The Prevalence and Risk Factors of Chronic Heart Failure in the Mongolian Population

Background: The prevalence of heart failure in the Mongolian population is unknown. Thus, in this study, we aimed to define the prevalence of heart failure in the Mongolian population and to identify significant risk factors for heart failure among Mongolian adults. Methods: This population-based study included individuals 20 years and older from seven provinces as well as six districts of the capital city of Mongolia. The prevalence of heart failure was based on the European Society of Cardiology diagnostic criteria. Results: In total, 3480 participants were enrolled, of which 1345 (38.6%) participants were males, and the median age was 41.0 years (IQR 30–54 years). The overall prevalence of heart failure was 4.94%. Patients with heart failure had significantly higher body mass index, heart rate, oxygen saturation, respiratory rate, and systolic/diastolic blood pressure than patients without heart failure. In the logistic regression analysis, hypertension (OR 4.855, 95% CI 3.127–7.538), previous myocardial infarction (OR 5.117, 95% CI 3.040–9.350), and valvular heart disease (OR 3.872, 95% CI 2.112–7.099) were significantly correlated with heart failure. Conclusions: This is the first report on the prevalence of heart failure in the Mongolian population. Among the cardiovascular diseases, hypertension, old myocardial infarction, and valvular heart disease were identified as the three foremost risk factors in the development of heart failure.


Introduction
Heart failure (HF) is one of the major public health concerns with an increasing incidence over the years and it remains to be one of the leading causes of mortality among cardiovascular (CV) diseases [1]. As of 2020, 64.3 million people were suffering from chronic HF worldwide [2]. In recent years, chronic HF has become more prevalent because of an aging population, increased cardiovascular risk factors caused by economic trends promoting unhealthy lifestyle behaviors, and modern therapeutic advances that have been extending the lifespan of patients with CV diseases. In the future, the number of patients with HF will continue to rise globally due to the above-mentioned reasons as well as the rise in related comorbidities [3].

Study Sample Size
The sample size was calculated based on the total population aged 20 years and older (n = 2,157,011) and an average prevalence from previous international studies (1.5%) assuming a 95% confidence interval (Z = 1.96) with a 2% acceptable margin of error (e = 0.02), which gave a sample size of 3600 subjects.

Sample Selection
Study clusters (n = 75) and subjects were randomly selected from 7 provinces in the Western, Mountain, Eastern, and Central regions according to geographical zoning and 6 districts of the Ulaanbaatar city. At each primary health care center, the subjects were enrolled in this study by using systematic sampling and were stratified into 10-year-interval age groups. The target sample size including 3600 subjects composed of 900 subjects in the 20-29 age group, 900 subjects in the 30-39 age group, 675 subjects in the 40-49 age group, 600 subjects in the 50-59 age group, 375 subjects in the 60-69 age group, and 150 subjects in the 70 years and over age group. Because the exclusion criterion was subjects with incomplete data, 120 subjects (3.3%) were excluded from the final analysis.

Data Collection
We performed quantitative survey methodology using standard questionnaires. Prior to the data collection, all research staff were provided with detailed instructions and trained for conducting interviews using study questionnaires. The questionnaire included subject's demographics, social characteristics, presence of CV risk factors, comorbidities, and HF-related symptoms. Educational level was divided into 3 groups, low, medium, and high. Marital status was categorized into 2 groups including married or cohabiting and divorced or single. Lifestyle characteristics such as smoking and alcohol consumption were classified as dichotomous variables: smoker or non-smokers, never or normal/abnormal use of alcohol. Diabetes mellitus was defined as self-reported physician-diagnosed diabetes and/or use of insulin and/or oral hypoglycemic medications. Coronary disease was defined as a prior myocardial infarction or revascularization (coronary bypass surgery or angioplasty).
Physical examinations were performed by well-trained physicians in order to identify HF-related signs. Blood pressure, heart rate, oxygen saturation, respiratory rate, and weight were measured by physicians. Body mass index (BMI, kg/m 2 ) was calculated by dividing the weight (kg) and height (m 2 ). Obesity was defined as a BMI of 30.0 kg/m 2 or greater. Hypertension was defined by a physician's diagnosis, systolic blood pressure

Study Sample Size
The sample size was calculated based on the total population aged 20 years and older (n = 2,157,011) and an average prevalence from previous international studies (1.5%) assuming a 95% confidence interval (Z = 1.96) with a 2% acceptable margin of error (e = 0.02), which gave a sample size of 3600 subjects.

Sample Selection
Study clusters (n = 75) and subjects were randomly selected from 7 provinces in the Western, Mountain, Eastern, and Central regions according to geographical zoning and 6 districts of the Ulaanbaatar city. At each primary health care center, the subjects were enrolled in this study by using systematic sampling and were stratified into 10-year-interval age groups. The target sample size including 3600 subjects composed of 900 subjects in the 20-29 age group, 900 subjects in the 30-39 age group, 675 subjects in the 40-49 age group, 600 subjects in the 50-59 age group, 375 subjects in the 60-69 age group, and 150 subjects in the 70 years and over age group. Because the exclusion criterion was subjects with incomplete data, 120 subjects (3.3%) were excluded from the final analysis.

Data Collection
We performed quantitative survey methodology using standard questionnaires. Prior to the data collection, all research staff were provided with detailed instructions and trained for conducting interviews using study questionnaires. The questionnaire included subject's demographics, social characteristics, presence of CV risk factors, comorbidities, and HFrelated symptoms. Educational level was divided into 3 groups, low, medium, and high. Marital status was categorized into 2 groups including married or cohabiting and divorced or single. Lifestyle characteristics such as smoking and alcohol consumption were classified as dichotomous variables: smoker or non-smokers, never or normal/abnormal use of alcohol. Diabetes mellitus was defined as self-reported physician-diagnosed diabetes and/or use of insulin and/or oral hypoglycemic medications. Coronary disease was defined as a prior myocardial infarction or revascularization (coronary bypass surgery or angioplasty).
Physical examinations were performed by well-trained physicians in order to identify HF-related signs. Blood pressure, heart rate, oxygen saturation, respiratory rate, and weight were measured by physicians. Body mass index (BMI, kg/m 2 ) was calculated by dividing the weight (kg) and height (m 2 ). Obesity was defined as a BMI of 30.0 kg/m 2 or greater. Hypertension was defined by a physician's diagnosis, systolic blood pressure ≥130 mmHg or diastolic blood pressure ≥80 mmHg, or use of antihypertensive medication.
We defined atrial fibrillation as a history or the presence of atrial fibrillation on an electrocardiography. CKD was defined as kidney damage or glomerular filtration rate (GFR) <60 mL/min/1.73 m 2 . Anemia was defined as serum hemoglobin levels <13.0 g/dL (<130 g/L) for men and <12.0 g/dL (<120 g/L) for women. The International Classification of Disease (ICD) 10 codes was used for the following comorbidities: COPD(J44), sleep apnea (G47.3), and thyroid disorders (E03 and E05).
Heart failure was defined as a syndrome recognized by the physician based on symptoms of exercise intolerance, signs of fluid retention, and response to therapy, according to the Guidelines of the ESC Working Group on Heart Failure. In our study, a diagnosis of chronic HF was based on the ESC clinical diagnostic criteria, including if the participant had chronic HF-related symptoms both at rest and during exercise (breathlessness, ankle swelling, and fatigue) or chronic HF-related signs (peripheral oedema, hepatomegaly, neck vein distention, third heart sound (S3) gallop rhythm, and pulmonary crepitations) and, in cases where there was doubt, the patient's response to diuretic treatment. An HF diagnosis was considered when the first and second criteria were both met [18].

Statistical Analysis
Patients' demographic characteristics and clinical variables were analyzed in the whole sample using descriptive statistics. Continuous variables were expressed as means ± standard deviations (for normal distribution) or medians with interquartile range (non-normal distribution). Categorical variables were shown as absolute numbers and percentages. The distribution of normality was based on visual assessment of a histogram and the Kolmogorov-Smirnov test. Categorical data were compared using a chi-square test. while continuous variables were compared using an independent sample t-test. Correlations between cardiovascular risk factors and both HF and non-HF groups were assessed using the Pearson's correlation coefficient. In addition, a logistic regression analysis was performed to calculate the odds ratio to assess associations between risk factors and covariates. Statistical significance was considered for two-sided p-values less than <0.05. The Statistical Package for the Social Sciences (SPSS 24.0) was used for data analysis.
Comparing the above-mentioned risk factors and comorbidities according to the main administrative groups, in the urban population there was significantly more smoking (22% vs. 19%) and diabetes mellitus (7% vs. 5%) than in the rural population (Table 3), while in the rural population there was significantly more abnormal alcohol consumption (10% vs. 7%) and obesity (25% vs. 21%). The overall prevalence of chronic HF was 4.94% for the total study population based on the ESC diagnostic criteria for HF. The prevalence of chronic HF strongly increased with age; it was 0.7% for the 20-29 year age group, while its frequency was 21.0% for the 70-87-year age group (Figure 2). Figure 3 shows the age-specific prevalence of overall HF for the 10-year-interval age groups in men and women. The prevalence of heart failure increased from 1.2% for men aged 20-29 years to 17.7% for men aged ≥70 years. For women, the prevalence increased from 0.5% in the lowest age group to 23.3% in the highest age group.
Abnormal alcohol consumption, n (%) 300 (9) 118 (7) 182 (10) 0.001 The overall prevalence of chronic HF was 4.94% for the total study population based on the ESC diagnostic criteria for HF. The prevalence of chronic HF strongly increased with age; it was 0.7% for the 20-29 year age group, while its frequency was 21.0% for the 70-87-year age group (Figure 2).

Figure 2.
Age-specific prevalence of HF. Note: The vertical axis shows prevalence rate expressed as a percentage. The horizontal axis shows age groups. Figure 3 shows the age-specific prevalence of overall HF for the 10-year-interval age groups in men and women. The prevalence of heart failure increased from 1.2% for men aged 20-29 years to 17.7% for men aged ≥70 years. For women, the prevalence increased from 0.5% in the lowest age group to 23.3% in the highest age group.
In the 40-49 and 50-59 year age groups, men and women showed comparable point prevalences.  The prevalences of HF in the urban and rural populations are shown in Figure 4. The prevalences of HF for males and females in the rural population were higher than those in the urban population. In the 40-49 and 50-59 year age groups, men and women showed comparable point prevalences.
The prevalences of HF in the urban and rural populations are shown in Figure 4. The prevalences of HF for males and females in the rural population were higher than those in the urban population. The prevalences of HF in the urban and rural populations are shown in Figure 4. The prevalences of HF for males and females in the rural population were higher than those in the urban population. The demographic and social characteristics of the study participants are shown in Table 4. The patients with HF were significantly older (median age 57 years), fewer had higher level education (18% vs. 36%), and more had low level education (30% vs. 13%) and were unemployed (58% vs. 40%) compared to subjects without HF. The remaining The demographic and social characteristics of the study participants are shown in Table 4. The patients with HF were significantly older (median age 57 years), fewer had higher level education (18% vs. 36%), and more had low level education (30% vs. 13%) and were unemployed (58% vs. 40%) compared to subjects without HF. The remaining variables including sex and marital status were comparable. Demographic and social characteristics of the study participants are shown in Table 4.  (20) 29 (17) For the logistics regression analysis, cardiovascular risk factors were included for analyzing the correlations between the variables and HF (Table 5). Among the cardiovascular risk factors, CAD, hypertension, valvular heart disease, abnormal alcohol consumption, and obesity significantly increased the risk of HF.  There were significant differences between the non-HF and HF groups regarding use of medications (Table 7). Diuretics were the most used medication, followed by reninangiotensin system (RAAS) inhibitors and β-blockers.

Discussion
First, this is the first study that revealed the prevalence of HF including both urban and rural populations using different clusters. Secondly, we found that Mongolian patients with HF had significantly higher frequencies of comorbidities and risk factors and poorer physical characteristics. Thirdly, hypertension, coronary heart disease, and valvular heart disease were leading CV causes of HF in Mongolian patients.
Based on our study results, the prevalence of HF (4.94%) in Mongolian adults is higher than that reported by studies from USA, some European countries (such as Italy, England, France, and Germany), some Asian countries (such as China and Japan), relatively comparable to Singapore, and lower than the prevalence in Malaysia [6,[19][20][21][22]. The present study findings suggest that the reason for the prevalence of HF was higher in the rural population than in the urban population, could be explained by disparities in economic levels (types of occupation), lifestyles (abnormal alcohol usage and obesity), education levels (p < 0.0001), and clinical conditions between urban and rural areas. These findings are consistent with those of a study in India [23].
Moreover, our findings show that unemployed and low education increase the risk of HF compared to participants without HF. A recent meta-analysis of 11 studies found that low socioeconomic status assessed by all common measures (education, income, occupation, and area) independently increased the incidence risk of heart failure by 62%, overall [24].

Discussion
First, this is the first study that revealed the prevalence of HF including both urban and rural populations using different clusters. Secondly, we found that Mongolian patients with HF had significantly higher frequencies of comorbidities and risk factors and poorer physical characteristics. Thirdly, hypertension, coronary heart disease, and valvular heart disease were leading CV causes of HF in Mongolian patients.
Based on our study results, the prevalence of HF (4.94%) in Mongolian adults is higher than that reported by studies from USA, some European countries (such as Italy, England, France, and Germany), some Asian countries (such as China and Japan), relatively comparable to Singapore, and lower than the prevalence in Malaysia [6,[19][20][21][22]. The present study findings suggest that the reason for the prevalence of HF was higher in the rural population than in the urban population, could be explained by disparities in economic levels (types of occupation), lifestyles (abnormal alcohol usage and obesity), education levels (p < 0.0001), and clinical conditions between urban and rural areas. These findings are consistent with those of a study in India [23].
Moreover, our findings show that unemployed and low education increase the risk of HF compared to participants without HF. A recent meta-analysis of 11 studies found that low socioeconomic status assessed by all common measures (education, income, occupation, and area) independently increased the incidence risk of heart failure by 62%, overall [24].
We observed that HF was more prevalent in men compared to women, despite a significantly higher prevalence of HF in women aged 70 years and older compared to men. These findings agreed with the results of a Chinese study [25].
HF is known primarily as a disease of the elderly. However, recent studies have indicated that the HF burden may be increasing in young individuals. Thus, the mean age for HF onset has been declining and the incidence of patients with HF aged below 50 years has increased by two-fold, particularly increasing from 3% to 6% [9]. In a Swedish study that linked national hospital discharge and death registries between 1987 and 2006, HF incidence increased in the last 5-year period by 50% among people aged 18-34 years and 43% among those aged 35-44 years [10]. In our study, the median age for a diagnosis of HF was 50 years, while the mean age at HF diagnosis was 73.7 ± 14.3 years in a UK population aged ≥30 years [27]. Overall, despite the South Asian and African ethnicity groups being significantly younger at HF onset than the Caucasian ethnicity group, they had similar or better cardiovascular risk profiles, which agreed with those previously reported in a younger UK general population [10].
The results of our study showed a highly age-specific prevalence of HF compared to other studies that have mostly included and examined populations aged 45 years and over and used various diagnostic approaches and criteria (see Table 8). Although medical records were reviewed for the definition of HF to identify HF diagnosis according to the Framingham Criteria in the Olmsted County Study [8], subjects in the Rotterdam Study were clinically examined to identify symptoms and signs suggestive of HF (e.g., shortness of breath, ankle oedema, and pulmonary crepitations) [18]. There is an ongoing debate regarding the definition of heart failure and there is a lack of a gold standard for assessing the presence of the syndrome in population-based studies. According to the ESC Guidelines on the diagnosis of HF, to establish the presence of heart failure, objective evidence of cardiac dysfunction must be present in addition to symptoms or medication for HF [28]. In the Framingham study, the overall prevalence of HF was 0.7% for those aged between 50 and 89 years, varying between 0.1% and 7.9% with age [28]. In the Rochester study, in 1986, the prevalence of HF in those over 35 years was 1.9%, increasing from 1% to 7.6% with age [29].
A recent randomized controlled trial suggested that the most common risk factors for HF were CAD, hypertension, and diabetes mellitus [30]. More specifically, the risk factors highly correlated with HF incidence included poorly controlled diabetes (HbA1c ≥ 8%), uncontrolled hypertension (SBP ≥ 160), and advanced obesity (BMI ≥ 35) [15]. Likewise, our study demonstrated that a previous history of CAD, hypertension, valvular heart disease, obesity, and abnormal alcohol consumption were main risk factors of HF. Because these findings were different compared to the NHANES study [12], we assume that it could be caused by the disparities of living standards and cultural differences.
The present study demonstrated that coronary artery disease (CAD) is the strongest risk factor for the development of HF among other risk factors with a prevalence of 29.1% (n = 50) in the total HF population. Secondly, having hypertension was also viewed as a major factor in the progression of HF with a prevalence of 84.9% (n = 149) in the HF population in this study. These results were in line with those of former studies such as a cardiovascular health study [31] and a Spanish study (81.8%) [32]. Our analysis also supports that valvular heart disease and obesity are important risk factors in the development of HF. This could be because of a higher incidence of rheumatic heart disease, lack of health education, and the cultural point of view among the Mongolian population.
The strength of our study is the population size which is representative enough for the overall population of Mongolia. Therefore, our study results could be generalizable to patients with and without HF in the general adult population. A limitation is, however, that the symptoms suggestive of HF as well as different disease prevalences (e.g., CHD, COPD, and diabetes) were self-reported, and therefore we were not able to validate this information. Another limitation of this study is that the participants who met the criteria for the clinical diagnosis of chronic HF were not further evaluated through echocardiography and natriuretic peptide testing for a confirmation of the diagnosis. Moreover, the diagnosis of HF was not validated in this study, which increases the risk of observer bias.

1.
This is the first investigation in Mongolia that describes the prevalence of HF among the general population. The prevalence of HF appears high (4.94%) in the Mongolian population compared with other studies.

2.
Our study revealed that coronary heart disease, hypertension, and valvular heart disease are the three foremost risk factors in the development chronic heart failure. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
Raw data that support the findings of this study are available from the corresponding author, upon reasonable request.