Framingham Risk Calculator
Using Canadian Lipid Guidelines
Case Finding (Who to Screen)
About This Document
To Calculator SI Units
This is not a Palm application. It will run from the web, or it can be downloaded to your own computer and run from there as a web page.
This calculator focuses on primary prevention of cardiovascular events in people without known heart disease. Derivation tables come from guidelines recently published by the Canadian Cardiovascular Society (1) and the U.S. Preventive Services Task Force (2) for lipids and ASA respectively. The revised 2009 lipid tables from the Framingham study were used as a source for the Canadian Cardiovascular Society recommendations (3). Guidelines are largely based on randomized controlled studies, however family history recommendations are of necessity based on cohort studies, and interpretation of the literature with respect to diabetics is still somewhat controversial. ASA recommendations for women are still open to interpretation in primary prevention, and therapy should be decided on a case by case basis. See the ASA derivation tables for discussion.
Source studies are done mainly in North America. It has been shown that results may be suspect in the following groups:
- The predictive value of the Framingham Risk Score (FRS) falls with age. Results for elderly patients must be interpreted with caution (4). There is no reliable information on the utility of ASA over the age of 80.
- Although the FRS holds well for populations in Australia and New Zealand, it is less reliable in European and Asian populations (5).
- There is evidence that disadvantaged populations in lower socioeconomic circumstances may have a falsely low FRS, and may therefore not receive appropriate therapy recommendations (6).
- ASA recommendations for women vary for primary prevention. This calculator uses the Framingham cardiovascular disease tables rather than stroke tables, but all ASA recommendations for women lack consensus.
Because the FRS is currently used as a guide to both lipid and ASA recommendations for primary prevention, ASA guidelines are included in this calculator. Decision analysis for ASA is now more complex than simply treating each person with a risk over 10%. The risk of GI hemorrhage must also be taken into account, and this risk is considered in the Preventive Task Force guidelines. The guidelines are firm for men, whose risk is primarily myocardial infarction. They are less firm for women between 55 and 65, whose primary risk is stroke.
Another choice to increase precision in those with 10-19% 10-year risk is the use of hsCRP. In males over age 50 and females over age 60 a hsCRP over 2 has been shown with good evidence to change risk status when LDL is below 3.5 mmol/L. This is also included in the calculator as an option, and the recommendation for this test will be flagged if the patient meets the criteria, and the recommendation would be to treat as high risk with a statin based on the JUPITER trial (7). Patients in the moderate risk category receive a treatment recommendation if LDL if above 3.5 mmol/L.
Currently diabetics are evaluated similarly to non-diabetics unless a woman is over 50 or a man over 45. Over these age cutoffs they are considered high risk (>20%). Under this age with one additional risk factor they are also considered high risk. The calculator will flag these conditions. Remember that there are risk factors which are not put into the calculator.
Decision analysis for primary prevention has become more evidence-based and complex. Use of tables is still possible, but it is prone to error and consumes a great deal of time. For decisions which have to be made several times a day by primary care providers, a more efficient means of calculation is necessary. Existing older calculators tend to underestimate risk substantially. Many of them are developed using U.S. units of measurement, which become confusing to those of us using SI units. This calculator incorporates the evidence in a Canadian context as of late 2008, and is available on the web or by download for unrestricted use. It is not recommended or adopted by any credible organization, but it seems to be accurate, and results can be checked against source tables by the user until there is confidence that it works properly. The javascript source code is available for those who wish to make their own adaptations. The author unfortunately does not have the programming skills to make this available in PDA format.