Clinical validation of the Rapid Cough Questionnaire across diverse lung diseases Authors: Hyeon-Kyoung Koo, Tai Joon An, Hyonsoo Joo, Chin Kook Rhee, Yee Hyung Kim, Sung-Kyoung Kim, Kyung Hoon Min, Deog Kyeom Kim, Jong Wook Shin, Hyoung Kyu Yoon, Seung Hun Jang, Yong Bum Park, Jin Woo Kim, Ji-Yong Moon, Hyeon-Kyoung Koo Published: 2026 ABSTRACT AbstractBackgroundThe Rapid Cough Questionnaire (RCQ) is a simplified three-item instrument developed as a pragmatic alternative to the Leicester Cough Questionnaire (LCQ). However, its performance across diverse respiratory diseases, particularly structural lung diseases, has not been sufficiently validated. This study evaluated the performance of the RCQ across major respiratory disease subgroups.MethodsA total of 300 adult patients with chronic cough were prospectively enrolled from multiple respiratory centres. Participants completed five cough-specific instruments: Numeric Rating Scale (NRS), Cough Symptom Score (CSS), COugh Assessment Test (COAT), LCQ and RCQ. Construct validity was evaluated using correlation analyses across disease subgroups, including chronic obstructive pulmonary disease, asthma, idiopathic pulmonary fibrosis (IPF) and bronchiectasis.ResultsThe RCQ showed strong correlations with the LCQ (r=0.93), COAT (r=−0.75), CSS (r=−0.62) and NRS (r=−0.59). These relationships remained robust among all aetiologies, with the correlation coefficients between RCQ and LCQ exceeding 0.88 in each subgroup. Bland-Altman plots and an intraclass correlation analysis indicated a high level of concordance between the RCQ and LCQ scores. The RCQ aligned closely with multidimensional symptom burden and maintained strong correlation with the LCQ even for IPF, where LCQ correlations with unidimensional tools (NRS or CSS) were attenuated.ConclusionsThe RCQ demonstrated strong cross-sectional concordance with the LCQ across heterogeneous respiratory conditions. Its brevity and consistent performance support its suitability in routine clinical practice and large-scale or longitudinal research settings. Further studies are warranted to evaluate its responsiveness and longitudinal performance. WHAT IS ALREADY KNOWN ON THIS TOPIC Patient-reported outcome measures, such as the Leicester Cough Questionnaire (LCQ), are widely used to assess chronic cough, but their length limits routine clinical use. Patient-reported outcome measures, such as the Leicester Cough Questionnaire (LCQ), are widely used to assess chronic cough, but their length limits routine clinical use. WHAT THIS STUDY ADDS The Rapid Cough Questionnaire (RCQ), a brief three-item instrument, demonstrated strong concordance with the LCQ across diverse respiratory diseases, including structural lung diseases. The Rapid Cough Questionnaire (RCQ), a brief three-item instrument, demonstrated strong concordance with the LCQ across diverse respiratory diseases, including structural lung diseases. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY These findings support the use of the RCQ as a pragmatic tool for assessing cough burden in routine clinical practice and large-scale studies, facilitating efficient and scalable evaluation of chronic cough. These findings support the use of the RCQ as a pragmatic tool for assessing cough burden in routine clinical practice and large-scale studies, facilitating efficient and scalable evaluation of chronic cough. Introduction Chronic cough is a common and burdensome symptom encountered in patients with various pulmonary conditions,1 including chronic obstructive pulmonary disease (COPD),25 asthma,68 idiopathic pulmonary fibrosis (IPF)913 and bronchiectasis.1416 In addition to the immediate physical discomfort it causes, chronic cough can significantly impair quality of life and contribute to pain, fatigue, sleep disturbance, emotional distress and social isolation.17 Owing to its multifactorial nature, accurate and practical assessment of cough severity and related burdens is essential in both clinical practice and research settings. Several clinical assessment instruments have been developed to quantify the severity and impact of chronic cough. The Leicester Cough Questionnaire (LCQ) is a widely validated instrument that assesses the multidimensional burden of chronic cough across physical, psychological and social domains.17 18 However, its length and complexity may limit its applicability in routine practice. To address these limitations, simpler tools such as the Numeric Rating Scale (NRS)19 and Cough Symptom Score (CSS)20 have been used, although they may not fully reflect the multidimensional aspects of chronic cough. Recently, the Rapid Cough Questionnaire (RCQ) was developed as a brief alternative designed to preserve the multidimensional assessment strength of the LCQ while improving its feasibility in routine clinical settings.21 The RCQ has demonstrated promising measurement properties and clinical relevance for use with patients with chronic cough and non-structural lung diseases. However, the psychometric validity and reliability of the RCQ across a broad range of diseases, particularly patients with structural lung diseases, have not been sufficiently established. As the symptom burden and impact of cough on the quality of life may vary according to the underlying aetiology of the condition, it is essential to evaluate the RCQ performance for a range of respiratory conditions. This study aimed to validate the RCQ and assess its performance relative to the LCQ in a heterogeneous cohort of patients with chronic cough from diverse respiratory backgrounds. The RCQ scores were compared with the results from established instruments, and their interrelationships were evaluated across disease-specific subgroups. This approach aimed to determine the RCQ generalisability and clinical utility as a brief and pragmatic measure of cough-related burden regardless of the underlying aetiology. Study design and methods Study population Adult patients (≥18 years old) with chronic cough who visited participating respiratory centres in the Republic of Korea during the study period between November 2021 and October 2023 were prospectively enrolled.22 The study population comprised patients with various underlying respiratory conditions. Diagnoses were determined by respiratory specialists at each participating centre based on clinical evaluation and relevant investigations, following established guideline-based criteria where applicable. COPD was diagnosed based on post-bronchodilator spirometry demonstrating a forced expiratory volume in 1 s to forced vital capacity ratio <0.723; asthma was diagnosed based on clinical features supported by evidence of variable airflow limitation24; IPF was diagnosed according to international guideline criteria with radiologic and multidisciplinary confirmation25; and bronchiectasis was confirmed by chest CT.26 27 Patients were classified according to the primary underlying diagnosis considered to be the main contributor to cough. As chronic cough may have multiple coexisting causes, cases with overlapping conditions were assigned to a single primary category based on overall clinical judgement. Patients in whom a primary aetiology could not be clearly determined were excluded from the analysis. The ‘other’ group included patients without structural lung disease, such as those with upper airway cough syndrome, eosinophilic bronchitis, gastro-oesophageal reflux disease, drug-induced cough and refractory unexplained chronic cough. The target sample size was determined based on methodological considerations for validation studies28 29 and empirical precedent.21 Previous literature suggests that sample sizes of 5–10 participants per item or >200 participants are generally sufficient to ensure stable estimation of measurement properties. The RCQ was originally developed in 255 patients with chronic cough without structural lung disease.21 In the present study, we included patients with structural lung diseases and selected a slightly larger sample size to ensure robust evaluation across heterogeneous disease groups. Patients and public involvement Patients and the public were not involved in the design, conduct, reporting or dissemination plans of this research. Participants were involved solely as respondents to study questionnaires. No patient or public input was sought in the development of the research question, outcome measures or interpretation of results. Cough questionnaires All the participants completed five cough-specific questionnaires: NRS,19 CSS,20 LCQ,17 18 Cough Assessment Test (COAT)30 and RCQ.21 The NRS is a self-reported measure that quantifies cough severity on a scale of 0–10, with higher scores indicating greater severity.19 The CSS evaluates cough severity based on two items assessing daytime and nighttime symptoms, each rated on a 0 to 5 Likert scale, with higher scores reflecting more severe coughing.20 The COAT is a concise, 5-item questionnaire evaluating cough frequency (COAT1), limitations on daily activities (COAT2), sleep disturbance (COAT3), fatigue (COAT4) and hypersensitivity to irritants (COAT5).22 The LCQ is a validated 19-item instrument designed to assess the impact of cough on the health-related quality of life across three domains: physical, psychological and social. Each item is rated on a 7-point Likert scale, with domain scores ranging from 1 to 7 and a total score between 3 and 21, with lower scores indicating greater impairment.17 18 The physical domain has eight items: chest or stomach pain (LCQ1), bothersome phlegm (LCQ2), tiredness (LCQ3), hypersensitivity to irritants (LCQ9), sleep disturbance (LCQ10), frequency of coughing bouts (LCQ11), voice hoarseness (LCQ14) and lack of energy (LCQ15). The psychological domain has seven items: feeling the cough cannot be controlled (LCQ4), embarrassment (LCQ5), anxiety (LCQ6), frustration (LCQ12), feeling fed-up (LCQ13), serious illness concern (LCQ16) and worries about others’ thoughts (LCQ17). The social domain has 4 items: interference with work or daily life (LCQ7), reduced enjoyment of life (LCQ8), interruptions during conversations or phone calls (LCQ18) and annoyance from family or close contacts (LCQ19). The RCQ consists of three items derived from the LCQ: tiredness (LCQ3: physical domain), feeling fed-up (LCQ13: psychological domain) and annoyance with partners/family/friends (LCQ19: social domain) (online supplemental figure S1).21 Statistical analyses Continuous variables are described as means with SD or medians with IQRs, and categorical variables are presented as counts and percentages. Inter-instrument agreement between the LCQ and RCQ was evaluated using the intraclass correlation coefficient (ICC). A two-way mixed-effects model with absolute agreement and a single-measure unit was applied, as recommended, to evaluate the agreement between two fixed measurement instruments. The ICCs were calculated for the overall cohort and further stratified by disease aetiology with 95% CIs. The minimal clinically important difference (MCID) for the RCQ was derived using a distribution-based approach and was defined as one-half of the SD of the baseline scores. Associations among the NRS, CSS, LCQ, COAT and RCQ scores were assessed using Spearman’s rank correlation. Correlation matrices were constructed using Spearman’s rank correlation, and only significant correlations with a coefficient ≥0.3 are presented.31 Correlation matrices and violin plots were generated using corrplot and vioplot packages, respectively. All statistical analyses were performed using R software (V.4.4.2; R Foundation, Vienna, Austria). Results Baseline characteristics A total of 300 patients were enrolled in the study. Their mean age was 62.5±15.3 years; 50.8% were male. Among the various aetiologies of chronic cough, asthma was the most prevalent (32.3%), followed by bronchiectasis (19.0%), COPD (17.3%) and IPF (11.3%). The mean scores for the NRS, CSS, COAT, LCQ and RCQ were 4.79±2.42, 3.96±1.83, 7.37±4.79, 13.9±4.2 and 13.8±4.2, respectively. The comparisons of demographics and cough-related measures by disease group are summarised in table 1. The mean total LCQ scores stratified by aetiology were as follows: COPD 15.7±3.6, asthma 13.1±4.1, IPF 13.9±4.2, bronchiectasis 15.0±4.3 and other conditions 12.6±3.8; the corresponding RCQ scores were 15.4±3.5, 12.6±4.3, 13.7±4.4, 15.4±3.8 and 12.7±3.9, respectively. The MCID for the RCQ was estimated to range from approximately 1.8 to 2.2 points across disease aetiologies. The distributions of the LCQ and RCQ scores across the disease groups are illustrated in online supplemental figure S2. Compared with the patients with COPD, those with asthma, IPF and other conditions had significantly higher NRS, CSS and COAT scores and lower LCQ and RCQ scores. Among the LCQ domains, physical and social domain scores were comparable between the COPD and IPF groups, whereas psychological domain scores were significantly lower in the IPF group. No significant differences were observed between the COPD and bronchiectasis groups. Table 1Baseline characteristics of patients according to chronic cough aetiologyCOPDAsthmaIPFBEOthers (n=52) (n=97) (n=34) (n=57) (n=60) Age70.7±7.555.5±17.4*71.5±7.067.7±8.656.6±17.2*Female sex4 (7.7%)63 (64.9%)*6 (17.6%)40 (70.2%)*34 (56.7%)*NRS3.9±2.35.2±2.3*5.1±2.4*4.0±2.55.3±2.4*CSS3.1±1.64.2±1.9*4.4±1.9*3.3±1.74.6±1.6*COAT total5.1±3.78.5±4.8*8.1±4.8*5.4±4.38.9±4.8*LCQ total15.7±3.613.1±4.1*13.9±4.2*15.0±4.312.6±3.8* Physical5.2±1.04.6±1.3*4.9±1.15.1±1.14.6±1.0* Psychological5.1±1.44.1±1.5*4.3±1.5*4.7±1.63.9±1.5* Social5.4±1.44.4±1.6*4.7±1.75.1±1.84.1±1.6*RCQ15.4±3.512.6±4.3*13.7±4.4*15.4±3.812.7±3.9**Significant difference compared with COPD population.BE, bronchiectasis; COAT, Cough Assessment Test; COPD, chronic obstructive pulmonary disease; CSS, Cough Symptom Score; IPF, idiopathic pulmonary fibrosis; LCQ, Leicester Cough Questionnaire; NRS, Numeric Rating Scale; RCQ, Rapid Cough Questionnaire. Baseline characteristics of patients according to chronic cough aetiology Significant difference compared with COPD population. BE, bronchiectasis; COAT, Cough Assessment Test; COPD, chronic obstructive pulmonary disease; CSS, Cough Symptom Score; IPF, idiopathic pulmonary fibrosis; LCQ, Leicester Cough Questionnaire; NRS, Numeric Rating Scale; RCQ, Rapid Cough Questionnaire. Detailed LCQ item-level comparisons across aetiologies are provided in online supplemental table S1. The LCQ1 (pain: physical), LCQ2 (sputum: physical) and LCQ9 (hypersensitivity to irritants: physical) scores did not differ significantly among the disease groups. In the comparison between asthma and IPF, most item scores were comparable, except for LCQ1 (chest or stomach pain), which was significantly higher in the IPF group. RCQ validation across respiratory diseases The correlations between the RCQ score and other cough-related measures, including NRS, CSS and LCQ, are presented in figure 1. The correlation matrix for these associations in the total population is shown in figure 2A. The RCQ was significantly correlated with the NRS (r=−0.59), total CSS (r=−0.62), COAT (r=−0.75) and LCQ (r=0.93). Stratified correlations of RCQ with NRS, CSS and COAT by disease group are presented in figure 1, and aetiology-specific correlation matrices in figure 2B–F. Although all correlations were significant, the strengths of the associations between NRS, CSS, COAT, and LCQ and RCQ, separately, were attenuated in the IPF group (LCQ and NRS, CSS, and COAT: r=−0.27, −0.46 and −0.58, respectively; RCQ and NRS, CSS, and COAT: r=−0.34, −0.39 and −0.53, respectively) compared with the other disease groups. Scatter plots showing pairwise correlations between RCQ and NRS, CSS, and COAT scores across disease groups. Each panel illustrates the relationship between the RCQ and another cough assessment tool across the entire study population. Higher NRS, CSS and COAT scores indicate more severe cough, whereas higher RCQ scores indicate better quality of life (lower symptom burden). COAT, Cough Assessment Test; COPD, chronic obstructive pulmonary disease; CSS, Cough Symptom Score; IPF, idiopathic pulmonary fibrosis; NRS, Numeric Rating Scale; RCQ, Rapid Cough Questionnaire. Correlation matrices between RCQ and other cough assessment tools across disease groups. Spearman’s correlation matrices illustrate the relationships between the RCQ and the other tested instruments (NRS, CSS, COAT and LCQ) within each disease group: (A) total population, (B) COPD, (C) asthma, (D) IPF, (E) bronchiectasis and (F) other diseases. Strong correlations across most groups supported the construct validity of the RCQ in diverse clinical contexts. Only significant associations are presented. COAT, Cough Assessment Test; COPD, chronic obstructive pulmonary disease; CSS, Cough Symptom Score; IPF, idiopathic pulmonary fibrosis; LCQ, Leicester Cough Questionnaire; NRS, Numeric Rating Scale; RCQ, Rapid Cough Questionnaire. The relationship between the RCQ and LCQ within each disease group is shown in figure 3. The correlation coefficients between the RCQ and total LCQ for COPD, asthma, IPF, bronchiectasis and other conditions were 0.93, 0.93, 0.93, 0.88, 0.95 and 0.95, respectively. The Bland-Altman plots indicated consistent agreement patterns across the range of scores within each disease group (figure 4). Correlations between RCQ and total LCQ scores across disease groups: (A) total population, (B) COPD, (C) asthma, (D) IPF, (E) bronchiectasis and (F) other diseases. Each dot colour represents a different aetiology group: red for COPD, green for asthma, blue for IPF, sky blue for bronchiectasis and black for other aetiologies. Regression lines indicate significance. COPD, chronic obstructive pulmonary disease; IPF, idiopathic pulmonary fibrosis; LCQ, Leicester Cough Questionnaire; RCQ, Rapid Cough Questionnaire. Bland-Altman plots comparing RCQ and total LCQ scores across disease groups. Plots display the agreement between RCQ and LCQ scores for each subgroup: (A) total population, (B) COPD, (C) asthma, (D) IPF, (E) bronchiectasis and (F) other diseases. The x-axis represents the mean of the RCQ and LCQ scores and the y-axis shows the difference between the two scores. The plots demonstrate even distributions without systematic bias, indicating good agreement between the two instruments across all aetiologies. COAT, Cough Assessment Test; COPD, chronic obstructive pulmonary disease; CSS, Cough Symptom Score; IPF, idiopathic pulmonary fibrosis; LCQ, Leicester Cough Questionnaire; NRS, Numeric Rating Scale; RCQ, Rapid Cough Questionnaire. The ICC analysis consistently demonstrated excellent agreement between RCQ and LCQ in the overall cohort (ICC=0.93, 95% CI 0.92 to 0.95) and across all disease groups, including COPD (ICC=0.95, 95% CI 0.87 to 0.96), asthma (ICC=0.93, 95% CI 0.89 to 0.95), IPF (ICC=0.89, 95% CI 0.78 to 0.94), bronchiectasis (ICC=0.94, 95% CI 0.91 to 0.97) and other conditions (ICC=0.95, 95% CI 0.92 to 0.97). An item-level correlation matrix between the RCQ and LCQ is displayed in online supplemental figure S3 shows significant associations between RCQ scores and individual LCQ items across all disease groups. RCQ scores were significantly correlated with all LCQ items across all disease groups, except for LCQ1 (chest/stomach pain), LCQ4 (feeling the cough cannot be controlled) and LCQ9 (hypersensitivity to irritants) in the IPF group. Notably, the LCQ9 did not correlate with the total LCQ score in the IPF group. Discussion In this study, we evaluated the clinical utility and validity of the RCQ in a heterogeneous population of patients with chronic cough caused by various underlying respiratory diseases. Our findings showed that the RCQ exhibited strong concordance with established cough assessment tools, including the LCQ, NRS, CSS and COAT, across a spectrum of aetiologies, such as asthma, COPD, IPF and bronchiectasis, thereby supporting its construct validity and clinical applicability. Notably, the RCQ score highly correlated with the total LCQ score for all disease groups, with correlation coefficients exceeding 0.88, indicating that the RCQ reliably captured the overall burden of cough symptoms and their impact on patient health-related quality of life. Given that the RCQ is derived from a subset of LCQ items and both instruments were administered contemporaneously, this high level of correlation should be interpreted with caution. The RCQ demonstrated strong correlations with all three domains of the LCQ (physical, psychological and social) indicating that it effectively captured the physical and psychosocial aspects of cough burden. This alignment with the multidimensional structure of the LCQ further supported the comprehensive assessment framework of the RCQ. RCQ performance was consistent across disease-specific subgroups, suggesting its robustness across varying symptom profiles. Furthermore, the Bland-Altman analysis confirmed good agreement between the RCQ and LCQ scores, with evenly distributed residuals across disease subgroups, indicating the absence of systematic bias. In the IPF subgroup, the strengths of the correlations between the RCQ or LCQ and other instruments such as NRS, CSS and COAT were comparatively reduced, suggesting that these unidimensional measures may have inadequately captured the multifaceted nature of cough-related impairments in fibrotic lung disease. In contrast, the RCQ maintained a strong correlation with the LCQ, even for patients with IPF, indicating closer alignment with the patient-experienced burden. These findings suggest that the RCQ, as a brief multidimensional instrument, may better reflect patient-experienced burden in specific disease contexts, particularly for patients with IPF. Importantly, the ICC analysis demonstrated a consistently high agreement between the RCQ and LCQ across all aetiologic subgroups. However, this should be interpreted as cross-sectional concordance rather than evidence of full psychometric equivalence. Although the ICC was slightly lower for patients with IPF, this finding likely reflects greater heterogeneity in symptom perception or disease-specific cough characteristics in fibrotic lung disease, rather than diminished measurement performance. Overall, the robustness of the agreement observed across disease categories supports the reliability of the RCQ as a complementary tool for assessing cough-related quality-of-life impairment for heterogeneous clinical populations. At the item level, most LCQ domains strongly aligned with the RCQ scores, although certain discrepancies were observed for patients with IPF. Specifically, LCQ1 (chest or stomach pain), LCQ4 (feeling the cough cannot be controlled) and LCQ9 (hypersensitivity to irritants) had weaker correlations with the RCQ in the IPF group. Additionally, the LCQ9 did not correlate significantly with the LCQ total score in IPF, suggesting that hypersensitivity-related symptoms may be less prominent or relevant in fibrotic lung disease. Thus, cough in this population may arise through different mechanisms compared with asthma, COPD or bronchiectasis and highlights the complex and possibly distinct symptom profiles of cough across different diseases.913 These findings underscore the importance of the context-specific interpretation of item-level data. Nevertheless, the RCQ performed consistently across all major aetiologies and may serve as a brief and pragmatic alternative to longer instruments such as the LCQ. Owing to its brevity, the RCQ may reduce respondent burden, improve compliance and facilitate repeated assessments in clinical practice and research.21 These features make it particularly advantageous for situations when longer questionnaires impose a higher response burden on patients or in longitudinal studies requiring repeated measurements over time. Overall, our results support the use of the RCQ as a valid, efficient and broadly applicable tool for assessing cough severity and impact in patients with chronic cough across diverse respiratory conditions. However, this study has several limitations. First, although the RCQ demonstrated strong cross-sectional validity, its longitudinal responsiveness to therapeutic interventions remains unclear. In addition, several important psychometric properties, including internal consistency, structural validity, known-groups validity based on prespecified hypotheses and test-retest reliability, were not evaluated in this study. Second, the sample population included a broad range of respiratory diseases; thus, the subgroup sizes, particularly for IPF and other less common aetiologies, were relatively small, which may have limited the statistical power for detecting subtle differences. Although a small number of patients with refractory unexplained chronic cough were included, the sample size was insufficient for subgroup analysis, and therefore, the applicability of the RCQ in this population cannot be directly established. Third, disease-specific variations in symptom profiles, particularly in IPF, suggest potential variability in symptom perception across disease subtypes and underscore the importance of further exploring the performance of the RCQ in subpopulations with atypical cough phenotypes. Fourth, the MCID was estimated using a distribution-based approach; therefore, additional anchor-based validation is required to confirm its clinical relevance. Finally, the study cohort was derived from specialised respiratory centres in the Republic of Korea, which may limit the generalisability of the findings to primary care centres and other populations. In conclusion, the RCQ demonstrated strong validity and cross-instrument consistency for assessing cough-related burden across a broad spectrum of respiratory conditions. The performance of the RCQ was comparable to that of the LCQ despite being markedly shorter and simpler. Therefore, the RCQ may be a practical alternative to longer instruments, particularly in settings requiring frequent or large-scale assessments. Supplementary material