Improving Risk of Coronary Heart Disease: Title and subTitle BreakCan a Picture Make the Difference?

Great progress has been made in understanding the causes of coronary heart disease (CHD) and in reducing the toll of the CHD epidemic.¹ Following identification of the major CHD risk factors (smoking, hypertension, hypercholesterolemia, diabetes, and family history) during the 1950s through the 1970s, clinical trials in the 1980s and beyond proved the clinical benefits of treating them. Prevention of CHD is no longer a dream; it is a reality.²

The major risk factors are implicated in as many as 85% of CHD cases.³ Conversely, individuals with favorable levels of all of the major risk factors are largely protected from clinical CHD. A constellation of healthy risk factor levels, including total blood cholesterol of less than 200 mg/dL (5.18 mmol/L), systolic blood pressure of no more than 120 mm Hg, and diastolic blood pressure of no more than 80 mm Hg, combined with nonsmoking and no diabetes, has defined groups of very low-risk individuals in several long-term cohort studies.⁴ Healthy lifestyle habits undoubtedly account for attainment of favorable risk factor status in many persons, as suggested by Nurses' Health Study data.⁵ A low-risk constellation of behaviors characterized by healthy eating patterns, daily aerobic exercise, nonsmoking, and body mass index of less than 25 was associated with a CHD event rate more than 60% lower than cohort members with none of these favorable characteristics.

Unfortunately, too few individuals have favorable CHD risk profiles. Less than 15% of the members of various cohorts were in the low-risk category.⁴ In the Nurses' Health Study, only 10% of participants had all 4 of the favorable lifestyle habits listed herein and less than 25% had at least 3.⁵ Thus, while prevention of CHD is feasible and scientifically definable by prevention or control of the major risk factors, the challenge to identify and treat those with major CHD risk factors is great. Nearly the entire US population can be considered at risk based on exposure to 1 or more unfavorable risk factor levels or unhealthy lifestyle habits. Thus, a population-wide approach is justified, as well as an intensive clinical approach to risk factor identification, patient activation and motivation, and targeted treatment for those found to be at high risk.

Despite well-defined strategies for reducing CHD risk factors,^{6 - 10} clinicians find that many patients at risk appear to lack the interest or motivation to undertake intensive risk factor treatment efforts. Awareness studies show that even now large proportions of Americans have never had risk factors measured, and many who have had these measured do not achieve acceptable levels of control.^{7 - 8 ,11} Knowing that many patients lack motivation to control CHD risk factors, clinicians have sought additional ways to activate apathetic patients. One method currently advocated by the National Cholesterol Education Program⁸ is comprehensive risk factor measurements to provide the patient with a global quantitative risk score. This approach is intuitively appealing on several grounds, including its capability to estimate the patient's absolute risk and thereby appropriately target intensity of therapy. For some patients, risk assessment alone might provide sufficient motivation to initiate or intensify treatment of abnormal risk factor levels.

However, risk factor control for virtually any health condition, which can be viewed as the end result of "patient compliance," is a multidimensional challenge. Patient motivation is only 1 element of the overall picture.¹² Compliance with medical treatments is a complex behavioral process strongly influenced by the environments in which patients live, practitioners work, and health care systems deliver care.¹² The patient is clearly central to the process of behavior change but cannot be expected to adhere to recommended treatments in a vacuum. Factors that influence compliance include knowledge of the disease and its precursors, previous levels of compliance, confidence in ability to follow recommended behaviors, perception of health and benefits of therapy or change in behavior, availability of social support, and complexity of the treatment regimen.¹² In addition to patient factors, practitioner factors and systems of care also influence patients' ability to respond to recommendations.¹² Many of these same barriers to behavior change apply to physicians in their struggle to adhere to clinical practice guidelines.¹³ Interestingly, behavior change is a challenge to patients and clinicians alike.

A number of tests have recently been proposed for improving the practice of preventive cardiology.^{14 - 16} The hope has been expressed that additional prognostic tests can simultaneously improve risk prediction, thereby improving the match between treatment intensity and patient risk, and also provide additional information to motivate patients to achieve improved risk factor control.¹⁵ As noted by O'Malley et al¹⁷ in this issue of THE JOURNAL, electron beam tomography (EBT) is commonly promoted directly to the general public for both of these purposes. Both goals of testing are vigorously contested and debated.¹⁸ Research is ongoing to assess whether EBT results can improve risk prediction in healthy, asymptomatic individuals beyond that provided by a global risk factor score alone. While this research proceeds, the role of EBT testing as a motivational tool has received relatively little attention prior to the study by O'Malley and colleagues.

Prima facie arguments for EBT testing suggest that "a picture is worth a thousand words" and, accordingly, an abnormal coronary calcium study could be expected to activate a previously unmotivated person to improve unhealthy behaviors. A similar argument was recently advanced in relation to testing for C-reactive protein (CRP) as a potential motivational device among patients who were overweight or hypertensive, smoked, or had other CHD risk factors.¹⁶ While CRP is not a direct image of the arteries, those who promote this test argue that CRP provides a biochemical "image" of the atherosclerotic process that might allow a clinician to target risk factor modification to those at greatest risk.¹⁹ Neither test has been proven to provide this motivational jolt to the average patient. Thus, the randomized trial by O'Malley et al is welcome evidence amid a sea of conjecture.

In the study by O'Malley et al,¹⁷ the use of coronary calcium screening in asymptomatic patients as part of a routine health examination did not lead to improved risk factor control as judged by the Framingham Risk Score 1 year after testing. However, the application of case management, which attempted to focus on the complex nature of behavior change in the clinical setting as discussed herein,¹² resulted in improved risk factor scores compared with usual care. The results apply most directly to the kinds of patients studied here: relatively healthy, asymptomatic, younger (average age, 42 years), well-educated individuals with comparatively low Framingham Risk Scores (average predicted risk of only 5% in 10 years). The results reinforce previous recommendations to avoid use of EBT testing in relatively low-risk people,^{14 ,18} especially as a "screening" tool. The findings, while not directly assessing other imaging modes or biochemical tests like CRP, are likely to apply to use of those tests in similar kinds of patients. The study also complements previous research on the utility of case management to improve risk factor control in primary and secondary prevention of CHD.^{2 ,12}

Additional research is needed to define whether EBT testing can improve patient outcomes in other clinical settings and other patient populations. Of note, there were trends toward improvement among patients with abnormal calcium scans, suggesting that high-risk patients might benefit from EBT testing. Another important question is whether EBT improves patient outcomes in individuals who seek testing rather than in those offered testing in a primary care setting. After all, this is how EBT testing is most commonly used now, and perhaps such testing can be more effective than was evident in the study by O'Malley et al.¹⁷ But EBT testing may also have adverse effects, as discussed by O'Malley et al,¹⁷ such as false reassurance among those with negative scans. Given that the median coronary calcium score in large samples of men younger than 50 years is in the range of 3 units by the Agatston method,²⁰ it is possible that certain patients with high Framingham Risk Scores will have a low calcium score and mistakenly conclude that they are at low long-term risk of CHD events. The data presented by O'Malley et al suggest that this may not be a serious problem. Definite answers to these and other questions regarding EBT and intensive case management will require carefully conducted trials such as that by O'Malley et al.¹⁷

To date, most research on EBT has been observational in nature, based entirely on self-referred patients. The randomized trial reported by O'Malley et al should set the stage for future work. Based on the trial results, intensive case management emerges as a viable primary prevention approach while the role of EBT remains uncertain. The report by O'Malley et al¹⁷ will hopefully encourage design and execution of additional randomized trials to define specific roles for EBT in risk prediction. It should also stimulate greater scrutiny of other new tests, including carotid ultrasonography for measurement of intimal-medial thickness, brachial artery reactivity studies to assess endothelial dysfunction, and CRP and other biochemical markers of inflammation.