Traditional formulas may overestimate heart risk
Last Updated: 2003-11-28 14:32:44 -0400 (Reuters Health)
By Karla Gale
NEW YORK (Reuters Health) - The traditional risk factors used to estimate the likelihood of developing heart disease can exaggerate a person's risk, investigators report today. However, a simple calculation can restore the accuracy of these predictions.
Physicians use equations to make an educated guess that a person will have a heart attack or die of heart disease in the future so that they can decide if and when treatment should be started. One of the most popular formulas is based on data collected from residents of Framingham, Massachusetts, between 1968 and 1975.
But since that time, "it seems that populations, in developed countries at least, are becoming relatively more low risk," lead researcher Dr. Peter Brindle told Reuters Health.
To see if the old-fashioned formula still worked, Brindle, at the University of Bristol, and colleagues examined the medical records of more than 6000 British men ages 40 to 59 years, starting in 1978. Information used to calculate risk included blood pressure, cholesterol levels, diabetes, smoking status, and ECG charts.
Over the next 10 years, 677 men had a fatal or nonfatal heart attack or angina. The Framingham formula predicted that that 1062 would suffer this fate, a relative overprediction of 57%.
Brindle explained that his group adjusted for the relative overestimation dividing the Framingham scores by the amount of overprediction. "Because the overestimations were fairly consistent throughout the risk range, by doing this across the board, the estimations seem to fit [the observed rate] much better," he said.
The changing pattern could be due to multiple factors, Brindle pointed out, such as increased exercise patterns and reduced stress. "Even water hardness and work conditions can affect heart risk," he said.
But "perhaps the biggest factor since that time is that lots of people are on treatment for blood pressure and for their cholesterol levels, but the prediction then was based on people who had very little intervention," he said.
By more accurately predicting patients' risk, doctors can better advise them about life style modifications and treatment options. This is particularly important, the authors note in their article, so that patients aren't unnecessarily worried and that they can make informed choices about treatment. Too high an estimate can also affect life insurance premiums.
Making better guesstimates should also save the health care system from the costs of drugs, lab tests, and dealing with side effects from the drugs.
In a related editorial, Dr. Hans-Werner Hense, an epidemiologist at the University of Munster in Germany, maintains that accurate prediction algorithms require "up to date epidemiological data - collected in surveys, registers, and when possible, cohorts from populations with varying risk levels."
SOURCE: British Medical Journal, November 29, 2003.
http://www.reutershealth.com/archiv...128elin021.html
Traditional formulas may overestimate heart risk
Hybrid Mode
