Understanding and Managing Cardiovascular Disease Risk Factors in Vietnam Integrating Clinical and Public Health Perspectives

Nguyen Ngoc Quang

Epidemiology and Global Health Department of Public Health and Clinical Medicine Umeå University, SE-90187 Umeå, Sweden.

Department of Cardiology Hanoi Medical University Hanoi, Vietnam.

Vietnam National Heart Institute Bachmai Hospital Hanoi, Vietnam.

Responsible publisher under Swedish law: the Dean of the Medical Faculty This work is protected by Swedish Copyright Legislation (Act 1960:729) © Nguyen Ngoc Quang ISBN: 978-91-7459-420-1 ISSN: 0346-6612Cover pictures: Photos taken by Dr Dinh Huynh Linh (HMU-VNHI) E-version available at http://umu.diva-portal.org/ Printed by: Print & Media, Umeå, Sweden 2012

“Superior doctors prevent the disease

Mediocre doctors treat the disease before it is evident

Inferior doctors treat the full blown disease”

Adapted from Huangdi Neijing, 2600 BC

To my family and my country

i

Prologue

My first involvement in epidemiology happened in 1994 when I received a diskette with Epi-Info from one of my old friends, who shared with me an interest in software. Acting on the advice of my older brother to apply it to my medical field, I tried to master this software, and use it to analyse data for my colleagues. I then learned more sophisticated procedures and got familiar with statistics. This experience also gave me more opportunities to work with clinical doctors and later contributed to my becoming a cardiologist. When training as a cardiology resident at the Vietnam National Heart Institute (VNHI), I joined in some surveys sponsored by World Health Organization (WHO), which were organized by Dr. Pham Thai Son, the secretary of WHO’s project at VNHI at that time (2001). Since then, we have worked together in many community projects and epidemiological surveys and intervention studies around the country. After becoming an official lecturer of Department of Cardiology of Hanoi Medical University (HMU) in 2002, I was recruited into the Health System Research Project (HSRP) at HMU and took the 3-week summer course in Umeå, June 2003. That was the first time I visited Umeå, met interesting teachers from the Division of Epidemiology and Public Health, and learned the basics in statistics, qualitative studies and public health.

Since 2001, I have been working as an intervention cardiologist at VNHI (an affiliated hospital of HMU). Returning to VNHI after nearly 18 months abroad on a fellowship in intervention cardiology (2004-2005), I focused more on the spectrum of vulnerable plaques in coronary arteries and percutaneous coronary interventions to fix these culprit lesions. These procedures always required careful consideration of the balance between latest clinical evidence and individual socioeconomic characteristics of any patients in order to achieve the best results in the short term as well as to protect them from later fatal complications. The more procedures I performed, the more I recognised the importance of preventive strategies to delay the onset of disease or to effectively control cardiovascular risk factors in order to avoid costly procedures, which can easily bankrupt any family without significantly changing mortality at advanced stages of cardiovascular disease. With this in mind, I tried to arrange time to participate in population studies at VNHI, which explored the burden of cardiovascular disease and potential solutions for the general population of Vietnam.

In 2007, I registered as a research student in Umeå University, a few months after Dr Son registered. Both of us started a new journey together in Sweden, spending 2-3 months a year studying new subjects in epidemiology and

ii

global health. Sharing a big project of prevention and management of hypertension in Vietnam under the same group of supervisors, we explored two different but complementary directions. Dr Son focused more on hypertension itself and the specific management of hypertension, in a vertical manner. I focused more on the general interactions of cardiovascular disease risk factors, including hypertension, in a more horizontal direction. Both approaches intersected at the point of hypertension and reflected our critical thinking pathway from epidemiological vision to interventional actions in the community. Working together very closely and becoming co-authors for almost all manuscripts, we used different datasets and study designs to scrutinize our approaches, leading to independent topics and themes for discussion not only in our own thesis but also in the papers in which Dr Son or I was separately the principal author. We expect readers will appreciate the divergence and the complementarity of our works, which should be considered as twinned visions and actions.

This thesis partially reflects the maturation in my perception from a purely clinical point-of-view to the comprehensive panorama of health science. Often lonely, being far away from my beloved, keeping away from my favourite challenging procedures in catheterization laboratory, being stuck sometimes in analysing data or writing manuscripts… if I say I am still very happy and never regret anything, I would be untruthful. However, being involved in community projects, contributing to social works, seeing the benefits and happiness people received from our completed actions, I feel fortunate to be on this right track. In addition to the inspiration from sharing and collaboration, I have gained a lot of invaluable experience and knowledge in methodology, anthropology and also philosophy. “Happiness is a journey, not a destination”. This thesis is a temporary stopover for me during my endless scientific journey to refine myself and bring benefits to other people.

iii

Abstract

Background: Vietnam, like other low-income countries, is facing an epidemic burden of cardiovascular disease risk factors (CVDRFs). The magnitude and directions of CVDRF progression are matters of uncertainty.

Objectives: To describe the epidemiological progression of CVDRFs and the preventive effects of community lifestyle interventions, with reference to the differences in progression of CVDRF patterns between men and women.

Methods: The study was conducted during 2001-2009 in nationally representative samples and in a local setting of rural areas of Ba-Vi district, Ha-Tay province. Both epidemiological and interventional approaches were applied: (i) a population-based cross-sectional survey of 2,130 people aged ≥25 years in Thai-Binh and Hanoi; (ii) an individual participant-level meta analysis of 23,563 people aged 24-74 years from multiple similar surveys in 9 provinces around Vietnam; (iii) a 17-month cohort study of 497 patients in a hypertension management programme; (iv) a quasi-experimental trial on community lifestyle promotion integrated with a hypertension management programme, evaluated by surveys of 4,645 people in both intervention and reference communes before and after a 3-year intervention.

Main findings: (i) in the general adult population ≥25 years, CVDRFs were common, often clustered within individuals, and increased with age; (ii) the Vietnamese population is facing a growing epidemic of CVDRFs, which are generally not well managed; (iii) it is possible to launch a community intervention in low-resource settings within the scope of a commune-based patient-targeted programme on hypertension management; (iv) community health intervention with comprehensive healthy lifestyle promotion improves blood pressure and some behavioural CVDRFs.

Conclusion: Alarming increases in CVDRFs in the general population need comprehensive multi-level prevention strategies, which combine both individual high-risk and population health approaches. The commune-based hypertension-centred management programmes integrated with community health promotion are the initial but essential steps towards comprehensive and effective management of CVDRFs and should be part of an integrated and co-ordinated national program on the prevention and control of chronic diseases in low-resource settings like Vietnam.

Keywords: cardiovascular disease risk factors, epidemiology, prevention, hypertension, Vietnam.

iv

Abbreviations

BMI Body Mass Index BP Blood Pressure CI Confidence Interval CHD Coronary Heart Disease CVD Cardiovascular Disease CVDRF Cardiovascular Disease Risk Factor DBP Diastolic Blood Pressure FilaBavi Epidemiological Field Laboratory in Ba-Vi District, Hanoi,

Vietnam HDL-C High-Density Lipoproteins Cholesterol HIV/AIDS Human Immunodeficiency Virus/Acquired Immune Deficiency

Syndrome LDL-C Low-Density Lipoproteins Cholesterol LMIC Low- and Middle-Income Country MI Myocardial Infarction NCD Non-Communicable Disease p p-value PAD Peripheral Artery Diseases PAR Population-Attributable Risk SBP Systolic Blood Pressure SE Standard Error STEP Stepwise Approach to Surveillance of Non-Communicable Risk

Factors USD United States Dollar VND Vietnamese Currency “Dong” (1USD ≈ 21,000 VND in 2011) VNHI Vietnam National Heart Institute WC Waist Circumference WHO World Health Organization WHR Waist-Hip Ratio

v

Original Papers

This thesis is based on the following papers, which will be referred to in the text by their Roman numerals.

I. Nguyen NQ, Pham TS, Do DL, Nguyen LV, Wall S, Weinehall L, Bonita R, Byass P. Cardiovascular disease risk factor patterns and their implications for intervention strategies in Vietnam. International Journal of Hypertension 2012, doi:10.1155/2012/ 560397

II. Nguyen NQ, Pham TS, Nguyen LV, Weinehall L, Bonita R, Byass P, Wall S. Time trends in blood pressure, body mass index and smoking in the Vietnamese population: a meta-analysis from multiple cross-sectional surveys. (Submitted)

III. Nguyen NQ, Pham TS, Nguyen LV, Wall S, Weinehall L, Bonita R, Byass P. Implementing a hypertension management programme in a rural area: local approaches & experiences from Ba-Vi district, Vietnam. BMC Public Health 2011, 11:325.

IV. Nguyen NQ, Pham TS, Nguyen LV, Wall S, Weinehall L, Bonita R, Byass P. Effectiveness of community-based comprehensive healthy lifestyle promotion on cardiovascular disease risk factors in a rural Vietnamese population: a quasi-experimental study. (Submitted)

vii

Table of Contents

Prologue i  Abstract iii  Abbreviations iv  Original Papers v  Table of Contents vii  Introduction 1  

Cardiovascular Diseases and Risk Factors 1  Challenges for Cardiovascular Disease Prevention 4  Conceptual Framework 6  

Objectives 8  Overall Objectives 8  Specific Objectives 8  

Materials and Methods 9  Study Settings and Designs 9  Main Variables 14  Statistical Methods 16  Ethical Considerations 16  

An Epidemic of CVDRF in Vietnam 17  Changes Over Time in Major CVDRFs in the General Population 17  Changes With Age and Clustering of CVDRFs in Adults 19  Gaps Between Burden And Management Capacity of CVDRFs 22  

Community-Based CVDRF Intervention 25  Tackle CVDRFs Separately or Comprehensively? 25  Commune Hypertension-Centred Management Programme 27  Perception Of Cardiovascular Diseases In Rural Areas 29  Community-Based Comprehensive Healthy Lifestyle Promotion 31  

Methodological Considerations 37  Study Limitations 37  Future Directions of the Study 38  Reflection on Concept Evolution 39  

Conclusion and Policy Implications 43  Acknowledgements 45  References 49  

Introduction

Cardiovascular Diseases and Risk Factors

Cardiovascular disease (CVD) is the term to embrace diseases of the heart, vascular diseases of the brain and diseases of blood vessels. CVDs are responsible for over 17.3 million deaths per year in 2008 and remain the leading causes of death and disabilities in the world [1, 2], especially in low and middle-income countries (LMICs) where 80% of the total CVD burden occurs [3, 4]. The percentage of premature deaths from CVDs in LMICs was 42%, ten times higher than that in high-income countries, but both included more than 3 million deaths before the age of 60, of which the majority can be prevented [1, 5].

Lifelong development of atherosclerosis is the main cause of fatal and non-fatal CVDs such as coronary heart disease (CHD), cerebrovascular disease and peripheral vascular or aortic diseases. Among factors that promote the process of atherosclerosis known as CVD risk factors (CVDRFs), a few traditional modifiable CVDRFs explain about 90% of the population-attributable risks (PAR) of both myocardial infarction (MI) and stroke [6, 7], the top two major adverse cardiac events and causes of CVD deaths, although the PARs are different for stroke (with greater impacts of hypertension and lesser impacts of smoking, diabetes and lipids) compared with the PARs for MI (Table 1).

Table 1. Population-attributable risks [%, (99%CI)] of major CVDRFs for MI and stroke in INTERHEART [6] and INTERSTROKE [7] studies

Major CVDRFs Acute MI Stroke

Hypertension 17.9 (15.7-20.4) 34.6 (30.4-39.1)* Current smoking 35.7 (32.5-39.1) 18.9 (15.3-23.1)*

Apo-lipoprotein B/A1 ratio 49.2 (43.8-54.5) 24.9 (15.7-37.1)

Diabetes 9.9 (8.5-11.5) 5.0 (2.6-9.5) Abdominal obesity (WHR) 20.1 (15.3-26.0) 26.5 (18.8-36.0)*

Lack of vegetables/fruits diet 13.7 (9.9-18.6) 18.8 (11.2-29.7)* No regular physical activity 12.2 (5.5-25.1) 28.5 (14.5-48.5)

Alcohol intake 6.7 (2.0-20.2) 3.8 (0.9-14.4)* Psychosocial stress 32.5 (25.1-40.8) 9.8 (4.8-19.4)

Cardiac causes - 6.7 (4.8-9.1)

All combined 90.4 (88.1-92.4) 88.1 (82.3-92.2) * significant for haemorrhagic stroke only

1

These major CVDRFs in Table 1 include hypertension, abnormal lipids, diabetes mellitus, tobacco use, obesity, diets with low intakes of fruits and vegetables, physical inactivity, excessive alcohol intake and psychosocial factors, which can divided into two groups: metabolic and behavioural CVDRFs, both accumulated mutually and linked in a life-long complex web with clinical endpoints, novel risk factors and environmental determinants of health (Figure 1) [8-11].

Figure 1. Pyramid of multi-level CVDRFs, adapted from Mozaffarian D [10]

When CHD and stroke were assessed as major contributors to global mortality in 2008, (7.3 and 6.2 million deaths, respectively [12]), there were considerable country-to-country variations in the burden of stroke versus CHD as well as variations in the relative contributions of CVDRFs over the world [13-19]. Lower national income, hypertension, heavy alcohol intake and advancing age were associated with higher stroke burden but smoking, dyslipidaemia, diabetes were linked to higher CHD burden, both highlighting the findings that CHD and stroke burden did not track uniformly and the differences in CVD burdens in each country required corresponding strategies/approaches in their own settings [19]. Along the various stages of epidemiological transition, inadequate health care resources in budget-constrained settings may directly impact on the stroke burden more than the CHD burden [20, 21]. Compared to other wealthier countries, in 7 Asian countries (except India) and other African and American LMICs, stroke is more prominent than CHD, probably due to a higher prevalence of hypertension and a lower prevalence of dyslipidaemia, both accompanying

Coronary Diseases

Stroke Cognitive Decline

Cardiac Arrhythmias 

Heart Failure

PAD

Death

Signs/Symptoms and Disability

Subclinical Vascular Diseases

Diabetes Hypertension Dyslipidemia Inflammation ThrombosisCoagulation

Adiposity Endothelium Dysfunction

OxidativeStress

Proinflammatory State

Stress Physical Inactivity

Excessive Alcohol Intake

SmokingUnhealthy Dietary Habits 

Education Health Services

Hazard/Toxic Exposures

Poverty/Social Class

Globalisation/Urbanisation

Environmental Determinants

Lifestyle Risk Factors

Established and Novel Risk Factors

Clinical Endpoints

Transport/Environment

MetabolicDysfunction

End‐organDamages

2

traditional habits of higher salt and lower fat intake as in the typical example of China [22]. However, recent westernization in the Asian region has increased fat and high-calorie food consumption, followed by faster peaking of the prevalence of obesity, dyslipidaemia and diabetes than elsewhere [23-28]: all these will make the CHD burden increase in addition to the existing stroke burden.

Progression of CVDs is a continuum, which requires lifelong prevention across all stages of disease development (Figure 2) [29]. Starting at the earliest stages of life (either genetically or in the perinatal period), and potentially interacting and accumulating through the life span [30], CVDRFs should be tackled by a sequence of preventive strategies including primordial, primary and secondary prevention, all of which should be combined flexibly and appropriately to reduce the CVD burden (morbidity and mortality), reducing recurrences and incidence and improving quality of life. Primordial prevention is focused on population improvement of socioeconomic and environmental health determinants to delay CVDRFs. Primary prevention is focused on CVDRF control using both population-based and clinic-based strategies while secondary prevention involves individual acute and chronic management or rehabilitation for CVDs.

Figure 2. Pyramid of CVD prevention, adapted from Gupta R [29]

Although high-risk people benefit more from preventive measures, CVD events occur more in low or medium-risk groups, which account for a higher proportion of the population [31]. Therefore, CVDRF prevention requires two complementary strategies: (i) a population (or public health) approach targeting the entire community to shift the CVDRF distribution in a

CVD risk factor controlSmoking cessation, 

Blood pressure control  Lipid, diabetes, obesity management

Acute and Chronic CVD Management, CVD Rehabilitation

Improving health promotion/educationHealthcare financing and health insurances

Policies for smoking, diet, physical activity modulationTackling social/environmental determinants of health

Environmental Determinants

Prenatal and Genetic Factors

Established Risk Factors

ChronicDiseases

Complicated End‐organ Diseases

Life‐course Cardiovascular Continuum 

Secondary Prevention

PrimaryPrevention

PrimordialPrevention

LifestyleRisk Factors

from childhood

3

favourable direction by population interventions, which are relatively low cost, but require extensive health education combined with national or policy-level measures, and (ii) a high-risk (or clinical) approach to identify and manage individuals at highest risk through healthcare systems, using cost-effective integrated methods, then reducing their susceptibility by multi-drug combinations [32-35].

CVD Burden in Vietnam

Vietnam has experienced a shift in mortality patterns from communicable to non-communicable diseases along with increasing longevity and an ageing population [36-39], resulting in a double financial burden from the established high burden of persistent infections (including the emerging burden of chronic infection such as HIV or tuberculosis) and the escalating burden of chronic diseases, especially CVDs. Similarly to other LMICs, stroke is more prominent than coronary heart disease in Vietnam, both accounting for nearly a quarter of total deaths [40]. A national population-based verbal autopsy survey showed that stroke-related mortality was five times as high as CHD-related deaths [39], which put stroke as the leading cause of death for both men and women, ranking first for men and second for women among the top causes of total disease burden in 2008 [40]. While CVDs were the largest contributor to the total burden of disease in Vietnam, the contribution of stroke and CHD was similar among men and women [40]. Stemming from environmental factors and unhealthy lifestyles, this emerging problem burdens families, impacts on the entire economy and social development, threatens the lives and health of millions of people, and compromises sustainable development efforts and poverty reduction initiatives [4, 27, 41]. However, the country had inadequate health system capacity (including monitoring and surveillance operations nationally) to effectively deal with the costly and complex clinical challenges of rapid rises in non-communicable diseases like CVDs and cancers [4, 42].

Challenges for Cardiovascular Disease Prevention

While CVD mortality in many richer countries has substantially reduced over the last few decades, CVD death rates in poorer countries are already higher and are rising [27, 43, 44]. CVD modelling from developed countries suggests that great reductions in mortality resulted from improving the CVDRF distribution in the population or enabling higher proportions of the population to receive evidence-based treatments [45]. The actual benefits would double if all eligible CVD patients could receive appropriate therapies at the right time [46]. However reductions in the prevalence of major CVDRFs, even modestly, had already accounted for more than twice as many

4

life-years gained as the treatments did [47, 48]. Favourable changes in CVDRFs such as smoking cessation and aggressive treatment of metabolic CVDRFs almost halved CVD mortality, separately from earlier diagnosis and life saving treatment [49, 50]. Three very cost-effective interventions in primary care are tobacco control, salt reduction and a multidrug strategy to treat people who are at high CVD risk [32, 51, 52]. Recently, more evidence emphasised the importance and effectiveness of interventions to tackle unhealthy diets, physical inactivity and obesity as well as policy-level solutions to create favourable healthy environments [52-55]. However, inadequate healthcare resources, insufficient access to healthcare services, physician inertia and “political failure” are main barriers to widely implementing these proven cost-effective interventions in the primary healthcare system [55-57].

Lessons and successful experiences from chronic HIV/AIDS epidemics can be usefully applied to the prevention and control of CVD [58]. Good surveillance systems for CVD and CVDRFs are crucial for measuring the problem’s magnitude and costs, identifying vulnerable groups, and evaluating intervention effects [58]. Low-cost detection approaches should be linked to cost-effective prevention and treatment that efficiently integrate behavioural and biomedical approaches (or public health and health care resources) [58, 59]. As in HIV/AIDS, besides the individual behavioural interventions and societal policies for healthier lifestyles, which mainly affected the most motivated adopters, it is critical to integrate low-cost biomedical interventions into CVD prevention as a complementary part of a holistic approach [58, 60]. To avoid costly therapies at more advanced stages, there will be overwhelming demands for wide availability of early detection and effective care and preventive services concurrently, which require prioritizing the development of delivery and financing models for integrated health systems and ensuring that supplies meet the demand [58].

In many LMICs, the existing healthcare systems confront the lack of resources, the low priority of CVD in systems geared towards acute and infectious care, and the lack of an integrated infrastructure involving communities, health services and local institutions [18, 61]. For prevention of CVD in LMICs, three urgent challenges must be met: (i) a redesigned primary care network with broad universal access; (ii) a surveillance system for key modifiable risk factors using repeated cross-sectional surveys, population-based cohorts or vital registration; (iii) a framework for social engagement, developing and implementing health policy to establish healthy living environments [18, 61]. Three relevant staircases of prevention are: (i) societal and policy levels (to address the upstream social determinants of health); (ii) health system level (comprehensive and integrated primary

5

healthcare); (iii) individual level (primary and secondary prevention or acute management) (Table 2) [62, 63].

Table 2. 3-level public health interventions, modified from Franco M [63]

CVDRFs Societal/policy level: built environments

Health system level

Individual level: behavioural change

Smoking - Tobacco tax/law - Programmes for children and adolescents

- Primary care programmes for smoking cessation

- Avoiding smoking uptake by young -Smoking cessation

Physical inactivity

- Recreational facilities - Promoting walking and cycling as means of transportation

- Primary care programmes for increasing physical activity

- Promotion of physical activity - Leisure versus utilitarian activity

Un-healthy dietary patterns

- Making vegetables and fruits more available, cheaper - Improving school cafeterias - Salt reduction and food labelling by the food industry

- Primary care unit or programmes for improving dietary quality

Nutritional education on TV and in schools

Metabolic CVDRFs

- See the above interventions - Health education to improve awareness, control of CVDRFs - National salt reduction

- Primary care unit (multi-disciplinary) to manage and control body weight, blood lipids and glucose levels, BP

- Nutritional education - Physical activity programmes - Develop home BP monitoring programmes

Conceptual Framework

The conceptual framework for this thesis is based on a dynamic, complex web of factors, in which adverse CVD endpoints are linked with health determinants and behavioural risk factors, and mediated through metabolic risk factors. CVDRFs can evolve intrinsically or be influenced by intentional health interventions. The two following themes are considered: (i) a epidemiological perspective on how CVDRFs progress over time and burden the general population and (ii) a prevention perspective on how CVDRFs can be influenced by community health interventions.

6

These themes are summarised in Figure 3 and covered by two chapters: the first chapter will describe the CVDRF patterns and progression in the general population while the following chapter will describe community-based interventions and the impacts of comprehensive lifestyle promotion on CVDRFs in a rural setting. Other chapters will discuss methodological issues, leading to conclusions and recommendations.

Figure 3. Conceptual framework for the study on the CVDRF progression

7

Objectives

Overall Objectives

The overall objective of this study is to investigate the natural progression of CVDRF patterns and impacts of community-targeted health interventions in the general adult population of Vietnam. The ultimate goal of the study is to contribute to the development and the optimization of future community-based health interventions for CVDRFs and consequently reduce their emerging epidemic burden in Vietnam and similar settings.

Specific Objectives

1. To estimate the magnitude of CVDRFs and to identify CVDRF patterns in the general adult population of Vietnam (I).

2. To explore the progression over time of CVDRFs in the general adult population of Vietnam (I, II)

3. To describe a model of community-targeted intervention incorporating comprehensive healthy lifestyle promotion campaigns integrated into a commune-based hypertensive-targeted programme on hypertension management by multidrug therapy (III).

4. To evaluate the effectiveness of the community-based comprehensive health promotion on the pattern of CVDRFs in the general adult population of a rural area in Vietnam (IV).

Main findings from four papers (I-IV), listed in page v, and additional analysis of data from the above-mentioned original studies are presented and summarized according to the themes in the conceptual framework (Figure 3).

8

Materials and Methods

Study Settings and Designs

Study settings

The studies in this thesis were conducted in two settings: (i) the national setting, based on eight provinces, which were representative of four geographical areas from northern and southern regions of Vietnam, where epidemiological surveys (study 1, 2) were done and (ii) a local setting incorporating two communes (Phu-Cuong and Phu-Phuong) from Ba-Vi district, Ha Tay province, where community-based intervention studies were done (study 3).

Figure 4. Study settings for the thesis

The setting for national level epidemiological surveys included 8 provinces randomly selected to be representative of city and three other geographical areas (lowland, highland and coastal) in the northern and southern part of Vietnam [64] (Figure 4). This setting was initiated by Vietnam National Heart Institute (VNHI) for the national survey on hypertension and its risk factors including diabetes (NESH), and subsequently used for the survey on heart failure and its risk factors in 4 northern provinces (HF-S) and the survey on diabetes and its risk factors in 2 northern provinces (DM-S). All these studies were designed and carried out by the VNHI core team.

1. Hanoi (city)2. Thai‐Binh (lowland)3. Thai‐Nguyen (highland)4. Nghe‐An (costal)5. Khanh‐Hoa (costal)6. Dac‐Lac (highland)7. Dong‐Thap (lowland)8. Ho‐Chi‐Minh City (city)

1. Ha‐Tay (lowland)

3

91

2

4

65

87

Phu‐Cuong

Phu‐Phuong

National settings for epidemiological surveys Local settings for community intervention studies

Map of Ba‐Vi district, Ha‐Tay provinceMap of 63 provinces of Vietnam* 

(* not include the Spratly and Paracel Islands, both belong to Vietnam) 

9

The local setting for community-based intervention studies was two lowland communes (Phu-Cuong and Phu-Phuong) from Ba-Vi district, Ha-Tay province. Ba-Vi is a typical rural area, 60 km west of Hanoi, with a population of about 260,000 (2010 data) covering an area of 410 km2, including lowland, highland and mountainous areas, experiencing a typical climate for northern Vietnam (predominantly a monsoon tropical climate with two wet and dry seasons). Agricultural production and livestock breeding were the main economic activities of the local people (81%), with major products being wet rice, cassava, corn, soybean, green beans and some fruit. Illiteracy was reportedly only for 0.4% of the adult population. The average income per person per year was around 48 USD in 1996 and increased to 750 USD in 2010. This district had already been purposively selected for a longitudinal population-based demographic surveillance FilaBavi (Field Laboratory for health system research) under a collaboration between the Health Strategy and Policy Institute, Hanoi Medical University and Swedish institutions (including the Division of International Health IHCAR, Karolinska Institute, Stockholm; the Division of Epidemiology and Global Health, Umeå University, Umeå and the Nordic School of Public Health, Gothenburg) [65].

Study designs and data collection

Figure 5. Overview of the study timeline for the all studies in the thesis

Cross-sectional study designs were used for all epidemiological surveys reported in this thesis to assess the major CVDRFs in the general population from 2001 to 2009, including some national-level surveys mentioned above (NESH, HF-S, DM-S) or the local survey on non-communicable disease risk factors in FilaBavi (NCDS) [66, 67] and some local screening surveys from the hypertension management project in rural communes (HMPS). All of these studies used the same protocol from the VNHI core team despite a few differences in sampling strategy (Table 1 paper II). Data from the DM-S

2001         ’02         ’03         ’04         ’05         ’06         ’07         ’08         ’09         2010          2001         ’02         ’03         ’04         ’05         ’06         ’07         ’08         ’09         2010          

CVDRF Clustering in adult population(Cross‐sectional survey ‘09)

CVDRF Trend over time in general adult population(Individual participant‐level meta analysis from multiple cross‐sectional surveys ‘01‐’09)

CVDRF Intervention on community (Quasi‐experimental study, evaluated by cross‐sectional surveys ’06‐‘09)

CVDRF Perception in rural area(Focus group discussions ‘07)

1

2

3

National setting

Local setting

10

survey was used to evaluate CVDRF burden in the general adult population (study 1). An individual participant-level collated dataset from all of these yearly cross-sectional surveys (including DM-S) was used to identify the time trends of CVDRFs in the general adult population (study 2). Figure 5 summarizes the timeline of all studies in the thesis.

Figure 6. Design of the quasi-experimental intervention study

Study 3 used a quasi-experimental design to evaluate the impact of community-based health promotion on CVDRFs in the general population in a rural setting (Figure 6). Two communes were purposively selected from Ba-Vi district: Phu-Phuong for reference and Phu-Cuong for intervention. A hypertension management programme nested with comprehensive health promotion campaigns was implemented simultaneously in Phu-Cuong while no programme was launched in Phu-Phuong, which retained local conventional healthcare services. The health promotion campaigns aimed at smoking cessation, reduction of alcohol consumption, encouragement of healthier less-salty diets and physical activity, all of which were targeted at all local people. The hypertension management programme focused on monthly check-ups, drug delivery and individual lifestyle consultations, all of which were targeted at hypertensives only. Repeated cross-sectional surveys (the baseline survey in 2006 and one after the 3-year intervention in 2009) were used to assess the changes in CVDRFs in the local communities to find out the impact of the health promotion campaigns (paper IV). The impact of the hypertension management programme on blood pressure (BP) control was evaluated separately by another cohort study following the same

Quasi‐experimental intervention study 

Intervention Commune

Reference Commune

Cross‐sectional survey at baseline

Comprehensive health promotion for all in local community

Cross‐sectional survey after interventions

Cross‐sectional survey at baseline

Cross‐sectional survey after interventions

Local management programme for 

hypertensive patients

Cohort study on hypertensive patients 

Refer to conventional health care services for hypertensives

Hypertension detected

11

participants with high BP detected from two baseline surveys (Figure 6). All steps to implement the management programme and data from the early follow-up of hypertensives in the intervention commune were described and analysed in paper III.

In all cross-sectional surveys, random sampling strategies were applied to recruit representative samples for the studied populations. The same multi stage stratified sampling methods were used in national surveys (NESH, HF-S and DM-S), in which 4 districts per province and 3 communes per district were randomly chosen. Then 110 persons were randomly selected from a list of all inhabitants in each randomized commune (as primary sampling units). [64]. In NCDS, sex and age group stratified random sampling was used with 250 persons per 10-year age group per sex [66, 67]. In HMPS and the all pre and post intervention surveys of study 3, simple random selection was used to select candidates from the list of all adult inhabitants in the local area, which possibly excluded a certain number of males who migrated out for economic reasons. Being initiated in NESH, the assumed sample size of 1,200 in each surveyed area was large enough to identify prevalence of both hypertension and diabetes in the general adult population accounting for a theoretical design effect of 2 (for multi-stage sampling). This sample size was followed by all surveys except NCDS with specific sampling strategy and bigger sample size (Table 1, paper II).

Data were collected at local health stations in each surveyed commune by well-trained surveyors. The STEPwise-based questionnaire [68, 69] was used to report behavioural CVDRFs, demographic background and medical history. An identical protocol [64] was applied to get anthropometric and BP measurements using standardized devices. In some surveys (DM-S or NESH [for one fifth of sample in cities] [64]), fasting participants were asked to have an oral glucose tolerance test with 75g glucose and a blood sample for lipid profiles (paper I). The VNHI team supervised all steps to guarantee the study quality and maintain an identical workflow through all surveys.

In addition to the quantitative surveys, a qualitative study was also carried out in Ba-Vi district to explore how people formulated their conceptions about CVDRFs, using thematized focus group discussions and content analysis approaches. Totally eight convenience-sampled group discussions were conducted (5-6 adults per group, separately for each sex, age group [25-64 years vs. ≥ 64 years] and living conditions [rural vs. semi-urban area]).

An overview of the four papers, in terms of study objectives, data sources, study sample, main variables and outcomes is given in Table 3. Paper I and II mainly addressed the epidemiological themes such as within-individual

12

clustering patterns and the time trends of CVDRFs in the general population. Paper III and IV addressed the prevention theme through a commune-based programme on hypertension management, focusing on hypertension and behavioural CVDRFs in rural settings. Paper III was mainly descriptive highlighting the implementation of the programme in a low-resource setting while paper IV was more analytical and addressed the effects of community health promotion interventions on blood pressure and behavioural CVDRFs.

Table 3. Overview of the designs of the four papers included in the thesis

Paper

Specific objective

Study designs and data collection methods

Sample size Main variables or outcomes

I

Pattern of CVDRFs in population (2009)

Cross-sectional surveys on diabetes and its risk factors

2,130 adults (aged ≥25) from DM-S (2 provinces)

Clustering of behavioural and metabolic CVDRFs for each sex or area

II

Time trend of major CVDRFs in population from 2001 to 2009

Individual participant-level meta analysis from multiple cross-sectional surveys

23,563 adults (aged 25-74) from NESH, HF-S, DM-S, NCDS, HPMS (9 provinces)

Blood pressure, smoking and body size: progression over time for each sex or area

III

Process to setup a BP manage-ment model in a rural area from 2006 to 2008

17-month cohort study on patients of the model of hypertension management at Phu-Cuong commune, Ba-Vi

497 adults (age ≥25) patients treated by the model

Factors influenced the adherence to the model of hypertension management at rural commune

IV

Impact of health promotion on CVDRFs in rural settings from 2006 to 2009

Quasi-experimental community health intervention in one reference and one intervention commune in Ba-Vi, evaluated by repeated cross-sectional surveys

4,645 adults (age ≥25) in four surveys pre and post in reference and intervention areas

Changes in behavioural CVD RFs and blood pressure: after 3-year intervention compared between reference and intervention areas

13

Main Variables

Occupational status was classified into 3 groups: government staff, manual workers (farmers, building workers, etc.) and other (housewives, jobless, disabled). Educational level, which was determined by years of schooling, was classified into 2 groups: incomplete secondary schooling (≤9 years of education) and higher (>9 years of education including high school or higher). Residential area consisted of urban and rural, which was identified on an administrative basis for each commune within each province.

Table 4. Main definition of behavioural CVDRFs in the thesis

CVDRFs Definition

Current smoking

Smoking tobacco products such as cigarettes, cigars or pipes over the previous month [69]

Heavy alcohol consumption

Having more than 2 standard units of drink per day (in women) or more than 3 per day (in men) [69]

Low vegetable or fruit consumption

Having on average less than five standard servings of fruit and/or vegetables per day [69]

Salty diet Having self-reported preference for daily foods that contained more salt than the similar foods chosen by other adult members in the family or people around them

Unhealthy diet Having either salty diet or diet with low fruit and vegetable consumption

Physical inactivity

Having total physical activity less than 3,000 metabolic equivalents (METs) minutes per week, which were estimated based on details of duration and type of all self-reported physical activities in a typical week [69]

Experiencing psychosocial stress

Having more than 2 moderate stressors, which were semi-quantified by some simple questions to evaluate any stress at work or at home, financial stress, major life events (marital separation/divorce, loss of crop/job, major intra-family conflict, death, illness of a close family member/spouse) or any major stress in the past year at different levels (none, mild, moderate and severe) [6].

Behavioural CVDRFs included current smoking, heavy alcohol consumption, unhealthy diet, physical inactivity and experiencing stress (Table 4). Metabolic CVDRFs included hypertension, abnormal lipids, obesity, and diabetes mellitus (Table 5). People who had ≥ 2/4 metabolic CVDRFs, ≥ 2/5 behavioural CVDRFs or total ≥ 4/9 of all metabolic and behavioural CVDRFs

14

were considered to have individual clusters of respective CVDRFs. Ideal Cardiovascular Health was defined as the simultaneous presence of 4 favourable behavioural CVDRFs (no smoking, ideal BMI <23 kgm-2, regular physical activity, healthy diet) and 3 favourable metabolic CVDRFs (untreated total cholesterol <5.17 mmol/l, untreated blood pressure <120/80 mmHg, fasting blood glucose <5.5 mmol/l) without any clinical CVD (CHD, stroke, heart failure, etc.) [70].

Table 5. Main definition of metabolic CVDRFs in the thesis

CVDRFs Definition Overweight Having body mass index (BMI) ≥23, which was calculated

as weight (kg) divided by height squared (m2) [71] Obesity Having BMI ≥25 or having central obesity (BMI ≥23 with

waist circumference ≥90 cm in men or ≥80 cm in women), as was specified for South-Asian populations [71]

Hypertension Having an average systolic BP (SBP) ≥ 140 mmHg or average diastolic BP (DBP) ≥ 90 mmHg or having self-reported current antihypertensive treatment [72-74]

Controlled hypertension

Having pharmacological treatment resulting in an average SBP < 140 mmHg and DBP < 90mmHg [64]

Dyslipidaemia Having self-reported current cholesterol-lowering treatment or having at least one following criteria: total cholesterol ≥ 5.17 mmol/l; HDL-C< 1.03 mmol/l; LDL-C ≥ 3.36 mmol/l and triglyceride ≥ 1.7 mmol/l [75]

Controlled dyslipidaemia

Having pharmacological treatment resulting in LDL-cholesterol < 3.36 mmol/l (for general population) [75]

Diabetes Having fasting plasma glucose ≥ 7.0 mmol/l or plasma glucose 2h-post oral glucose tolerance test ≥ 11.1 mmol/l or having self-reported current diabetes treatment [76-78]

Controlled diabetes

Having pharmacological treatment resulting in HbA1C < 7.0% and fasting plasma glucose < 7.0 mmol/l [78]

The overall 10-year risk of developing coronary heart disease (MI, coronary death) and other important adverse cardiac events (stroke, heart failure) in people aged 30-74 years without baseline CVDs was estimated by the Framingham general cardiovascular risk score using the following variables: age, sex, current smoking, treated and untreated SBP, diabetes, and lipid profile (total cholesterol, HDL-C) or BMI (replacing lipids if blood test was unavailable). People with overall CVD 10-year risk ≥20% were classified as having a high overall CVD risk [79].

15

Statistical Methods

The mean levels or prevalence of each CVDRF were presented as means ± standard errors (SE) or proportions with 95% confidence intervals (CI), standardized for age structure and sex. In the study of CVDRF time trends, age distribution for each surveyed province from the Vietnam Population and Housing Census in 2009 [80] was used to weight and age-standardise for the prevalence or mean level of each CVDRF so that the adjusted values for each year could be comparable over time.

Univariate ANOVA tests were used to identify significant differences among these prevalence or mean levels of CVDRFs between comparable groups. Impacts of explanatory variables on the outcome were considered using multivariable linear or logistic regression models. Multilevel mixed-effect models were used to manage the potential heteroskedasticity and the cluster effect among surveys. Interactions between some explanatory variables were added to the model to identify the specific difference in time trends (study 2) or the differences-in-differences (i.e. impact) of intervention (study 3).

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 Considerations

The protocols of each survey were approved by Scientific Ethical Committee for Biomedical Research at the Vietnam National Heart Institute (VNHI) and at the corresponding involved partners. The protocol for the community-based intervention study was approved by Scientific Ethical Committees for Biomedical Research at the Ministry of Health, at VNHI and also by the People’s Committee of Ba-Vi district, Ha-Tay province.

All human subjects in any surveys were asked for their written consent before the collection of data or their acceptance for giving blood sample. Any participants with hypertension or other disorders detected during the surveys were referred to appropriate facilities for further investigation and treatment. In the quasi-experimental intervention study, any patients detected with hypertension in the reference commune were referred for conventional care and treatment, while all detected patients in the intervention commune were invited to join the newly initiated local programme on hypertension management to get monthly check-ups and drug provision during the study period. All participants had complete rights to withdraw from any study at any time without any threat or disadvantage.

16

An Epidemic of CVDRF in Vietnam

Changes Over Time in Major CVDRFs in the General Population

Results from the 9-year meta-analysis show increasing trends over time in BP (both SBP and DBP) and body size (weight and waist circumference) in the population of Vietnam aged 25-74 years (Table 6, Figure 7). On average, the SBP and DBP increased 0.9 and 0.4mmHg per year, similarly between men and women and between rural and urban areas (paper II).

Table 6. Estimated annual changes (95% CI) of blood pressure and body size in the general population, by sex and residential area (study 2)

Annual changes during 2001-2009 Rural area Urban area Women Men Women Men

Mean systolic BP (mmHg) 0.7 0.5-0.9

1.0 0.8-1.3

0.9 0.6-1.1

1.1 0.8-1.5

Mean diastolic BP (mmHg) 0.4 0.3-0.5

0.6 0.4-0.7

0.2 0.1-0.3

0.3 0.1-0.5

Men weight (kg) 0.3 0.3-0.4

0.6 0.5-0.7

0.2 0.1-0.3

0.4 0.3-0.6

Mean waist circumference (cm) 0.6 0.5-0.7

0.6 0.5-0.7

0.3 0.2-0.4

0.4 0.2-0.5

Figure 7a. Time trends of systolic blood pressure in the general population

0.0

1.0

2.0

30

.01

.02

.03

100 150 200 250 100 150 200 250

Female, Rural Area Female, Urban Area

Male, Rural Area Male, Urban Area

2001−2003 2004−2006 2006−2009

Den

sity

of S

ysto

lic B

lood

Pre

ssur

e in

pop

ulat

ion

Systolic Blood Pressure (mmHg)

17

Figure 7b. Time trends of weight in the general population

Figure 7c. Time trends of waist circumference in the general population

The average weight and waist circumference increased 0.4 kg and 0.5 cm per year respectively, with similar trends in men and women and faster in rural than in urban areas. Consequently, the prevalence of hypertension and

0.0

25.0

50

.025

.05

20 40 60 80 100 20 40 60 80 100

Female, Rural Area Female, Urban Area

Male, Rural Area Male, Urban Area

2001−2003 2004−2006 2006−2009

Den

sity

of W

eigh

t in

popu

latio

n

Weight (kg)

0.0

25.0

50

.025

.05

40 60 80 100 120 40 60 80 100 120

Female, Rural Area Female, Urban Area

Male, Rural Area Male, Urban Area

2001−2003 2004−2006 2006−2009

Den

sity

of W

aist

Circ

umfe

renc

e in

pop

ulat

ion

Waist Circumference (cm)

18

obesity in adults annually 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 (paper II). While smoking did not reduce over time in men, the progression of average blood pressure and weight in the population resulted in an increase of overall CVD 10-year risk of 0.2% in women and 0.4% in men. This rate was higher in men than in women and higher in urban than in rural areas (Figure 8). In a span of less than 10 years (2001-2009), our data showed that the increase in SBP and BMI in the Vietnamese general population far exceeded the global estimation of changes in these CVDRFs for Southeast Asian region [81-83]. Combined with smoking, which was unchanged during this period and other CVDRFs, which often clustered at individual level (paper I), without prompt reactions these trends would rapidly lead to a catastrophic projected CVD epidemic after a short time lag and threaten national wealth [84, 85].

Figure 8. Time trend of overall CVD 10-year risk, estimated by BMI-based Framingham general cardiovascular risk score in the general population

Changes With Age and Clustering of CVDRFs in Adults

Results from the most recent general population survey (paper I) showed that each CVDRF was more common in men than in women except for physical inactivity and experiencing stress (Table 7). CVDRFs were often individually clustered, increasing with age by sex-different patterns (Figure 9).

0.0

2.0

4.0

60

.02

.04

.06

0 50 100 0 50 100

Female, Rural Area Female, Urban Area

Male, Rural Area Male, Urban Area

2001−2003 2004−2006 2006−2009

Den

sity

of o

vera

ll C

VD ri

sk in

pop

ulat

ion

Estimated overall CVD 10−year risk (%)

19

Table 7. Prevalence of CVDRFs (± SE) in the general population in 2009, stratified by sex and residential area (study 1)

CVDRFs Rural area Urban area

Women Men Women Men Hypertension 17.3 ± 1.5 26.2 ± 2.3 30.8 ± 1.8 35.5 ±2.8 Diabetes 3.8 ± 0.8 7.4 ± 1.4 8.0 ± 1.1 8.5 ± 1.6 Obesity 6.3 ± 0.9 4.3 ± 1.0 25.8 ± 1.8 23.3 ± 2.7 Dyslipidaemia 44.6 ± 2.2 55.0 ± 2.8 58.2 ± 2.1 69.7 ±3.0 ≥ 2/4 metabolic CVDRFs 16.1 ± 1.5 21.9 ± 2.2 37.0 ± 2.0 41.0 ±3.0

Current smoking 2.6 ± 0.7 61.7 ± 2.6 4.6 ± 0.8 56.3 ± 3.1 Excessive alcohol intake 1.0 ± 0.4 26.6 ± 2.5 0.9 ± 0.4 28.5 ±2.8 Unhealthy diet 66.8 ± 2.0 66.5 ± 2.6 40.9 ± 2.1 53.2 ± 3.1 Physical inactivity 3.6 ± 0.7 5.7 ± 1.2 31.1 ± 2.0 32.9 ±2.9 Having stress 31.3 ± 2.0 26.9 ± 2.4 24.0 ± 1.9 20.6 ±2.6 ≥ 2/5 behaviour CVDRFs 29.1 ± 2.0 62.4 ± 2.6 25.4 ± 1.9 61.6 ±3.0

≥4/9 major CVDRFs 6.1±1.0 25.6±2.3 18.2±1.6 42.1±3.1

Figure 9. Changes of number of clustered CVDRFs with age in the general population in 2009, stratified by sex and residential area (study 1)

01

23

Female Male Female Male25−3435−4445−5455−6465−7475++

25−3435−4445−5455−6465−7475++

25−3435−4445−5455−6465−7475++

25−3435−4445−5455−6465−7475++

Rural Urban

Metabolic CVD risk factors Behavioural CVD risk factors

Aver

age

num

ber o

f CVD

risk

fact

ors

20

Among behavioural CVDRFs, having stress reduced while unhealthy diet or physical inactivity remained unchanged with age in both sexes. Smoking and excessive alcohol intake was traditionally male-dominated but also reduced with age. The number of behavioural CVDRFs in men reduced with age but was always higher than the number in women (Figure 9) (Table 2, paper I).

All four metabolic CVDRFs increased with age in women while in men only diabetes and hypertension increased. The age of 55 can be considered as an inflection point where the increasing trend of clustered metabolic CVDRF in women crossed and surpassed the trend in men, especially in urban areas (Table 2 & 4, paper I). This also highlighted a neglected subgroup of elderly women with alarming increases in clustered metabolic CVDRFs (Figure 9). Having more CVDRFs resulted in a higher overall CVD 10-year risk, regardless of sex or residential area (Figure 10).

Figure 10. Proportional relationship between number of CVDRFs and overall CVD 10-year risk, estimated by lipid profile-based Framingham general cardiovascular risk score

Estimation from study 1 showed that clustered major CVDRFs were found in 13.0% of women and 34.4% of men of the general population of Vietnam. These findings raised more concerns about cardiovascular population health, with the total proportion of ideal cardiovascular health in population only consisting of 7.3%, being especially low in men where smoking or heavy alcohol consumption was prevalent (1.7% in men versus 10.2% in women (Table 8). Our analysis in paper II also highlighted that urban living

010

2030

Aver

age

Ove

rall

CVD

10−

year

risk

0 RF 1 RF 2 RFs 3 RFs 4 RFs 5 RFs 6−7 RFsNumber of CVD risk factors

Urban Men Rural Men Urban Women Rural Women

21

conditions were significantly related to a higher prevalence of metabolic disorders after adjusting for age and other social factors [86-88].

Table 8. Prevalence of clustered CVDRFs in the general population in 2009, stratified by sex and residential areas (summarized from study 1)

Area Sex Major CVDRFs Ideal Cardiovascular Health

Rural Female 6.1 ± 1.0 12.0 ± 1.5

Male 25.6 ± 2.4 2.3 ± 0.8

Urban Female 18.2 ± 1.6 8.8 ± 1.3

Male 42.1 ± 3.1 1.0 ± 0.6

Total Female 13.0 ± 1.0 10.2 ± 1.0

Male 34.4 ± 2.0 1.7 ± 0.6

Gaps Between Burden And Management Capacity of CVDRFs

Hypertension, diabetes and dyslipidaemia are routinely treated by multiple drug therapy and lifestyle consultation in the existing Vietnamese health care system but not in primary health care. Our meta-analysis showed a clear trend in improvement of awareness, treatment and control of hypertension through the years 2001-2009 but highlighted the gaps between the potential burden of CVDRFs and the real awareness or capacity to solve these in the community, which were even bigger in rural areas (Table 3, paper II).

Table 9. Prevalence (± SE) of awareness, treatment and control among people who have hypertension, diabetes, dyslipidaemia or at least ≥1/3 mentioned metabolic disorders in the general population in 2009 (study 1)

Residential area

Hyper-tension

Diabetes Dys-lipidemia

Metabolic disorders

Awareness Rural 30.5 ± 3.0 22.5 ± 5.8 3.4 ± 0.8 14.0 ± 1.4 Urban 36.6 ± 2.6 44.5 ± 5.4 13.8 ± 1.4 26.8 ± 1.7

Total 34.6 ± 2.0 37.3 ± 4.2 9.8 ± 0.9 21.8 ± 1.2 Treatment Rural 19.6 ± 2.5 22.5 ± 5.8 1.2 ± 0.5 9.3 ± 1.1

Urban 30.2 ± 2.4 43.5 ± 5.4 3.0 ± 0.6 18.6 ± 1.4 Total 26.7 ± 1.8 36.6 ± 4.2 2.3 ± 0.4 15.0 ± 1.0

Control Rural 4.1 ± 1.3 13.2 ± 4.7 0.9 ± 0.4 3.2 ± 0.7

Urban 6.4 ± 1.3 15.9 ± 3.7 2.1 ± 0.5 6.2 ± 0.8 Total 5.7 ± 1.0 15.0 ± 2.9 1.6 ± 0.3 5.0 ± 0.6

In 2009, although being the most common among 3 metabolic disorders, the prevalence of awareness, treatment for dyslipidaemia were lowest (Table 9).

22

Only a minority of people diagnosed with hypertension, diabetes or dyslipidaemia achieved targeted outcomes (5.7%, 15.0% or 1.6% respectively (Table 9). This situation became even worse when diagnosis lagged far behind prevalence although the management of these disorders likely improved over time (paper II, I). Consequently, very little proportion of the adult population with high overall CVD risk (1.9%) had received optimal treatment for all metabolic disorders. The proportion of being treated at least 1/3 mentioned metabolic disorders in 2009 only consisted of 9.6% in the general population (5.3% in rural versus 13.1% in urban area), far behind the treatment demand for three main metabolic disorders (Figure 11).

Figure 11. Gaps between having and being treated for at least one among three disorders including hypertension, diabetes and dyslipidaemia in 2009

While the modifiable CVDRFs increased in the region due to globalisation [89], unplanned urbanisation, marketing of unhealthy food, and inadequacies in public health policies [18], access to medical care was a common barrier for people needing effective CVDRF management, especially in rural low-resource settings where local cardiac care capacity was insufficient. Self-treatment for common mild illness [90], no treatment for asymptomatic metabolic disorders and higher hospitals for advanced stages of disease (Table 10) were the main options for seeking healthcare

20%

40%

60%

80%

0%Hav

ing

at le

ast 1

/3 m

etab

olic

dis

orde

rs

Rural UrbanFemale Male Female Male

Treated at least one disorder Untreated all disorders

23

rather than local commune health care station even though costly medical care services can easily lead to a “poverty trap” [91].

Table 10. Healthcare access for treating at least one metabolic disorder including hypertension, diabetes and dyslipidaemia in 2009 (study 1)

Area Sex Self-

treatment

Commune Health Station

Private Clinics or Pharmacy

District or Higher

Hospitals

Rural Female 4.7 ± 3.6 10.7 ± 4.3 0.5 ± 0.5 84.2 ± 5.4

Male 0.0 22.9 ± 7.4 2.0 ± 2.0 75.1 ± 7.6

Urban Female 2.2 ± 1.4 11.2 ± 3.4 9.9 ± 2.6 76.7 ± 4.1

Male 1.0 ± 1.0 4.9 ± 2.2 8.5 ± 5.2 85.6 ± 5.5

Total Female 2.7 ± 1.3 11.1 ± 2.8 8.1 ± 2.1 78.2 ± 3.5

Male 0.7 ± 0.7 11.2 ± 3.1 6.2 ± 3.5 81.9 ± 4.4

Among available healthcare service options for treatment of the three mentioned metabolic disorders, most people preferred district or higher-level hospitals (79.3%, higher in urban than in rural areas) compared to commune health station (11.2%), private clinics or pharmacy (7.5%) and self-treatment (2.0%), in which the last three options were similar between urban and rural areas (Table 10). It reflected the existing condition of insufficient resources on demands for CVDRF management at the commune level, even though only 15% of people having these disorders were treated at all.

In summary, the general adult population in Vietnam are facing an alarming increase in the prevalence of CVDRFs. CVDRFs are often clustered within individuals, increase with age and result in high demands for management, while the capacity to handle such large numbers of at-risk people is not available in a resource constrained country like Vietnam.

24

Community-Based CVDRF Intervention

Tackle CVDRFs Separately or Comprehensively?

People with high risk of CVD in the general population can be screened by: (i) a single prominent proximal CVDRF linking directly to a major adverse CVD event (e.g. hypertension for stroke); or (ii) a holistic score which combines major CVDRFs to estimate the overall CVD risk in certain period of time (e.g. in lifetime, in 10 years or shorter) [73, 92-95]. Then the decision on how to manage and follow up on a particular person will depend on the magnitude of this single CVDRF or the overall CVD risk. The holistic approach is preferred as CVDRFs are usually clustered within individuals and interact with each other, so that moderate reductions in several risk factors will be more effective than major reductions in a single factor [96-102].

Using the CVDRF pattern in the general population of Vietnam in 2009, our extrapolation suggests that targeting a single CVDRF (such as hypertension) without considering other modifiable CVDRFs (such as unhealthy diet or smoking) is not an efficient approach for achieving a high health impact for the community. The high-risk approach using the individual Framingham general cardiovascular risk score [79] would benefit the entire population more than only approaching a group of hypertensives. Our results also confirmed that a combination of population-wide and high-risk approaches is necessary, with the first priority being tobacco use reduction and the second being salt reduction in food (Figure 12).

Figure 12. Estimation of potential benefits of intervention strategies for the population of Vietnam (extract from Figure 3, paper II)

25

Moreover, using a single CVDRF approach to prevent adverse CVD events is sometimes more expensive than using the overall risk approach, which enables more accurate targeting of high-risk groups in the population [95]. Under the single risk approach, a person having high overall CVD risk but mild to moderate elevation of multiple CVDRFs could be easily ignored while many people having low overall CVD risk but high elevation of a single factor could be treated even if the risk-benefit ratio is low and the treatment is costly [95]. The implementation of overall risk is more cost-effective as the threshold of high CVD risk is raised regardless of drug efficacy and treatment cost [98, 103-108], which has important policy implications for low-resource settings where resource constraints are a major barrier for prevention [95].

There are several risk scores to estimate overall CVD risks at different certain periods of time, mostly including some basic parameters such as age, sex, blood pressure level, smoking status, diabetes and lipid profiles. Some recent risk scores integrate novel risk factors such as C-reactive protein, haemoglobin A1c, social deprivation, family history of premature CHD and so on [79, 94]. Actually, physicians in daily practice have to face the difficult clinical challenge of decision making based on many seemingly discordant risk estimations, which always have some uncertainty especially when the risk score is applied to different target populations compared to the population from which the original score was derived. While still being imperfect and inappropriate for promoting complacency [94], overall CVD risk assessment in asymptomatic people is increasingly advocated [75, 109-113], in which the overall 10-year or lifetime CVD risk scores are recommended for exploring individual future risk, to identify who is benefiting from proven preventive measures, and to engage more patients at an earlier stage by highlighting the necessity of early and lifelong interventions on their CVDRF profiles [94, 114].

Among these algorithms, the effective risk scores, which does not require expensive blood tests, are preferred in limited resource settings [79, 115, 116]. Our calculations in paper I showed the concordance between the traditional Framingham general CVD risk score using lipid profiles and the simplified version using BMI instead of blood lipid tests. However, the validity and applicability of each suggested CVD risk score should be carefully modified or calibrated in any target population to avoid the over- [117, 118] or under-estimation of CVD risk by the original risk score [119-126] and should be the first priority where no validated risk score exists, for example in Vietnam.

26

Commune Hypertension-Centred Management Programme

Community-based CVD interventions are programmes that attempt to modify the prevalence of one or more CVDRFs, CVD mortality or both within a defined community [127]. Four distinct generations of community-based CVD intervention programmes have already demonstrated the effectiveness of population-based approaches to address specifically targeted communities or sub-populations but have also highlighted the importance of multisectoral partnerships within the health care system, the involvement of non-health sectors and orchestrated multilevel interventions (i.e. social or physical environments and relevant public policies) [128-131]. However, the modest impacts of community interventions on CVDRFs also reflect the challenges in selecting the most potent or cost-effective strategies, the most likely-to-respond population, the appropriate model for evaluation, and especially the contextual factors required for effective intervention delivery [132-136]. Key factors for succeeding in a community-based intervention include a good understanding of the targeted population, the deep engagement of all involved partners, an effective dose of interventions in supportive environments and a reliable monitoring system [131, 136-138].

Table 11. Effects of public health services in early or late-adopter community, modified from Pearson TA [137]

Public health service

Early-adopter community Late-adopter community

Surveillance Sufficient to arouse community and improve knowledge

Insufficient to arouse community

Media/ Education

Enough to creates a social norm, to activate behaviour changes

Insufficient to activate community

Community organization/ partnerships

Community organizations for heart health often ready in place

Need to rearrange existing organizations for partnerships

Assuring health services

Services available; high demand for preventive services

Need to reorient toward prevention

Policy/environ-mental change

Community will initiate policy/ environmental change

Needed to create milieu for behaviour change

Among three dimensions of a community intervention comprising targeted CVDRFs, a community setting and public health services, the last component

27

is critical for optimizing and maintaining the effectiveness in a late-adopter community, which are the communities lagging behind in terms of receipt of the ideas or opportunities for example in rural areas (Table 11) [131, 137, 138].

In addition to the prominent roles of health education in community-based CVD interventions, the systematic involvement of the primary healthcare system can be the most effective approach in the long run, especially when the programme targets metabolic CVDRFs such as hypertension, diabetes or dyslipidaemia [131, 139, 140]. In Vietnam, while communicable diseases currently overload the primary health care system and the knowledge of local health care teams on CVDs or CVDRFs is inadequate, health service improvement is one crucial component of any community intervention so that the local primary healthcare system has enough capacity to carry out the essential public health services as well as provide chronic care for people requiring treatment [18, 42, 56].

Implementing a commune-level hypertension-centred management programme such as the model in Ba-Vi district was an initial step to strengthen the primary health care system so that local people could promptly receive the necessary cardiac care services on site. After successful implementation, each local commune health station would become an independent node and join in the national cardiac care network with steady support from reference cardiac centres (paper III). The three complementary targets in the model are: (i) entire commune for health promotion; (ii) hypertensive patients for monthly check-ups, drug delivery and lifestyle consultation; (iii) local health care system by for upgrading (Figure 13).

Figure 13. Process for implementing a hypertension management programme

Training &  Infrastructure Upgrading  to Improve Local Healthcare Capacity

Community Health Promotion to Improve CVDRF Perception and Lifestyle Modification

Hypertension Management by Multi‐Drug & Lifestyle Consultation

Hypertension management programme at commune‐level

Independent node of national cardiac care network 

Cardiac reference centres

Professional supports and supervision

Hub connection

28

Paper III summarized both top-down and bottom-up approaches on how to engage all involved sectors in the hypertension management programme, which is a new community-based intervention on CVDRFs implemented in rural low-resource settings where people were only familiar with traditional infectious diseases. The implementation also highlighted the importance of the close collaboration between local authorities, a supportive cardiac care network and national health authorities for sustaining the intervention activities and its national implications [131]. In the context of a pilot model in a small community, it is very difficult to include collaboration with industry or businesses to create supportive environments, but this policy may be very cost-effective in national-level interventions [131].

One main goal of the hypertension management programme was to improve the capacity for cardiac care, especially hypertension care, among a local team of health workers. Among CVDRFs, hypertension was chosen as a trigger for the programme because it was so prevalent in the local community; it directly linked to stroke that was the leading cause of CVD mortality and morbidity in the rural area; it was easy to implement and train due to simple and reproducible measurements using automatic devices; changes in BP can be self-witnessed to improve the adherence to the life-long multidrug treatment or individual behavioural changes.

Management capacity and knowledge of the local team on CVD and CVDRFs is also very important because local health workers are mainly responsible for chronic care management and the delivery of preventive services for local inhabitants [139]. Supported adequately, primary health workers could deliver lifestyle consultations to the patients; tailor the community intervention to specific individuals in local areas; consolidate the impacts of health promotion and deal promptly with incidental CVD events, thus becoming a feasible and highly cost-effective approach in low-resource settings [141]. The severity of hypertension and effectiveness of treatment were the main factors influencing the decision of local people to join or leave the management programme, which support the initial selection of hypertension and primary care itself as the targets for a community-based comprehensive CVD intervention (paper III).

Perception Of Cardiovascular Diseases In Rural Areas

The preliminary results from our qualitative study in Ba-Vi district showed how local people formed the concept of new non-communicable diseases emerging in the community, like hypertension and stroke from the existing norm like “wind” and the vivid examples of experienced cured infection. Interacting with the traditional beliefs, social norms and information

29

updated from professional healthcare workers, the knowledge of local people on CVD or CVDRFs grew from the Awareness to the Assessment and finally the Action stage (Figure 14). Local people often perceived any CVDRF as a dichotomous variable, ignored the continuum of the CVD prevention; linked the adverse outcomes unavoidably with ageing or fate [142-145]; became “unrealistic optimists” in comparison with some local bad examples [143]; and preferred their own emotion-related experiences to evidence-based ones in building up a new reflection (“seen and confirmed”) [146]. Moreover, local people failed to appreciate the effectiveness of lifestyle changes or lifelong treatment, refused to take responsibility themselves for the progression of CVDRFs, expected immediate healing therapy for chronic problems, kept a fatalistic view of any CVD outcomes and mistrusted local care services. All of these incorrect and fragmentary perceptions demonstrated the ambivalence [147] and the complexity involved in creating or adopting a new concept of disease [146].

Figure 14. Perception pathway for cardiovascular diseases in rural areas

Similar to a widely known infection like tuberculosis [148], the stigma of an ambiguous chronic disease, the fear of high medical expenses and the lack of local health facilities were key factors behind the delays in health seeking for CVD including diagnosis and initiation of appropriate treatment. Data from our qualitative decoding revealed potential gender-gaps on health seeking behaviour for CVD in the area, which was strongly influenced by gender and family structure. In Vietnam, there are two gender constructions: the Confucian overlaps with the Socialist model [149, 150]. Confucianism traditionally assigned to women the values of hard work, chastity and proper behaviour, and focused on female subordination (i.e. moral code of “Three Obediences”). The Socialist women’s emancipation facilitated women being involved widely in social-political life (i.e. 'Three Criteria Women Campaign”) and blended in traditional ideas related to women’s roles in the family [150].

Awareness (Recognition)Education ‐ Communication

Assessment (Reflection)Expression ‐ Comparison

Action (Reinforcement) Experience ‐ Credence

Local social belief/norms

“Wind” concept and sym

bol

Professional consultationHealth education, care services

Traditional infectious diseasesTuberculosis as reference

Non‐communicable diseasesEmerging CVD as example

30

However, women are expected to work diligently to better their families and always put the family’s interest ahead of their own, while men are expected to be the ‘family pillar’ with higher social roles, respect and priority, even in the allocation of healthcare expenses among family members [148, 150]. Even when there was a commune-based hypertension management programme locally available, women still deviated from the programme, possibly explained by the traditional roles of women in the family that tended to disadvantage them and lead to neglect of chronic diseases until obvious symptoms occurred (paper III). The gender hierarchy embodied in tradition and culture seems to create barriers against women accessing quality preventive and curative measures and thus widens the gaps in gender equity for effective management of communicable [151] and non-communicable diseases [152-155].

Community-Based Comprehensive Healthy Lifestyle Promotion

The risk of having a CVD event later in life is influenced in a cumulative fashion by social and environmental health determinants together with behavioural and metabolic CVDRFs acting throughout the life course [8, 156-160]. The incidence and mortality of CVD increase and the quality of life decreases progressively with the number of CVDRFs [161-164]. Often co-occurring as clustered risk factors, lifestyle and cultural habits influence metabolic disorders more than genetic factors [165]. A composite measure of multiple behavioural CVDRFs can predict mortality, above and beyond the predictive value of any single behavioural CVDRF [166-169]. Behavioural risk factors accounted for 50-80% of contributors to chronic disease including CVD [170]. Lifestyle changes such as smoking cessation, healthy diet, and increasing physical activity show excellent cost effectiveness in lowering the CVD burden [171]. Compared to most proven clinical preventive measures, lifestyle intervention results in more than fivefold return even though being imperfect [172]. Prevention by lifestyle changes works better than drugs and lasts longer than percutaneous coronary interventions or surgery, but only if it is done early in life [170]. So, comprehensive healthy lifestyle promotion should be a centrepiece of any efforts to improve cardiac health [173].

The health promotion in our quasi-experimental intervention study applied recommended top-priority, population-wide, highly cost-effective and affordable interventions [32, 35, 51, 53, 174] including smoking cessation, reduction of alcohol consumption, encouragement towards healthier diets and increased physical activity. All these components encouraged local people to adopt a healthy lifestyle to prevent or delay the negative effects of CVDRFs at low cost [175]. Sharing the same messages, health promotion was

31

also combined with individual lifestyle consultations for hypertensives to improve patient adherence to chronic care management of hypertension. Contents for health promotion were also tailored for local demands and requests.

Our intervention study showed that community-based health promotion significantly decreased the prevalence of salty diets (6.4-7.4%), had no impact on prevalence of smoking, and had inconclusive impacts on the prevalence of heavy alcohol consumption (decreased in reference and intervention areas) and the prevalence of physical inactivity (increased in intervention commune) (Figure 15).

Figure 15. Changes of prevalence of behavioural CVDRFs in the general population, after a 3-year health promotion intervention

Our data also recognized the increase of physical inactivity with significant growth of waist circumference, indicative of a sedentary trend in the general population. The decrease of BP and alcohol consumption in the reference community, both of which were unexpected in relation to the CVDRF trends in the general population, might be explained by a Hawthorne effect from the baseline survey or diffusion effects due to population mobility (paper IV).

0%

10%

20%

30%

40%

50%

Phys

ical

Inac

tivity

Reference InterventionFemale Male Female Male

0%

10%

20%

30%

40%

50%

Dai

ly S

mok

ing

Reference InterventionFemale Male Female Male

0%

10%

20%

30%

40%

50%

Hea

vy A

lcoh

ol C

onsu

mpt

ion

Reference InterventionFemale Male Female Male

Baseline After 3 years

0%

10%

20%

30%

40%

50%

Salty

Die

t

Reference InterventionFemale Male Female Male

Baseline After 3 years

32

As in other studies [176, 177], the prevalence of smoking did not change with our health promotion campaigns despite being very common in the local male population in rural areas [66, 178, 179]. Deeply rooted and symbolized in social norms as a masculinity metaphor or a transformation indicator, smoking is accepted as a culturally internalized habit [180], which cannot be easily changed by gender-neutral CVD-oriented messages in opposition to other negative motivations such as the difficulty of struggling against dependency [180] or the fear of weight gain after smoking cessation [181]. Being a strong context-bound behavioural CVDRF, any changes in smoking can be influenced and maintained by supportive social-cultural environments such as tobacco tax and legislation to ban smoking in public areas. Lifestyle interventions theoretically require long-lasting social, political, and economic will to forge multi-level changes among individuals, communities, health systems, and health policy makers [182].

A comprehensive lifestyle intervention should include both primordial and primary prevention to avoid CVDRFs, control existing CVDRFs and defer CVD onset which can bring desired outcomes such as reductions in CVD incidence, mortality and morbidity, and improved quality of life [183]. Individual adherence to comprehensive healthy lifestyles is associated with lower long-term CVD risk [184-186]. However, due to some dilution biases such as social diffusion, population mobility, time lag and multifactorial natural progression of CVDRFs, the effects of any population comprehensive intervention are usually weaker and less certain in relation to mortality and morbidity outcomes, rather than on intermediate outcomes such as changes in intervened CVDRFs [187].

In our intervention study, changes in blood pressure and CVDRF patterns were chosen as the outcomes because they are closely linked to the contents of the interventions, sensitive to the marginal effects and can be self-witnessed to motivate local people to become more involved in CVD prevention activities. However, there are still a lot of unanswered questions on how to optimize the comprehensive health promotion at the individual or population level [169] such as (i) the ideal number of behavioural CVDRFs on which to intervene, (ii) whether intervene on several CVDRFs simultaneously or sequentially [188], and (iii) how to get positive synergistic effects across individuals and the population as a whole [189].

Besides the modest impacts on behavioural CVDRFs, health promotion activities in our study brought significant reductions of 3.0-3.3mmHg in SBP and 4.6-4.7mmHg in DBP after accounting of the effectiveness of the management programme (paper IV). Consequently, the commune-based integrated intervention model significantly improved the awareness,

33

treatment and control of hypertension in the intervention area (Figure 16 and 17).

Figure 16. Changes in blood pressure distribution after 3-year intervention

0.0

1.0

2.0

30

.01

.02

.03

80 120 160 200 240 80 120 160 200 240

Female, Reference Female, Intervention

Male, Reference Male, Intervention

Baseline After 3 years

Kern

el d

ensi

ty o

f SBP

in p

opul

atio

n

Systolic Blood Pressure (mmHg) in general population

0.0

1.0

2.0

3.0

40

.01

.02

.03

.04

40 60 80 100 120 140 160 40 60 80 100 120 140 160

Female, Reference Female, Intervention

Male, Reference Male, Intervention

Baseline After 3 years

Kern

el d

ensi

ty o

f DBP

in p

opul

atio

n

Diastolic Blood Pressure (mmHg) in general population

34

Figure 17. Effectiveness of commune-based management programme on prevalence of hypertension and treatment after a 3-year intervention

In summary, our results suggested a model of comprehensive community healthy lifestyle promotion combined with a commune-based programme on hypertension management was effective in improving CVDRFs and feasible for integration into existing primary health care facilities and then expansion to the national level. Both population-wide and high-risk individual approaches were simultaneously implemented at the commune level to optimize and deal with local demands of CVDRF management. Table 12 summarises the different characteristics of two main approaches.

In 2009, the Vietnamese Prime Minister approved a National Targeted Programme on Prevention and Management of Hypertension, which was proposed and submitted by VNHI. It was a remarkable milestone when the data and experience from our model were applied nationwide, strengthening the primary healthcare system towards better prevention and management of CVDRFs at commune level. In the future, this model can be used as a framework for commune-level interventions for more advanced CVD or other chronic diseases.

0%

20%

40%

60%

Hyp

erte

nsio

n

Reference InterventionFemale Male Female Male

0%

20%

40%

60%

Awar

enes

s of

Hyp

erte

nsio

n

Reference InterventionFemale Male Female Male

0%

20%

40%

60%

Trea

ted

Hyp

erte

nsio

n

Reference InterventionFemale Male Female Male

Baseline After 3 years

0%

20%

40%

60%

Wel

lcon

trolle

d H

yper

tens

ion

Reference InterventionFemale Male Female Male

Baseline After 3 years

35

Table 12. Summary of CVDRF management at commune level

Approach Population (Public Health) High-risk (Clinical)

Target Entire population People with high CVD risk

Preventive focus Primordial and primary Primary and secondary Overall aims Reduce incidence or delay

the onset of CVDRFs Control existing CVDRFs and delay the onset of CVD

Specific measures

Healthy lifestyle: stop smoking, reduce alcohol consumption, encourage healthy diet and physical activity, reduce stress

- Stop smoking - Control hypertension, diabetes, dyslipidaemia - Maintain optimal weight, BMI, waist circumference

Benefits/ Costs +++/+ +/+++ Requirements Aware health policy makers

and involved intersectoral partnership

Qualified primary health care system supported by cardiac care network

Evaluation Cardiac health population metric by repeated cross-sectional surveys

Changes in CVDRFs or in the overall CVD risk by periodical screening and follow-up activities

Resistance - Vague outcomes - Politician inertia - Social health norm/belief

- Individual perception - Physician inertia - Budget constraint - Incapable or overload healthcare system

Priority strategy Policy-level intervention esp. smoking & salty diet

Individual management by multi-drug combination

Priority task Gender-specific contents for behavioural change intervention in community

“cheap, simple, valid” score and threshold to identify almost high risk people

36

Methodological Considerations

Study Limitations

Cross-sectional surveys were the main approach in our study for collecting data on CVDRFs in the general population of Vietnam in different areas at different times. Using random sampling strategies with standardized STEPwise-based questionnaires and identical protocols, these surveys were able to capture the representative characteristics of the studied populations. During each survey, measured biological CVDRFs such as blood pressure or body size (weight, waist and hip circumference) at a single visit [190] sometimes failed to reflect the “real” values of these parameters, which can vary in their own natural circadian rhythms or depend seasonally on weather or harvest cycles, especially in rural areas. Behavioural CVDRFs and medical history of CVD or other chronic diseases are assessed by self-reported questionnaires, which can be easily misclassified by recall bias. However, these inherent data-driven biases of any surveys can be minimized with well-trained surveyors and strong quality control procedures by identical core teams from VNHI. Specific stratification (by sex or residential area), age standardization and multi-level analyses are complementary statistical procedures applied to manage the heteroskedasticity across the surveys at different times or areas. Due to time and budget constraints, some data was missed in some surveys (such as or dyslipidaemia or diabetes diagnosed by blood tests), resulting in increased uncertainty of our estimates or making it impossible to further explore CVDRF interactions. With acceptable and predictable limitations, cross-sectional surveys can be considered as feasible, flexible and appropriate tools for assessing CVDRFs in the general population. Repeated cross-sectional studies can be carried out periodically to monitor the prevalence of variables of interest over certain periods of time (as a pseudo-longitudinal study) and these results could be used in public health policy planning or the development of intervention strategies [191].

The quasi-experimental study design was applied in our study to evaluate the effectiveness on CVDRFs of the local hypertension management model and community health promotion campaigns. Initial cross-sectional surveys were used as a screening study to assess baseline CVDRF patterns as well as to recruit hypertensive patients for the local hypertension management programme. The inherent weakness of the lack of random allocation between intervention and reference groups make the quasi-experimental design able only to inform rather than to establish any cause-effect link, which can be confirmed in the true experimental studies. In low-resource settings like Ba-Vi area, a quasi-experimental design is most appropriate due to the logistics

37

for random assignment not being affordable, as well as the impossibility of avoiding “contamination” of the reference group (by effects from the screening survey itself, the awareness of CVD knowledge in local area being low and the demand being high) and the hypertension management programme being developed over time. However, this design is very valuable by providing information on whether some outcomes are changing while others are not as well as the magnitude of changes over time and which subgroups of the population could benefit from the programme.

The economic and social structure of Vietnam has changed dramatically from that of a poor to a lower-middle income country in the last decade, followed by rapid changes in the CVDRF pattern in the general population. This period of epidemiological transition could amplify changes of CVDRFs in the population so that the secular progression could be easier to identify through repeated cross-sectional surveys without a traditional longitudinal cohort study, which is more costly, requires more time or resources and is even unsuitable sometimes for areas with high proportion of missing data due to local migration. However, this amplification effect can also distort the real effects from community interventions and make the interpretation of these results more difficult especially when there are no reference groups.

The commune-based hypertension management programme integrated with comprehensive health promotion in our study showed effectiveness on CVDRFs based only on outcomes evaluated in the short-term. Due to budget constraints, the data on blood lipid profiles in the general population were missed, as well as hypertension being the only target linked to multi-drug combinations at the current stage. Sharing the same risk factors with other chronic diseases, besides CVD outcomes, components of comprehensive lifestyle promotion potentially influence other chronic diseases in the general population such as cancer or chronic lung diseases. Further studies are needed to evaluate the sustainability of reduced CVD outcomes as well as the potential impacts on other prominent chronic diseases in the community.

Future Directions of the Study

Major CVDRFs increased at similar rates between urban and rural areas while the gaps between CVD burdens and management capacity were bigger in rural areas. The decision to pilot, develop and expand the hypertension management model in rural communes targeted people living in rural areas, which constitute nearly 70% of the population in Vietnam [80], although the rapid urbanisation is occurring widely. People living in urban areas have easier and quicker access to public or private health care services while

38

people in rural areas depend much more on the local primary health care system. Modest impacts of the comprehensive lifestyle interventions in rural areas demonstrated the necessity to apply more intensively the qualitative approaches on these settings. Focusing health education at different perspectives based on gender, social, and cultural aspects is a promising strategy for engaging more social partners, ultimately changing the social norms of many behavioural CVDRFs or health seeking habits, and then motivating individual behavioural changes. Further studies are needed in urban areas where the hypertension management programme could be implemented in different settings such as workplaces, etc. with various properties and impacts [138]. Policy-level interventions are also other research focuses for the future.

The hypertension management programme at commune level combined with comprehensive health promotion in our study demonstrated feasibility for expansion over the country and the ability to integrate this model into the existing primary healthcare system. Besides the population intervention measures, which often have a low cost of implementation [175], the high-risk approach requires periodic screening tests, which would be more costly (for example blood tests to detect hypercholesterolaemia), and multi-drug combinations, which potentially put new financial and resource loads on the primary healthcare system, already overloaded with infectious diseases. Cost-effectiveness studies are needed in the future to evaluate the economic aspects of the programme, identify subgroups which can benefit most and the most appropriate intervention measure in the context of budget constraints.

Reflection on Concept Evolution

With my background as a consultant in intervention cardiology, firstly I joined in the epidemiological surveys on hypertension (NESH) in 2001 as a surveyor and then joined in other studies as member of the VNHI team for the population epidemiological and intervention research. On experiencing and adapting to strange research fields like statistics and public health, I have had to reappraise, add or sometimes integrate a lot of concepts, which have driven the directions of my studies as well as this thesis.

From “disease” to “risk factor” concept

Initially, for myself as well as other doctors in my team, hypertension is considered as a chronic disease - the original concept of “disease” from our antecedent doctors in medical school. We were taught the idea that people with hypertension, defined by some arbitrary cut-point, regardless of their

39

age, gender, or the presence of other CVDRFs, should be detected and controlled with multi antihypertensive drug therapy. Being a disease, we assumed, hypertension should be targeted for intervention and other risk factors such as obesity, smoking, excessive alcohol intake, physical inactivity, stress and high levels of dietary salt were considered as risk factors for hypertension. At that time, the concept of “risk factor” appeared not far from the cause-effect link and the concept of CVDRFs or “high CVD risk” was confusing, difficult to grasp and therefore easily ignored.

In my daily clinical practice, a lot of my patients presented with severe end-stage CVDs such as stroke or CHD but with only mildly elevated hypertension, while quite a number of others, especially young patients, had very high blood pressure without apparent signs or symptoms of CVDs. Many patients cannot have their BP controlled under 140/90 mmHg with many antihypertensive drug combinations although they may still benefit from some blood pressure reduction. Slowly, hypertension lost its role as a CVD outcome and became a convenient surrogate for following up treated patients. The actual demand shifted to the challenging assessment and management of future risk of sudden major adverse CVD events, which are becoming prominent in many booming towns around the country. Motivated by these trends and the overall CVD risk assessment embedded in international guidelines [73, 93, 98], we are adopting gradually, but increasingly widely, the concept of “high CVD risk” and accepting hypertension as a risk factor rather than a manifestation of CVD.

However, when the model of hypertension management was implemented in the primary healthcare system, the local healthcare worker team had to struggle with this conceptual shift. People tend to ignore a risk factor but fear a disease, even when both concepts refer to the same thing, like hypertension. In my proposal’s defence, I have been asked several times to differentiate between the “disease” and the “risk factor” concept for hypertension. For low-resource settings like ours, with low levels of knowledge and perception of hypertension, the “disease” concept of hypertension could be used to motivate the population to accept and adopt healthy lifestyles including the reduction of salt, which causes about one third of new cases of hypertension. At a later phase, when the management programme is running smoothly and people already recognize the importance of hypertension, the target will be all major CVDRF and CVD outcomes, and then hypertension can return to its original role as a “risk factor”.

In the National Targeted Programme of Prevention and Management of Hypertension, after launching quite a number of mass-media health

40

education campaigns on hypertension and healthy lifestyle, we step-by-step introduced the “risk factor” concept of hypertension and merged it with other CVDRFs. The data entry form for the hypertension management software or the paper sheets used in screening surveys was embedded with simple overall CVD risk assessment so that the health workers in the primary care system could be acquainted with and follow the “risk factor” concept.

From “poly-pill” to “policy” intervention

Analysing community-based data, I have recognized the huge burden of CVDRFs on the general population and the potential benefits of proven cost-effective interventions. Compared to the costly and time-consuming procedures I have done daily in order to gain some quality of life years for individuals, this has been at the expense of stressing that population interventions are much more likely to have a large health impact. Clearly both are required and a balance is needed.

Implementing the programme on hypertension management at the level of the commune health station, I have recognized the limitations of the existing primary health care system to manage CVD/CVDRFs. This management model targeted at hypertensives will be highly dependent on well-functioning primary healthcare system, which requires comprehensive approaches and coordinated health policy changes [56]. Moreover, the model challenges the current capacity of the outmoded healthcare systems, with the system soon becoming overloaded by emerging treatment demands, which require corresponding health policies in order to reasonably allocate financial resources (such as the application of insurance at primary heath system) or professional health worker resources (such as the development of a cardiac care network from primary to tertiary or regional hospitals).

In conducting some qualitative research and then evaluating the community intervention study, I have recognized the context-bound aspects of CVDRFs especially the behavioural ones. All elements of the agro-eco-socio-system have hidden but very high impacts on population health, not only through the population dietary intake [54] but also through complex interactions between economic cultural determinants from childhood or even earlier [8, 30]. The negative impacts of CVDs (or chronic diseases) on human and economic development of individuals, families or entire countries should be solved comprehensively and alongside other pressing national problems, which requires broad political commitment, leadership at high level and coordinated coherent strategies for action.

41

Following the commune management model, the National Targeted Programme of Prevention and Management of Hypertension indicated our advocacy approach to address CVD burden in the general population: bringing community-level evidence for health policy makers to determine the most appropriate health policies in order to mobilize all social resources to improve the health of the population.

42

Conclusion and Policy Implications

The general population of Vietnam is facing an epidemic of cardiovascular risk factors (CVDRFs) which are increasing rapidly, and clustered within individuals. While the management capacity for these CVDRFs is low especially in primary healthcare systems and in low-resource settings, these trends in CVDRFs at population level threaten a considerable projected CVD epidemic in a short time in the absence of rapid and widespread appropriate reactions.

Two approaches should be simultaneously applied to tackle this emerging problem, including a population-wide approach with community health promotion and a high CVD-risk individual approach with multi-medication. Policy-level interventions targeted at smoking and reduced salt diets should be the first priority for population interventions, while multi-drug strategies could be reasonable for high-risk individual interventions. Calibration or modification of an existing score for identifying high CVD-risk individuals in the general population and applying a gender perspective to behavioural change interventions are necessary tasks, which need to be done first in order to optimize and maintain the long-term results of CVD intervention.

Implementing the commune-based hypertension management programme in combination with community comprehensive health promotion campaigns is feasible in rural low-resource settings and effective in improving CVDRFs, especially blood pressure, in the general population. Hypertension is a likely key trigger-point for advancing the programme on managing other CVDRFs in future stages. This implementation is also a necessary step for strengthening the primary health care system towards a better capacity for chronic cardiac care management and enabling poor people to access the necessary services promptly and locally. Combined with surveillance activities via repeated cross-sectional surveys, this model can be used as an intervention framework for advanced CVD or other chronic diseases. The modest impact of health promotion on smoking and the sedentary trend in the community suggest that more supportive environments (e.g. health policy) or customized contents of health education (e.g. gender-specific) are needed to optimize the effectiveness and maintain the long-term impacts of population comprehensive healthy lifestyle promotion.

In developing countries like Vietnam, the prevention and management of CVDs or other chronic diseases cannot be separate from, but must cooperate and collaborate with, approaches for dealing with infectious diseases.

43

Besides the strengthening of the well-functioning health care system and the implementation of the robust evidence-based interventions, advancing the prevention of CVDs or chronic diseases requires coordinated political and stepwise health system actions, in which enlightened leadership at high levels and priority setting for using the most cost-effective interventions first are critical. Political oblivion or inertia rather than practical failure are likely to be the main barrier to addressing the CVDs or other chronic disease pandemics. Using a proximal surrogate like hypertension rather than other real fatal or non-fatal CVD endpoints, the model of community-based prevention and management of hypertension is reaching only a modest goal at its starting point. The National Targeted Programme on Prevention and Control of Hypertension in Vietnam, proposed initially by VNHI and approved subsequently by the Prime Minister in 2009, is needed not only to apply and expand nationwide the experience and estimation from the model of hypertension management at commune level but also to maximize resource mobilisation and fully awaken stakeholders and political leaders to the danger of the epidemic of CVDRFs. After extending the size, this programme will be broadened in its scope to target ultimate endpoints, such as a reduction in premature death due to CVDs or other non-communicable diseases.

44

Acknowledgements

I wish to acknowledge financial support from Department for Research Cooperation of the Swedish International Development Cooperation Agency (SIDA/SAREC) for providing a scholarship through the Health System Research Project (HSRP), and a long-term collaboration between Vietnamese and Swedish institutions. I wish to acknowledge the support for population studies from World Heart Organization Western Pacific Region Office through VNHI, from SIDA/SAREC through HSRP, from Ministry of Health Vietnam through VNHI, from the International Network of field sites for continuous Demographic Evaluation of Populations and Their Health in developing countries (INDEPTH) and FAS (the Swedish Council for Social and Work Life Research); from a grant for International Health Research (20A-6), Japanese Ministry of Health, Labour and Welfare and a grant of National Centre for Global Health and Medicine, Japan (22A-10). Support from the Umeå Centre for Global Health Research, funded from FAS (grant no. 2006-1512) is appreciated.

I would like to express my deep gratitude to everyone who has contributed to my studies and the thesis either directly or spiritually with their generous support, advice or inspiration. Without their valuable contributions, this study would have never been completed.

Especially, I would like to show my sincere thanks and great appreciation to the following people:

Professor Peter Byass, my main supervisor for his continuous and sympathetic support for all my research, especially as I was a part-time student busy with clinical responsibilities and living far away in Vietnam. Geographical or time zone distance was not a hindrance in receiving practical instruction and brilliant suggestions at the right time. I am very impressed with his patience with my poorly written English manuscripts, which he always generously corrected.

Professor Nguyen Lan Viet, my co-supervisor for his continuous support enabling me to do all the necessary studies. Being Director of VNHI, former Dean of HMU and Head of the HSRP, he helped me to get involved in the HSRP, become a research student at Umeå University, carry out community research in Vietnam and implement the hypertension management model over the country through the National Targeted Programme on Prevention and Control of Hypertension. I have always been learning from his huge experience and knowledge since I was a resident.

45

Professor Stig Wall, my co-supervisor for his support immeasurable beyond words. I always feel his warm concern for my personal and professional status whenever I was staying in Umeå. I remember his strategic guidance and sharp and clear comments, sometimes demanding on my data analysis skills, which usually awakened me and raised the academic level of the manuscripts. He made me recognize the importance of public health as well as interdisciplinary and cross-cultural collaboration.

Professor Ruth Bonita, my co-supervisor for her superior comments, suggestions and corrections for all the manuscripts and this cover story. Her immense knowledge, experience and perspicacious judgement especially in non-communicable diseases or health policy benefited me a lot in my research work. Time zone differences between Auckland, Hanoi and Umeå enabled me to receive her quick responses: these tricks I sometimes overused.

Professor Lars Weinehall, my co-supervisor. I am always grateful for his timely and great support and comments on my thesis or my study time in Umeå. His experience and knowledge in community programmes on CVD prevention is very useful for my research work.

Uncles Le Quoc Ty and Dinh Tran Toan, my childhood mentors in foreign language and oriental philosophy. I always remember their humorous and mysterious lessons, which made me have faith in the beauty of Vietnamese language, in the perfection and in the systematic induction.

A/Professor Nguyen Thi Kim Chuc, coordinator of the HSRP in Vietnam for initiating and facilitating all my research work within FilaBavi. Even though she is not a formal tutor of my study, I am always grateful for her continuously invaluable advice and encouragement to my studies and the thesis.

Professor Pham Gia Khai at VNHI, HMU for his continuous support and advice. I always admire his vision and have learnt a lot from his enormous experience and knowledge in cardiology since I was a cardiology resident.

Professor Thach Nguyen, my “big friend” in Merrillville, Indiana, USA for the inspiration to me from his work on the integration of basic science into clinical cardiology, international leadership and collaboration.

A/Professor Do Doan Loi at VNHI, HMU for his continuous support and suggestions in academic research. He is the one who opened the gate of cardiology for me when I was a medical student.

46

A/Professor Pham Manh Hung, my “big brother” at VNHI, HMU for his continuous support and encouragement. He is the one who drove me to the field of cardiology as well as intervention cardiology. I have learnt a lot from his inspiration in working, his confidence in handling fatal situations and his great experience in clinical medicine.

Dr Pham Thai Son, my co-researcher, colleague and close friend for his collaborating, sharing and enjoying work and life with me in VNHI, around Vietnam and in Umeå.

People and friends from Division of Epidemiology and Global Health, Department of Public Health and Clinical Medicine, Umeå University. I always received their warm hospitality, which becomes vital for me in wintertime in Umeå. Birgitta Åström, Anna-Lena Johansson, Karin Johansson, Lena Mustonen, Göran Lönnberg, Jerzy Pilch for their kindness and valuable help in administration or logistics arrangements. Professor Lars Lindholm and Dr Curt Löfgren for invaluable discussions on evaluations in public health when I was fumbling with my proposal. Dr Hans Stenlund, Dr Kjerstin Dahlblom, Dr Klas-Göran Sahlen, Dr Nawi Ng, Prof. Ingrid Mogren, Prof. Ann Öhman, Dr Isabel Goicolea, Dr Kristina Lindvall, Dr Miguel San Sebastian, Fredrik Norström, Felix Kisanga, for their help to my work and their kind sharing of my life in Umeå.

My friends, colleagues from the HSRP, the officers, the surveyors and field supervisors from FilaBavi, the healthcare teams at Phu-Cuong, Phu-Phuong communes and Ba-Vi district hospital: Nguyen Binh Minh, Dr Hoang Minh Hang, Dr Hoang Van Minh, Dr Nguyen Thi Thanh, Dr Tran Thanh Do, Dr Mattias Larsson, Dang Thi Tuyen, Nghiem Thi Minh Trang, Nguyen Nguyet Minh, Dao Dinh Sang, Truong Hoang Long, Tran Khanh Toan, Dr Nguyen Van Lam for their tireless contributions to the fieldwork; organizing local surveys; entering data and arranging the logistics.

My colleagues from the research team, the cardiology residents, involved medical students, staff from the secretariat of the National Targeted Programme on Prevention and Control of Hypertension and other involved partners: Dr Nguyen Thi Bach Yen, Dr Nguyen Quoc Khanh, Dr Nguyen Thi Thu Hoai, Dr Pham Tran Linh, Dr Tran Hai Yen, Dr Do Thuy Can, Dr Nguyen Xuan Tu, Dr Le Thanh Binh, Dr Van Duc Hanh, Dr Pham Quang Huy, Dr Duong Ngoc Long, Dr Bui The Long, Nguyen Thi Thanh, Nguyen Van Sinh, Nguyen Dinh Tam for their contributions and support to my work.

My teachers and colleagues from Department of Cardiology, HMU and my senior and junior doctors, nurses from Cathlab, CTSICU (C1) and other units

47

(C2-C8), VNHI: Dr Tran Van Duong, A/Prof. Dinh Thi Thu Huong, A/Prof. Truong Thanh Huong, A/Prof. Nguyen Quang Tuan, Dr Nguyen Lan Hieu, Dr Nguyen Tuan Hai, Dr Phan Dinh Phong, Dr Dinh Huynh Linh, Dr Phan Tuan Dat, Dr Nguyen Thi Minh Ly, Dr Tran Van Dong, Dr Duong Duc Hung, Dr Pham Nhu Hung, Dr Vu Kim Chi for their support and encouragement for my study in Umeå and for sharing a lot of work when I am busy or away for my studies.

Most of all, my deepest thanks go to my big family, my dearest wife Vu Ngoc Diep and my son Nguyen Quang Lam, my parents, my parents in law, my brother, my sister in law, my nephew and niece for their boundless love and everlasting support. Their encouragement and sharing are invaluable for me in overcoming all difficulties and challenges in my life and career. All my beloved deserve more for their suffering from my thoughtless life due to the unpremeditated emergency tasks in hospital or my long commitment to study and research in the medical field. I always feel guilty for being away so much from them although this thought makes me stronger to continue my journey.

Friends and colleagues, and all of those I did not mention by name for their support. Lack of space in this thesis to display your name does not mean you are forgotten, but engraved in my memory.

Last, but not least, I would like to thank my patients and their families, who have put their trust in the healthcare services provided by my team and myself, especially people in FilaBavi for their cooperation and participation in the FilaBavi project. I wish that this small work might be of some use for them, and the people of Vietnam in return.

48

References 1. WHO: Global atlas on cardiovascular disease prevention and

control. Geneva: World Health Organization; 2011. 2. WHO: Causes of death 2008: data sources and methods. Geneva:

World Health Organization; 2011. 3. WHO: The global burden of disease: 2004 update. Geneva: World

Health Organization; 2008. 4. Alwan A, Maclean DR, Riley LM, et al: Monitoring and surveillance of

chronic non-communicable diseases: progress and capacity in high-burden countries. Lancet 2010, 376(9755):1861-1868.

5. WHO: Global status report on non-communicable diseases 2010. Geneva: World Health Organization; 2010.

6. Yusuf S, Hawken S, Ounpuu S et al: Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet 2004, 364(9438):937-952.

7. O'Donnell MJ, Xavier D, Liu L et al: Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study. Lancet 2010, 376(9735):112-123.

8. Smith GD, Hart C: Life-course socioeconomic and behavioral influences on cardiovascular disease mortality: the collaborative study. Am J Public Health 2002, 92(8):1295-1298.

9. WHO: Preventing chronic diseases : a vital investment. Geneva: World Health Organization; 2005.

10. Mozaffarian D, Wilson PW, Kannel WB: Beyond established and novel risk factors: lifestyle risk factors for cardiovascular disease. Circulation 2008, 117(23):3031-3038.

11. Sassi F, Cecchini M, Lauer J, et al: Improving lifestyles, tackling obesity: the health and economic impact of prevention strategies. In: OECD Health Working Paper 48. Paris: OECD; 2009.

12. WHO: Cardiovascular diseases (CVDs), Fact sheet Number 317. In. Geneva: World Health Organization; 2011.

13. Ni Mhurchu C, Rodgers A, Pan WH, et al: Body mass index and cardiovascular disease in the Asia-Pacific Region: an overview of 33 cohorts involving 310 000 participants. Int J Epidemiol 2004, 33(4):751-758.

14. Barzi F, Patel A, Woodward M, et al: A comparison of lipid variables as predictors of cardiovascular disease in the Asia Pacific region. Ann Epidemiol 2005, 15(5):405-413.

15. Martiniuk AL, Lee CM, Lam TH et al: The fraction of ischaemic heart disease and stroke attributable to smoking in the WHO Western Pacific and South-East Asian regions. Tob Control 2006, 15(3):181-188.

49

16. Martiniuk AL, Lee CM, Lawes CM, et al: Hypertension: its prevalence and population-attributable fraction for mortality from cardiovascular disease in the Asia-Pacific region. J Hypertens 2007, 25(1):73-79.

17. Karthikeyan G, Teo KK, Islam S, et al: Lipid profile, plasma apolipoproteins, and risk of a first myocardial infarction among Asians: an analysis from the INTERHEART Study. J Am Coll Cardiol 2009, 53(3):244-253.

18. Dans A, Ng N, Varghese C, et al: The rise of chronic non-communicable diseases in southeast Asia: time for action. Lancet 2011, 377(9766):680-689.

19. Kim AS, Johnston SC: Global variation in the relative burden of stroke and ischemic heart disease. Circulation 2011, 124(3):314-323.

20. Omran AR: The epidemiologic transition. A theory of the epidemiology of population change. 1971. Bull World Health Organ 2001, 79(2):161-170.

21. Johnston SC, Mendis S, Mathers CD: Global variation in stroke burden and mortality: estimates from monitoring, surveillance, and modelling. Lancet Neurol 2009, 8(4):345-354.

22. Ueshima H, Sekikawa A, Miura K, et al: Cardiovascular disease and risk factors in Asia: a selected review. Circulation 2008, 118(25):2702-2709.

23. Yoon KH, Lee JH, Kim JW, et al: Epidemic obesity and type 2 diabetes in Asia. Lancet 2006, 368(9548):1681-1688.

24. Asia Pacific Cohort Studies Collaboration: Cholesterol, diabetes and major cardiovascular diseases in the Asia-Pacific region. Diabetologia 2007, 50(11):2289-2297.

25. Ramachandran A, Ma RC, Snehalatha C: Diabetes in Asia. Lancet 2010, 375(9712):408-418.

26. Moran A, Gu D, Zhao D, et al: Future cardiovascular disease in China: markov model and risk factor scenario projections from the coronary heart disease policy model-China. Circ Cardiovasc Qual Outcomes 2010, 3(3):243-252.

27. Gaziano TA, Bitton A, Anand S, et al: Growing epidemic of coronary heart disease in low- and middle-income countries. Curr Probl Cardiol 2010, 35(2):72-115.

28. Huffman MD, Prabhakaran D, Osmond C, et al: Incidence of cardiovascular risk factors in an Indian urban cohort results from the New Delhi birth cohort. J Am Coll Cardiol 2011, 57(17):1765-1774.

29. Gupta R, Deedwania P: Interventions for cardiovascular disease prevention. Cardiol Clin 2011, 29(1):15-34.

30. Ben-Shlomo Y, Kuh D: A life course approach to chronic disease epidemiology: conceptual models, empirical challenges and interdisciplinary perspectives. Int J Epidemiol 2002, 31(2):285-293.

50

31. Rose G: Sick individuals and sick populations. Int J Epidemiol 2001, 30(3):427-432; discussion 433-434.

32. Lim SS, Gaziano TA, Gakidou E, et al: Prevention of cardiovascular disease in high-risk individuals in low-income and middle-income countries: health effects and costs. Lancet 2007, 370(9604):2054-2062.

33. Cooney MT, Dudina A, Whincup P, et al: Re-evaluating the Rose approach: comparative benefits of the population and high-risk preventive strategies. Eur J Cardiovasc Prev Rehabil 2009, 16(5):541-549.

34. Mendis S, Fuster V: National policies and strategies for noncommunicable diseases. Nat Rev Cardiol 2009, 6(11):723-727.

35. Beaglehole R, Bonita R, Horton R, et al: Priority actions for the non-communicable disease crisis. Lancet 2011, 377(9775):1438-1447.

36. Minh HV, Byass P, Wall S: Mortality from cardiovascular diseases in Bavi district, Vietnam. Scand J Public Health Suppl 2003, 62:26-31.

37. Huong DL, Minh HV, Byass P: Applying verbal autopsy to determine cause of death in rural Vietnam. Scand J Public Health Suppl 2003, 62:19-25.

38. Hoang VM, Dao LH, Wall S, et al: Cardiovascular disease mortality and its association with socioeconomic status: findings from a population-based cohort study in rural Vietnam, 1999-2003. Prev Chronic Dis 2006, 3(3):A89.

39. Ngo AD, Rao C, Hoa NP, et al: Mortality patterns in Vietnam, 2006: Findings from a national verbal autopsy survey. BMC Res Notes 2010, 3:78.

40. Nguyen TTN, Tran KL, Bui ML, et al: Vietnam burden of diseases and injury study 2008. Hanoi: Medical Publishing House; 2011.

41. WHO: ECOSOC high-level segment 2009 discussion paper: noncommunicable diseases, poverty and the development agenda. Geneva: World Health Organization; 2009.

42. Leowski J, Krishnan A: Capacity to control noncommunicable diseases in the countries of South-East Asia. Health Policy 2009, 92(1):43-48.

43. Yusuf S, Reddy S, Ounpuu S, et al: Global burden of cardiovascular diseases: part I: general considerations, the epidemiologic transition, risk factors, and impact of urbanization. Circulation 2001, 104(22):2746-2753.

44. Feigin VL, Lawes CM, Bennett DA, et al: Worldwide stroke incidence and early case fatality reported in 56 population-based studies: a systematic review. Lancet Neurol 2009, 8(4):355-369.

45. Ford ES, Capewell S: Proportion of the decline in cardiovascular mortality disease due to prevention versus treatment: public health versus clinical care. Annu Rev Public Health 2011, 32(1):5-22.

51

46. Capewell S, O'Flaherty M, Ford ES, et al: Potential reductions in United States coronary heart disease mortality by treating more patients. Am J Cardiol 2009, 103(12):1703-1709.

47. Kabir Z, Bennett K, Shelley E, et al: Life-years-gained from population risk factor changes and modern cardiology treatments in Ireland. Eur J Public Health 2007, 17(2):193-198.

48. Capewell S, Hayes DK, Ford ES, et al: Life-years gained among US adults from modern treatments and changes in the prevalence of 6 coronary heart disease risk factors between 1980 and 2000. Am J Epidemiol 2009, 170(2):229-236.

49. Capewell S, Beaglehole R, Seddon M, et al: Explanation for the decline in coronary heart disease mortality rates in Auckland, New Zealand, between 1982 and 1993. Circulation 2000, 102(13):1511-1516.

50. Ford ES, Ajani UA, Croft JB, et al: Explaining the decrease in U.S. deaths from coronary disease, 1980-2000. N Engl J Med 2007, 356(23):2388-2398.

51. Asaria P, Chisholm D, Mathers C, et al: Chronic disease prevention: health effects and financial costs of strategies to reduce salt intake and control tobacco use. Lancet 2007, 370(9604):2044-2053.

52. Gaziano TA, Galea G, Reddy KS: Scaling up interventions for chronic disease prevention: the evidence. Lancet 2007, 370(9603):1939-1946.

53. Cecchini M, Sassi F, Lauer JA, et al: Tackling of unhealthy diets, physical inactivity, and obesity: health effects and cost-effectiveness. Lancet 2010, 376(9754):1775-1784.

54. Lock K, Smith RD, Dangour AD, et al: Health, agricultural, and economic effects of adoption of healthy diet recommendations. Lancet 2010, 376(9753):1699-1709.

55. Geneau R, Stuckler D, Stachenko S, et al: Raising the priority of preventing chronic diseases: a political process. Lancet 2010, 376(9753):1689-1698.

56. Samb B, Desai N, Nishtar S, et al: Prevention and management of chronic disease: a litmus test for health-systems strengthening in low-income and middle-income countries. Lancet 2010, 376(9754):1785-1797.

57. Deedwania P: Preface cardiovascular disease prevention: the continuum of primary and secondary prevention. Cardiol Clin 2011, 29(1):xi-xiii.

58. Narayan KM, Ali MK, del Rio C, et al: Global non-communicable diseases - Lessons from the HIV–AIDS experience. N Engl J Med 2011, 365(10):876-878.

59. Rabkin M, El-Sadr WM: Why reinvent the wheel? Leveraging the lessons of HIV scale-up to confront non-communicable diseases. Glob Public Health 2011, 6(3):247-256.

52

60. Dabhadkar KC, Kulshreshtha A, Ali MK, et al: Prospects for a cardiovascular disease prevention polypill. Ann Rev Public Health 2011, 32:23-38.

61. Mayosi BM, Flisher AJ, Lalloo UG, et al: The burden of non-communicable diseases in South Africa. Lancet 2009, 374(9693):934-947.

62. Gersh BJ, Sliwa K, Mayosi BM, et al: Novel therapeutic concepts: the epidemic of cardiovascular disease in the developing world: global implications. Eur Heart J 2010, 31(6):642-648.

63. Franco M, Cooper RS, Bilal U, et al: Challenges and opportunities for cardiovascular disease prevention. Am J Med 2011, 124(2):95-102.

64. Son PT, Quang NN, Viet NL, et al: Prevalence, awareness, treatment and control of hypertension in Vietnam - results from a national survey. J Hum Hypertens 2012, 26(4):268-280.

65. Chuc NT, Diwan V: FilaBavi, a demographic surveillance site, an epidemiological field laboratory in Vietnam. Scand J Public Health Suppl 2003, 31(Suppl. 62):3-7.

66. Hoang VM, Byass P, Dao LH, et al: 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 2007, 4(2): A22.

67. Minh HV, Huong DL, Giang KB: Self-reported chronic diseases and associated sociodemographic status and lifestyle risk factors among rural Vietnamese adults. Scand J Public Health 2008, 36(6):629-634.

68. Bonita R, de Counter M, Jamrozi K, et al: Surveillance of risk factors for non-communicable disease: The WHO STEPwise approach. Geneva: World Health Organization; 2001.

69. WHO: WHO STEPS surveillance manual : the WHO STEPwise approach to chronic disease risk factor surveillance. Geneva: World Health Organization; 2005.

70. Lloyd-Jones DM, Hong Y, Labarthe D, et al: Defining and setting national goals for cardiovascular health promotion and disease reduction: the American Heart Association's strategic Impact Goal through 2020 and beyond. Circulation 2010, 121(4):586-613.

71. WHO Western Pacific Region: The Asia-Pacific perspective: redefining obesity and its treatment: Health Communications Australia Pty Limited; 2000.

72. The sixth report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood pressure. Arch Intern Med 1997, 157(21):2413-2446.

73. Chobanian AV, Bakris GL, Black HR, et al: Seventh report of the Joint National Committee on prevention, detection, evaluation, and

53

treatment of high blood pressure. Hypertension 2003, 42(6):1206-1252.

74. MOH: Guidelines for management of hypertension. Department of Therapeutics, Ministry of Health, Vietnam; 2010.

75. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III): Third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) final report. Circulation 2002, 106(25):3143-3421.

76. Genuth S, Alberti KG, Bennett P, et al: Follow-up report on the diagnosis of diabetes mellitus. Diabetes Care 2003, 26(11):3160-3167.

77. American Diabetes Association: Diagnosis and classification of diabetes mellitus. Diabetes Care 2010, 33 Suppl 1:S62-69.

78. American Diabetes Association: Standards of medical care in diabetes - 2012. Diabetes Care 2012, 35 Suppl 1:S11-63.

79. D'Agostino RB Sr, Vasan RS, Pencina MJ, et al: General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation 2008, 117(6):743-753.

80. The 2009 Vietnam population and housing census: completeted results [http://www.gso.gov.vn/default.aspx?tabid=512&idmid=5&ItemID=10798]

81. Danaei G, Finucane MM, Lin JK, et al: 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 2011, 377(9765):568-577.

82. Finucane MM, Stevens GA, Cowan MJ, et al: 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 2011, 377(9765):557-567.

83. Farzadfar F, Finucane MM, Danaei G, et al: 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 2011, 377(9765):578-586.

84. Tolonen H, Mähönen M, Asplund K, et al: 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. Stroke 2002, 33(10):2367-2375.

85. Anand SS, Yusuf S: Stemming the global tsunami of cardiovascular disease. Lancet 2011, 377(9765):529-532.

54

86. Wu Z, Yao C, Zhao D, et al: Sino-MONICA Project: a collaborative study on trends and determinants in cardiovascular diseases in China, part I: morbidity and mortality monitoring. Circulation 2001, 103(3):462-468.

87. Gu D, Gupta A, Muntner P, et al: Prevalence of cardiovascular disease risk factor clustering among the adult population of China: results from the International Collaborative Study of Cardiovascular Disease in Asia (InterAsia). Circulation 2005, 112(5):658-665.

88. Teo KK, Liu L, Chow CK, et al: Potentially modifiable risk factors associated with myocardial infarction in China: the INTERHEART China study. Heart 2009, 95(22):1857-1864.

89. Stuckler D: Population causes and consequences of leading chronic diseases: a comparative analysis of prevailing explanations. Milbank Q 2008, 86(2):273-326.

90. Giang KB, Allebeck P: Self-reported illness and use of health services in a rural district of Vietnam: findings from an epidemiological field laboratory. Scand J Public Health Suppl 2003, 62:52-58.

91. Thanh NX, Hang HM, Chuc NT, et al: Does "the injury poverty trap" exist? A longitudinal study in Bavi, Vietnam. Health Policy 2006, 78(2-3):249-257.

92. Elliott WJ: Systemic hypertension. Curr Probl Cardiol 2007, 32(4):201-259.

93. Mancia G, de Backer G, Dominiczak A, et al: 2007 Guidelines for the management of arterial hypertension: The Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). Eur Heart J 2007, 28(12):1462-1536.

94. Berger JS, Jordan CO, Lloyd-Jones D, et al: Screening for cardiovascular risk in asymptomatic patients. J Am Coll Cardiol 2010, 55(12):1169-1177.

95. Mendis S, Lindholm LH, Anderson SG, et al: Total cardiovascular risk approach to improve efficiency of cardiovascular prevention in resource constrain settings. J Clin Epidemiol 2011, 64(12):1451-1462.

96. De Backer G, Ambrosioni E, Borch-Johnsen K, et al: European guidelines on cardiovascular disease prevention in clinical practice. Third joint task force of European and other societies on cardiovascular disease prevention in clinical practice. Eur Heart J 2003, 24(17):1601-1610.

97. Jackson R, Lawes CM, Bennett DA, et al: Treatment with drugs to lower blood pressure and blood cholesterol based on an individual's absolute cardiovascular risk. Lancet 2005, 365(9457):434-441.

55

98. WHO: Prevention of cardiovascular disease: guidelines for assessment and management of total cardiovascular risk. Geneva: World Health Organization; 2007.

99. Arcucci O, de Simone G, Izzo R, et al: Association of suboptimal blood pressure control with body size and metabolic abnormalities. J Hypertens 2007, 25(11):2296-2300.

100. Weycker D, Nichols GA, O'Keeffe-Rosetti M, et al: Risk-factor clustering and cardiovascular disease risk in hypertensive patients. Am J Hypertens 2007, 20(6):599-607.

101. Belletti DA, Zacker C, Wogen J: Effect of cardiometabolic risk factors on hypertension management: a cross-sectional study among 28 physician practices in the United States. Cardiovasc Diabetol 2010, 9:7.

102. Ohta Y, Tsuchihashi T, Onaka U, et al: Clustering of cardiovascular risk factors and blood pressure control status in hypertensive patients. Intern Med 2010, 49(15):1483-1487.

103. Baker S, Priest P, Jackson R: Using thresholds based on risk of cardiovascular disease to target treatment for hypertension: modelling events averted and number treated. BMJ 2000, 320(7236):680-685.

104. Murray CJ, Lauer JA, Hutubessy RC, et al: Effectiveness and costs of interventions to lower systolic blood pressure and cholesterol: a global and regional analysis on reduction of cardiovascular-disease risk. Lancet 2003, 361(9359):717-725.

105. Emberson J, Whincup P, Morris R, et al: Evaluating the impact of population and high-risk strategies for the primary prevention of cardiovascular disease. Eur Heart J 2004, 25(6):484-491.

106. Gaziano TA, Opie LH, Weinstein MC: Cardiovascular disease prevention with a multidrug regimen in the developing world: a cost-effectiveness analysis. Lancet 2006, 368(9536):679-686.

107. Qin X, Jackson R, Marshall R, et al: Modelling the potential impact of population-wide and targeted high-risk blood pressure-lowering strategies on cardiovascular disease in China. Eur J Cardiovasc Prev Rehabil 2009, 16(1):96-101.

108. Batsis JA, Lopez-Jimenez F: Cardiovascular risk assessment--from individual risk prediction to estimation of global risk and change in risk in the population. BMC Med 2010, 8:29.

109. Grundy SM, Pasternak R, Greenland P, et al: Assessment of cardiovascular risk by use of multiple-risk-factor assessment equations: a statement for healthcare professionals from the American Heart Association and the American College of Cardiology. Circulation 1999, 100(13):1481-1492.

110. Pearson TA, Blair SN, Daniels SR, et al: AHA guidelines for primary prevention of cardiovascular disease and stroke: 2002 Update: consensus panel guide to comprehensive risk reduction for adult

56

patients without coronary or other atherosclerotic vascular diseases. Circulation 2002, 106(3):388-391.

111. Grundy SM, Cleeman JI, Merz CN, et al: Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation 2004, 110(2):227-239.

112. Graham I, Atar D, Borch-Johnsen K, et al: European guidelines on cardiovascular disease prevention in clinical practice: executive summary. Eur Heart J 2007, 28(19):2375-2414.

113. US Preventive Services Task Force: Aspirin for the prevention of cardiovascular disease: US preventive services task force recommendation statement. Ann Intern Med 2009, 150(6):396-404.

114. Kones R: Primary prevention of coronary heart disease: integration of new data, evolving views, revised goals, and role of rosuvastatin in management. A comprehensive survey. Drug Des Devel Ther 2011, 5:325-380.

115. Barzi F, Patel A, Gu D, et al: Cardiovascular risk prediction tools for populations in Asia. J Epidemiol Community Health 2007, 61(2):115-121.

116. Gaziano TA, Young CR, Fitzmaurice G, et al: Laboratory-based versus non-laboratory-based method for assessment of cardiovascular disease risk: the NHANES I follow-up study cohort. Lancet 2008, 371(9616):923-931.

117. Liu J, Hong Y, D'Agostino RB Sr, et al: Predictive value for the Chinese population of the Framingham CHD risk assessment tool compared with the Chinese Multi-Provincial Cohort Study. JAMA 2004, 291(21):2591-2599.

118. Wu Y, Liu X, Li X, et al: Estimation of 10-year risk of fatal and nonfatal ischemic cardiovascular diseases in Chinese adults. Circulation 2006, 114(21):2217-2225.

119. Laurier D, Nguyen PC, Cazelles B, et al: Estimation of CHD risk in a French working population using a modified Framingham model. The PCV-METRA Group. J Clin Epidemiol 1994, 47(12):1353-1364.

120. Liao Y, McGee DL, Cooper RS, et al: How generalizable are coronary risk prediction models? Comparison of Framingham and two national cohorts. Am Heart J 1999, 137(5):837-845.

121. Menotti A, Puddu PE, Lanti M: Comparison of the Framingham risk function-based coronary chart with risk function from an Italian population study. Eur Heart J 2000, 21(5):365-370.

122. Thomsen TF, Davidsen M, Ibsen H, et al: A new method for CHD prediction and prevention based on regional risk scores and randomized clinical trials; PRECARD and the Copenhagen risk score. J Cardiovasc Risk 2001, 8(5):291-297.

123. Diverse Populations Collaborative Group: Prediction of mortality from coronary heart disease among diverse populations: is there a common predictive function? Heart 2002, 88(3):222-228.

57

124. Brindle P, Emberson J, Lampe F, et al: Predictive accuracy of the Framingham coronary risk score in British men: prospective cohort study. BMJ 2003, 327(7426):1267.

125. Empana JP, Ducimetiere P, Arveiler D, et al: Are the Framingham and PROCAM coronary heart disease risk functions applicable to different European populations? The PRIME Study. Eur Heart J 2003, 24(21):1903-1911.

126. Hense HW, Schulte H, Löwel H, et al: Framingham risk function overestimates risk of coronary heart disease in men and women from Germany--Results from the MONICA Augsburg and the PROCAM cohorts. Eur Heart J 2003, 24(10):937-945.

127. Parker DR, Assaf AR: Community interventions for cardiovascular disease. Prim Care 2005, 32(4):865-881.

128. Papadakis S, Moroz I: Population-level interventions for coronary heart disease prevention: what have we learned since the North Karelia project? Curr Opin Cardiol 2008, 23(5):452-461.

129. Eriksson C: Learning and knowledge-production for public health: a review of approaches to evidence-based public health. Scand J Public Health 2000, 28(4):298-308.

130. Pennant M, Davenport C, Bayliss S, et al: Community programs for the prevention of cardiovascular disease: a systematic review. Am J Epidemiol 2010, 172(5):501-516.

131. Nissinen A, Berrios X, Puska P: Community-based noncommunicable disease interventions: lessons from developed countries for developing ones. Bull World Health Organ 2001, 79(10):963-970.

132. Shea S, Basch CE: A review of five major community-based cardiovascular disease prevention programs. Part II: intervention strategies, evaluation methods, and results. Am J Health Promot 1990, 4(4):279-287.

133. Mittelmark MB, Hunt MK, Heath GW, et al: Realistic outcomes: lessons from community-based research and demonstration programs for the prevention of cardiovascular diseases. J Public Health Policy 1993, 14(4):437-462.

134. Stone EJ, Pearson TA, Fortmann SP, et al: Community-based prevention trials: challenges and directions for public health practice, policy, and research. Ann Epidemiol 1997, 7(Suppl):S113-120.

135. Schooler C, Farquhar JW, Fortmann SP, et al: Synthesis of findings and issues from community prevention trials. Ann Epidemiol 1997, 7(Suppl):S54-68.

136. McLaren L, Ghali LM, Lorenzetti D, et al: Out of context? Translating evidence from the North Karelia project over place and time. Health Educ Res 2007, 22(3):414-424.

137. Pearson TA, Wall S, Lewis C, et al: Dissecting the "black box" of community intervention: lessons from community-wide

58

cardiovascular disease prevention programs in the US and Sweden. Scand J Public Health Suppl 2001, 56:69-78.

138. Pearson TA, Bazzarre TL, Daniels SR, et al: American Heart Association guide for improving cardiovascular health at the community level: a statement for public health practitioners, healthcare providers, and health policy makers from the American Heart Association expert panel on population and prevention science. Circulation 2003, 107(4):645-651.

139. van Lieshout J, Wensing M, Campbell SM, et al: Primary care strength linked to prevention programs for cardiovascular disease. Am J Manag Care 2009, 15(4):255-262.

140. Schmittdiel JA, Shortell SM, Rundall TG, et al: Effect of primary health care orientation on chronic care management. Ann Fam Med 2006, 4(2):117-123.

141. Farzadfar F, Murray CJ, Gakidou E, et al: Effectiveness of diabetes and hypertension management by rural primary health-care workers (Behvarz workers) in Iran: a nationally representative observational study. Lancet 2012, 379(9810):47-54.

142. Weinstein ND: Why it won't happen to me: perceptions of risk factors and susceptibility. Health Psychol 1984, 3(5):431-457.

143. Weinstein ND: Unrealistic optimism about susceptibility to health problems: conclusions from a community-wide sample. J Behav Med 1987, 10(5):481-500.

144. Howitt A, Armstrong D: Implementing evidence based medicine in general practice: audit and qualitative study of antithrombotic treatment for atrial fibrillation. BMJ 1999, 318(7194):1324-1327.

145. van Steenkiste B, van der Weijden T, Timmermans D, et al: Patients’ ideas, fears and expectations of their coronary risk: barriers for primary prevention. Patient Educ Couns 2004, 55(2):301-307.

146. Emmelin M, Weinehall L, Stenlund H, et al: To be seen, confirmed and involved--a ten year follow-up of perceived health and cardiovascular risk factors in a Swedish community intervention programme. BMC Public Health 2007, 7:190.

147. Kehler D, Christensen B, Lauritzen T, et al: Ambivalence related to potential lifestyle changes following preventive cardiovascular consultations in general practice: a qualitative study. BMC Fam Pract 2008, 9:50.

148. Johansson E, Long NH, Diwan VK, et al: Gender and tuberculosis control: perspectives on health seeking behaviour among men and women in Vietnam. Health Policy 2000, 52(1):33-51.

149. Zhang HX, Locke C: Contextualising reproductive rights challenges: The Vietnam situation. Women's Studies International Forum 2002, 25(4):443-453.

59

150. Schuler SR, Hoang TA, Vu SH, et al: Constructions of gender in Vietnam: in pursuit of the 'Three Criteria'. Cult Health Sex 2006, 8(5):383-394.

151. Theobald S, Tolhurst R, Squire SB: Gender, equity: new approaches for effective management of communicable diseases. Trans R Soc Trop Med Hyg 2006, 100(4):299-304.

152. Regitz-Zagrosek V: Therapeutic implications of the gender-specific aspects of cardiovascular disease. Nat Rev Drug Discov 2006, 5(5):425-438.

153. Rosano GM, Collins P: Gender differences in treatment of cardiovascular disease: a task force on gender of the ESC proposal on gender specific studies in cardiovascular pharmacology. Fundam Clin Pharmacol 2010, 24(6):662-663.

154. McSweeney JC, Pettey CM, Souder E, et al: Disparities in women's cardiovascular health. J Obstet Gynecol Neonatal Nurs 2011, 40(3):362-371.

155. Mosca L, Barrett-Connor E, Wenger NK: Sex/gender differences in cardiovascular disease prevention: what a difference a decade makes. Circulation 2011, 124(19):2145-2154.

156. Blane D, Hart CL, Smith GD, et al: Association of cardiovascular disease risk factors with socioeconomic position during childhood and during adulthood. BMJ 1996, 313(7070):1434-1438.

157. Kittleson MM, Meoni LA, Wang NY, et al: Association of childhood socioeconomic status with subsequent coronary heart disease in physicians. Arch Intern Med 2006, 166(21):2356-2361.

158. Kivimäki M, Smith GD, Juonala M, et al: Socioeconomic position in childhood and adult cardiovascular risk factors, vascular structure, and function: Cardiovascular risk in young Finns study. Heart 2006, 92(4):474-480.

159. Senese LC, Almeida ND, Fath AK, et al: Associations between childhood socioeconomic position and adulthood obesity. Epidemiol Rev 2009, 31:21-51.

160. Lee DS, Chiu M, Manuel DG, et al: Trends in risk factors for cardiovascular disease in Canada: temporal, socio-demographic and geographic factors. CMAJ 2009, 181(3-4):E55-66.

161. Yusuf HR, Giles WH, Croft JB, et al: Impact of multiple risk factor profiles on determining cardiovascular disease risk. Prev Med 1998, 27(1):1-9.

162. Lowe LP, Greenland P, Ruth KJ, et al: Impact of major cardiovascular disease risk factors, particularly in combination, on 22-year mortality in women and men. Arch Intern Med 1998, 158(18):2007-2014.

163. Daviglus ML, Liu K, Pirzada A, et al: Favorable cardiovascular risk profile in middle age and health-related quality of life in older age. Arch Intern Med 2003, 163(20):2460-2468.

60

164. Asia Pacific Cohort Studies Collaboration: Joint effects of systolic blood pressure and serum cholesterol on cardiovascular disease in the Asia Pacific region. Circulation 2005, 112(22):3384-3390.

165. Chiu JF, Bell AD, Herman RJ, et al: Cardiovascular risk profiles and outcomes of Chinese living inside and outside China. Eur J Cardiovasc Prev Rehabil 2010, 17(6):668-675.

166. Tamakoshi A, Tamakoshi K, Lin Y et al: Healthy lifestyle and preventable death: findings from the Japan Collaborative Cohort (JACC) study. Prev Med 2009, 48(5):486-492.

167. Kvaavik E, Batty GD, Ursin G, et al: Influence of individual and combined health behaviors on total and cause-specific mortality in men and women: the United Kingdom health and lifestyle survey. Arch Intern Med 2010, 170(8):711-718.

168. McCullough ML, Patel AV, Kushi LH et al: Following cancer prevention guidelines reduces risk of cancer, cardiovascular disease, and all-cause mortality. Cancer Epidemiol Biomarkers Prev 2011, 20(6):1089-1097.

169. Spring B, Moller AC, Coons MJ: Multiple health behaviours: overview and implications. J Public Health (Oxf) 2012, 34 Suppl 1:i3-i10.

170. Kones R: Is prevention a fantasy, or the future of medicine? A panoramic view of recent data, status, and direction in cardiovascular prevention. Ther Adv Cardiovasc Dis 2011, 5(1):61-81.

171. Russell LB: Preventing chronic disease: an important investment, but don't count on cost savings. Health Aff (Millwood) 2009, 28(1):42-45.

172. Trust for America's Health: Prevention for a healthier America: investments in disease prevention yields significant savings, stronger communities. In. Washington: Trust for America's Health; 2008.

173. Folsom AR, Yatsuya H, Nettleton JA, et al: Community prevalence of ideal cardiovascular health, by the American Heart Association definition, and relationship with cardiovascular disease incidence. J Am Coll Cardiol 2011, 57(16):1690-1696.

174. Anderson P, Chisholm D, Fuhr DC: Effectiveness and cost-effectiveness of policies and programmes to reduce the harm caused by alcohol. Lancet 2009, 373(9682):2234-2246.

175. Ha DA, Chisholm D: Cost-effectiveness analysis of interventions to prevent cardiovascular disease in Vietnam. Health Policy Plan 2011, 26(3):210-222.

176. Moher M, Hey K, Lancaster T: Workplace interventions for smoking cessation. Cochrane Database Syst Rev 2005(2):CD003440.

177. Secker-Walker RH, Gnich W, Platt S, et al: Community interventions for reducing smoking among adults. Cochrane Database Syst Rev 2002(3):CD001745.

61

178. Ashraf A, Quaiyum MA, Ng N et al: Self-reported use of tobacco products in nine rural INDEPTH Health and Demographic Surveillance Systems in Asia. Glob Health Action 2009, 2.

179. Ng N, Hakimi M, Van Minh H, et al: Prevalence of physical inactivity in nine rural INDEPTH Health and Demographic Surveillance Systems in five Asian countries. Glob Health Action 2009, 2.

180. Ng N, Weinehall L, Ohman A: 'If I don't smoke, I'm not a real man'--Indonesian teenage boys' views about smoking. Health Educ Res 2007, 22(6):794-804.

181. Donny EC, Caggiula AR, Weaver MT, et al: The reinforcement-enhancing effects of nicotine: implications for the relationship between smoking, eating and weight. Physiol Behav 2011, 104(1):143-148.

182. Nishtar S: Prevention of coronary heart disease in south Asia. Lancet 2002, 360(9338):1015-1018.

183. Weintraub WS, Daniels SR, Burke LE, et al: Value of primordial and primary prevention for cardiovascular disease: a policy statement from the American Heart Association. Circulation 2011, 124(8):967-990.

184. Stampfer MJ, Hu FB, Manson JE, et al: Primary prevention of coronary heart disease in women through diet and lifestyle. N Engl J Med 2000, 343(1):16-22.

185. Kurth T, Moore SC, Gaziano JM, et al: Healthy lifestyle and the risk of stroke in women. Arch Intern Med 2006, 166(13):1403-1409.

186. Chiuve SE, Rexrode KM, Spiegelman D, et al: Primary prevention of stroke by healthy lifestyle. Circulation 2008, 118(9):947-954.

187. Lindholm L, Rosén M: What is the "golden standard" for assessing population-based interventions?--problems of dilution bias. J Epidemiol Community Health 2000, 54(8):617-622.

188. Spring B, Pagoto S, Pingitore R, et al: Randomized controlled trial for behavioral smoking and weight control treatment: effect of concurrent versus sequential intervention. J Consult Clin Psychol 2004, 72(5):785-796.

189. Spring B, Schneider K, McFadden HG et al: Make better choices (MBC): study design of a randomized controlled trial testing optimal technology-supported change in multiple diet and physical activity risk behaviors. BMC Public Health 2010, 10:586.

190. Bovet P, Gervasoni JP, Ross AG et al: Assessing the prevalence of hypertension in populations: are we doing it right? J Hypertens 2003, 21(3):509-517.

191. Levin KA: Study design III: Cross-sectional studies. Evid Based Dent 2006, 7(1):24-25.

62