P r e v e n t i o n
The Prevention Of Cardiovascular Disease: Have We Really Made Progress? A balance of community and medical approaches holds the most promise for preventing CVD. by Thomas A. Pearson ABSTRACT: Despite reductions in cardiovascular disease (CVD) mortality, current evidence suggests that CVD is not being prevented but, rather, is being made less lethal. Evidencebased guidelines have been developed for secondary, primary, and community-based prevention. To improve compliance with secondary prevention guidelines, programs must better organize and monitor care. Primary prevention requires assessment of risk in asymptomatic people, to yield cost-effective benefits. CVD prevention at the societal level should target deleterious behavior in community settings, using effective public health interventions. Policy options that involve multiple preventive approaches offer the best opportunity to minimize the economic and social burdens of CVD. [Health Affairs 26, no. 1 (2007): 49– 60; 10.1377/hlthaff.26.1.49]
C
a r d i o va s c u l a r d i s e a s e (CVD) is largely preventable, based on several lines of evidence. First, heart disease and stroke mortality rates vary greatly among countries.1 Second, several studies of people without established CVD risk factors (cigarette smoking, diabetes mellitus, elevated blood pressure, elevated blood cholesterol, and so forth) demonstrate exceptionally low rates of CVD incidence.2 Third, studies of people without deleterious health behavior (tobacco use, sedentary lifestyle, high saturated fat and cholesterol in the diet, and excess body weight) suggest that a large proportion of CVD incidence could be prevented by lifestyle modifications alone.3 Therefore, nihilistic assumptions of the inevitability of CVD development in individuals or societies are unwarranted. The intent of this paper is to describe the various approaches to CVD prevention, examine the extent to which they have been implemented, and examine policy approaches to improving that implementation.
Thomas Pearson (
[email protected]) is the Albert D. Kaiser Professor and Chair, Department of Community and Preventive Medicine, and senior associate dean for clinical research at the University of Rochester School of Medicine and Dentistry in New York. He currently directs the Rochester Prevention Research Center and the Cardiology Clinic at the University of Rochester Medical Center.
H E A L T H A F F A I R S ~ Vo l u m e 2 6 , N u m b e r 1 DOI 10.1377/hlthaff.26.1.49 ©2007 Project HOPE–The People-to-People Health Foundation, Inc.
49
Tr e n d s
Are We Really Preventing CVD? The reductions in mortality from CVD have been appropriately lauded as one of the major health accomplishments of the twentieth century. However, these promising results have been tempered by the large annual reductions in ageadjusted CVD mortality rates slowing to 1.5 percent per year and stroke mortality not declining altogether.4 To understand mortality rates, one must know two parameters: the incidence (rate of new cases) and their case-fatality rates (percentage of cases dying of the disease). Dramatic declines in case-fatality rates have been documented over the past thirty-five years. However, at least two welldesigned, population-based studies found no change in the incidence of acute myocardial infarction (AMI)—in Worchester, Massachusetts, and in Olmstead County, Minnesota—from 1990 onward.5 The recent decline in mortality has been mostly attributed to improvements in acute cardiac care and secondary prevention after CVD onset.6 In this case, policy development has been impaired by limited surveillance data, which highlights the need for improved systems of surveillance for incidence and prevalence of CVD. A stable incidence in the setting of an improving case-fatality rate can translate into only one thing: an increasing prevalence, as people previously succumbing to CVD now survive the acute presentation of the disease. These people then return to their communities as CVD cases. In 2000 the American College of Cardiology (ACC) estimated that the population of 12.5 million Americans with the diagnosis of heart disease in that year would double in size by 2050, on the basis of the aging of the population.7 Continued case-fatality reductions would further accelerate this rise in prevalence. The implications for health care costs and for public health are enormous. Some of these survivors will have sequelae of their disease— namely, disability and reduced quality of life. Almost all will require numerous medications, increased care by medical specialists, recurrent testing, and, for some, recurrent hospitalizations and invasive procedures.
Approaches To CVD Prevention: Secondary, Primary, And Primordial n Definitions of types of prevention. Secondary prevention has been classically defined as the prevention of disease recurrence and death after the onset of symptomatic disease. Because the strongest predictor for death or a recurrent CVD event is having a history of CVD, patients with such a history are all considered to be at very high risk. Primary prevention traditionally has been the prevention of the onset of symptomatic disease through the treatment of risk factors for CVD, such as treating hypertension to prevent stroke. Primordial prevention describes efforts to reduce the onset of the risk factors known to predispose people to CVD. For example, lifestyle modifications to maintain ideal body weight and to limit sodium consumption are means of preventing the development of high blood pressure. n Blurring of distinctions between types of prevention. The secondary, pri-
50
Januar y/ Fe br uar y 2007
P r e v e n t i o n
mary, and primordial prevention distinctions are of decreasing usefulness. The advent of a variety of imaging and noninvasive testing modalities have demonstrated the ability to identify asymptomatic people with apparently established CVD.8 Tests such as exercise electrocardiography, coronary calcium scoring using computed tomography (CT) scans, ankle-brachial blood pressure indexes, and so forth provide evidence for the asymptomatic presence of atherosclerotic disease. Whether this early identification of disease will lead to prevention of symptoms is not frequently investigated, which makes it difficult for policymakers to accept or prioritize the use of these new technologies. Epidemiologic studies such as the Framingham Heart Study have provided support for the identification of high-risk people using risk-factor profiles.9 In the instance of diabetes mellitus, history of a stroke, or symptomatic peripheral arterial disease, the risk is considered so high that these patients are considered to have a “coronary heart disease equivalent,” with a risk similar to a myocardial infarction (MI) survivor.10 In addition, the Framingham study developed equations to estimate absolute risk, on the basis of the level and number of risk factors, to quantify individuals’ risks over the next ten years for death from MI or coronary disease.11 Using these tools, today’s approach is to stratify people on the basis of risk. This may be viewed from a cost-effectiveness perspective, in which the total cost of a prevention program would be minimized by limiting costly and side effect–prone treatments to people at high enough risk to prevent adequate numbers of CVD cases to balance treatment costs. Several interventions have attractive cost-effectiveness ratios of $50,000–$70,000 per year of life saved.12 A small (but growing) number of people would be at high risk (arbitrarily defined as more than 20 percent risk per ten years), demanding aggressive risk-factor management, prophylactic medications, and treatment of symptoms of CVD.13 A larger group would be considered to be at moderate risk (10–20 percent risk), benefiting from risk-factor management with less costly drugs and devices. Finally, low-risk people (less than 10 percent risk) would benefit from health education, modification of harmful health behavior, and a healthier environment, at little direct personal cost. n Risk-factor paradigm in CVD prevention. Fundamental to any prevention of CVD is the role of risk factors in the etiology of CVD and in their modification to reduce CVD risk. Risk factors may be classified into three groups (Exhibit 1): nonmodifiable, behavioral, and physiological.14 Nonmodifiable risk factors are still useful in risk assessment. Behavioral risk factors may be direct causes of CVD but may also cause physiological risk factors. These common factors might be addressed not only in clinical settings through individual counseling, but might be better addressed at the population level through societal change and public health initiatives. Physiological risk factors are those measured in a clinical setting. As such, they might be addressed with behavioral change advice but frequently require pharmacologic interventions by a care provider. n Need for a combined approach to CVD risk reduction. The risk-factor para-
H E A L T H A F F A I R S ~ Vo l u m e 2 6 , N u m b e r 1
51
Tr e n d s
EXHIBIT 1 The Risk-Factor Paradigm Showing The Relationship Between Nonmodifiable, Behavioral, And Physiological Risk Factors And Cardiovascular Disease Endpoints Nonmodifiable risk factors
Physiological risk factors
Age Male sex Family history
Hypertensive heart disease
Behavioral risk factors Sedentary lifestyle Diet: - Saturated fat - Salt - Cholesterol - Total energy content Heavy alcohol consumption Smoking
Endpoints
Hypertension
Hemorrhagic stroke
Elevated LDL cholesterol
Coronary heart disease
Decreased HDL cholesterol
Atherothrombotic stroke
Diabetes
Peripheral vascular disease
Obesity
Insulin resistance
SOURCE: T.A. Pearson, D.T. Jamison, and J. Trego-Gutierrez, “Cardiovascular Disease,” in Disease Control Priorities in Developing Countries (New York: World Bank and Oxford University Press, 1993), 577–599. NOTES: LDL is low-density lipoprotein. HDL is high-density lipoprotein.
digm in Exhibit 1 identifies at least three opportunities for intervention to reduce CVD mortality: treatment of the symptomatic patient; treatment of behavioral and physiologic risk factors to reduce risk of disease onset or progression in individuals; and modification of deleterious risk behavior at the population level to prevent the onset of risk factors themselves. In reality, a balance of the three is required, with priorities set by the efficacy and cost-effectiveness of the interventions. Treatment of symptomatic disease does not inhibit the atherosclerotic disease process, which, left unimpeded, will reemerge with symptomatic disease from new lesions in other arterial beds. Treatment of risk factors has excellent efficacy and cost-effectiveness as part of the long-term therapy for CVD.15 Perhaps the best evidence for the merits of a balance of primordial, primary, and secondary prevention approaches is illustrated by Lee Goldman and Francis Cook, who analyzed factors explaining the reduction in U.S. mortality from coronary heart disease between 1968 and 1978.16 They concluded that only a small part of the decline could not be explained. Populationwide changes in blood cholesterol (presumably from changes in the U.S. diet) and reductions in cigarette smoking accounted for 54 percent of the decline. Changes in health care, including treatment of hypertension, emergency coronary care, and revascularization, accounted for another 39.5 percent. This illustrates the contribution of a broad and balanced CVD prevention effort, rather than one focused on a single risk factor or stage of disease.
52
Januar y/ Fe br uar y 2007
P r e v e n t i o n
Secondary Prevention For Adult Patients With Coronary And Other Atherosclerotic Vascular Disease n American Heart Association (AHA)/ACC guidelines: 2006 update. The enormous evidence base regarding intervention to prevent vascular disease recurrence and death was first distilled into secondary prevention guidelines, with the most recent in 2006.17 These guidelines set down goals for risk behavior, physiologic risk factors, and prophylactic interventions in CVD patients (Exhibit 2). n Policy issues in secondary prevention. Unresolved scientific issues. The secondary prevention guidelines of the AHA/ACC have not met with much controversy, given the depth of evidence supporting them. One unresolved issue is the lowdensity lipoprotein cholesterol (LDL-C) goal in CVD patients. Several clinical trials have suggested that patients whose LDL-C levels are lowered below 100 mg/dl have fewer events than those just reaching that LDL-C goal.18 A lowering to less than 70 mg/dl is considered reasonable. This degree of reduction often requires multiple cholesterol-lowering drugs with implications for drug costs, side effects, and the need to engage a lipid specialist. The lowering of the LDL-C goal is likely less of an efficacy issue and more of a cost-effectiveness issue. Second, interventions on other lipid fractions (for example, high-density lipoprotein cholesterol [HDL-C], triglyc-
EXHIBIT 2 Goals Of Secondary Prevention For Adult Patients With Coronary And Other Vascular Disease: 2006 AHA/ACC Guidelines Risk management
Goal
Smoking
Complete cessation, no exposure to environmental tobacco smoke
Blood pressure control
Blood pressure