Prediction of Lifetime Risk for Cardiovascular Disease by Risk Factor ...

Prediction of Lifetime Risk for Cardiovascular Disease by Risk Factor Burden at 50 Years of Age Donald M. Lloyd-Jones, Eric P. Leip, Martin G. Larson, Ralph B. D’Agostino, Alexa Beiser, Peter W.F. Wilson, Philip A. Wolf and Daniel Levy Circulation 2006;113;791-798; originally published online Feb 6, 2006; DOI: 10.1161/CIRCULATIONAHA.105.548206 Circulation is published by the American Heart Association. 7272 Greenville Avenue, Dallas, TX 72514 Copyright © 2006 American Heart Association. All rights reserved. Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/cgi/content/full/113/6/791

Subscriptions: Information about subscribing to Circulation is online at http://circ.ahajournals.org/subscriptions/ Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Fax: 410-528-8550. E-mail: [email protected] Reprints: Information about reprints can be found online at http://www.lww.com/reprints

Downloaded from circ.ahajournals.org by on April 24, 2007

Epidemiology Prediction of Lifetime Risk for Cardiovascular Disease by Risk Factor Burden at 50 Years of Age Donald M. Lloyd-Jones, MD, ScM; Eric P. Leip, PhD; Martin G. Larson, ScD; Ralph B. D’Agostino, PhD; Alexa Beiser, PhD; Peter W.F. Wilson, MD; Philip A. Wolf, MD; Daniel Levy, MD Background—Lifetime risk for atherosclerotic cardiovascular disease (CVD) has not previously been estimated, and the effect of risk factor burden on lifetime risk is unknown. Methods and Results—We included all Framingham Heart Study participants who were free of CVD (myocardial infarction, coronary insufficiency, angina, stroke, claudication) at 50 years of age. Lifetime risks to 95 years of age were estimated for men and women, with death free of CVD as a competing event. We followed up 3564 men and 4362 women for 111 777 person-years; 1757 had CVD events and 1641 died free of CVD. At 50 years of age, lifetime risks were 51.7% (95% CI, 49.3 to 54.2) for men and 39.2% (95% CI, 37.0 to 41.4) for women, with median survivals of 30 and 36 years, respectively. With more adverse levels of single risk factors, lifetime risks increased and median survivals decreased. Compared with participants with ⱖ2 major risk factors, those with optimal levels had substantially lower lifetime risks (5.2% versus 68.9% in men, 8.2% versus 50.2% in women) and markedly longer median survivals (⬎39 versus 28 years in men, ⬎39 versus 31 years in women). Conclusions—The absence of established risk factors at 50 years of age is associated with very low lifetime risk for CVD and markedly longer survival. These results should promote efforts aimed at preventing development of risk factors in young individuals. Given the high lifetime risks and lower survival in those with intermediate or high risk factor burden at 50 years of age, these data may be useful in communicating risks and supporting intensive preventive therapy. (Circulation. 2006;113:791-798.) Key Words: cardiovascular disease 䡲 epidemiology 䡲 risk factors 䡲 survival

D

espite 4 decades of declining mortality from cardiovascular disease (CVD) in the United States, CVD remains by far the leading cause of morbidity and mortality,1 and it is soon to be the leading cause of morbidity and mortality in the developing world.2 Recent data suggest disturbing increases in the prevalence of CVD risk factors such as diabetes, obesity, and the metabolic syndrome,1,3,4 which may reverse downward trends in CVD mortality. In the face of the enormous public health burden imposed by CVD, renewed efforts are needed to promote prevention.

Clinical Perspective p 798 One tool that may be useful in public health education is an understanding of the lifetime risk for CVD, which has not been estimated in any population to date. Given that lifetime risk estimates provide an absolute risk assessment, they may be more easily understood by clinicians and patients than

relative risks, and they may help to motivate beneficial changes in lifestyle or health behaviors. The best example of lifetime risk data being used effectively to change behavior is the wide dissemination of data on lifetime risk for breast cancer (1 in 8 for women at 40 years of age),5 which appears to have contributed to markedly increased rates of screening for breast cancer in the early 1990s.6,7 The Framingham Heart Study, with its well-defined cohorts, long-term follow-up, and careful documentation of risk factors and events, provides a unique opportunity to examine factors that may modify remaining lifetime risk for CVD in the context of overall survival. Factors that increase short-term risk for CVD are well known. However, the effect of risk factors on long-term and lifetime risk may be unpredictable because some risk factors also increase risk for non-CVD death, resulting in competing risks. We sought to estimate the lifetime risk for CVD and to examine overall survival in the presence and absence of established risk factors.

Received March 9, 2005; revision received December 7, 2005; accepted December 14, 2005. From the Department of Preventive Medicine, Feinberg School of Medicine, Northwestern University, Chicago, Ill (D.M.L.-J.); National Heart, Lung, and Blood Institute’s Framingham Heart Study, Framingham, Mass (D.M.L.-J., E.P.L., M.G.L., R.B.D., A.B., P.W.F.W., P.A.W., D.L.); Departments of Epidemiology and Preventive Medicine (M.G.L., R.B.D., P.W.F.W., D.L.) and Neurology (P.A.W.), Boston University School of Medicine, Boston, Mass; Department of Epidemiology and Biostatistics, Boston University School of Public Health, Boston, Mass (E.P.L., R.B.D., A.B.); and National Heart, Lung, and Blood Institute, Bethesda, Md (D.L.). Guest Editor for this article was Donna K. Arnett, PhD. Correspondence to Donald M. Lloyd-Jones, MD, ScM, Department of Preventive Medicine, Feinberg School of Medicine, Northwestern University, 680 N Lake Shore Dr, Ste 1102, Chicago, IL 60611. E-mail [email protected] © 2006 American Heart Association, Inc. Circulation is available at http://www.circulationaha.org

DOI: 10.1161/CIRCULATIONAHA.105.548206

791 Downloaded from circ.ahajournals.org by on April 24, 2007

792

Circulation

February 14, 2006

Methods Study Sample Study design and entry criteria for the Framingham Heart Study cohorts have been detailed elsewhere.8,9 All study protocols and procedures have been approved by the Institutional Review Board of Boston Medical Center. For the present analysis, to reflect more contemporary experience, we included all participants who were free of CVD before their earliest examination between 1971 and 2002, provided that they were examined at least once between 50 and 94 years of age and they had follow-up after their earliest eligible examination. Participants who were ⬍50 years of age at the beginning of the study period entered the sample on attainment of 50 years of age. Only 32 participants (0.4%) were lost to follow-up during the study period.

Risk Factor Measurement During routine examinations at the Heart Study, participants underwent standardized anthropometric measurements to determine height and weight, from which body mass index (BMI) was calculated. Current smoking was defined as self-report of active smoking within the last year before the examination. Blood pressure was determined from the average of 2 separate readings taken by a physician at least 5 minutes apart, as described previously.10 Blood, drawn in ethylene diaminotetracetic acid plasma for all cholesterol measurements, was measured as described previously.11 Diabetes was defined as the use of insulin or hypoglycemic agents, a fasting blood glucose ⬎6.88 mmol/L (⬎125 mg/dL) in offspring participants, or a casual blood glucose ⱖ11 mmol/L (ⱖ200 mg/dL) in original cohort participants.

Case Ascertainment Events during follow-up were ascertained from review of interim medical records of those who appeared and those who failed to appear for scheduled examinations. Atherosclerotic CVD events were defined by the occurrence of myocardial infarction, coronary insufficiency, death resulting from coronary heart disease, angina pectoris, atherothrombotic stroke, intermittent claudication, or other cardiovascular death. We did not include congestive heart failure events (for which the lifetime risk has previously been published12) because they often are not mediated by atherosclerosis12,13; we also did not include revascularization procedures occurring in isolation without a preceding clinical event because those typically were initiated by knowledge of the presence of risk factors rather than clinically overt disease. All deaths and suspected cardiovascular events were reviewed by a panel of 3 physicians who applied established criteria for such events.14

Statistical Analysis All statistical analyses were performed with SAS statistical software.15 To calculate lifetime risk, a modified technique of survival analysis was used, as described previously.16 In this type of analysis, participants contributed information on the incidences of CVD and death free of CVD for each age they attained during follow-up. Participants could enter the sample at ⬎50 years of age and contribute person-time to follow-up if they were still free of CVD (and therefore were free of CVD at 50 years of age) and if risk factor data had been measured at 50 years of age. Age-specific hazards, incidence rates, cumulative incidence, and survival probabilities were calculated as in a Kaplan-Meier analysis.17 Because the Kaplan-Meier cumulative incidence does not reflect the competing risk for death from other causes before the development of CVD, adjustment was made for this competing risk to yield a true remaining lifetime risk for CVD.18 Each subject in the study sample was followed up from entry through 2002 until the occurrence of a first CVD event, death, attainment of 95 years of age, or the last clinic examination or medical contact for which follow-up data were available. Lifetime risk estimates for CVD were calculated for men and women beginning at 50 years of age and extended only to 95 years of age because few participants survived past 94 years of age.

We repeated the analyses, ignoring angina and claudication as first events and counting only “hard” CVD events. We grouped participants according to risk factor levels as measured at the examination most closely preceding and up to 5 years before 50 years of age. Lifetime risks for CVD and Kaplan-Meier overall survival were then calculated as above for men and women in each risk factor stratum. Separate analyses indicated no substantial difference when we included and excluded participants receiving antihypertensive therapy, so they were included and defined as being hypertensive. Rates of cholesterol-lowering therapy were low until near the end of the study period, so we ignored treatment status in assigning cholesterol strata. Because HDL cholesterol levels were measured only from 1971 on, we did not have follow-up to 94 years of age by strata of HDL cholesterol; we also used HDL cholesterol levels measured up to 10 years before 50 years of age to increase the sample size for those analyses. For the analysis examining lifetime risk among diabetics, lifetime risk was calculated to 75 years of age because of limited follow-up among diabetics ⬎75 years of age. Finally, we stratified participants a priori into mutually exclusive categories based on whether they had all optimal risk factor levels, ⱖ1 low risk factors, ⱖ1 intermediate risk factors, 1 major risk factor only, or ⱖ2 major risk factors (see Table 1 footnote for definitions). The authors had full access to the data and take responsibility for its integrity. All authors have read and agree to the manuscript as written.

Results Baseline Characteristics We followed up 3564 men and 4362 women for a total of 111 777 person-years. Baseline characteristics for participants at 50 years of age are shown in Table 1. During follow-up, 1757 participants suffered an incident CVD event, and 1641 died free of overt CVD.

Lifetime Risk for CVD and Median Survival at 50 Years of Age Among men free of CVD at 50 years of age, the lifetime risk to 95 years of age for developing CVD was 51.7% (95% CI, 49.3 to 54.2). Median overall survival in men was 30 years. Among women, lifetime risk to 95 years of age was 39.2% (95% CI, 37.0 to 41.4), with median overall survival of 36 years. Lifetime risks for hard CVD to 95 years of age were 41.2% (95% CI, 38.8 to 43.7) in men and 28.8% (95% CI, 26.6 to 30.8) in women.

Effect of Individual Risk Factors As shown in Table 2 and Figure 1, increasing blood pressure and total cholesterol were associated with increased lifetime risk for CVD and with shorter median survival in both men and women. The presence of diabetes at 50 years of age conferred the highest lifetime risk for CVD of any single risk factor, at 67.1% for diabetic men and 57.3% for diabetic women through 75 years of age. Median survival was substantially lower among diabetic compared with nondiabetic men. Overweight and obesity were associated with modest increases in lifetime risk and reductions in survival compared with normal weight. Lifetime risk for CVD was similar for smokers and nonsmokers (Table 2 and Figure 1). Smokers had CVD events much earlier than nonsmokers; eg, through 70 years of age, smokers had almost twice the adjusted cumulative incidence of CVD as nonsmokers. However, the competing risk of death from other smoking-related causes shortened


Lloyd-Jones et al TABLE 1.

Lifetime Risk for CVD

793

Characteristics of Study Participants at 50 Years of Age

Characteristic

Men (n⫽3564)

Women (n⫽4362)

Total cholesterol, mean⫾SD, mmol/L (mg/dL)

5.66⫾1.03 (219⫾40)

5.72⫾1.14 (221⫾44)

HDL cholesterol, mean⫾SD, mmol/L (mg/dL)

1.14⫾0.31 (44⫾12)

1.47⫾0.41 (57⫾16)

Systolic blood pressure, mean⫾SD, mm Hg

130⫾17

127⫾19

Diastolic blood pressure, mean⫾SD, mm Hg

82⫾10

79⫾11

103 (2.9)

61 (1.4)

1679 (47.1)

1649 (37.8)

BMI, mean⫾SD, kg/m

27.1⫾3.8

25.4⫾4.6

All optimal risk factors,* n (%)

114 (3.2)

196 (4.5)

ⱖ1 Not optimal risk factor,† n (%)

392 (11.0)

602 (13.8)

ⱖ1 Elevated risk factor,‡ n (%)

827 (23.2)

1051 (24.1)

1 Major risk factor,§ n (%)

1508 (42.3)

1767 (40.5)

ⱖ2 Major risk factors,§ n (%)

723 (20.3)

746 (17.1)

Diabetes, n (%) Current smoker, n (%) 2

*Defined as total cholesterol ⬍4.65 mmol/L (⬍180 mg/dL), blood pressure ⬍120/⬍80 mm Hg, nonsmoker, and nondiabetic. †Defined as total cholesterol 4.65 to 5.15 mmol/L (180 to 199 mg/dL), systolic blood pressure 120 to 139 mm Hg, diastolic blood pressure 80 to 89 mm Hg, nonsmoker, and nondiabetic. ‡Defined as total cholesterol 5.16 to 6.19 mmol/L (200 to 239 mg/dL), systolic blood pressure 140 to 159 mm Hg, diastolic blood pressure 90 to 99 mm Hg, nonsmoker, and nondiabetic. §Defined as total cholesterol ⱖ6.20 mmol/L (ⱖ240 mg/dL), systolic blood pressure ⱖ160 mm Hg, diastolic blood pressure ⱖ100 mm Hg, smoker, or diabetic.

median survival by 5 years and limited lifetime risk for CVD among smokers. As shown in Table 2, the relative effect of some risk factors on lifetime risks for CVD differed through 75 years of age compared with effects through 95 years of age. For example, smoking and elevated blood pressure at 50 years of age were associated with greater relative increases in lifetime risk for CVD through 75 years of age than through 95 years of age. Conversely, elevated cholesterol was associated with a fairly constant relative effect on lifetime risk for CVD. Lifetime risk estimates through 75 years of age in effect assume that the last individual in the cohort would die at this age. However, the preponderance of mortality occurred after 75 years of age (median survival was 33 years at 50 years of age). Because smoking and elevated blood pressure increase risk for nonatherosclerotic and non-CVD death but cholesterol does not, the different effects of these risk factors on short-term and lifetime risks for CVD become evident with longer follow-up.

Effect of Overall Risk Profile When we stratified participants by their burden of risk factors at 50 years of age, the magnitude of lifetime risk rose steeply from those with optimal risk factor levels to those with ⱖ2 major risk factors, whereas median survival declined substantially. As shown in Table 3 and Figure 2, compared with participants with ⱖ2 major risk factors, participants with optimal levels had substantially lower lifetime risks (5.2% versus 68.9% in men, 8.2% versus 50.2% in women) and markedly longer median survivals (by ⬎11 years in men, ⬎8 years in women). For both men and women ⬎50 years of age, the adjusted cumulative incidence curves across aggregate risk strata (Figure 2) separated early and continued to diverge throughout the remaining lifespan.

When low HDL cholesterol (⬍1.03 mmol/L [⬍40 mg/dL] in men, ⬍1.29 mmol/L [⬍50 mg/dL] in women) and obesity (BMI ⱖ30 kg/m2) were included as major risk factors, lifetime risks for CVD were similar to those shown in Table 3 and Figure 2, indicating that low HDL cholesterol and obesity were equivalent to major risk factors. There were too few participants with otherwise optimal risk factor profiles and high HDL cholesterol or normal BMI (⬍25 kg/m2) to provide meaningful estimates of lifetime risk in this stratum.

Discussion Principal Findings These results represent the first estimates of the lifetime risk for developing CVD; for people free of CVD at 50 years of age, more than half of men and nearly 40% of women will develop CVD during their remaining lifespan. Diabetes at 50 years of age confers the highest lifetime risk for CVD of any single risk factor. Participants with optimal risk factor levels at 50 years of age (only 3.2% of men and 4.5% of women in our sample) have very low remaining lifetime risk for CVD and markedly longer survival, whereas participants with ⱖ2 major risk factors (20.2% of men and 17.1% of women) have very high lifetime risk even in the face of substantially shorter survival.

Clinical and Public Health Implications The lifetime risks for CVD are among the highest published for any disease to date. Compared with the present results, the lifetime risks for cancers are substantially lower. At 50 years of age, the most common types are breast cancer in women, with a lifetime risk of 12.5%, and prostate cancer in men, with a lifetime risk of 19%.19 Lifetime risks for lung cancer are 8% and 6% in men and women, respectively, and for colorectal cancer are 6% and 6%, respectively.19


794 TABLE 2.

Circulation

February 14, 2006

Lifetime Risk for CVD and Median Survival for Men and Women in Selected Risk Factor Strata at 50 Years of Age Men

Women

Lifetime Risk for CVD (95% CI), %

Lifetime Risk for CVD (95% CI), %

To 75 y

To 95 y

Median Survival (IQR), y

35.0 (32.9–37.2)

51.7 (49.3–54.2)

30 (22–37)

⬍4.65 (⬍180)

26.2

38.7

31

9.1

19.4

37

4.65–5.15 (180–199)

29.2

46.9

32

11.3

31.1

37

5.16–6.19 (200–239)

34.5

49.2

30

16.7

37.6

36

ⱖ6.20 (ⱖ240)

45.3

64.6

30

30.0

48.3

34

Not low: ⱖ1.03 (ⱖ40) men/ⱖ1.29 (ⱖ50) women

23.6

䡠䡠䡠

⬎33

11.0

䡠䡠䡠

⬎33

Low: ⬍1.03 (⬍40) men/⬍1.29 (⬍50) women

34.0

䡠䡠䡠

29

15.9

䡠䡠䡠

⬎33

26.6

47.3

33

10.5

29.3

37

120–139 or 80–89

31.8

47.9

32

17.9

37.0

36

140–159 or 90–99

46.4

61.6

29

28.8

52.3

35

ⱖ160 or ⱖ100 or treated

51.3

65.1

28

35.0

50.6

31

Nondiabetic

30.2

䡠䡠䡠

32

16.3

䡠䡠䡠

36

Diabetic

67.1

57.3

27.8

䡠䡠䡠 47.8

23

Nonsmoker

34

14.2

䡠䡠䡠 39.0

38

Smoker

34.0

51.5

29

20.6

36.8

33

⬍25

27.5

47.8

32

14.7

35.3

37

25–25.9

30.4

49.3

29

18.1

42.5

36

ⱖ30

41.8

58.0

29

21.9

43.0

33

Overall

To 75 y

To 95 y

Median Survival (IQR), y

19.2 (17.5–20.8)

39.2 (37.0–41.4)

36 (28–42)

Total cholesterol, mmol/L (mg/dL)

HDL cholesterol, mmol/L (mg/dL)

Systolic or diastolic blood pressure, mm Hg ⬍120 or ⬍80

34

BMI, kg/m2

IQR indicates interquartile range.

Despite the fact that the risks of developing and dying from CVD are substantially higher than for cancer,1,19 surveys continue to indicate that most Americans perceive that their risk of cancer is higher. In a 2003 survey of a nationally representative sample of women, fewer than half correctly identified CVD as the leading cause of death among women in the United States. Furthermore, only 13% identified CVD as their greatest personal health risk, whereas 51% identified cancer as their greatest health risk.20 Thus, a substantial deficit in awareness of CVD risk exists, and there is a paradoxical gap between perceived and actual personal risk.20 Greater media coverage of cancer risks (especially breast cancer risks) during the 1990s has been well documented21 and may have perpetuated some of the knowledge gap among women. The American Heart Association and National Heart, Lung, and Blood Institute have recently initiated national programs attempting to improve women’s awareness of CVD, and there are now specific guidelines published for CVD prevention in women.22 It is hoped that the current data can help galvanize media and public interest in preventing CVD in both women and men. Clearly, prevention efforts need to begin decades before 50 years of age because even the presence of a single major risk factor at 50 years of age is associated with substantially

increased lifetime risk for CVD and markedly shorter survival. Lifestyle measures focused on diet and exercise in young adulthood and middle age could prevent the development of obesity, diabetes, hypertension, and dyslipidemia in large numbers of individuals,23 but few segments of the population have been successful in implementing them. Among individuals who already have ⱖ1 intermediate or major risk factors at 50 years of age, our data suggest that aggressive global risk factor modification should be considered, given the associated high lifetime risks for CVD. Special attention should be paid to diabetics, who have the highest lifetime risk and shortest median survival. Lifetime risk estimates represent an average experience derived from large cohorts, so applying them to an individual patient should be done with caution. Each patient has unique risk for developing CVD, depending on his or her levels of risk factors and genetic predisposition, that should be considered in determining the need for primary prevention strategies. In addition, patients have differing risks for death from non-CVD causes. Nonetheless, the absolute risks and median survivals provided here may be extremely useful for guiding clinicians and patients in decisions about how intensively to focus on prevention and modification of risk factors. Our findings support the concept of estimating global risk rather than focusing on single risk factors to make treatment


Lloyd-Jones et al


795

Figure 1. Cumulative incidence of CVD adjusted for the competing risk of death for men and women according to individual risk factor strata at 50 years of age. The numbers at the right of each graph represent the adjusted cumulative incidence to 95 years of age (75 years of age for diabetes) or the lifetime risk for CVD.

decisions because we observed better discrimination of lifetime risk with aggregate risk factor burden than with any single risk factor alone. The current European guidelines24 and National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (NCEP-ATP III) guidelines25 recommend estimation of global 10-year risk for fatal CVD or major CHD events in determining the need for preventive therapy. However, relying solely on estimates of short-term absolute risk is problematic. Any single risk factor can produce cumulative damage and high risk if left untreated for many years.11,26 –28 Long-term risk assessment is particularly relevant for younger patients, in whom exclusive attention to low shortterm risk may discourage initiation of or adherence to lifestyle modification and treatment. Hence, there is a need to

consider lifetime risk in addition to short-term risk, as recommended in the NCEP-ATP III guidelines.25 These lifetime risk data, in conjunction with the median survival data, could provide a useful adjunct to short-term risk estimation in the clinical setting. For example, using the ATP-III online risk estimator (http://hin.nhlbi.nih.gov/atpiii/ riskcalc.htm) indicates that a 50-year-old nonsmoking, nondiabetic man with total cholesterol of 6.47 mmol/L (250 mg/dL), HDL cholesterol of 1.55 mmol/L (60 mg/dL), and untreated systolic blood pressure of 160 mm Hg has an estimated 10-year risk for hard CHD of 7%. In contrast, his average lifetime risk for CVD is nearly 70%, and his median survival is ⬎11 years shorter than that for a man at the same age with optimal risk factors. A woman at 50 years of age with identical risk factor levels has an estimated 10-year risk


796 TABLE 3.

Circulation

February 14, 2006

Lifetime Risk for CVD and Median Survival for Men and Women by Aggregate Risk Factor Status at 50 Years of Age Men

Women

Lifetime Risk for CVD (95% CI), % Risk Stratum* Overall All optimal risk factors

To 75 y

To 95 y

35.0 (32.9–37.2)

51.7 (49.3–54.2)

5.2 (0–12.2)

5.2 (0–12.2)

Lifetime Risk for CVD (95% CI), % Median Survival (IQR), y

To 75 y

To 95 y

30 (22–37)

19.2 (17.5–20.8)

39.2 (37.0–41.4)

⬎39 (32–⬎45)

8.2 (0–22.3)

8.2 (0–22.3)

Median Survival (IQR), y 36 (28–42) ⬎39 (28–⬎45)

ⱖ1 Not optimal risk factor

17.6 (10.9–24.4)

36.4 (23.1–49.6)

36 (29–42)

6.9 (3.3–10.5)

26.9 (18.4–35.5)

39 (33–43)

ⱖ1 Elevated risk factor

26.0 (21.0–31.0)

45.5 (38.0–53.1)

35 (26–42)

14.6 (11.2–17.9)

39.1 (33.0–45.1)

39 (32–44)

1 Major risk factor

37.6 (33.8–41.5)

50.4 (46.2–54.5)

30 (23–36)

18.0 (15.3–20.7)

38.8 (35.0–42.6)

35 (28–42)

ⱖ2 Major risk factors

53.2 (47.1–59.3)

68.9 (61.7–73.2)

28 (18–35)

37.7 (32.5–43.0)

50.2 (44.7–55.7)

31 (23–38)

IQR indicates interquartile range. *Risk factor levels defined as for Table 1.

of only 2% compared with a lifetime risk for CVD of 50% and ⬎8-years-shorter median survival compared with a woman at the same age with optimal risk factors. These data, placed in the clinical context, may be much more useful in motivating therapeutic lifestyle changes and promoting adherence to therapy.

Effect of Risk Factors The most striking findings in the present analysis were the enormous differences in lifetime risk and survival between participants with optimal risk factor levels and those with ⱖ2 major risk factors. Increasing interest is being focused on participants with optimal or low levels of traditional risk factors. In addition to substantially lower risks for CVD, cardiovascular death, and total mortality,29 –31 individuals with low risk factor levels also appear to age more success-

fully. Daviglus et al32 recently reported that individuals with favorable risk profiles in middle age had better health-related quality of life after ⬇25 years of follow-up compared with those who had intermediate or major risk factors. These data underscore the importance of preventing the development of traditional risk factors at younger ages to increase healthy longevity in older adults. However, national survey data indicate that between 1991 and 2001, the percentage of American adults with no known major risk factors declined from 42% to 36%,3 a trend driven by increases in the prevalence of high cholesterol, hypertension, diabetes, and especially obesity.3,4 Our data should not be interpreted as indicating that smoking has no overall effect on CVD or health in general. To the contrary, consistent with the overwhelming weight of epidemiological evidence, smoking was associated with a

Figure 2. Cumulative incidence of CVD adjusted for the competing risk of death for men and women according to aggregate risk factor (RF) burden at 50 years of age. The numbers at the right of each graph represent the adjusted cumulative incidence to 95 years of age or the lifetime risk for CVD. Optimal risk factors are defined as total cholesterol ⬍4.65 mmol/L (⬍180 mg/dL), blood pressure ⬍120/ ⬍80 mm Hg, nonsmoker, and nondiabetic. Not optimal risk factors are defined as total cholesterol of 4.65 to 5.15 mmol/L (180 to 199 mg/dL), systolic blood pressure of 120 to 139 mm Hg, diastolic blood pressure of 80 to 89 mm Hg, nonsmoker, and nondiabetic. Elevated risk factors are defined as total cholesterol of 5.16 to 6.19 mmol/L (200 to 239 mg/dL), systolic blood pressure of 140 to 159 mm Hg, diastolic blood pressure of 90 to 99 mm Hg, nonsmoker, and nondiabetic. Major risk factors are defined as total cholesterol ⱖ6.20 mmol/L (ⱖ240 mg/dL), systolic blood pressure ⱖ160 mm Hg, diastolic blood pressure ⱖ100 mm Hg, smoker, and diabetic.


Lloyd-Jones et al substantially higher short-term risk for CVD and with shortened survival (by 5 years in men and women). However, in the context of the competing risk for death over the longer term, the force of mortality from noncardiovascular causes associated with smoking (ie, cancers, lung diseases) had the effect of constraining lifetime risk for CVD among smokers. Conversely, nonsmokers lived longer and had CVD events later in life.33 In addition, participants who subsequently ceased smoking may have lowered their risk for CVD and thereby lowered the lifetime risk estimate for smokers as a group. One-time assignment of risk factor status at the index age without updating for later crossover may appear to be a limitation of the current methods, but it best represents the data available to clinicians and patients in the office for prediction of future risk. To the best of our knowledge, these are the first data to examine lifetime risks for CVD according to BMI status. Of note, higher BMI at 50 years of age was associated with increased lifetime risk for CVD and with reduced median survival (by 3 to 4 years).

Potential Study Limitations The Framingham cohorts, which are exclusively white, have consistently had risk factor levels, treatment rates, and disease incidence rates that are quite similar to US whites as a group. Blacks tend to have a heavier burden of some CVD risk factors compared with whites,34,35 but Framingham risk equations have been shown to be broadly applicable in ordering risk, although absolute risk estimates may differ.36 Data from another cohort37 suggest that lifetime risks for CVD death are similar among blacks and whites. Whereas blacks have somewhat higher age-specific incidences for CVD, they also have substantially higher competing risks for non-CVD mortality at all ages,38 limiting the lifetime risks for CVD. Application of these absolute lifetime risk estimates results to other race/ethnic groups is uncertain. The high prevalence of smoking among study participants at 50 years of age, which likely reflects several decades of exposure, was associated with increased intermediate-term incidence of CVD, as observed in the lifetime risk estimates through 75 years of age (ie, 25-year follow-up). However, this increased risk is constrained when lifetime CVD risk is extended through 95 years of age, probably because of the substantial competing risk for non-CVD death associated with smoking (eg, chronic lung disease and cancer) and the preponderance of mortality from these causes occurring after 75 years of age. Lifetime risk estimates may be subject to birth cohort effects, particularly when disease incidence rates are changing rapidly. Individuals born in different decades may experience diverse risks for CVD based on temporal changes in environment and exposure to risk factors. Given the decline in CVD mortality rates in the United States over the last 4 decades,1 which has been mirrored in the Framingham cohort,39 there may have been changes in lifetime risk also. Nonetheless, these lifetime risk estimates represent an important adjunct to understanding population risks that can add to short-term risk estimates.

Acknowledgments Dr Lloyd-Jones is supported by grant K23 HL04253 from the National Heart, Lung, and Blood Institute. The Framingham Heart Study is supported by NIH/NHLBI contract N01-HC-25195.


797

Disclosures None.

References 1. American Heart Association. Heart Disease and Stroke Statistics: 2005 Update. Dallas, Tex: American Heart Association; 2004. 2. Murray CJL, Lopez AD. Alternative projections of mortality and disability by cause 1990 –2020: Global Burden of Disease Study. Lancet. 1997;349:1498 –1504. 3. Centers for Disease Control and Prevention. Declining prevalence of no known major risk factors for heart disease and stroke among adults: United States, 1991–2001. MMWR Morbid Mortal Wkly Rep. 2004; 53:4 –7. 4. Mokdad AH, Ford ES, Bowman BA, Dietz WH, Vinicor F, Bales VS, Marks JS. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA. 2003;289:76 –79. 5. Feuer EJ, Wung L, Boring CC, Flanders WD, Timmel MJ, Tong T. The lifetime risk of developing breast cancer. J Natl Cancer Inst. 1993;85: 892– 897. 6. National Center for Health Statistics. Health, United States, 1995. Hyattsville, Md: Public Health Service; 1996. 7. Blackman DK, Bennett EM, Miller DS. Trends in self-reported use of mammograms (1989 –1997) and Papanicolaou (1991–1997). MMWR CDC Surveill Summ. 1999;48:1–22. 8. Dawber TR, Kannel WB, Lyell LP. An approach to longitudinal studies in a community: the Framingham study. Ann N Y Acad Sci. 1963;107: 539 –556. 9. Kannel WB, Feinleib M, McNamara PM, Garrison RJ, Castelli WP. An investigation of coronary heart disease in families: the Framingham Offspring Study. Am J Epidemiol. 1979;110:281–290. 10. Lloyd-Jones DM, Evans JC, Larson MG, Levy D. Treatment and control of hypertension in the community: a prospective analysis. Hypertension. 2002;40:640 – 646. 11. Lloyd-Jones DM, Wilson PWF, Larson MG, Leip EP, Beiser A, D’Agostino RB, Cleeman JI, Levy D. Lifetime risk for coronary heart disease by cholesterol levels at selected ages. Arch Intern Med. 2003; 163:1966 –1972. 12. Lloyd-Jones DM, Larson MG, Leip EP, Beiser A, D’Agostino RB, Kannel WB, Murabito JM, Vasan RS, Benjamin EJ, Levy D. Lifetime risk for developing congestive heart failure: the Framingham Heart Study. Circulation. 2002;106:3068 –3072. 13. Levy D, Larson MG, Vasan RS, Kannel WB, Ho KK. The progression from hypertension to congestive heart failure. JAMA. 1996;275: 1557–1562. 14. Abbott RD, McGee DL. The Framingham Study: An Epidemiological Investigation of Cardiovascular Disease, Section 37: The Probability of Developing Certain Cardiovascular Diseases in Eight Years at Specified Values of Some Characteristics. Bethesda, Md: National Heart, Lung, and Blood Institute; 1987. 15. SAS/STAT User’s Guide, Version 8. Cary, NC: SAS Institute, Inc; 1999. 16. Beiser A, D’Agostino RB, Seshadri S, Sullivan LM, Wolf PA. Computing estimates of incidence, including lifetime risk. Stat Med. 2002;19: 1495–1522. 17. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53:457– 481. 18. Gaynor JJ, Feuer EJ, Tan CC, Wu DH, Little CR, Straus DJ, Clarkson BD, Brennan MF. On the use of cause-specific failure and conditional failure probabilities. J Am Stat Assoc. 1993;88:402– 409. 19. SEER Cancer Statistics Review, 1975–2000. Available at: http:// seer.cancer.gov/csr/1975_2000/results_merged/topic_lifetime_risk.pdf. Accessed September 26, 2005. 20. Mosca L, Ferris A, Fabunmi R, Robertson RM. Tracking women’s awareness of heart disease: an American Heart Association national study. Circulation. 2004;109:573–579. 21. Blanchard D, Erblich J, Montgomery GH, Bovbjerg DH. Read all about it: the over-representation of breast cancer in popular magazines. Prev Med. 2002;35:343–348. 22. Mosca L, Appel LJ, Benjamin EJ, Berra K, Chandra-Strobos N, Fabunmi RP, Grady D, Haan CK, Hayes SN, Judelson DR, Keenan NL, McBride P, Oparil S, Ouyang P, Oz MC, Mendelsohn ME, Pasternak RC, Pinn VW, Robertson RM, Schenck-Gustafsson K, Sila CA, Smith SC Jr, Sopko G, Taylor AL, Walsh BW, Wenger NK, Williams CL. Evidence-based guidelines for cardiovascular disease prevention in women. Circulation. 2004;109:672– 693.


798

Circulation

February 14, 2006

23. Slentz CA, Duscha BD, Johnson JL, Ketchum K, Aiken LB, Samsa GP, Houmard JA, Bales CW, Kraus WE. Effects of the amount of exercise on body weight, body composition, and measures of central obesity: STRRIDE: a randomized controlled study. Arch Intern Med. 2004;164: 31–39. 24. Third Joint Task Force of European and Other Societies on Cardiovascular Disease Prevention in Clinical Practice. European guidelines on cardiovascular disease prevention in clinical practice. Atherosclerosis. 2004;173:381–391. 25. 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;106:3143–3421. 26. Grundy SM, Bazzarre T, Cleeman J, D’Agostino RB, Hill M, HoustonMiller N, Kannel WB, Krauss R, Krumholz HM, Lauer RM, Ockene IS, Pasternak RC, Pearson TA, Ridker PM, Wood D. Prevention Conference V: beyond secondary prevention: identifying the high-risk patient for primary prevention: medical office assessment. Circulation. 2000;101:e3– e11. 27. Klag MJ, Ford DE, Mead LA, He J, Whelton PK, Liang K-Y, Levine DM. Serum cholesterol in young men and subsequent cardiovascular disease. N Engl J Med. 1993;328:313–318. 28. Stamler J, Daviglus ML, Garside DB, Dyer AR, Greenland P, Neaton JD. Relationship of baseline serum cholesterol levels in 3 large cohorts of younger men to long-term coronary, cardiovascular, and all-cause mortality and to longevity. JAMA. 2000;284:311–318. 29. Stamler J, Stamler R, Neaton JD, Wentworth D, Daviglus ML, Garside D, Dyer AR, Liu K, Greenland P. Low risk-factor profile and long-term cardiovascular and noncardiovascular mortality and life expectancy. JAMA. 1999;282:2012–2018. 30. Stampfer MJ, Hu FB, Manson JE, Rimm EB, Willett WC. Primary prevention of coronary heart disease in women through diet and lifestyle. N Engl J Med. 2000;343:16 –22.

31. Daviglus ML, Stamler J, Pirzada A, Yan LL, Garside DB, Liu K, Wang R, Dyer AR, Lloyd-Jones DM, Greenland P. Favorable cardiovascular risk profile in young women and long-term risk of cardiovascular and all-cause mortality. JAMA. 2004;292:1588 –1592. 32. Daviglus ML, Liu K, Pirzada A, Yan LL, Garside DB, Feinglass J, Guralnik JM, Greenland P, Stamler J. Favorable cardiovascular risk profile in middle age and health-related quality of life in older age. Arch Intern Med. 2003;163:2460 –2468. 33. Mamun AA, Peeters A, Barendregt J, Willekens F, Nusselder W, Bonneux L. Smoking decreases the duration of life lived with and without cardiovascular disease: a life course analysis of the Framingham Heart Study. Eur Heart J. 2004;25:409 – 415. 34. Jones DW, Chambless LE, Folsom AR, Heiss G, Hutchinson RG, Sharrett AR, Szklo M, Taylor HA Jr. Risk factors for coronary heart disease in African Americans: the Atherosclerosis Risk in Communities Study, 1987–1997. Arch Intern Med. 2002;162:2565–2571. 35. Keil JE, Sutherland SE, Knapp RG, Lackland DT, Gazes PC, Tyroler HA. Mortality rates and risk factors for coronary disease in black as compared with white men and women. N Engl J Med. 1993;329:73–78. 36. D’Agostino RB, Grundy SM, Sullivan LM, Wilson P. Validation of the Framingham coronary heart disease prediction scores: results of a multiple ethnic groups investigation. JAMA. 2001;286:180 –187. 37. Lloyd-Jones DM, Wang R, Dyer AR, Yan L, Daviglus ML, Greenland P. Lifetime risks for cardiovascular or coronary death in AfricanAmericans and whites. Circulation. 2005;111:e245. 38. Davey Smith G, Neaton JD, Wentworth D, Stamler R, Stamler J. Mortality differences between black and white men in the USA: contribution of income and other risk factors among men screened for the MRFIT. Lancet. 1998;351:934 –939. 39. Fox CS, Evans JC, Larson MG, Kannel WB, Levy D. Temporal trends in coronary heart disease mortality and sudden cardiac death from 1950 to 1999: the Framingham Heart Study. Circulation. 2004;110:522–527.

CLINICAL PERSPECTIVE Current practice in primary prevention of CVD involves estimation of short-term (typically 10-year) risks for developing CVD to identify individuals at high risk who may benefit most immediately from the institution of drug therapy. However, focusing solely on short-term risk results in treatment only for older individuals with substantial risk factor burden. Younger and middle-age individuals with moderate or clearly adverse risk factor levels may have low short-term but substantial lifetime risks for development of CVD. Here, we estimated the lifetime risks for CVD and median survival associated with different clinical strata of individual risk factors and with aggregate risk factor burden at 50 years of age. Compared with participants with ⱖ2 major risk factors at 50 years of age, those with optimal levels had substantially lower lifetime risks (5.2% versus 68.9% in men, 8.2% versus 50.2% in women) and markedly longer median survivals (⬎39 versus 28 years in men, ⬎39 versus 31 years in women). It is hoped that these findings will galvanize public health and clinical efforts to prevent the development of risk factors, to increase awareness of the dangers of cardiovascular risk factors in the long term, to improve risk communication, and to promote efforts at primary prevention. We suggest that lifetime risk estimation should be used as an adjunct to 10-year risk estimation to improve patient understanding of cardiovascular risk, to identify new segments of the population who may merit preventive therapy, and to motivate lifestyle changes and adherence to therapy.
