Validation of cough evaluation test in subacute cough
Highlight box
Key findings
• Cough evaluation test (CET) is a reliable, valid and responsive subjective tool in subacute cough.
• The minimal important clinical difference of CET is two.
What is known and what is new?
• CET, composed of five items, has been developed and validated for chronic cough. CET is able to evaluate physical, psychological and social impact of chronic cough in a simple way based on our previous study.
• This study suggests that CET is reliable, valid and responsive when applied to patients with subacute cough.
What is the implication, and what should change now?
• CET is a promising subjective tool for cough assessment in subacute and chronic cough, offering a simple yet reliable way to evaluate the multidimensional impact of cough. It may facilitate more efficient assessment of cough in clinics and researches.
Introduction
Cough is a common symptom in respiratory clinics, and has a multidimensional negative impact on patients, including physical, psychological and social aspects. Cough can be classified into acute (<3 weeks), subacute (3–8 weeks) and chronic cough (>8 weeks) (1).
Well-rounded and reliable assessment of cough is fundamental to clinical management. Subjective assessments, based on patient-reported outcomes, are widely used in both clinical and research settings. However, there is a lack of validated scales for assessment of subacute cough. Existing researches have largely focused on chronic cough, while subacute cough has received comparatively less attention. Although various cough scales are available, most of them are designed for chronic cough and have not been validated in patients with subacute cough. So, it is necessary to identify appropriate tools for this population.
Cough evaluation test (CET) was developed and validated for chronic cough in our previous study (2). It consists of five items rated on a five-point Likert scale and covers physical, psychological and social domains, with a total score ranging from 5 to 25. Compared to Leicester Cough Questionnaire (LCQ), CET offers a simpler approach for evaluating multidimensional impact of cough. However, its applicability in subacute cough remains unclear. Therefore, the present study aims to evaluate the reliability, validity, and responsiveness of CET in patients with subacute cough. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-661/rc).
Methods
Subject
Patients with subacute cough were recruited from the outpatient department of The First Affiliated Hospital of Guangzhou Medical University between November 2022 to March 2024. Eligible subjects were those aged >14 years who presented with cough as the sole or predominant symptom lasting for 3 to 8 weeks, with no abnormal findings of chest X-ray or computed tomography scan. Exclusion criteria were as follow: (I) current smokers; (II) ex-smokers who had quit within the past six months or had a smoking history of >10 pack-years.
All patients provided demographic information and completed CET, cough severity Visual Analogue Scale (VAS), Cough Symptom Score (CSS) and Mandarin Chinese version of the Leicester Cough Questionnaire (LCQ-MC) at baseline. Sixty-five patients were followed up via telephone after a 7-day interval and completed CET, VAS, CSS and Global Rating of Change Questionnaire (GRCQ). GRCQ, a 15-point Likert scale commonly used to evaluate the change of condition, ranged from −7 (a very great deal worse) to 7 (a very great deal better) with integer scores. According to GRCQ scores, change of condition was categorized into no change (0, ±1), small change (±2, ±3), moderate change (±4, ±5) and large change (±6, ±7) (3).
The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of The First Affiliated Hospital of Guangzhou Medical University (No. 201777). All patients or their guardians provided written informed consent.
Reliability and item analysis
Internal reliability was assessed using Cronbach’s alpha coefficient, which reflected the internal consistency. A Cronbach’s alpha coefficient greater than 0.7 was acceptable. Furthermore, Cronbach’s alpha coefficient was expected to decrease if any item was removed, indicating its contribution to the overall reliability. Item-item correlation was used to explore item homogeneity. Correlation coefficient greater than 0.70 suggested strong similarity or redundancy, in which case the items were considered for combination or removal. An item-total correlation was acceptable if it was greater than 0.4. Otherwise, the item might not be sufficiently related to the overall construct. Ceiling or floor effects were deemed present if more than 50% of participants selected the highest or lowest score for a given item, respectively. These effects were considered undesirable.
Concurrent validity
To evaluate concurrent validity, correlation between CET and other three tools was calculated using the data collected at the same time point. Moreover, correlation comparisons were conducted, for example, comparing correlation between VAS and CET with that between VAS and LCQ-MC/CSS total/CSS daytime/CSS nighttime. A total of 30 correlation comparisons were performed at baseline and 18 at follow-up. Correlation strength was interpreted as weak (<0.5), moderate (≥0.5 and <0.7) and strong (≥0.7) based on the absolute value of coefficient.
Repeatability
The change of condition was determined based on GRCQ. Intraclass correlation coefficient (ICC) was calculated to assess repeatability of CET in patients with no and small change (GRCQ =0, ±1, ±2, ±3). Repeatability was interpreted as poor (ICC <0.5), moderate (0.5≤ ICC <0.75), good (0.75≤ ICC <0.9) and excellent (ICC ≥0.9).
Responsiveness
Patients with changed condition were included and the score changes were evaluated for responsiveness of CET. Anchor-based approach was applied to determine minimal important clinically difference (MICD) (4) and GRCQ served as the anchor. Briefly, the score changes compared to baseline were calculated for the patients reporting a small change (GRCQ =±2, ±3).
Statistical analysis
IBM SPSS 26 was used for data analysis. Normally distributed data and skewed distribution data were presented as mean ± standard deviation and median (interquartile range), respectively. Independent quantitative data were compared by independent samples t-test or Mann-Whitney U test. Paired quantitative data were compared using paired t-test or Wilcoxon signed-rank test. Qualitative data were compared using Chi-squared test. Spearman’s rank correlation coefficients were calculated and correlation comparisons were performed by cocor (5). ICC was calculated based on a single-rater, consistency, two-way random effects model for repeatability analysis. A P value <0.05 was statistically significant.
Results
Subject
A total of 158 patients were recruited for this study, and 65 patients completed follow-up. All required data were fully collected at baseline, and all follow-up data, except for LCQ-MC, were also obtained. Overall, the median age and cough duration were 36.0 (30.0, 49.0) years and 4.0 (4.0, 6.0) weeks respectively, with females of 56.3% (Table 1). The scores of VAS, LCQ-MC, CSS total and CET at baseline were 50.0 (30.0, 70.0) mm, 12.5±3.5, 4.0 (3.0, 5.0) and 14.5 (12.0, 17.0), respectively. There was no significant difference of all these baseline characteristics or scales between patients with follow-up and those without follow-up.
Table 1
| Characteristics | All patients (n=158) | Patients without follow-up (n=93) | Patients with follow-up (n=65) | t/Z/χ2 | P value† |
|---|---|---|---|---|---|
| Age, years | 36.0 (30.0, 49.0) | 36.0 (30.0, 49.0) | 35.0 (28.5, 49.0) | −0.649 | 0.52 |
| Gender | |||||
| Female | 89 (56.3) | 52 (55.9) | 37 (56.9) | 0.016 | 0.90 |
| Cough duration, weeks | 4.0 (4.0, 6.0) | 4.0 (4.0, 6.0) | 5.0 (4.0, 6.0) | −0.843 | 0.40 |
| LCQ-MC | 12.5±3.5 | 12.5±3.9 | 12.5±3.0 | 0.082 | 0.93 |
| VAS, mm | 50.0 (30.0, 70.0) | 50.0 (30.0, 72.5) | 60.0 (40.0, 70.0) | −0.960 | 0.34 |
| CSS | |||||
| Daytime | 3.0 (2.0, 3.0) | 3.0 (2.0, 4.0) | 3.0 (2.0, 3.0) | −1.157 | 0.25 |
| Nighttime | 1.0 (1.0, 2.0) | 1.0 (1.0, 2.0) | 1.0 (1.0, 2.5) | −1.023 | 0.31 |
| Total | 4.0 (3.0, 5.0) | 4.0 (3.0, 5.0) | 4.0 (4.0, 5.0) | −0.121 | 0.90 |
| CET | 14.5 (12.0, 17.0) | 14.0 (11.0, 17.0) | 15.0 (12.0, 17.0) | −0.909 | 0.36 |
Data were presented as mean ± standard deviation, median (interquartile range) or n (%). †, comparisons of characteristics between patients without follow-up and with follow-up. CET, cough evaluation test; CSS, Cough Symptom Score; LCQ-MC, Mandarin Chinese version of the Leicester Cough Questionnaire; VAS, Visual Analogue Scale.
Internal consistency and item analysis
Cronbach’s alpha for CET was 0.805. There was no ceiling effect or floor effect (Figure S1). None of the item-item correlation coefficients exceeded 0.7, indicating no redundancy (Figure 1). Corrected item-total correlations for each item were above 0.4, ranging from 0.454 to 0.706. Additionally, Cronbach’ s alpha decreased when any item was removed. Therefore, no modifications to the items were necessary when applying CET in patients with subacute cough.
Concurrent validity
At baseline, CET showed a strong negative correlation with LCQ-MC (r=−0.732), and moderate positive correlations with VAS, CSS total, CSS daytime and CSS nighttime (r=0.662, r=0.672, r=0.558 and r=0.509, respectively) (Figure 2). The correlation between VAS and CET was significantly stronger than that between VAS and LCQ-MC/CSS nighttime (both P<0.001). In addition, the correlation between LCQ-MC and CET was greater than that between LCQ-MC and VAS/CSS total/CSS daytime/CSS nighttime (all P<0.001). Furthermore, CET showed a stronger correlation with CSS total than with CSS daytime/CSS nighttime (P=0.01 and P<0.001, respectively).
At follow-up, CET demonstrated strong positive correlations with VAS and CSS total (r=0.764 and r=0.823, respectively), and moderate positive correlations with CSS daytime and CSS nighttime (r=0.696 and r=0.694, respectively) (Figure 3). VAS correlated more closely with CET than with CSS nighttime (P=0.03), which was also found at baseline. CET also showed a stronger correlation with CSS total than with CSS daytime/CSS nighttime (P=0.01 and P=0.008, respectively). Complete results of correlation comparisons were presented in Appendix 1.
Repeatability
A total of 28 patients with GRCQ scores of 0, ±1, ±2, ±3 were included, owing to the limited number of stable patients with subacute cough. The ICC of CET, VAS and CSS total was 0.571, 0.677 and 0.644 (all P<0.001), indicating a moderate repeatability.
It was likely that the true ICC was higher, as the analysis included patients whose condition was not entirely stable.
Responsiveness
A total of 56 patients with changed condition, whose GRCQ scores were ±2, ±3, ±4, ±5, ±6 or ±7, were included in responsiveness analysis. CET, VAS and CSS total all decreased significantly from baseline to follow-up, demonstrating strong responsiveness (Table 2). There were 19 patients reporting a small change in condition. The MICD for CET, VAS and CSS total in subacute cough were 2.0 (1.0, 5.0), 20.0 (10.0, 30.0) mm and 1.0 (0, 2.0) respectively.
Table 2
| Scales | Baseline | Follow-up | Z | P value |
|---|---|---|---|---|
| CET | 15.0 (13.0, 17.0) | 9.0 (7.0, 12.0) | −5.711 | <0.001 |
| VAS, mm | 60.0 (40.0, 75.0) | 22.5 (10.0, 47.5) | −5.482 | <0.001 |
| CSS total | 4.0 (4.0, 5.8) | 3.0 (2.0, 4.0) | −5.887 | <0.001 |
Data were presented as median (interquartile range). CET, cough evaluation test; CSS, Cough Symptom Score; VAS, Visual Analogue Scale.
Discussion
CET was originally designed for chronic cough as a simpler and quicker tool for assessing the multidimensional impact of cough compared to the LCQ, which was demonstrated in our previous study (2). However, its applicability in subacute cough remained unclear. In the present study, CET showed robust internal consistency, strong concurrent validity, good repeatability and responsiveness in patients with subacute cough. Additionally, the MICD for CET in subacute cough was determined to be 2.0 (1.0, 5.0).
The shorter duration of cough did not reduce the internal consistency of CET. For concurrent validity, we selected VAS, CSS total and LCQ-MC to perform correlation analysis, which were commonly used subjective tools (6). VAS and LCQ were widely recognized for evaluating cough severity and cough-related quality of life, respectively. As was mentioned above, CET showed moderate to strong correlations with VAS, LCQ-MC and CSS total. Notably, significant stronger correlations were observed between VAS and CET, compared to that between VAS and LCQ-MC/CSS nighttime. Moreover, LCQ-MC correlated more strongly with CET than with VAS/CSS total/CSS daytime/CSS nighttime. These findings suggested that CET could effectively measure both cough severity and cough-related quality of life in patients with subacute cough, consistent with previous findings in chronic cough. Importantly, CET consisted of only five items and could be finished in a short time, making it well-suited for use in clinical practice. In addition, the correlation between CET and CSS total was stronger than that between CET and CSS daytime/nighttime, implying that CET captured the impact of cough throughout the day rather than focusing on specific time periods.
Postinfectious cough was a main etiology in subacute cough (1). It was considered as self-limited and symptoms would improve after symptomatic treatment in most cases. Consequently, unlike chronic cough, a lack of patients with stable condition was almost inevitable in subacute cough, which posed a challenge for assessing the repeatability of CET. This point also explained why the ICC of LCQ was not determined for acute cough in Yousaf’s research (7). One possible approach to address this issue was to shorten the follow-up interval to 24 hours (8). However, in our view, an interval of 24 hours was too short and it was possible that patients just repeated the previous response without proper reflection. Another study included the patients that still had a cough to performed test-retest analysis (9). Therefore, we included the patients with no or small change in condition for ICC analysis. Although the ICC for CET was 0.571, it was likely underestimated due to the inclusion of some unstable patients. Bland-Altman plot was not generated due to the skewed distribution of the data. We tied to make comparison with other studies about validation of subjective tools in acute or subacute cough, but the available data was quite limited. In acute cough, cough severity VAS showed an ICC more than 0.7 with an interval of 24 hours in Sunger’s study (8). The ICC of Cough Severity Diary, another subjective tool, was reported as 0.68 and 0.94 at day 8 and day 15 visit, respectively (10). However, in the study of Cough Severity Diary, the subjects for test-retest analysis included stable patients with both chronic cough and subacute cough, with only two cases of subacute cough. Therefore, those results might not fully represent the subacute cough population. Interestingly, the ICC of VAS was reported as 0.51 in patients with refractory or unexplained chronic cough, possibly due to the absence of a gold standard for defining patient stability or differences in patient populations (11).
CET declined significantly from 15.0 (13.0, 17.0) at baseline to 9.0 (7.0, 12.0) at follow-up, indicating its good responsiveness. Effect size was wildly used to quantify the extent of change, but it was not calculated because of the skew nature of data. The MICD of CET was determined to be 2.0 (1.0, 5.0) by the anchor-based approach in subacute cough, which was similar in chronic cough based on our previous study (2).
In most cases, acute and subacute cough were considered as self-limited, which has resulted in greater research focus on chronic cough. Although VAS, LCQ and CSS total were commonly used (6), there were limited data about the application of these tools in subacute cough. This study partly filled the gap by providing data about the correlation, repeatability and responsiveness. Besides, the MCID of VAS and CSS total in subacute cough was 20.0 (10.0, 30.0) mm and 1.0 (0, 2.0), respectively. In acute cough, the MICD of VAS was reported as 13 or 17 mm in two different studies (3,7).
Our findings suggested that CET was a reliable and valid instrument for evaluating subacute cough. It was widely acknowledged that cough had a broad and multidimensional impact. While most available scales focused either on cough severity or cough-related quality of life, CET offered a more comprehensive evaluation of both dimensions. There were only five items in CET, in contrast to LCQ and Cough-specific Quality of Life Questionnaire, which contained 19 and 28 items (6), respectively. Therefore, CET might enable clinicians to quickly and comprehensively evaluate the change of cough severity after treatment, guiding clinical decisions-making in subacute cough.
There were some limitations of this study. First, the study population was relatively young with a median age of 36 years old, which might limit the generalizability of the findings to older populations. Second, there was a lack of LCQ-MC in follow-up. Most patients would improve after treatment, even without treatment, which leaded to a low willingness to seek medical help again. As a result, follow-up was performed via telephone and LCQ-MC could not be reliably collected during follow-up. Due to the absence of LCQ-MC in follow-up, its responsiveness and MCID in subacute cough could not be evaluated. However, CET and LCQ-MC were strongly correlated, and responsiveness of CET was clearly demonstrated. In addition, the responsiveness of LCQ in acute cough has been previously validated (3,7). Therefore, we speculated that LCQ-MC would also exhibit significant changes in follow-up. This assumption, however, required further confirmation.
Conclusions
This study suggests that CET is a well-designed, reliable, valid and responsive subjective tool for assessing subacute cough. CET effectively captures multidimensional impact of cough in a simpler and quicker way. We believe that CET is a promising tool for subacute cough in clinics or researches.
Acknowledgments
We would like to thank the patients for participating in this study. We also would like to thank Yawen He, Xinqi Wu and Jiayi Liao for helping the data collection. The abstract has been presented as a poster in the 4th International Cough Conference, Guangzhou China, December 2023.
Footnote
Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-661/rc
Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-661/dss
Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-661/prf
Funding: This work was supported by
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-661/coif). K.L. serves as an unpaid editorial board member (Respiratory Medicine) of Journal of Thoracic Disease. The other authors have no conflicts of interest to declare.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the research ethics committee of The First Affiliated Hospital of Guangzhou Medical University (No. 201777) and informed consent was obtained from all individual participants or their guardians.
Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.
References
- Chinese national guideline on diagnosis and management of cough(2021). Zhonghua Jie He He Hu Xi Za Zhi 2022;45:13-46. [Crossref] [PubMed]
- Zhan W, Zhang L, Jiang M, et al. A new simple score of chronic cough: cough evaluation test. BMC Pulm Med 2020;20:68. [Crossref] [PubMed]
- Lee KK, Matos S, Evans DH, et al. A longitudinal assessment of acute cough. Am J Respir Crit Care Med 2013;187:991-7. [Crossref] [PubMed]
- Braido F, Candeliere MG, Bondi B, et al. Chronic Cough Patient Perspective: questionnaire validation and symptom impact. ERJ Open Res 2025;11:00221-2024. [Crossref] [PubMed]
- Diedenhofen B, Musch J. cocor: a comprehensive solution for the statistical comparison of correlations. PLoS One 2015;10:e0121945. [Crossref] [PubMed]
- Mackay EC, Turner RD, Cho PSP, et al. Patient-reported assessments of chronic cough in clinical trials: accessory or primary endpoints? J Thorac Dis 2024;16:7165-81. [Crossref] [PubMed]
- Yousaf N, Lee KK, Jayaraman B, et al. The assessment of quality of life in acute cough with the Leicester Cough Questionnaire (LCQ-acute). Cough 2011;7:4. [Crossref] [PubMed]
- Sunger K, Powley W, Kelsall A, et al. Objective measurement of cough in otherwise healthy volunteers with acute cough. Eur Respir J 2013;41:277-84. [Crossref] [PubMed]
- Anderson-James S, Newcombe PA, Marchant JM, et al. An acute cough-specific quality-of-life questionnaire for children: Development and validation. J Allergy Clin Immunol 2015;135:1179-85.e1-4.
- Vernon M, Kline Leidy N, Nacson A, et al. Measuring cough severity: development and pilot testing of a new seven-item cough severity patient-reported outcome measure. Ther Adv Respir Dis 2010;4:199-208. [Crossref] [PubMed]
- Martin Nguyen A, Bacci ED, Vernon M, et al. Validation of a visual analog scale for assessing cough severity in patients with chronic cough. Ther Adv Respir Dis 2021;15:17534666211049743. [Crossref] [PubMed]
