Honolulu Heart Program

BACKGROUND:

The Honolulu Heart Program was initiated to verify reports that men of Japanese ancestry living in the United States had a much higher risk of coronary heart disease and a lower risk of stroke than Japanese men in Japan. The Honolulu Heart Program was actually a component of the larger NI-HON-SAN Study. The NI-HON-SAN Study (from the acronyms Nippon, Honolulu, and San Francisco) was initiated in 1965. The populations under investigation included members of the Honolulu Heart Program as well as men of Japanese ancestry born between 1900 and 1919 and residing in Japan and San Francisco.

In the Honolulu Heart Program, 11,148 men were identified of whom 8,006 participated in baseline examinations between 1965 and 1968. The 1965 - 1968 examination included a detailed medical and social history with specific inquiries about smoking, alcohol consumption, diet, and physical activity. A physical examination focussed on the cardiovascular system; laboratory tests included hematocrit, serum cholesterol, triglyceride, uric acid, glucose determination, and urinalysis. A twelve lead electrocardiogram and lung function testing revealed that 301 men had evidence of coronary heart disease and 111 had a stroke. The prevalence of coronary heart disease was slightly higher than in the comparable population in Japan, but the incidence rate during the first two-year period of follow-up was twice as high in Honolulu as in Japan. Stroke prevalence was much higher in Japan than in the Honolulu cohort, with a threefold difference. The Honolulu Japanese men were more obese, ate a more Westernized diet with a higher amount of animal protein, saturated fat, and sucrose and a smaller amount of complex carbohydrates. They had higher levels of serum cholesterol, triglyceride, uric acid, and glucose than Japanese men in Japan. Cigarette smoking was more common in the Japanese cohort, while the proportion of heavy smokers was greater in the Honolulu cohort. Alcohol consumption in the Japanese cohort was twice that of the Honolulu cohort. There were no differences in mean blood pressure or prevalence of hypertension. Repeat examinations were performed near the second and sixth anniversaries of the initial examinations. The six year follow-up data identified 294 new cases of coronary heart disease and 133 new cases of stroke. Between 1970 and 1972 some 2,800 of the cohort were examined and had lipoprotein determinations performed as part of the NHLBI Lipoprotein Phenotyping Project. Survivors of the lipoprotein cohort were reexamined between 1975 and 1978 near the tenth anniversary of their initial examination and again between 1980 and 1982.

In 1985, the program was enlarged to make a comparative analysis of the 12-year incidence of coronary heart disease in the Japanese and Hawaii study cohort of the original NI-HON-SAN Study, to sample 400 offspring of the study cohort ages 45 to 60 years and to complete nutrient indicators for association with blood pressure in a manner consistent with the United States-Japan study of nutrition and blood pressure. Resulting data were compared with a comparable selected group of Japanese men and with the parents of the offspring to examine familial associations. Pulmonary function data were recorded and analyzed for association with other risk factors, respiratory disease, and other major chronic diseases.

Beginning in March 1993 and continuing through December 1994, the National Institute on Aging, with its own funds, investigated aging and dementia among Honolulu Heart Program participants.

Beginning in March 1997, the study was supported by U01HL56274.

DESIGN NARRATIVE:

In this prospective cohort study, the hospital discharge rosters of all general hospitals on the Island of Oahu were reviewed twice yearly as part of the morbidity and mortality surveillance. The appropriate chart materials were abstracted for review by study physicians for evidence of myocardial infarction and stroke. Beginning in 1986, both old and new hospital discharge records were reviewed for the additional endpoints of congestive heart failure, aortic aneurysm, bronchitis, emphysema, asthma, chronic obstructive lung disease, dementia, organic brain syndrome, and Parkinson's disease. Deaths were monitored by following obituary columns as well as reviewing death certificates from the State Health Department. There was daily contact with hospitals to find out about impending autopsies in which study personnel could participate. However, only 20-25 percent of the deaths had protocol autopsies.

The fifth and final complete examination of the surviving cohort concluded in Spring, 1996 under a contract with the National Institute on Aging. While the emphasis of this exam was the investigation of dementia, NHLBI funded the recording of blood pressures and ECGs as well as the continued surveillance for morbidity and mortality. A cooperative agreement extends the surveillance for an additional five years. As part of the close-out phase of the study, frozen blood samples collected since the study's inception have been sent to a repository in Rockville, MD. A review board for screening prospective users of this resource holds quarterly meetings.

As the study group includes both migrants from Japan and Hawaii-born men, there is a spectrum of cultural lifestyles and varying grades of acculturation which are being examined as risk factors for cardiovascular and other chronic diseases. The existing baseline and follow-up examination data and the surveillance information on the incidence of most major diseases provide opportunities for a great variety of research projects concerning the predictors of cardiovascular disease, as well as studies of the broader concepts of health and disease in the aging population. The low out-migration rate and the well-defined sources of medical care have provided an ideal setting for maintaining disease morbidity and mortality surveillance; nearly 100 per cent of hospital discharge episodes have been recorded. As the study participants range in age from 75-95 in 1996, the rates of coronary heart disease, stroke, and other chronic diseases are increasing rapidly and thus are providing a large number of disease endpoints for study. Other Institutes are also currently supporting activities which utilize the original cohort.

Stored sera provide an opportunity to test new ideas and etiologic hypotheses using both prospective and case-control study designs. Protocol autopsy information can now be utilized to examine relationships among risk factors, clinical manifestations of CHD, and stroke and pathological evidence of atherosclerosis.

Beginning in 1997, a cooperative agreement extends the surveillance for an additional five years through February, 2002. As part of the close-out of the study, frozen blood samples collected since the study's inception have been sent to a repository in Rockville, Maryland.