N Engl J Med 2012; 366:1641-1643April 26, 2012

Article

To the Editor:

In their article (Jan. 26 issue), Berry et al.1 reported on the lifetime risks of cardiovascular disease in the United States based on a total cohort of 257,384 men and women. Our own research estimating the lifetime risk of cardiovascular disease in England and Wales was based on a cohort of 2,343,759 persons, including 113,925 who were nonwhite.2 We validated our algorithms in a separate cohort of 1,267,159 people from England and Wales. Like Berry et al., we found that diabetes, smoking, and higher levels of cholesterol and blood pressure were associated with increased lifetime risks of cardiovascular disease. We also took into account body-mass index, socioeconomic variables, family history of heart disease, rheumatoid arthritis, atrial fibrillation, and renal disease, which were associated with increased risk. Unlike Berry et al., we found differences in risk between ethnic groups even after accounting for differing levels of risk factors, with substantially higher risks in South Asians and marginally lower risks in black populations as compared with white ones. Overall estimates were broadly similar in the two studies.

Carol Coupland, Ph.D.
Julia Hippisley-Cox, M.B., Ch.B., M.D.
University of Nottingham, Nottingham, United Kingdom
carol. coupland@nottingham. ac. uk

Peter Brindle, M.B., Ch.B., M.D.
NHS Bristol, Bristol, United Kingdom

Dr. Coupland reports being a consultant statistician for ClinRisk. Dr. Hippisley-Cox reports being co-director of QResearch and director of ClinRisk. No other potential conflict of interest relevant to this letter was reported.

2 References

2. 1

   Berry JD, Dyer A, Cai X, et al. Lifetime risks of cardiovascular disease. N Engl J Med 2012;366:321-329
   Full Text | Web of Science | Medline

3. 2

   Hippisley-Cox J, Coupland C, Robson J, Brindle P. Derivation, validation, and evaluation of a new QRISK model to estimate lifetime risk of cardiovascular disease: cohort study using QResearch database. BMJ 2010;341:c6624-c6624
   CrossRef | Web of Science

To the Editor:

Berry et al. present a meta-analysis of lifetime risk for cardiovascular disease, calculated with the use of a Practical Incidence Estimator, which uses "age" as the time scale (as opposed to "calendar period," as used in Kaplan–Meier analyses). It allows the combination of subject information enrolled at different ages, with varied follow-up.1 However, some germane points remain to be addressed. First, the proportions of risk factors measured at one time point (subject to inaccuracies) versus those averaged over a number of measurements were not reported. Second, given that the average follow-up was just 10 years, most of the data contributing to lifetime-risk estimates at age 55 years were not from the actual 55-year-olds themselves but rather from the 65-year-olds and 75-year-olds initially allocated to the same risk-factor category. Related to this, the allocation to risk-factor categories was (unidirectionally) fixed in this analysis. For example, 70-year-olds initially enrolled in the category "all risk factors optimal" are presumed to have always been in this category throughout their prior lifespan. Clearly this is not always the case in real life. How this consideration influenced the results was not discussed. Finally, would the authors speculate as to why the percentage of men in the all-risk-factors-optimal category remained constant with age (at about 2.9%)?

John W. McEvoy, M.B., B.Ch.
Johns Hopkins Ciccarone Center for the Prevention of Heart Disease, Baltimore, MD
jmcevoy1@jhmi. edu

No potential conflict of interest relevant to this letter was reported.

1 References

2. 1

   Vasan RS, Beiser A, Seshadri S, et al. Residual lifetime risk for developing hypertension in middle-aged women and men: the Framingham Heart Study. JAMA 2002;287:1003-1010
   CrossRef | Web of Science | Medline

To the Editor:

Berry et al. performed a meta-analysis using data from 18 unique cohort studies that focused on lifetime risks of cardiovascular disease. Their findings convincingly reinforce the importance of traditional coronary risk factors with respect to the lifetime risks of cardiovascular disease. Beside age, sex, and race, risk-factor profiles included diabetes, smoking, total cholesterol level, and systolic blood pressure. In my view it is a real pity that some measures of physical activity consistently remain unconsidered in otherwise large and excellent studies. There is ample evidence that physical activity is an independent and important factor in assessing the risk of cardiovascular disease and that increasing levels of physical fitness protect against elevations in most risk factors in subjects with or without cardiovascular disease, thus reducing mortality.1,2 Despite these facts, physical activity is often ignored in large epidemiologic studies, as well as in the general cardiovascular risk–assessment profile.3 A consideration of physical activity would also help focus attention on important nonpharmacologic preventive measures, such as the change from a sedentary to a more active lifestyle.

Martin Burtscher, M.D., Ph.D.
University of Innsbruck, Innsbruck, Austria
martin. burtscher@uibk. ac. at

No potential conflict of interest relevant to this letter was reported.

3 References

2. 1

   LaMonte MJ, Eisenman PA, Adams TD, Shultz BB, Ainsworth BE, Yanowitz FG. Cardiorespiratory fitness and coronary heart disease risk factors: the LDS Hospital Fitness Institute cohort. Circulation 2000;102:1623-1628
   Web of Science

3. 2

   Myers J, Prakash M, Froelicher VF, Do D, Partington S, Atwood JE. Exercise capacity and mortality in men referred for exercise testing. N Engl J Med 2002;346:793-801
   Full Text | Web of Science | Medline

4. 3

   Vasan RS, Kannel WB. Strategies for cardiovascular risk assessment and prevention over the life course: progress amid imperfections. Circulation 2009;120:360-363
   CrossRef | Web of Science

Author/Editor Response

We are aware of the valuable study by Hippisley-Cox et al., in which information on both traditional and nontraditional risk factors was obtained between 1994 and 2010 and participants were followed for up to 16 years.1 Their approach should be distinguished from ours, in which we report lifetime risks on the basis of directly measured risk factors and observed outcomes among participants in the United States over the past 50-plus years. Our approach allows us to provide insight into an important clinical question: What is the effect of a given risk-factor profile in middle age on cardiovascular disease events decades later? Cardiovascular disease in later life reflects the cumulative burden of risk factors across the lifespan, and the benefit of a low risk-factor burden is greatest at younger ages. Nevertheless, the findings from the QResearch database and those from our study are, overall, quite similar.

We would like to make several important clarifications with respect to McEvoy's comments about our methodologic approach. The choice of time scale for Kaplan–Meier and other techniques for survival analysis can vary, and it is well established that the "time on study" approach and the "attained age" approach yield similar results when persons who are disease-free are followed after a selected index age.2 Our approach uses age as the time scale, but we did not average risk-factor levels across the lifespan, as we clearly state in our article: "Participant data were stratified according to risk-factor levels or status as assessed within 5 years of each index age. For example, risk factors measured for participants between 40 and 49 years of age were included in the analyses for the age of 45 years." All reported lifetime-risk estimates for each age group are derived only from the participants with risk factors measured at that reported age. The relatively constant prevalence of low-risk status in men reflects the effects of changes in individual risk factors across the age spectrum. For example, when men with risk factors measured at 45 and 75 years of age are compared, smoking is more prevalent at the younger age (51.0% vs. 20.7%), whereas diabetes is more prevalent at the older age (2.8% vs. 11.7%).

Finally, we appreciate the insightful comments from Burtscher on the importance of physical activity on lifetime risk for cardiovascular disease. Recently, we reported on the association between physical-fitness levels and lifetime risk for cardiovascular disease in the Cooper Center Longitudinal Study using this same analytic approach.3 We observed that physical fitness represents an important determinant of long-term risk, particularly among persons with established risk factors.

Jarett D. Berry, M.D.
University of Texas Southwestern Medical Center, Dallas, TX

Donald M. Lloyd-Jones, M.D.
Northwestern University Feinberg School of Medicine, Chicago, IL
dlj@northwestern. edu

Since publication of their article, the authors report no further potential conflict of interest.

3 References

2. 1

   Hippisley-Cox J, Coupland C, Robson J, Brindle P. Derivation, validation, and evaluation of a new QRISK model to estimate lifetime risk of cardiovascular disease: cohort study using QResearch database. BMJ 2010;341:c6624-c6624
   CrossRef | Web of Science

3. 2

   Pencina MJ, Larson MG, D'Agostino RB. Choice of time scale and its effect on significance of predictors in longitudinal studies. Stat Med 2007;26:1343-1359
   CrossRef | Web of Science

4. 3

   Berry JD, Willis B, Gupta S, et al. Lifetime risks for cardiovascular disease mortality by cardiorespiratory fitness levels measured at ages 45, 55, and 65 years in men. J Am Coll Cardiol 2011;57:1604-1610
   CrossRef | Web of Science