Nurses' Health Study (Cardiovascular Component)

BACKGROUND:

The Nurses' Health Study began in 1976, when 121,700 female registered nurses living in eleven states completed a mailed questionnaire that included items about their medical history, diet, and other risk factors for cancer. The National Cancer Institute funded the first three years of the study and subsequent years on data pertaining to cancer. In 1980 the National Heart, Lung, and Blood Institute began funding the cardiovascular portion of the study. Follow-up questionnaires have been administered every two years subsequently. Blood samples were collected on a subsample of the women between 1989-1990 and stored for analysis of germline DNA and circulating biomarkers.

The cardiovascular component is funded through May 2029.

DESIGN NARRATIVE:

In 1976, 121,700 nurses ages 30-55 completed and returned the initial Nurses' Health Study questionnaire and comprise the Nurses' Health Study cohort. The cohort has been followed by means of biennial mailed questionnaires. A total of 109,413 participants responded to the 1978 questionnaire. In addition, 400 women who replied in 1976 stated that they did not wish to continue in the study and 390 women died. In 1980, the first major dietary assessment was added to the questionnaire. In the 1978-1980 interval, there were 624 deaths, 300 women declined further participation, approximately 99,000 completed the full questionnaire including the dietary questionnaire, and another 4,000 completed an abbreviated questionnaire, for a total of about 103,000 responses. In 1982, the overall response was 101,174. The approximately 19,000 living non-respondents were followed by telephone, yielding 71 percent of the non-respondents. Thus, at the end of 1982, current information was available on 116,150 or 95.4 percent of the original cohort. A subset of 32,826 women provided blood samples between 1989 and 1990. Genotyping on 1,186 women (382 with coronary heart disease and 804 matched controls) was performed using Affymetrix 6.0 platform.

Deaths in the cohort were usually reported by next-of-kin or postal authorities. At the completion of each mailing cycle, the National Death Index was searched for names of non-respondents who might have died. By comparing deaths ascertained from independent sources, the study ascertained an estimated 98 percent of deaths. Death certificates were obtained from state vital statistics departments to confirm all reported deaths. For all death certificates indicating possible cardiovascular disease, permission to obtain further information was requested from family members. Confirmed coronary heart disease death required additional information beyond the death certificate such as autopsy reports, ECG and enzyme changes of myocardial infarction prior to death, classical chest pain immediately prior to death, prior documented myocardial infarction or angina, or prior cardiac catheterization showing severe coronary disease.

Data were collected on date of birth, weight, cigarette smoking, menopausal status and current interim use of menopausal hormone therapy. Data were also collected on incident cases of nonfatal myocardial infarction, stroke, pulmonary embolism, and angina pectoris. These items appeared on each questionnaire. In addition, nutritional parameters were assessed by a self-administered semi-quantitative food frequency questionnaire in 1986 and again in 1990.

The study was renewed in 1993 with particular attention given to dietary antioxidants, other nutritional factors, physical activity, regional fat distribution, menopausal estrogen and progestin therapy, and biochemical markers including plasma lipids and apoproteins. Follow-up for nonfatal cardiovascular disease events is over 92 percent for the original 121,700 participants. Fatal cardiovascular events are documented by death certificates and confirmed and classified by review of hospital records, autopsy reports, and interviews with next of kin. Searches of the National Death Index for all nonrespondents ensure the identification of remaining deaths, resulting in mortality follow-up that is more than 98 percent complete.

The study was renewed in FY 2002 to continue follow-up through March 2007. A major aim of that renewal was to compare the predictive capability of several biochemical and genetic markers of inflammation and endothelial activation for coronary heart disease versus stroke in women: C-reactive protein (CRP), E-selectin, intercellular adhesion molecule-1, endothelin-1, and polymorphisms of the CRP and E-selectin genes. The study also continues the investigation of lifestyle determinants of cardiovascular disease, including menopausal hormone therapy (dose, formulation, and duration of use) and alcohol consumption (dose and beverage type), in the full cohort, and interactions of these exposures with the above biomarkers and with novel genetic markers (prothrombin and alcohol dehydrogenase-3 gene polymorphisms). Genotypes from 1,186 women typed as part of a genome-wide association study of coronary heart disease were used to study genetic factors associated with cardiovascular disease and the interplay of genes and diet on cardiovascular risk. Subsequent renewal cycles included study of plasma adipokines and metabolomic predictors, including gut flora metabolites such as trimethylamine N-oxide, choline, and L-carnitine, of coronary heart disease; and comprehensive metabolomic profiling of coronary heart disease cases and controls and development of metabolomic risk scores for coronary heart disease. During the current funding cycle, researchers will conduct comprehensive proteomic profiling of coronary heart disease risk and the relationship to diet and other lifestyle factors.