Characteristics of the Study Participants
The baseline characteristics of the enrolled participants are shown in Table 1. The 4,217 participants had a mean age of 59.5 ± 11.8 years, of which 42.5% were male. The mean BMI was 24.1 ± 3.4 kg/m2, and only 2.8% of all participants had a normal body weight. A total of 1,651 (39.2%) participants had an ever-smoking history, and 629 (14.9%) participants were current smokers. Of the total participants, 2.9%, 11.6%, and 5.5% had a history of asthma, allergic rhinitis, and sinusitis, respectively. Moreover, 31.8%, 26.2%, and 42.0% had normal blood pressure, pre-hypertension, and hypertension, respectively. Diabetes and pre-diabetes were observed in 18.6% and 49.1% of participants, respectively. The pulmonary function characteristics of the participants were as follows: the mean forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and FEV1/FVC values were 87.3 ± 12.7% of the predicted value, 87.4 ± 13.5% of the predicted value, and 0.77 ± 0.07, respectively. Regarding socioeconomic factors, low household income accounted for 22.9% of the total, and the unemployed or economically inactive population accounted for 41.1% of the total participants. Of the participants, 37.3% had less than 9 years of lower educational level.
Table 1
Variables | Persons without chronic cough (n = 4,094) | Persons with chronic cough (n = 123) | P value |
Age (years) | | | < 0.001*** |
40–49 | 1,025 (25.0) | 21 (17.1) | |
50–59 | 1,098 (26.8) | 15 (12.2) | |
60–69 | 1,020 (24.9) | 39 (31.7) | |
≥ 70 | 951 (23.2) | 48 (39.0) | |
Sex | | | 0.116 |
Male | 1,731 (42.3) | 61 (49.6) | |
Female | 2,363 (57.7) | 62 (50.4) | |
BMI (kg/m2) | | | 0.118 |
Underweight (< 18.5) | 1,508 (36.8) | 48 (39.0) | |
Normal (18.5–22.9) | 113 (2.8) | 6 (4.9) | |
Overweight (23–24.9) | 999 (24.4) | 37 (30.1) | |
Obese I (25.0–29.9) | 1,266 (30.9) | 26 (21.1) | |
Obese II (≥ 30) | 207 (5.1) | 6 (4.9) | |
Smoking habit | | | |
Never-smoker | 2,493 (61.0) | 64 (52.0) | |
Ever-smoker | 1,592 (39.0) | 59 (48.0) | 0.049* |
Current smoker | 592/4085 (14.5) | 37/123 (30.1) | < 0.001*** |
Hypertension | | | 0.191 |
Normal | 1,312 (32.0) | 30 (24.4) | |
Pre-hypertension | 1,067 (26.1) | 37 (30.1) | |
Hypertension | 1,715 (41.9) | 56 (45.5) | |
Diabetes mellitus | | | < 0.001*** |
Normal | 1,272 (32.6) | 28 (24.6) | |
Pre-diabetes | 1,911 (48.9) | 62 (54.4) | |
Diabetes mellitus | 724 (18.5) | 24 (21.1) | |
Asthma history | | | < 0.001*** |
No asthma history | 3984 (97.3) | 109 (88.6) | |
Asthma history | 110 (2.7) | 14 (11.4) | |
Allergic rhinitis history | | | 1.000 |
No Allergic rhinitis history | 3618 (88.4) | 109 (88.6) | |
Allergic rhinitis history | 476 (11.6) | 14 (11.4) | |
Sinusitis history | | | 0.547 |
No sinusitis history | 3871 (94.6) | 115 (93.5) | |
Sinusitis history | 223 (5.4) | 8 (6.5) | |
Lung function tests | | | |
FVC (L) | 3.28 ± 0.85 | 3.04 ± 0.87 | 0.004** |
FVC (% of predicted) | 87.5 ± 12.7 | 82.6 ± 13.8 | < 0.001*** |
FEV1 (L) | 2.53 ± 0.67 | 2.15 ± 0.72 | < 0.001*** |
FEV1 (% of predicted) | 87.6 ± 13.3 | 78.6 ± 17.7 | < 0.001*** |
≥ 80 | 2,656 (74.4) | 55 (53.4) | |
≥ 50–<80 | 887 (24.8) | 42 (40.8) | |
≥ 30–<50 | 24 (0.7) | 4 (3.9) | |
< 30 | 4 (0.1) | 2(1.9) | |
House income | | | < 0.001*** |
High | 1,117 (27.4) | 20 (16.4) | |
Middle–high | 989 (24.2) | 19 (15.6) | |
Middle–low | 1,058 (25.9) | 36 (29.5) | |
Low | 916 (22.5) | 47 (38.5) | |
Economic activity | | | < 0.001*** |
Employed | 2,431 (59.4) | 53 (43.1) | |
Unemployed, inactive | 1,662 (40.6) | 70 (56.9) | |
Educational level | | | 0.038* |
High (> 9years) | 2,576 (63.0) | 66 (53.7) | |
Low (≤ 9 years) | 1,515 (37.0) | 57 (46.3) | |
Abbreviations: |
BMI, body mass index; FVC, forced vital capacity; FEV1, forced expiratory volume in 1 s |
Patients in the chronic cough group were older than those in the group without chronic cough (P < 0.001). There was no difference in the distribution of sex, BMI, and prevalence of hypertension between the two groups (P > 0.05). Participants with a chronic cough had a higher smoking history (P = 0.049) and a higher prevalence of diabetes (P < 0.001). A history of asthma was more common in the chronic cough group (P < 0.001), but the frequencies of allergic rhinitis and sinusitis did not differ between the two groups. Overall, compared with 74.4% of participants without chronic cough, only 53.4% of participants with chronic cough had normal lung function (P < 0.001). Participants with chronic cough had a lower educational background (P = 0.038), household income level (P < 0.001), and economic activity (P < 0.001) than those without chronic cough.
Group at High-risk of OSA
The distribution of the STOP-Bang questionnaire results is shown in Fig. 2 and Table 2. The total score of the STOP-Bang questionnaire was mean 2.33 ± 1.38 and the chronic cough group had a higher score than the other group without chronic cough (P < 0.001). The group at high-risk of OSA, with a score of 3 or higher, comprised 52% and 40.4% of all participants, depending on the presence or absence of chronic cough (P = 0.012). In participants with chronic cough, the frequency of group at very high-risk for OSA was higher than in participants without the symptom (P = 0.005).
Table 2
Group at high-risk for OSA by the STOP-BANG questionnaire
STOP-BANG questionnaire variables | Person without chronic cough (n = 4,094) | Person with chronic cough (n = 123) | P value |
STOP-Bang | | | |
Total score (0–8) | 2.32 ± 1.38 | 2.80 ± 1.39 | < 0.001*** |
Group at high-risk of obstructive sleep apnea | | | |
≥ 3 points in STOP-Bang | 1,656 (40.4) | 64 (52.0) | 0.012* |
≥ 5 points in STOP-Bang | 270 (6.6) | 17 (13.8) | 0.005** |
Note: Data are presented as number (%) or mean ± standard deviation, unless otherwise indicated. |
*P < 0.05, **P < 0.01, ***P < 0.001 |
Among the eight items of the STOP-Bang questionnaire, the presence of snoring, tiredness (or sleepiness), and older age (over 50 years old) showed differences between the groups with and without chronic cough. Persons with these three items were more common in the chronic cough group. However, there were no significant differences in the observed apneas (choking), hypertension, obesity, thick neck circumference (> 40 cm), and sex.
Association between High-risk of OSA and Chronic Cough
The risk factors associated with chronic cough are shown in Table 3. In the univariate analysis, old age (60–69 years, P = 0.023; ≥70, P = 0.001), current smoking (P < 0.001), asthma history (P < 0.001), poor lung function (FEV1 ≥ 50–<80, P < 0.001; <50, P < 0.001), low household income (P < 0.001), low-grade educational level (P = 0.036), reduced economic activity (P < 0.001), and a very high-risk of OSA (P = 0.001) were associated with chronic cough. There was no relationship between chronic cough and sex, BMI, history of allergic rhinitis or sinusitis, hypertension, or diabetes.
Table 3
Risk factors for chronic cough.
| Univariate analysis | | Multivariate analysis | |
Variables | Exp. (95% CI) | P value | Exp. (95% CI) | P value |
Age (years) | | | | |
40–49 | 1.000 | | 1.000 | |
50–59 | 0.667 (0.342–1.301) | 0.234 | 0.594 (0.284–1.244) | 0.167 |
60–69 | 1.866 (1.090–3.195) | 0.023* | 1.315 (0.673–2.567) | 0.423 |
≥ 70 | 2.464 (1.464–4.145) | 0.001** | 2.396 (1.192–4.816) | 0.014* |
Sex | | | | |
Female | 1.000 | | 1.000 | |
Male | 1.343 (0.938–1.923) | 0.107 | 0.603 (0.349–1.042) | 0.070 |
BMI (kg/m2) | | | | |
Underweight (< 18.5) | 1.668 (0.699–3.982) | 0.249 | 1.727 (0.613–4.865) | 0.301 |
Normal (18.5–22.9) | 1.000 | | 1.000 | |
Overweight (23–24.9) | 1.164 (0.752–1.800) | 0.496 | 0.925 (0.548–1.559) | 0.769 |
Obese I (25.0–29.9) | 0.645 (0.398–1.046) | 0.075 | 0.517 (0.293–0.913) | 0.023* |
Obese II (≥ 30) | 0.911 (0.385–2.154) | 0.831 | 0.549 (0.194–1.557) | 0.260 |
Smoking habit | | | | |
No-smoker | 1.000 | | 1.000 | |
Current-smoker | 2.539 (1.710–3.768) | < 0.001*** | 3.080 (1.829–5.186) | < 0.001*** |
Hypertension | | | | |
Normal | 1.000 | | 1.000 | |
Pre-hypertension | 1.517 (0.931–2.471) | 0.095 | 1.190 (0.674–2.099) | 0.549 |
Hypertension | 1.428 (0.911–2.238) | 0.120 | 0.591 (0.318–1.096) | 0.095 |
Diabetes mellitus | | | | |
Normal | 1.000 | | | |
Pre-diabetes | 1.474 (0.938–2.316) | 0.092 | | |
Diabetes mellitus | 1.506 (0.866–2.617) | 0.147 | | |
Asthma history | | | | |
No asthma history | 1.000 | | 1.000 | |
Asthma history | 4.652 (2.584–8.374) | 0.000*** | 2.606 (1.124–6.042) | 0.026* |
Allergic rhinitis history | | | | |
No AR history | 1.000 | | | |
AR history | 0.976 (0.555–1.717) | 0.934 | | |
Sinusitis history | | | | |
No sinusitis history | 1.000 | | | |
Sinusitis history | 1.208(0.582–2.504) | 0.612 | | |
FEV1 (% of predicted) | | | | |
≥ 80 | 1.000 | | 1.000 | |
≥ 50–<80 | 2.287 (1.519–3.441) | < 0.001*** | 1.936 (1.249–3.001) | 0.003** |
< 50 | 10.348 (4.119–25.999) | < 0.001*** | 5.327 (1.853–15.314) | 0.002** |
House income | | | | |
High | 1.000 | | | |
Middle–high | 1.073 (0.569–2.022) | 0.828 | | |
Middle–low | 1.900 (1.093–3.304) | 0.023 | | |
Low | 2.866 (1.686–4.871) | < 0.001*** | | |
Educational level | | | | |
High (> 9years) | 1.000 | | | |
Low (≤ 9 years) | 1.468 (1.024–2.105) | 0.036* | | |
Economic activity | | | | |
Employed | 1.000 | | 1.000 | |
Unemployed, inactive | 1.932 (1.345–2.775) | < 0.001*** | 1.660 (1.047–2.633) | 0.031* |
STOP-Bang | | | | |
0–2 | 1.000 | | 1.000 | |
3–4 | 1.401 (0.950–2.067) | 0.089 | 2.230 (1.251–3.975) | 0.007** |
5–8 | 2.602 (1.495–4.527) | 0.001** | 6.614 (2.816–15.537) | < 0.001*** |
Abbreviations: AR, allergic rhinitis; BMI, body mass index; FEV1, forced expiratory volume in 1 s |
*P < 0.05, **P < 0.01, ***P < 0.001 |
Subsequently, multivariate analysis showed that advanced age (≥ 70 years, P = 0.014), current smoking status (P < 0.001), asthma history (P = 0.026), poor lung function (FEV1 ≥ 50 to < 80, P = 0.003; <50, P < 0.002), impaired economic activity (P = 0.031), and a high-risk of OSA (STOP-Bang score 3–4, P = 0.007; 5–8, P < 0.001) were associated with chronic cough. Conversely, obesity I showed a negative relationship with the occurrence of chronic cough (P = 0.023). The risk-adjusted analysis showed that household income and education levels lost their relationship with chronic cough.