Factors Associated with Hypertension Among Adults: A Cross-Sectional Analysis of the Indonesian Family Life Survey

Introduction

Hypertension (HTN) is a serious cardiovascular disease that can cause several potentially fatal conditions.^1,2 Stroke, dementia, chronic kidney disease, coronary artery disease, coronary heart disease and heart failure are significantly exacerbated by HTN.³ Pre-hypertensives (120–139 mmHg systolic; 80–89 mmHg diastolic) and stage 1 hypertensives (140–159 mmHg systolic; 90–99 mmHg diastolic) account for most of the rising prevalence of HTN-linked events.³ It is estimated to be responsible for 13 to 19% of all deaths in the world annually,^2,4–6 and it is predicted that by 2025, 1.56 billion people will be hypertensive.⁷ Several factors contribute to the rising incidence of the disease including the ongoing nutritional transition and increasingly sedentary lifestyle. Populations in low- and middle-income countries may experience a greater burden of the disease,⁸ reflecting a shift that has occurred over the past three decades. Raised Blood Pressure (BP) was initially a significant burden in high-income countries, now it is prevalent in South and East Asia and Sub-Saharan Africa.⁹ In Southeast Asia, HTN is currently diagnosed in one-third of adults and causes over one million deaths a year.¹⁰ Indonesia, the largest and most populous country in Southeast Asia, has the highest prevalence of HTN in the region (42.7% of men and 39.2% of women). Additionally, there is still a lack of understanding and access to care, and the management of HTN is poor.^11–13 A study using data from the 2007 Indonesia Family Life Survey-5 (IFLS-5) examined socio-demographic disparities in the diagnosis and treatment of HTN in Indonesian adults. It found that 67% of those with HTN were unaware of their condition, less than 30% were taking medication for it and only 25% had their BP under control.¹⁴ The early detection of BP levels enables classification into normal, prehypertension or hypertension. This classification highlights the possibility of early detection of prehypertension and advanced hypertension and illustrates the progressive nature of HTN.¹⁵

Identifying factors that influence HTN risk may improve decision-making.¹⁶ High BP has several influences, including lifestyle and demographic factors.¹⁷ Some studies indicate that significant alterations in lifestyle habits have an impact on the prevalence of HTN;^18,19 these include smoking,²⁰ poor sleep quality,²¹ and lack of physical activity.²² Moreover, recent studies demonstrate that socio-demographic factors are responsible for gaps in the management of HTN.^23–25 An earlier study suggested that racial differences in HTN prevalence were influenced by demographic factors,²⁶ such as age,²⁷ gender,²⁸ marital status,²⁹ ethnicity,³⁰ and occupation status.³¹

Despite global recognition of the significant roles that these factors play in HTN development, few studies have investigated their specific impact in the Indonesian context.³² Furthermore, there are critical gaps in the existing literature. Notably, few previous studies have examined the association between sleep quality and HTN. A multifaceted approach that addresses the root causes of HTN will be essential to reducing its prevalence in Indonesia and improving the overall health of the population.

To address the existing gaps, this study aimed to identify the association between socio-demographic factors (age, gender, marital status, ethnicity and occupation status) and lifestyle factors (smoking, sleep quality, physical activity) with HTN, using data from the IFLS-5. By understanding these factors, policymakers can formulate and implement effective policies to address HTN’s incidence.

Methods

Study Design

A cross-sectional study design using secondary data from IFLS-5 was conducted in 2015.³² The community surveys gathered information on households, individuals and communities using a multistage stratified sample. The IFLS-5 accurately represents 83% of Indonesians.³³ The data can be accessed free of charge at https://www.rand.org/labor/FLS/IFLS.html. The research was sponsored by Survey METER and RAND Corporation (US).¹² More in-depth information about the survey and sampling methodology can be accessed at https://www.rand.org/well-being/social-and-behavioural-policy/data/FLS/IFLS/ifls5.html.³⁴

Variables and Measurements

We limited our analysis to the IFLS-5 cross-sectional survey of adults, which is the most current national survey available providing a cross-sectional assessment of HTN. Data of respondents younger than 15 years old or incomplete responses on the HTN factor were excluded from this study (Figure 1). In constructing our multiple regression model for HTN, we chose variables based on theoretical and practical significance. We drew insights from previous studies outside Indonesia to validate their relevance. We prioritized variables with low missing data percentages to enhance analysis robustness. Additionally, we exclude variables like cholesterol levels due to reliance on self-reporting, opting for more objectively measured data to uphold model integrity.

Figure 1 Flow diagram of the sample-selection process.

Hypertension

Respondents older than 15 years had their BP checked three times at home by specially trained nurses using Omron digital self-inflating sphygmomanometers.³³ The American Heart Association standard was used as the basis for the IFLS-5 BP measurement database.³⁵ Before the first measurement, respondents sat in silence in a reclined chair with their legs uncrossed, both feet flat on the floor and their forearms resting on a table at heart level.³³ The first measurement was taken at the start of the interview, and the next two were taken during the interview. The average of the three measurements was used for this study.¹² While both systolic and diastolic measurements are crucial for diagnosis and treatment, clinical studies indicate that among individuals aged 50 and above, systolic BP might be a more significant predictor of adverse events. Conversely, in individuals under 50, diastolic BP may exhibit slightly greater predictiveness.³⁶ Elevated systolic BP without a concurrent increase in diastolic BP is prevalent in the elderly, yet other research emphasizes the essential role of diastolic BP in younger populations.^37–43 Due to the varying emphasis on systolic and diastolic BP in the literature, Kundu et al⁴⁴ propose the use of mean arterial pressure (MAP) for a more accurate prediction of BP compared to relying on systolic or diastolic BP alone. MAP, representing the average arterial BP over a single cardiac cycle, emerges as a potentially superior option for diagnosing and statistically analyzing BP.⁴⁵ This suggestion finds support in studies examining the association of BP with potential causative factors,⁴⁴ demonstrating an improved ability to identify mild cases of hypertension.⁴⁵ For categorizing the respondents as hypertensive or non-hypertensive, a MAP value less than 100 is considered non-hypertensive.⁴⁴ The HTN variable was categorized as hypertensive or non-hypertensive.

Lifestyle Habits

Three health risk behaviours were investigated: smoking, sleep quality and physical activity. Smoking was assessed with the question “Have you ever chewed tobacco, smoked a pipe, smoked self-rolled cigarettes or smoked cigarettes/cigars?”³³ Sleep quality was assessed with the statement “My sleep quality is very bad, fair, good, and very good”.³³ The first three categories were coded as (0) indicating poor sleep quality while the last category was coded as (1) indicating good sleep quality.⁴⁶ Physical activity over the previous seven days was assessed using a modified version of the International Physical Activity Questionnaire (IPAQ), short version (IPAQ-S7S). The IPAQ manual instructions were followed, and physical activity was classified into low, moderate and high categories based on the official IPAQ scoring protocol.⁴⁷ The question was “During the last 7 days, did you do any vigorous, moderate or walking activities for at least 10 minutes continuously?”³³ A ‘Yes’ response was categorised as good physical activity, while a “No” response was categorised as a lack of physical activity.

Socio-Demographic Factors

The socio-demographic factors assessed were age, gender (male and female), marital status, (currently unmarried/married), ethnicity (Javanese and non-Javanese) and occupational status (unemployed and employed).³³

Statistical Analysis

All responses were grouped into two categories, except for age. The categories were collapsed to ensure sufficient numbers in each and enable meaningful analysis.⁴⁸ The variable characteristics were analyzed using descriptive statistics, while the univariate association between the potential factors and outcome was evaluated using a chi-square test. Multivariable logistic regression was used to evaluate the association between assessed factors and HTN. The odds ratio (OR) with a 95% confidence interval (CI) was reported. The analysis was performed using the SPSS program (SPSS version 26, IBM Corporation, Armonk, NY, USA).⁴⁹

Results

Baseline Characteristics

In total, 32,670 respondents were included in this study (Figure 1). Approximately half were female (53.3%), non-Javanese (56.1%) and employed (57.9%). Most respondents were married (71.4%), non-smokers (63.9%), had poor sleep quality (95.9%), had no physical activity (78%) and were aged 25–34 years (26.1%). Among all respondents, 31.2% had HTN (Table 1).

Table 1 Characteristics of Respondents (n=32,670)

Association Between Potential Factors and Hypertension

In the multivariable logistic regression analysis, all age groups were found to have statistically significant associations with HTN. Specifically, respondents aged <65 years had higher odds of developing HTN than those aged 65 or older. Marital status was also associated with HTN, with being married showing a significant association (OR 1.257, 95% CI 1.170–1.352). Furthermore, smokers had higher odds of having HTN (OR 1.297, 95% CI 1.198–1.404) than non-smokers. Conversely, being male was less associated with an increased risk of HTN (OR 0.677, 95% CI: 0.625–0.733) compared to females. Additionally, not doing physical activity reduced the risk of HTN (OR 0.870, 95% CI: 0.813–0.930) (Table 2).

Table 2 Multivariable Regression Analysis of Study Variables Associated with Hypertension

Discussion

Of all 32,670 respondents, more than a quarter experienced HTN, which was highest in the 45–54-year age group. All age groups, marital status and smoking factors showed an increased risk of having HTN, while gender and physical activity factors reported fewer associations with HTN.

This study showed a relatively high prevalence of HTN. Similarly, a study conducted in Myanmar in 2009 found that the prevalence of HTN in 15–64-year-olds was 30.1%.⁵⁰ Another study conducted in China in individuals aged 18 years and older had a prevalence of 29.6%.⁵¹ A study conducted in Indonesia, using IFLS-4, found that the prevalence of HTN among Indonesian adults ≥40 Years was 47.8%,¹² which is comparable to the age-specific prevalence reported in our own study.

Indonesia is one of the largest markets in the world for tobacco.⁵² Like ours, a study in China with the same measurements of smoking and HTN revealed that former smokers had an increased risk of having HTN with ORs 1.48 (95% CI: 1.01–2.18). Also, in line with our findings, a Swedish cohort study demonstrated that women who currently smoke have higher adjusted mean BP than non-smokers have.⁵³ According to another study, smoking causes elevated BP mostly by stimulating the sympathetic nervous system.⁵⁴ Our study also found that individuals who performed no physical activity had a lower likelihood of experiencing HTN. The relationship between physical activity and HTN is complex. The specific mechanisms by which physical activity prevents HTN are not yet fully understood due to the multifactorial nature of the condition.⁵⁵ The authors of a study of the potential mechanisms underlying the blood pressure-lowering effects of physical activity propose that physical activity can reduce BP by decreasing vascular resistance, involving the sympathetic nervous system and the renin-angiotensin system. This suggests that physiological pathways influenced by physical activity play a role in modulating BP levels.⁵⁶ These studies show that physical activity’s impact on HTN is a complex interplay of many factors.

Ages less than 65 years exhibited a statistically significant association with HTN in our study. The involvement of various social, nutritional and psychological factors in HTN has been reported.⁵⁷ A longitudinal study reporting an in-depth analysis of psychosocial variables, including elevated levels of time urgency, impatience and hostility revealed an association with the risk of developing HTN over a 15-year period in a cohort of young adults aged 18–30 years.⁵⁸ Regarding gender, our study did not find a significant association with HTN. However, one study found that Chinese and Indian people but not Malays, showed significant associations between gender and HTN.²³ Since part of the Indonesian population consists of Malays, these findings may explain the absence of gender-related associations in our study. Furthermore, another study exploring the impact of gender on HTN, explained that gender can be a risk factor for developing HTN due to the influence of sex hormones on renal sodium regulation and vascular resistance. The study suggested that male sex hormones, such as androgens, may contribute to higher BP levels in men, while female sex hormones appear to have a protective effect against salt-induced increases in BP.⁵⁹ Surprisingly, the marital status factor in this study showed an association with HTN; we do not fully understand the mechanics underlying this. Psychopathological factors, neuroendocrine pathways, health habits (exercise, nutrition and adherence), biological mediators and immunological pathways have all been proposed in studies as potential explanations for the impacts of marital status on BP.⁶⁰

This study had many respondents, which provided robust findings and bolstered the statistical power of the analysis. Additionally, the HTN data was gathered by measuring the respondents’ right and left arms three times each to make sure the data collected was more reliable. Furthermore, previous studies did not address quality of sleep as a potential risk factor for HTN. Conversely, this study was cross-sectional and causal association cannot be proven. Moreover, it is important to consider the influence of social desirability bias on the results, as individuals may provide answers that conform to societal norms or expectations, rather than providing truthful or precise responses.

The development of suitable interventions for HTN in Indonesia is a big concern. Interventions should focus on promoting healthy lifestyles and early detection and treatment of HTN in all age groups. Social support networks, stress reduction techniques and counselling may be effective in reducing HTN risk in married individuals. Smoking cessation programs and nicotine replacement therapy may help reduce HTN risk in smokers. Healthcare providers should emphasize the importance of these interventions and provide resources to assist in their implementation to promote healthier lifestyles in Indonesia. Future studies should also explore the potential impact of additional factors, such as obesity and sedentary habits, as these lifestyle factors may play a significant role in contributing to HTN.

Conclusion

Our study findings suggest an association between certain socio-demographic and lifestyle factors and HTN, indicating a need for policy interventions aimed at reducing HTN rates. These interventions should focus on addressing the factors identified in this study.