Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 44
5 Priorities for Global Research and Development CONTEXT As developing countries undergo the epidemiologic transition, cardiovascular disease (CVD) epidemics are emerging or accelerating (Murray and Lopez, 1997a; Reddy and Yusuf, 1998; WHO, 1996, 1997). Depending on the level of development, the pace of the transition varies, although the general process and direction is similar for all countries. The challenge of the epidemiologic transition, therefore, is not whether it will happen in developing countries, but whether it is possible to traverse quickly from the early stage of nutritional and infectious disorders that affect the young to the later stage of noncommunicable diseases (NCDs) that affect mainly the elderly, thus averting or abbreviating the midstage of NCD epidemics when there is a major impact on individuals in their productive middle years. Knowledge of CVD and other NCDs, gained from research conducted primarily in developed countries, can help speed the transition. Local research is required, however, to adapt current knowledge to specific developing country situations. Policies related to CVD control, as well as priorities for research, must address the emerging epidemic from a long-term perspective. Increases in CVD risk factors precede the CVD epidemic so that strategies for CVD control must focus on the prevention of risk factors as well as their reduction. Control of risk factors may involve: prevention of risk factor development or progression in communities and age groups currently at low risk (primordial prevention); recognition and reduction of high risk in populations that have already acquired an adverse risk profile (primary prevention); and risk factor modification to minimize further complications in those who have clinical disease (secondary prevention). In addition, cost-effective clinical care is needed to improve both the
OCR for page 45
survival and the quality of life of persons who have developed CVD. Appropriately focused research will be necessary to meet each of these goals. This chapter discusses key areas of research for effective CVD prevention and control in developing countries. Recommendations are presented in bold type both here and in the Executive Summary. RESEARCH PRIORITIES There are four criteria to be used in setting priorities for investing in CVD R&D in developing countries: Investments should have a large-scale impact on populations that include men and women, all socioeconomic groups, and various regions of a country. Incremental implementation may be necessary in many countries. Investments in a country should involve methods and processes (but not necessarily results) that are broadly transferable to other low- and middle-income countries. Investments should yield results within a time frame of 5-10 years, although evaluation may be desirable over a longer term. Investments should focus on measurable data that use, for the most part, established epidemiologic, health policy, economic, and social behavioral methodologies. The main categories of research activity and key recommendations within these categories follow. Improve Knowledge of the Size of the CVD Burden in Developing Countries Vital Registration A sample registration system collects vital statistics on a sample of the population in each state. Such systems are incomplete or nonexistent for many developing countries. About 18 percent of global CVD mortality occurs in China, where the sample registration covers only about 10 percent of the population or about 100 million people in rural and urban settings. About 17 percent of global CVD mortality occurs in India, which has no complete registration system. Registration of vital statistics and cause of death is generally lacking for Africa, Middle Eastern countries, and Asian islands (Murray and Lopez, 1997b). There are several options for vital registration, the key for each being complete coverage of all deaths, including those from CVD, classified at least by age and sex. Priority assistance should be given to countries that currently lack vital
OCR for page 46
registration systems. Appropriate sampling frames should be built into each system to allow identification of geographic, ethnic, and rural and urban differences in CVD mortality and morbidity. In developing countries, most vital registries are government funded and lack a research component. These need to develop improved methods for vital registration. In countries with inadequate registration systems, this may include consideration of the capture-recapture method for deriving better estimates of total deaths. Cause-of-Death Statistics These provide information on the underlying reasons for death. They are available for developed countries and, to a lesser extent, for Latin America. In China, cause-of-death data are derived from a systematic follow-up of one million deaths to determine their causes and from the district surveillance points system (DSP), which includes about 145 communities in urban and rural settings. In India, cause-of-death data are collected through verbal autopsies on 0.5 percent of rural deaths (Murray and Lopez, 1997b). Both the Chinese and the Indian systems suffer from methodological problems. In China, death rates may be underreported by as much as 30 percent, although unofficial estimates of 8 percent in urban areas and 15 percent in rural areas have been suggested (unpublished data). The Indian system has up to 25 percent of deaths classified as senile or ill-defined. These deaths are likely to include ischemic heart disease, stroke, or other CVD deaths, which would lead to an underestimation of CVD mortality. Ascertaining cause-specific mortality could be strengthened through the use of community-based random samples, sentinel sites, and surveillance systems modeled after those used in the MONICA study, as well as verbal autopsy techniques to determine the cause of death of adults (especially CVD-related deaths). Data from Tanzania suggest that most deaths occur at home (United Republic of Tanzania, 1997), so including these will require a special effort. Establishing and evaluating sentinel registration sites, along with validating and using verbal autopsy techniques, are also priority areas for research support. There is a need to assess the utility of cause-of-death statistics available through the extensive missionary health care network throughout Asia and Africa. Umbrella organizations, such as the Christian Medical and Dental Society, which has more than 400 physician members, have expressed interest in cooperating in this effort. Church groups account for 36 percent of health care in Kenya and more than 50 percent in Cameroon.
OCR for page 47
Disability Estimates Most developing countries lack culturally relevant estimates of disability due to CVD and other causes. Estimates of morbidity are not standardized across populations. Focused pilot studies could generate such estimates through validation of a standard quality-of-life measure. To improve knowledge of the size of the CVD burden in developing countries, the committee recommends the following: creating standardized surveys using networks such as the MONICA model (including selected sentinel sites) to monitor cardiovascular mortality levels and trends of clinical events, and the CINDI and CARMEN models to monitor cardiovascular risk factors; expanding national and regional systems for vital registration; improving the accuracy and completeness of cause-of-death statistics; and developing better estimates of disability. Establishing the Levels, Determinants, and Consequences of Risk Factors Cross-Sectional Surveys A major goal of cross-sectional surveys of risk factors is to describe the prevalence and distribution of risk factors, by age, sex, and ethnicity, in representative samples of the population. Such studies of CVD would require moderately large sample sizes (several tens of thousands or hundreds of thousands) and focused, simple data collection. Like vital registration, cross-sectional surveys require a valid sampling frame. Ideally, the data from local cross-sectional studies can be linked to local estimates of the magnitude of various risk factors that are drawn from case-control or prospective studies. Survey data can also be linked to local mortality data. Cross-sectional surveys are repeated at regular intervals to assess trends in the levels and distribution of risk factors. These surveys could also quantify the association of a CVD risk factor with disease. For this purpose the surveys would be designed to repeat observations of many individuals and to take independent samples during successive surveys. Follow-up of patients in many developing country populations is difficult due to the lack of vital registration systems, the mobility of the population, and variable access to medical care. In these populations it may be easier to classify mortality status than to record non-
OCR for page 48
fatal CVD events. Cross-sectional studies of CVD would have to be sufficiently large to permit follow-up for mortality outcomes. Analytic Studies Retrospective case-control or prospective studies can provide estimates of the magnitude of disease risk associated with various CVD risk factors. Such studies should focus on major conditions such as myocardial infarction and on established risk factors such as tobacco use, high blood pressure, high serum cholesterol, diabetes-related syndromes, physical inactivity, and obesity. There is also a need for biochemical studies of risk factors such as apolipoproteins, fibrinogen and other clotting factors, albumin, homocysteine, folate, dietary antioxidants, and infectious agents such as Chlamydia, Helicobacter, and cytomegalovirus. Such studies, conducted selectively, could confirm the role of CVD and its risk factors in previously underinvestigated populations. Case-control studies of the incidence of disease can identify the strength of association of a risk factor with CVD and may also uncover new risk factors. Although prospective studies are more robust methodologically because exposure to a risk factor demonstrably precedes disease, retrospective case-control studies can usually generate needed data more quickly and at lower cost. Ideally, cases of disease incidence (e.g., initial myocardial infarction) should be studied to avoid biases in the recording of risk factors. Key conditions for study by case-control methods are acute myocardial infarction, acute stroke, transient ischemic attacks, congestive heart failure, and peripheral vascular disease. Prospective studies have the advantage of measuring exposure to a risk factor prior to the development of disease. By necessity, these studies must be large-with at least 100,000 to 200,000 middle-age individuals-if they are to reliably measure end points in multiple population groups, within a 5- to 10-year period. These studies should focus on questionnaires and physical measurements. Biological sample collection may be included. Because prospective studies require rigorous follow-up over a number of years, they are best done when the required infrastructure and long-term research units are in place. Systematic review of past epidemiologic studies of CVD in developing countries may provide information useful for policy planning. Methodological advances in meta-analysis of observational studies may help develop useful data on the age-specific prevalence of CVD deaths and risk factors and on their trends. Since most of the earlier studies are small and not often found in computerized databases such as Medline, efforts to collate previous studies would rely on a combination of journal searches and contact with researchers. These systematic reviews could be a useful way to organize networks of researchers. Such approaches could also be applied to evaluations of past randomized trials, although few of these have been done in developing countries. Experience with
OCR for page 49
meta-analysis of randomized controlled trials suggests that international collaborative efforts using original patient data provide more valid results. Finally, judicious use of hospital admission and discharge data could provide limited, but useful, information on the age, sex, and other characteristics of CVD cases. After evaluating the respective strengths and limitations of the different analytic studies described above, the committee recommends that immediate support be directed to determining the morbidity and mortality of conventional and new risk factors for CVD and assessing their interactions through the use of incident case-control studies. Public Health Interventions Public health interventions are directed at entire populations or subpopulations. Appropriate research for these may range from randomized trials of the efficacy of treatment in individuals to quasi-random or other systematic evaluations at the community level to assess the effectiveness of interventions in actual-use situations. Randomized Trials Randomized trials are the best method for establishing the efficacy and magnitude of the impact of intervention on CVD control. Evidence from such trials has led to dramatic changes in the use of medications and the practice of cardiology in developed countries. Randomized trials studying salt reduction through dietary guidance or substitution of potassium may be useful in populations with a high mortality from stroke. The proportion of hemorrhagic stroke is higher in developing country populations (Reddy and Yusuf, 1998). Since these events are known to be sensitive to blood pressure, even moderately successful interventions to reduce blood pressure could have a significant impact on CVD incidence. Alternative applications of randomized trials would be to randomly assign clinicians or clinical sites to a package of interventions or an algorithm for CVD care (see below) and compare these with usual practice. Such trials could yield results that are generalizable and would be made more acceptable by the fact that they derive from studies directly involving the clinical community. Another possibility is to increase participation from developing countries in international trials of low-cost treatments.
OCR for page 50
Clinical Interventions Essential Vascular Package A wealth of evidence from randomized trials indicates that several clinical treatments can provide cost-effective care for established vascular patients. Aspirin, beta-blockers, ACE (angiotensin-converting enzyme) inhibitors, and cholesterol-lowering statin drugs reduce the probability of death and subsequent nonfatal, major vascular events in patients with established ischemic heart disease. However, to have a sustained effect in developing countries, these drug packages must be low cost and widely accessible. Although cholesterol-lowering drugs such as statins are currently expensive compared to aspirin or beta-blockers, their costs may decrease in the next 5 years when their patents expire. These would then be key components of the essential vascular package, or EVP. The committee considers that such an EVP could benefit large numbers of people in developing countries. Because its delivery relies on self-presentation it has no screening costs and could be highly cost-effective. Packaging EVPs into single, daily formulations could substantially improve compliance, as it did for treatment of tuberculosis. The goals of the EVP should be (1) to select low-cost, generic versions of these drugs; (2) to achieve near-universal access; and (3) to price these packages so they are affordable for developing world clinics and patients. The EVPs should first be assessed for acceptability, use, and outcome to ensure their successful adoption. Such testing could be in the form of randomized trials that assess the EVP versus standard clinical care. If acceptable, the EVP should be included in a publicly financed, universally available package of clinical services and on insurance treatment lists. Efforts to educate physicians and increase patient awareness about the value of such packages should also be supported. The second and third goals of the EVP can be addressed by pricing reimbursements of treatments to the lowest-cost basis of the EVP. In summary, the committee recommends that research be directed to the following: evaluating the responses of different ethnic populations to cardiovascular drugs and interventions, and their implications for rational drug treatment. expanding the participation of developing country centers in multicenter collaborative clinical trials of EVPs and other potentially affordable, widely applicable interventions.
OCR for page 51
Algorithms for Effective Diagnosis and Clinical Management Algorithms for the clinical care of CVD could improve the awareness and use of effective treatments. To be maximally effective, they should be adapted to different cultural needs and be widely applicable and usable by nonphysicians and physicians for various levels of care. Each algorithm should define clinical diagnostic or presumptive criteria, along with the steps for administering and evaluating simple medical treatments. The algorithms should, at a minimum, address the following clinical problems: treatment of acute coronary ischemia, acute stroke, and transient ischemic attack, chronic coronary ischemia, chronic peripheral vascular disease, and congestive heart failure; management of low-cost, home-based rehabilitation following a stroke or myocardial infarction; and guidelines on the detection and treatment of elevated blood pressure and on the screening and treatment of elevated cholesterol. For each algorithm, it will be important to measure its acceptability and use, to determine how to market it to the public and private sectors, and to cooperate with the pharmaceutical industry in this process. Algorithms should be developed and implemented according to the local characteristics of disease, cultural norms, and health service settings. Monitoring the effects of algorithms on disease outcome is essential. In summary, the committee recommends that research be undertaken to evaluate algorithms for the clinical diagnosis and affordable management of the following: (1) hypertension, (2) dyslipidemia, (3) diabetes, (4) acute myocardial infarction, (5) angina, (6) stroke, (7) transient ischemic attacks, (8) congestive heart failure, (9) peripheral vascular disease, (10) post-myocardial infarction rehabilitation and risk management, and (11) poststroke rehabilitation and risk management. Randomized Trials of Other Low-Cost Treatments There is growing evidence that homocysteine is a risk factor for coronary heart disease and that low-cost folate supplementation may reduce homocysteine levels and thus the risk of heart disease. Randomized trials assessing the cardiovascular benefit of folate supplementation are needed, and several trials are being planned or are under way in developed country populations. Since folate intake and blood levels among adults in developing countries are believed to be
OCR for page 52
lower than those in industrial countries, randomized trials of this intervention should also be considered for developing country populations. Inclusion of Developing Countries in International Collaborative Clinical Trials The evaluation of lifesaving technologies through randomized clinical trials often involves collaboration among investigators in several countries with diverse populations. To date, however, these trials have been conducted primarily in developed countries. The inclusion of populations and scientists from developing countries in future trials will help to (1) extend the generalizability of trial results; (2) understand different ethnic responses to cardiovascular drugs and other interventions; (3) compare the cost-effectiveness of interventions in developing country populations; and (4) enhance research capacity. Other Public Health Interventions Many other issues of program effectiveness can be addressed through systematic demonstration studies designed to yield rigorous results through standardized evaluation and serial implementation in successive communities with comparable pre- and postevaluation strategies. Such approaches can address effects at the community (rather than the individual) level where random allocation of larger numbers of observational units may not be feasible, affordable, or acceptable. RESEARCH TO CONTROL SPECIFIC RISK FACTORS Research on CVD control should focus on the risk factors that contribute most to the current and projected burden of disease in developing countries. Such research has several goals: periodically estimate the rates or levels of exposure to a risk factor in the population; develop and validate cost-effective methods for identifying exposed individuals at high risk of CVD; assess the capacity of health care delivery systems to implement programs for the detection and control of these risk factors; and evaluate the cost-effectiveness of programs aimed at: reducing the population-wide risk, the risk to persons already at high risk, and the risk to persons who have clinically manifest CVD.
OCR for page 53
Attention should be given to selecting the risk factors for early intervention. These would include risk factors that contribute to more than one CVD or NCD. Knowledge of population-specific conditions can make such public health actions more effective. The Global Burden of Disease Study identifies tobacco use and hypertension among the major risk factors contributing to the present and projected mortality and disability from NCDs (Murray and Lopez, 1997b). Interventions aimed at reducing the levels of tobacco use and hypertension in developing country populations deserve high priority. Tobacco Use Nearly three-quarters of the more than 1 billion people who regularly use tobacco live in developing countries (Jha, 1997). By the year 2030, tobacco is predicted to kill approximately 10 million people annually in developing countries, which is triple the current mortality caused by tobacco (Peto, 1997). In most developed countries, the epidemiologic transition described in Chapter 2 has been followed by a behavioral transition to unhealthy behaviors, including smoking. Because this transition has not occurred yet or has begun recently in some developing countries, it is still possible to stop the tobacco epidemic at an early stage. As noted by the Institute of Medicine (IOM, 1998), ''The success of tobacco control efforts in developed countries has largely been due to the cultivation of a receptive social and political climate through the availability of information about the real risks of tobacco use, supported by research on appropriate pricing and regulation." There is a similar need for developing countries to establish knowledge on the extent of tobacco-related CVD mortality; the determinants of tobacco use, including the impact of advertising, promotion, and price; the costs of tobacco use; and cost-effective strategies for reducing tobacco use. Local knowledge will allow for political, social, and cultural differences among developing country populations, while striving for comparability across studies. Research on tobacco control can help reduce the substantial burden of disease projected for several tobacco-related diseases, including certain cancers and chronic obstructive pulmonary disease as well as CVD (see Chapters 1 and 2; CDC, 1990, 1994; WHO, 1997). Thus, research on tobacco control in developing countries should be undertaken to (1) estimate the prevalence of regular tobacco use in population samples; (2) monitor tobacco consumption trends in especially vulnerable groups, such as children, adolescents, and women; (3) evaluate the cost-effectiveness of community-based intervention programs that promote abstinence from tobacco; (4) evaluate the cost-effectiveness of tobacco cessation programs aimed
OCR for page 54
at changing the behavior of current smokers; and (5) evaluate the economic impact of tobacco control on developing countries that grow, manufacture, or export tobacco products in order to encourage alternative crops. Hypertension High blood pressure is widely prevalent in developing countries and a major contributor to coronary heart disease and to hemorrhagic and thrombotic stroke. More than twice as many deaths from stroke occur in the developing countries as in the developed countries (see Chapters 1 and 2). Even small changes in the population distribution of blood pressure can have a profound impact on CVD rates. In high-risk persons who have developed clinical hypertension, the benefits of blood pressure reduction through drug therapy are substantial. Evidence from clinical trials demonstrates that a 5-6 mm Hg reduction in diastolic blood pressure will reduce stroke death by 35-40 percent and coronary deaths by 15-20 percent (WHO, 1996a). Hypertension control programs are an effective starting point for CVD prevention and control for the following reasons: hypertension is a risk factor for both coronary heart disease and stroke; such programs have a "clinical" appeal to both the care providers and the community; the goals are easily measurable; the impact on hypertension awareness, treatment status, and level of control can be measured in a relatively short time (five years); the programs provide a natural coalition of various categories of health care providers, each with an important role in the detection or management of hypertension and its sequelae; and the concept of "comprehensive cardiovascular reduction" as a part of hypertension management makes it possible to incorporate strategies aimed at modifying other CVD risk factors such as tobacco use, high blood lipids, diabetes, and obesity. Thus, the committee recommends that research on hypertension control in developing countries be undertaken to: estimate the distribution of elevated blood pressure and prevalence of hypertension in population samples; evaluate the cost-effectiveness of community-based, life-style-linked interventions such as improved nutrition and exercise, and reduced smoking to decrease the incidence of high blood pressure;
OCR for page 55
assess the cost-effectiveness of programs to detect and treat hypertension by improving awareness, treatment initiation and adherence, and level of control; and evaluate the effectiveness of low-cost combination drug therapies developed by countries such as China. Diet and Physical Activity Profound demographic changes are increasing the proportion of populations in middle- and older-age groups where adult CVD becomes common. In addition, as cultures adapt to become more "Western," diet and physical activity are adversely affected. Global R&D is needed to identify successful ways to "delink" social and economic development from adverse changes in diet and physical activity. These changes will determine in large part the course of the CVD epidemic (Labarthe, 1998). HEALTH SERVICES RESEARCH Review of Patterns of Clinical Practice A systematic review of medical practice patterns should identify effective ways of diagnosing CVD and of treating it effectively. These reviews could be conducted in hospital settings in several countries, with standardized diagnostic methods. The CVD-related conditions to be included are acute coronary ischemia, acute stroke and transient ischemia, chronic coronary ischemia, chronic peripheral vascular disease, and congestive heart failure. Research on CVD control should include consideration of the influence of professional training, prescribing behavior, and payment systems, such as prepayment versus fee for service on the treatments recommended. This would also review the influences on government decisions to fund hospital-based curative services versus population-based CVD prevention. The Economic Burden and Cost-Effectiveness of Interventions There are few data on the economic costs of CVD in developing countries. However, the rapid escalation in costs can be anticipated from the emerging epidemic of CVD (see Chapter 2). This could be a powerful motivation for policymakers to implement preventive CVD strategies. Studies addressing the economic burden caused by. CVD in developing countries should focus on the effect of CVD on production, earnings, and household health. These studies will have to standardize definitions of cost across populations and among different health
OCR for page 56
care systems. These studies could identify the costs borne by income group, sex, and rural or urban status. There are limited data from developing countries on the effectiveness of CVD expenditures. Preliminary estimates from India suggest that population-based risk factor reduction and low-cost secondary treatments are as cost-effective as other interventions included in minimal essential packages of care (Jha et al., forthcoming). These studies should be incorporated into broader studies of the cost-effectiveness of medical care. GOVERNMENT'S POLICY ROLE There is a strong need for governments to review the policies guiding specific health interventions and to encourage evaluation of products and tools for their effectiveness. The role of government is to regulate, legislate, and provide information. Tobacco Control Policy The key elements of effective tobacco control are (1) setting high prices; (2) displaying serious and prominent health warnings on all tobacco products; (3) banning advertising and promotion of all tobacco-associated products or trademarks; (4) using cost-effective mass media to counter product advertising; (5) supporting research on the causes, consequences, and costs of smoking; (6) controlling smuggling; and (7) developing the capacity to monitor the burden of disease caused by tobacco and to lobby for tobacco control (IOM, 1998; Jha et al., forthcoming). These policies were effective in reducing tobacco consumption by more than 40 percent in developed countries from 1960 to 1990 (Peto et al., 1994; World Bank, 1997b). Developing countries have not instituted effective tobacco controls and are projected to have a rapid increase in smoking. The problem is likely to be greatest in China, East Asia, South Africa, and the former Soviet states (Peto et al., 1994). National programs are needed to reduce current and future tobacco use. Food and Nutrition Policy The establishment of food and agriculture policies and nutrition goals has helped control consumption in developed countries of diets high in fat, sugar, and salt (Carleton et al., 1991; Grundy et al., 1997). These approaches can be effective in developing countries as well. The production, processing, packaging, labeling, and marketing of food items, as well as subsidies for their production, have to be assessed for their influence on dietary intakes. These goals should be
OCR for page 57
coordinated with broad public health goals so that, for example, reducing salt consumption does not undermine programs for iodine supplementation through the use of iodized salt. Pharmaceutical Policy Provision by the health care system of low-cost pharmaceuticals for CVD treatment and the promotion of their rational use by health care providers require coordination of policies and practices in the manufacture, regulation, and pricing of pharmaceuticals, essential drug lists, and the influence of the pharmaceutical industry on prescribing practices. Private-Sector Participation The private sector can participate in population-based prevention programs, as well as in cost-effective case management programs. Partnerships with the private sector may also attract research funding to evaluate population-based interventions for risk factor modification and those testing clinical care algorithms. BUILDING CAPACITY FOR RESEARCH AND DEVELOPMENT To undertake research for the control of CVD in developing countries requires a strengthening of existing capacities for conducting research. This will require training investigators in (1) population-based epidemiologic research, (2) clinical research, (3) health policy research, and (4) economic evaluation of health care interventions. It will also require strengthening linkages among basic, epidemiologic, clinical, and policy research areas so that CVD can be targeted with a problem-solving, multidisciplinary approach. There are programs that provide such training. The Scientific Council on Epidemiology and Prevention of the International Society and Federation of Cardiology (ISFC) conducts an annual seminar on CVD epidemiology and prevention for 30-35 international fellows drawn from diverse health disciplines. About one-third to one-half of these fellows are from developing countries. The International Clinical Epidemiology Network (INCLEN) has a training program for clinical researchers, social scientists, statisticians, and health economists from selected academic institutions in developing countries. Although it is not specifically directed to CVD-related research, this program has augmented the capacity for health research in developing countries through multidisciplinary and international collaboration. There are also formal training programs for epidemiologists and clinicians in various academic institutions are accessible
OCR for page 58
through fellowship programs sponsored by governments, the World Health Organization (WHO), and international aid agencies. More trained investigators are needed for CVD-relevant research. The unfinished health agenda of the developing world and the emergence of the HIV-AIDS epidemic have restricted the number of epidemiologists available to investigate CVD epidemiology. Nutritional research on NCDs has not engaged attention due to continuing and justified concerns about deficiency disorders. Clinical investigators have been more preoccupied with applications of high-technology interventions than with testing widely needed, low-cost, clinical algorithms for diagnosis and management of CVD. Hence, there is a growing need to train individuals and equip institutions to undertake research relevant to CVD control. Regional training programs, modeled after the ISFC seminars and the INCLEN program, could be established in developing countries to focus on the control of CVD and other NCDs. Institutional capacity for conducting integrated, problem-oriented research could be strengthened with training and equipment for institutions that have the potential to undertake epidemiologic, clinical, and policy research. Much of the research methodology already developed, tested, and applied can be transferred to developing countries. Agencies such as WHO can facilitate this process, both as a central repository of information and by providing subsidized pharmaceuticals. There are mutual benefits to be derived from twin-center programs between developed and developing countries. Regional research networks can effectively link institutions within and among developing countries to leverage scientific expertise and financial resources. In summary, the committee recommends establishing or expanding the following capacity-strengthening capabilities in developing countries: (1) training programs in cardiovascular epidemiology, clinical research methodology, health policy research, and health economics; (2) the institutional capacity for undertaking integrated research relevant to CVD control; and (3) channels of collaboration through twin-center programs and regional research networks.
Representative terms from entire chapter: