Epidemiology of Stroke in a General Japanese Population: The Hisayama Study

Abstract

The Hisayama Study is an ongoing epidemiological study of stroke, coronary artery disease (CAD), and other noncommunicable diseases in a general Japanese population established in 1961. According to the longitudinal data from the Hisayama Study, average levels of systolic blood pressure among hypertensive individuals have decreased with time since 1961. In contrast, the prevalence of metabolic risk factors such as obesity, hypercholesterolemia, and glucose intolerance has increased with time. The incidence rates of ischemic stroke in this population have declined significantly as a result of improvement in hypertension management, but the proportion of atherothrombotic brain infarction (ATBI) and embolic stroke among the total ischemic stroke cases have increased probably due to the increased prevalence of metabolic risk factors and the increased number of patients with atrial fibrillation (AF) with super-aging population. Therefore, a strategy to reduce the risks of ATBI and embolic stroke by comprehensive management of their risk factors is necessary.

In this review, we first show the secular trends in the incidence of stroke and the prevalence of its risk factors using the data from the Hisayama Study. Then, the studies for the association of traditional risk factors with stroke development in the Hisayama Study are introduced. Finally, we developed risk prediction models to estimate the absolute risk of atherosclerotic cardiovascular disease (ASCVD; including ATBI and CAD) and AF, that may be used for the stratification of future risk of ATBI and AF-related stroke in clinical practice or health examination.

Introduction

Stroke is one of the major causes of physical disability and cognitive dysfunction and is the fourth leading cause of mortality in Japan. To make a clinical strategy to reduce the burden of stroke, epidemiological studies in general populations are essential to elucidate the current status of stroke incidence and its risk factors. The Hisayama Study is an ongoing observational study for stroke, coronary artery disease (CAD), and other noncommunicable diseases established in 1961 in Hisayama town, Fukuoka Prefecture, Japan¹⁾. This town in the eastern suburb of the Fukuoka metropolitan area has a population of 9,312 (estimated in April 2023). According to the National Census in Japan, the age and occupational distributions in this town have been similar to those in Japan over the past half century²⁾. Since 1961, comprehensive health examinations for the residents of Hisayama aged ≥ 40 years have been repeated annually by the town’s government and Kyushu University. We attempted to examine ＞70% of the residents in this age group in health examinations every 2–5 years to establish new cohorts for the cross-sectional and longitudinal studies. Such high participation rates enable us to accurately estimate the prevalence, incidence, and mortality rates of diseases with minimal selection bias. Each cohort was comprehensively followed up by repeated physical examinations, mail and telephone questionnaires, and an established follow-up system organized by the study team, local general practitioners and hospitals, and the town’s government. Almost all the cohort participants have completed their follow-up surveys (follow-up rates: approximately 98%). All available information about potential events of stroke among the study participants, such as neurological symptoms, imaging data (brain CT and MRI), and autopsy findings (if available), was collected and reviewed by physician members of the study team to determine the occurrence of stroke events and clinical subtypes under the standardized diagnostic criteria throughout the study period. As a result of such a comprehensive data collection and adjudication system, the Hisayama Study has offered high-quality evidence of stroke and its risk factors in a general Japanese population over more than 60 years.

In this review article, we first show the secular trends in the incidence of stroke and the prevalence of its risk factors using the data from the Hisayama Study. Then, we demonstrate some findings from the Hisayama Study that examined the association of traditional risk factors with stroke development. Finally, we introduce a risk prediction model for the estimation of the absolute risk of atherosclerotic cardiovascular disease (ASCVD), including atherothrombotic brain infarction (ATBI) and CAD, and a risk prediction model for atrial fibrillation (AF) as an important risk factor for cardioembolic stroke.

Secular Trends in Stroke Incidence and its Risk Factors

Trends in the Prevalence of Traditional Risk Factors

We compared the age-standardized prevalence rates of traditional risk factors for stroke using the data from the health examinations in 1961, 1973–1974, 1983, 1993, and 2002–2003 (Table 1)²⁾. The age-standardized prevalence of hypertension (blood pressure ≥ 140/90 mmHg or use of antihypertensive agents) in men increased from 38.4% in 1961 to 47.7% in 1983, but then demonstrated a declining trend, with a prevalence of 41.3% in 2002. In women, the age-adjusted prevalence of hypertension increased from 35.9% in 1961 to 41.2% in 1983, and then consistently decreased to 30.8% in 2002. With the spread use of antihypertensive drugs, the average systolic blood pressure among hypertensive individuals dropped significantly from 161 mmHg in 1961 to 148 mmHg in 2002 for men and from 163 mmHg in 1961 to 149 mmHg in 2002 for women. The age-standardized proportion of current smokers also decreased from 75.0% to 47.4% for men and 16.6% to 8.5% for women during the same period. In contrast, the age-standardized prevalence of obesity (body mass index ≥ 25 kg/m²) increased consistently from 7.0% in 1961 to 29.2% in 2002 for men and from 12.9% in 1961 to 23.8% in 2002 for women. Similar increasing trends were observed for hypercholesterolemia (serum total cholesterol ≥ 220 mg/dL [5.7 mmol/L]: from 2.8% to 22.2% for men; from 6.6% to 35.3% for women) and glucose intolerance (diabetes or prediabetes: from 11.6% to 54.0% for men; from 4.8% to 35.1% for women) during the same period. Such significant increases in metabolic risk factors may be explained by the changes in lifestyles, such as the westernization of dietary patterns and physical inactivity due to motorization. In this population, the age-standardized prevalence of AF, the most important embolic source of cardioembolic stroke, did not show a clear secular change (from 0.7% in 1961 to 1.1% in 2002 for men; from 0.5% in 1961 to 0.6% in 2002 for women)³⁾.

Table 1. Age-standardized prevalence rates or mean values of traditional risk factors for stroke among 5 baseline examinations of the Hisayama Study

	1961 (n= 1,618)	1973-74 (n= 2,038)	1983 (n= 2,459)	1993 (n= 1,983)	2002-03 (n= 3,108)	P for trend
Men
Number of participants	705	855	1,048	747	1,305
Age, years	55	56	57＊	61＊	61＊	＜0.001
Hypertension, %	38.4	43.1＊	47.7＊	43.7＊	41.3	0.71
Antihypertensive agents, %	2.0	8.4＊	10.9＊	14.7＊	17.5＊	＜0.001
SBP in hypertensive individuals, mmHg	161	157＊	152＊	152＊	148＊	＜0.001
DBP in hypertensive individuals, mmHg	91	90	92	88＊	89	0.01
Glucose intolerance, %	11.6	14.1	14.3＊	29.9＊	54.0＊	＜0.001
Hypercholesterolemia, %	2.8	12.2＊	23.0＊	25.2＊	22.2＊	＜0.001
Obesity, %	7.0	11.6＊	20.2＊	26.7＊	29.2＊	＜0.001
Current smoker, %	75.0	73.3	57.2＊	47.0＊	47.4＊	＜0.001
Women
Number of participants	913	1,183	1,411	1,236	1,803
Age, years	57	58＊	58	61＊	62＊	＜0.001
Hypertension, %	35.9	40.1＊	41.2＊	34.6	30.8＊	＜0.001
Antihypertensive agents, %	2.1	7.4＊	11.5＊	15.2＊	16.2＊	＜0.001
SBP in hypertensive individuals, mmHg	163	161	155＊	155＊	149＊	＜0.001
DBP in hypertensive individuals, mmHg	88	87	87	84＊	86＊	＜0.001
Glucose intolerance, %	4.8	7.9＊	7.0＊	21.0＊	35.1＊	＜0.001
Hypercholesterolemia, %	6.6	19.9＊	33.5＊	35.7＊	35.3＊	＜0.001
Obesity, %	12.9	21.5＊	23.5＊	26.2＊	23.8＊	＜0.001
Current smoker, %	16.6	10.2＊	7.4＊	4.6＊	8.5＊	＜0.001

Abbreviations: SBP, systolic blood pressure; DBP, diastolic blood pressure.

^＊P＜0.05 compared with the examination in 1961.

The table was created by the author using data presented in Hata J et al.²⁾

Trends in the Incidence of Stroke and its Subtypes

The above-mentioned secular changes in the prevalence of risk factors have been likely to affect the incidence rates of stroke and its major subtypes (i.e., ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage). Therefore, we compared the age-standardized incidence rates of stroke subtypes among 5 cohorts with different time periods (Fig.1)²⁾. The age-standardized incidence rates of ischemic stroke for both men and women and the incidence rate of intracerebral hemorrhage for men decreased significantly during the overall study periods (all P＜0.001 for trend). The main reason for the decline in stroke incidence can be explained by the improvement in blood pressure control among hypertensive individuals through the spread of antihypertensive treatment and the reduction in smoking rates.

Fig.1. Age-standardized incidence rates of stroke subtypes among 5 cohorts of the Hisayama Study

^＊P＜0.001 for trend.

The figure was created by the author using data presented in Hata J, et al.²⁾

Ischemic stroke is further classified into some clinical subtypes such as lacunar infarction, ATBI, and cardioembolic or other embolic stroke. We compared the proportion of each clinical subtype among the ischemic stroke cases developed during 3 different time periods (Fig.2)³⁾. In the first period (1961–1974), the most common clinical subtype of ischemic stroke was lacunar stroke (62%). However, the proportion of lacunar stroke among the total ischemic stroke cases has decreased with time (48% in the third period: 1988–2001). The major pathological finding of lacunar stroke is known as microatheroma or lipohyalinosis related to hypertension in perforating arteries. The improvement in blood pressure control may contribute to a decrease in lacunar infarction. In contrast, the proportion of ATBI and embolic stroke among the total ischemic stroke cases have increased with time (from 21% to 27% for ATBI; from 13% to 25% for embolic stroke from the first to the third period). These findings can be explained by the increased risk of atherosclerosis in carotids and intracranial brain arteries due to an increase in the prevalence of metabolic risk factors, and the increased number of patients with AF due to the super-aging population.

Fig.2. The proportion of ischemic stroke subtypes among the total ischemic stroke cases among 3 cohorts of the Hisayama Study

The figure was created by the author using data presented in Kubo M, et al.³⁾

The Association of Traditional Risk Factors with Incident Stroke

Hypertension and Prehypertension

Hypertension is known as the most important risk factor for cardiovascular disease (CVD) such as stroke and CAD. In addition, recent epidemiological evidence has suggested that a positive association between blood pressure levels and the risk of CVD is observed even in individuals without hypertension⁴⁾. In other words, not only hypertension but also prehypertension can be a risk factor for stroke and CAD. To elucidate this hypothesis in a general Japanese population, the association between blood pressure levels and the risks of stroke and CAD (including myocardial infarction, coronary artery revascularization, and cardiac sudden death) was evaluated using the data from the Hisayama Study (Fig.3)⁵⁾. In this study, a total of 2,634 participants without a history of CVD (stroke or CAD) who aged ≥ 40 years in 1988 were followed up for 19 years until 2007. As a result, the risks of ischemic and hemorrhagic stroke and CAD increased with higher blood pressure categories. Compared to the participants with the lowest blood pressure levels (＜120/80 mmHg), the risk of ischemic stroke increased significantly among the hypertensive participants (≥ 140/90 mmHg) and the risk of CAD was significantly higher among the severe hypertensive participants (≥ 160/100 mmHg). In contrast, the risk of hemorrhagic stroke increased significantly not only in the hypertension groups but also in the prehypertension groups (120–139/80–89 mmHg). These findings suggest that the impact of high blood pressure on ischemic and hemorrhagic stroke may be different, and lifestyle modifications, such as dietary treatment with reduced salt intake, physical exercise, and reduced body weight, are important to reduce the risk of hemorrhagic stroke among the prehypertensive individuals.

Fig.3. The multivariable-adjusted hazard ratio for stroke and coronary artery disease according to blood pressure categories: the Hisayama Study, 1998–2007

^＊P＜0.05 vs. blood pressure levels of ＜120/80 mmHg.

Adjusted for age, sex, diabetes, serum total cholesterol, serum high-density lipoprotein cholesterol, body mass index, chronic kidney disease, electrocardiogram abnormalities, smoking habits, drinking habits, and regular exercise.

The figure was created by the author using data presented in Fukuhara M, et al.⁵⁾

Hyperglycemia

The association between hemoglobin A1c (HbA1c) levels and the risk of stroke and CAD was estimated using the Hisayama cohort data (Fig.4)⁶⁾. In this study, a total of 2,851 participants without a history of CVD who aged 40–79 years in 2002 were followed up for 7 years until 2009. While no significant association was observed between HbA1c levels and the risk of hemorrhagic stroke, a significantly positive association of HbA1c levels was observed with the risks of ischemic stroke and CAD. Compared to the lowest HbA1c group (≤ 5.0%), the risk of ischemic stroke increased significantly even in the group with a prediabetic level of HbA1c (5.5%–6.4%) as well as in the group with a diabetic level of HbA1c (≥ 6.5%), indicating that hyperglycemia even within a non-diabetic level is a possible risk factor for ischemic stroke. Dietary and physical intervention for the individual with prediabetes may be important to reduce the risk of ischemic stroke.

Fig.4. The multivariable-adjusted hazard ratio for stroke and coronary artery disease according to hemoglobin A1c levels and group with treated diabetes: the Hisayama Study, 2002–2009

^＊P＜0.05 vs. hemoglobin A1c levels of ≤ 5.0% without antidiabetic medication.

Adjusted for age, sex, hypertension, serum total cholesterol, serum high-density lipoprotein cholesterol, body mass index, electrocardiogram abnormalities, smoking habits, drinking habits, and regular exercise

The figure was created by the author using data presented in Ikeda F, et al.⁶⁾

Hypercholesterolemia

Hypercholesterolemia is known as an established risk factor for ischemic stroke and CAD in Western countries. However, the results from the observational studies which examined the association between hypercholesterolemia and ischemic stroke have been inconsistent in Japan^{7, 8)}. We assessed the association of serum non-high-density lipoprotein (non-HDL) cholesterol levels and the risk of stroke and its subtypes as well as CAD using the data from the Hisayama Study (Fig.5)⁹⁾. A total of 2,452 participants aged ≥ 40 years without a history of CVD in 1983 were followed up for 24 years until 2007. While a higher non-HDL cholesterol level was a significant risk factor for CAD, no clear association of non-HDL cholesterol was observed with both ischemic and hemorrhagic stroke. However, further analysis indicated that the association of non-HDL cholesterol with the risk of ischemic stroke was different among its clinical subtypes: a higher non-HDL cholesterol level was a significant risk factor only for ATBI and an inverse association between non-HDL cholesterol level was observed for embolic stroke. Similar subtype-specific associations were observed for serum low-density lipoprotein (LDL) cholesterol levels¹⁰⁾. These findings indicate that hypercholesterolemia defined by non-HDL cholesterol or LDL cholesterol is a common risk factor for ASCVD such as ATBI and CAD, but is not associated with an increased risk of other subtypes of stroke.

Fig.5. The multivariable-adjusted hazard ratio for stroke and coronary artery disease according to serum non-HDL cholesterol levels: the Hisayama Study, 1983–2007

Abbreviation: HDL, high-density lipoprotein.

^＊P＜0.05 vs. serum non-HDL cholesterol levels of ≤ 3.14 mmol/L.

Adjusted for age, sex, systolic blood pressure, diabetes, body mass index, electrocardiogram abnormalities, smoking habits, drinking habits, and regular exercise

The figure was created by the author using data presented in Imamura T, et al.⁹⁾

Atrial Fibrillation

AF is a common arrhythmia in older people and an established risk factor for ischemic stroke. According to the 32-year follow-up data (1961–1993) with a total of 1,621 stroke-free participants of the Hisayama Study (aged ≥ 40 years), the multivariable-adjusted hazard ratios (95% confidence intervals) in the AF group compared with the non-AF group for developing ischemic stroke was 3.7 (1.8–7.3) for men and 2.8 (1.3–6.1) for women¹¹⁾. The corresponding values for embolic stroke (including cardioembolic stroke and other embolic strokes) were extremely high: 17.8 (7.3–43.2) for men and 5.9 (1.9–18.7) for women.

Although the age-standardized prevalence of AF in the Hisayama Study did not show a clear secular change³⁾, the National Surveys on Cardiovascular Diseases reported that the estimated number of Japanese people with AF has increased and is expected to increase (391 thousand in 1980; 729 thousand in 2000; and 1,081 thousand in 2030) because of the super-aging population¹²⁾. The strategy for early diagnosis and early intervention of AF is important to reduce the burden of cardioembolic stroke and other AF-related complications.

Risk Prediction Models for Atherosclerotic ASCVD and AF

As shown in the secular trends data in the Hisayama Study, the improvement of blood pressure control by the spread of antihypertensive treatment might contribute to a reduced risk of lacunar infarction and intracerebral hemorrhage. However, the proportion of ATBI and embolic stroke among the total ischemic stroke cases have increased with time, suggesting that the clinical importance of atherosclerosis and AF is increasing. In this section, we introduce risk prediction models for ASCVD (including ATBI and CAD) and AF that were developed recently using the data from the Hisayama Study.

Risk Prediction Model for ASCVD

Ischemic stroke and CAD share their risk factors, such as hypertension, diabetes, dyslipidemia, and smoking. However, the impact of risk factors on incident ischemic stroke is considered to be different among ischemic stroke subtypes. For example, hypercholesterolemia (elevated non-HDL or LDL cholesterol) is considered to be a significant risk factor for ATBI and CAD but not to be a risk factor for other stroke subtypes^{9, 10)}. Since ATBI and CAD share a common pathological mechanism (i.e., atherosclerosis in major arteries), it is reasonable to combine ATBI and CAD into a single entity as ASCVD for risk prediction. However, previously-reported risk prediction tools have not distinguished ATBI from other ischemic stroke subtypes. Therefore, we developed a risk prediction model for incident ASCVD including ATBI and CAD using the follow-up data from the Hisayama Study.

For the development of the ASCVD risk prediction model, a total of 2,454 participants of the comprehensive health examination in 1988 aged 40–84 years without a history of CVD were followed up for 24 years until 2012 ¹³⁾. During the follow-up periods, 270 participants experienced the first ASCVD events (62 cases of ATBI and 216 cases of CAD). A total of 14 potential cardiovascular risk factors were included in the first multivariable Cox proportional hazards model. After the variable selection process using the backward elimination procedure, 9 risk factors remained in the final risk prediction model (age, sex, systolic blood pressure, diabetes mellitus, serum HDL cholesterol, serum LDL cholesterol, proteinuria, smoking status, and regular exercise). This model exhibited good discrimination ability (Harrell’s C statistics 0.786). The calibration between the observed and predicted risks of ASCVD during 10 years was good (Greenwood-Nam-D’Agostino Chi-square test: P value=0.29). Based on the final risk prediction model, a simple point-based risk score for the 10-year risk of ASCVD was developed (Fig.6).

Fig.6. Simple risk score for ASCVD: the Hisayama Study (original version)

Abbreviation: ASCVD, atherosclerotic cardiovascular disease; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

The figure was reproduced by the author using data presented in Honda T, et al.¹³⁾

The Japan Atherosclerosis Society has published the Guidelines for Prevention of ASCVD (JAS Guidelines) every 5 years since 1997 ^14-16). In 2012, the risk prediction model was first applied in the JAS Guidelines for the risk stratification to determine the management or treatment targets of dyslipidemia. The risk score for CAD mortality developed by the longitudinal observational data from NIPPON DATA80 was adopted in the JAS Guidelines 2012 ¹⁴⁾ and the risk score for incident CAD from the Suita Study was applied in the JAS Guidelines 2017 ¹⁵⁾. These risk scores were not developed for the estimation of the risk of ischemic stroke or ATBI. Therefore, in the latest JAS Guidelines 2022 ¹⁶⁾, the risk score developed by the Hisayama Study¹³⁾ was adopted with some modifications for the estimation of the absolute risk of ASCVD including both ATBI and CAD comprehensively. In the modified Hisayama risk score for the JAS Guidelines, diabetes mellitus and proteinuria were removed from the original Hisayama risk score because the patients with diabetes mellitus or chronic kidney disease have been categorized in the high-risk group regardless of the total risk score in the previous JAS Guidelines^{14, 15)}. Instead, impaired glucose tolerance was added to the modified Hisayama risk score (1 point for patients with impaired glucose tolerance). Regular exercise was also removed from the risk score because the information for physical activities is not always obtained in health examination or clinical practice. As a result, the modified Hisayama risk score for the JAS Guidelines 2022 ¹⁶⁾ included 7 risk factors (age, sex, systolic blood pressure, impaired glucose tolerance, serum HDL cholesterol, serum LDL cholesterol, and smoking status). In this guideline, the estimated absolute risk of ASCVD was categorized into 3 groups according to the modified Hisayama risk score to determine the management or treatment targets of dyslipidemia.

Risk Prediction Model for AF

The data from the Hisayama Study indicated that the proportion of embolic stroke among the total ischemic stroke have increased with time. Since AF is the most important embolic source of cardioembolic stroke and the estimated number of patients with AF is increasing with the super-aging population in Japan¹²⁾, the strategy to reduce the burden of AF-related stroke is important. Earlier detection of AF is important to start intervention for AF as soon as possible and to reduce the future risk of cardioembolic stroke. Therefore, we developed a risk prediction model for incident AF using the data from the Hisayama Study.

A total of 2,442 participants of the health examination in Hisayama in 1988 aged ≥ 40 years without a history of AF were followed up for 24 years¹⁷⁾. During the follow-up periods, 230 participants experienced AF events. A total of 18 potential risk factors of AF were analyzed and 8 risk factors remained in the final risk prediction model after the variable selection process (age, sex, systolic blood pressure, waist circumference, estimated glomerular filtration rate, abnormal cardiac murmur, high R-wave amplitude on electrocardiogram, and arrhythmia other than AF). This model exhibited good discrimination ability (Harrell’s C statistics 0.785) and good calibration between the observed and predicted risks of AF (Greenwood-Nam-D’Agostino chi-square test: P value=0.87). A simple point-based risk score for the absolute risk of AF was also developed (Fig.7).

Fig.7. Simple risk score for AF: the Hisayama Study

Abbreviation: eGFR, estimated glomerular filtration rate; AF, atrial fibrillation.

The figure was created by the author using data presented in Hata J, et al.¹⁷⁾

The stratification for the future risk of AF may be possible by using this risk score in clinical practice or health examination. The comprehensive strategy for early AF detection, such as repeated electrocardiogram examinations, Holter monitoring electrocardiogram, and self-palpation or the use of a wearable pulse meter, may be recommended for individuals at a higher risk of AF to start treatment for AF itself and stroke prevention (i.e., antiarrhythmic and anticoagulant treatments and catheter ablation) as soon as possible and to reduce the future risk of cardioembolic stroke.

Conclusions

According to the longitudinal data from the Hisayama Study, the incidence rates of stroke have decreased over the past half century mainly due to improvement in blood pressure control by antihypertensive treatment and a decrease in the smoking rate. However, the prevalence rates of metabolic risk factors have increased and the proportions of ATBI and embolic stroke among the total ischemic stroke cases have increased. Comprehensive evaluation and management of various risk factors are needed to reduce ischemic stroke risk near future. The risk scores for ASCVD and AF developed in the Hisayama Study may support the strategy for the primary prevention of ischemic stroke.

Acknowledgements

We thank the residents of the Hisayama Town for their participation in the Hisayama Study and the staff of the Division of Health of Hisayama government for their cooperation. The Hisayama Study was supported in part by the Ministry of Education, Culture, Sports, Science and Technology of Japan (JSPS KAKENHI: JP21H03200, JP21K07522, JP21K11725, JP21K10448, JP22K07421, JP22K17396, JP23K09692, JP23K09717, JP23K16330, JP23K06787, and JP23K09060); by the Health and Labour Sciences Research Grants of the Ministry of Health, Labour and Welfare of Japan (JPMH23FA1006 and JPMH23FA1022); by the Japan Agency for Medical Research and Development (JP23dk0207053 and JP23km0405209); and by Japan Science and Technology Agency (JPMJPF2210). None of the funding sources had any role in the study design, data analysis, data interpretation, or manuscript preparation.

Conflict of Interest

The Authors declare that there is no conflict of interest.

References

• 1)  Ninomiya T. Japanese legacy cohort studies: the Hisayama Study. J Epidemiol, 2018; 28: 444-451
• 2)  Hata J, Ninomiya T, Hirakawa Y, Nagata M, Mukai N, Gotoh S, Fukuhara M, Ikeda F, Shikata K, Yoshida D, Yonemoto K, Kamouchi M, Kitazono T, Kiyohara Y. Secular trends in cardiovascular disease and its risk factors in Japanese: half-century data from the Hisayama Study (1961-2009). Circulation, 2013; 128: 1198-1205
• 3)  Kubo M, Hata J, Doi Y, Tanizaki Y, Iida M, Kiyohara Y. Secular trends in the incidence of and risk factors for ischemic stroke and its subtypes in Japanese population. Circulation, 2008; 118: 2672-2678
• 4)  Vasan RS, Larson MG, Leip EP, Evans JC, O’Donnell CJ, Kannel WB, Levy D. Impact of high-normal blood pressure on the risk of cardiovascular disease. N Engl J Med, 2001; 345: 1291-1297
• 5)  Fukuhara M, Arima H, Ninomiya T, Hata J, Yonemoto K, Doi Y, Hirakawa Y, Matsumura K, Kitazono T, Kiyohara Y. Impact of lower range of prehypertension on cardiovascular events in a general population: the Hisayama Study. J Hypertens, 2012; 30: 893-900
• 6)  Ikeda F, Doi Y, Ninomiya T, Hirakawa Y, Mukai N, Hata J, Shikata K, Yoshida D, Matsumoto T, Kitazono T, Kiyohara Y. Haemoglobin A1c even within non-diabetic level is a predictor of cardiovascular disease in a general Japanese population: the Hisayama Study. Cardiovasc Diabetol, 2013; 12: 164
• 7)  Cui R, Iso H, Yamagishi K, Saito I, Kokubo Y, Inoue M, Tsugane S; JPHC Study Group. High serum total cholesterol levels is a risk factor of ischemic stroke for general Japanese population: the JPHC study. Atherosclerosis, 2012; 221: 565-569
• 8)  Tanabe N, Iso H, Okada K, Nakamura Y, Harada A, Ohashi Y, Ando T, Ueshima H; Japan Arteriosclerosis Longitudinal Study Group. Serum total and non-high-density lipoprotein cholesterol and the risk prediction of cardiovascular events: the JALS-ECC. Circ J, 2010; 74: 1346-1356
• 9)  Imamura T, Doi Y, Ninomiya T, Hata J, Nagata M, Ikeda F, Mukai N, Hirakawa Y, Yoshida D, Fukuhara M, Kitazono T, Kiyohara Y. Non-high-density lipoprotein cholesterol and the development of coronary heart disease and stroke subtypes in a general Japanese population: the Hisayama Study. Atherosclerosis, 2014; 233: 343-348
• 10)  Imamura T, Doi Y, Arima H, Yonemoto K, Hata J, Kubo M, Tanizaki Y, Ibayashi S, Iida M, Kiyohara Y. LDL cholesterol and the development of stroke subtypes and coronary heart disease in a general Japanese population: the Hisayama Study. Stroke, 2009; 40: 382-388
• 11)  Tanizaki Y, Kiyohara Y, Kato I, Iwamoto H, Nakayama K, Shinohara N, Arima H, Tanaka K, Ibayashi S, Fujishima M. Incidence and risk factors for subtypes of cerebral infarction in a general population: the Hisayama Study. Stroke, 2000; 31: 2616-2622
• 12)  Ohsawa M, Okayama A, Sakata K, Kato K, Itai K, Onoda T, Ueshima H. Rapid increase in estimated number of persons with atrial fibrillation in Japan: an analysis from national surveys on cardiovascular diseases in 1980, 1990 and 2000. J Epidemiol, 2005; 15:194-196
• 13)  Honda T, Chen S, Hata J, Yoshida D, Hirakawa Y, Furuta Y, Shibata M, Sakata S, Kitazono T, Ninomiya T. Development and validation of a risk prediction model for atherosclerotic cardiovascular disease in Japanese adults: the Hisayama Study. J Atheroscler Thromb, 2022; 29: 345-361
• 14)  Teramoto T, Sasaki J, Ishibashi S, Birou S, Daida H, Dohi S, Egusa G, Hiro T, Hirobe K, Iida M, Kihara S, Kinoshita M, Maruyama C, Ohta T, Okamura T, Yamashita S, Yokode M, Yokote K; Japan Atherosclerosis Society. Executive summary of the Japan Atherosclerosis Society (JAS) guidelines for the diagnosis and prevention of atherosclerotic cardiovascular diseases in Japan: 2012 version. J Atheroscler Thromb, 2013; 20: 517-523
• 15)  Kinoshita M, Yokote K, Arai H, Iida M, Ishigaki Y, Ishibashi S, Umemoto S, Egusa G, Ohmura H, Okamura T, Kihara S, Koba S, Saito I, Shoji T, Daida H, Tsukamoto K, Deguchi J, Dohi S, Dobashi K, Hamaguchi H, Hara M, Hiro T, Biro S, Fujioka Y, Maruyama C, Miyamoto Y, Murakami Y, Yokode M, Yoshida H, Rakugi H, Wakatsuki A, Yamashita S; Committee for Epidemiology and Clinical Management of Atherosclerosis. Japan Atherosclerosis Society (JAS) Guidelines for Prevention of Atherosclerotic Cardiovascular Diseases 2017. J Atheroscler Thromb, 2018; 25: 846-984
• 16)  Japan Atherosclerosis Society. Japan Atherosclerosis Society (JAS) Guidelines for Prevention of Atherosclerotic Cardiovascular Diseases 2022. Japan Atherosclerosis Society, Tokyo, Japan, 2022 [in Japanese]
• 17)  Hata J, Nagata T, Sakata S, Oishi E, Furuta Y, Hirakawa Y, Honda T, Yoshida D, Kitazono T, Ninomiya T. Risk prediction model for incident atrial fibrillation in a general Japanese population: the Hisayama Study. Circ J, 2021; 85: 1373-1382