BACKGROUND:

In 1985 when the study was initiated, there were surprisingly little data on the distribution of risk factors among subgroups of the population and the changes occurring over time. These issues assumed greater importance as the decline in coronary heart disease mortality focused interest on the role of primary prevention through alteration of risk factors. In review of the potential contributions to this decline, it was noted that cessation of cigarette smoking, better blood pressure control and decrease in total serum cholesterol may have accounted partially for the decline in coronary heart disease mortality. However, these observations overlooked the possibility that observations in the total population or confined to specific socioeconomic groups may fail to disclose differential changes in particular population segments. For example, Americans in the higher educational and income groups experienced rapid and marked changes in risk behaviors while less educated, lower income Americans or specific racial/ethnic groups experienced no beneficial changes. The majority of cohort studies followed rather select groups and most cross-sectional surveys also had little sociodemographic variability. However, focusing public health, educational and medical intervention efforts required that particular groups at greater risk or experiencing less beneficial secular change be identified and targeted for greater efforts.

DESIGN NARRATIVE:

Trends in blood pressure, serum lipids, and cigarette smoking were described in nationally representative samples of adolescents ages 12 to 17 and in young adults ages 18 to 24 using data from surveys conducted from 1960 to 1984 by the National Center for Health Statistics (NCHS). Trends in blood pressure, serum lipids, cigarette smoking, and obesity were described in nationally representative samples of older adults ages 35 to 75 using NCHS survey data. The national surveys used included National Health Examination Surveys (NHES-I and III), the National Health and Nutrition Examination Surveys (NHANES I and II) and Hispanic NHANES. The secular trends in these risk factors were contrasted by subgroupings of age, race, sex, family income, and education. The correlates of these risk factors were determined and secular changes over time were identified and quantified. Risk factor values and their correlates in the national population sample were contrasted with the three Hispanic populations surveyed in the Hispanic Health and Nutrition Examination Survey (HHANES). Statistical methodologies were refined including statistical software for the analysis of complete sample survey data, model fitting strategies to characterize the secular trends across four national surveys, and guidelines for the choice of analytical strategies.