataCorp). was 3% in never smokers, 4% in former smokers, and 8% in current smokers. The prevalence was highest among current smokers 60 to 69 years of age (11%), and lowest among never smokers 20 to 39 years of age (1%)

[ 1 5 2 # 3 C H E S T S E P T E M B E R 2 0 1 7 ]

Original Research

85