Prevalence, Awareness, and Associated Factors of Airflow Obstruction in Russia: The Ural Eye and Medical Study

Background: Although chronic obstructive pulmonary disease and asthma belong to the most important causes of disability and death in all world regions, data about the prevalence of airflow obstruction and asthma in Russia and the associated parameters have been scarce so far. We therefore assessed the prevalence of airflow obstruction and asthma in a Russian population. Methods: The population-based Ural Eye and Medical Study, conducted in a rural and urban region of Bashkortostan/Russia, included 5,392 participants (mean age: 58.6 ± 10.6 years; range: 40–94 years) out of 7,328 eligible individuals. Airflow obstruction was defined spirometrically and asthma by self-reported diagnosis. Results: Airflow obstruction was present in 369 individuals (6.8%; 95% confidence interval (CI): 6.2, 7.5) with an awareness rate of 63.4% (95%CI: 58.5, 68.4) and known duration of 19.5 ± 15.8 years (median: 16 years). Prevalence of undiagnosed airflow obstruction was 2.6% (95%CI: 2.2, 3.1). Higher prevalence of airflow obstruction was associated (multivariable analysis) with higher prevalence of current smoking [P < 0.001; odds ratio (OR): 2.91; 95%CI: 1.76, 4.83] and number of cigarette package years (P < 0.001; OR: 1.03; 95%CI: 1.02, 1.08), female gender (P = 0.03; OR: 1.42; 95%CI: 1.04, 1.93), urban region (P = 0.003; OR: 1.43; 95% CI: 1.12, 1.79), higher prevalence of cardiovascular diseases/stroke (P < 0.001; OR: 1.86; 95%CI: 1.45, 2.39), higher depression score (P = 0.002; OR: 1.05; 95%CI: 1.02, 1.08), and lower physical activity (P = 0.01; OR: 0.71; 95%CI: 0.54, 0.93). Asthma prevalence (2.6%; 95%CI: 2.0, 3.1; known duration: 17.2 ± 15.0 years) was associated with less alcohol consumption (OR: 0.53; 95%CI: 0.33, 0.87; P = 0.01), higher depression score (OR: 1.08; 95%CI: 1.03, 1.12; P < 0.001), and urban region (OR: 0.68; 95CI: 0.49, 0.95; P = 0.0.03). Conclusions: In this Russian population aged 40+ years, the prevalence of airflow obstruction was 6.8% with an awareness rate of 63.4% and smoking as main risk factor. Asthma prevalence was 2.6%.

In a similar manner, 0.40 million people died from asthma in 2015 worldwide, and the prevalence of asthma increased globally by 12.6%, whereas the age-standardized prevalence decreased by 17.7% (4). The main predisposing factors for COPD were smoking and ambient particulate matter while household air pollution, occupational particulates, ozone and secondhand smoke also played a role. Together, these risks explained 73.3% of DALYs due to COPD. In the same survey, smoking and occupational asthmagens as the only quantified risk factors for asthma accounted for 16.5% of DALYs due to asthma (4).
Despite the high importance of COPD and asthma for public health, data about the prevalence of COPD or airflow obstruction and asthma in Russia and parameters associated with the occurrence of both diseases in Russia has remained scarce so far (5)(6)(7). We therefore explored the prevalence of airflow obstruction and asthma in a population in Russia and assessed associations between these diseases and other major risk factors. With the population of Russia including many ethnicities, we chose the Russian republic of Bashkortostan as study site, since the population of Bashkortostan includes Russians and other ethnic groups with different cultural backgrounds (8,9).

METHODS
We conducted the Ural Eye and Medical Study in the city of Ufa as the capital of the republic of Bashkortostan and in a rural region in a distance of 65 km to Ufa. Ufa is located about 1,400 km East of Moscow at the Southwestern end of the Ural Mountains. An age of 40 years or older and living in the study regions were the only inclusion criteria of the study. The Ethics Committee of the Academic Council of the Ufa Eye Research Institute approved the study protocol and all participants gave an informed written consent.
Trained social workers conducted an interview which included more than 250 standardized questions on the socioeconomic background, smoking habits and alcohol consumption, physical activity, depression and anxiety, and known diagnosis and therapy of major diseases (Supplemental Table 1). The questionnaire in particular included questions on chronic cough, breathlessness on exertion, sputum production, frequent exacerbations of bronchitis, and a history of exposure to risk factors, especially tobacco smoke, occupational dusts, home cooking and biomass fuels ( Table 1) (10). The study methods further included anthropometry, blood pressure measurement, handgrip dynamometry, spirometry, and biochemical analysis of blood samples taken under fasting conditions. We assessed the presence of depression by applying the Center for Epidemiologic Studies Depression Scale (CES-D) Scoresheet, and we explored trait and state anxiety by using the State-Trait Anxiety Inventory (STAI). We defined the levels of blood pressure using the guidelines of the American Heart Association, and diabetes mellitus by a fasting serum glucose concentration of ≥7.0 mmol/L or a self-reported history of physician-based diagnosis or therapy of diabetes mellitus (11). We applied the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER statement guidelines) (12). We have described the study design in detail recently (8,9,13,14).
All participants underwent a pulmonary function test by spirometric measurement (spirometry device: Riester spirotest, Riester Company, Jungingen, Germany) of the forced expiratory volume in 1 s (FEV1) and of the forced vital capacity (FVC). The FEV1 was defined as the greatest volume of air that could be breathed out in the first second after in-breathing as deep as possible, and FVC was defined as the greatest volume of air that could be breathed out in a single breath as hard as possible, for as long as possible, preferably for at least 6 s. We additionally measured the tidal volume ( Table 1) (14). We applied the recommendations made by the American Thoracic Society and European Respiratory Society Task Force (15)(16)(17), except for not using a bronchodilator. Since we did not apply a bronchodilator, the measurements were pre-bronchodilator readings. In agreement with the Global Initiative of Chronic Obstructive Lung Disease, airflow obstruction was defined applying a spirometry-based definition with a cut-off value of the FEV1/FVC ratio of <0.7 (10). Using the recommendation made by Hankinson et al. we additionally calculated the lower limit of normal (LLN) for the spirometric measurements to define the presence of airflow obstructions (18). Asthma was defined by a self-reported diagnosis of physician-made diagnosis of asthma. Using a statistical software program [SPSS (Statistical Package for Social Science) version 25.0; IBM-SPSS Inc., Chicago, USA], we first calculated the prevalence of airflow obstruction and showed the results as mean and 95% confidence intervals (CI). We then assessed differences between the group of participants with airflow obstruction and the individuals without airflow obstruction. We finally conducted a multivariable binary regression analysis with the prevalence of airflow obstruction as dependent variable and as independent variables all those parameters, which differed significantly between the group with vs. without airflow obstruction. We determined the odds ratios (OR) and their 95% confidence intervals (CI). All P-values were two-sided and we considered them statistically significant if their values were <0.05.

RESULTS
Out of 7,328 eligible individuals, the Ural Eye and Medical Study included 5,899 (80.5%) study participants. Ranging between 40 and 94 years, the mean age was 59.0 ± 10.7 years. The demographic data of the study population did not vary significantly from the data of the Russian census carried out in 2010 (http://www.gks.ru/). The present investigation consisted of 5,392 (91.4%) individuals [2,449 (45.4%) men] for whom spirometric measurements were available. The group of study participants with spirometric measurements as compared with the group of individuals without spirometric measurements was significantly younger (58.6 ± 10.6 years vs. 63.1 ± 11.3 years; P < 0.001), and had a significantly higher proportion of men [2,449 (45.4%) men/2,943 (54.6%) women vs. 131 (25.8%) men/376 (74.2%) women; P < 0.001]. We assessed the symptoms of airflow obstruction, biomass use and workplace exposures for only 3,198 participants, while we performed spirometry for all 5,391 individuals ( Table 1). The reason was that the questions on obstructive airflow symptoms were included into the questionnaire after the study had started.
In univariate analysis, the airflow obstruction group differed from the group without airflow obstruction in the demographic parameters such as gender and region of habitation, anthropometric factors such as body mass index, socioeconomic parameters such as level of education, results of the biochemical blood analysis, and parameters related to physical activity, medical history, diet, smoking, and subjective hearing loss (Supplemental Table 1). In addition, the airflow obstruction group as compared to the group without airflow obstruction showed a higher prevalence of coughing (P < 0.001) and having sputum during coughing (P < 0.001), higher prevalence of catching a cold in the winter (P < 0.001), dyspnea (P < 0.001), and higher occurrence rate of dust at the working place (Supplemental Table 1).
The multivariable regression analysis included the occurrence of airflow obstruction as dependent variable and as independent variables all those parameters that were significantly associated with the prevalence of airflow obstruction in the univariate analysis. After dropping in a step-by-step manner those independent parameters which had lost the statistical significance of their association with the airflow obstruction prevalence, a higher airflow obstruction prevalence was associated with a higher prevalence of current smoking and higher number of cigarette package years, female gender, urban region of habitation, higher depression score, higher prevalence of history of arthritis, and cardiovascular diseases including stroke, and higher prevalence of ownership of a telephone, lower prevalence of vigorous activities during leisure time, and lower serum concentration of creatinine ( Table 4). If the parameter of cigarette package years was dropped, the odds ratio for current smoking increased to 9.12 (95%CI: 6.77, 12.3). If age or body height were added to model, each parameter was not significantly (P = 0.73 and P = 0.56, respectively) associated with the prevalence of airflow obstruction. If the LLN-based diagnosis of airflow obstruction was used a higher prevalence of airflow obstruction was associated with a higher prevalence of current smoking (OR: 4.66; 95%CI: 3.01, 7.20; P < 0.001) and higher number of cigarette package years (OR: 1.02; 95%CI: 1.01, 1.03; P < 0.001), higher depression score (OR: 1.06; 95%CI: 1.03, 1.09; P < 0.001), higher prevalence of history of arthritis (OR: 1.58; 95%CI: 1.25, 1.98; P < 0.001), and cardiovascular diseases including stroke (OR: 1.86; 95%CI: 1.48, 2.33; P < 0.001), and higher prevalence of ownership of a telephone (OR: 1.56; 95%CI: 1.10, 2.21; P = 0.01), lower prevalence of vigorous activities during leisure time (OR: 0.76; 95%CI: 0.60, 0.97; P = 0.03), and lower serum concentration of creatinine (OR: 0.99; 95%CI: 0.98, 0.99; P < 0.001). If age or body height were added to model, each parameter was not significantly (P = 0.16 and P = 0.88, respectively) associated with the prevalence of airflow obstruction.
If the parameter of a history of menopause was added to the model in women, telephone ownership (P = 0.22), cigarette package years (P = 0.51), and vigorous activity in leisure time (P = 0.054) were no longer significantly associated with airflow obstruction prevalence, so that in the final model, a higher prevalence of airflow obstruction in women was associated with higher prevalence of current smoking (P < 0.001), urban region of habitation, higher prevalence of history of menopause, arthritis and cardiovascular diseases including stroke, higher depression score, lower serum concentration of creatinine ( Table 5). If age or body height were added to model, each parameter was not significantly (P = 0.38 and P = 0.85, respectively) associated with the prevalence of airflow obstruction.
Asthma was present in 142 individuals indicating a prevalence of 2.6% (95%CI: 2.0, 3.1) in the study population ( Table 6). For 132 individuals who could remember when their asthma was diagnosed, known duration of asthma was 17.2 ± 15.0 years (median: 14 years; range: 0-76 years). For 10 individuals, the known duration of asthma could not be well-remembered.
Within the Russian ethnic group (n = 1,185), asthma was present in 27 individuals [2.5% (95%CI: 2.0, 3.0)] with a known duration of asthma of 14.9 ± 13.4 years (median: 11 years; range: 1-58 years). The Russian group and the non-Russian group within the study population did not differ significantly (P = 0.92) in the prevalence of asthma.

DISCUSSION
The figure of the prevalence of airflow obstruction of 6.8% (using the definition of FEV1/FVC < 0.70) in our Russian study  (29). The COPD prevalence was associated with rural region of habitation, older age, smoking, lower body mass index, lower level of education, and poor ventilation in the kitchen. van Gemert et al. performed a prospective observational cross-sectional study in rural Uganda and defining COPD as FEV1/FVC ratio of less than the lower limit of normal, found a prevalence of COPD of 16.2% (men: 15.4%, women: 16.8%) in a population of 588 individuals with a mean age of 45 ± 13.7 years, with 546 (93%) of the individuals being exposed to biomass smoke (27). The prevalence was highest in people aged 30-39 years, with the major risk factors of biomass smoke for both sexes and tobacco smoke for men. Jaganath et al. examined a population-based sample of 2,957 adults aged ≥35 years from four resource-poor settings in Peru (22). Defining COPD as a post-bronchodilator FEV1/FVC <0.70, they found an overall prevalence of COPD of 6.0% (95%CI: 5.1%, 6.8%) with the major risk factors of post-treatment tuberculosis and daily exposure to biomass fuel smoke. The prevalence of COPD (defined as post-bronchodilator FEV1/FVC ratio <0.7 and FEV1 <80% predicted) in Australians aged 40+ years was reported to be 7.5% among people with an age of 40+ years, and 29.2% among individuals with an age of 75+ years (25). Among individuals with an age of 40+ years, the prevalence of wheeze in the past 12 months was 30.0%, and the frequency of shortness of breath when hurrying on the level or climbing a slight hill was 25.2%. Other studies found a COPD prevalence of 16.2% in Uppsala/Sweden with the main risk factors of older age (OR: 2.08 per 10 years) and smoking (OR: 1.33 per 10 pack years), while higher education was protective (OR: 0.70 per 5 years of education) (21 (4). It roughly corresponds to a prevalence of a prevalence 6.8% in individuals aged 40+years as in our study population.
In the meta-analysis by the Global Burden of Diseases Study, smoking and ambient particulate matter were the main risks for COPD followed by household air pollution, occupational particulates, ozone, and second-hand smoke. These findings were congruent to those obtained in other investigations as well as in our study in which current smoking was by far the most important risk factor for airflow obstruction (OR: 9.07) ( Table 1). In our study population, household air pollution and occupational particulates (questions: "Do you have a fireplace or stove with an open fire at home?"; "Is there smoke at your working place"; "Is there dust at your working place") generally had a low frequency so that these parameters did not play a major role as risk factors for airflow obstruction in our study population, neither in the urban region or the rural area (Supplemental Table 1). In the investigation by Andreeva and associates, a higher prevalence of COPD was associated only with smoking (OR: 2.47; 95%CI: 1.60-3.82) (5).
The prevalence of asthma as found in our study population with a figure of 2.6% (95%CI: 2.0, 3.1) fits with previous in the literature (4). De Roos et al. found in the Rotterdam Study a prevalence of physician-diagnosed asthma of 3.6% (95% CI: 3.3%, 3.9%) in 14,621 participants with a mean age of 65.5 years, with a higher prevalence in women than in men (4.2 vs. 2.8%) (30). In a meta-analysis about the asthma prevalence in Iran, Varmaghani et al. reported an overall asthma prevalence of 4.56% (95% CI: 3.76%, 5.36%) among men, and of 4.17% (95% CI: 3.42-4.91%) among women (31). Factors associated with a higher asthma prevalence in our study population were a higher depression score and urban region of habitation. In the Global Burden of Disease Study, smoking and occupational asthmagens as the only quantified risk factors for asthma accounted for 16.5% of DALYs due to asthma (4).
Interestingly, the prevalence of airflow obstruction in men decreased from 17.2% (95% CI: 12.1, 22.4) in the age group of 40-44 years to 5.10% (95% CI: 3.10, 7.09) in the age group of 55-59 years, before it increased again ( Table 3). A similar U-shaped form the association between airflow obstruction prevalence and age was not seen in the female study population ( Table 3). The reasons for the lower airflow obstruction prevalence in middleaged men have remained unclear so far. Potential factors might be an increased mortality of these individuals or measures taken to reduce the burden of airflow obstruction in elderly men.
Limitations of our study should be mentioned. First, the main limitation is that the definition we used for airflow obstruction was based only on the ratio FEV1/FCV < 0.7 without that a bronchodilator test was performed. It is in contrast to the recommendation described in the GOLD Executive Summary 2017 and 2019 in which "a post-bronchodilator fixed ratio of FEV 1 /FVC <0.70 is the spirometric criterion for airflow limitation" (32,33). In a similar manner, the GOLD Executive Summary 2013 stated that "although postbronchodilator spirometry is required for the diagnosis and assessment of severity of COPD, the degree of reversibility of airflow limitation (e.g., measuring FEV 1 before and after bronchodilator or corticosteroids) is no longer recommended." (10). In the framework of a population-based study as ours it was however technically and logistically difficult to include the application of a bronchodilator into the routine procedures of a study on more than 5,000 participants. Second, since we did not apply bronchodilators we could not use the definition of GINA (Global Initiative for Asthma) for the definition of asthma (34). The difference in the methodology and definition of asthma might have influenced the results of the prevalence of asthma in our study population. Third, a point of concern is that the definition of COPD varies between studies (35). In our investigation we applied the definition of the Global Initiative of Chronic Obstructive Lung Disease with a ratio of FEV1 to FVC of <0.70, used however pre-bronchodilator values. Other studies used the lower limit of normal (LLN) method of deriving a threshold as the fifth percentile of the FEV1/FVC ratio in a healthy reference population. In our study, we additionally applied the LLN-based definition of airflow obstruction and, as compared to using the definition of FEV1/FVC <0.70, we arrived at similar results with respect to the associations between the prevalence of airflow obstruction and other parameters, while the prevalence of airflow obstruction defined by the LLN method was slightly lower than the prevalence of airflow obstruction defined by a value of FEV1/FVC <0.70 [5.8% (95%CI: 5.2, 6.5) vs. 6.8% (95%CI: 6.2, 7.5)]. One may also take into account that, both the GOLD definition and the LLN method may lead to misclassifications. In a study performed by Güder et al. the diagnostic accuracy and prognostic capability of the GOLD and LLN definition were compared to an expert-based diagnosis (36). It revealed that compared to the expert panel diagnosis, the GOLD-based definition of COPD led to a misclassification rate of 28%, and the three LLN-based definitions of COPD were associated with a misclassification rate of 46, 39, and 98%, respectively. In general, the GOLD-based definition was correlated with more false positive results, while the LLNbased definitions were associated with more false negative decisions. Fourth, most surveys of asthma used a case definition based on self-report of a diagnosis of asthma by a physician and wheeze (with other respiratory symptoms) in the past 12 months (37). In our study, asthma was defined by selfreported diagnosis of physician-made diagnosis of asthma. Toelle et al. and others have suggested that a case definition for clinically relevant asthma should preferably include wheezing symptoms occurring the preceding year and a bronchial hyperresponsiveness to inhalation of methacholine or histamine that is reversible with a bronchodilator (38). Several surveys have applied this definition to measure the asthma prevalence, but it has not been universally adopted. Reasons were logistical factors and the concern about a poor specificity and poor prediction of the future risk of asthma in individuals without symptoms (39). Pattemore et al. and Pekkanen and Pearce pointed out that the application of biological measurements to improve the validity of the asthma definition is associated with the goal of the investigation. To cite an example, the bronchial hyperresponsiveness has a similar or better specificity, but a worse sensitivity, than symptom questionnaires, so that it may be less suitable for the measurement of the prevalence (40,41). Also, an overlap between COPD and asthma may have to be considered as assessed in the CHAIN study (42). Fifth, some subgroups of the study population were relatively small, such as the age subgroups with an age of 75-79 years or with an age of 80+years and the subgroup of individuals with asthma, so that the statistical power might not have been sufficient to detect the significance of associations for these subgroups (43).
In conclusion, in this typically ethically mixed urban and rural Russian population aged 40+ years, airflow obstruction prevalence was 6.8% (95%CI: 6.2, 7.5) with an awareness rate of 63.4% (95%CI: 58.5, 68.4) and the main risk factor of smoking. Household air pollution and occupational particulates did not play a major role as risk factors. Asthma prevalence was 2.6% (95%CI: 2.0, 3.1).

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding author.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by According to the Declaration of Helsinki, the Ethics Committee of the Academic Council of the Ufa Eye Research Institute approved the study and all participants gave informed written consent. The ethics committee confirmed that all methods were performed in accordance with the relevant guidelines and regulations. The patients/participants provided their written informed consent to participate in this study.