Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Time Trends in Blood Pressure, Body Mass Index and Smoking in the Vietnamese Population: A Meta-Analysis from Multiple Cross-Sectional Surveys

  • Quang Ngoc Nguyen ,

    quangtm@gmail.com

    Affiliations Department of Cardiology, Hanoi Medical University, Hanoi, Vietnam, Vietnam National Heart Institute, Bach Mai Hospital, Hanoi, Vietnam, Department of Public Health and Clinical Medicine, Umeå Centre for Global Health Research, Umeå University, Umeå, Sweden

  • Son Thai Pham,

    Affiliations Vietnam National Heart Institute, Bach Mai Hospital, Hanoi, Vietnam, Department of Public Health and Clinical Medicine, Umeå Centre for Global Health Research, Umeå University, Umeå, Sweden

  • Viet Lan Nguyen,

    Affiliations Department of Cardiology, Hanoi Medical University, Hanoi, Vietnam, Vietnam National Heart Institute, Bach Mai Hospital, Hanoi, Vietnam

  • Lars Weinehall,

    Affiliation Department of Public Health and Clinical Medicine, Umeå Centre for Global Health Research, Umeå University, Umeå, Sweden

  • Ruth Bonita,

    Affiliation School of Population Health, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand

  • Peter Byass,

    Affiliation Department of Public Health and Clinical Medicine, Umeå Centre for Global Health Research, Umeå University, Umeå, Sweden

  • Stig Wall

    Affiliation Department of Public Health and Clinical Medicine, Umeå Centre for Global Health Research, Umeå University, Umeå, Sweden

Abstract

Introduction

Data for trends in cardiovascular disease (CVD) risk factors are needed to set priorities and evaluate intervention programmes in the community. We estimated time trends in blood pressure (BP), anthropometric variables and smoking in the Vietnamese population and highlighted the differences between men and women or between rural and urban areas.

Methods

A dataset of 23,563 adults aged 25–74 from 5 cross-sectional surveys undertaken within Vietnam from 2001 to 2009 by the Vietnam National Heart Institute was used to estimate mean BP, weight, waist circumference (WC), body mass index (BMI), the prevalence of hypertension, adiposity or smoking, which were standardised to the national age structure of 2009. Multilevel mixed linear models were used to estimate annual changes in the variables of interest, adjusted by age, sex, residential area, with random variations for age and surveyed provinces.

Findings

Among the adult population, the age-standardised mean systolic and diastolic BP increased by 0.8 and 0.3 mmHg in women, 1.1 and 0.4 mmHg in men, while the mean BMI increased by 0.1 kgm−2 in women, 0.2 kgm−2 in men per year. Consequently, the prevalence of hypertension and adiposity increased by 0.9 and 0.3% in women, 1.1 and 0.9% in men with similar time trends in both rural and urban areas, while smoking prevalence only increased in women by 0.3% per year. A U-shaped association was found between age-adjusted BP and BMI in both sexes and in both areas.

Conclusions

From 2001 to 2009, mean BP, weight and WC significantly increased in the Vietnamese population, leading to an increased prevalence of hypertension and adiposity, suggesting the need for the development of multi-sectoral cost-effective population-based interventions to improve CVD management and prevention. The U-shaped relationship between BP and BMI highlighted the hypertension burden in the underweight population, which is usually neglected in CVD interventions.

Introduction

Three recent analyses of the trends in systolic blood pressure (BP), body mass index (BMI) and serum total cholesterol over the past generation raised concern about a global epidemic of risk factors for cardiovascular diseases (CVD) and called for actions to avoid catastrophic outcomes, especially in low and middle income countries (LMICs) where 80% of the total CVD burden occurs [1], [2], [3], [4]. However, the calculations for developing countries contained some uncertainties due to the lack of longitudinal, national-level, high quality, standardised or comparable surveys from LMICs. In Vietnam, a number of population-based surveys in different areas of the country showed an increasing prevalence of hypertension, obesity and smoking but none assessed the time trends of these major CVD risk factors [5], [6], [7]. Using data from published and unpublished studies, a comprehensive individual participant-level dataset from 2001 to 2009 was collated. Taking into account the variations in age patterns and geographical areas for each component survey, we estimated the time trends in BP, BMI, and smoking status in adult Vietnamese population over a 9-year period and highlighted the differences between men and women as well as the differences between urban and rural areas. These findings will be important for policy planners to coordinate the intervention strategies required to reduce the emerging CVD burden in Vietnam.

Methods

Study Population and Data Source

The integrated individual participant-level dataset used in this study came from several population-based cross-sectional surveys with similar designs and protocols from 2001 to 2009. All studies were designed and carried out by the Vietnam National Heart Institute (VNHI). In total, five datasets were used for this study and included the following surveys:

  1. The national epidemiological survey on hypertension and its risk factors in 8 provinces around Vietnam from 2001 to 2008 (NESH) [7]: multi-stage sampling strategies were applied to randomly select 110 participants from each commune, 3 communes from each district and 4 districts from each province. In total 9,823 participants (overall response rate of 93.0%) were recruited from 8 provinces (including 4 northern provinces: Hanoi, Thai Binh, Thai Nguyen, Nghe An, and 4 southern provinces: Khanh Hoa, Dac Lac, Dong Thap and Ho Chi Minh City).
  2. The survey on heart failure and its risk factors in 4 northern provinces of Vietnam from 2003 to 2004 (HF-S) (unpublished data): a similar sampling strategy to NESH was applied to select 4,840 participants (overall response rate of 91.7%) in 4 northern provinces of Vietnam: Hanoi, Thai Binh, Thai Nguyen and Nghe An.
  3. The survey on diabetes and its risk factors from 2008 to 2009 in 2 northern provinces of Vietnam (DM-S) [8]: a similar sampling strategy to NESH was applied to select 2,306 participants (overall response rate of 87.3%) in 2 northern provinces of Vietnam: Hanoi, Thai Binh.
  4. The survey on non communicable disease risk factors in FilaBavi in 2005 (NCDS) [5]: stratified random sampling was applied to select 250 people in each sex and 10-year group, which followed WHO’s STEPwise approach [9] and was based on the FilaBavi sampling frame [10]. Total 2,362 participants (overall response rate of 94.5%) were recruited from Ba Vi District of Ha Tay province.
  5. The screening surveys at all communes from the project on hypertension management in rural communes (HMPS) before 2010 when the National Target Project on Prevention and Control of Hypertension was started in Vietnam (unpublished data): in these surveys, simple random sampling strategy was applied to select participants from a list of all local inhabitants at the moment of the surveys. Total 5,855 participants (overall response rate of 97.6%) were recruited. These datasets included the pre-intervention cross-sectional surveys at Xuan Canh commune of Dong Anh District, Hanoi in 2004 and at Phu Cuong commune of Ba Vi District, Ha Tay province in 2006. It also included the cross-sectional surveys in 2007 and 2009 at Phu Phuong and Phu Cuong communes, which were control and intervention communes in a quasi-experimental study in Ba Vi District, Ha Tay Provinces.

To maximize comparability across the 5 surveys, we restricted analyses to examined, non-pregnant adults who were aged 25 to 74 years. This consisted of 9,403, 4,494, 2,098, 2,357 and 5,210 persons or 95.7%, 92.9%, 91.0%, 99.8% and 89.0% of studied samples in the 5 datasets, respectively.

CVD Risk Factor Assessments

In all surveys, physical measurements such as BP, weight, height, waist circumference (WC) and hip circumference were taken using the same standardised protocol which has been described elsewhere [7]. BP was measured at least twice in a resting, sitting position using an automatic digital sphygmomanometer (OMRON Healthcare Inc.®, Bannockburn, Illinois), with an appropriately sized cuff. Hypertension was defined as an average systolic BP (SBP) ≥140 mmHg, and/or average diastolic BP (DBP) ≥90 mmHg, and/or self-reported current treatment with antihypertensive medications [11], [12]. Awareness of hypertension was defined as self-report of any prior diagnosis of hypertension by a health care professional. Treatment of hypertension was defined as use of a prescription medication for management of high BP at the time of the interview. Controlled hypertension was defined as pharmacological treatment of hypertension associated with an average SBP<140 mmHg and DBP<90 mmHg.

All other anthropometric measurements were performed at least twice with the participants wearing light clothing and no footwear. BMI was calculated as weight (kg) divided by height squared (m2). In this study, we used a BMI cut-off point of <18.5 kgm−2 for underweight, 18.5–22.9 kgm−2 for normal range, ≥23 kgm−2 for overweight, and ≥25 kgm−2 for generalized obesity. Abdominal obesity was defined as WC either ≥90 cm in men or ≥80 cm in women. Adiposity was defined as having either generalized obesity or abdominal obesity or both. These criteria have been specified for South-Asian populations by WHO Regional Office for Western Pacific [13], [14]. People who smoked tobacco products such as cigarettes, cigars or pipes over the previous month were classified as current smokers. Urban or rural residential area was identified on an administrative basis for each commune within each province.

CVD Risk Factor Measurements in Studied Population

The prevalence of hypertension and obesity as well as the mean levels of individual BP and anthropometry measurements for each year between 2001 to 2009 were weighted and age-standardised for both sexes and residential areas, using age distribution data for related provinces from the Vietnam Population and Housing Census in 2009 [15].

Estimation of the annual incidence of hypertension was based on annual trends in consecutive prevalence of hypertension in the population. Using demographic data for the Vietnamese population structure and age distributions from the General Office for Population and Family Planning [16] and the age-standardised surveyed prevalence of hypertension from studied samples for each year during the period from 2001 to 2009, we estimated the number of people with hypertension among the general adult population aged 25–74. Assuming that hypertension is a life-long progressive disease until death; the prevalence of hypertension was very low at the age of 25 years; the general population changed consistently every year and hypertension-related case-fatality was constant, the difference between two consecutive values for current cases of hypertension in the population was used to extrapolate the number of new cases of hypertension every year. Then we estimated the annual average number of new cases of hypertension during the period 2001–2009 and the average population at risk from 2001 to 2009 to calculate the average annual incidence of hypertension among the adult population aged 25–74 years, separately for each sex and each residential area.

Time Trends Analysis

The prevalence of hypertension, adiposity and smoking were aggregated and analysed by 3-year periods (2001 to 2003, 2004 to 2006, and 2007 to 2009) to yield clearer time trends across sex and residential areas. For the visual demonstrations of trends across time, we plotted trends using linear fit plots plus smooth fractional-polynomial fit plots based on the age-adjusted mean of relevant variables over surveyed years with corresponding estimated 95% confidence intervals (CI).

The unadjusted p-value for trend was calculated using univariable linear regression analysis (function regress in STATA) for each relevant variable, using year as a single continuous independent variable. For the multivariable analyses of the time trends based on all individual-level observations, we modelled trends using multilevel mixed-effect linear regression (function xtmixed in STATA), fitted via maximum likelihood, containing both fixed-effects and random-effects parts in order to compensate the potential heteroskedasticity among component surveys. Using the variables of studied year (treated as a continuous variable), participant’s age, sex and residential areas for the fixed-effects, this part allowed estimation of the changes in relevant variables by year after adjusting for age, sex and residential areas. Taking into account the fact that each survey was carried out in a specific geographical area (province or district) at a specific time of year in a selected sample representative of a specific population with possible different variations, the random-effects were random slope models adjusted for age and surveyed year and grouped by the studied provinces. For example, the mixed-effect model for SBP (model 1) was: SBP  = α+β1×survey year + β2×age + β3×sex + β4×residential area + Z× province× (b1×survey year + b2×age). The value of β1 in model 1 is the estimated annual mean change in SBP (or annual change in other relevant variables in other similar models) after adjusting for age, sex, and residential area with random effect part of Z grouped by each province at a random slope (b1 and b2 were the intercepts for age and surveyed year). A significant β coefficient (i.e. its 95% CI did not cover zero) with corresponding p-value (adjusted p for trend) suggested a significant trend over time.

To highlight the difference between women and men, we used similar models separated by sex or the original models with the addition of a sex-by-survey year interaction term as an independent variable, like the following model (model 2 for SBP for example): SBP  = α+β1×survey year + β2×age + β3×sex + β4×residential area + β5× (sex×survey year) + Z×province× (b1×survey year + b2×age). A negative coefficient for the interaction term β5 suggested a negative association between sex and SBP (or other relevant variable) over calendar years (i.e. a decrease in the slope). A positive coefficient suggested a positive association. A statistically significant p-value for the coefficient for the interaction (i.e. its 95% CI did not cover zero) suggested a difference in time trends between two sexes. Similar models including interaction analysis were applied to highlight the difference in time trend between urban and rural areas.

Statistical Analysis

A p-value <0.05 was considered to represent statistical significance. Both descriptive and analytical statistical analyses were carried out using STATA 11 software (Stata Corporation®, Texas, USA).

Ethical Statement

The protocols of each survey were all approved by both Scientific Ethical Committees in Biomedical Research at Vietnam National Heart Institute and the ones at the corresponding involved partners. All human subjects in any surveys were asked for their written consent before collection of data, and all had complete rights to withdraw from the study at any time without any threat, or disadvantage. Any participants with hypertension or other disorders were referred to appropriate facilities for further investigation and treatment.

Results

Our analysis included 23,563 adults aged 25–74, or 93.6% of the collated dataset (after excluding 1,624 cases aged over 74 or who were pregnant). All surveys used province-based sampling, in which multi-stage cluster sample surveys (NESH, HF-S, DM-S) provided 68% of all data, stratified random sampling (NCDS) provided 10% while simple random sample surveys (HMPS) provided 22% of all data (Table 1). Each year from 2001 to 2009 contained 9–10% of the data except for 2003 (18%) and 2004 (14%). In the final collated dataset, 61.7% of participants lived in rural areas. The proportion of men was 36–41% in almost all surveys (except for NCDS, which applied a sex-stratified sampling strategy) due to the lower proportion of men in rural populations, resulting from working-age migration (especially men to urban areas in Vietnam (Table 1)).

thumbnail
Table 1. General characteristics of subpopulations aged 25–74 from multiple cross-sectional community surveys from 2001 to 2009 around Vietnam.

https://doi.org/10.1371/journal.pone.0042825.t001

Time Trends between Men and Women

The yearly variations in age-standardised mean BP (both SBP and DBP), weight, WC, BMI, and the age-standardised prevalence of hypertension (as well as prevalence of awareness, treatment and control of hypertension), adiposity (both generalized and abdominal obesity) and current smoking are summarized in Table 2 and Table 3, stratified by sex, and residential areas. The diversification by year for each variable of interest shown in Table 2 and 3 reflected the real differences of general characteristics of potentially different studied population at certain specific time or area.

thumbnail
Table 2. Trends in mean age-standardised systolic and diastolic blood pressure, weight, waist circumference, body mass index and waist-hip ratio in Vietnam population aged 25–74 between 2001 and 2009, stratified by sex and residential areas.

https://doi.org/10.1371/journal.pone.0042825.t002

thumbnail
Table 3. Trends in age-standardised prevalence of hypertension, adiposity, and current smoking in Vietnam population aged 25–74 between 2001 and 2009, stratified by sex, and residential areas.

https://doi.org/10.1371/journal.pone.0042825.t003

Each year on average, age-adjusted mean systolic and diastolic BP increased 0.9 mmHg and 0.4 mmHg respectively in general population, at a considerably higher rate in men than in women (p<0.001). Consequently, during the period of 2001–2009, the prevalence of hypertension significantly increased by 0.9% per year in general adult population (p<0.001) with similar time trends between both sexes (p = 0.498). The mean BMI increased 0.1 kgm−2 annually in the adult population (p<0.001) with non-significant differences in time trends between both sexes (p = 0.589). However, the prevalence of abdominal obesity increased more rapidly than the prevalence of generalized obesity in women while this trend was reversed in men (Table 4). In general, except for the prevalence of current smoking in men (non-significant changes over time, p = 0.224), all other variables (either mean or prevalence) increased in a linear fashion from 2001 to 2009 in both sexes (Table 4, Figure 1 and Figure 2).

thumbnail
Table 4. Adjusted annual changes of CVD risk factors, stratified by sex.

https://doi.org/10.1371/journal.pone.0042825.t004

thumbnail
Figure 1. Trends in age-standardised mean of blood pressure (systolic and diastolic), weight, waist circumference, body mass index and waist-hip ratio between 2001 and 2008 for woman and men in rural and urban areas.

The solid line represents fractional-polynomial prediction plots and the shaded area the 95% uncertainty interval.

https://doi.org/10.1371/journal.pone.0042825.g001

thumbnail
Figure 2. Changes over time in age-adjusted prevalence of hypertension, adiposity and smoking from 2001 to 2009 by urban and rural areas, demonstrated by 3-year moving averages.

https://doi.org/10.1371/journal.pone.0042825.g002

Using two consecutive extrapolations of the prevalence of hypertension in Vietnamese adults, the estimated annual incidence of hypertension among the population aged 25–74 was 65 per 10,000 in women and 165 per 10,000 in men, or the projected burden of hypertension in adult population of Vietnam was 140,000 new hypertensive women and 330,000 new hypertensive men every year (Table 5).

thumbnail
Table 5. Estimation of yearly incidence of hypertension based on the trends in age-standardised prevalence of hypertension, in Vietnam population aged 25–74 between 2001 and 2009, stratified by sex.

https://doi.org/10.1371/journal.pone.0042825.t005

Time Trends between Urban and Rural Area

Mean systolic and diastolic BP in adult population both increased significantly in both rural and urban areas (p<0.001) with significant time trends between areas. High blood pressure was detected and controlled significantly more with similar time trends in both urban and rural areas (p<0.001) although the prevalence of controlled hypertension in rural areas (p = 0.253) and the prevalence of aware hypertension in urban areas (p = 0.568) did not change considerably over time (year). Other anthropometric indexes such as mean weight, WC, BMI or WHR significantly increased over time in both areas although all the rates (time trend slope) in rural areas were higher than in urban areas (all p<0.05). Consequently, the prevalence of adiposity rose significantly overtime in both areas although the prevalence of generalized obesity in rural areas (p = 0.063) and the prevalence of abdominal fatness in urban areas (p = 0.153) did not change in a clear linear fashion with time (Table 6).

Aggregated age-standardised prevalence in each triennium shows more clearly the sex-specific time trends of hypertension, adiposity and smoking in urban and rural area (Figure 2). Although there was a breakdown in trend of adiposity prevalence likely resulting from BMI variations in the surveyed provinces, overall the prevalences of hypertension and adiposity were higher in urban than rural areas and both prevalences trended upwards over time (Figure 2). While the time trends in smoking prevalence were similar between urban and rural areas (p = 0.546), smoking increased significantly over time among women (starting from a very low baseline) but did not reduce significantly among men (Figure 2 and Table 6).

thumbnail
Table 6. Adjusted annual changes of CVD risk factors, stratified by residential areas.

https://doi.org/10.1371/journal.pone.0042825.t006

The age-standardised prevalence of hypertension increased 0.9–1.0% per year, equally in both sexes and in both residential areas (Table 4 and 6), both of which arose mostly due to a shifting to the right of the BP distribution in the general population rather than small increases among hypertensives on treatment. Among hypertensives, the prevalence of awareness, treatment and control of hypertension were higher in women than in men and higher in urban than rural areas and all tended to improve over time (Figure 3). Even though the prevalences of treatment and control of hypertension increased significantly, there was obviously a big gap between the prevalence of hypertension (representing the demand) and the prevalence of treatment or control (representing the capacity) (Table 3 & Figure 3).

thumbnail
Figure 3. Change over time in age-standardised prevalence of awareness, treatment and control of hypertension by urban and rural areas.

https://doi.org/10.1371/journal.pone.0042825.g003

U-shaped Relationship between BMI and Hypertension

Analysing the relationship between BP and BMI, our data also showed that an increase in BMI was generally associated with a significant increase either in mean BP (both SBP and DBP) or prevalence of hypertension in an observable, linear fashion as BMI levels increased in both sexes. However, in the underweight population group, the age-standardised prevalence of hypertension significantly increased inversely with BMI in both sexes, making a U-shaped relationship between BMI and hypertension (Figure 4).

thumbnail
Figure 4. U-shaped relationship between age-adjusted systolic blood pressure or hypertension and body mass index, stratified by sex and residential areas.

The solid line represents fractional-polynomial prediction plots and the shaded area the 95% uncertainty interval.

https://doi.org/10.1371/journal.pone.0042825.g004

Discussion

Time Trends in Major CVD Risk Factors in Vietnam

Our results, based on data from multiple cross-sectional surveys across the country, demonstrated that the time trends in hypertension and adiposity, both major CVD risk factors, steadily increased from 2001 to 2009, confirming the global trends of SBP and BMI in the Vietnamese community and reflecting the contribution to increasing CVD mortality recorded in Vietnam and the region. [2], [3], [4], [17] However, using the available individual participant-level data in each year and taking into account large variations between the surveyed provinces, the estimated increases of mean SBP in our study were 0.8 mmHg per year in women and 1.1 mmHg per year in men, which far exceeded the global estimations of SBP increases in Southeast Asian region that ranged from 1.0–2.7 mmHg per decade in women and 0.8–1.6 mmHg per decade in men [3]. With similar trends of lower magnitude, our results estimated the increases of average BMI were 0.7–1.6 kgm−2 per decade in women and men in Vietnam while the mean global BMI only increased by 0.4–0.5 kgm2 per decade in women and men together with the fact that BMI increased most slowly in the Southeast Asia region [2]. These trends in CVD risk factors at the population level, without rapid and widespread appropriate reactions, would lead to trends in CVD events in the same direction, threatening a projected CVD epidemic after a time lag of about 3–4 years [18].

The time trends also highlighted the bigger gaps between the potential burden of CVD risk factors and the real awareness of these and the capacity to solve them in the community (such as the hypertension issue demonstrated in Table 3, Table 6 & Figure 3), especially in rural areas, where there were unfavourable factors such as poorer access to medical services, inadequate supply of essential drugs, less health education and lower income [19]. So far, LMICs including Vietnam have a dearth of capacity to effectively deal with the existing and projected burden of cardiovascular diseases when health care systems are facing the challenges of reform to cope with the double burden of both relevant communicable and emerging non-communicable diseases [20]. On the other hand, in developed countries, CVD mortality/morbidity have reduced rapidly and substantially in recent decades, following both improvements in cardiac treatment and decreases in some but not all major CVD risk factors (such as blood pressure, cholesterol, and smoking) [2], [3], [4], [21]. These facts suggest that among the efforts to stem and reverse the tide of major CVD risk factors [17], endeavours to control blood pressure, smoking and cholesterol should be the prime concern. In low-resource settings like Vietnam, more resources should be allocated for cost-effective population-level interventions targeted at some similar major CVD risk factors (such as tobacco cessation or salt reduction to control BP) with multi-dimensional approaches (such as health education to promote healthy lifestyles in the population or policy changes to create favourable social environments) [2], [21], [22]. In addition, the existing healthcare system should be improved to promote efficient access to local healthcare services and be integrated with comprehensive surveillance systems so that any later interventions in local areas can be actively implemented, closely monitored and thoughtfully evaluated. Consequently any cost-benefit interventions can be tailored to tackle both communicable and non-communicable diseases at either community or individual level.

Limitations of this Study

Using multi cross-sectional surveys with different sampling strategies and diverse geographical areas (provinces) and times during each year, our findings must be considered within the context of these limitations. Similar to other population-based cross-sectional surveys on hypertension and CVD risk factor, hypertension status may have been overestimated when it was assessed only during a single visit in the survey [23] and the overweight status may have been underestimated depending on harvest cycles, especially in rural areas. However, this should have minimal effects on within-sample comparisons and it is an inherent problem of large epidemiological investigations. Using a standard protocol with strong quality control procedures for any physical measurement and questionnaire by nearly identical core teams from VNHI for all surveys [7], the potential protocol-related heterogeneity and data-driven bias should have been minimized; consequently, the data variations could be explained mostly by the effects of time (year) and geographical area (province). In order to manage the heteroskedasticity across all component surveys, multilevel mixed linear modelling with random slope at province level for age and surveyed year was used to adjust the random variations within geographic areas and specific time of each year as well as the random variations of relevant variables with age. Moreover, age-standardised prevalences or means were used to compensate for the potential bias due to differences of age structure in each dataset. The availability of real individual observation-level data almost in every year of analysis rather than using assumed equations allowed us to capture more variation to reflect population health and to estimate annual changes more accurately. However, due to specific aims or protocols of the component surveys and their budget constraints, some yearly data were missing in particular areas (such as data in urban areas for 2005 or in rural areas for 2008), which could possibly increase the uncertainty of our estimations. For the same reason, our study also could not explore other CVD risk factors, their interactions or adjust for more potential confounders which were not fully collected in all component studies.

U-shaped Interaction between Obesity and Hypertension

Although obesity and hypertension are considered as major independent CVD risk factors [24], [25], the interactions between these two factors and their impacts on CVD outcomes are complex [26], [27]. Our results revealed a U-shaped relationship between SBP or prevalence of hypertension and BMI in both sexes in the general population (in both urban and rural areas). In hypertensives, the CVD outcomes became excessive when patients were too thin or too fat, suggesting a similar curvilinear shape in relationship between CVD mortality and BMI [28], [29]. However, the hypothetical paradoxical link between BMI and CVD outcomes in the lean population group reflected in most cross-sectional surveys [30], including ours, does not compete against the demonstrated effectiveness of weight reductions in overweight populations or the impacts of significant weight gain on prevalence or incidence of CVD risk factors (such as hypertension, diabetes), and CVD events (including cardiac death) [26], [31], [32], [33], [34], [35], [36]. The data still support the need for weight reduction in the prevention and treatment of CVDs but further well-conducted studies with long-term follow-up are necessary to examine the effects of weight loss or weight gain on hypertension outcomes due to the complexity of the weight/hypertension relationship. Our data also highlighted the important and potential CVD burden of underweight status, which is often neglected in non-communicable disease areas but which is still widely prevalent in LMICs at early stages of epidemiologic transition where there is huge cumulative exposure to poverty and diseases, and nutritional deprivation [30].

Conclusions

Over 9 years from 2001 to 2009, blood pressure, weight and waist circumference increased steadily in the adult population of Vietnam in both sexes and in both urban and rural areas, resulting in a growing epidemic of hypertension and obesity while the awareness, treatment and control of hypertension lagged far behind the prevalence. Concurrently, the prevalence of smoking increased in women while hardly decreasing in men. In addition to the interaction between obesity and hypertension, this study also highlights the potential burden of hypertension in the neglected underweight subpopulation. Altogether, the sheer volume of prospective cardiovascular diseases accruing from modifiable CVD risk factors indicates that a higher priority for both swift and effective population-based cardiovascular health interventions is needed.

Acknowledgments

We would like to express our sincere thanks to all staff from Vietnam National Heart Institute and the corresponding partners involving in all surveys. Support from the Umeå Centre for Global Health Research, funded from FAS, the Swedish Council for Working Life and Social Research (grant no. 2006-1512) and through the SIDA Health Systems Research Programme is appreciated.

Author Contributions

Conceived and designed the experiments: QNN SW STP VLN LW RB PB. Performed the experiments: QNN SW LW PB. Analyzed the data: QNN SW LW PB. Contributed reagents/materials/analysis tools: QNN SW PB. Wrote the paper: QNN STP VLN SW LW RB PB.

References

  1. 1. Abegunde DO, Mathers CD, Adam T, Ortegon M, Strong K (2007) The burden and costs of chronic diseases in low-income and middle-income countries. Lancet 370: 1929–1938.
  2. 2. Finucane MM, Stevens GA, Cowan MJ, Danaei G, Lin JK, et al. (2011) National, regional, and global trends in body-mass index since 1980: systematic analysis of health examination surveys and epidemiological studies with 960 country-years and 9.1 million participants. Lancet 377: 557–567.
  3. 3. Danaei G, Finucane MM, Lin JK, Singh GM, Paciorek CJ, et al. (2011) National, regional, and global trends in systolic blood pressure since 1980: systematic analysis of health examination surveys and epidemiological studies with 786 country-years and 5.4 million participants. Lancet 377: 568–577.
  4. 4. Farzadfar F, Finucane MM, Danaei G, Pelizzari PM, Cowan MJ, et al. (2011) National, regional, and global trends in serum total cholesterol since 1980: systematic analysis of health examination surveys and epidemiological studies with 321 country-years and 3.0 million participants. Lancet 377: 578–586.
  5. 5. Minh HV, Byass P, Huong DL, Chuc NTK, Wall S (2007) Risk Factors for Chronic Disease Among Rural Vietnamese Adults and the Association of These Factors With Sociodemographic Variables: Findings From the WHO STEPS Survey in Rural Vietnam, 2005. Prev Chronic Dis 4: 2.
  6. 6. Walls HL, Peeters A, Son PT, Quang NN, Hoai NT, et al. (2009) Prevalence of underweight, overweight and obesity in urban Hanoi, Vietnam. Asia Pac J Clin Nutr 18: 234–239.
  7. 7. Son PT, Quang NN, Viet NL, Khai PG, Wall S, et al. (2012) Prevalence, awareness, treatment and control of hypertension in Vietnam - Results from a national survey. J Hum Hypertens 26: 268–80.
  8. 8. Nguyen QN, Pham ST, Do LD, Nguyen VL, Wall S, et al. (2012) Cardiovascular disease risk factor patterns and their implications for intervention strategies in Vietnam. Int J Hypertens 2012: 560397.
  9. 9. World Health Organization (2005) WHO STEPS surveillance manual: the WHO STEPwise approach to chronic disease risk factor surveillance. Geneva, Switzerland: World Health Organization: Noncommunicable Diseases and Mental Health.
  10. 10. Minh HV, Huong DL, Giang KB (2008) Self-reported chronic diseases and associated sociodemographic status and lifestyle risk factors among rural Vietnamese adults. Scand J Public Health 36: 629–634.
  11. 11. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, et al. (2003) Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 42: 1206–1252.
  12. 12. Department of Therapeutics, Ministry of Health (2010) Guidelines for Management of Hypertension. In: Department of Therapeutics MoH, Vietnam, editor. Vietnam..
  13. 13. Inoue S ZP, Caterson I, Chen C, Ikeda Y, Khalid AK, et al. (2000) TheAsia-Pacific Perspective: Redefining Obesity and Its Treatment: Health Communications Australia Pty Limited.
  14. 14. World Health Organization (WHO) (2004) Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 363: 157–163.
  15. 15. The General Statistics Office (2010) The 2009 Vietnam Population And Housing Census: Completeted Results. Hanoi, Vietnam: The General Statistics Office - Steering Commitee of Central Population and Housing Census.
  16. 16. GOPFP - General Office for Population and Family Planning MoH, Vietnam (2011) Population Distribution and Structure.
  17. 17. Anand SS, Yusuf S (2011) Stemming the global tsunami of cardiovascular disease. The Lancet 377: 529–532.
  18. 18. Tolonen H, Mahonen M, Asplund K, Rastenyte D, Kuulasmaa K, et al. (2002) Do trends in population levels of blood pressure and other cardiovascular risk factors explain trends in stroke event rates? Comparisons of 15 populations in 9 countries within the WHO MONICA Stroke Project. World Health Organization Monitoring of Trends and Determinants in Cardiovascular Disease. Stroke 33: 2367–2375.
  19. 19. Wang Z, Wu Y, Zhao L, Li Y, Yang J, et al. (2004) Trends in prevalence, awareness, treatment and control of hypertension in the middle-aged population of China, 1992–1998. Hypertens Res 27: 703–709.
  20. 20. Alwan A, MacLean DR, Riley LM, d’Espaignet ET, Mathers CD, et al. (2010) Monitoring and surveillance of chronic non-communicable diseases: progress and capacity in high-burden countries. Lancet 376: 1861–1868.
  21. 21. Ford ES, Capewell S (2011) Proportion of the Decline in Cardiovascular Mortality Disease due to Prevention Versus Treatment: Public Health Versus Clinical Care. Annu Rev Public Health 32: 5–22.
  22. 22. Anand SS, Yusuf S (2011) Stemming the global tsunami of cardiovascular disease. Lancet 377: 529–532.
  23. 23. Bovet P, Gervasoni JP, Ross AG, Mkamba M, Mtasiwa DM, et al. (2003) Assessing the prevalence of hypertension in populations: are we doing it right? J Hypertens 21: 509–517.
  24. 24. Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, et al. (2004) Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet 364: 937–952.
  25. 25. O’Donnell MJ, Xavier D, Liu L, Zhang H, Chin SL, et al. (2010) Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study. Lancet 376: 112–123.
  26. 26. Lavie CJ, Milani RV, Ventura HO (2009) Obesity and cardiovascular disease: risk factor, paradox, and impact of weight loss. J Am Coll Cardiol 53: 1925–1932.
  27. 27. Kotchen TA (2010) Obesity-related hypertension: epidemiology, pathophysiology, and clinical management. Am J Hypertens 23: 1170–1178.
  28. 28. Tuomilehto J (1991) Body mass index and prognosis in elderly hypertensive patients: a report from the European Working Party on High Blood Pressure in the Elderly. Am J Med 90: 34S–41S.
  29. 29. Stamler R, Ford CE, Stamler J (1991) Why do lean hypertensives have higher mortality rates than other hypertensives? Findings of the Hypertension Detection and Follow-up Program. Hypertension 17: 553–564.
  30. 30. Tesfaye F, Nawi NG, Van Minh H, Byass P, Berhane Y, et al. (2007) Association between body mass index and blood pressure across three populations in Africa and Asia. J Human Hypertens 21: 28–37.
  31. 31. Lavie CJ, Milani RV (1997) Effects of cardiac rehabilitation, exercise training, and weight reduction on exercise capacity, coronary risk factors, behavioral characteristics, and quality of life in obese coronary patients. Am J Cardiol 79: 397–401.
  32. 32. Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, et al. (2002) Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 346: 393–403.
  33. 33. Milani RV, Lavie CJ (2003) Prevalence and profile of metabolic syndrome in patients following acute coronary events and effects of therapeutic lifestyle change with cardiac rehabilitation. Am J Cardiol 92: 50–54.
  34. 34. Neter JE, Stam BE, Kok FJ, Grobbee DE, Geleijnse JM (2003) Influence of weight reduction on blood pressure: a meta-analysis of randomized controlled trials. Hypertension 42: 878–884.
  35. 35. Eilat-Adar S, Eldar M, Goldbourt U (2005) Association of Intentional Changes in Body Weight with Coronary Heart Disease Event Rates in Overweight Subjects Who Have an Additional Coronary Risk Factor. Am J Epidemiol 161: 352–358.
  36. 36. Aucott L, Poobalan A, Smith WC, Avenell A, Jung R, et al. (2005) Effects of weight loss in overweight/obese individuals and long-term hypertension outcomes: a systematic review. Hypertension 45: 1035–1041.