Goal 2: Reduce Human Disease

Should clinical primary prevention of ASCVD be guided by subclincal disease or estimated risk?

Current approaches to guiding use of clinical primary prevention interventions, e.g., statins and aspirin, are based on treating patients who exceed a specific risk threshold. The performance of risk estimation is good, but not outstanding, and results from clinical and population studies continue to support the value of new biomarkers. Given the widespread use of preventive therapies, the lack of untreated cohorts is a major barrier to observational evaluation of new risk assessment approaches and biomarkers. A randomized trial is needed to test whether clinical primary prevention interventions are more cost-effectively guided by data on presence of subclinical disease (e.g., coronary artery calcium by CT) than by estimated risk.

Tags (Keywords associated with the idea)

Is this idea a Compelling Question (CQ) or Critical Challenge (CC)?: Compelling Question (CQ)

Details on the impact of addressing this CQ or CC:

The size of the US and global population qualifying for treatment with a statin or aspirin for primary prevention of ASCVD is immense. Given the performance of risk estimation, even if risk estimation were universally implemented, patients would be misclassified with the consequence of being under or over treated. If treatment based on presence of subclinical disease is more cost-effective, the benefits of preventive therapies can be enjoyed by larger proportions of our population and more ASCVD can be averted. Given the ionizing radiation, albeit low intensity, associated with CT scanning, it is incumbent on the biomedical research community to document the advantages, if any, of a subclinical disease guided approach to provision of clinical primary prevention services for ASCVD.

Feasibility and challenges of addressing this CQ or CC:

Many people will be concordant for the two methods of guiding provision of therapy, about 65% of middle aged and older adults. That is, many people will be high risk and have subclinical disease and many people will below risk and not have subclinical disease. It is only the discordant people, i.e., high risk people without subclinical disease and low risk people with subclinical disease, who will be informative study participants. Hence, many people will need to be screened to identify the roughly 35% who are discordant, and would be treated differently by the two approaches.


People may be unwilling to accept randomization once they know the information about their estimated risk and presence or absence of subclinical disease. If a low participation rate among eligible persons is observed, an even larger population of screenees would be needed.


A vanguard phase could provide information about these potential challenges.

Name of idea submitter and other team members who worked on this idea: David Goff, Donald Lloyd-Jones, Phil Greeland.....


-3 net votes
6 up votes
9 down votes
Idea No. 546