CARDIA study

CARDIA study
Name	Coronary Artery Risk Development in Young Adults
Acronym	CARDIA
Established	1985
Location	United States
Type	Longitudinal cohort study
Founders	National Heart, Lung, and Blood Institute
Participants	~5,115 (baseline)
Duration	multi-decade

CARDIA study

The Coronary Artery Risk Development in Young Adults study is a long-term, multicenter longitudinal cohort investigation initiated in the mid-1980s to investigate the development of cardiovascular risk from young adulthood into middle age. Funded and coordinated by the National Heart, Lung, and Blood Institute and conducted at multiple clinical centers, the study has generated influential evidence on cardiovascular disease, metabolic disorders, lifestyle factors, and social determinants across decades.

Overview

The study was launched with collaboration among the National Heart, Lung, and Blood Institute, the University of Alabama at Birmingham, the Northwestern University, the University of Minnesota, the University of California, Oakland (later Kaiser Permanente centers), and the Johns Hopkins University clinical research teams. Its multidisciplinary design engaged investigators from institutions including the Harvard University–affiliated centers, the University of Michigan, and the Yale University School of Medicine, incorporating expertise from cardiology departments, epidemiology units, and community health programs. CARDIA produced landmark cohort data that intersect with major public health initiatives such as the Framingham Heart Study, the Nurses' Health Study, and population surveys by the Centers for Disease Control and Prevention.

Study Design and Methods

CARDIA used a prospective cohort design with repeated in-person examinations, telephone follow-ups, laboratory assays, imaging studies, and validated questionnaires. Baseline recruitment in 1985–1986 enrolled approximately 5,115 participants balanced by age, sex, race, and education, with follow-up visits at fixed intervals (years 2, 5, 7, 10, 15, 20, 25, 30, and beyond). Measurements included fasting lipids, blood pressure, electrocardiography, carotid intima-media thickness, coronary artery calcium by computed tomography, and oral glucose tolerance testing, complemented by behavioral inventories referencing smoking status, diet, and physical activity instruments used in studies at Columbia University, University of California, San Francisco, and the University of North Carolina at Chapel Hill. Biostatistical approaches involved Cox proportional hazards models, mixed-effects regression, and mediation analysis methods applied in collaboration with statisticians from the University of Washington and the Stanford University School of Medicine.

Participant Recruitment and Cohorts

Participants were recruited from four metropolitan areas with clinical center sites tied to academic institutions: Birmingham, Alabama (linked to the University of Alabama at Birmingham), Chicago, Illinois (linked to Northwestern University), Minneapolis, Minnesota (linked to the University of Minnesota), and Oakland, California (linked to Kaiser Permanente centers). The cohort intentionally sampled Black and White men and women aged 18–30 at baseline to permit analyses of racial, sex, and age-related trajectories comparable to cohorts from the Framingham Heart Study and the Bogalusa Heart Study. Retention strategies mirrored methods used by investigators at Johns Hopkins University and Harvard T.H. Chan School of Public Health, with community outreach influenced by practices from the Robert Wood Johnson Foundation programs.

Key Findings and Publications

Investigators published influential articles demonstrating that risk factors in young adulthood—such as elevated systolic blood pressure, adverse lipid profiles, smoking, and insulin resistance—predict subclinical and clinical cardiovascular outcomes in middle age, paralleling findings from the Framingham Heart Study and extending knowledge from the Bogalusa Heart Study. Publications have highlighted associations between early-life socioeconomic status, neighborhood characteristics studied alongside work from the Kaiser Family Foundation, and later cardiometabolic risk, and have linked diet patterns assessed with tools developed at the Harvard School of Public Health to trajectories of weight gain and diabetes. Major papers appeared in journals where researchers affiliated with Johns Hopkins University, University of Michigan, and Yale University School of Medicine frequently publish, contributing to meta-analyses coordinated with groups at the World Health Organization and policy syntheses referenced by the American Heart Association.

Impact on Public Health and Policy

CARDIA findings informed guidelines and prevention strategies advocated by the American College of Cardiology and the American Heart Association by emphasizing prevention beginning in young adulthood. Results influenced risk prediction frameworks that intersect with recommendations from the U.S. Preventive Services Task Force and informed surveillance priorities echoed by the Centers for Disease Control and Prevention. The study’s evidence on disparities contributed to policy discussions in forums involving the Robert Wood Johnson Foundation, the Institute of Medicine (now the National Academy of Medicine), and municipal public health departments in cities such as Chicago and Oakland.

Criticisms and Limitations

Critiques cite limited racial and geographic diversity relative to national samples used by the Behavioral Risk Factor Surveillance System and note cohort attrition that resembles challenges faced by long-term studies like the Nurses' Health Study. Some investigators have debated generalizability to other racial and ethnic groups beyond Black and White participants, contrasting CARDIA with broader studies supported by the National Institutes of Health and international cohorts coordinated with the World Health Organization. Measurement changes across decades and evolving assay methods—issues also encountered in the Framingham Heart Study—introduce potential for measurement bias, and residual confounding remains a caution in causal inference despite sophisticated methods developed with collaborators at Stanford University and the University of Washington.

Category:Epidemiological studies