Original Paper Chronic Respiratory Disease Volume 19: 1–10 Prevalence of chronic cough and patient © The Author(s) 2022 Article reuse guidelines: characteristics in adults in Spain: sagepub.com/journals-permissions DOI: 10.1177/14799731221098722 journals.sagepub.com/home/crd A population-based cross-sectional survey Javier Domı́nguez-Ortega1,2, Vicente Plaza2,3,4, Vicky W Li5, Eileen Fonseca6, Luis Cea-Calvo7, Ashley Martin5, Marta Sánchez-Jareño7, Joanne E Brady6 and Jonathan Schelfhout6 Abstract Background: Chronic cough (CC) represents a significant health burden. This study assessed the prevalence of CC (defined as per international guidelines as cough duration >8 weeks) in Spanish adults and compared characteristics between CC and non-CC cohorts. Methods: CC cohorts were compiled using data from adult respondents to the 2020 Spanish cross-sectional online National Health and Wellness Survey (NHWS). Using propensity scores, respondents experiencing CC during their lifetime and the previous 12 months were matched 3:1 to respondents without CC and their health characteristics were compared. The number of Spanish adults affected with CC was estimated using weighted CC prevalence. Results: CC during their lifetime or the previous 12 months was experienced by 579 (8.2%) and 389 (5.5%) of 7074 NHWS respondents, of whom 233 (38.5%) and 171 (44.0%), respectively, had physician-diagnosed CC. Based on weighted prevalence rates, lifetime and 12-month CC were estimated to affect ≈3.3 million and ≈2.2 million Spanish adults, re- spectively. Relative to the non-CC cohort, the 12-month CC cohort consistently demonstrated poorer health status, poorer mental health, greater healthcare utilization, and lower productivity at work and home. Conclusion: This study contributes novel data regarding the prevalence of CC in Spain, suggests that CC is under- diagnosed, and reflects that CC and related comorbidities inflict a significant health burden in the affected population. Keywords Chronic cough, epidemiology, prevalence, survey Date received: 19 December 2021; accepted: 17 April 2022 1 Allergy Department, La Paz University Hospital, IdiPAZ, Madrid, Spain 2 CIBER of Respiratory Diseases, CIBERES, Madrid, Spain 3 Respiratory Medicine Department, Santa Creu i Sant Pau Hospital, Introduction Barcelona, Spain 4 Cough is a common reason for visiting primary healthcare Medicine Department, University Autonomous of Barcelona (UAB), Barcelona, Spain professionals (HCPs) worldwide1 and a significant cause of 5 Cerner Enviza, New York, NY, USA morbidity.2 The estimated global prevalence of chronic 6 Merck & Co., Inc. Rahway, NJ, USA cough (CC), regardless of definition, is 9.6%, ranging from 7 Medical Affairs, MSD, Madrid, Spain 2.3% and 4.4% in Africa and Asia to 11.0–18.1% in Corresponding author: America, Europe, and Oceania.2 Using the international Javier Domı́nguez-Ortega, Allergy Department, La Paz University guideline definition of CC (cough lasting ≥8 weeks,3,4 CC Hospital, Paseo de la Castellana, 261, Madrid 28046, Spain. was more prevalent in Europe (12.0%5 and 7.2%6 in the Email: javier.dominguez@idipaz.es Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 Chronic Respiratory Disease UK5 and Finland6) than in Asia (2.2%7 and 2.6%8 in Japan7 Health status was assessed using the Medical Outcomes and Korea8), and Africa (1.1% in Nigeria9). Study 12-item Short Form Survey (SF-12v2).19 Better health The absence of International Classification of Diseases status is represented by higher domain (physical functioning, (ICD) codes for CC prevents the use of health record da- physical role limitations, bodily pain, general health, vitality, tabases for quantifying CC prevalence. Research conducted social functioning, emotional role limitations, and mental in clinical settings and specialist cough centers may provide health), physical component summary (PCS), and mental different results than in the overall population. In Spain, component summary (MCS) scores. A score of 50 (±10), was evidence regarding the epidemiology of CC at the assumed to be the population norm. The SF-12v2 response population-level is lacking. Because CC negatively impacts pattern was converted to SF-6D Utility Scores, which range health status and quality of life (QoL),2,7,10–15 emotional well- from 0 (health state equivalent to death) to 1 (perfect health). being (associated with anger/frustration,14 depression,14–16 Depression and anxiety symptoms over the prior 2 weeks anxiety14,15) daily living7,15 and social activities,14 insights were evaluated respectively by the 9-item Patient Health on the frequency and impact of CC are of interest for Questionnaire (PHQ-9)20 and 7-item Generalized Anxiety healthcare planning. Disorder scale (GAD-7).21 Each item is scored from 0–3 for This study aimed to estimate the prevalence of CC in a total PHQ-9 score of 0–27 (scores ≥10 suggest major Spanish adults using data from a web-based survey ad- depressive disorders)22 and a total GAD-7 score of 0–21 ministered to a panel representative of the general adult (scores of 5, 10, and 15 are the cutoff for mild, moderate, population of Spain, and compare characteristics between and severe anxiety symptoms, respectively). CC and matched non-CC cohorts. Work productivity over the previous 7 days was assessed by the 6-item Work Productivity and Activity Impairment (WPAI) questionnaire, which measured health-related ab- Methods senteeism (percentage of work time missed), presenteeism The National Health and Wellness Survey (NHWS) is a (percentage of impairment experienced while at work), cross-sectional survey conducted in numerous countries, overall work productivity loss (absenteeism plus pre- including Spain, using a self-administered, internet-based senteeism), and activity impairment (percentage of questionnaire addressed to a sample of the general pop- impairment in daily activities outside of work).23 All re- ulation aged ≥18 years. Post-hoc analyses have shown that spondents provided activity impairment data, but only full- the NHWS typically captures data from ≈0.02% of adults or part-time respondents provided work-related data. from a given country. The NHWS protocol and questionnaire Sleep problems (self-reported or physician-diagnosed) in were approved by the Pearl Institutional Review Board. the previous 12 months were assessed via direct questions to Potential NHWS respondents were recruited through an the respondents. To assess healthcare utilization, respon- existing, general-purpose (non-healthcare-specific) web- dents were asked to report any-cause hospitalizations, and based panel. All panelists explicitly agreed to be panel any visits to HCPs, relevant specialists, and emergency members. A targeted quota sampling procedure was im- rooms (ER) during the previous 6 months. plemented to ensure that the sex and age of the NHWS sample were representative of the Spanish adult population. Cohort selection Inclusion criteria for the NHWS were age ≥18 years, the ability to read and write in Spanish, and informed consent. NHWS respondents were asked questions pertaining to CC The 2020 Spanish NHWS was conducted between 30 (defined in the survey as daily cough lasting >8 weeks, December 2019 and 20 April 2020, using a Spanish-language according to the European Respiratory Society guidelines survey version translated from original and reviewed by expert available at the time of study design and data collection4). physicians. The covariates captured by the NHWS included Individuals reporting that they had experienced CC at any demographic variables, general health characteristics, health point in their lifetime or during the previous 12 months status, experienced and/or diagnosed comorbidities, depres- formed the lifetime and 12-month CC cohorts, respectively. sion and anxiety symptoms, work productivity, and healthcare Individuals reporting CC diagnosed by a physician in their utilization. All information was self-reported by respondents, lifetime or the previous 12 months formed the diagnosed and no clinical charts or medical reports were reviewed. lifetime and diagnosed 12-month CC cohorts, respectively. Comorbidity burden was assessed using the Charlson comorbidity index (CCI),17,18 which represents a weighted sum of multiple comorbid conditions predictive of mortality Statistical analysis (higher scores indicate a greater comorbidity burden). The Descriptive data are reported as means (± standard deviation) for original CCI, plus a modified version excluding chronic quantitative variables, and frequencies for categorical variables. obstructive pulmonary disease (COPD), a condition fre- The prevalence of CC was calculated using the total number of quently associated with CC, were used. NHWS respondents as the denominator. The weighted Domı́nguez-Ortega et al. 3 prevalence of CC was estimated using post-stratification The mean respondent age was 46.6 (±15.0) years; re- Horvitz-Thompson sampling weights based on the age spondents were generally well distributed across age cat- and sex distribution in Spain according to data from the egories (Table 1), except the ≥75-year age category, which US Census’ International Database (https://www.census. had 131 respondents (1.9%). Most (62.1%) respondents gov/programs-surveys/international-programs/about/ were employed. Current, former and never smokers ac- idb.html#). The weighted prevalence of CC was pro- counted for 31.1%, 31.2%, and 37.7% of respondents. One- jected to the Spanish adult population to estimate the third (34.3%) of respondents were overweight (body mass number of Spanish adults affected by CC. index [BMI] 25 to <30 kg/m2), and 14.6% were obese To compare outcomes, respondents of the 12-month CC (BMI ≥30 kg/m2). cohort were matched 1:3 to NHWS respondents without CC in the previous 12 months using a propensity score calculated using age, sex, modified CCI score, marital status, household Chronic cough prevalence income, and a marital status/household income interaction term. These variables were selected to account for differences During their lifetime, CC was experienced by 579 re- in healthcare and general health status between adults of spondents (8.2%), and 233 (3.2%) reported physician- different sexes, ages, and socioeconomic status, and to ac- diagnosed CC (Table 1). CC during the previous count for comorbid conditions not associated with CC. 12 months was experienced by 389 respondents (5.5%), and Chi-squared tests were used to compare proportions, and 171 (2.4%) had physician-diagnosed CC during this period. Student t-tests were used to compare continuous variables. A physician diagnosis of CC was received by 38.5% and Two-tailed p-values of <0.05 were considered statistically 44.0% of respondents who experienced CC during their significant. Statistical analyses were performed using SPSS lifetime or the previous 12 months, respectively (Table 1). (v23), R (v4.0.2), and SAS (v9.4). For lifetime CC, the weighted prevalence of experi- encing CC was 8.2%, estimated to affect ≈3.35 million adults in Spain, and of being diagnosed with CC was 3.2%, Results estimated to affect ≈1.30 million (Table 2). For 12-month CC, the weighted prevalence of experiencing CC was 5.5%, Sample estimated to affect ≈2.2 million Spanish adults, and of being The NHWS Spanish sample included 7074 respondents, diagnosed with CC was 2.4%, affecting ≈1 million. The 3487 males (49.3%) and 3587 females (50.7%) (Table 1). weighted prevalence of 12-month CC was slightly higher in Table 1. Unweighted prevalence of chronic cough (experienced or diagnosed by a physician) in Spain. Lifetime chronic cough 12-month chronic cough a a b NHWS respondents (n) EXP, n (%) DIA, n (%) DIA/EXP (%) EXP, n (%)a DIA, n (%)a DIA/EXP (%)b Total (7074) 579 (8.2) 223 (3.2) 38.5 389 (5.5) 171 (2.4) 44.0 Gender Female (3587) 320 (8.9) 123 (3.4) 38.4 222 (6.2) 97 (2.7) 43.7 Male (3487) 259 (7.4) 100 (2.9) 38.6 167 (4.8) 74 (2.1) 44.3 p-value 0.022 0.177 0.010 0.111 Age (years) 18–29 (1109) 115 (10.4) 35 (3.2) 30.4 64 (5.8) 26 (2.3) 40.6 30–39 (1289) 133 (10.3) 40 (3.1) 30.1 81 (6.3) 28 (2.2) 34.6 40–49 (1634) 124 (7.6) 51 (3.1) 41.1 92 (5.6) 42 (2.6) 45.7 50–64 (1865) 142 (7.6) 70 (3.8) 49.3 105 (5.6) 56 (3.0) 53.3 65–74 (1046) 53 (5.1) 21 (2.0) 39.6 38 (3.6) 15 (1.4) 39.5 ≥75 (131) 12 (9.2) 6 (4.6) 50.0 9 (6.9) 4 (3.1) 44.4 p-value <0.001 0.181 0.098 0.171 Smoking status Current/former (4409) 419 (9.5) 169 (3.8) 40.3 293 (6.6) 135 (3.1) 46.1 Never (2665) 160 (6.0) 54 (2.0) 33.8 96 (3.6) 36 (1.4) 37.5 p-value <0.001 <0.001 <0.001 <0.001 DIA: diagnosed by a physician; EXP: experienced; NA: not available; NHWS: National Health and Wellness Survey. a Values indicate, by variable, the number and percentage (in brackets) of cohort members with chronic cough; percentages are based on the number of NHWS respondents with that variable. b Values indicate the percentage of DIA members within the EXP cohort. 4 Chronic Respiratory Disease women than in men (6.0% vs. 4.9%; p = 0.049), but no versus 32.6% of cohort members were overweight, and other significant between-gender differences were shown 16.7% versus 13.3% were obese. The percentage of em- (Table 2). ployed respondents was similar between the 12-month The non-weighted (Table 1) and weighted (Table 2) CC cohort (65.8%) and the matched non-CC cohort prevalence of lifetime and 12-month CC were significantly (61.7%). (p < 0.001) higher in current/former smokers than in never- Smoking status differed significantly (p < 0.001) be- smokers. In the cohorts self-reporting CC, more current/ tween the 12-month CC and non-CC cohorts. Current former smokers than non-smokers had physician-diagnosed smokers accounted for 50.1% versus 30.9% of the 12- CC (Table 1). month CC and non-CC cohorts, respectively, previous The weighted prevalence of lifetime CC differed sig- smokers for 25.2% versus 30.8%, and never smokers for nificantly between age categories (p<0.001), being higher in 24.7% versus 38.3%. those aged 18–39 and ≥75 years than in those aged 40–74 CCI scores indicated that comorbidities were signifi- years (≈10% vs. ≈7%; Table 2). In the cohort self-reporting cantly (p < 0.05) more common in the 12-month CC than in lifetime CC, physician-diagnosed CC was less common in the non-CC cohort (Table 3). Digestive disorders (specifi- individuals aged 18–39 years than in those aged >40 years cally gastroesophageal reflux disease/acid reflux, heart- (≈30% vs. ≈40–50%; Table 1). burn), neurologic disease, respiratory diseases (specifically allergies, asthma, COPD, chronic bronchitis), and rheu- matoid arthritis were significantly more common in the CC General health and comorbidities cohort than in the non-CC cohort. The cohort self-reporting CC in the past 12 months and the Respondents with past 12-month CC had significantly matched non-CC cohort of NHWS respondents had similar poorer health status than those in the non-CC cohort, as distributions of all covariates included in the propensity shown by their 3.8 and 3.3 point lower mean SF-12v2 PCS score algorithm. Mean ages in the 12-month CC and the and MCS scores, and 2.7–4.3 point lower mean domain matched non-CC cohorts were similar (45.1 and 45.3 years), scores (Table 4; all p < 0.001). as was the gender distribution (percentage of females: Depression and anxiety symptoms were more common 57.1% and 58.2%, respectively). There were no significant in the 12-month CC cohort than in the non-CC cohort, as differences between the 12-month CC and non-CC cohorts indicated by their significantly (p < 0.001) higher mean concerning weight-related parameters: mean BMI of 25.5 PHQ-9 and GAD-7 scores (Table 4). Significantly (p < (±5.2) kg/m2 versus 25.2 (±5.2) kg/m2 (p=0.394), 30.3% 0.001) more individuals in the CC cohort than in the non- Table 2. Weighted prevalence of chronic cough (experienced or diagnosed by a physician) in Spain. Lifetime chronic cough 12-month chronic cough Spanish population (n) EXP, n (%)a DIA, n (%)a EXP, n (%)a DIA, n (%)a Adults (40,663,139) 3,345,015 (8.2) 1,296,104 (3.2) 2,236,153 (5.5) 980,074 (2.4) Gender Female (20,767,414) 1,792,882 (8.6) 714,049 (3.4) 1,256,228 (6.0) 551,295 (2.7) Male (19,895,725) 1,552,133 (7.8) 582,055 (2.9) 979,925 (4.9) 428,779 (2.2) p-value 0.229 0.247 0.049 0.191 Age, years 18–29 (6,025,059) 652,365 (10.8) 196,785 (3.3) 351,639 (5.8) 145,701 (2.4) 30–39 (6,790,933) 719,511 (10.6) 215,580 (3.2) 437,452 (6.4) 149,101 (2.2) 40–49 (8,600,286) 634,864 (7.4) 261,698 (3.0) 470,306 (5.5) 214,450 (2.5) 50–64 (10,149,558) 779,304 (7.7) 382,797 (3.8) 577,281 (5.7) 306,462 (3.0) 65–74 (8,181,336) 464,631 (5.7) 192,074 (2.3) 337,713 (4.1) 135,911 (1.7) ≥75 (915,967) 94,340 (10.3) 47,170 (5.1) 61,762 (6.7) 28,449 (3.1) p-value <0.001 0.341 0.267 0.372 Smoking status Current/former (25,229,919) 2,428,625 (9.6) 964,116 (3.8) 1,696,750 (6.7) 764,453 (3.0) Never smoked (15,433,220) 916,390 (5.9) 331,988 (2.2) 539,403 (3.5) 215,621 (1.4) p-value <0.001 <0.001 <0.001 <0.001 DIA: diagnosed; EXP: experienced; NA: not available; NHWS: National Health and Wellness Survey. a Values indicate, by variable, the estimated number of adults in Spain with chronic cough and (in brackets) the weighted prevalence in the overall NHWS population. Domı́nguez-Ortega et al. 5 Table 3. Physician-diagnosed comorbidities during the previous 12 months. 12-month chronic cough Matched non-chronic cough Comorbidity-related outcome cohort (n = 389) cohort (n = 1164) p value CCI Original Mean score (SD) 0.77 (1.19) 0.47 (1.06) <0.001 No. of comorbidities, n (%)a 0 214 (55.0) 869 (74.7) <0.001 1 108 (27.8) 162 (13.9) ≥2 67 (17.2) 133 (11.4) CCI Modified (without COPD) Mean score (SD) 0.50 (1.05) 0.43 (1.01) 0.235 No. of comorbidities, n (%)a 0 277 (71.2) 893 (76.7) 0.042 1 69 (17.7) 149 (12.8) ≥2 43 (11.1) 122 (10.5) Comorbidity (responder nb), n (%)c Cancer, anyd (1539) 14 (3.6) 54 (4.7) 0.457 Digestive disease, anye (1530) 124 (32.4) 193 (16.8) <0.001 GERD or acid reflux (1544) 33 (8.5) 42 (3.6) <0.001 Heartburn (1544) 76 (19.6) 95 (8.2) <0.001 Liver disease (1535) 5 (1.3) 14 (1.2) 0.893 Cardiovascular disease, anyf (1540) 40 (10.4) 124 (10.7) 0.876 Neurological disease, anyg (1540) 14 (3.6) 14 (1.2) 0.005 Rheumatoid arthritis (1540) 24 (6.2) 39 (3.4) 0.025 Respiratory disease, anyh (1542) 267 (69.0) 391 (33.6) <0.001 Allergies, any (1542) 148 (38.2) 342 (29.6) 0.002 Asthma (1549) 66 (17.0) 78 (6.7) <0.001 Rhinitis (1541) 17 (4.4) 27 (2.3) 0.055 COPD (1544) 21 (5.4) 13 (1.1) <0.001 Chronic bronchitis (1543) 86 (22.3) 33 (2.7) <0.001 Emphysema (1544) 4 (1.0) 8 (0.7) 0.742 Chronic cough, diagnosed (1541) 171 (44.2) 10 (0.9) <0.001 CCI: Charlson Comorbidity Index; COPD: chronic obstructive pulmonary disease; GERD: gastroesophageal reflux disease; NHWS: National Health and Wellness Survey; SD: standard deviation. a Values in the cohort columns indicate the absolute number and percentage (in brackets) of cohort members with variable; percentages were calculated based on the total number of members in the cohort. b Provides the total number of responders to each question in both cohorts (maximum possible 1553); the remaining respondents declined to respond to this question. c Values in the cohort columns indicate the absolute number and percentage (in brackets) of cohort members with variable; percentages were based on the number of cohort members who responded to the specific comorbidity question. d Includes any tumor, leukemia, lymphoma, and metastatic solid tumors. e Includes GERD, acid reflux, heartburn, irritable bowel syndrome (with constipation and/or diarrhea), ulcerative colitis, and active/peptic stomach or duodenal ulcer. f Includes congestive heart failure, ischemic heart disease, cerebrovascular disease (transient ischemic attack, stroke), and peripheral vascular disease. g Includes dementia, current hemiplegia, multiple sclerosis, and Parkinson’s disease. h Includes asthma, chronic bronchitis, COPD, emphysema, hay fever, physician-diagnosed chronic cough and allergies. CC cohort had moderate-to-severe depression (39.9% vs. Activity/work impairment 21.6%) or moderate-to-severe anxiety symptoms (26.8% vs. 13.8%). As assessed by the WPAI, mean activity impairment in the Sleeping disorders (experienced or physician-diagnosed) previous 7 days was higher in the 12-month CC cohort than were significantly more common in the 12-month CC cohort in the non-CC cohort (38.5% [±28.2] vs. 25.1% [±27.4]; than in the non-CC cohort, as were insomnia, narcolepsy p<0.001). Among actively working cohort members, the (experienced only), sleep apnea, and other sleep difficulties CC cohort had significantly (p<0.001) higher mean per- (Table 5). centages of absenteeism (11.9% [±20.0] vs. 6.5% [±16.6]), 6 Chronic Respiratory Disease Table 4. Health status, depression, and anxiety in the previous 12 months. 12-month chronic cough Matched non-chronic cough Outcome, instrument cohort (n = 389) cohort (n = 1164) p value Health status, SF-12-v2 Physical Component Summary score, mean (SD) 47.68 (8.25) 51.44 (8.03) <0.001 Mental Component Summary score, mean (SD) 42.42 (9.14) 45.69 (9.31) <0.001 Domain scores, mean (SD) Physical functioning score 47.69 (9.79) 51.87 (8.40) <0.001 Physical role limitation score 43.79 (8.76) 47.94 (8.82) <0.001 Bodily pain score 45.07 (9.77) 49.20 (9.45) <0.001 General health score 46.35 (9.91) 49.25 (9.02) <0.001 Vitality score 49.80 (9.08) 52.51 (8.99) <0.001 Social functioning score 42.38 (9.10) 46.40 (9.23) <0.001 Emotional role limitations score 39.23 (10.75) 43.58 (10.82) <0.001 Mental health score 44.57 (9.09) 47.82 (8.71) <0.001 Health utility, SF-6D Utility score, mean (SD) 0.65 (0.11) 0.70 (0.12) <0.001 Depression, PHQ-9a Total score, mean (SD) 8.84 (6.64) 5.78 (5.88) <0.001 Classification of depression, n (%)a None–minimal (score 0–4) 118 (30.3) 624 (53.6) <0.001 Mild (score 5–9) 116 (29.8) 289 (24.8) Moderate (score 10–14) 84 (21.6) 136 (11.7) Moderately severe (score 15–19) 40 (10.3) 74 (6.4) Severe (score 20–27) 31 (8.0) 41 (3.5) Anxiety, GAD-7 Mean score (SD) 6.91 (5.30) 4.56 (4.53) <0.001 Classification of anxiety, n (%) a None (score 0–4) 145 (37.3) 660 (56.7) <0.001 Mild (score 5–9) 140 (36.0) 344 (29.6) Moderate (score 10–14) 66 (17.0) 110 (9.5) Severe (score 15–21) 38 (9.8) 50 (4.3) GAD-7: Generalized Anxiety Disorder-7 scale; NHWS: National Health and Wellness Survey; PHQ-9: Patient Health Questionnaire-9; SD: standard deviation; SF-12v2: Medical Outcomes 12-item Short Form Survey Instrument; SF-6D Short-Form Six-Dimension health index. a Values indicate the number and percentage (in brackets) of cohort members with that classification; percentages are based on the total number of NHWS respondents in the cohort. presenteeism (34.6% [±27.9] vs. 21.6% [±26.3]), and total Discussion work productivity impairment (38.4% [±31.1] vs. 24.1% [±28.7]) than the non-CC cohort. Given the lack of epidemiological studies of CC in Spain, this study, based on the NHWS cross-sectional survey, contributes novel data regarding the estimated prevalence of CC in adults in Spain and the significant health burden Healthcare utilization experienced by this population. In this study, the weighted During the previous 6 months, significantly (p<0.001) more prevalence of lifetime and 12-month CC was 8.2% and 12-month CC cohort than of non-CC cohort members used 5.5% respectively. These figures are in line with recently healthcare resources, including all-cause hospitalization, published NHWS surveys from the USA24 and Japan25 and visits to any HCP, relevant specialists, and ER using similar methodology, which reported a weighted (Figure 1). The CC cohort, relative to the non-CC cohort, prevalence of 12-month CC of 5.0%24 and 4.3%,25 had a significantly (p<0.001) higher mean number of respectively. visits to any HCP (7.6 [±9.9] vs. 4.1 [±4.7]), relevant Similar to prior CC studies,5,6 being a current/former specialists (4.4 [±5.8] vs. 2.3 [±3.2]), primary-care phy- smoker was consistently associated with an increased sicians (2.6 [±3.3] vs. 1.5 [±1.9]), and ER (1.4 [±2.1] vs. prevalence of CC. A significantly higher weighted preva- 0.7 [±2.2]). lence of CC in females than in males was shown only in the Domı́nguez-Ortega et al. 7 Table 5. Sleep disorders in the previous 12 months. 12-month chronic Matched non-chronic Both cohorts (n = 1553) cough cohort (n = 389) cough cohort (n = 1164) Sleep disorder Responded to question (n)a n (%)b n (%)b p value Experienced in the previous 12 months Any 1545 286 (73.6) 513 (44.3) <0.001 Specific disorders 1545 Insomnia 1545 186 (48.1) 296 (25.6) <0.001 Narcolepsy 1545 9 (2.3) 5 (0.4) 0.002 Sleep apnea 1545 59 (15.2) 79 (6.8) <0.001 Other sleep difficulties 1545 154 (39.8) 243 (21.0) <0.001 Diagnosed by a physician in the previous 12 months Any 1540 143 (37.1) 209 (18.1) <0.001 Specific disorders Insomnia 1539 99 (25.8) 138 (11.9) <0.001 Narcolepsy 1540 1 (0.3) 1 (0.1) 1.000 Sleep apnea 1539 35 (9.1) 58 (5.0) 0.004 Other sleep difficulties 1542 51 (13.2) 71 (6.1) <0.001 a The remaining cohort members declined to answer these questions. b Percentage based on the number of members of that cohort who responded to that specific question. underdiagnosis in this age group, who are less often in contact with the health system and less likely to seek medical care. Matching the 12-month CC cohort to a non-CC cohort allowed this study to explore differences in these pop- ulations while controlling for differences in age, gender, CCI, marital status, and income. Comorbidities were more common in individuals with CC during the previous 12 months than those without CC, both overall and with regard to particular types of comorbidities, including dis- eases affecting the digestive, neurologic, and respiratory systems, and rheumatoid arthritis. Potential relationships Figure 1. Use of all-cause healthcare resources in the previous have been suggested between CC and many other condi- 6 months. Any specialist is a summary of any of a group of specialties most relevant to persons with chronic cough, tions, primarily gastrointestinal symptoms5,6,26–31 and re- specifically general/family practitioner, internist, allergist, spiratory conditions,6,11,26–31 as well as sleep apnea32–34 cardiologist, gastroenterologist, neurologist, nurse practitioner/ and laryngeal hypersensititivy.35 physician assistant, otolaryngologist, psychiatrist, psychologist/ Individuals with CC during the previous 12 months had therapist, and pulmonologist. All p < 0.001. poorer overall, physical, and mental health status than those without CC, as shown by their significantly lower SF-12v2 12-month CC cohort; similar between-gender differences PCS, MCS, and domain scores. Similar results have been have been reported in other studies,9–11, 14,26 including the shown in other studies. For example, the condition was NHWS survey from the USA.24 Age significantly influ- burdensome and affected daily living measures in CC pa- enced only the weighted prevalence of lifetime CC. Of tients in a US study,10 adversely affected the QoL of 96% of interest, although respondents aged 18–39 years had higher CC patients in a European survey,11 and decreased QoL rates of lifetime CC than those aged 40–74 years, rates of scores in CC patients relative to healthy volunteers in a diagnosed CC were lower in younger individuals. Studies Chinese study12 and matched non-CC respondents in frequently describe CC as being more common in middle- Japanese13 and Korean15 surveys. aged populations (aged 40–65 years),6,10,11,14,26 but this Furthermore, depression, anxiety, and sleep problems study suggests that younger populations may have a higher were significantly more common in individuals with CC in relative prevalence of CC than previously described. This the previous 12 months than in those without CC, as has finding could be due in part to a higher percentage of been previously shown. Depression was significantly 8 Chronic Respiratory Disease associated with CC in elderly patients in Korea,16 depres- the survey. Respondents were asked about smoking habits, sion, anxiety, and sleep disturbances more common in but not specifically ACEi use, therefore, its impact on CC patients with CC than in matched non-CC respondents in prevalence could not be analyzed. The calculation of CCI Japan,13 and depression, anxiety, and disturbed sleep were may be slightly low, as the NHWS does not capture the each reported in 55–77% of patients with CC in the UK.14 comorbidities of moderate/severe liver disease and para- These aspects must be taken into account when studying plegia (usually included in the definition of hemiplegia). patients with CC. Finally, it should be noted that many of the outcomes that In addition, CC impacted the daily life and work of were worse in patients with CC (health status, depression, individuals to a significant extent. Relative to individuals anxiety, sleep disorders, healthcare utilization, and activity without CC, those with 12-month CC had 1.4-fold higher and work-productivity impairment) may be influenced by impairment of daily activities and, in individuals who were existing comorbidities, beyond those accounted for in the currently working, a 1.6-fold higher impairment of total modified CCI used in matching, rather than by CC itself. work productivity. This confirms the findings of other Future research is warranted specifically in populations with studies, in which CC significantly impaired daily activities refractory CC or unexplained CC. Patients with these and work productivity by 1.4-and 1.5-fold in Japan,13 sig- conditions constitute a subset of the overall CC population nificantly affected the employment, social life, hobbies, and with a greater burden of cough, and exploration of outcomes holidays of Swedish patients,36 and affected the daily ac- in this population where cough has been adequately diag- tivities and social life of >60% of UK patients.14 nosed and managed may reveal additional insights on the Finally, healthcare utilization for any reason was more burden of cough not related to underlying conditions. common in patients with than without CC, including a 1.2-. Further studies evaluating the influence of comorbidities 1.7-, and 2.0-fold higher proportion who visited any HCP or and other potential confounding factors on these outcomes specialist, visited the ER, or who were hospitalized, re- would be of value. spectively. Likewise, the Japanese survey showed that CC patients had higher numbers of visits to HCPs and the ER than matched non-CC respondents13 Conclusion The incremental burden of CC on comorbidities, health This study showed that CC is frequent in Spanish adults, status, depression, anxiety, sleep problems, daily life and with 8.2% (≈3.2 million) and 5.5% (≈2.2 million) experi- work activities, and healthcare utilization, indicate that there encing CC lifetime or in the past 12 months, but it appears is an unmet need to improve overall health in patients with underdiagnosed, especially in younger age groups. Re- CC. Furthermore, CC may be difficult to treat successfully spondents afflicted with CC experience a significant health and is often persistent.11,37,38 Given the estimated preva- and QoL burden relative to those without CC, that might be lence of CC in Spain and other countries, CC represents a related to CC itself, and also to comorbidities not included considerable burden to healthcare payers, and, when work in the modified CCI. impairment is included, to society. Better interventions and targeted treatments are needed to manage CC,38–40 which Acknowledgements have the potential to improve the overall health status and Medical writing assistance was provided by Katherine A. Lyseng- QoL of patients, as well as reduce CC-related depression, Williamson, Content Ed Net, and funded by MSD Spain. anxiety, sleep problems, daily living, and work impairment, and healthcare utilization. Author contributions The current study has limitations that should be con- sidered. In common with other internet-based, patient- All authors have contributed significantly to the conception, de- reported surveys, the NHWS sample likely under- sign, or acquisition of data, or analysis and interpretation of data. represents people without access to or familiarity with All authors have participated in drafting, reviewing, and/or re- electronic media, including the elderly, institutionalized vising the manuscript and have approved its submission. individuals, and those with severe comorbidities and dis- abilities that may make access to, and ability to participate Declaration of conflicting interests in, online survey panels more difficult. As all information The author(s) declared the following potential conflicts of interest was self-reported by participants, it may be subject to recall with respect to the research, authorship, and/or publication of this and self-representation bias and cannot be verified. article: JDO has received payment for consultancy services and Our sample can be considered representative of the adult presentations from Leti Pharma, Mundipharma, AstraZeneca, Spanish population aged 18–75 years, but not of those Chiesi, Novartis, GSK, MSD, Sanofi, and Teva. VP has received aged ≥75 years, who accounted for only 1.9% of NHWS payment for consultancy services from Sanofi, for presentations respondents. Smoking or the use of angiotensin-converting from AstraZeneca, Chiesi, MSD and Boehringer Ingelheim, enzyme inhibitors (ACEi) did not exclude participation in meeting travel assistance from Novartis and research grants from Domı́nguez-Ortega et al. 9 AstraZeneca, Chiesi and Menarini. VWL and AM are full time Clin Immunol Pract 2021; 9: 1624–1637, e1610. DOI: 10. employees at Kantar Health LLC, New York, NY, USA. EF, JB and 1016/j.jaip.2020.11.018 JS are full-time employees of Merck Sharp & Dohme Corp., a 11. Chamberlain SA, Garrod R, Douiri A, et al. The impact of subsidiary of Merck & Co., Inc., Rahway, NJ, USA and share- chronic cough: a cross-sectional European survey. Lung holders in Merck & Co., Inc., Rahway, NJ, USA. LCC and MSJ are 2015; 193: 401–408. DOI: 10.1007/s00408-015-9701-2 full-time employees of MSD Spain. 12. Ma W, Yu L, Wang Y, et al. Changes in health-related quality of life and clinical implications in Chinese patients with chronic Funding cough. Cough 2009; 5: 7. DOI: 10.1186/1745-9974-5-7 13. Kubo T, Tobe K, Okuyama K, et al. Disease burden and The author(s) disclosed receipt of the following financial support quality of life of patients with chronic cough in Japan: a for the research, authorship, and/or publication of this article: This population-based cross-sectional survey. BMJ Open Respir study was funded by Merck Sharp & Dohme Corp., a subsidiary of Res 2021; 8: e000764. DOI: 10.1136/bmjresp-2020-000764 Merck & Co., Inc., Rahway, NJ, USA. 14. Everett CF, Kastelik JA, Thompson RH, et al. Chronic persistent cough in the community: a questionnaire survey. References Cough 2007; 3: 5. DOI: 10.1186/1745-9974-3-5 1. Finley CR, Chan DS, Garrison S, et al. What are the most 15. Won HK, Lee JH, An J, et al. Mechanisms and rationale for common conditions in primary care? Systematic review. Can targeted therapies in refractory and unexplained chronic Fam Physician 2018; 64: 832–840. cough: the Korean National Health and Nutrition Examina- 2. Song WJ, Chang YS, Faruqi S, et al. The global epidemiology tion survey 2010-2016. Allergy Asthma Immunol Res 2020; of chronic cough in adults: a systematic review and meta- 12: 964–979. DOI: 10.4168/aair.2020.12.6.964 analysis. Eur Respir J 2015; 45: 1479–1481. DOI: 10.1183/ 16. Sohn KH, Song WJ, Kim SH, et al. Chronic cough, not 09031936.00218714 asthma, is associated with depression in the elderly: a 3. Pratter MR, Brightling CE, Boulet LP, et al. An empiric community-based population analysis in South Korea. Ko- integrative approach to the management of cough: ACCP rean J Intern Med 2019; 34: 1363–1371. DOI: 10.3904/kjim. evidence-based clinical practice guidelines. Chest 2006; 129: 2018.187 222s–231s. DOI: 10.1378/chest.129.1_suppl.222S 17. Charlson ME, Pompei P, Ales KL, et al. A new method of 4. Morice AH, Fontana GA, Belvisi MG, et al. ERS guidelines classifying prognostic comorbidity in longitudinal studies: on the assessment of cough. Eur Respir J 2007; 29: development and validation. J Chronic Dis 1987; 40: 373–383. 1256–1276. DOI: 10.1183/09031936.00101006 DOI: 10.1016/0021-9681(87)90171-8 5. Ford AC, Forman D, Moayyedi P, et al. Cough in the 18. Witt EA and Goren A. A comparison of the utility of variants community: a cross sectional survey and the relationship to of the charlson comorbidity index (CCI) in predicting patient- gastrointestinal symptoms. Thorax 2006; 61: 975–979. DOI: reported health outcomes. Value in Health 2014; 17: A194, 10.1136/thx.2006.060087 DOI: 10.1016/j.jval.2014.03.1134. 6. Lätti AM, Pekkanen J and Koskela HO. Defining the risk 19. Maruish ME. User’s manual for the SF-36v2 Health Survey. factors for acute, subacute and chronic cough: a cross- Lincoln, RI: Quality Metric Incorporated, 2011. sectional study in a Finnish adult employee population. 20. Kroenke K, Spitzer RL and Williams JB. The PHQ-9: validity BMJ Open 2018; 8: e022950. DOI: 10.1136/bmjopen-2018- of a brief depression severity measure. J Gen Intern Med 2001; 022950 16: 606–613. DOI: 10.1046/j.1525-1497.2001.016009606.x 7. Fujimura M. Frequency of persistent cough and trends in 21. Spitzer RL, Kroenke K, Williams JB, et al. A brief measure seeking medical care and treatment-results of an internet for assessing generalized anxiety disorder: the GAD-7. Arch survey. Allergol Int 2012; 61: 573–581. DOI: 10.2332/ Intern Med 2006; 166: 1092–1097. DOI: 10.1001/archinte. allergolint.11-OA-0368 166.10.1092 8. Kang MG, Song WJ, Kim HJ, et al. Point prevalence and 22. Manea L, Gilbody S and McMillan D. Optimal cut-off epidemiological characteristics of chronic cough in the score for diagnosing depression with the Patient Health general adult population: the Korean National Health and Questionnaire (PHQ-9): a meta-analysis. CMAJ 2012; 184: Nutrition Examination Survey 2010-2012Point prevalence E191–E196. DOI: 10.1503/cmaj.110829 and epidemiological characteristics of chronic cough in the 23. Reilly MC, Zbrozek AS and Dukes EM. The validity and general adult population. Medicine (Baltimore) 2017; 96: reproducibility of a work productivity and activity impair- e6486. DOI: 10.1097/md.0000000000006486 ment instrument. Pharmacoeconomics 1993; 4: 353–365. 9. Desalu OO, Salami AK and Fawibe AE. Prevalence of cough DOI: 10.2165/00019053-199304050-00006 among adults in an urban community in Nigeria. West Afr J 24. Meltzer EO, Zeiger RS, Dicpinigaitis P, et al. Prevalence and Med 2011; 30: 337–341. burden of chronic cough in the United States. J Allergy Clin 10. Zeiger RS, Schatz M, Hong B, et al. Patient-reported burden Immunol Pract 2021; 9: 4037–4044. DOI: 10.1016/j.jaip. of chronic cough in a managed care organization. J Allergy 2021.07.022 10 Chronic Respiratory Disease 25. Tobe K, Kubo T, Okuyama K, et al. Web-based survey to 33. Sundar KM and Daly SE. Chronic cough and OSA: an un- evaluate the prevalence of chronic and subacute cough and derappreciated relationship. Lung 2014; 192: 21–25. DOI: 10. patient characteristics in Japan. BMJ Open Respir Res 2021; 1007/s00408-013-9534-9 8: e000832. DOI: 10.1136/bmjresp-2020-000832 34. Good JT Jr., Rollins DR, Kolakowski CA, et al. New insights 26. Campi G, Noale M, Fabbrizzi A, et al. The demographic and in the diagnosis of chronic refractory cough. Respir Med clinical characteristics of an Italian population of adult out- 2018; 141: 103–110. DOI: 10.1016/j.rmed.2018.06.024 patients with chronic cough. Aging Clin Exp Res 2020; 32: 35. Sundar KM, Stark AC, Hu N, et al. Is laryngeal hypersen- 741–746. DOI: 10.1007/s40520-019-01464-4 sitivity the basis of unexplained or refractory chronic cough? 27. Kanemitsu Y, Niimi A, Matsumoto H, et al. Gastroesophageal ERJ Open Res 2021; 7: 00793–02020. DOI: 10.1183/ dysmotility is associated with the impairment of cough- 23120541.00793-2020 specific quality of life in patients with cough variant 36. Ternesten-Hasséus E, Larsson S and Millqvist E. Symptoms asthma. Allergol Int 2016; 65: 320–326. DOI: 10.1016/j.alit. induced by environmental irritants and health-related quality 2016.02.014 of life in patients with chronic cough-A cross-sectional study. 28. Shirai T, Mikamo M, Tsuchiya T, et al. Real-world effect of Cough 2011; 7: 6. DOI: 10.1186/1745-9974-7-6 gastroesophageal reflux disease on cough-related quality of 37. Koskela HO, Lätti AM and Purokivi MK. Long-term prognosis life and disease status in asthma and COPD. Allergol Int 2015; of chronic cough: a prospective, observational cohort study. BMC 64: 79–83. DOI: 10.1016/j.alit.2014.08.001 Pulm Med 2017; 17: 146. DOI: 10.1186/s12890-017-0496-1 29. Morice A. Chronic cough: epidemiology. Chron Respir Dis 38. Kardos P, Blaiss M and Dicpinigaitis P. Addressing unmet 2008; 5: 43–47. DOI: 10.1177/1479972307084252 needs for diagnosis and management of chronic cough in the 30. Chung KF and Pavord ID. Prevalence, pathogenesis, and primary care setting. Postgrad Med 2021; 133: 481–488. causes of chronic cough. Lancet 2008; 371: 1364–1374. DOI: DOI: 10.1080/00325481.2021.1914944 10.1016/s0140-6736(08)60595-4 39. Mazzone SB and McGarvey L. Mechanisms and rationale for 31. D’Urzo A and Jugovic P. Chronic cough. Three most common targeted therapies in refractory and unexplained chronic cough. Clin causes. Can Fam Physician 2002; 48: 1311–1316. Pharmacol Ther 2021; 109: 619–636. DOI: 10.1002/cpt.2003 32. Gouveia CJ, Yalamanchili A, Ghadersohi S, et al. Are chronic 40. Muccino D and Green S. Update on the clinical development cough and laryngopharyngeal reflux more common in ob- of gefapixant, a P2X3 receptor antagonist for the treatment of structive sleep apnea patients? Laryngoscope 2019; 129: refractory chronic cough. Pulm Pharmacol Ther 2019; 56: 1244–1249. DOI: 10.1002/lary.27557 75–78. DOI: 10.1016/j.pupt.2019.03.006