Background
Methods
Swedish cohort studies
Meta-analysis
Search strategy
Inclusion criteria
Data extraction and quality assessment
Statistical analysis
Results
Swedish cohort studies
Meta-analysis
No. of stroke cases | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|
Study | Cohort name, country | No. of subjects | Age, years | Men, % | Follow-up, years | IS | ICH | SAH | NOS score | Adjustments |
Donahue et al., 1986 [14] | Honolulu Heart Program, USA | 8006 | 45–69 | 100 | 12 | 190 | 44 | 32 | 8 | Age, smoking, BMI, hypertensive status, serum cholesterol, uric acid glucose, and hematocrit concentrations |
Stampfer et al., 1988 [15] | Nurses’ Health Study, USA | 87,526 | 34–59 | 0 | 3.8 | –a | – | 28 | 7 | Age, smoking, obesity, exercise, family history of myocardial infarction, menopausal status and hormone use, history of hypertension, high cholesterol levels, diabetes, intake of cholesterol, saturated fat, and polyunsaturated fat |
Iso et al., 1995 [16] | Rural Japanese Cohorts, Japan | 2890 | 40–69 | 100 | 10.5 | 104 | – | – | 8 | Age, smoking, hypertension, serum total cholesterol, and diabetes |
Kiyohara et al., 1995 [17] | The Hisayama Study, Japan | 1621 | ≥ 40 | 44 | 26 | 244 | 60 | – | 9 | Age, sex, BMI, hypertension, electrocardiographic abnormalities, heart rate, glucose intolerance, serum cholesterol, and smoking |
Leppälä et al., 1999 [18] | Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study, Finland | 26,556 | 50–69 | 100 | 6.1 | 733 | 95 | 83 | 7 | Age, randomized treatment assignment, education, smoking, BMI, physical activity, serum total cholesterol, history of heart disease, diabetes |
Sankai et al., 2000 [19] | Six Japanese Communities, Japan | 12,372 | 40–69 | 40 | 9.4 | – | – | 71 | 8 | Age, sex, smoking, BMI, blood pressure, serum total cholesterol, diabetes |
Suh et al., 2001 [20] | Korea Medical Insurance Corporation Study, Korea | 114,793 | 35–59 | 100 | 5.4 | – | 373 | 98 | 6 | Age, smoking, BMI, blood pressure, fasting blood glucose, total serum cholesterol |
Klatsky et al., 2001 [21] | Kaiser Permanente Medical Care Program Cohort, USA | 128,934 | < 40–70+ | 44 | 18 | 2014 | – | – | 6 | Age, sex, race, education, smoking, BMI |
Klatsky et al., 2002 [22] | Kaiser Permanente Medical Care Program Cohort, USA | 128,934 | < 40–70+ | 44 | 18 | – | 299 | 133 | 6 | Age, sex, race, education, smoking, BMI |
Djousse et al., 2002 [23] | Framingham Study, USA | 9171 | ≥ 50 | 42 | 8.8 | 441 | – | – | 7 | Age, smoking, BMI, diabetes |
Iso et al., 2004 [24] | Japan Public Health Centre-Based Prospective Study, Japan | 19,356 | 40–59 | 100 | 11.0 | 319 | 219 | 73 | 6 | Age, education, public health centers, smoking, BMI, sports at leisure time, history of diabetes, intakes of fruit, vegetable, and fish |
Iwashita et al., 2005 [25] | Kyushu Lipid Intervention Study, Japan | 4349 | 45–74 | 100 | 5 | 81 | – | – | 6 | Age, smoking, BMI, serum total cholesterol, high-density lipoprotein cholesterol, prior use of cholesterol-lowering drugs, history of angina, hypertension, and diabetes |
Mukamal et al, 2005 [26] | Cardiovascular Health Study, USA | 4410 | ≥ 65 | 36 | 9.2 | 434 | – | – | 7 | Age, sex, geographic region, parental history of myocardial infarction, smoking, BMI, physical activity, hypercholesterolemia, diabetes, aspirin use, intakes of energy, vitamin E, folate, saturated fat, trans fats, omega-3 fatty acids, potassium, magnesium, and dietary fiber |
Nielsen et al., 2005 [27] | Copenhagen City Heart Study, Denmark | 12,096 | 21–98 | 44 | 15 | 786 | – | – | 8 | Age, sex, education, smoking, BMI, physical activity, systolic blood pressure, history of myocardial infarction and diabetes, Forced Expiratory Volume 1 |
Elkind et al., 2006 [28] | Northern Manhattan Study, USA | 3176 | ≥ 40 | 37 | 5.9 | 172 | – | – | 8 | Age, sex, race/ethnicity, education, smoking, high-density lipoprotein cholesterol, history of hypertension, atrial fibrillation, and diabetes |
Bazzano et al., 2007 [29] | China National Hypertension Survey Epidemiology Follow-up Study, China | 64,338 | ≥ 40 | 100 | 7.7 | 1724 | – | – | 8 | Age, geographic region, urban or rural residence, education, smoking, BMI, physical activity, diabetes |
Sturgeon et al, 2007 [30] | Atherosclerosis Risk in Communities Study and Cardiovascular Health Study, USA | 21,680 | ≥ 45 | 45 and 42 | 12.2 | – | 135 | – | 5 | Age |
Chiuve et al., 2008 [31]b | Health Professionals Follow-up Study, USA | 43,685 | 40–75 | 100 | 18 | 600 | – | – | 6 | Age, calendar year, family history of myocardial infarction, aspirin use, vitamin E supplementation |
Lu et al., 2008 [32] | Swedish Women’s Lifestyle and Health Cohort Study, Sweden | 45,449 | 30–50 | 0 | 11 | 111 | 47 | – | 7 | Age, education, BMI, smoking, parity and age at first birth, oral contraceptive use |
Sandvei et al., 2009 [33] | Nord-Trøndelag Health Study, Norway | 61,371 | ≥ 20 | 49 | 22 | – | – | 132 | 4 | Age, sex |
Bos et al., 2010 [34] | European Investigation into Cancer and Nutrition-Netherlands Cohort, The Netherlands | 10,530 | 20–70 | 0 | 9.4 | 165 | – | – | 8 | Age, cohort, education, smoking, BMI, physical activity, menopausal status, hypercholesterolemia, diabetes, antihypertensive medication, intakes of energy, vitamin E, vitamin C, saturated fat, and fiber |
Rist et al., 2010 [35] | Physicians’ Health Study, USA | 21,860 | 40–84 | 100 | 21.6 | 1157 | – | – | 8 | Age, smoking, BMI, exercise, systolic blood pressure, treatment for hypertension, diabetes, and migraine |
Zhang et al., 2011 [36] | Six geographic areas of Finland | 36,686 | 25–74 | 47 | 13.7 | 1167 | – | – | 8 | Age, study year, sex, education, smoking, BMI, physical activity, systolic blood pressure, total cholesterol level, family history of stroke, diabetes, intake of fruit and vegetables |
Higashiyama et al., 2011 [37] | Suita Study, Japan | 2336 | 30–79 | 100 | 12.5 | 78 | – | – | 7 | Age, smoking, BMI, high-density lipoprotein cholesterol, triglycerides, history of hypercholesterolemia, hypertension, and diabetes |
Drogan et al., 2012 [38] | European Investigation into Cancer and Nutrition-Potsdam cohort, Germany | 2175 | 35–65 | 37 | 8.2 | 169 | – | – | 7 | Age, education, smoking, BMI, waist circumference, physical activity, hypertension, diabetes, plasma total cholesterol levels, and non-alcohol energy intake |
Jimenez et al., 2012 [39] | Nurses’ Health Study, USA | 83,578 | 30–55 | 0 | 20.3 | 1206 | – | – | 8 | Age, education, marital status, family history of heart disease, smoking, BMI, physical activity, postmenopausal status, hormone therapy use, aspirin use, high cholesterol, history of heart disease and diabetes, bilateral oophorectomy, multivitamin use, diet score |
Ikehara et al., 2013 [40] | Japan Public Health Centre-Based Prospective Study, Japan | 47,100 | 40–69 | 0 | 16.7 | 964 | 532 | 338 | 8 | Age, area, smoking, BMI, sports at leisure-time, history of hypertension and diabetes, flushing after alcohol drinking, mental stress, menopausal status |
Kadlecová et al., 2015 [41] | Swedish Twin Registry, Sweden | 11,644 | ≤ 60 | 30 | 29.1 | 1846 | – | – | 8 | Age, sex, smoking, BMI, exercise, history of coronary heart disease, hypertension, diabetes and depression, stress reactivity |
Jones et al., 2015 [42] | Atherosclerosis Risk in Communities study, USA | 12,433 | 45–64 | 45 | 22.6 | 773 | –c | – | 6 | Age, sex, center-race interaction, education, smoking; IS analyses were further adjusted for low-density lipoprotein cholesterol, history of coronary heart disease and diabetes, marital status, diet score |
Larsson et al., 2016 (current study) | Cohort of Swedish Men, Sweden | 39,941 | 45–79 | 100 | 11.9 | 2216 | 350 | 82 | 8 | Age, education, family history of myocardial infarction, smoking, BMI, exercise, walking/bicycling, history of hypertension, hypercholesterolemia, atrial fibrillation, and diabetes, aspirin use, diet score |
Larsson et al., 2016 (current study) | Swedish Mammography Cohort, Sweden | 33,646 | 49–83 | 0 | 11.9 | 1608 | 205 | 94 | 8 | Age, education, family history of myocardial infarction, smoking, BMI, exercise, walking/bicycling, history of hypertension, hypercholesterolemia, atrial fibrillation, and diabetes, aspirin use, diet score |
Overall analyses
Subgroup analyses
Alcohol consumption | ||||||
---|---|---|---|---|---|---|
Light-to-moderate (≤ 2 drinks/day) | High-to-heavy (> 2 drinks/day) | |||||
No.a | RR (95 % CI) |
I
2b
| No.a | RR (95 % CI) |
I
2b
| |
Ischemic stroke | ||||||
All studies | 24 | 0.91 (0.88–0.94) | 9.1 % | 23 | 1.09 (1.03–1.16) | 0 % |
Geographic area | ||||||
United States | 9 | 0.90 (0.85–0.94) | 7.7 % | 8 | 1.12 (1.01–1.24) | 0 % |
Europe | 9 | 0.93 (0.88–0.98) | 11.1 % | 8 | 1.14 (1.04–1.25) | 0 % |
Asia | 5 | 0.87 (0.78–0.98) | 10.6 % | 6 | 1.07 (0.93–1.23) | 30.6 % |
Sex | ||||||
Men | 12 | 0.94 (0.88–1.00) | 18.4 % | 10 | 1.11 (1.00–1.23) | 15.2 % |
Women | 7 | 0.88 (0.83–0.95) | 0 % | 6 | 1.15 (0.96–1.36) | 0 % |
Womenc | 7 | 0.88 (0.79–0.98) | 32.2 % | 7 | 0.96 (0.85–1.08) | 7.9 % |
Study quality | ||||||
NOS ≥ 7 | 18 | 0.90 (0.86–0.93) | 0 % | 17 | 1.09 (1.02–1.16) | 0 % |
NOS < 7 | 5 | 0.87 (0.81–0.94) | 13.6 % | 4 | 1.10 (0.97–1.25) | 9.3 % |
Intracerebral hemorrhage | ||||||
All studies | 11 | 0.95 (0.84–1.07) | 10.2 % | 9 | 1.45 (1.18–1.78) | 55.0 % |
Geographic area | ||||||
United States | 3 | 0.94 (0.75–1.17) | 33.4 % | 2 | 1.38 (0.86–2.21) | 33.6 % |
Europe | 4 | 0.92 (0.77–1.10) | 0 % | 3 | 1.39 (0.98–1.96) | 40.2 % |
Asia | 4 | 1.09 (0.79–1.49) | 34.1 % | 4 | 1.56 (1.13–2.17) | 69.7 % |
Sex | ||||||
Men | 5 | 0.98 (0.78–1.24) | 26.4 % | 5 | 1.35 (1.06–1.72) | 59.4 % |
Women | 3 | 0.95 (0.76–1.19) | 0 % | 2 | 2.23 (1.47–3.38) | 0 % |
Womenc | 3 | 0.81 (0.55–1.19) | 37.6 % | 3 | 1.54 (1.08–2.18) | 29.6 % |
Study quality | ||||||
NOS ≥ 7 | 7 | 0.99 (0.84–1.18) | 7.5 % | 6 | 1.68 (1.32–2.14) | 25.4 % |
NOS < 7 | 4 | 0.90 (0.76–1.07) | 17.0 % | 3 | 1.20 (0.92–1.56) | 51.9 % |
Subarachnoid hemorrhage | ||||||
All studies | 10 | 1.16 (0.98–1.37) | 0 % | 9 | 1.57 (1.18–2.09) | 39.0 % |
Geographic area | ||||||
United States | 3 | 1.46 (1.02–2.10) | 21.4 % | 2 | 2.04 (1.12–3.73) | 24.4 % |
Europe | 4 | 1.03 (0.81–1.30) | 0 % | 3 | 1.17 (0.58–2.37) | 58.9 % |
Asia | 3 | 1.26 (0.81–1.97) | 19.8 % | 4 | 1.57 (1.10–2.24) | 31.3 % |
Sex | ||||||
Men | 5 | 1.06 (0.69–1.60) | 29.9 % | 5 | 1.48 (0.96–2.28) | 51.0 % |
Women | 3 | 1.38 (1.04–1.85) | 0 % | 2 | 1.90 (1.16–3.13) | 0 % |
Womenc | 3 | 1.56 (1.01–2.38) | 19.5 % | 3 | 1.54 (1.08–2.20) | 0 % |
Study quality | ||||||
NOS ≥ 7 | 7 | 1.25 (0.97–1.62) | 13.4 % | 3 | 1.44 (1.01–2.06) | 16.5 % |
NOS < 7 | 4 | 1.10 (0.87–1.39) | 0 % | 6 | 1.57 (1.03–2.40) | 50.0 % |