Sage Journals: Discover world-class research

Abstract

Background:

Sex-related disparities in the prevalence of chronic cough have been consistently reported globally, with varying male-to-female ratios.

Objectives:

This study aimed to evaluate sex-related differences by comparing correlations between cough-related symptoms in males and females of different age groups.

Design:

Adult patients with chronic cough who completed the Leicester Cough Questionnaire (LCQ) were recruited from 16 respiratory centers.

Methods:

Correlation networks were constructed based on Spearman’s correlation coefficients in males and females of various age groups. The distinct relationships of cough-related symptoms between subgroups were validated by an independent cohort.

Results:

A total of 255 patients were enrolled in this study (male-to-female ratio, 1:1.71). The following LCQ items were highly correlated: embarrassment and interference with daily work, anxiety, and interference with overall life enjoyment/feeling of being fed up, interference with daily work and overall life enjoyment, interference with overall life enjoyment and feeling of being fed up, and feeling of being fed up and annoyance to partner/family/friends. The patterns of these correlations between LCQ items varied in males and females of different ages. The strongest interrelationship was observed in male patients aged >50 years old, which was similar to those in the validation cohort.

Conclusion:

The correlation patterns between cough-related symptoms vary significantly according to age and sex. Understanding the mechanisms underlying the development of cough-related symptoms may facilitate sex- and age-specific strategies for chronic cough.

Keywords

age cough gender interrelationship network quality of life

Background

Cough is a vital defense mechanism that protects the respiratory tract and lungs^1,2; however, it can become bothersome when it persists, regardless of age or sex.^3,4 With varying male-to-female ratios, sex imbalances regarding the prevalence of chronic cough have been consistently reported globally,⁵ indicating that there are sex-specific differences in the pathophysiology of chronic cough. Nonetheless, the mechanisms underlying sex-related disparities remain poorly understood.

For assessing the severity of cough and its impact on health-related quality of life, several tools are available.^6–11 Among these, patient-reported outcome measurements using validated and reliable questionnaires are considered the gold standard.¹² The Leicester Cough Questionnaire (LCQ) was developed to evaluate the effect of cough on a patient’s quality of life,⁶ and its Korean version has been validated.^7,13 The LCQ is a collection of disparate items and includes physical, psychological, and social domains.⁷ The physical domain includes items about the presence and severity of pain (LCQ1), phlegm (LCQ2),^14,15 tiredness (LCQ3), hypersensitivity to irritants (LCQ9),¹⁶ sleep disturbance (LCQ10),¹⁷ coughing bout frequency (LCQ11), voice hoarseness (LCQ14), and loss of energy (LCQ15). The psychological domain includes items that indicate the patient’s perceived severity of cough, such as feeling in control of cough (LCQ4), embarrassment (LCQ5), anxiety (LCQ6),¹⁸ frustration (LCQ12), feeling of being fed up (LCQ13), worries about serious illness (LCQ16), and concern about other peoples’ thoughts (LCQ17). The social domain includes items on interference with social activities, such as daily work (LCQ7), overall life enjoyment (LCQ8), conversation (LCQ18), and annoyance to others (LCQ19).⁴ Detailed evaluation of the characteristics of each item may provide valuable information reflecting the pathophysiology of cough.¹⁹

A potential strategy for better understanding the mechanism for cough development and identifying the effects of age and sex on cough-related symptoms is to analyze the characteristics of each LCQ item and their interconnections. The aim of this study was to examine the distinct patterns of cough-related symptoms, including correlations between LCQ items, in males and females of various age groups. In addition, we attempted to validate our findings using an external cohort, which was the independent cohort of a multicenter, prospective study designed to validate the Korean version of the LCQ.¹³

Methods

Study population and data collection

Between March 2016 and February 2018, adult patients aged ⩾18 years old diagnosed with chronic cough were prospectively recruited from 16 respiratory centers in the Republic of Korea and retrospectively analyzed. Chronic cough was defined as cough lasting longer than 8 weeks.²⁰ The inclusion criterion for this study was completion of the LCQ⁶ at baseline. Pulmonary specialists at each hospital evaluated the possible causes of chronic cough according to the Korean cough guidelines.²¹ Exclusion criteria included patients with abnormal chest radiographs and those with known chronic respiratory diseases, such as overt asthma, chronic obstructive pulmonary disease, bronchiectasis, tuberculosis-destroyed lung, or lung cancer. At the initial visit, demographic information, the LCQ score,⁶ and the identified causes of cough were recorded.

Validation was performed using an independent preexisting cohort of a multicenter prospective study designed to validate the Korean version of the LCQ.¹³ Details of the inclusion and exclusion criteria for this cohort have been described previously.¹³

The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.²² The checklist from STROBE was provided in the Supplemental Material.

Statistical analysis

Data are presented as mean ± standard deviation or median with interquartile range for continuous variables and as number (percentage) for categorical variables. Correlation networks were constructed based on correlation matrices using Spearman’s rank correlation, and the Bonferroni correction (p value <0.05) was applied to adjust the correlation matrices. In the correlation network, each LCQ item is represented as a node in a color that corresponds to the domain of the item: red, green, and blue for physical, psychological, and social domains, respectively. The size of each node represents the mean score of each item. The links between the nodes demonstrate the existence of a significant correlation with a coefficient >0.6. The links with stronger correlation (coefficient > 0.7) were marked thicker and darker. The igraph package was used to visualize the correlation networks. To compare the strength of correlations between the symptoms of LCQ items⁶ in male and female patients, we compared two Spearman’s correlations based on independent groups using Zou’s method,²³ with an alpha level of 0.05 and a confidence level of 0.95. The strength of the correlation of the paired variables between males and females was compared using the cocor package. All statistical analyses were performed using R software (version 4.1.3).

Results

Characteristics of the study population stratified according to sex

A total of 255 patients were included in this study. The baseline patient characteristics are summarized in Table 1. The mean age of the patients was 47.7 ± 14.3 years, and 94 (36.9%) and 161 (63.1%) of them were male and female, respectively (male-to-female ratio, 1:1.71). The female patients were older than the males (43.3 ± 13.8 versus 50.3 ± 14.0; p < 0.001). The histogram and density plot for age distribution stratified according to sex are shown in Supplemental Figure S1.

Table 1.

Comparison of the baseline characteristics of male and female patients.

	Total (N = 255)	Male (N = 94)	Female (N = 161)	p Value
Age	47.7 ± 14.3	43.3 ± 13.8	50.3 ± 14.0	<0.001
Etiology
UACS	116 (45.5%)	45 (47.9%)	71 (44.1%)	0.650
CVA	66 (25.9%)	15 (16.0%)	51 (31.7%)	0.009
EB	41 (16.1%)	14 (14.9%)	27 (16.8%)	0.828
GERD	34 (13.3%)	14 (14.9%)	20 (12.4%)	0.712
Tobacco smoking	15 (5.9%)	13 (13.8%)	2 (1.2%)	<0.001
Idiopathic cough	16 (6.3%)	8 (8.5%)	8 (5.0%)	0.391
Multiple cause	32 (12.5%)	15 (16.0%)	17 (10.6%)	0.289
LCQ score
Physical domain	4.1 ± 0.9	4.2 ± 0.8	4.0 ± 1.0	0.121
Psychological domain	3.5 ± 1.2	3.5 ± 1.1	3.5 ± 1.2	0.728
Social domain	3.6 ± 1.3	3.7 ± 1.2	3.6 ± 1.4	0.536
Total score	11.2 ± 3.1	11.4 ± 2.9	11.0 ± 3.2	0.392

CVA, cough variant asthma; EB, eosinophilic bronchitis; GERD, gastroesophageal reflux disease; LCQ, Leicester cough questionnaire; UACS, upper airway cough syndrome.

The proportion of female patients increased with age before the 70–90 years age group (Figure 1). In addition, the number of female patients in their 50s (OR = 2.71; 95% CI = 1.06–6.95) and 60s (OR = 5.78; 95% CI = 1.97–16.98) was significantly higher than those in the 10–20 years age group. Female patients were more likely to have cough-variant asthma than male patients (16.0% versus 31.7%; p = 0.009). Also, the proportion of current smokers was higher in males than females (13.8% versus 1.2%; p < 0.001). Since the use of cannabis is strictly prohibited by law in Korea, only tobacco use among smokes was considered in our study. The severity of cough, measured using the LCQ score, was not significantly different between sexes (Table 1 and Supplemental Table S1).

Figure 1.

Changes in the sex-specific proportions of patients in age groups categorized in 10-year age increments.

Correlations between the LCQ items

A correlation matrix was constructed to evaluate the differences in the relationships between LCQ items and the Bonferroni correction method was employed to adjust the correlation matrices (Supplemental Figure S2A). Strong correlations with Spearman’s coefficients >0.7 were observed between LCQ5 (embarrassment) and LCQ7 (interference with daily work), LCQ6 (anxiety) and LCQ8 (interference with overall life enjoyment), LCQ6 (anxiety) and LCQ13 (feeling of being fed up), LCQ7 (interference with daily work) and LCQ8 (interference with overall life enjoyment), LCQ8 (interference with overall life enjoyment) and LCQ13 (feeling of being fed up), and LCQ13 (feeling of being fed up) and LCQ19 (annoyance to partner/family/friends).

Differences in correlation networks according to sex and age

Correlation matrices were constructed separately for males (Supplemental Figure S2B) and females (Supplemental Figure S2C) to compare the patterns of the correlations between LCQ items. In addition, correlation networks for males [Figure 2(a)] and females [Figure 2(b)] were drawn based on the correlation matrices. The pattern of correlation networks of males was distinct from that of females. LCQ5 (embarrassment), LCQ6 (anxiety), and LCQ13 (feeling of being fed up) for males demonstrated features of node centrality, as did LCQ7 (interference with daily work) and LCQ13 (feeling of being fed up) for females (Supplemental Table S2A and S2B). The p values for the comparison of Spearman’s correlation coefficients between males and females are summarized in Supplemental Figure S3. Only male patients exhibited strong correlations (coefficient > 0.7) between LCQ3 (tiredness) and LCQ8 (interference with overall life enjoyment), LCQ5 (embarrassment) and LCQ6 (anxiety), and LCQ5 (embarrassment) and LCQ13 (feeling of being fed up). On the contrary, only female patients exhibited strong correlations between LCQ7 (interference with daily work) and LCQ13 (feeling of being fed up), and LCQ18 (interruption of conversation/phone calls) and LCQ19 (annoyance to partner/family/friends).

Figure 2.

Correlation networks showing the differences in patterns between (a) males and (b) females. Each LCQ item was represented in the correlation network as a node whose color reflected the characteristic domain to which it belonged: physical, psychological, and social domains as red, green, and blue, respectively. The size of each node represented the mean score for each item. Links between the nodes demonstrated the existence of statistically significant correlations. The strength of the correlation (Spearman’s coefficient) was represented by the thickness and darkness of the links. Only correlations with Spearman’s correlation coefficient ⩾0.6 are drawn.

The patients were divided into two groups based on their mean age: <50 years and >50 years. Comparison of the baseline characteristics of the two age groups is summarized in Supplemental Table S3. The proportion of females in the >50 years group was higher than that in the <50 years group (54.3% versus 73.5%, p = 0.002). There was no difference in the prevalence of each cause of cough between the groups. However, the subtotal scores of the psychological and social domains were higher in the >50 years group. Physical domain scores did not differ between the groups. Separate correlation networks were drawn for the two groups, and denser correlation links were observed in the >50 years group (Supplemental Figure S4), particularly between items in the psychological and social domains. LCQ5 (embarrassment), LCQ6 (anxiety), LCQ7 (frustration), LCQ8 (interference with overall life enjoyment), LCQ12 (frustration), LCQ13 (feeling of being fed up), LCQ18 (interruption of conversation/phone calls), and LCQ19 (annoyance to partner/family/friends) demonstrated node centrality features in the >50 years age group (Supplemental Table S2C and S2D). The p values for the comparison of Spearman’s correlation coefficients between the <50 and >50 years groups are summarized in Supplemental Figure S5.

Consequently, the patients were divided into four groups according to sex and age (Supplemental Table S4). The correlation networks between LCQ items exhibited distinct patterns in the four groups (Figure 3). Out of 342 total connections, males under 50 years, males over 50 years, females under 50 years, and females over 50 years had 30 (8.8%), 124 (36.3%), 54 (15.8%), and 70 (20.5%) links, respectively, and the strongest interrelationship between items was observed in male patients aged >50 years old.

Figure 3.

Correlation networks showing the distinct patterns according to age and sex. (a) Male under 50 years, (b) Male over 50 years, (c) Female under 50 years, and (d) Female over 50 years.Chest/stomach pain (LCQ1); presence of bothersome phlegm (LCQ2); tiredness (LCQ3); feeling in control of cough (LCQ4); embarrassment (LCQ5); anxiety (LCQ6); interference with daily work (LCQ7); interference with overall life enjoyment (LCQ8); hypersensitivity to irritants (LCQ9); sleep disturbance (LCQ10); cough bout frequency (LCQ11); frustration (LCQ12); feeling of being fed up (LCQ13); voice hoarseness (LCQ14); loss of energy (LCQ15); worries about serious illness (LCQ16); concern about other people’s thoughts (LCQ17); interruption of conversation/phone calls (LCQ18); and annoyance to partner/family/friends (LCQ19).LCQ, Leicester cough questionnaire.

External validation

Validation was performed using an independent cohort of 203 patients with chronic cough. The baseline characteristics of the two cohorts are summarized in Supplemental Table S5. The mean age of the patients in the validation cohort was 49.3 ± 14.3 years, and 80 (39.4%) and 123 (60.6%) of them were male and female, respectively, indicating that the sex distributions of the study and validation cohorts are comparable. Correlation networks between LCQ items were drawn for four groups: males <50 years old, males >50 years old, females <50 years old, and females >50 years old. Comparisons of the patterns of the correlation networks in the four groups are shown in Supplemental Figure S6. The strongest interrelationship between items was observed in male patients older than 50 years.

Discussion

In this study, we investigated the complex correlations between cough-related symptoms and compared the correlation patterns between sexes across different age groups. Several items from the psychological and social domains of the LCQ were found to be very strongly correlated with each other. Although the LCQ score did not differ between males and females, correlation patterns varied according to sex and age. LCQ scores increased with increasing age, as did the strength of the correlations between items, particularly those in the psychological and social domains. The densest and strongest correlations between items were identified in male patients older than 50 years.

Sex-related disparities in the prevalence of chronic cough have been consistently reported in epidemiological studies worldwide.⁵ Chronic cough is reported to be most prevalent in the 50–60 years age group.⁵ Similar to previous reports, the male-to-female ratio in the present study was 1:1.71, and the age distribution showed twin peaks: the first in the 30–39 years age group, dominated by men, and the second in the 60–69 years age group, dominated by women.

Several explanations have been proposed for the pathogenesis of sex-related differences in the prevalence of cough. Fujimura et al. conducted a capsaicin cough challenge test on healthy nonsmokers and found that female participants exhibited a lower threshold, regardless of age.²⁴ Kelsall et al. measured cough frequency and cough reflex sensitivity in patients with chronic cough and found that female patients had a higher nighttime cough frequency and cough reflex.²⁵ These results have been used to support the theory that females are susceptible to chronic cough. However, similar to our results, total and subtotal LCQ scores in the aforementioned study did not differ between males and females.²⁵ Few studies have focused on the differences in quality of life between men and women with chronic cough.

The pathophysiology of sex-related disparities in the prevalence of chronic cough may be attributable to genetic, hormonal, and socioeconomic factors.^26,27 In the present study, the proportion of females with cough increased with age in all age groups younger than 70–90 years. Considering this consistent upward trend in prevalence, which was neither concave nor convex, it is difficult to attribute these sex disparities in prevalence to hormonal mediators or socioeconomic effects.

Female patients had a stronger association between interruption of conversation/phone calls and annoyance to partner/family/friends than male patients, whereas male patients had a stronger association between embarrassment and anxiety or feelings of being fed up than female patients. Correlation patterns may indicate the pathophysiologic mechanism underlying the development of cough-associated symptoms. Given the significance of psychological factors on the development of various cough-related symptoms, it may be necessary to conduct a comprehensive evaluation, including patients’ perceptions of their psychosocial impairment in the management of chronic cough.

Interestingly, females are often mistreated as having hypochondriasis and somatization; however, the present study indicated that the associations between physical and psychological or physical and social items in females were neither stronger nor significantly different from those in males.

In our study, the severity of cough decreased with age. However, the interrelationships between LCQ items became more intricate with age, which was the most complex in male patients aged >50 years, suggesting the impact of socioeconomic factors. Interactions between psychosocial items became more prominent in patients >50 years old, particularly in male patients, who are usually in charge of providing for their families. Emotional symptoms, including frustration, embarrassment, worries about serious illness, feeling of being fed up, and social symptoms, including interference with overall life enjoyment, appeared to be central to all associations in male patients >50 years old. Coughing is a complex reflex involving the peripheral nervous system, brainstem, and higher brain, under volitional and cognitive control. Functional brain imaging studies in humans have discovered extensive brain activity in sensory cortical areas, cingulate, insula, and orbitofrontal cortices during the inhalation of cough-inducing stimuli, which reflects the various autonomic responses with emotional and discriminative sensory experiences accompanying cough.^16,28 These processes may partially explain the impact of psychological factors on physical and social symptoms.

The key strength of our study is the comprehensive analysis of cough-related quality of life impairments and their interactions, which have not been studied to date. Our findings emphasize the different mechanisms by which cough affects quality of life in various demographic groups, highlighting the need for specific assessments to improve patient management.

This study has several limitations. First, we could not assess the effects of sex-related disparities on treatment outcomes or longitudinal effects. Additional follow-up data are necessary to determine the clinical relevance of our study and sex-specific treatment outcomes in patients with chronic cough. Second, we could not collect additional clinical information, such as mental health conditions, detailed medication histories, and laboratory findings, because we used predefined data. Further research is necessary to understand the complex linkage of cough-associated symptoms with other mental health status indices. Third, different social determinants of health may have varying effects on disease across age groups. However, we could not assess the impact of other health disparity indices, such as socioeconomic status. Fourth, sample size power calculation was not performed for our study. Although we performed external validation using an independent cohort, additional research is required to determine the generalizability of our results to other populations.

Conclusion

This study demonstrated that there are sex-related disparities in the correlation patterns of cough-associated symptoms across age groups. The findings of this study pave the way for future research to understand the mechanisms underlying the development of cough and cough-related symptoms, as well as their impact on the overall quality of life. On the basis of our findings, sex- and age-specific strategies for precision treatment should be developed.

Supplemental Material

sj-docx-1-tar-10.1177_17534666241252545 – Supplemental material for Sex-related disparities in cough-associated symptoms across different age groups

Supplemental material, sj-docx-1-tar-10.1177_17534666241252545 for Sex-related disparities in cough-associated symptoms across different age groups by Jiyeon Kang, Woo Jung Seo, Jung Gon Kim, Ji-Yong Moon, Deog Kyeom Kim, Jin Woo Kim, Seung Hun Jang, Jae-Woo Kwon, Byung-Jae Lee and Hyeon-Kyoung Koo in Therapeutic Advances in Respiratory Disease

Supplemental Material

sj-docx-2-tar-10.1177_17534666241252545 – Supplemental material for Sex-related disparities in cough-associated symptoms across different age groups

Supplemental material, sj-docx-2-tar-10.1177_17534666241252545 for Sex-related disparities in cough-associated symptoms across different age groups by Jiyeon Kang, Woo Jung Seo, Jung Gon Kim, Ji-Yong Moon, Deog Kyeom Kim, Jin Woo Kim, Seung Hun Jang, Jae-Woo Kwon, Byung-Jae Lee and Hyeon-Kyoung Koo in Therapeutic Advances in Respiratory Disease

Footnotes

Acknowledgements

None.

Declaration

ORCID iD

Hyeon-Kyoung Koo

Supplemental material

Supplemental material for this article is available online.

References

Andrani

Aiello

Bertorelli

, et al. Cough, a vital reflex. Mechanisms, determinants and measurements. Acta Biomed 2019; 89: 477–480.

Irwin

Boulet

Cloutier

, et al. Managing cough as a defense mechanism and as a symptom. A consensus panel report of the American College of Chest Physicians. Chest 1998; 114: 133s–181s.

Spanevello

Beghé

Visca

, et al. Chronic cough in adults. Eur J Intern Med 2020; 78: 8–16.

Won

Song

. Impact and disease burden of chronic cough. Asia Pac Allergy 2021; 11: e22.

Bai

Sha

, et al. Gender difference in chronic cough: are women more likely to cough? Front Physiol 2021; 12: 654797.

Baiardini

Braido

Fassio

, et al. A new tool to assess and monitor the burden of chronic cough on quality of life: Chronic Cough Impact Questionnaire. Allergy 2005; 60: 482–488.

Birring

Prudon

Carr

, et al. Development of a symptom specific health status measure for patients with chronic cough: Leicester Cough Questionnaire (LCQ). Thorax 2003; 58: 339–343.

Faruqi

Thompson

Wright

, et al. Quantifying chronic cough: objective versus subjective measurements. Respirology 2011; 16: 314–320.

Zhang

Sykes

Brindle

, et al. Chronic cough-the limitation and advances in assessment techniques. J Thorac Dis 2022; 14: 5097–5119.

10.

Barton

Gaydecki

Holt

, et al. Data reduction for cough studies using distribution of audio frequency content. Cough 2012; 8: 12.

11.

Spinou

Birring

. An update on measurement and monitoring of cough: what are the important study endpoints? J Thorac Dis 2014; 6: S728–S734.

12.

Lee

Davenport

Smith

, et al.; CHEST expert cough panel. Global physiology and pathophysiology of cough: Part 1: cough phenomenology – CHEST Guideline and Expert Panel Report. Chest 2021; 159: 282–293.

13.

Kwon

Moon

Kim

, et al. Reliability and validity of a Korean version of the Leicester cough questionnaire. Allergy Asthma Immunol Res 2015; 7: 230–233.

14.

Martin

Harrison

. Causes of chronic productive cough: an approach to management. Respir Med 2015; 109: 1105–1113.

15.

Morice

Dicpinigaitis

McGarvey

, et al. Chronic cough: new insights and future prospects. Eur Respir Rev 2021; 30: 210127.

16.

Chung

McGarvey

Song

, et al. Cough hypersensitivity and chronic cough. Nat Rev Dis Primers 2022; 8: 45.

17.

Goldsobel

Kelkar

. The adult with chronic cough. J Allergy Clin Immunol 2012; 130: 825–825.e826

18.

Hulme

Deary

Dogan

, et al. Psychological profile of individuals presenting with chronic cough. ERJ Open Res 2017; 3(1): 00099–2016.

19.

Koo

Bae

Moon

, et al. Differential features of chronic cough according to etiology and the simple decision tree for predicting causes. Sci Rep 2021; 11: 10326.

20.

Joo

Moon

, et al. Revised Korean Cough Guidelines, 2020: recommendations and summary statements. Tuberc Respir Dis (Seoul) 2021; 84(4): 263–273.

21.

Rhee

Jung

Lee

, et al. The Korean cough guideline: recommendation and summary statement. Tuberc Respir Dis (Seoul) 2016; 79: 14–21.

22.

Vandenbroucke

von Elm

Altman

, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med 2007; 4: e297.

23.

Zou

. Toward using confidence intervals to compare correlations. Psychol Methods 2007; 12: 399–413.

24.

Fujimura

Kasahara

Kamio

, et al. Female gender as a determinant of cough threshold to inhaled capsaicin. Eur Respir J 1996; 9: 1624–1626.

25.

Kelsall

Decalmer

McGuinness

, et al. Sex differences and predictors of objective cough frequency in chronic cough. Thorax 2009; 64: 393–398.

26.

Bewley

McCartney

Meads

, et al. Sex, gender, and medical data. BMJ 2021; 372: n735.

27.

Regensteiner

Libby

Huxley

, et al. Integrating sex and gender considerations in research: educating the scientific workforce. Lancet Diabetes Endocrinol 2019; 7: 248–250.

28.

Mazzone

Chung

McGarvey

. The heterogeneity of chronic cough: a case for endotypes of cough hypersensitivity. Lancet Respir Med 2018; 6: 636–646.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.66 MB

0.03 MB