Globorisk expands CVD risk estimates to 182 countries
medwireNews: The Globorisk project has published charts for the cardiovascular disease (CVD) risk assessment of people in 182 countries.
“Until now, CVD risk formulae were only developed for high-income country populations,” writes Andrew Moran (Columbia University, New York, USA) in a commentary, which accompanies the publication in The Lancet Diabetes & Endocrinology.
Globorisk previously used a laboratory-based model to produce charts for 11 high-income countries. The group has now extended this by recalibrating the risk model using country-specific CVD data to produce individual charts for those countries, which are available here, along with risk calculators.
The original risk model was based on age, blood pressure, smoking, diabetes, and total cholesterol level, but to facilitate the use of Globorisk charts in low-resource settings, the team created a separate office-based model that does not require any blood tests.
And “[t]he two risk scores are designed in a way that allows and necessitates updating as new data on mean risk factor levels and CVD rates become available,” say Goodarz Danaei (Harvard School of Public Health, Boston, Massachusetts, USA) and co-researchers.
The loss of diabetes had the largest impact on the accuracy of the office-based risk model, so that in countries such as Mexico, where diabetes is very common, the lack of this information led to an underestimation of the proportion of people at high CVD risk.
The team therefore advocates a two-stage strategy in resource-limited settings, in which the office-based risk model is used to identify definite high-risk people for interventions, plus a borderline risk group who can then undergo blood tests, thus optimizing resource use.
Separating the countries by relative affluence revealed that living in a high-income country mitigated the CVD risk associated with having risk factors. The team gives the example of a nonsmoking 65-year-old man with diabetes, systolic blood pressure of 160 mmHg, and a total cholesterol level of 6 mmol/L. If living in Japan or the USA, this man would have a 21% 10-year CVD risk, but if living in China he would have a 53% risk.
In his commentary, Moran observes that recalibrating, which pairs exposure and incidence data from the target country with risk factor–outcome associations from the original population, produces a less reliable model than would be obtained by following up a population-based cohort in the target country.
“[B]ut undertaking Framingham-equivalent cohort studies in over 100 countries is not feasible,” he observes.
Additional issues arise from the fact that target country incidence data are subject to bias, being “indirectly estimated from two sources,” says Moran.
But he stresses that, irrespective of how accurately the charts pinpoint absolute risk, they have “excellent ability” to discriminate high-risk from low-risk groups, giving them “the potential to channel countries’ scarce preventive resources more efficiently and equitably.”
medwireNews is an independent medical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2017