Introduction

Non-communicable diseases (NCDs) cause premature deaths in 41 million people each year [1], accounting for a staggering 74% of mortality worldwide [2]. The higher prevalence of NCDs in low- and middle-income countries (LMIC) [3,4], particularly their outcomes, pose a threat with respect to economic sustainability [5]. Most NCDs in these countries are associated with common preventable risk factors related to four main behavioral variables: tobacco use, physical inactivity, unhealthy diet, and excessive alcohol use [6–11]. These behavioral risk factors commonly result in five key metabolic/physiological changes: raised blood pressure (hypertension); being overweight/obese; raised blood glucose (diabetes); elevated cholesterol; and hyperlipidemia [12].

In the context of resource-constrained settings, the lack of hypertension detection [13] and discrepancies between physician-perceived and calculated risk (e.g., using a standardized scoring system for cardiovascular risk) are concerns for health policymakers [12]. Regarding such discrepancies, Cesaro et al. [12] observed that physician-perceived cardiovascular risk was underestimated in 37% of cases. They argued that this discrepancy underscores a potential gap in clinical practice, indicating that reliance on clinical judgement alone may be insufficient for accurate risk estimation and should be complemented by an evidence-based risk score.

In a recent systematic meta-analysis, Ranzani et al. [14] observed that hypertension in LMICs increased during 1990–2020 in both urban and rural areas, with a stronger trend in rural areas. In this context, the multifaceted disparities in rural vs. urban healthcare facilities, infrastructure, and socioeconomic conditions have been highlighted by numerous studies, including Huang et al. [15], Zeba et al. [16], and Zhang et al. [17]. Furthermore, the growing ‘urbanicity’ of the countryside, driven by the rapid adoption of lifestyle changes among the rural population, is contributing to the epidemiological shift towards NCDs (including hypertension) in the LIMICs [18,19], particularly in Bangladesh [20,21].

Bangladesh is an LMIC in South Asia, home to more than 170 million people. Research focusing on NCDs in Bangladesh is important, as the country is undergoing an “epidemiological transition” wherein chronic non-communicable degenerative conditions such as hypertension, diabetes, and obesity are rapidly displacing infectious diseases as the major cause of death [6,7,10,22–24]. Hypertension is one of the leading causes of death in Bangladesh [5], with many such deaths being premature [6,7]. To compound the problem, the prevalence of hypertension in Bangladesh was projected to grow from 26% in 2000 to 29% in 2025 [25]. Additionally, hypertension and diabetes share a complex interconnection, with both being high-risk factors for heart disease [6,8,9]. The literature on the role of hypertension in the development of coronary artery disease and myocardial ischemic syndromes, and other cardiovascular diseases is voluminous. Elaborating on this relationship, Volpe and Gallo [26] noted that the associations between hypertension and coronary artery disease are complex, involving overactivation of neurohormones, accelerated atherosclerotic plaque development, endothelial dysfunction, and altered intramyocardial coronary circulation. Furthermore, a nationwide cohort study in Taiwan found that both type 2 diabetes mellitus with subsequent hypertension and hypertension with subsequent diabetes were associated with higher cardiovascular disease risk [27].

Bangladesh has a higher comorbidity of diabetes-hypertension compared to other LMICs [6,8,24] and, by extension, a greater risk for heart disease. Recent studies have confirmed that the prevalence of hypertension in Bangladesh is on the rise among adults aged 35 years and older [9,28–30], while others have identified an increase in risk factors, such as lifestyle and behavioral factors (e.g., obesity, diet, and physical activity levels), among both men and women [6,8,28,31–34]. Indeed, findings have revealed a significant association between obesity/overweight and hypertension [6,31,35,36] and have identified physical activity as an important risk factor [6–8,10] for hypertension. In LMICs, differential stages of demographic and nutritional transitions have been found to be associated with hypertension prevalence [9,13,37–42]. Several studies in Bagladesh have also demonstrated that socioeconomic status and inequalities in wealth distribution pose significant risk factors for prehypertension and hypertension [6–8,10,24,43], and that these risk factors vary based on place of residence, educational attainment, occupational level, and tobacco use. The STEPwise Surveillance for NCD risk factors (STEPS) survey of 2018 and two nationally representative datasets, namely, the Bangladesh Demographic and Health Survey (BDHS) 2011 and 2017–18, have highlighted the growing prevalence of NCDs and various risk factors [9,44]. However, the literature on these subjects exhibits an urban bias, with very little research examining the rural population. Furthermore, existing studies typically do not consider the broader population and are instead targeted to a specific area, age group, or gender [32,35,43,45,46]. Moreover, studies examining and comparing the association between certain risk factors for hypertension are also rare. While precisely measuring hypertension and its risk factors in the general population is challenging, access to such data is critically important, as it would help researchers and policymakers understand rates and patterns of hypertension and, thus, how to formulate effective policy interventions.

Bangladesh has high rates of hypertension that are associated with risk factors such as dietary habits and lifestyle. Considering that hypertension is one of the most important NCDs in Bangladesh, this study investigates its distribution patterns among rural populations, with a particular focus on physical, behavioral, and dietary risk factors. This research aims to achieve two specific objectives: i) to identify and analyze the physical, behavioral, and dietary risk factors associated with prehypertension and hypertension among adult males and females in rural Bangladesh, and ii) to determine the magnitude of the effects of the identified risk factors upon prehypertension and hypertension among this cohort.

Methods

Data sources, study population, and survey design

A multi-stage sampling method was used to select 149 villages across eight districts of Bangladesh: Moulvibazar, Sunamganj, Sherpur, Jamalpur, Munshiganj, Pabna, Satkhira, and Khulna. Two Upazilas (sub-districts) were randomly selected from each district, for a total of 16 Upazilas, from which villages were sampled. As the study was observational and involved no intervention, a control group was not used. This sample comprised 12% of the rural population across Bangladesh’s 64 districts, with the household as the Primary Sampling Unit (PSU). The following formula was used to calculate the optimal sample size for the study:

(1)

where is the quantile of the standard normal distribution, is the margin of error, and represents the proportion of the target indicator, .

To ensure a confidence level of 95%, it was assumed that , and the margin of error was 5 (i.e., ). Finally, the design effect was set as 1, as the simple random sampling method was applied to select households within the village.

Since the value of p was unknown, it was assumed that p = 0.5 would yield the maximum number of samples appropriate for measuring multiple indicators. Ultimately, a sample size of 384 samples were deemed necessary for each sampling unit (i.e., rural Upazila) to achieve a 95% confidence interval. The rural areas consisted of 384 PSUs x 16 Upazilas, yielding a total required sample size of 6144 PSUs. This sample was then extrapolated to 7,384 to obtain better coverage. Sample respondents were recruited for a four-month period, starting from 20 August 2017 till 21 December 2017. Data collection took place between August and December 2017 through semi-structured interviews with household heads (HHs) and their spouses in each PSU.

The survey process consisted of two components: i) data collection using a semi-structured questionnaire, with informed consent (written and verbal with a witness where necessary), and ii) a physical examination (conducted with participant consent). The questionnaires, formulated in Bangla (Bengali), translated into English, and validated for analysis, were administered via direct interviews in Bangla. Details of these instruments are provided in the studies by Shahidullah et al. [47] and Badat [48]. The standardized physical examinations measured height, weight, blood pressure, and blood sugar/glucose levels. The equipment used in the examinations included measuring tapes, digital scales, blood pressure monitors, stethoscopes, and glucometers. Field organizers and Information Technology personnel from Bangladesh’s Centre for Natural Resource Studies (CNRS) provided necessary technical support and troubleshooting throughout our fieldwork. The questionnaire consisted of multiple modules, with the preliminary questionnaire being modified to incorporate additional questions based on pre-testing in communities outside the study areas.

Statistical analysis

To investigate the risk factors associated with hypertension, we conducted a series of statistical analyses using SAS 9.4 (SAS Institute Inc., Cary, NC). First, we described the sociodemographic characteristics of the study population and the respondents’ household characteristics (Tables 1 and 2). Next, we calculated the prevalence of hypertension and examined its relationship with different age groups, genders, and wealth quintiles (Tables 3 and 4). A chi-square test was then employed to assess any statistically significant relationships. Finally, we conducted a multinomial logistic regression analysis using a stepwise selection method to determine how these risk factors, both independently and collectively, influenced the prevalence of hypertension (Table 5). Our response variables were pre-hypertensive vs. non-hypertensive and hypertensive vs. non-hypertensive. The odds ratio was used to quantify the strength of association between two categorical variables, with p-values <0.05 considered statistically significant throughout the analysis.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Background characteristics of the respondents (n = 14,343).

https://doi.org/10.1371/journal.pone.0351569.t001

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Household characteristics of respondents by age group and gender.

https://doi.org/10.1371/journal.pone.0351569.t002

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Prevalence and ORs of risk factors of hypertension by age in rural Bangladesh.

https://doi.org/10.1371/journal.pone.0351569.t003

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Prevalence and ORs of risk factors of hypertension by wealth quintile in rural Bangladesh.

https://doi.org/10.1371/journal.pone.0351569.t004

Download:

• PNG
  larger image
• TIFF
  original image

Table 5. Logistic regression model identifying risk factors influencing hypertension in rural Bangladesh.

https://doi.org/10.1371/journal.pone.0351569.t005

Results

Discussion

The prevalence of hypertension among adults in low- and middle-income countries (LMICs) is substantial and has been steadily increasing over the past few decades. A systematic review and meta-analysis by Ranzani et al. [14] examined 299 population-based surveys from 66 LMICs, covering nearly 20 million adults aged 15 and older. The authors noted a pooled prevalence of hypertension of 30.5% in urban areas and 27.9% in rural areas between 1990 and 2020. The mean age of participants was approximately 45 years. Several global analyses have reported that the number of adults with hypertension nearly doubled from 1990 to 2019, with the majority of this increase occurring in LMICs due to population growth and aging. Despite this rise, awareness, treatment, and control rates remain low in many of these countries [13,51].

Our field investigation examined the prevalence and associated risk factors of hypertension (including physical, behavioral, dietary habits, and socioeconomic status) among individuals living in rural Bangladesh. Our findings indicated that one-third of respondents aged 18 years or older (33.7% of males and 28.6% of females) had elevated blood pressure, classified as either prehypertensive or hypertensive. This finding indicates a significant increase in the prevalence of hypertension among residents of rural Bangladesh, as the 2018 national STEPS survey only found 21% of respondents to be hypertensive [52]. Likewise, we found that the prevalence of hypertension had increased considerably compared to the WHO’s 2016 study [7], which found that diabetes and hypertension were prevalent among the rural adult population at a rate of 8% and 26.4%, respectively [7]. In recent decades, the findings of population-based studies have demonstrated a significant global increase in the prevalence of NCDs, particularly hypertension and diabetes. Similarly, previous studies have reported that hypertension has become a major health concern in Bangladesh [53], which they have ascribed to changes in dietary patterns, food insecurity, poverty, lifestyle factors, and urbanization [29,54–56].

Age has been consistently shown to be a significant risk factor for hypertension, with findings showing a direct correlation between age and hypertension prevalence [57–59]. For instance, one recent study found the prevalence of hypertension to be especially high among participants in the 50–53 age group [9]. Furthermore, findings suggest that, after the age of 59, hypertension rates tend to be higher among women compared to men [57–59]. Our results confirm these findings, revealing a steady increase in hypertension prevalence with age, particularly among older individuals in rural Bangladesh. Indeed, we observed the highest rates of hypertension in women aged 45–54 years and above 65 years. While previous studies have found lower hypertension rates among adults younger than 39 [10], our findings indicate a shift in this trend. In addition, we also find that individuals under 45 years old are both at greater risk of hypertension, with men and women being 1.8 and 2.2 times more likely to develop hypertension, respectively, compared to older age groups. Several recent studies attributed this pattern to changes in lifestyle, occupation and effects of urbanicity in the rural areas by the younger age group. [29,60,61]. Our findings align with Fottrell’s [6] data, which shows a high prevalence of hypertension in the elderly population (aged 60 and over). These results underscore the growing problem of hypertension among older age groups and emphasize the need for targeted interventions and preventative measures.

Gender disparities in hypertension rates have been well-established [5,57–59,62–65]. Our data indicated that the overall prevalence of hypertension differed significantly between sexes (33.7% of males and 28.6% of females were prehypertensive or hypertensive; p-value < 0.0001). This difference is due to the predominance of pre-hypertensive. It was further revealed that some of the risk factors related to hypertension among the sexes are also significant. This result is consistent with previous studies of Bangladesh, which have found a close association between risk factors such as age, BMI, occupation, tobacco consumption, and fruit and vegetable consumption and high blood pressure across genders [6,8,32,44]

Evidence from the 2017–18 Bangladesh Demographic and Health Survey (BDHS) indicates that the prevalence of hypertension increases markedly with age: approximately 39.4% of adults aged 35 and older are affected, compared with 12.2% among those aged 18–34 [66]. This trend aligns with global patterns and underscores the need for early age-targeted screening initiatives, ideally initiated before age 35. Such proactive approaches can help identify pre-hypertensive individuals, enabling timely lifestyle interventions and reducing the risk of long-term cardiovascular complications. Alarmingly, recent studies also report growing hypertension prevalence among younger adults, particularly those in urban and peri-urban regions. This shift is attributed to increasing sedentary behavior, high dietary sodium intake, and reduced physical activity, which are linked to occupational transitions and urbanization [60,61]. These findings suggest the need for youth-focused interventions, such as awareness campaigns, workplace wellness programs, and the integration of blood pressure checks into routine care services. Unlike trends in many high-income countries, where men typically exhibit a higher prevalence of the condition, BDHS data reveal that Bangladeshi women bear a higher burden of hypertension (29.5%) than men (26.2%)—especially women with low physical activity and limited access to health information [66]. This gap warrants gender-sensitive strategies, including mobile health clinics, female community health workers, and the integration of hypertension screening into reproductive and maternal health services. Conversely, studies show that although men may have a slightly lower prevalence, they are often underdiagnosed and less likely to adhere to treatment, primarily due to poor health-seeking behavior and limited engagement with primary health services [5].

Studies of LMICs have identified education as one of the most important risk factors for hypertension, with many demonstrating a strong inverse relationship between educational attainment and hypertension in women [29,30,45,59,67–70]. For instance, Rahman [7] found significantly higher rates of hypertension among women with little or no education. Interestingly, although 80.7% of the females in our sample had a maximum education level of Grade 5, our findings did not indicate a significant association between their level of education and prehypertension or hypertension rates. This finding contradicts the trend observed in other LMIC studies, including findings from studies in Bangladesh [28,29,31,36,45,52,71–73]. The patterns in our study are likely attributable to other confounding factors, such as age, socio-economic status, and BMI, similar to prior findings reported by Chowdhury et al. [28] and Zahangir et al. [74].

The impact of limited education on dietary choices cannot be ignored and remains a crucial factor in dietary and health outcomes. Previous studies of rural Bangladesh have found that female household heads tend to make healthier food choices for their families, and that their understanding of “healthy” and “unhealthy” diets is significantly influenced by their own educational background [33,74,75]. This lack of awareness, coupled with potential disparities in access to adequate quantities and quality of food, can negatively affect the health outcomes of women and their dependents [75–77], regardless of their control over household decisions.

In our study of rural Bangladesh, we found that across all age groups and socioeconomic levels, women were significantly more likely than men to consume adequate amounts of fruits and vegetables. Specifically, wealthy women were 2.5 times as likely to eat sufficient vegetables and 3 times as likely to consume sufficient fruit. In contrast, wealthy men exhibited a minimal increase, with odds ratios for adequate vegetable and fruit consumption rising by 1.7 and 1.8, respectively. In many cultures, food consumption patterns are highly symbolic, reflecting underlying power differentials along gender lines. For instance, diets that are high in meat are often perceived as masculine, while the consumption of fruits and vegetables is typically seen as being more feminine [77], which can be protective against hypertension. This gendered lens influences food availability and distribution, particularly in LMICs, where gender preference and bias can impact dietary intake from early childhood [78]. These factors, along with lifestyle and SES differences between urban and rural populations, can intensify the risk factors for NCDs in women [45].

For the male participants, the odds of being prehypertensive or hypertensive increased with higher educational attainment and employment in non-physical occupations. This aligns with previous studies of Bangladesh conducted by Sathi [9], Fottrell [6], and Kibria [24], whose findings indicate that educated populations experience higher rates of diabetes, hypertension, and obesity due to sedentary lifestyles. We also found a notable positive association between wealth and risk of obesity/overweight, as men who qualified as being “rich” were 1.2 and 1.5 times more likely to be obese and overweight, respectively, compared to the poorest respondents. Our findings align with those of Ali et al. [32], who also observed increased obesity and hypertension rates among Bangladeshi men from middle to higher SES backgrounds.

Similarly, being overweight or obese also significantly increased the odds of hypertension among women, with obese women being 4.2 times more likely to be hypertensive compared to those with normal body weight. In one study of young adults across various regions of Bangladesh, Ali [32] identified abdominal obesity as a major risk factor for hypertension in women, particularly among those living in Dhaka, Barisal, and Rangpur Divisions. However, as Zahangir [74] notes, underweight women are also at risk of hypertension. Research focusing on rural Bangladesh has revealed a unique trend wherein wives of migrant laborers, both in urban Bangladesh and abroad, tend to be overweight or obese rather than undernourished. This phenomenon has been attributed to these women’s increased purchasing power, which enables them to access higher-quality food and hire domestic help [79]. In addition, the literature indicates that women from higher-income households generally have lower levels of physical activity [9,31]. This trend is reflected in our results, which show a significant inverse relationship between SES and physical activity (p = 0.02) among the female participants. It is notable that reduced physical activity could be both cause and consequence of hypertension [10,32]. However, the association between SES and BMI (p = 0.10) was not found statistically significant in our study. This finding runs counter to those of previous studies showing a strong positive relationship between SES and overweight/obesity among female participants [6,24,31,32,74].

Tobacco use, including smoking and chewing, is a significant health burden in South Asian countries such as India, Pakistan, and Bangladesh, as such behavior contributes to oral and lung cancers, as well as increased rates of NCDs such as hypertension and cardiovascular diseases, particularly among men [70,79,80–83]. However, our results could not establish a positive association between tobacco consumption and hypertension among the rural adults in Bangladesh, which is likely due to Type II error (i.e., false-negative rejection of the null hypothesis). However, our findings showed a significant positive correlation between tobacco consumption and age among women, with older women (65+) being 2.5 times more likely to use tobacco products compared to women in the 18–34 age group.

Notably, a study by Rahman [46] found a strong association between cigarette smoking and hypertension in Bangladeshi men. While tobacco use has declined among younger generations in recent years, older men and women continue to engage in these habits. The social stigma associated with women using tobacco is likely a major reason for the higher rates of tobacco use observed among men compared to women [28,65]. Although our study revealed a significant association between SES and tobacco consumption (p = 0.02) in men, usage rates were found to be lower among men qualifying as “rich.” Similar results have been reported in previous studies, where findings show tobacco consumption being prevalent in all SES categories, but lower among higher SES cohorts [6,34,42].

Our study found no direct association between SES and hypertension prevalence. This diverges from the findings of numerous prior studies, which have consistently demonstrated a positive association between household wealth and hypertension [6–8,24,32,43]. However, our findings indicate that SES affects lifestyle choices, particularly physical activity and diet. Consistent with prior findings, we found that individuals from higher SES groups exhibited healthier dietary habits, including greater consumption of fruits and vegetables [6,74]. This suggests that, while lifestyle factors play an important role in risk for hypertension, other complex factors, such as age and gender, are likely to contribute to its prevalence [24]. Kirschbaum’s [5] study of 76 LMICs found that significant regional differences in NCD prevalence were largely driven by socioeconomic disparities. While increased per capita GDP generally correlated with decreased hypertension risk factors, particularly among men, this pattern did not hold true for South-Asian countries, where risk factors continued to rise alongside SES [24]. Our findings indicate significant socioeconomic inequality in the prevalence of hypertension risk factors, specifically physical activity levels, BMI, fruit and vegetable intake, and tobacco consumption, across genders and age categories among residents of rural Bangladesh.

For long-term disease control, these patterns reinforce the importance of monitoring systems that track hypertension prevalence and outcomes by age and sex. Community-based monitoring, facilitated by trained health workers, can provide an accessible and cost-effective strategy for follow-up care, particularly in rural areas with limited access to clinics [32]. Furthermore, disaggregating national health data by demographic groups enables the development of responsive and equitable programming. Targeted interventions and nutrition-sensitive policies should be developed to address the socioeconomic determinants of diet and lifestyle.

Interventions to promote healthy, balanced diets, particularly fruit and vegetable intake, are urgently needed. Mass media campaigns promoting healthy dietary practices and highlighting their association with hypertension and diabetes are recommended. Place-of-residence-specific interventions should be incorporated into national policy formulation to address urban-rural disparities. Prioritizing these specific interventions in national policy formulation will help Bangladesh fulfill its commitment to the 2030 UN Sustainable Development Goals [84]. The national health policies of Bangladesh, particularly those addressing hypertension, should be guided by the World Health Organization’s evidence-based recommendations for the initiation and management of hypertension treatment in adults [84].

The major recommendations in the WHO guidelines include the following: i) developing a standardized treatment protocol through adopting a simplified algorithm-based treatment protocol to ensure consistency and scalability; ii) lowering treatment thresholds by initiating pharmacological treatment at >140/90 mmHg for most adults, and at lower thresholds for those with higher cardiovascular risk; iii) monitoring and establishing regular follow-up intervals and clear targets for blood pressure control (generally >140/90 mmHg); iv) leveraging mobile health and telemedicine to support adherence and remote monitoring; and v) developing equity-focused strategies by addressing social determinants and ensuring access for disadvantaged populations.

Conclusion

Consistent with prior studies, our findings revealed that one-third of Bangladesh’s rural adult population is living with elevated blood pressure, with hypertension rates being highest among older men and women. High blood pressure was also found to be associated with inadequate intake of fruits and vegetables. Obesity and overweight were also found to significantly influence the prevalence of hypertension in the study sample. While prior studies have shown tobacco consumption and educational attainment to be important risk factors among both sexes and women, respectively, our findings could not establish a significant positive association between smoking and hypertension. However, unadjusted analysis indicated some degree of correlation between tobacco consumption and hypertension. Results also showed no significant association between education and hypertension among women. Moreover, although our findings show no direct association between SES and hypertension prevalence, they did show a clear link between SES and hypertension via various related behavioral and physical risk factors.

Our findings contribute to the existing understanding of hypertension epidemiology in rural Bangladesh by demonstrating the multi-layered and interconnected nature of the risk factors. While access to information and health resources may reduce risk factors among advantaged individuals, the relationship between SES and risk factors underscores the need for equity-based techniques to minimize their impact.

Targeted efforts should focus on promoting healthy aging strategies, enhancing access to cardiovascular screening and management, and advocating for lifestyle modifications, including increased physical activity and healthier dietary practices. Customized interventions that consider sex-specific risk factors and socioeconomic contexts are essential to reducing the prevalence of hypertension in this population effectively. Additionally, addressing rural-urban disparities in healthcare services, infrastructure, and socioeconomic conditions requires serious policy attention and effective interventions.

Supporting information

Acknowledgments

We thank the study participants for having confidence and trust in the interviewers and for sharing sensitive information with the research team. Our gratitude also goes to the Centre for Natural Resource Studies (CNRS), Dhaka, Bangladesh, for implementing this research project.