Article Text

Download PDFPDF

Predicting the future: a matter of time
  1. Jarett D Berry1,2
  1. 1Division of Cardiology, Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, Texas, USA
  2. 2Department of Clinical Sciences, University of Texas Southwestern Medical Center, Dallas, Texas, USA
  1. Correspondence to Dr Jarett D Berry, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390-9047,USA; jarett.berry{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Cardiovascular disease (CVD) risk prediction has been the cornerstone of prevention practice for decades because it identifies the individuals most likely to benefit from therapeutic intervention. While the relative risk reduction observed with lipid-lowering therapy is similar for patients with low and high absolute risk, the clinical impact of lipid-lowering therapy depends on the baseline absolute risk and consequently, the absolute risk reduction. It is for this reason that both US1 and European2 prevention guidelines emphasise global CVD risk estimation as a guide to treatment decisions.

While there is broad agreement on the importance of absolute risk, the clinical application of these risk estimates has been the subject of considerable debate. In the past, statin therapy was more expensive and therefore guidelines recommended treatment for only those at the higher end of the risk distribution. However, with generic statins and decades of safety data, guidelines have lowered risk thresholds with many more patients now candidates for therapy, giving rise to new concerns about treating too many, rather than too few patients.

Despite these changing trends in short-term risk treatment thresholds, other more fundamental challenges persist—namely, age remains the dominant determinant of short-term absolute risk estimates. In particular, most young individuals are considered ‘low risk’ by short-term CVD risk equations regardless of the burden of traditional risk factors.3 In a previous study in men aged less than 30 years, we observed that the highest risk group with a total cholesterol of 226 mg/dL, a systolic blood pressure of 146 mm Hg and a 93% prevalence of smoking were low risk with a Framingham Risk Score< 10%.3 Despite the fact that virtually all young adults are ‘low risk’ regardless of risk factor burden, CVD remains the number one cause of death. Thus, this approach to CVD risk prediction—while justifiable—remains at odds with the cumulative, lifelong nature of the atherosclerotic process.

A variety of novel approaches have been identified to address this problem of age, including several adjuncts to short-term risk equations such as relative risk estimates, lifetime risk and vascular age.4 In the present paper by Jorstad,5 the authors explore an alternative strategy proposed by the European Guidelines to address some of the particular limitations to the CVD mortality risks derived from the Systematic COranary Risk Evaluation. More specifically, the guidelines suggest that multiplying the CVD mortality risk estimate by three can serve as a rough estimate for the risk for total CVD. This is particularly relevant for younger adults in whom the distribution of total CVD events is weighted towards non-fatal events. Nevertheless, the guidelines note a word of caution, suggesting that the use of a fixed multiplier to estimate total CVD can be unreliable because of the impact of age on the distribution of fatal and non-fatal CVD events. The guidelines also caution that the multiplier can be greater than physicians expect because of the impact of subsequent, downstream events that occur after the incident event.

In the present paper,5 the authors confirm these concerns with a more comprehensive analysis of the distribution of fatal and non-fatal CVD events using data from the large and well-characterised European Prospective Investigation of Cancer and Nutrition-Norfolk cohort. In this study, the authors observed that the ratio of total to fatal CVD was dramatically higher in younger women (28.5-fold) and in younger men (11.7-fold). It was only in the oldest age groups that the authors observed the more expected threefold gradient between total and fatal CVD events. The authors conclude that the use of the threefold multiplier to estimate total CVD events should be avoided, particularly in younger adults. The authors also suggest more broadly that the use of CVD mortality in younger adults underestimates the total burden of CVD, and therefore may not represent the best approach to guide preventive practice in these patients.

While this appears to be the case in the short term, the discordance between fatal and total atherosclerotic cardiovascular disease (ASCVD) is less apparent across the lifespan. In the Cardiovascular Lifetime Risk Pooling Project,6 for example, among men and women aged 45 years, we observed that the ratio of the lifetime risks for total CVD to CVD mortality ranged from 1.5 to 2 across the risk spectrum (see figure 1). Therefore, when viewed across the long term, CVD mortality represents a reasonable outcome to guide preventive therapy, suggesting that the more fundamental challenges with risk estimation in younger adults is not the outcome but rather the time horizon that is considered relevant for estimating risk.

Figure 1

Ratios of lifetime risk for total CVD to the lifetime risk for CVD mortality across selected risk factor groups in men and women aged 45 years in the cardiovascular lifetime risk pooling project. CVD, cardiovascular disease; RF, risk factor.6

The application of lifetime risk estimates as a risk communication strategy is a non-controversial approach that is now part of the new American Heart Association/American College of Cardiology (AHA/ACC) guidelines.1 What is controversial—but also desperately needed—is more clarity regarding the use of lifetime risk estimates to guide treatment decisions among younger adults not already candidates for lipid-lowering therapy. With emerging data confirming stagnation in the decline in coronary heart disease in younger adults,7 more thoughtful approaches are needed to address the inevitable burden of CVD among younger adults with low short-term risk but high lifetime risk for CVD.


View Abstract


  • Contributors JDB is solely responsible for the present work. The corresponding author had full access to all data in the study and had final responsibility for the decision to submit for publication. The author has read and agrees to the manuscript as written.

  • Funding JDB receives funding from (1) the Dedman Family Scholar in Clinical Care endowment at the University of Texas Southwestern Medical Center and (2) 14SFRN20740000 from the American Heart Association prevention network.

  • Competing interests JDB reports receiving speakers bureau honoraria from Merck.

  • Provenance and peer review Commissioned; internally peer reviewed.

Linked Articles