National trends in ideal cardiovascular health among adults in Bhutan from three cross-sectional surveys in 2007, 2014, and 2019

The study aimed to estimate the prevalence, distribution, and correlates of ideal cardiovascular health (CVH) among individuals (20–69 years) across three cross-sectional surveys in 2007, 2014 and 2019 in Bhutan. Cross-sectional data were analysed from 9712 individuals (20–69 years, mean age = 37.6 years) who participated in the 2007, 2014 or 2019 Bhutan STEPS surveys, had complete measurement of CVH metrics, and had no history of a cardiovascular disease in 2014 and 2019. Ideal CVH measures included non-smoking, healthy diet, physical activity, body mass index (< 23 kg/m2), blood pressure < 120/ < 80 mmHg, total cholesterol < 200 mg/dL, and fasting blood glucose < 100 mg/dL). The prevalence of five to seven ideal CVH metrics increased from 11.6% in 2007 to 37.9% in 2019. Trend analyses showed that ideal physical activity, ideal total cholesterol, ideal blood pressure, and ideal fasting glucose increased from 2007 to 2019, while ideal fruit and vegetable intake, ideal smoking, and ideal body mass index decreased from 2007 to 2019. Five to seven ideal CVH metrics, 0–7 ideal CVH metrics, and 0–14 poor, intermediate, and ideal CVH metrics increased from 2007 to 2019. In the adjusted logistic regression analysis, older age decreased the odds of having 5–7 ideal CVH metrics in all three study years. Male sex increased the odds, and employment decreased the odds of 5–7 ideal CVH metrics in 2007, and urban residence increased the odds of 5–7 ideal CVH metrics in 2014 and decreased the odds in 2019. The proportion of meeting 5–7 ideal CVH metrics increased in Bhutan. Primary and secondary prevention programmes must be strengthened to improve CVH in Bhutan, considering identified associated factors.

We could not find any trend study on ideal CVH in East and South Asian countries, with the exception of a study in North China, which however, does not report the total ideal CVH metrics 18 . In a trend study over ten years in Iran, the prevalence of 5-7 ideal CVH metrics slightly reduced from 22.7% in 2007 to 19.5% in 2016, the mean CVH metrics increased from 4.7 in 2007 to 5.0 in 2016 19 . In a trend study among adults in USA over seven years the prevalence of the ideal CVH score significantly increased from 3.89 in 2011 to 3.96 in 2017 20 , and a trend study over 20 years in France found a decrease in ideal CVH 21 .

Methods
Study design and participants. Secondary data from three cross-sectional STEPwise approach to surveillance (STEPS) surveys in Bhutan in 2007, 2014, and 2019 26 with complete CVH metrics measurements and no history of CVD in 2014 and 2019 were analyzed; the overall response rate was > 93% in 2014 > 96% in 2019 [27][28][29] . "A multi-stage stratified sampling process was carried out to randomly select participants from the target population. An individual within the age range of the survey" was selected per household (25-74 years in 2007, 18-69 in 2014, and 15-69 years in 2019) [27][28][29] . In 2007 the study was restricted to the capital city of Thimphu, and in 2014 and 2019 the whole country of Bhutan. We restricted our data analyses to persons 20-69 years and those with no missing data on "smoking status, BMI, PA, diet, total TC, FBG and BP" measurements. The Research Ethics Board for Health (REBH), Bhutan, approved the study and written informed consent was obtained from all participants [27][28][29] . All methods were performed in accordance with the relevant guidelines and regulations.
Data collection followed the "WHO three STEPS methodology: step 1 included administration of a structured questionnaire (sociodemographics, medical history, medication use, and health risk behaviour) step 2 consisted of blood pressure and anthropometric measurements, and step 3 included biochemical tests (blood glucose and blood lipids)" 26 . Anthropometric measurements were taken with a portable digital weighing scale (SECA 843), constant tension Myotape tape (SecaTMbrand) [27][28][29] . Prior to taking blood pressure measurements, participants were asked to sit quietly and rest for 15 min with legs uncrossed. Three readings of systolic and diastolic blood pressure were obtained, with participants resting for three minutes between each reading. Of the three blood pressure measurements using "Omron BP apparatus automatic blood pressure monitor" [27][28][29] ; the last two readings following recommendations by WHO were averaged 26 . "Blood glucose, and total cholesterol were measured in peripheral (capillary) blood at the data collection site using dry chemical methods, biochemical analysis with a Hitachi 912 bio-analyzer" [27][28][29] .
In STEPS surveys 2014 and 2019 but not in 2007, "history of CVDs included self-reported coronary heart disease; angina, also called angina pectoris; a heart attack (also called myocardial infarction); stroke; any kind of heart condition or heart disease (other than the ones I just asked about) (Yes, No)" 26 .
Sociodemographic covariates included age (years), sex (male, female), education in years, number of adult household members, employment, and residence status 24 .
Data analysis. All statistical analyses were conducted with "STATA software version 14.0 (Stata Corporation, College Station, TX, USA)". "Analysis weights were calculated by taking the inverse of the probability of selection of each participant. These weights were adjusted for differences in the age-sex composition of the sample population as compared to the target population". Taylor linearization methods were utilized to account for complex study design. Descriptive statistics are used to describe CVH metrics (ideal, intermediate, and poor). Chi-square tests were used to test for differences in proportion. The significance of linear trends was analysed by using study year as categorical variable in logistic, linear, and ordinal regression analyses depending on the outcome variable. Logistic and linear regressions were used to assess the associations between sociodemographic factors and meeting 5-7 CVH metrics and continuous CVH metrics, overall and stratified by sex. Covariates in the multivariable logistic and linear regression models were age group, sex, educational level, number of adult household members, work and residence status. P values < 0.05 were considered significant and missing values were excluded from the analysis.  Table 1 (see Table 1).

Distribution of cardiovascular health metrics by study year.
Trend analyses showed that ideal physical activity, ideal total cholesterol, ideal blood pressure, and ideal fasting glucose increased from 2007 to 2019, while ideal fruit and vegetable intake, ideal smoking, and ideal body mass index decreased from 2007 to 2019. Five to seven ideal CVH metrics, 0-7 ideal CVH metrics and 0-14 poor, intermediate, and ideal CVH metrics increased from 2007 to 2019. The prevalence of five to seven ideal CVH metrics increased from 11.6% in 2007 to 37.9% in 2019, the seven-mean ideal CVH metrics increased from 2.9 in 2007 to 4.2 in 2019, and the 12-mean ideal CVH metrics increased from 8.7 in 2007 to 10.2 in 2019 (see Table 2). Table 3 shows CVH metrics by age group (20-34 years, 35-49, and 50-69 years). The proportion of ideal smoking increased with higher age group during each study year. The prevalence of ideal BMI decreased by age group during each study year. Ideal fruit and vegetable intake and ideal physical activity did not differ significantly by age group in each study year. The proportion of ideal total cholesterol, ideal blood pressure and ideal fasting glucose decreased with higher age groups in each study year (see Table 3). Table 4 shows CVH metrics by sex. Ideal smoking, and ideal blood pressure were higher among women than among men in all study years, while ideal BMI and ideal physical activity were higher among men than women, and ideal fruit and vegetable consumption, ideal FBG and ideal total cholesterol did not have clear sex differences (see Table 4).

Distribution of cardiovascular health metrics by sex and study year.
Associations with meeting 5-7 ideal CVH metrics. In adjusted logistic regression analysis, older age was negatively associated with meeting 5-7 ideal CVH metrics in all three study years. Male sex increased the odds, and employment decreased the odds of 5-7 ideal CVH metrics in 2007, and urban residence increased the odds of 5-7 ideal CVH metrics in 2014 and decreased the odds in 2019. Furthermore, in the sex stratified analysis, higher education increased the odds of 5-7 ideal CVH metrics in 2019 among women (see Table 5).

Associations with continuous CVH metrics.
In the adjusted linear regression analysis, older age was negatively associated with continuous CVH metrics in the three study years. Male sex was positively associated  Table 6).

Discussion
The researchers found that the prevalence of ideal CVH (5-7 ideal metrics) in Bhutan increased from 11.6% in 2007 to 37.9% in 2019, and the mean CVH metrics increased from 2.9 in 2007 to 4.2 in 2019. A similar increase in the ideal CVH score was found in a trend study in the USA (increase from 3.89 in 2011 to 3.96 in 2017 20 , while the mean CVH metrics increased in Iran from 4.7 in 2007 to 5.0 in 2016 19 . The increase in ideal CVH metrics in this study was mainly attributed to ideal physical activity, ideal total cholesterol, ideal blood pressure, and ideal blood glucose. However, we saw decreases in the proportion of ideal fruit and vegetable, ideal smoking, and ideal body mass index from 2007 to 2019. The prevalence of ideal CVH (5-7 ideal metrics) (ranging from 11.6% in 2007 to 37.9% in 2019), were similar to global estimates, mainly in high-income countries, ideal CVH (having 5-7 ideal metrics) (19.6%) CVH 8 , and in China (33.0%, 5-7 ideal metrics) 15 , but lower than in Nepal (51.6%, 5-7 ideal metrics) 16 . The proportion of ideal CVH metrics (all 7 metrics) (0.6% in 2007, 0.1% in 2014, and 0.0% in 2019), was similar to in previous studies, such as in urban India (< 0.1% had 7 ideal metrics) 17 , Shandon in China (0.05% all 7 ideal metrics) 12 , in rural area Northwest China all 7 ideal metrics (0.0%) 13 .
Like the three best global estimates 8 , this survey found that TC (92.7% in 2019), smoking (89.6% in 2019), and FGP (91.9% in 2019) had the highest proportion of ideal CVH status. A healthy diet (16.2% in 2019) had the poorest prevalence of ideal CVH status in this study which compares with the poorest global estimates 8 . The proportion of ideal fruit and vegetable consumption compares with global rates in low-and middle-income countries (18.0%) 33 . The estimates of ideal PA (84.7% in 2019) in this study were higher than global estimates of PA (40.6%) and ideal BMI (34.3% in 2019) are lower than global figures of ideal BMI (40.3%) 8 . Ideal blood pressure significantly decreased from 35.8% in 2007 to 30.1% in 2019 and poor blood pressure or hypertension significantly increased from 19.9% in 2007 to 26.6% in 2019, which is higher than global estimates in low-and middle-income countries (17.5%) 34 . Previous research in Bhutan also found a high prevalence of poor diet, overweight, and high blood pressure among adults calling for expanded supplementation and dietary interventions 3 .
Consistent with previous research 8, 16,17,22,23 , ideal CVH was higher among younger age groups (20-34 years) and among men in 2007. Some previous research 12,22,23 showed an association between higher education and ideal CVH metrics, while we only found this result, in particular among women, in 2019. Other research 24,25 , www.nature.com/scientificreports/ found an association between rural residence and ideal CVH metrics, which was also found in this study in 2019, while we found higher ideal CVH metrics in residents of urban than rural areas in 2014.
Results may inform Bhutan's CVH promotion policy. The increase in ideal physical activity may be attributed to the "implementation of community wide public education and awareness campaigns for physical activity", including a nationwide "Move for Health Campaign" is conducted routinely and open-air gyms in Thimphu and a number of other districts 3,35 . Diabetes and raised total cholesterol may have reduced due to the universal health coverage, including the early diagnosis and management of NCDs in Bhutan 3 . To improve CVH in Bhutan, multifactorial interventions are needed 36 , targeting the promotion of body weight control, smoking cessation, healthy diets, and screening and control of high levels of blood pressure 5 .
In addition to the use of large representative survey samples and standardized assessment measures, the study limitations included the cross-sectional design and self-report of some of the data. In addition, our healthy diet measure only consisted of one item (fruit and vegetable consumption) and not as in the original AHA healthy www.nature.com/scientificreports/ diet definition "(≥ 4.5 cups/day fruits and vegetables, 2: ≥ 3.5-oz servings/week of fish, 3: < 1500 mg/day sodium, 4: < 450 cal/week sweets/sugar, and 5: ≥ 3 1-oz servings/day whole grains)" 8,9 . Another limitation was that the history of any previous CVD was not evaluated in the 2007 and only in the 2014 and 2019 survey.  www.nature.com/scientificreports/

Conclusion
The proportion of meeting 5-7 ideal CVH metrics increased in Bhutan. Primary and secondary prevention programmes should be strengthened to improve CVH in Bhutan, considering identified associated factors. Future research may include more comprehensive measures on a healthy diet. www.nature.com/scientificreports/