Can a “calculator” really tell doctors who should be on Lipitor?
November 13, 2013- By Steven E. Greer, MD
The AHA released the long-awaited new guidelines that recommend which patients should be on a cholesterol-lowering statin, such as Lipitor or Crestor. The panel of experts incorporated evidence-based medicine into the guidelines, perhaps better than any other medical society guideline panel has done previously. However, the panel also used some very dubious statistical tools as well.
According to the American Heart Association’s website, “The American Heart Association and the American College of Cardiology worked with other professional groups in finalizing these guidelines, and multiple stakeholder organizations were invited to review and endorse the final documents.”.
The new guidelines indicating who might benefit from statin therapy are summarized in bullet points, from the AHA website:
“The guideline recommends statin therapy for the following groups:
- People without cardiovascular disease who are 40 to 75 years old and have a 7.5 percent or higher risk for heart attack or stroke within 10 years.
- People with a history of heart attack, stroke, stable or unstable angina, peripheral artery disease, transient ischemic attack, or coronary or other arterial revascularization (also known as secondary prevention).
- People 21 and older who have a very high level of bad cholesterol (190 mg/dL or higher) (which indicates a genetic familial disease).
- People with Type 1 or Type 2 diabetes who are 40 to 75 years old.”
To the panel’s credit, they dropped the use of LDL cholesterol as a treatment target, which was never proven to be a predictor of cardiac disease outcomes, and the wrong parameter to use for statin therapy (See our interview with Dr. Harlan Krumholz of Yale, video above).
Arbitrarily chasing rigid LDL goals was never shown in clinical trials to benefit patients by reducing the chance of myocardial infraction or increasing lifespan. But the LDL goal did succeed in creating drugs, such as Lipitor, that generated tens of billions in revenue per year for Pfizer, etc.
For several years, certain respected academic cardiologists, such as Sanjay Kaul, Harlan Krumholz, and others have questioned the reliance on LDL treatment targets as not being evidence-based. To finally have it removed from the guidelines might not seem like a big deal to the outside reader, but was a monumental accomplishment for cardiologists, even if it did come decades too late.
The rest of the new guidelines are open to debate. In the first bullet point above, the decision to place people on statins who have “…a 7.5 percent or higher risk for heart attack or stroke within 10 years.” is completely arbitrary. There are no clear rationale described for making 7.5% the magical cutoff. This new component of the guidelines will potentially double the number of people on statin therapy, despite no clinical trials showing conclusive evidence that these low-risk patients will live longer.
Also new to the guidelines is a black box “calculator”. Dr. Sanjay Kaul, a prominent member of the AHA and ACC, stated, “Overall, the new guidelines are faithful to the evidence, and they are patient-centered rather than disease-focused. The risk calculator, although based on empirical evidence and sound statistical modeling, is perhaps not as evidenced-based as other recommendations. Whether clinicians will use the calculator to base their treatment decisions is going to be a big challenge. The Framingham Risk Score has been out for over 20 years and less than 5% of clinicians use it. However, if the risk calculator is embedded in the EHR to be used at the point of care, and if it is coded as a ‘billable’ activity, then I can see greater adoption of its use. Bottom line, I see these guidelines as a step in the right direction, and in compliance with the IOM standards for trustworthy guidelines.”
Indeed, few people train to become a medical doctor so that they can rely upon black box calculators to determine patient therapy.
Cardiology, Cedars Sinai LA, Late Breaking Data, Policy | apples49 | 
August 16, 2015 11:52 am
