Risk Factors for Heart Failure and Coronary Artery Disease Mortality Based on the National Vital Statistics During a 25-Year Follow-up in Japan　― NIPPON DATA90 ―

Abstract

Background: Prevention of heart failure (HF) is a public health issue. Using the National Vital Statistics, we explored risk factors for HF and coronary artery disease (CAD) mortality.

Methods and Results: Altogether, 7,556 Japanese individuals aged ≥30 years in 1990 were followed over 25 years; of these, 139 and 154 died from HF and CAD, respectively. In multivariable Cox proportional hazard analysis, common risk factors for CAD and HF mortality were hypertension (hazard ratio [HR] 1.48 [95% confidence interval {CI} 1.00–2.20] and 2.31 [95% CI 1.48–3.61], respectively), diabetes (HR 2.52 [95% CI 1.63–3.90] and 2.07 [95% CI 1.23–3.50], respectively), and current smoking (HR 2.05 [95% CI 1.27–3.31) and 1.86 [95% CI 1.10–3.15], respectively). Specific risk factors for CAD were male sex, chronic kidney disease, history of cardiovascular disease, and both abnormal T and Q waves, with HRs (95% CIs) of 1.75 (1.05–2.92), 1.78 (1.19–2.66), 2.50 (1.62–3.88), and 11.4 (3.64–36.0), respectively. Specific factors for HF were current drinking (HR 0.43; 95% CI 0.24–0.78) and non-high-density lipoprotein cholesterol (non-HDL-C; HR 0.81; 95% CI 0.67–0.98). There was an inverse association between non-HDL-C and HF in those aged ≥65 years (HR 0.71; 95% CI 0.56–0.90), but not in those aged <65 years.

Conclusions: We identified common risk factors for HF and CAD deaths; a history of cardiovascular disease was a specific risk for CAD.

In Western countries, the incidence of and mortality from coronary artery disease (CAD) and stroke have decreased, whereas those for heart failure (HF), a terminal state of some types of heart disease, such as CAD, have increased.^1,2 Improvements in the treatment of patients with acute heart disease and HF, as well as population aging, are driving these trends.^3,4 The most current knowledge of risk factors for HF is based on North American and European population-based studies.^4–6 A previous study reported that HF mortality differed by race;⁷ thus, population-specific findings are needed to effectively manage the risk factors for HF according to population characteristics.

Meanwhile, a few studies in Asian countries, including Japan, have also examined risk factors for HF.^8–10 The burden of HF has also been increasing in the Asia-Pacific region, where most traditional cardiovascular risk factors, such as elevated blood pressure, obesity, and smoking, increase the risk of death due to HF in Asian and non-Asian populations alike.¹¹ Evidence also shows that HF due to ischemic etiologies is less common in Japan than in Western populations.¹² Regarding CAD, more evidence in the Japanese population is needed, because the incidence of CAD is lower in Japan than in Western populations.¹³ Furthermore, CAD mortality is lower in East Asian than other Asian populations.¹⁴

Accordingly, in the present study we explored the risk factors for HF and CAD mortality based on National Vital Statistics with a 25-year follow-up of the general Japanese population.

Methods

Study Participants

NIPPON DATA 90 (National Integrated Project for Prospective Observation of Noncommunicable Disease and Its Trends in the Aged 1990) is a cohort study based on the National Survey on Cardiovascular Diseases, baseline surveys of which were conducted in 1990. The details of the study have been reported previously.^15–18 Altogether, 8,383 people (3,504 men, 4,879 women) aged ≥30 years were randomly selected from 300 residential areas in 47 prefectures and were followed until November 15, 2015. Of these people, 827 were excluded because of loss to follow-up (n=182) and missing data on certain parameters used in the statistical analysis in the present study, such as cholesterol (n=645). The remaining 7,556 people were included in the present study.

Because of the voluntary participation of community dwellers to the National Survey on Cardiovascular Diseases, the need for informed consent for this research survey was waived. Permission to use the National Vital Statistics was acquired from the Management and Coordination Agency of the government of Japan, and both data matched and the anonymous nature of the data for the analysis. Approval for the study was obtained from the institutional review boards of Shiga University of Medical Science (No. 12-18, 2000; No. 17-21-1, 2010) and Keio University (2018108). Participants are provided with the opportunity to refuse to participate in the study on the Shiga Medical University website, where it is also disclosed that Keio University is a collaborating research organization.

Baseline Examination

The baseline survey for this study was conducted as the National Survey of Circulatory Disorders Cardiovascular Diseases according to the manual of the Ministry of Health and Welfare (currently, Ministry of Health, Labour and Welfare).¹⁹ To obtain information on medical history, questionnaires were designed to ask about the time of onset, symptoms, hospitalization, current hospital attendance, activities of daily living, and whether the participant needed care, for stroke. Similarly, for CAD, the questionnaire asked about symptoms at the time of the attack and current hospital attendance for myocardial infarction, whereas for angina pectoris asked about symptoms, response after using sublingual nitroglycerine tablets, and current hospital attendance. The questionnaires were also checked through face-to-face interviews by doctors or public health workers in the field as part of health checkups. In addition, public health workers collected lifestyle information, including smoking and alcohol drinking status. The final decision was then made by an expert committee in the Ministry of Health and Welfare, together with the descriptive information in the free-text section. Furthermore, baseline data information was checked under the auspices of the Japanese Society for Cerebrovascular Disease Control (currently, the Japanese Society for Cardiovascular Disease Prevention) and the Japan Cardiovascular Disease Research Foundation.

Height and weight were measured and body mass index (BMI) was calculated by dividing weight (kg) by height squared (m²). Using a standard mercury sphygmomanometer, trained observers measured participants’ systolic and diastolic blood pressure. Non-fasting blood samples were collected and assayed at SRL Corporation (Tokyo, Japan). Serum total cholesterol (TC) was measured by an enzymatic method, whereas high-density lipoprotein cholesterol (HDL-C) was measured by a precipitation method using heparin. Non-HDL-C was obtained by subtracting the HDL-C value from the TC value. Given that most of the study participants were in a non-fasting status during the baseline survey, we used non-HDL-C as a candidate risk factor for CAD, instead for low-density lipoprotein cholesterol (LDL-C). HbA1c was measured by high-performance liquid chromatography. The Japan Diabetes Society (JDS) HbA1c values were converted to National Glycohemoglobin Standardization Program (NGSP) values using the following formula:¹⁵

HbA1c (NGSP) = HbA1c (JDS) × 1.02 + 0.25

In the present study, HbA1c (NGSP) values were used in all analyses. Considering that serum creatinine (SCr) was measured using the alkaline picric acid method (Jaffe method), serum SCr data were converted to SCr values based on the enzymatic method using the following formula:^16,17

SCr (enzymatic method) = SCr (Jaffe method) − 0.2 mg/dL

Furthermore, the estimated glomerular filtration rate (eGFR; mL/min/1.73 m²) was calculated using the following formula for the Japanese population proposed by the Japanese Working Group on Chronic Kidney Disease:¹⁸

eGFR = 194 × SCr^−1.094 (enzyme method) × age^−0.287 × 0.739 (for women)

Blood hemoglobin (Hb) was measured using the oxyhemoglobin method.

Electrocardiogram (ECG) findings were independently evaluated by 2 trained investigators at each of the 12 centers according to the previously reported Minnesota Code (mc),²⁰ and the main ECG findings included Q wave abnormality (mc 1-1 to 1-2), high R wave (mc 3-1 to 3-4), ST depression (mc 4-1 to 4-4), T abnormality (mc 5-1 to 5-4), and atrial fibrillation (mc 8-3).

The definitions use in this study were as follows. Hypertension was defined as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, or a history of the use of antihypertensive drugs. Diabetes was defined as non-fasting blood glucose ≥200 mg/dL, HbA1c (NGSP) ≥6.5%, or a history of diabetes treatment. Chronic kidney disease (CKD) was defined as eGFR <60 mL/min/1.73 m² or a positive result within 1+ to 4+ on a urine protein paper qualitative test.^17,21 Anemia was defined as Hb <13 g/dL in men and <12 g/dL in women.²² Obesity was defined as BMI ≥25 kg/m²; normal weight was defined as BMI ≥18.5 kg/m² and BMI <25 kg/m²; and underweight was defined as BMI <18.5 kg/m². A history of cardiovascular disease (CVD) was defined as a history of angina pectoris, myocardial infarction, or stroke.

Follow-up Survey

We identified the causes of death at the follow-up survey using information from the National Vital Statistics of Japan. All death certificates issued by physicians were forwarded centrally to the Ministry of Health and Welfare via the public health center in each area of residence. The primary cause of death was coded using International Classification of Diseases, Ninth Revision (ICD-9; until the end of 1994) and International Classification of Diseases, Tenth Revision (ICD-10; from the beginning of 1994) codes as follows: CAD, ICD-9 codes 410–414 and ICD-10 codes I20–I25; and HF, ICD-9 code 428 and ICD-10 code I50.

Statistical Analysis

Mean±SD values are reported for continuous variables, whereas categorical variables are reported as numbers and proportions. Hazard ratios (HRs) and 95% confidence intervals (CIs) of the candidate risk factors for HF and CAD mortality were calculated using Cox proportional hazards models. Age, sex, BMI (obesity, normal, and underweight), hypertension, diabetes, smoking status (never, former, and current smoker), alcohol drinking status (never, former, and current drinker), CKD, non-HDL-C, HDL-C, treatment for dyslipidemia, anemia, history of CVD, and ECG findings were examined as candidate risk factors. Taking into consideration competing risks for HF and CAD deaths, HRs and 95% CIs for the aforementioned candidate risk factors were calculated using Fine-Gray models. In addition, these analyses were conducted and stratified by age (≥65 and <65 years). Furthermore, a sensitivity analysis using imputation was conducted using attribute mean assignment. Two-sided P<0.05 was considered significant. Statistical analyses were performed using JMP15 and 17 (SAS Institute Inc., Cary, NC, USA).

Results

Baseline data are presented in Table 1. The mean age of participants was 52.8±0.2 years, and the prevalence of obesity was 23.8%. The prevalence rates of hypertension and diabetes were 42.8% and 5.8%, respectively.

Table 1.

Baseline Characteristics of the Trial Participants Enrolled in 1990

	All (n=7,556)	Men (n=3,167)	Women (n=4,389)
Age (years)	52.8±0.2	53.3±0.2	52.5±0.2
Male sex	3,167 (41.9)	–	–
BMI (kg/m2)
<18.5	485 (6.4)	193 (6.1)	292 (6.7)
18.5≤, <25.0	5,274 (69.8)	2,221 (70.1)	3,053 (69.5)
≥25.0	1,797 (23.8)	753 (23.8)	1,044 (23.8)
HypertensionA	3,237 (42.8)	1,469 (46.4)	1,768 (40.3)
DiabetesB	175 (5.8)	200 (6.3)	189 (4.3)
Smoking status
Never smoked	4,527 (60.0)	649 (20.5)	3,878 (88.4)
Former smokerC	880 (11.6)	770 (24.3)	110 (2.5)
Current smoker	2,149 (28.4)	1,748 (55.2)	401 (9.1)
Alcohol drinking status
Never drinker	5,167 (68.4)	1,108 (35.0)	4,059 (92.5)
Former drinker	259 (3.4)	215 (6.8)	44 (1.0)
Current drinker	2,130 (28.2)	1,844 (58.2)	286 (6.5)
eGFR (mL/min/1.73 m2)	99.7±0.5	94.1±0.4	103.8±0.5
Urine proteinD	414 (5.5)	208 (6.6)	206 (4.7)
Non-HDL-C (mg/dL)	149.3±0.5	148.3±0.7	150.2±0.6
HDL-C (mg/dL)	54.1±0.2	50.4±0.3	56.8±0.2
Treatment for dyslipidemia	246 (3.3)	76 (2.4)	170 (3.9)
AnemiaE	1,056 (14.0)	231 (7.3)	825 (18.8)
History of CVDF	334 (4.4)	166 (5.2)	168 (3.8)
ECG findingsG
T wave abnormality	449 (5.9)	153 (4.8)	296 (6.7)
Q wave abnormality	44 (0.6)	26 (0.8)	18 (0.4)
High R wave + ST depression or T + Q abnormality	90 (1.2)	42 (1.3)	48 (1.1)
Atrial fibrillation	139 (1.8)	65 (2.1)	74 (1.7)

Data are presented as the mean±SD or n (%). ^AHypertension was defined as systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg or a history of antihypertensive medication use. ^BDiabetes was defined as HbA1c >6.5% or non-fasting blood glucose >200 mg/dL or a history of treatment for diabetes. ^CFormer smoking was defined as past smoking or current smoking <20 cigarettes/day. ^DUrine protein above ≥1+. ^EAnemia was defined as hemoglobin <13 g/dL in men and <12 g/dL in women. ^FHistory of CVD was defined as a history of myocardial infarction, stroke or angina pectoris. ^GECG findings were categorized using MC as follows: T abnormality, mc 5-1 to 5-4; Q abnormality, mc 1-1 to 1-2; high R wave, mc 3-1 to 3-4; ST depression, mc 4-1 to 4-4; atrial fibrillation, mc 8-3. BMI, body mass index; CKD, chronic kidney disease; CVD, cardiovascular disease; ECG, electrocardiogram; eGFR, estimated glomerular filtration rate; HDL-C, high-density lipoprotein cholesterol; MC, Minnesota codes.

Altogether, the data of 157,897 person-years (63,120 men, 94,777 women) were recorded, with a mean follow-up duration of 20.9 years. During the follow-up period, there were 154 deaths (87 men, 67 women) due to CAD and 139 deaths (46 men, 93 women) due to HF. Moreover, there were 70 CAD deaths among patients aged <65 years and 84 among those aged ≥65 years; and 40 HF deaths among patients aged <65 years and 99 among those aged ≥65 years.

Table 2 presents multivariable-adjusted HRs and 95% CIs for CAD and HF mortality. Common risk factors for CAD and HF mortality were age (HR 1.10 [95% CI 1.08–1.11] and 1.15 [95% CI 1.13–1.17], respectively), hypertension (1.48 [95% CI 1.00–2.20] and 2.31 [95% CI 1.48–3.61], respectively), diabetes (2.52 [95% CI 1.63–3.90] and 2.07 [95% CI 1.23–3.50], respectively), and current smoking (2.05 [95% CI 1.27–3.31] and 1.86 [95% CI 1.10–3.15], respectively). Specific risk factors for CAD mortality were male sex (HR 1.75; 95% CI 1.05–2.92), CKD (HR 1.78; 95% CI 1.19–2.66), a history of CVD (HR 2.50; 95% CI 1.62–3.88), and T and Q wave abnormalities on ECG (HR 11.4; 95% CI 3.64–36.0). Specific factors for HF mortality were current drinking (HR 0.43; 95% CI 0.24–0.78) and non-HDL-C (HR 0.81; 95% CI 0.67–0.98), which was inversely associated with HF mortality.

Table 2.

Multivariable-Adjusted HRs and 95% CIs for CAD and HF Mortality: Model 1

	CAD death (all; n=154)	P value	HF death (all; n=139)	P value
Age	1.10 (1.08–1.11)*	<0.001	1.15 (1.13–1.17)*	<0.001
Sex	1.75 (1.05–2.92)*	0.03	0.91 (0.52–1.58)	0.73
18.5≤BMI<25.0 (kg/m2)	Ref.		Ref.
BMI <18.5	1.31 (0.70–2.46)	0.39	1.56 (0.83–2.93)	0.17
BMI ≥25.0	0.87 (0.59–1.27)	0.47	1.26 (0.86–1.85)	0.24
Hypertension	1.48 (1.00–2.20)*	0.05	2.31 (1.48–3.61)*	<0.001
Diabetes	2.52 (1.63–3.90)*	<0.001	2.07 (1.23–3.50)*	0.007
Never smoked	Ref.		Ref.
Former smoker	1.18 (0.66–2.09)	0.58	1.40 (0.73–2.68)	0.31
Current smoker	2.05 (1.27–3.31)*	0.003	1.86 (1.10–3.15)*	0.02
Never drinker	Ref.		Ref.
Former drinker	1.09 (0.55–2.14)	0.81	0.69 (0.27–1.79)	0.45
Current drinker	0.89 (0.57–1.39)	0.60	0.43 (0.24–0.78)*	0.006
CKD	1.78 (1.19–2.66)*	0.005	1.16 (0.74–1.83)	0.51
Non-HDL-C	1.13 (0.97–1.32)	0.13	0.81 (0.67–0.98)*	0.03
HDL-C	1.01 (0.84–1.21)	0.94	0.94 (0.78–1.14)	0.52
Treatment for dyslipidemia	1.73 (1.00–2.98)	0.05	0.95 (0.46–1.97)	0.89
Anemia	1.08 (0.68–1.69)	0.75	1.05 (0.68–1.63)	0.83
History of CVD	2.50 (1.62–3.88)*	<0.001	0.70 (0.36–1.35)	0.29
T + Q wave abnormality	11.4 (3.64–36.0)*	<0.001	–	1.00
High R wave + ST depression or T + Q abnormality	0.82 (0.31–2.17)	0.69	1.89 (0.82–4.37)	0.14
Atrial fibrillation	1.83 (0.80–4.16)	0.15	1.31 (0.53–3.24)	0.55

*P≤0.05. Unless indicated otherwise, data show hazard ratios with 95% CIs in parentheses. Variables were standardized with mean=0 and standard deviation=1. Age is shown as unit risk with continuous variables, and non-HDL-C and HDL-C are shown as standardized risk. Sex (male vs. female), BMI (<18.5 or ≥25 vs. 18.5–25 kg/m²), smoking status (past smoking, current smoking <20 cigarettes/day vs. never smoking), alcohol drinking status (past drinking, current drinking vs. never drinking), and other nominal valuables (yes=1 vs. no=0). CAD, coronary artery disease; CI, confidence interval; CKD, chronic kidney disease (defined as eGFR <60 mL/min/1.73 m² or urine protein ≥1+); HF, heart failure; HR, hazard ratio. Other abbreviations as in Table 1.

Table 3 summarizes results stratified according to age (i.e., <65 and ≥65 years). In this stratified analysis, for CAD mortality, common risk factors in younger (<65 years) and older (≥65 years) people were age (HR 1.10 [95% CI 1.06–1.13] and 1.07 [95% CI 1.03–1.12], respectively), diabetes (HR 2.34 [95% CI 1.21–4.53] and 2.50 [95% CI 1.37–4.57], respectively), and a history of CVD (HR 3.08 [95% CI 1.49–6.35] and 2.26 [95% CI 1.29–3.94], respectively). Specific risk factors for CAD mortality among younger people were hypertension (HR 1.80; 95% CI 1.04–3.13), current smoking (HR 3.02; 95% CI 1.52–6.05), and CKD (HR 2.34; 95% CI 1.24–4.41). Specific risk factors for CAD mortality among older people were sex (HR 2.01; 95% CI 1.00–4.06), T and Q wave abnormality (HR 13.7; 95% CI 3.97–47.1), and atrial fibrillation (HR 2.99; 95% CI 1.28–6.96). In the stratified analysis for HF mortality, common risk factors in younger and older people were age (HR 1.12 [95% CI 1.07–1.18] and 1.18 [95% CI 1.14–1.23], respectively), hypertension (HR 3.22 [95% CI 1.46–7.10] and 1.96 [95% CI 1.15–3.33], respectively), and current drinking (HR 0.36 [95% CI 0.14–0.93] and 0.44 [95% CI 0.20–0.99], respectively). Specific risk factors for HF mortality among younger people were obesity (HR 2.00; 95% CI 1.02–3.92) and high R wave + ST depression or T + Q wave abnormality (HR 5.41; 95% CI 1.51–19.4). The specific factor for HF mortality among older people was non-HDL-C (HR 0.71; 95% CI 0.56–0.90).

Table 3.

Multivariable-Adjusted HRs and 95% CIs for CAD and HF Mortality: Stratified Analysis by Age <65 and ≥65 Years

	CAD death				HF death
	Age <65 years (n=70)	P value	Age ≥65 years (n=84)	P value	Age <65 years (n=40)	P value	Age ≥65 years (n=99)	P value
Age	1.10 (1.06–1.13)*	<0.001	1.07 (1.03–1.12)*	0.001	1.12 (1.07–1.18)*	<0.001	1.18 (1.14–1.23)*	<0.001
Sex	1.44 (0.68–3.06)	0.35	2.01 (1.00–4.06)*	0.05	1.48 (0.57–3.87)	0.42	0.72 (0.36–1.44)	0.35
18.5≤BMI<25.0 (kg/m2)	Ref.		Ref.		Ref.		Ref.
BMI <18.5	1.67 (0.59–4.77)	0.34	1.13 (0.51–2.48)	0.77	0.87 (0.11–6.70)	0.90	1.73 (0.88–3.42)	0.11
BMI ≥25.0	0.90 (0.51–1.57)	0.71	0.82 (0.48–1.40)	0.46	2.00 (1.02–3.92)*	0.04	1.00 (0.61–1.62)	0.98
Hypertension	1.80 (1.04–3.13)*	0.04	1.09 (0.63–1.89)	0.75	3.22 (1.46–7.10)*	0.004	1.96 (1.15–3.33)*	0.01
Diabetes	2.34 (1.21–4.53)*	0.01	2.50 (1.37–4.57)*	0.003	2.22 (0.92–5.33)	0.08	1.95 (0.99–3.85)	0.05
Never smoked	Ref.		Ref.		Ref.		Ref.
Former smoker	1.68 (0.68–4.15)	0.26	0.93 (0.44–1.95)	0.84	0.86 (0.21–3.43)	0.83	1.66 (0.78–3.52)	0.19
Current smoker	3.02 (1.52–6.05)*	0.002	1.40 (0.64–3.23)	0.34	2.32 (0.98–5.51)	0.06	1.59 (0.80–3.14)	0.18
Never drinker	Ref.		Ref.		Ref.		Ref.
Former drinker	0.62 (0.17–2.22)	0.46	1.43 (0.64–3.23)	0.39	–	1.00	1.00 (0.37–2.74)	1.00
Current drinker	0.80 (0.42–1.53)	0.50	0.87 (0.46–1.68)	0.69	0.36 (0.14–0.93)*	0.03	0.44 (0.20–0.99)*	0.05
CKD	2.34 (1.24–4.41)*	0.009	1.55 (0.91–2.65)	0.11	1.42 (0.56–3.61)	0.46	1.06 (0.63–1.78)	0.84
Non-HDL-C	1.17 (0.91–1.49)	0.22	1.11 (0.90–1.37)	0.33	1.09 (0.78–1.54)	0.61	0.71 (0.56–0.90)*	0.005
HDL-C	1.08 (0.83–1.40)	0.57	0.96 (0.75–1.24)	0.75	0.96 (0.66–1.40)	0.83	0.95 (0.75–1.19)	0.64
Treatment for dyslipidemia	1.93 (0.86–4.32)	0.11	1.44 (0.67–3.06)	0.35	0.31 (0.04–2.36)	0.26	1.23 (0.56–2.70)	0.60
Anemia	0.61 (0.19–2.00)	0.42	1.25 (0.75–2.08)	0.40	2.14 (0.79–5.80)	0.13	0.89 (0.55–1.46)	0.65
History of CVD	3.08 (1.49–6.35)*	0.002	2.26 (1.29–3.94)*	0.004	0.86 (0.20–3.75)	0.84	0.65 (0.31–1.36)	0.26
T + Q wave abnormality	–	1.00	13.7 (3.97–47.1)*	<0.001	–	1.00	–	1.00
High R wave + ST depression or T + Q abnormality	0.99 (0.22–4.41)	0.99	0.73 (0.20–2.65)	0.64	5.41 (1.51–19.4)*	0.009	1.09 (0.34–3.54)	0.84
Atrial fibrillation	–	1.00	2.99 (1.28–6.96)*	0.01	–	1.00	1.86 (0.74–4.66)	0.18

*P≤0.05. Abbreviations as in Tables 1,2.

Table 4 shows the result of the competing risk analysis. The common risk factor for CAD and HF mortality was age (HR 1.09 [95% CI 1.07–1.10] and 1.14 [95% CI 1.11–1.16], respectively). The specific risk factors for CAD mortality were male sex (HR 1.69; 95% CI 1.03–2.75), diabetes (HR 2.43; 95% CI 1.56–3.79), current smoking (HR 1.99; 95% CI 1.27–3.11), CKD (HR 1.64; 95% CI 1.06–2.54), history of CVD (HR 2.63; 95% CI 1.70–4.05), and T and Q wave abnormalities on ECG (HR 12.3; 95% CI 4.53–33.4). The specific risk factors for HF mortality were hypertension (HR 2.32; 95% CI 1.46–3.69) and current drinking (HR 0.47; 95% CI 0.26–0.86).

Table 4.

Competing Risks and Multivariate-Adjusted HRs and 95% CIs for CAD and HF Mortality: Model 2

	CAD death (all; n=154)	P value	HF death (all; n=139)	P value
Age	1.09 (1.07–1.10)*	<0.001	1.14 (1.11–1.16)*	<0.001
Sex	1.69 (1.03–2.75)*	0.04	0.82 (0.46–1.44)	0.48
18.5≤BMI<25.0 (kg/m2)	Ref.		Ref.
BMI <18.5	1.14 (0.58–2.27)	0.70	1.53 (0.80–2.95)	0.20
BMI ≥25.0	0.89 (0.62–1.29)	0.55	1.23 (0.83–1.83)	0.30
Hypertension	1.46 (0.96–2.22)	0.08	2.32 (1.46–3.69)*	<0.001
Diabetes	2.43 (1.56–3.79)*	<0.001	1.67 (0.93–2.98)	0.08
Never smoked	Ref.		Ref.
Former smoker	1.10 (0.64–1.91)	0.73	1.43 (0.74–2.78)	0.29
Current smoker	1.99 (1.27–3.11)*	0.003	1.71 (0.96–3.02)	0.07
Never drinking	Ref.		Ref.
Former drinking	1.11 (0.56–2.21)	0.76	0.68 (0.26–1.76)	0.43
Current drinking	0.95 (0.60–1.53)	0.84	0.47 (0.26–0.86)*	0.01
CKD	1.64 (1.06–2.54)*	0.03	1.13 (0.71–1.81)	0.62
Non-HDL-C	1.15 (0.98–1.35)	0.08	0.83 (0.68–1.00)	0.06
HDL-C	1.02 (0.84–1.24)	0.86	0.93 (0.76–1.12)	0.43
Treatment for dyslipidemia	1.63 (0.92–1.63)	0.09	0.97 (0.47–1.98)	0.93
Anemia	1.11 (0.71–1.11)	0.66	1.00 (0.62–1.60)	0.99
History of CVD	2.63 (1.70–4.05)*	<0.001	0.63 (0.32–1.22)	0.17
T + Q wave abnormality	12.3 (4.53–33.4)*	<0.001	–	1.00
High R wave + ST depression or T + Q abnormality	0.90 (0.32–2.52)	0.83	1.70 (0.66–4.36)	0.27
Atrial fibrillation	1.59 (0.67–3.78)	0.29	1.15 (0.46–2.87)	0.77

*P≤0.05. Abbreviations as in Tables 1,2.

Table 5 presents the results for competing risk analysis stratified according to age (i.e., age <65 and ≥65 years). In this analysis, for CAD mortality, the common risk factors in younger and older people were age (HR 1.10 [1.06–1.13] and 1.06 [1.02–1.10], respectively), diabetes (HR 2.32 [1.12–4.81] and 2.36 [1.32–4.22], respectively), and a history of CVD (HR 3.14 [1.53–6.43] and 2.31 [1.35–3.95], respectively). The specific risk factors for CAD mortality among younger people were hypertension (HR 1.78; 95% CI 1.00–3.16), current smoking (HR 2.96; 95% CI 1.47–5.95), and CKD (HR 2.31; 95% CI 1.21–4.40). Specific factors for CAD mortality among older people were sex (HR 1.96; 95% CI 1.03–3.78), T and Q wave abnormality (HR 13.8; 95% CI 4.59–41.7), and atrial fibrillation (HR 2.75; 95% CI 1.18–6.41). In the stratified analysis, the common risk factors for HF mortality in younger and older people were age (HR 1.12 [95% CI 1.07–1.18] and 1.16 [95% CI 1.12–1.20], respectively) and hypertension (HR 3.18 [95% CI 1.40–7.23] and 1.97 [95% CI 1.15–3.38], respectively). The specific factors for HF mortality among younger people were obesity (HR 2.02; 95% CI 1.05–3.90) and high R wave with ST depression or T + Q wave abnormality (HR 5.62; 95% CI 1.26–25.0). The specific factor for HF mortality among older people was non-HDL-C (HR 0.72; 95% CI 0.56–0.91), which was inversely associated with HF mortality.

Table 5.

Competing Risks and Multivariate-Adjusted HRs and 95% CIs for CAD and HF Mortality: Stratified Analysis by Age <65 and ≥65 Years

	CAD death				HF death
	Age <65 years (n=70)	P value	Age ≥65 years (n=84)	P value	Age <65 years (n=40)	P value	Age ≥65 years (n=99)	P value
Age	1.10 (1.06–1.13)*	<0.001	1.06 (1.02–1.10)*	0.006	1.12 (1.07–1.18)*	<0.001	1.16 (1.12–1.20)*	<0.001
Sex	1.46 (0.66–3.19)	0.35	1.96 (1.03–3.78)*	0.05	1.45 (0.58–3.63)	0.43	0.66 (0.32–1.35)	0.25
18.5≤BMI<25.0 (kg/m2)	Ref.		Ref.		Ref.		Ref.
BMI <18.5	1.68 (0.57–4.89)	0.35	1.02 (0.45–2.33)	0.96	0.83 (0.10–7.03)	0.86	1.70 (0.84–3.42)	0.14
BMI ≥25.0	0.89 (0.51–1.56)	0.69	0.85 (0.51–1.42)	0.54	2.02 (1.05–3.90)*	0.04	0.96 (0.57–1.59)	0.86
Hypertension	1.78 (1.00–3.16)*	0.05	1.06 (0.60–1.87)	0.83	3.18 (1.40–7.23)*	0.006	1.97 (1.15–3.38)*	0.01
Diabetes	2.32 (1.12–4.81)*	0.02	2.36 (1.32–4.22)*	0.004	2.18 (0.89–5.38)	0.09	1.61 (0.79–3.27)	0.19
Never smoked	Ref.		Ref.		Ref.		Ref.
Former smoker	1.66 (0.67–4.13)	0.27	0.88 (0.46–1.67)	0.69	0.86 (0.21–3.42)	0.83	1.65 (0.75–3.63)	0.21
Current smoker	2.96 (1.47–5.95)*	0.002	1.35 (0.72–2.51)	0.35	2.29 (0.93–5.59)	0.07	1.46 (0.70–3.05)	0.32
Never drinker	Ref.		Ref.		Ref.		Ref.
Former drinker	0.63 (0.16–2.47)	0.51	1.46 (0.67–3.17)	0.34	–	1.00	0.94 (0.34–2.59)	0.91
Current drinker	0.82 (0.42–1.60)	0.56	0.92 (0.46–1.85)	0.81	0.37 (0.13–1.05)	0.06	0.47 (0.22–1.01)	0.05
CKD	2.31 (1.21–4.40)*	0.01	1.47 (0.81–2.68)	0.21	1.40 (0.48–4.04)	0.54	1.03 (0.61–1.75)	0.90
Non-HDL-C	1.17 (0.90–1.51)	0.24	1.13 (0.92–1.39)	0.24	1.10 (0.76–1.59)	0.63	0.72 (0.56–0.91)*	0.007
HDL-C	1.08 (0.81–1.43)	0.61	0.98 (0.74–1.29)	0.86	0.96 (0.68–1.37)	0.84	0.93 (0.74–1.18)	0.56
Treatment for dyslipidemia	1.96 (0.85–4.52)	0.11	1.33 (0.61–2.89)	0.47	0.30 (0.04–2.26)	0.24	1.24 (0.58–2.68)	0.58
Anemia	0.61 (0.19–2.00)	0.43	1.29 (0.79–2.10)	0.32	2.15 (0.74–6.23)	0.16	0.84 (0.50–1.41)	0.51
History of CVD	3.14 (1.53–6.43)*	0.002	2.31 (1.35–3.95)*	0.002	0.84 (0.18–3.95)	0.82	0.58 (0.28–1.22)	0.15
T + Q wave abnormality	–	1.00	13.8 (4.59–41.7)*	<0.001	–	1.00	–	1.00
High R wave + ST depression or T + Q abnormality	0.98 (0.17–5.57)	0.98	0.78 (0.20–3.05)	0.73	5.62 (1.26–25.0)*	0.02	1.02 (0.32–3.21)	0.97
Atrial fibrillation	–	1.00	2.75 (1.18–6.41)*	0.02	–	1.00	1.62 (0.63–4.11)	0.32

*P≤0.05. Abbreviations as in Tables 1,2.

A sensitivity analysis using imputation with a method of attribute mean assignment showed almost similar results (Table 6).

Table 6.

Sensitivity Analysis

	CAD death (all; n=171)	P value	HF death (all; n=154)	P value
Age	1.09 (1.07–1.11)*	<0.001	1.15 (1.12–1.17)*	<0.001
Sex	1.74 (1.06–2.87)*	0.03	0.87 (0.51–1.49)	0.62
18.5≤BMI<25.0 (kg/m2)	Ref.		Ref.
BMI <18.5	1.35 (0.72–2.53)	0.35	1.53 (0.82–2.88)	0.18
BMI ≥25.0	0.87 (0.59–1.27)	0.47	1.26 (0.86–1.85)	0.24
Hypertension	1.52 (1.03–2.25)*	0.04	2.47 (1.59–3.84)*	<0.001
Diabetes	2.44 (1.57–3.79)*	<0.001	2.07 (1.23–3.51)*	0.007
Never smoked	Ref.		Ref.
Former smoker	1.17 (0.66–2.07)	0.59	1.45 (0.77–2.76)	0.25
Current smoker	1.99 (1.24–3.20)*	0.004	1.85 (1.10–3.10)*	0.02
Never drinker	Ref.		Ref.
Former drinker	0.99 (0.50–1.96)	0.98	0.66 (0.25–1.71)	0.39
Current drinker	0.90 (0.58–1.41)	0.65	0.43 (0.24–0.78)*	0.006
CKD	1.72 (1.15–2.57)*	0.009	1.11 (0.71–1.75)	0.65
Non-HDL-C	1.14 (0.98–1.34)	0.09	0.82 (0.68–1.00)*	0.05
HDL-C	1.01 (0.84–1.21)	0.93	0.93 (0.76–1.13)	0.45
Treatment for dyslipidemia	1.64 (0.95–2.83)	0.08	0.91 (0.44–1.90)	0.81
Anemia	1.09 (0.69–1.72)	0.72	1.01 (0.64–1.57)	0.98
History of CVD	2.47 (1.59–3.84)*	<0.001	0.72 (0.37–1.38)	0.31
T + Q wave abnormality	10.8 (3.49–33.5)*	<0.001	–	1.00
High R wave + ST depression or T + Q abnormality	0.87 (0.33–2.29)	0.78	1.87 (0.81–4.33)	0.15
Atrial fibrillation	1.80 (0.79–4.11)	0.16	1.36 (0.55–3.35)	0.51

*P≤0.05. Abbreviations as in Tables 1,2.

Discussion

The present study showed that the risk factors for CAD mortality include older age, male sex, hypertension, diabetes, current smoking, CKD, a history of CVD, and abnormal ECG (Q and T wave abnormalities), with further analysis stratified by age indicating atrial fibrillation was also a risk factor in those aged ≥65 years. The risk factors for CAD mortality detected in the present study are consistent with previous studies in general European and US populations,^23,24 as well as in Japan.²⁵ Meanwhile, the risk factors for HF mortality were older age, diabetes, hypertension, and current smoking, which is consistent with the findings of previous studies involving general populations in Europe and the US.^2,4–6,26 In contrast, current drinking and non-HDL-C were factors inversely associated with HF mortality. In particular, in the present study, lower non-HDL-C was identified as a risk factor for HF mortality in individuals aged ≥65 years, but not in those aged <65 years.

A unique finding of the present study is the significant inverse association observed between non-HDL-C and HF mortality in older people, whereas higher non-HDL-C was positively correlated with CAD, as reported previously,^27,28 although it did not reach statistical significance in the multivariable model. Prior studies have shown that low total cholesterol is related to all-cause and stroke mortality²⁹ and, in older people, low cholesterol is associated with lower activities of daily living, especially when combined with a low serum albumin.³⁰ In older people, low cholesterol is associated with nutritional status, and it is possible that it may be confused with frailty or malnutrition when diagnosing deaths from HF.

In contrast, in CAD mortality, low-density lipoprotein cholesterol (LDL-C) is an important risk factor³¹ and, in older people, high LDL-C was also reported to be a risk factor for atherosclerotic disease and an important therapeutic target.³² Furthermore, higher non-HDL-C levels are generally considered to be a risk factor for CAD because they reflect strongly atherogenic particles, such as small-dense LDL-C. Many observational studies in Western populations have shown a clear association between non-HDL-C and fatal and/or non-fatal coronary artery disease.^3,4 Similar findings were reported in previous prospective cohort studies conducted in Japanese and Western populations.^9,11 In the present study, non-HDL-C was not found to be a risk factor after adjusting for other confounders. The reason non-HDL-C was not a significant risk factor for CAD mortality in the multivariate analysis is probably due to the fact that the present study was conducted after the launch of statins, even though dyslipidemia medication at baseline was adjusted for. Specifically, the effect of non-HDL-C at baseline as a risk factor for CAD mortality may have been reduced because the use of statins became more widespread during the follow-up period of this study. Given that non-HDL-C includes very low-density lipoprotein (VLDL), adjusting for diabetes, hypertension, and BMI related to elevated VLDL may possibly have made the relationship less visible.

In older people in the present study, although not a significant risk factor, low BMI tended to increase the risk of HF mortality. Similar to low non-HDL-C, which was associated with an increased risk of HF mortality in the present study, low BMI may also be associated with frailty and undernutrition. Hence, these findings suggest that improvements in frailty and nutritional status may be important along, with control of arterial stiffness. A possibility is the presence of cardiac cachexia. A previous study demonstrated the presence of cardiac cachexia in relatively underweight populations and the association of obesity with improved survival in patients with compensated HF.³³ Furthermore, obese patients with other chronic disabling conditions, including chronic obstructive pulmonary disease, cancer, renal failure, or cirrhosis, also showed better survival.³⁴ The Japanese population tends to have a low BMI as a population characteristic,³⁵ which may predispose them to compensatory HF from underweight and undernutrition, especially in older people. These background characteristics may have an effect on the association between lower non-HDL-C levels and HF mortality among older people in the present study.

Past studies have reported that hypertension is a risk factor for left ventricular hypertrophy and HF.²⁶ Diastolic HF has been observed to be more common in older people,³⁶ which could be explained by potential cardiac hypertrophy caused by elevated blood pressure. In the present study, hypertension was a risk factor for HF mortality, consistent with the findings of previous studies. In addition, an increased risk of CAD mortality due to hypertension was also seen in the younger age group. These findings suggest that blood pressure control from a young age is important.

A history of CVD was significantly associated with CAD mortality. In addition, in the present study, T and Q wave abnormalities were significantly higher among those who died because of CAD. In an additional analysis in this study (Supplementary Table), T wave abnormalities alone, excluding Q wave abnormalities, were also found to be a risk factor for CAD mortality, which is consistent with the findings of a previous study.³⁷ In contrast, a history of CVD, including CAD, was not associated with an increased risk of HF mortality, although a history of CAD is known to be a major risk factor for HF.³⁸ In comparing the present study with our previous study,⁸ a similar baseline survey that preceded the present study by 10 years (i.e., in 1980), we found that hypertension, diabetes, and current smoking were risk factors for both HF and CAD mortality in the past and throughout the present study. In addition, in both studies, current drinking was inversely associated with HF mortality. Although low BMI was a risk factor for HF mortality in the past study,⁸ it was not a significant risk factor in the present study. Finally, a history of CVD was a risk factor for CAD mortality, but not HF mortality, in the present study, which is inconsistent with the findings reported in our previous study.⁸

These differences are thought to be due to a social influence because, in 1993, the Japanese Ministry of Health, Labour and Welfare changed the rule to:

In cases of terminal stage causes from other disease, heart failure or respiratory failure should NOT be used as the diagnoses of death.

Compared with the 1980 cohort, the present cohort was more affected by government guidelines for this type of death diagnosis during a greater part of the follow-up period. After the rule change, “HF” as the cause of death decreased markedly. A history of CVD may have led to the use of CAD as the cause of death because it may be the only evidence for establishing a diagnosis (e.g., when cases of sudden death or unexplained deaths in geriatric hospitals are encountered).

The present study has several limitations. First, if the hypothetical association between each risk factor considered is unclear, the interpretation of each factor may be difficult, a phenomenon referred to as “table 2 fallacy”.³⁹ Such concerns are undeniably always present in multivariate models. However, many of the risk factors used in the present study, such as hypertension and smoking, have been used as independent risk factors in the risk prediction models in many guidelines.^40,41

Second, we only measured attributes at baseline; thus, information on medication and lifestyle changes during follow-up was not available. Third, the causes of death obtained from the demographic study may have been misclassified. Fourth, these findings may not be consistent with those reported in previous studies. For example, previous studies, including one by our group,^8,42 found that low BMI was associated with an increased risk of death from HF. However, the present study showed that low BMI was associated with an increased HR for HF mortality, although the increased risk was not statistically significant. This could be attributed to differences in age range, mean BMI, or the sample size of each cohort. Finally, some potential confounders, such as lower physical activity,⁴³ cellular growth factors,⁴⁴ and fat-soluble vitamins,⁴⁵ were not considered in this study.

In conclusion, our study data showed that the major risk factors of hypertension, diabetes, and smoking were common for HF and CAD mortality. A history of CVD and ECG abnormalities were specific risk factors for CAD mortality. This may suggest the underlying cause of death due to CAD, but not HF, in the National Vital Statistics is influenced by past history or clinical findings. Our findings warrant further research in the future.

Acknowledgments

The authors express their deepest gratitude to A.H. and T.O. for the tremendous guidance as advisors during the investigation. The authors also thank the members of the NIPPON DATA Research Team for their substantial advice as subreviewers during the preparation of this paper.

Sources of Funding

This study was supported by a Grant-in-Aid from the Ministry of Health, Labour and Welfare (Tokyo, Japan) under the auspices of the Japanese Association for Cerebro-Cardiovascular Disease Control (Tokyo, Japan), a Research Grant for Cardiovascular Diseases (7A-2) from the Ministry of Health, Labour and Welfare, Grants-in-Aid from Japan Society for the Promotion of Science (Tokyo, Japan), Japan Agency for Medical Research and Development grants (Tokyo, Japan), and Health and Labour Sciences Research Grants, Japan (Comprehensive Research on Aging and Health H11-Chouju-046, H14-Chouju-003, H17-Chouju-012, H19-Chouju-Ippan-014 and Comprehensive Research on Life-Style Related Diseases including cardiovascular Diseases and Diabetes Mellitus H22-Junkankitou-Seishuu-Sitei-017, Junkankitou-Seishuu-Sitei-24FA2002, H30Junkankitou-Sitei-002, Kiban-Kenkyuu (A) 21H04854, 21FA2002, 22FA1006, AMED-JP23rea522009, JP23rea522010).

Disclosures

A.F. is a member of Circulation Journal’s Editorial Team. All other affiliations, financial involvement, or financial conflicts with the topics discussed in this paper are fully disclosed in the Acknowledgments section. The remaining authors have no conflicts of interest to disclose.

IRB Information

Institutional Review Board (IRB) approval was obtained from the IRBs of Shiga University of Medical Science (No. 12-18, 2000; No. 17-21-1, 2010) and Keio University School of Medicine (2018108). This study was performed in accordance with the Declaration of Helsinki and the ethical standards of the responsible committee on human experimentation.

Data Availability

The data used in this study were provided by the Ministry of Health, Labour and Welfare of Japan, and the authors do not have the right to share them. In order to obtain access to the data, contact the Health Service Division in the Health Service Bureau of the Ministry of Health, Labour and Welfare ([e-mail]).

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-23-0847

References

• 1.   Schocken DD, Benjamin EJ, Fonarow GC, Krumholz HM, Levy D, Mensah GA, et al. Prevention of heart failure: A scientific statement from the American Heart Association Councils on Epidemiology and Prevention, Clinical Cardiology, Cardiovascular Nursing, and High Blood Pressure Research; Quality of Care and Outcomes Research Interdisciplinary Working Group; and Functional Genomics and Translational Biology Interdisciplinary Working Group. Circulation 2008; 117: 2544–2565.
• 2.   Kannel WB, D’Agostino RB, Silbershatz H, Belanger AJ, Wilson PW, Levy D. Profile for estimating risk of heart failure. Arch Intern Med 1999; 159: 1197–1204.
• 3.   Kenchaiah S, Narula J, Vasan RS. Risk factors for heart failure. Med Clin North Am 2004; 88: 1145–1172.
• 4.   Levy D, Kenchaiah S, Larson MG, Benjamin EJ, Kupka MJ, Ho KK, et al. Long-term trends in the incidence of and survival with heart failure. N Engl J Med 2002; 347: 1397–1402.
• 5.   Curtis JP, Selter JG, Wang Y, Rathore SS, Jovin IS, Jadbabaie F, et al. The obesity paradox: Body mass index and outcomes in patients with heart failure. Arch Intern Med 2005; 165: 55–61.
• 6.   Dries DL, Exner DV, Domanski MJ, Greenberg B, Stevenson LW. The prognostic implications of renal insufficiency in asymptomatic and symptomatic patients with left ventricular systolic dysfunction. J Am Coll Cardiol 2000; 35: 681–689.
• 7.   Nayak A, Hicks AJ, Morris AA. Understanding the complexity of heart failure risk and treatment in Black patients. Circ Heart Fail 2020; 13: e007264.
• 8.   Nakamura Y, Tanvir CT, Nahid R, Miura K, Kita Y, Takashima N, et al. Risk factors for heart failure and coronary heart disease mortality over 24-year follow-up period in Japan: NIPPON DATA80. Global Heart 2010; 5: 97–103.
• 9.   Horibe H, Kasagi F, Kagaya M, Matsutani Y, Okayama A, Ueshima H, et al. A nineteen-year cohort study on the relationship of electrocardiographic findings to all cause mortality among subjects in the national survey on circulatory disorders, NIPPON DATA80. J Epidemiol 2005; 15: 125–134.
• 10.   Roh JH, Park JH, Lee H, Yoon YH, Kim M, Kim YG, et al. Higher fatty liver index is associated with increased risk of new onset heart failure in healthy adults: A nationwide population-based study in Korea. BMC Cardiovasc Disord 2020; 20: 204.
• 11.   Huxley RR, Barzi F, Woo J, Giles G, Lam TH, Rahimi K, et al. A comparison of risk factors for mortality from heart failure in Asian and non-Asian populations: An overview of individual participant data from 32 prospective cohorts from the Asia-Pacific Region. BMC Cardiovasc Disord 2014; 14: 61.
• 12.   Konishi M, Ishida J, Springer J, von Haehling S, Akashi YJ, Shimokawa H, et al. Heart failure epidemiology and novel treatments in Japan: Facts and numbers. ESC Heart Fail 2016; 3: 145–151.
• 13.   Ueshima H, Sekikawa A, Miura K, Turin TC, Takashima N, Kita Y, et al. Cardiovascular disease and risk factors in Asia: A selected review. Circulation 2008; 118: 2702–2709.
• 14.   Zhao D. Epidemiological features of cardiovascular disease in Asia. JACC Asia 2021; 1: 1–13.
• 15.   Kashiwagi A, Kasuga M, Araki E, Oka Y, Hanafusa T, Ito H, et al. International clinical harmonization of glycated hemoglobin in Japan: From Japan Diabetes Society to National Glycohemoglobin Standardization Program values. J Diabetes Investig 2012; 3: 39–40.
• 16.   Coresh J, Astor BC, McQuillan G, Kusek J, Greene T, Van Lente F, et al. Calibration and random variation of the serum creatinine assay as critical elements of using equations to estimate glomerular filtration rate. Am J Kidney Dis 2002; 39: 920–929.
• 17.   Imai E, Horio M, Watanabe T, Iseki K, Yamagata K, Hara S, et al. Prevalence of chronic kidney disease in the Japanese general population. Clin Exp Nephrol 2009; 13: 621–630.
• 18.   Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009; 53: 982–992.
• 19.  Ministry of Health and Welfare, Japanese Society for Cerebrovascular Disease Control (currently, Japanese Society for Cardiovascular Disease Prevention). National survey of circulatory disorders 1990 [in Japanese]. Japan Cardiovascular Research Foundation, 1993.
• 20.   Blackburn H, Keys A, Simonson E, Rautaharju P, Punsar S. The electrocardiogram in population studies: A classification system. Circulation 1960; 21: 1160–1175.
• 21.   Matsushita K, van der Velde M, Astor BC, Woodward M, Levey AS, de Jong PE, et al. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: A collaborative meta-analysis. Lancet 2010; 375: 2073–2081.
• 22.   Kubo K, Okamura T, Sugiyama D, Hisamatsu T, Hirata A, Kadota A, et al. Effect of chronic kidney disease or anemia or both on cardiovascular mortality in a 25-year follow-up study of Japanese general population (from NIPPON DATA90). Am J Cardiol 2022; 184: 1–6.
• 23.   Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation 1998; 97: 1837–1847.
• 24.   D’Agostino RB Sr, Grundy S, Sullivan LM, Wilson P; CHD Risk Prediction Group. Validation of the Framingham coronary heart disease prediction scores: Results of a multiple ethnic groups investigation. JAMA 2001; 286: 180–187.
• 25.   Nishimura K, Okamura T, Watanabe M, Nakai M, Takegami M, Higashiyama A, et al. Predicting coronary heart disease using risk factor categories for a Japanese urban population, and comparison with the Framingham risk score: The Suita study. J Atheroscler Thromb 2014; 21: 784–798.
• 26.   Rismiati H, Arima H, Fujiyoshi A, Miura K, Takashima N, Ohkubo T, et al. Hypertensive heart failure in Asia. Pulse (Basel) 2021; 9: 47–56.
• 27.   Ito T, Arima H, Fujiyoshi A, Miura K, Takashima N, Ohkubo T, et al. Relationship between non-high-density lipoprotein cholesterol and the long-term mortality of cardiovascular diseases: NIPPON DATA 90. Int J Cardiol 2016; 220: 262–267.
• 28.   Sugiyama D, Turin TC, Yeasmin F, Rumana N, Watanabe M, Higashiyama A, et al. Hypercholesterolemia and lifetime risk of coronary heart disease in the general Japanese population: Results from the Suita Cohort Study. J Atheroscler Thromb 2020; 27: 60–70.
• 29.   Chen L, Li X, Lv Y, Tan X, Zhong VW, Rong S, et al. Physical frailty, adherence to ideal cardiovascular health and risk of cardiovascular disease: A prospective cohort study. Age Ageing 2023; 52: afac311.
• 30.   Okamura T, Hayakawa T, Hozawa A, Kadowaki T, Murakami Y, Kita Y, et al. Lower levels of serum albumin and total cholesterol associated with decline in activities of daily living and excess mortality in a 12-year cohort study of elderly Japanese. J Am Geriatr Soc 2008; 56: 529–535.
• 31.   Abdullah SM, Defina LF, Leonard D, Barlow CE, Radford NB, Willis BL, et al. Long-term association of low-density lipoprotein cholesterol with cardiovascular mortality in individuals at low 10-year risk of atherosclerotic cardiovascular disease. Circulation 2018; 138: 2315–2325.
• 32.   Ouchi Y, Sasaki J, Arai H, Yokote K, Harada K, Katayama Y, et al. Ezetimibe Lipid-Lowering Trial on Prevention of Atherosclerotic Cardiovascular Disease in 75 or Older (EWTOPIA 75): A randomized, controlled trial. Circulation 2019; 140: 992–1003.
• 33.   Kistorp C, Faber J, Galatius S, Gustafsson F, Frystyk J, Flyvbjerg A, et al. Plasma adiponectin, body mass index, and mortality in patients with chronic heart failure. Circulation 2005; 112: 1756–1762.
• 34.   Springer J, von Haehling S, Anker SD. The need for a standardized definition for cachexia in chronic illness. Nat Clin Pract Endocrinol Metab 2006; 2: 416–417.
• 35.   Finucane MM, Stevens GA, Cowan MJ, Danaei G, Lin JK, Paciorek CJ, et al. National, regional, and global trends in body-mass index since 1980: Systematic analysis of health examination surveys and epidemiological studies with 960 country-years and 9.1 million participants. Lancet 2011; 377: 557–567.
• 36.   McDonald K. Diastolic heart failure in the elderly: Underlying mechanisms and clinical relevance. Int J Cardiol 2008; 125: 197–202.
• 37.   Kurl S, Mäkikallio TH, Laukkanen JA. T-wave inversion and mortality risk. Ann Med 2015; 47: 69–73.
• 38.   Tromp J, Paniagua SMA, Lau ES, Allen NB, Blaha MJ, Gansevoort RT, et al. Age dependent associations of risk factors with heart failure: Pooled population based cohort study. BMJ 2021; 372: n461.
• 39.   Westreich D, Greenland S. The table 2 fallacy: Presenting and interpreting confounder and modifier coefficients. Am J Epidemiol 2013; 177: 292–298.
• 40.   Okamura T, Tsukamoto K, Arai H, Fujioka Y, Ishigaki Y, Koba S, et al. Japan Atherosclerosis Society (JAS) guidelines for prevention of atherosclerotic cardiovascular diseases 2022. J Atheroscler Thromb 2024; 31: 641–853.
• 41.   Grundy SM, Stone NJ, Bailey AL, Beam C, Birtcher KK, Blumenthal RS, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of blood cholesterol: Executive summary: A report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation 2019; 139: e1046–e1081.
• 42.   Cui R, Iso H, Tanabe N, Watanabe Y, Tamakoshi A; JACC Study Group. Association between weight change since 20 years of age with mortality from myocardial infarction and chronic heart failure in the Japan collaborative cohort (JACC) study. Circ J 2014; 78: 649–655.
• 43.   Kushima T, Yamagishi K, Kihara T, Tamakoshi A, Iso H. Physical activity and risk of mortality from heart failure among Japanese population. J Atheroscler Thromb 2022; 29: 1076–1084.
• 44.   Eshak ES, Maruyama K, Iso H, Tamakoshi A. The prospective association between plasma concentrations of cellular growth factors and risk of heart failure mortality in Japanese population. J Epidemiol 2019; 29: 104–109.
• 45.   Eshak ES, Iso H, Yamagishi K, Cui R, Tamakoshi A. Dietary intakes of fat soluble vitamins as predictors of mortality from heart failure in a large prospective cohort study. Nutrition 2018; 47: 50–55.