In November, the American Heart Association and the American College of Cardiology released new guidelines for managing cardiovascular (CV) risk in adult patients. The new set of guidelines are driving physicians to use a new risk calculator that is built on complex algorithms based on a large number of historic randomized clinical trials. But still missing from the equation is the use of big data analytics to tailor that risk to individual patients.
The 2013 American Heart Association (AHA)/the American College of Cardiology (ACC) Pooled Cohort Equations CV Risk Calculator is designed to more accurately quantify heart disease risk compared to older cardiovascular risk calculators. However, some critics have commented that the new risk calculator may overestimate risk and suggest that many people might need cholesterol-lowering statin drugs, even if they do not really need it. The medical community is confused about the clinical implications behind the Pooled Cohort Equations CV Risk Calculator and the AHA and ACC have said that they may need to make modifications to it.
The future of heart disease prevention and treatment will include [personalized genetic profiles], but we are not there yet.
Today’s health risk calculators, like the 2013 AHA/ACC Pooled Cohort Equations CV Risk Calculator, do not incorporate one of the most significant elements of human health: personalized genetic profiles. The future of heart disease prevention and treatment will include such information, but we are not there yet. Everyone can fill out a risk questionnaire and enter their age, height, weight and smoking status. But most patients do not have access to their personal genetic profile information. Moreover, clinical trials have not studied the relationships between all of our genetic profile information to major cardiovascular outcomes such as heart attacks and stroke.
If we look at the 2013 calculator, we will see that it only captures nine patient variables as of Nov. 18, 2013, and the AHA/ACC may make modifications to this calculator based on comments and feedback. The old AHA Heart Attack Risk Calculator was based on 17 patient variables including age, weight, waist size, LDL cholesterol and more. Keep in mind that the old calculator was also based on a 1948 research study called the Framingham Heart Study. The newer 2013 risk calculator attempts to combine information from multiple research studies.
By capturing only nine variables, the 2013 calculator fails to quantify CV risk based on some important pieces of personal information. For example, the calculator only provides two designations for race. It will give you one score if you select African American and another score if you select white or other. It is hard to believe that the calculator is stratifying cardiovascular risk based on only two categories of race.
The CV risk calculator does not ask about body weight and height, so it does not take into account whether a person may be overweight or obese. Something seems to be missing there given that the medical community has known for many years that overweight people have a higher risk of developing heart disease.
The calculator does not ask about family history of heart disease. Medical students learn early in their training that a strong family history of heart disease may increase the risk of heart disease. Genetics play a strong role in our overall health.
The calculator does not ask about alcohol consumption. The medical community knows that heavy drinking increases the risk of heart disease and stroke.
I could go on about the new CV risk calculator, but my point is that future health calculators will capture more than nine patient variables, and they will incorporate our personal genomic profile to calculate CV risk. Life insurance companies have developed very elaborate predictive models based on hundreds of variables. This is why life insurance application forms have so many questions about your health, your family and your habits. This is also why online health risk assessments include a very long list of questions.
We are living in a data-driven world. I would like to think that future versions of health risk calculators should be incorporating more patient variables, not fewer. We are entering an era where we will be using aggregate patient data to build more accurate predictive models while protecting individual patient privacy. Once medical researchers have genetic profile information on millions of heart disease patients, supercomputers will be able to analyze and interpret that big data to generate a highly accurate risk calculator that incorporates all of these variables.
I would suspect that future CV risk calculators should ask for more than nine patient variables. Asking for 50 or even 100 variables may seem laborious, but if most of the patient data is stored on an electronic health record, then a CV risk calculator that uses 50 or 100 variables could get populated automatically with a click of a button.
About the author:
Joseph Kim is a physician technologist who has a passion for leveraging health IT to improve public health. Dr. Kim is the founder of NonClinicalJobs.com and is an active social media specialist. Let us know what you think about the story; email firstname.lastname@example.org or contact @SearchHealthIT on Twitter.
This was first published in November 2013