The Framingham risk score (FRS)

• American (National Cholesterol Education Program) and European guidelines uses the FRS to classify people for primary prevention. Those with score <10% are classified as low risk, 10-20% as moderately high risk and >20% as high risk.

• People with established CVD (cardiovascular disease) or diabetes are already at high risk of CVD events and are targeted for maximum aggressive therapy with an LDL goal of <100 mg/dl, with an optional goal of LDL<70 mg/dl in very high risk patients.¹ Those who have high triglycerides should also meet the non-HDL-C goal which is set at 30 mg/dl higher than the LDL goal.

• The FRS is derived from the landmark Framingham Heart Study (FHS) that has provided valuable insights into CAD (coronary artery disease) risk prediction. The FRS is based on the clinical experience and coronary events of 5,209 men and women (aged 28-62 years) followed since 1948 and have guided the delivery of preventive cardiology for half a century.²

• Launched in 1948, blood pressure readings for each subject were taken at two year intervals for a period of 14 years and the findings dispelled several myths that existed then and shared by some even today.³ One such myth was “ The greatest danger to a man with blood pressure lies in its discovery because some fool may try to treat it.”⁴ The first key finding from the study was that both systolic blood pressure (SBP) and diastolic blood pressure (DBP) are key markers of cardiovascular risk, especially stroke, heart failure, and heart attacks.³

• Weighted scores are assigned to 6 easily measurable constructs─ total cholesterol (TC), HDL-C, systolic blood pressure, treatment of hypertension, smoking status, age and gender.⁵

 Overestimation and Underestimation of the Risk       

• D’Agostino et al ⁸  applied FRS calculations to 6 prospectively studied, ethnically diverse populations within the US. It was discovered that the FHS performed well for whites and blacks. Among Japanese American and Hispanic men and Native American women, the FRS systematically overestimated the risk. After recalibration, taking into account different prevalence of risk factors and underlying rates of developing CAD, the authors were able to construct more accurate models for those subgroups whose risk was overestimated by the conventional method. ⁸

• The prediction of absolute risk was not very accurate when a model derived from one study was applied to a different study or population.⁹ The FRS has been shown to substantially overestimate the risk, when applied to northern Europeans (including UK, Belfast) with high risk of CAD and particularly in southern Europeans with low risk of CAD. ^10-13  Over 50% of the variance in CAD death rates in 25 years was accounted for by the difference in mean serum cholesterol.¹⁴ The overestimation was as high as 2-fold among Spaniards and 4-fold among Chinese, but the accuracy could be improved by recalibration.¹²

• The FRS systematically overestimated the absolute CAD risk in the Chinese Multi-provincial Cohort Study (CMCS). This cohort consisted of 30,121 adults aged 35-65 years at baseline and was followed for 12 years. The 10-year CAD rate of the CMCS cohort was one-fifth that of the Framingham cohort.

• For example, in the 10th risk decile, the predicted rate of CAD death in men was 20% CMCS vs. an actual rate of 3%. The proportion of Chinese people with 10-year risk exceeding 10% was 9.9% by FRS estimate but only 0.35 by the CMCS functions. ⁶ The recalibration of the FRS using the mean values of risk factors and mean CAD incidence rates of the CMCS cohort substantially improved the performance of the FRS functions in the CMCS cohort. ⁶ Nevertheless, application of recalibrated FRS models again significantly overestimated the CAD risk in both men (by approximately 97%) and women (by approximately 228%).

• Stroke is much more prevalent than CAD in China; thus, any risk prediction model only for CAD may not be appropriate in application. In another study involving 9903 Chinese followed for 17 years, of the 371 CVD events, 266 were ischemic strokes and 105 were coronary events.¹⁵

• Recently a simplified point score model has been developed and tested for estimating the 10-year integrated cardiovascular risk (Ischemic stroke and heart attack) in Chinese, in whom stroke is the predominant form of CVD. This model uses age, systolic blood pressure, smoking, total cholesterol, diabetes and BMI (body mass index).¹⁵ Since overestimation of the risk in Chinese population could result in over treatment, the Framingham risk function cannot be used without calibration.

• In sharp contrast to the Chinese the FRS underestimate the risk of CAD among Indians by 200%.⁷ See Underestimation of risk among Asian Indians.

• Recalibrating the Framingham equations to the event rate and the prevalence of the risk among Spaniards and Chinese have been shown to improve the prediction of the risk. After calibration, the FRS  function became an effective method of estimating the risk .⁸ The FRS-based coronary risk chart overestimates absolute coronary risk in populations characterized by a lower incidence of CAD events and should be used with caution.

Limitations of the FRS 

• The FRS is recommended only for patients with > 2 risk factors and is not accurate in patients with extreme end of a particular risk factor such as heavy smoking or very high cholesterol level. 

• FRS uses TC (instead of LDL-C) and does not include risk factor modification interventions, except for hypertension.  It does not include family history or south Asian ethnicity both of which are known to confer 2-fold risk of future CAD ^{16, 17}  It also does not include other identified risk factors such as overall and abdominal obesity, physical inactivity, metabolic syndrome, high lipoprotein(a), high triglycerides, non-HDL-C, apolipoprotein B, C-reactive protein, fibrinogen, left ventricular hypertrophy, estimated glomerular filtration rates etc.

• The FRS includes only CAD events but excludes other vascular events such as stroke, which are more common than heart attack in Asian populations. The scoring was developed by studying a predominantly white and middle class population and particularly useful in the elderly but not be accurate for young and people of other ethnic origin. ^{18, 19}

• The young subjects in FHS were not only underrepresented but had very few events, which led to its heavy emphasis on age and is unsuitable for predicting premature CAD (defined as 55 years and younger in men and 65 years and younger in women). In a study of 220 patients with premature heart attack (mean age 50, 25% women), almost 70% were classified as low risk by FRS and only 25% met criteria to qualify for medication prior to the heart attack. For women in this population, only 18% met criteria for treatment.²⁰

• People with metabolic syndrome have recently been found to have a 53% higher mortality than is found in people without diabetes or metabolic syndrome. The metabolic syndrome and its components help predict the risk independent of Framingham risk score in people with in diabetics and nondiabetics alike.

• 10-year risk and lifetime risk: A major limitation of FRS is that it estimates the risk of developing CAD within only a 10-yr period. Among young individuals, there is marked disparity between 10-year risk and lifetime risk in the FHS.  For example, a 50-year old man in the lowest risk tertile had a 10-year risk of only 4% but a life time risk of 50%. Ford et al  have recently estimated that only 3% of the US population would be classified as CAD risk equivalents and therefore qualify for intensive lipid therapy.²¹ Thus, FRS may not identify subjects with low short-term but high lifetime risk for CAD ²²

• These data underscore the need for continued vigilance in those who have been found to be at low short-term risk.  Likewise it may underestimate the risk in populations highly prone for premature CAD.

Sources

 1. Grundy SM, Cleeman JI, Merz CN, et al. Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation. Jul 13 2004;110(2):227-239.

2. Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation. May 12 1998;97(18):1837-1847.

3. Turnbull F., Pascal Kengne A, MacMahon S. Blood pressure and cardiovascular disease: tracing the steps from Framingham. Prog Cardiovasc Dis. Jul-Aug 2010;53(1):39-44.

4. Moser M. Historical perspectives on the management of hypertension. J Clin Hypertens (Greenwich). Aug 2006;8(8 Suppl 2):15-20; quiz 39.

5. NCEP III. Third Report of the National Cholesterol Education Program(NCEP) Adult Treatment Panel III: National Institute of Health; September2002, 2002. 02-5215.

6. Liu JL, Hong Y, D’Agostino RB, Sr., et al. Predictive value for the Chinese population of the Framingham CHD risk assessment tool compared with the Chinese Multi-Provincial Cohort Study. Jama. Jun 2 2004;291(21):2591-2599.

7. Enas  EA, Singh V, Munjal YP, et al. Recommendations of the second Indo-U.S. health summit on prevention and control of cardiovascular disease among Asian Indians. Indian Heart J. May-Jun 2009;61(3):265-274.

8. D’Agostino RB Sr, Grundy S, Sullivan LM, Wilson P. Validation of the Framingham coronary heart disease prediction scores: results of a multiple ethnic groups investigation. Jama. 2001;286(2):180-187.

9. Prediction of mortality from coronary heart disease among diverse populations: is there a common predictive function? Heart. Sep 2002;88(3):222-228.

10. Brindle P, Emberson J, Lampe F, et al. Predictive accuracy of the Framingham coronary risk score in British men: prospective cohort study. Bmj. Nov 29 2003;327(7426):1267.

11. Empana JP, Ducimetiere P, Arveiler D, et al. Are the Framingham and PROCAM coronary heart disease risk functions applicable to different European populations? The PRIME Study. Eur Heart J. Nov 2003;24(21):1903-1911.

12. Marrugat J, D’Agostino R, Sullivan L, et al. An adaptation of the Framingham coronary heart disease risk function to European Mediterranean areas. J Epidemiol Community Health. Aug 2003;57(8):634-638.

13. Menotti A, Puddu PE, Lanti M. Comparison of the Framingham risk function-based coronary chart with risk function from an Italian population study. Eur Heart J. Mar 2000;21(5):365-370.

14. Menotti A, Keys A, Kromhout D, et al. Inter-cohort differences in coronary heart disease mortality in the 25-year follow-up of the seven countries study. Eur J Epidemiol. Sep 1993;9(5):527-536.

15. Wu Y, Liu X, Li X, et al. Estimation of 10-year risk of fatal and nonfatal ischemic cardiovascular diseases in Chinese adults. Circulation. Nov 21 2006;114(21):2217-2225.

16. Lloyd-Jones DM, Nam BH, D’Agostino RB, Sr., et al. Parental cardiovascular disease as a risk factor for cardiovascular disease in middle-aged adults: a prospective study of parents and offspring. Jama. May 12 2004;291(18):2204-2211.

17. Forouhi NG, Sattar N, Tillin T, McKeigue PM, Chaturvedi N. Do known risk factors explain the higher coronary heart disease mortality in South Asian compared with European men? Prospective follow-up of the Southall and Brent studies, UK. Diabetologia. Nov 2006;49(11):2580-2588.

18. Grundy SM, Pasternak R, Greenland P, Smith S, Jr., Fuster V. Assessment of cardiovascular risk by use of multiple-risk-factor assessment equations: A statement for healthcare professionals from the American Heart Association and the American College of Cardiology. Circulation. 1999;100(13):1481-1492.

19. Grundy SM, Balady G, Criqui M, et al. Primary prevention of coronary heart disease: guidance from Framingham: A statement for healthcare professionals from the AHA Task Force on Risk Reduction. American Heart Association. Circulation. 1998;97(18):1876-1887.

20. Akosah KO, Schaper A, Cogbill C, Schoenfeld P. Preventing myocardial infarction in the young adult in the first place: how do the National Cholesterol Education Panel III guidelines perform? J Am Coll Cardiol. May 7 2003;41(9):1475-1479.

21. Ford ES, Giles WH, Mokdad AH. The distribution of 10-Year risk for coronary heart disease among US adults: findings from the National Health and Nutrition Examination Survey III. J Am Coll Cardiol. May 19 2004;43(10):1791-1796.

22. Lloyd-Jones DM, Wilson PW, Larson MG, et al. Framingham risk score and prediction of lifetime risk for coronary heart disease. Am J Cardiol. Jul 1 2004;94(1):20-24.