Letters Prediction models for cardiovascular risk

Author’s reply to Woodward

BMJ 2016; 354 doi: https://doi.org/10.1136/bmj.i4485 (Published 16 August 2016) Cite this as: BMJ 2016;354:i4485
  1. Johanna A A G Damen, PhD fellow1 2
  1. 1Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, Netherlands
  2. 2Cochrane Netherlands, University Medical Center Utrecht, PO Box 85500, Str 6.131, 3508 GA Utrecht, Netherlands
  1. j.a.a.damen{at}umcutrecht.nl

We highly appreciate Woodward’s comment on our systematic review of prediction models for cardiovascular disease (CVD) risk in the general population.1 2 Woodward refers to the development and validation of a relatively new prediction model in this field, Globorisk.

At first sight this is just another development of a model to predict CVD, which can be added to the list of 363 prediction models already described in our systematic review. As already stated in a previous comment,3 the Globorisk model does indeed have some features that make it an interesting candidate for further research. During its development Globorisk has been directly stratified by sex and cohort (from various countries across the globe) and can therefore be more easily tailored to different countries but ones similar to those included in the development set. This tailoring is possible even using only routinely available, country specific data on overall CVD risk and on population averages of the predictors. The authors clearly describe how this updating should be done, and they demonstrate this for 11 countries from various continents. As discussed in our review, to our knowledge no prediction models have yet been developed from data specifically documented from African or South American countries, which makes Globorisk an interesting candidate for this gap.

Globorisk builds on the assumption that associations between predictor effects and CVD are similar across Western and Asian populations.4 This is a unique assumption which, to our knowledge, has never been used in previous models. There is conflicting evidence for this assumption, however.5 6 7 8 9 Gijsberts and colleagues describe, for example, that the association between total cholesterol levels and CVD risk differs between race/ethnicity groups.5

External validation studies, in our view, are therefore still necessary to investigate whether this assumption truly holds and to study the transportability of the model to populations other than the cohorts used in developing Globorisk. Furthermore, we believe that head to head comparisons using individual participant data are still warranted, to compare Globorisk with other existing (and frequently advocated and validated) CVD risk prediction models, such as the Pooled Cohort Equations,10 Framingham prediction models,11 12 and SCORE.13


  • Competing interests: None declared.


View Abstract

Log in

Log in through your institution


* For online subscription