Many traditional ways of measuring your risk for having a heart attack are not nearly as accurate as many believe. For example, many believe that our total cholesterol and LDL are the best indicators for heart disease, but the association is weak. As far back as 1964, Garrett, Horning, and Creech found no definite correlation between serum cholesterol levels and the nature and extent of atherosclerotic disease. They published their findings in the Journal of the American Medical Association. The famous Framingham heart study concluded in 1977 that HDL was a much better predictor of heart disease than LDL, and total cholesterol was not associated with any risk. In 1997, The Lancet published a paper on the elderly that showed for each 18 mg/dl increase of cholesterol, the chance of death went down by 15%. Amazingly, they also found that mortality from cancer and infection was significantly lower among the participants in the highest total cholesterol category. We believe these readings to be the best way to predict heart disease because we have been told to. Today I will show you some more statistical hocus pocus that helped promote these tests, and tell you about a much more accurate one you should get.