Trends for Cardiovascular Risk Factors and Diseases in Japan

We presented trends of cardiovascular risk factors and diseases in a rural population, northeastern Japan between the 1960s and the 1990s. Population blood pressure levels declined substantially with improved medical treatment for hypertension and reduction of mean dietary sodium intake from 20g/day to 14g/day. Serum total cholesterol increased from 152mg/dl to 185mg/dI for men, and 163mg/dI to 195mg/dI for women with an increased intake of meat, egg, milk and dairy products. Age-adjusted incidence rate of stroke declined 75% while that of coronary heart disease was low and stable. The prevalence of stroke and the number of stroke patients also declined in spite of an increase of the elderly. Furthermore, the number of severely- disabled stroke patients, so called bed-ridden patients decreased a half. The substantial decline in stroke incidence, prevalence along with the decline of blood pressure levels suggest an effect of population hypertension control program conducted since 1963. In nationwide, however, an increase in the prevalence of stroke patients and the number of severely disabled patients among the elderly were reported. Primary prevention of stroke through control and prevention of hypertension should be further augmented nationwide to reduce a public health burden from the disabled stroke patients in the elderly. J Epidemiol, 1996 ; 6 : S183-S188.

We presented trends of cardiovascular risk factors and diseases in a rural population, northeastern Japan between the 1960s and the 1990s. Population blood pressure levels declined substantially with improved medical treatment for hypertension and reduction of mean dietary sodium intake from 20g/day to 14g/day. Serum total cholesterol increased from 152mg/dl to 185mg/dI for men, and 163mg/dI to 195mg/dI for women with an increased intake of meat, egg, milk and dairy products. Age-adjusted incidence rate of stroke declined 75% while that of coronary heart disease was low and stable. The prevalence of stroke and the number of stroke patients also declined in spite of an increase of the elderly. Furthermore, the number of severelydisabled stroke patients, so called bed-ridden patients decreased a half. The substantial decline in stroke incidence, prevalence along with the decline of blood pressure levels suggest an effect of population hypertension control program conducted since 1963. In nationwide, however, an increase in the prevalence of stroke patients and the number of severely disabled patients among the elderly were reported. Primary prevention of stroke through control and prevention of hypertension should be further augmented nationwide to reduce a public health burden from the disabled stroke patients in the elderly. J Epidemiol,1996 ; 6 : S183-S188.
stroke, coronary heart disease, blood pressure, serum total cholesterol, fat intake, sodium intake, hypertension control Japan had the highest mortality from stroke and the lowest mortality from coronary heart disease among developed countries in 19601). The epidemic of stroke in Japan lead to launch population-based hypertension control programs in several communities, and contributed the development of stroke epidemiology 2). We have combined stroke prevention program and observational epidemiology for community residents and occupational populations, which is the best practical way to maintain long-term studies under the support of municipal governments and companies.
This article deals with trends in cardiovascular risk factors and diseases in a northeastern rural community between 1963 and 1991 to provide past and present pictures and future perspectives of cardiovascular disease in Japan. mmHg and/or taking antihypertensive medication. Blood was drawn from seated participants into a plain, siliconized glass tube and the serum was separated. Serum cholesterol was measured by the Zak-Henly method in 1963-66, and later by the Liebermann-Burchard direct method using the Autoanalyzer II (Technicon, Terrytown, U.S.A) at Osaka Medical Center for Cancer and Cardiovascular Diseases4). The laboratory has been standardized by the Lipid Standardization Program, Centers for Disease Control, Atlanta, and successfully met the criteria of precision and accuracy of triglycerides and cholesterol measurements4).

METHODS
Height in stocking feet and weight in light clothing were measured. Body mass index was calculated as weight divided by the square of height (kg/m2). An interview was conducted to ascertain the number of cigarettes smoked per day, usual weekly intake of ethanol in unit of go (a Japanese traditional unit of volume corresponding to 23g ethanol), and use of diabetes mellitus medications.
For nutrition assessment, a weighing method was used for 12 male volunteers aged 40-59 in 1969. After the 1970s, 24hour dietary recalls were conduced for men aged 40 and 59, systematically sampled from the participants in risk factor surveys. Comparability of these two surveys was discussed else-where3), and trend in the percentage of fat intake to the total calories was examined. Stroke and coronary heart disease in men and women aged >=30 were identified by reviewing all possible cases reported from hospitals, national insurance claims, local physicians, ambulance records, death certificates, public health nurses, health volunteers, and blood pressure surveys. The completeness of the combined case-finding sources was established at the beginning of surveillance when a single, detailed household survey revealed no undetected cases 3 ). To validate the diagnoses, all living cases were visited to obtain a history from cases and/or their families, and for stroke, systematic neurological examinations were conducted by study physicians. Medical records of the local clinics and hospitals were reviewed. For deaths, histories were obtained from families and/or attending physicians and medical records were reviewed.
Stroke was defined as a focal neurological disorder with rapid onset which persisted 24 hours or more, based on Millikan's criteria') . Transient ischemic attack was not included. Final standardized diagnoses of stroke were made by a panel of three or four physician-epidemiologists, blinded to the diagnoses of other panelists. The criteria for coronary heart disease were modified from those of a WHO Expert Committee6). Painless types of coronary heart disease were not investigated because of difficulty with complete ascertainment. 'Definite' myocardial infarction was indicated by typical chest pain (lasting for 30 minutes or longer) with the appearance of abnormal and persistent Q or QS waves, or changes in cardiac enzyme activity or both. 'Suspect' myocardial infarction was indicated by typical chest pain without positive electrocardiogram and enzyme activity findings. Angina pectoris was defined as repeated episodes of chest pain during effort, especially when walking, usually disappearing rapidly after the cessation of effort or by use of sublinguinal nitroglycerin. Final diagnoses were made by a panel of three or four study physician-epidemiologists, blinded to the data of the baseline examination. Definite or suspect myocardial infarction, were combined and presented as myocardial infarction. Five-year incidence rates were calculated as the number of new cases per 1,000 census population. Table 1. Mean values (SE) of systolic and diastolic blood pressure by sex and age in 1963-1966,1972-1975,1980-1983, and 1987-1991.

Figure 1 illustrated changes in lifestyles in Ikawa between
the 1960s and the 1980s. Mean intake of meat was extremely low in the 1960s and increased double in the 1980s.
Consumption of egg, milk and other dairy foods also increased.
Mean intake of total fat (% of total calories) increased from 12% in the 1960s to 17% in the 1980s, with the doubled animal fat intake . Mean salt intake declined from 20 g/day to 14g/day. Working conditions in farming and other Table 3. The prevalence of normotension according to past or present antihypertensive medication use by sex and age in 1963-1966, 1972-1975, 1980-1983, and 1987-1991  manual work changed substantially by the wide use of machines, leading the reduction of physical activities.
Mean systolic and diastolic blood pressures declined in all age groups of both men and women between the 1960s and the 1990s ( Table 1). The blood pressure decline was larger in ages 60-79 than in the younger ages for both sexes. Distributions of systolic blood pressure for men and women aged 60-69 were shown in Figure 2. The prevalence of high systolic blood pressure (>= 200mmHg) declined first between the 1960s and the 1970s. Between the 1970s and the 1980s, the distribution shifted downward with a large decrease in the prevalence of systolic blood pressure >= l60mmHg. After the 1980s, there was no further shift of the distribution, but the prevalence of systolic blood pressure of 150-169 mmHg decreased. A similar trend was seen for other age groups and for diastolic blood pressure.
The frequency of medication use among hypertensives was 6% to 22% in all sex-age groups in the 1960s, and increased in a stepwise fashion between the 1960s and the 1980s, reaching saturation towards the 1990s ( Table 2). The frequency of medication use among all participants was 12% or less in the 1960s and increased similarly.
The prevalence of normotensives with past or present medication use increased in all sex-age groups between the 1960s and the 1980s, reaching saturation for men aged 40-59 and for  Table 3). The proportion of normotensives without antihypertensive medication use showed no change between the 1960s and the 1970s, but it did increase for all sex-age groups thereafter.
Blood pressure levels for participants and their offsprings when both of parents and offsprings were aged 40-49 were available because of long-term risk factor surveys. If either parents or offsprings used antihypertensive medication, the parent-offspring pair was excluded from the analysis. Mean values of systolic and diastolic blood pressure were significantly lower in offsprings than in parents for men and women (Table 4).
There was a significant rise in mean serum cholesterol in every age-sex group between the 1960s and the 1990s. Approximately 30 mg/dl rise was found for both men and women (Table 5).
Stroke incidence in ages 40-69 declined over 78% for men and 70% between the 1960s and the 1990s (Figure 3). The decline was larger for cerebral hemorrhage than for cerebral infarction. For ages 70 and over, stroke incidence did not decline until the end of the 1970s, but decreased thereafter. A similar trend was seen for cerebral infarction, and a consistent Table 5. Mean values (SE) of serum total cholesterol by sex and age in 1963-1966, 1972-1975, 1980-1983, and 1987-1991. (mg/dl) Differences from the 1963-1966 values: *p< .001.  -1966,1972-1975,1980-1983, and 1987-1991.  decline was seen for cerebral hemorrhage.
The prevalence of stroke per 1,000 declined significantly between 1976 and 1986 ( Figure 4). In spite of an increase in persons aged 65 and over, the number of stroke patients also decreased.
The number of severely disabled stroke patients, so called bed-ridden patients declined a half for both age groups, 50-69 and 70+ between 1976 and 1991 ( Figure 5).
Incidence of coronary heart disease was 10 to 15% of that of stroke in the 1960s and did not change substantially between the 1960s and the 1990s.

DISCUSSION
A substantial decline in stroke incidence was attributable to a large decline in blood pressure levels for both men and women of ages 40-79.
The decline in blood pressure levels between the 1960s and the 1970s was mostly due to an improvement of detection and treatment of hypertension. The larger blood pressure decline with a shift of the whole blood pressure distributions between the 1970s and the 1980s were difficult to be attributable solely to the increase of antihypertensive medication use, because there was an increase in the frequency of normotensives without med-ication use between the 1970s and the 1980s. Thus, the blood pressure decline was probably due to improvements of environmental factors such as sodium intake reduction, improvements of house heating and working conditions7 ). These non-pharmacologic effects on the blood pressure decline were also supported by the result of our parent-offspring study8). Blood pressure decline along with reduction of salt intake was observed in the national representative samples, too. According to the National Nutrition Survey, mean systolic blood pressure level declined between 1960 and 19909). Salt intake declined from 13.7g/day in 1976 to 12.5g/day in 1990 9).
The increase of serum cholesterol levels was observed across all sex-age groups, which was readily explained by the increase of fat intake. The increase of fat intake was primarily due to an increase of meat, and to a lesser extent due to an increase of milk, egg and dairy products. In the national representative samples, serum total cholesterol levels increased 13-16 mg/dl for both men and women between 1980 and 1990 9.10 ). Fat intake per capita doubled from 11% of total calory in 1960 to 25% in 19909) The decline of intracerebral hemorrhage was possibly due to the increase of serum cholesterol levels, because we observed an inverse association between low serum cholesterol levels and the increased risk of intracerebral hemorrhage 3). The incidence of coronary heart disease, however, was 10 to 15% of the stroke incidence in the 1960s, and did not change over time 3) . We assume that relatively low serum cholesterol level, although increased, has not shown any impact of increasing coronary heart disease. There has been no evidence on an increase of age-adjusted mortality of coronary heart disease in Japan, although we reported an increasing trend in the incidence of coronary heart disease among urban male employees 10) Surveillance on cardiovascular disease and its risk factors is underway in both rural and urban Japanese populations to clarify trends in the incidence of coronary heart disease as well as stroke. However, an increase in the incidence among urban male employees whose serum total cholesterol level has been around 200 mgldl warrants further surveillance and evaluation of primary prevention.
Lifestyles in rural Japan, characterized by a low intake of animal foods, a low serum cholesterol, a high sodium intake and heavy manual work, are associated with the development of severe hypertension which causes intracerebral hemorrhage, lacunar infarction and subendocardial infarction. Lifestyles in urban Japan, characterized by a high intake of animal foods, a high serum cholesterol, obesity, physical inactivity which cause large artery occlusive infarction and transmural myocardial infarction. The rural surveyed population is shifting from the pattern of rural Japan to the pattern of urban Japan, but does not show any increase of coronary heart disease.
The successful decline in the stroke incidence and prevalence in the rural community warrants the importance of a long-teen hypertension control and prevention in the community. In nationwide, however, an increase in the prevalence of stroke patients and the number of severely disabled patients in the elderly were reported 11 ). Primary prevention of stroke through control and prevention of hypertension should be further augmented to reduce public health burden from the disabled stroke patients in the elderly