摘要：Objectives. We estimated the distribution of predicted long-term cardiovascular disease (CVD) risk among young adults in the United States. Methods. Our data were derived from National Longitudinal Study of Adolescent Health participants (n = 14 333; average age: 28.9 years). We used a Framingham-derived risk prediction function to calculate 30-year risks of “hard” and “general” CVD by gender and race/ethnicity. Results. Average 30-year risks for hard and general CVD were 10.4% (95% confidence interval [CI] = 10.1%, 10.7%) and 17.3% (95% CI = 17.0%, 17.7%) among men and 4.4% (95% CI = 4.3%, 4.6%) and 9.2% (95% CI = 8.9%, 9.5%) among women. Average age-adjusted risks of hard and general CVD were higher among Blacks and American Indians than among Whites and lower among Asian/Pacific Islander women than White women. American Indian men continued to have a higher risk of general CVD after adjustment for socioeconomic status. Four percent of women (95% CI = 3.6%, 5.0%) and 26.2% of men (95% CI = 24.7%, 27.8%) had a 20% or higher risk of general CVD. Racial differences were detected but were not significant after adjustment for socioeconomic status. Conclusions. Average CVD risk among young adults is high. Population-based prevention strategies and improved detection and treatment of high-risk individuals are needed to reduce the future burden of CVD. Estimated costs of cardiovascular disease (CVD) in the United States are expected to be more than $1 trillion by 2030. 1 This fact, along with growing evidence of the cost effectiveness of primordial and primary prevention, 2 has prompted increasing attention to the extent of CVD risk early in life. Elevated CVD risk is detectable in childhood 3 and pronounced by young adulthood; among young adults 18 to 25 years of age, 34.2% report current smoking, 4 and 23.8% and 16.8% are overweight and obese, respectively. 5 Hypertension rates are estimated to be as high as 7.2% among those 18 to 24 years of age 5 and may be as high as 20.9% among those 25 to 32 years of age. 6 These individual statistics, although alarming, do not account for the co-occurrence of risk factors and the differing strength of their relationship to CVD as is done with risk prediction. The application of risk prediction functions to large epidemiological data sets could serve as a useful indicator of the burden of CVD among young adults. 7 However, most functions were designed to be used with middle-aged or older adults, and they predict risk over a 5- to 10-year time period. This time frame is too short for most young adults given that incident disease will occur over decades, rather than years, after assessment. Existing research involving data from the National Health and Nutrition Examination Survey has shown that the overwhelming majority (82%) of US adults have a low 10-year risk of coronary heart disease (CHD), but nearly two thirds of these individuals have a high long-term risk of CVD. 8 Longer-term risk assessments are a better predictor of subclinical 9,10 and clinical CVD 11 than shorter-term risk prediction functions and take account of competing causes of death, thereby providing a more realistic assessment of the overall burden of CVD. 7,11 Existing longer-term estimates suggest a high lifetime risk of CVD among both men (60%) and women (56%). 12 However, these estimates have been made in reference to individuals 45 years of age. Although considerable data are available on levels of individual risk factors among young adults, concurrent consideration of multiple risk factors better discriminates longer-term risk than any single risk factor. 13 To our knowledge, no estimates of this type currently exist for the US young adult population. Thus, estimates of CVD risk among young adults are needed to more accurately estimate the extent of risk in this population and to more accurately predict future disease burden. The 30-year Framingham risk score (30-year FRS) is the only longer-term risk prediction function designed to be used with young adults. The function was developed with the Framingham offspring cohort, and analyses of these data represent the only estimates of the extent of long-term risk among young adults. However, the participants in that study were recruited in the 1970s, and CVD risk factors have since changed significantly; there has been an increase in average body mass index (BMI) 14 and in the prevalence of diabetes, 15 whereas the prevalence of smoking has decreased. 16 A more accurate estimate of the extent of long-term CVD risk among young adults requires an application of the 30-year risk prediction function to a contemporary sample of young adults. In this study, we began to fill this gap by using the 30-year FRS to provide the first, to our knowledge, nationally representative estimates of long-term CVD risk by gender and race/ethnicity. We also examined racial/ethnic differences in the prevalence of “high” CVD risk (a risk score of 20% or higher).
