The prevalence of rhinitis and its association with smoking and obesity in a nationwide survey of Japanese adults
Edited by: Wytske Fokkens
Satoshi Konno, First Department of Medicine, School of Medicine, Hokkaido University, Kita-15 Nishi-7 Kita-Ku, Sapporo 060-8638, Japan.
Tel.: +81 11 716 1161 (Ext. 5911)
Fax: +81 11 706 7899
Rhinitis is a common disease, and its prevalence is increasing worldwide. Several studies have provided evidence of a strong association between asthma and rhinitis. Although smoking and obesity have been extensively analyzed as risk factors of asthma, associations with rhinitis are less clear.
The aims of our study were (i) to evaluate the prevalence of rhinitis using the European Community Respiratory Health Survey (ECRHS) questionnaire in Japanese adults and (ii) to evaluate the associations of smoking and body mass index (BMI) with rhinitis.
Following our study conducted in 2006–2007 to determine the prevalence of asthma using the ECRHS questionnaire, our present analysis evaluates the prevalence of rhinitis and its association with smoking and BMI in Japanese adults 20–79 years of age (N = 22819). We classified the subjects (20–44 or 45–79 years) into four groups as having (i) neither rhinitis nor asthma; (ii) rhinitis without asthma; (iii) asthma without rhinitis; or (iv) rhinitis with asthma. We then evaluated associations with smoking and BMI in each group.
The overall age-adjusted prevalence of rhinitis was 35.1% in men and 39.3% in women. A higher prevalence was observed in the younger population than in the older population. Active smoking and obesity were positively associated with asthma without rhinitis. In contrast, particularly in the 20- to 44-year age-group, active smoking and obesity were negatively associated with rhinitis without asthma.
The results of the present study suggest that smoking and obesity may have different effects on the development of rhinitis and asthma.
Rhinitis, particularly allergic rhinitis, is one of the most common respiratory disorders. Although rhinitis is not a life-threatening condition, it has a huge socioeconomic impact, as it affects people at the ages normally associated with active employment or attending school. Accordingly, epidemiologic studies for evaluating the prevalence of rhinitis and associated factors are required.
Several epidemiologic studies of the general population have provided evidence of a strong association between asthma and rhinitis [1-3]. Practice guidelines for asthma care reinforce the importance of identifying and treating rhinitis in patients with asthma [4, 5]. Although it was initially believed that a common atopic background accounts for the increased risk of patients with rhinitis developing asthma, recent studies have demonstrated that rhinitis is also a risk factor for asthma in nonatopic patients . These results suggest that rhinitis and asthma could be manifestations of the same disease, irrespective of atopic status, and that they represent a continuum, sharing common pathologic and physiologic characteristics. This phenomenon has been labeled ‘one airway one disease’ or ‘united airways disease’.
Conversely, there also exist major differences between the upper and lower airways [6, 7]. Important differences between pulmonary and nasal responses include the residence time of inflammatory cells, chemical mediators, and the mechanisms regulating repair of the epithelium following an inflammatory event. Bronchi are characterized by the presence of smooth muscle, which is responsible for bronchoconstriction. The remodeling of the airways is less extensive in the nose than in the bronchial wall.
Both genetic and environmental factors play important roles in the etiology of rhinitis and asthma. Among various environmental factors, smoking and obesity have been extensively analyzed as risk factors of asthma [8-13]. There is accumulating evidence that exposure to environmental tobacco smoke is associated with the development of asthma in children [8, 9]. Several studies have also suggested that active smoking is associated with the development of asthma in adults [10, 11]. However, the association of rhinitis with smoking and obesity is less clear. Of note, recent studies have shown an inverse association of rhinitis with smoking and obesity [3, 14, 15].
In the present analysis, following our study conducted in 2006–2007 for evaluation of the prevalence of asthma and its risk factors [16, 17], we evaluated the prevalence of rhinitis using the European Community Respiratory Health Survey (ECRHS) questionnaire, which has been used in several studies for the nationwide analysis of the prevalence of rhinitis [18-22]. In addition, we evaluated the association of smoking and body mass index (BMI) with rhinitis, considering the coexistence of rhinitis and asthma. This study was approved by the Ethics Committee of Sagamihara National Hospital and Hokkaido University.
Study design and questionnaire
A population-based, cross-sectional study was conducted with Japanese subjects 20–79 years of age, living in ten different areas of Japan. Detailed methods for selecting areas, participants, questionnaire distribution, and response rates in each area were described in our previous reports [16, 17]. We classified subjects as having rhinitis if they responded affirmatively to the question, ‘Do you have any nasal allergies including hay fever?’ We classified subjects as having a wheeze if they responded affirmatively to the question, ‘Have you had wheezing or whistling in your chest at any time in the last 12 months?’ This question is used worldwide for the evaluation of the prevalence of asthma for subjects aged 20–44 years [23-26]. We also classified patients as having asthma if they met following two criteria [17, 25]: (i) an affirmative response to the question, ‘Have you ever had asthma?’ followed by ‘Was this confirmed by a doctor?’ and (ii) having at least one asthma-related symptom in the last 12 months. A subject was considered to have asthma-related symptoms if he or she answered in the affirmative to at least one of the following four questions: (i) ‘Have you had wheezing or whistling in your chest at any time in the last 12 months?’, (ii) ‘Have you woken up with a feeling of tightness in your chest at any time in the last 12 months?’, (iii) ‘Have you been woken up by an attack of shortness of breath at any time in the last 12 months?’, and (iv) ‘Have you been woken up by an attack of coughing at any time in the last 12 months?’ We classified subjects according to their smoking habits as ‘active smokers’ if they were current smokers and had smoked at least one cigarette per day or one cigar a week for the past year, or ‘past smokers’ if they had smoked for at least 1 year (as defined above) but not during the last month. All other subjects were considered nonsmokers. Body mass index was categorized into four groups: <18.50, 18.50–24.99, 25.00–29.99, and ≥30.00 kg/m2. This study was conducted from July to October 2006, except in Mitake Town, Gifu, where the study was conducted from January to February 2007.
A total of 22819 subjects who responded to all questions regarding gender, age, the presence of wheeze, asthma, and rhinitis, smoking habit, height, and body weight were analyzed. Statistical analyses were performed using the statistical software package systat for Windows, version 11 (SYSTAT, San Jose, CA, USA). Fisher's exact tests were carried out to assess the differences in prevalence between men and women. Univariate and multivariate logistic regression models were used to calculate adjusted odds ratios (OR) and 95% confidence intervals (CI). To assess the differences in prevalence between areas, the prevalence in each area was adjusted to a standard population with an equal distribution by age using age-groups 20–29, 30–39, 40–49, 50–59, 60–69, and 70–79 for men and women separately. Age-specific prevalence of rhinitis with 95% CI was also calculated to explore the effects of age. For all statistical analyses, a P value <0.05 was considered significant.
Table 1 shows the age-standardized prevalence of rhinitis in ten areas, and Table 2 shows the prevalence of rhinitis by age and gender in all subjects. The overall age-adjusted prevalence of rhinitis was 35.1% (95%CI, 34.2–35.9) for men and 39.3% (38.4–40.2) for women. A gradual decrease in the prevalence of rhinitis in subjects >20 years of age in men and in subjects >50 years of age in women was observed (Table 2). The prevalence was lowest in Kamishihoro Town, Hokkaido, probably due to reduced levels of cedar pollen in Hokkaido (the north island of Japan), which is a major cause of pollinosis in Japan (Table 1). In the age-groups 30–39, 40–49, 50–59, and 60–69, the prevalence of rhinitis was significantly higher among women than among men (Table 2). To adjust for the effect of gender differences in smoking prevalence on prevalence of rhinitis , the prevalence of rhinitis was calculated after limiting the subjects to nonsmokers (N = 13610). The prevalence of rhinitis among nonsmokers was not significantly different between genders in all age-groups (Table 2).
Table 1. Age-standardized prevalence of rhinitis per area
|Kamishihoro Town, Hokkaido||3016||17.7 (15.8–19.6)||18.8 (16.8–20.8)||23.1 (21.0–25.2)||24.0 (21.8–26.1)|
|Fuchu Town, Toyama||2814||30.0 (27.6–32.4)||29.1 (26.7–31.5)||29.0 (26.6–31.4)||28.7 (26.4–31.1)|
|Setagaya Ward, Tokyo||1774||48.9 (45.6–52.2)||47.3 (44.0–50.6)||49.3 (46.0–52.6)||48.3 (45.0–51.5)|
|Sagamihara City, Kanagawa||3523||43.3 (41.0–45.6)||44.3 (42.0–46.6)||50.3 (48.0–52.6)||50.4 (48.0–52.7)|
|Fujieda City, Shizuoka||2534||38.1 (40.8–45.4)||38.3 (35.6–41.0)||46.8 (44.1–49.5)||46.4 (43.7–49.1)|
|Mitake Town, Gifu||1490||40.7 (37.1–44.3)||40.0 (36.5–43.5)||43.7 (40.2–47.2)||41.5 (38.0–45.0)|
|Nagakute Town, Aichi||1232||41.7 (37.6–45.8)||41.3 (37.3–45.4)||47.4 (43.6–51.2)||44.9 (41.1–48.6)|
|Akiohta Town, Hiroshima||1959||31.2 (28.2–34.2)||32.6 (29.6–35.6)||34.0 (31.1–36.9)||40.2 (37.2–43.2)|
|Kurashiki Town, Okayama||2259||33.0 (30.2–35.8)||32.9 (30.1–35.7)||37.9 (35.1–40.7)||37.8 (35.0–40.6)|
|Nangoku City, Kochi||2218||31.2 (28.4–34.0)||31.0 (28.3–33.8)||34.0 (31.3–36.7)||34.7 (32.0–37.5)|
Table 2. Prevalence of rhinitis by gender and age
|All subjects (N = 22819)|
|20–29||48.2 (45.7–50.7)||47.8 (45.2–50.4)||0.83|
|30–39||45.2 (42.9–47.4)||49.7 (47.5–51.9)||0.0050|
|40–49||40.9 (38.6–43.2)||50.0 (47.8–52.2)||<0.001|
|50–59||32.9 (31.0–34.9)||38.7 (36.8–40.6)||<0.001|
|60–69||25.2 (23.4–27.1)||29.6 (27.7–31.4)||0.0015|
|70–79||18.0 (16.1–19.9)||20.1 (18.2–22.0)||0.13|
|Nonsmokers (N = 13610)|
|20–29||49.5 (45.9–53.1)||48.2 (45.2–51.2)||0.60|
|30–39||49.6 (45.6–53.6)||50.0 (47.5–52.6)||0.85|
|40–49||49.7 (45.4–54.0)||50.1 (47.6–52.6)||0.87|
|50–59||40.6 (36.8–44.3)||39.1 (37.0–41.1)||0.50|
|60–69||27.1 (23.9–30.3)||29.7 (27.7–31.7)||0.19|
|70–79||17.5 (14.5–20.5)||20.0 (18.0–22.0)||0.19|
As rhinitis and asthma have been reported to coexist [1-3], associations between rhinitis and the presence of wheeze/asthma were analyzed in this population of survey respondents. Positive associations between rhinitis and wheeze/asthma were shown among all respondents 20–44 years of age (P < 0.001) and 45–79 years of age (P < 0.001) (Table 3). A similar association was also observed in nonsmokers (P < 0.001) (Table 3).
Table 3. Association of rhinitis and the presence of wheeze or asthmaa
|All subjects (N=8563)||2.29 ||1.96–2.67||2.90 ||2.38–3.55|
|Nonsmokers (N=4892)||2.74 ||2.16–3.47||3.14 ||2.34–4.20|
|All subjects (N=14256)||2.05 ||1.83–2.29||3.36 ||3.04–4.37|
|Nonsmokers (N=8718)||2.50 ||2.14–2.92||4.43 ||3.48–5.62|
Because significant coexistence of rhinitis and wheeze/asthma was observed, we classified all subjects into four groups and evaluated the pure association of rhinitis with smoking and obesity: (i) none of the conditions [Rhinitis(−) Wheeze(−); N = 13267], [Rhinitis(−) Asthma(−); N = 14025]; (ii) rhinitis without wheeze/asthma [Rhinitis(+) Wheeze(−); N = 7263], [Rhinitis(+) Asthma(−); N = 7767]; (iii) wheeze/asthma without rhinitis [Rhinitis(−) Wheeze(+); N = 1126], [Rhinitis(−) Asthma(+); N = 368]; (iv) rhinitis with wheeze/asthma [Rhinitis(+) Wheeze(+); N = 1163], [Rhinitis(+) Asthma(+); N = 659]. Table 4 shows the characteristics of the subjects in each group.
Table 4. Characteristics of subjects in the eight groups
|55 (20–79)||45 (20–79)||59 (20–79)||47 (20–79)||56 (20–79)||46 (20–79)||53 (20–79)||43 (20–79)|
|Smoking status (n)|
Tables 5, and 6, Tables S1 and S2 show the results of the multivariate logistic regression analysis for the association of smoking status and BMI using the outcome variables ‘rhinitis without wheeze/asthma’, ‘wheeze/asthma without rhinitis’, and ‘rhinitis and wheeze/asthma’ in the 20–44-year (Table 5, Table S1) and 45–79-year (Table 6, Table S2) age-groups. Active smoking was positively associated with wheeze without rhinitis in both age-groups (20–44 years: OR = 2.36, P < 0.001; 45–79 years: OR = 2.12, P < 0.001). In contrast, being an active smoker was negatively associated with rhinitis without wheeze (20–44 years: OR = 0.78, P < 0.001; 45–79 years: OR = 0.57, P < 0.001). Being overweight (BMI, 25.00–29.99) or obese (BMI ≥ 30.00) was positively associated with wheeze without rhinitis (obesity: OR = 2.40, P < 0.001), but negatively associated with rhinitis without wheeze (overweight: OR = 0.80, P = 0.0028) in the 20–44-year age-group. When the same analysis was performed after limiting subjects to lifetime nonsmokers, a similar association of high BMI and rhinitis without wheeze (overweight: OR = 0.77, P = 0.015; obesity: OR = 0.56, P = 0.0043) was observed. When asthma was used instead of wheeze, similar results were obtained regarding the positive association between smoking and asthma without rhinitis, and the negative association between obesity and rhinitis without asthma in subjects 20–44 years of age (Table 5, Table S1).
Table 5. Association of smoking status or BMI with the presence of rhinitis and/or asthma (20–44 years)
|All subjects (N = 8563)|
|Nonsmokers (N = 4892)|
Table 6. Association of smoking status or BMI with the presence of rhinitis and/or asthma (45–79 years)
|All subjects (N = 14256)|
|Nonsmokers (N = 8718)|
Consistent with the results from a number of earlier studies [1-3], a strong association between rhinitis and the presence of wheeze/asthma was observed in this study, supporting the hypothesis that rhinitis and asthma represent a continuum of the same disease. Of note, significant association between two conditions was observed even in elderly subjects (45–79 years). Although it was initially believed that a common atopic background accounts for the increased risk of asthma development in patients with rhinitis, recent studies have demonstrated that rhinitis is also a risk factor for asthma in nonatopic patients . Moreover, recent studies report a higher prevalence of sinonasal symptoms even in chronic obstructive pulmonary disease (COPD) [27-29]. Collectively, the co-occurrence of rhinitis and the presence of wheeze/asthma observed both in younger (20–44 years) and older (45–79 years) adults in this study could be explained not only by the underlying common allergic components of rhinitis and asthma, but also by the common pathogenesis of upper and lower airway inflammation unrelated to atopy and allergy.
Although this study revealed the coexistence of rhinitis and wheeze/asthma, their association with the potential risk factors of active smoking and obesity opposed one another. In contrast to the results from several studies showing positive association between smoking and asthma [10, 11], the present study demonstrates that active smoking has a negative correlation with rhinitis. Consistent with our results, the protective effect of smoking on the development of cedar allergy has been reported in the Japanese population . Tobacco smoke has been shown to suppress human immunity, including inhibition of cytokine production and T-cell responsiveness [30, 31]. Accordingly, the immunosuppressive effect of smoking might result in the low prevalence of rhinitis in active smokers.
Several in vivo studies have shown that short-term smoking or administration of nicotine attenuated allergic airway inflammation in mice [32, 33], whereas smoking enhanced airway hyperresponsiveness in other reports [34, 35]. Botelho et al. have recently shown that smoking attenuated eosinophilic airway inflammation, while enhancing airway remodeling in a house dust mite (HDM)-induced asthma model in mice . Taken together, factors other than allergic responses, such as the induction of airway hyperreactivity or remodeling, which could directly influence the airways, might account for the effect of smoking on asthma.
In the present study, the presence of rhinitis was higher among women than in men in all subjects. However, when subjects were limited to lifetime nonsmokers, no significant difference was observed between genders (Table 2), suggesting that the higher prevalence of rhinitis among women in all subjects was influenced by the lower smoking rate among women. Therefore, when the prevalence of rhinitis is evaluated in epidemiologic surveys, smoking status needs to be considered.
Another interesting and unexpected finding of this study is that obesity had a negative association with rhinitis in subjects 20–44 years of age. Although the association between obesity and asthma has been gaining more attention, few studies have been conducted concerning the relationship between obesity and other allergic diseases. Of note, a recent large survey of school children in Japan showed that the obesity was negatively associated with the prevalence of rhinitis , which is consistent with our results. A recent report by Johnston et al.  showed enhanced airway responsiveness with attenuation of airway inflammation in obese mice. These results may support our findings of opposing effects of obesity on asthma and rhinitis, similar to the opposite effects of smoking on both disease conditions.
Several previous reports have shown results contradicting those in our study regarding the association between smoking status/obesity and rhinitis [37-39]. The exact reasons for this inconsistency are unclear; however, unlike the current study, most previous studies did not consider the coexistence of rhinitis and asthma. One previous study that did consider the coexistence of rhinitis and asthma was conducted in Italy using the ECRHS questionnaire and, similar to our study, found a negative association between active smoking and rhinitis . Thus, consideration of the co-occurrence of asthma and rhinitis might have contributed to revealing the interesting negative association between rhinitis and active smoking/obesity in the current study.
A number of limitations to this study exist, as previously described [16, 17]. The cross-sectional nature of the study precludes establishing temporal relationships or inferring causality. Thus, the possibility that subjects with rhinitis might be less likely to smoke or tend not to exercise resulting in obesity could be another explanation for the negative association between rhinitis and active smoking/obesity. The number of risk factors was limited, and we did not consider the possible contributions of other relevant variables, such as economic status, occupation, and passive smoking. Questions regarding the presence of nasal allergies in the ECRHS questionnaire are ambiguous, as some patients with nonallergic rhinitis might respond affirmatively, and this question cannot distinguish between seasonal and perennial rhinitis. In addition, subjects who previously had, but do not currently have, rhinitis symptoms might also have been included in the rhinitis group. Furthermore, as the reliability of the ECRHS questionnaire with subjects >45 years of age has not been widely evaluated, the associations between rhinitis and active smoking/obesity among older subjects in this study need to be cautiously interpreted. To confirm our results, further studies are needed using a prospective design with precise definitions of asthma and rhinitis, particularly for older subjects, and more confounding factors need to be considered for their possible associations with asthma and rhinitis.
We wish to emphasize that our results are not meant to change policies regarding smoking and weight management. Because of the numerous possible adverse effects, smoking and obesity should not be considered therapeutic options. Our findings could provide clues to understanding the pathogenesis of rhinitis and for therapeutic strategies for protection against rhinitis. The results of this study also encourage a more intensive investigation into the relationship between smoking and rhinitis.
In conclusion, this cross-sectional study using the ECRHS questionnaire determined the prevalence of rhinitis in Japanese population. The results of the present study suggest that smoking and obesity may have different effects on the development of rhinitis and asthma.
This work was supported by Health and Labor Science Research Grants from the Ministry of Health, Labor, and Welfare, Japan.
Masami Taniguchi, Kiyoshi Takahashi, Akira Akasawa, Kazuo Akiyama, and Masaharu Nishimura involved in the design of the protocol. Satoshi Konno and Nobuyuki Hizawa involved in the analysis and writing. The other authors participated in the design of the protocol and drafting of the paper.
Conflicts of interests
The authors declare no conflicts of interest.