Article Text

Download PDFPDF
OP03 Predicting cardiovascular disease mortality rates in the united states in 2030: prospective modelling approaches
  1. M Guzman-Castillo1,
  2. J Pearson-Stuttard2,
  3. J Penalvo3,
  4. C Rehm3,
  5. A Afshin3,
  6. G Danaei4,
  7. T Gaziano5,
  8. D Mozaffarian3,
  9. M O’Flaherty1,
  10. S Capewell1
  1. 1Department of Public Health and Policy, University of Liverpool, Liverpool, UK
  2. 2Division of Medical Sciences, University of Oxford, Oxford, UK
  3. 3Friedman School of Nutrition Science and Policy, Tufts University, Boston, USA
  4. 4Harvard T. H. Chan School of Public Health, Harvard University, Boston, USA
  5. 5Division of Cardiovascular Medicine, Brigham and Women’s Hospital, Boston, USA


Background While cardiovascular disease (CVD) mortality has been reducing in the US since the 1970s, CVD remains the leading cause of mortality. This reflects reductions in population level risk factors assisted by evidence based medical treatments. Rates of reduction in CVD mortality may be slowing and large shifts in population characteristics are forecasted, creating uncertainty as to future trends in CVD mortality rates as well as uncertainty regarding race CVD inequalities.

We therefore aimed to compare US CVD mortality projections to 2030 under contrasting scenarios.

Methods We used a well-established hierarchical Bayesian Age Period Cohort (BAPC) model to model the age, period and cohort effects in CVD mortality from 1982 to 2012 stratified by age, gender and race. Data sources included the National Vital Statistics System, SEER single year population estimates, and the US Bureau 2012 National Population projections

Using US population projections to 2030, we then projected coronary and stroke deaths to 2030 under contrasting scenarios.

In SCENARIO A, we held the age, period and cohort effects components constant at 2012 values.

In SCENARIO B, we allowed the age, period and cohort effects to continue along their recent, linear trends, and extrapolated the resulting trajectories to 2030.

We then compared the number of CVD deaths forecasted in 2030.

Results In scenario A, assuming that mortality rates would persist at 2012 levels, coronary and stroke deaths were projected to INCREASE by approximately 17.1% and 50.3% respectively by 2030 (equating to some 63,400 and 64,400 additional deaths per year).

By contrast, scenario B, optimistically assuming that recent declines continue, coronary and stroke mortality might FALL by approximately 21.2% and 2.7%, equating to some 78,200 and 3,100 fewer annual deaths respectively by 2030. The burden of CVD mortality is unequal by race. These inequalities vary by age and gender, however CVD mortality rates in younger (35–55 years) black Americans is double that of age matched white Americans.

Conclusion The large US declines in coronary heart disease mortality rates observed over recent decades are likely to continue. However, stroke mortality rates will fall more slowly. This has major health implications.

Our BAPC models simultaneously estimate age, period and cohort effects to project future trends. They therefore offer more plausible future predictions than the conventional indirect standardisation approaches. These more sophisticated projection methods may therefore be potentially valuable to future researchers and to policy makers.

  • Cardiovascular disease
  • mortality projections
  • inequalities

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.