Article Text
PDF
Abstract
Objectives This study estimated the prevalence of hypertension, in accordance with the American College of Cardiology and American Heart Association’s 2017 guidelines, and examined the association between various socioeconomic factors and systolic blood pressure (SBP), diastolic blood pressure (DBP) and hypertension.
Setting and design We used nationally representative data from the 2016 Nepal Demographic and Health Survey. Multivariate analysis was used to study the association of hypertension with socioeconomic factors: logistic regression was used for hypertension and linear regression was used for DBP and SBP.
Participants Our sample consisted of 9827 adults between the ages of 15 and 49 years.
Results The prevalence of hypertension was 36%. The mean DBP and SBP were 76.4 and 111.5, respectively. Janjatis (adjusted OR (AOR): 1.34, CI: 1.12 to 1.59), Other Terai castes (AOR: 1.38, CI: 1.03 to 1.84), Muslim and other ethnicities (AOR: 1.64, CI: 1.15 to 2.33) and Dalits (AOR: 1.26, CI: 1.00 to 1.58) had higher odds of hypertension. Individuals employed in professional, technical and managerial professions collectively (AOR: 1.62; CI: 1.18 to 2.21) also had higher odds of hypertension. Moderately food insecure household had lower odds of hypertension (AOR: 0.84; CI: 0.72 to 0.99) compared with households with no issue of food insecurity. Results were similar for SBP and DBP. When stratified by sex, there were differences mainly in terms of occupation and ethnicity.
Conclusion There are substantial disparities in hypertension prevalence in Nepal. These disparities extend across ethnic groups, occupational status and food security status. Differences also persist across different provinces. As hypertension continues to be increasingly more significant, more research is needed to better understand the disparities and gradients that exist across various socioeconomic factors.
- hypertension
- public health
- health policy
Data availability statement
Data are available upon reasonable request. All datasets in this analysis are available at http://www.measuredhs.com.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
Strengths and limitations of this study
This is the first study, to the best of our knowledge, to use nationally representative data to examine specifically the association of hypertension/blood pressure with ethnicity, occupation and food security status.
In addition to using hypertension as a binary outcome, this study uses systolic and diastolic blood pressure as a continuous outcome measure, thereby addressing cutoff-point issues linked with recent changes in definition of hypertension.
Study sample only includes adults between the ages of 15 and 49 years.
This is an associational study based on cross-sectional data, and therefore, the findings cannot be used make causal claims.
Introduction
Hypertension is a major risk factor for cardiovascular diseases (CVDs) and a significant contributor to the growing burden of non-communicable diseases (NCDs) in low-income and-lower-middle-income countries (LMICs). Nearly two-thirds of the people with hypertension live in LMICs and about one in every third person in LMICs has hypertension.1 2 With the increasing growth in the share of elderly populations, hypertension also stands to be a major risk factor for cognitive impairment and dementia in LMICs.3
Nepal is no exception to the hypertension trend that has been observed in other LMICs. A recent analysis found that the prevalence of hypertension among Nepalese adults is 44.2%.4 The number varies by subgroups. It is nearly 30 percentage points higher for those with body mass index (BMI) of 30 or more, 20 percentage points higher for people older than 70 and 6 percentage points higher for those from the richest quintile. Existing studies on hypertension in Nepal have focused predominantly on the prevalence of and factors associated with hypertension.4–8 Among the most studied factors include age, sex, BMI, education level, marital status, wealth index and smoking and drinking status. Existing studies of hypertension in Nepal have not explored the role of ethnicity, occupation and household food security status. Studies outside of Nepal have looked at the role of unmet social needs (eg, food security) and health outcomes,9 and suggests a high positive correlation between unmet social needs and poor health outcomes.
Our study contributes to the literature in a few ways. First, this is the first Nepal-based multivariate study, to the best of our knowledge, that uses a nationally representative data to understand the association of hypertension with ethnicity, occupation and food security. Using food security as an important determinant for hypertension and blood pressure, the study also contributes to the growing literature on the role of unmet social needs on health and health outcomes. Second, this study looks at three different outcome measures. In addition to the binary measure of hypertension, the study uses continuous measures of blood pressure (corresponding to diastolic blood pressure (DBP) and systolic blood pressure (SBP) measurements). Few studies have examined the change in hypertension prevalence using the new definition but the evidence on Nepal is still limited overall.4 10 Using the continuous measure of SBP and DBP allows us to study the risk of high blood pressure without relying too much on a single binary outcome. SBP and DBP are measures of risk that do not rely on a particular cut-off. The use of these measures mitigates the cut-off point issues associated with the change in the definition of hypertension. Additionally, there is evidence that SBP is a predictor of risk for CVDs.11 We also stratified our analysis by sex to capture if there are substantial differences by sex.
Finally, our study focuses on a much younger population (15–49 years) than the earlier studies. Even though the younger population is less likely to suffer from hypertension, it is still susceptible to the risks of hypertension due to changes in lifestyle, unmet social needs, occupational status and other correlates including sex and BMI.
Methods
Data source
We used data from Nepal Demographic and Health Survey (NDHS) 2016. Data were collected from June 2016 to January 2017 by New ERA with support from the Ministry of Health. The NDHS provides up-to-date estimates of the basic demographic and health indicators as well as a comprehensive overview of population, maternal and child health issues in Nepal.
Survey design
The sampling frame used for the 2016 NDHS was an updated version of the frame from the 2011 National Population and Housing Census. The frame was updated since the urban/rural classification changed at the ward level in 2014 and 2015. The sample was stratified and selected in two stages in rural areas and three stages in urban areas. In rural areas, wards were selected as the primary sampling units (PSUs) and households were selected from the sample PSUs. In urban areas, wards were selected as the PSUs. One enumeration area (EA) was selected from each PSU and then households were selected from the sample EAs. Our sample consisted of a total of 9827 individuals aged 15–49 years. All these individuals had complete information on our variables of interest.
Outcome measures
Our main outcome variable was hypertension, defined in line with the 2017 American College of Cardiology (ACC)/American Heart Association (AHA) guidelines (SBP≥130 mm Hg and/or DBP≥80 mm Hg). Individuals were also considered hypertensive if they reported taking antihypertensive medications. Our additional outcomes included measures for SBP and DBP. The survey measured respondents’ blood pressure three times. The average of the second and the third reading was used.
Explanatory variables
Our main explanatory variables were ethnicity, occupation and food security status. Ethnicity was divided into six categories.12 Occupational status was divided into seven categories. Household food security status was divided into four categories: food secure, mildly food insecure, moderately food insecure and severely food insecure. The four categories were created using the questionnaires that focused on food insecurity for the household as a unit. The questions, arranged in order of severity and frequency of occurrence, captured households’ perceptions of food vulnerability or stress and behavioural responses to food insecurity. The questions followed the Household Food Insecurity Access Scale developed by USAID’s Food and Nutrition Technical project.13
Other explanatory variables included provincial location, rural/urban status, ecological region, sex, age group, BMI category, highest level of education, sex of the household head and household wealth index. Nepal has seven provinces and three ecological regions. Age was categorised into seven groups (15–19, 20–24 and so on). Similarly, BMI was categorised into four groups: thin (BMI<18.5), normal (BMI≥18.5 and BMI<25), overweight (BMI≥25 and BMI<30) and obese (BMI≥30). Educational attainment was grouped into four categories: no education and/or preschool, primary, secondary and higher secondary. The household wealth index, derived using data on household assets and other household characteristics such as access to water and sanitation facilities, consists of five quintiles (from poorest to the richest). Additional independent variables included current work/employment status, media use, marital status and tobacco use. A dummy variable for media use was created based on the frequency of the use of internet, radio, television or reading of newspaper or magazine. Tobacco use, also a dummy variable, included any use of tobacco.
Statistical analysis
We used multivariate regression analysis to understand and quantify the magnitude of the association between the outcome measures (ie, hypertension, DBP and SBP) and the independent variables. We used logit model for our binary outcome variable (hypertension) and linear models for the two continuous outcome variables (DBP and SBP). Key independent variables were ethnicity, occupation and food security status. We also included other independent variables (such as sex, BMI, education and others) that have been widely used in the hypertension literature. Additionally, when the outcome was hypertension, we stratified the analysis by sex. All tests were two-tailed and a p-value of <0.05 was considered statistically significant. We applied sample weight calculated in the DHS and adjusted for clustering using the ‘svy’ command in Stata.
All the statistical analysis was performed in Stata V.15.1 (StataCorp). Approval to access and use the DHS data was obtained from the ICF International.
Patient and public involvement
Study participants or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.
Results
We present descriptive statistics in table 1. In our study sample comprising of 9827 people between the ages of 15 and 49 years, 36% had hypertension. The mean values of SBP and DBP were 112 and 76, respectively. At 32%, Janjatis were the largest ethnic group. Nearly 41% of the study population was self-employed in agriculture, while 25% of the population did not work or held other occupations. Around 50% of the people lived in food-secure households. Nearly 73% of the study population was married and 26% used tobacco products. Additionally, 50% of the people lived in the Terai region, 63% resided in urban areas, 61% were females and 26% lived in households headed by females. Approximately 70% of the study population used media of some kind in the week prior to the survey.
Characteristics of the study population
Table 2 shows results for multivariate regression models for all three outcome measures: hypertension (measured as a binary variable), DBP (measured as a continuous variable) and SBP (measured as a continuous variable). Our results for hypertension, presented in terms of adjusted OR (AOR) in panel A of table 2, suggest the following: compared with Brahmins/Chhetris, Janjatis (AOR: 1.34; CI: 1.12 to 1.59; p<0.01), Other Terai castes (AOR: 1.38; CI: 1.03 to 1.84; p<0.05), Muslim and other ethnicities (AOR: 1.64; CI: 1.15 to 2.33; p<0.01) and Dalits (AOR: 1.26; CI: 1.00 to 1.58; p<0.05) had higher odds of hypertension. Individuals employed in professional, technical and managerial professions collectively (AOR: 1.62; CI: 1.18 to 2.21; p<0.01) also had higher odds of hypertension compared with those who did not work or were employed in other jobs. Moderately food insecure household had lower odds of hypertension (AOR: 0.84; CI: 0.72 to 0.99; p<0.05) compared with households with no issue of food insecurity. Compared with the youngest population in the study sample (ie, 15-year to 19-year-olds), older adults had higher adjusted odds of hypertension. More notably, there was an age gradient in the adjusted odds of having hypertension. As implied by the magnitude of the AOR, the association between hypertension and age was stronger for older age groups. Compared with individuals in the lowest BMI category, those in the higher BMI categories had higher odds of hypertension. In terms of the geographic location, individuals residing in Gandaki province (AOR: 1.85; CI: 1.35 to 2.54; p<0.001) and Lumbini province (AOR: 1.80; CI: 1.33 to 2.44; p<0.001) had higher odds of hypertension. Individuals in richer households had lower odds of hypertension compared with individuals in the poorest households. For example, individuals residing in the richest (OR: 0.60; CI: 0.46 to 0.80; p<0.001) and the second richest (OR: 0.66; CI: 0.50 to 0.87; p<0.01) households had lower odds of hypertension compared with individuals living in the poorest households. Compared with males, females had lower odds (OR: 0.64, CI: 0.55 to 0.73; p<0.001). No significant association was recorded between hypertension and the use of tobacco products, current working status, marital status, media use, and the place of residence (in terms of ecological zones).
Association of hypertension, diastolic blood Pressure (DBP) and systolic blood Pressure (SBP) with social, economic and geographic variables
Our results on the association between DBP, presented in panel B table 2, suggest the following: compared with Brahmins/Chhetris, Dalit (estimate: 1.69; CI: 0.59 to 2.80; p<0.01), Janjatis (estimate: 1.27; CI: 0.42 to 2.12; p<0.01) and Muslim and other (estimate: 2.94; CI: 1.45 to 4.44; p<0.001) individuals had higher readings of blood pressure. Occupation wise, individuals working in sales and services (estimate: 1.39; CI: 0.30 to 2.49; p<0.05) had higher readings for blood pressure. Food insecurity was negatively associated with blood pressure. Individuals from severely food insecure (estimate: −1.46; CI: −2.54 to −0.38; p<0.01) households had lower levels of blood pressure. Blood pressure was increasing in age groups as well as in BMI—older people and people with higher BMI had higher levels. Some education (regardless of primary, secondary or higher) was associated with a higher level of DBP. Individuals living in Gandaki province (estimate: 2.76; CI: 1.15 to 4.38; p<0.001) and Lumbini province (estimate: 2.66; CI: 1.18 to 4.14; p<0.001) had higher levels of blood pressure. Females had lower levels of blood pressure compared with males. No significant association was recorded between DBP and the use of tobacco products, current working status, marital status, media use and the place of residence (in terms of ecological zones).
Finally, we present results with SBP as the outcome measure. These results, which are very similar to the results for hypertension and DBP, are presented in panel C of table 2. The results suggest the following: compared with Brahmins/Chhetris, all the other ethnic groups (including the Newars) had higher levels of SBP. Food insecure households had lower levels of SBP compared with food secure households. As with the other two outcome measures (ie, hypertension and DBP), SBP was increasing in age and BMI. Individuals who lived in Gandaki province and Lumbini province had higher levels of SBP compared with individuals who lived in province 1. Females had lower levels of SBP compared with males. No significant association was recorded between SBP and tobacco use, current working status, education, marital status, media use and the place of residence (in terms of the ecological zones).
Table 3 presents results stratified by sex. We see that occupation is positively associated with hypertension for males only. Other categories that suggest somewhat differential association for the two sexes include ethnicity, food security, wealth, education, provincial and ecological residence, tobacco use and female headed household. Increasing age and BMI is positively associated with hypertension for both males and females. Compared with individuals who are thin, those who are normal, overweight or obese are more likely to be associated with hypertension.
Association of hypertension with social, economic and geographic variables stratified by sex
Discussion
Using cross-sectional data, our study examined hypertension and its correlates for Nepal. Our analysis, based on the 2016 DHS data, found that around 36% of individuals between the ages of 15 and 49 in Nepal are hypertensive, suggesting a significant fraction of adults with an elevated risk of CVDs. Given the existing evidence that younger population is more likely to have undiagnosed hypertension,14 population at risk for CVD in Nepal may be higher than what is reported, further underscoring the treatment gap and the need for addressing the gap.
Our study identified multiple factors associated with hypertension (a binary measure), DBP (a continuous measure) and SBP (a continuous measure). Across all three outcome measures, we found that relative to Brahmins/Chhetris, other ethnic groups have higher odds of hypertension. This finding could be explained by a few factors. First, access issues may be especially pronounced for certain ethnic groups, as suggested by some of the existing evidence.11 Brahmins/Chhetris have always been two of the most privileged groups in Nepal. Relative to other ethnic groups, these two groups have historically enjoyed, on average, greater social and political advantage in the country. This advantage likely translates into access to healthcare as well as other social determinants of health (such as education, employment, housing and others). Second, the differences in hypertension rates across the groups may also reflect, to some extent, ethnicity-specific food and cultural practices and preferences.15
We found that, for the most part, food insecurity was associated with a lower rate of hypertension, DBP and SBP. This finding is somewhat counterintuitive because available evidence from LMICs suggests that food insecurity is associated with unfavourable health outcomes such as obesity.16 Since obese people are more likely to be hypertensive than non-obese people, we would expect to see a positive association between food insecurity and hypertension. Our seemingly discrepant finding may be a function of the underlying mechanism, which operates not through the ‘high-energy, processed foods’ but likely through the food unavailability route, among others. We found that jobs that are more office-oriented were associated with high blood pressure and hypertension. The reasons might be that individuals who work on these jobs are mentally more stressed but end up doing limited physical work. In terms of geographic locations, Gandaki province and Lumbini province show higher rates of hypertension and blood pressure. It is unclear, ex-ante, the reason for geographic variation in the prevalence of hypertension. We did not find any evidence of urban residence, marital status, household head, exposure to media and the use of tobacco products affecting blood pressure or hypertension.
When stratified by sex, we found both similarities and differences in terms of the association between socioeconomic several factors. Our results complement earlier work that has also shown that hypertension and/or prehypertension were associated with being overweight or obese, tobacco use, alcohol consumption, age group, education and the place of residence.7 17 While earlier studies included individuals that were older, our study included only individuals who were no more than 49 years. Compared with other studies, our study instead uses occupation, ethnicity and food security as additional covariates that influence hypertension.
Few studies have looked at the change in hypertension status in Nepal using the new definition.4 5 10 These studies have reported higher risks of hypertension among people who are older, males and have higher BMI.18 19 Our results suggest similar association. In addition, DBP and SBP also show clear positive association with age, gender and BMI. Furthermore, our results suggest that hypertension and blood pressure (diastolic and systolic) increase monotonically with age and increasing BMI.
It is important to understand the differences in hypertension prevalence among different ethnic groups and the underlying causes of these differences. Similarly, more research needs to be done to better understand geographic disparities. An important dimension to consider on this could be the ethnic composition of the different provinces and relatedly, the food and cultural practices that accompany different ethnic groups. Nepal has been seeing an increasing trend in BMI over time.17 Any future policy should target individuals with high BMI in Nepal.
Limitations
There are limitations to this study. First, though we found certain ethnicities had higher rates of hypertension, we were unable to disentangle as to why this might be the case. Second, we could not rule out the selection of individuals into certain occupations. Relatedly, we also could not measure the type of work done in each occupational setting and how these affected measures of blood pressure. Type of work done may hide variation in blood pressure due to the demands of jobs within occupations. Third, food insecurity likely did not sufficiently capture the spectrum of unmet social needs that might be driving the variation. Finally, this study was based on individuals between the ages of 15 and 49 years. It is therefore important to recognise that hypertension for this group was lower than the overall adult population (which comprises of the elderly population).
Conclusion
Our study showed that certain ethnic groups and occupations were associated with higher rates of hypertension. These groups also had higher levels of SBP and DBP. We found that food insecurity was associated with reduced likelihood of hypertension, and lower levels of DBP and SBP. Consistent with the findings of the existing research, increasing age, male sex and higher BMI were correlated with hypertension as well as SBP and DBP. Geographically, two provinces (Gandaki and Lumbini) had higher rates of hypertension. More research is required to understand the impact of NCDs among different occupations and ethnic groups. It is also important to understand geographic disparities in terms of hypertension prevalence and the related risks.
Data availability statement
Data are available upon reasonable request. All datasets in this analysis are available at http://www.measuredhs.com.
Ethics statements
Patient consent for publication
Ethics approval
This study was based on the analysis of deidentified, secondary data that are publicly available. We obtained permission to use the data from ICF International/Demographic Health Survey (DHS) team.
References
Footnotes
Contributors BT and SJ jointly conceived and planned the overall project. SJ sought approval for data use, led the data analysis, and takes responsibility for the final content. BT and SJ jointly wrote the draft and revised the paper. BT and SJ reviewed and approved the final manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.