Table 1.
Characteristics of the included cohort publications
| Author, year, country [reference] | Study name | Study period, follow-up | Number of participants, sex, age: number of cases/deaths | Exposure vs. comparator, subgroup | Outcome | RR (95% CI) | Adjustment |
|---|---|---|---|---|---|---|---|
| Snowden, 1984, USA [47] | Adventist Mortality Study (AMS) | 1960–1980, 21 years follow-up |
22362 M/W, age 30–84 years: 1243 IHD deaths |
Vegetarians vs. nonvegetarians, M Vegetarians vs. nonvegetarians, W |
IHD mort. IHD mort. |
0.67 (0.56–0.77)a 0.73 (0.63–0.83)a |
Age |
| Fraser, 1995, USA [46] | Adventist Health Study 1 (AHS-1) | 1974–1976 – 1982, 5.14 years follow-up |
27321 M/W, age > 24 years: 448 IHD cases |
Vegetarians vs. nonvegetarians, M Vegetarians vs. nonvegetarians, W |
IHD inc. IHD inc. |
0.56 (0.40–0.78)a 1.02 (0.74–1.41)a |
Smoking, BMI, exercise, diabetes mellitus, hypertension, nut consumption |
| Key, 1999, USA [20] | Adventist Mortality Study (AMS) | 1959–1960 – 1965, 5.6 years follow-up |
24538 M/W, age 16–89 years: 780 CVD deaths 598 IHD deaths 182 CBVD deaths |
Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians |
CVD mort.b IHD mort. CBVD mort. |
0.72 (0.62–0.83)b 0.74 (0.63–0.88) 0.65 (0.48–0.87) |
Age, sex, and smoking status |
| Key, 1999, USA [20] | Adventist Health Study 1 (AHS-1) | 1976–1980 – 1988, 11.1 years follow-up |
28952 M/W, age 16–89 years: 1238 CVD deaths 921 IHD deaths 317 CBVD deaths |
Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians |
CVD mort.b IHD mort. CBVD mort. |
0.70 (0.61–0.80)b 0.62 (0.53–0.73) 0.93 (0.73–1.19) |
Age, sex, and smoking status |
| Key, 1999, Germany [20] | Heidelberg Study (HBS) | 1978–1981 – 1989, 9.9 years follow-up |
1757 M/W, age 16–89 years: 31 CBVD deaths |
Vegetarians vs. nonvegetarians | CBVD mort. | 1.69 (0.69–4.15) | Age, sex, and smoking status |
| Key, 1999, USA, Germany, UK (pooled analysis) [20] |
Adventist Mortality Study (AMS) + Adventist Health Study 1 (AHS-1) + Heidelberg Study (HBS) + Oxford Vegetarian Study (OVS) |
11.7 years follow-up |
76172 M/W, age 16–89 years: 3173 CVD deaths 2264 IHD deaths 909 CBVD deaths |
Vegans vs. regular meat-eaters Vegans vs. regular meat-eaters Vegans vs. regular meat-eaters Vegetarians vs. regular meat-eaters Vegetarians vs. regular meat-eaters |
CVD mort.b,c IHD mort.c CBVD mort.c IHD mort.d− IHD mort.d+ |
0.73 (0.47–1.14)b,c 0.74 (0.46–1.21)c 0.70 (0.25–1.98)c 0.64 (0.53–0.77)d− 0.66 (0.55–0.79)d+ |
Age, sex, and smoking dAge, sex, smoking status, alcohol use, education, exercise ± BMI |
|
43048 M/W (detailed confounders available): 1047 IHD deaths | |||||||
| Appleby, 2002, UK [26] | Oxford Vegetarian Study (OVS) | 1980–1984 – 2000, 17.6 years follow-up |
11045 M/W, 16–89 years: 469 circulatory disease deaths 375 CVD deaths 250 IHD deaths 125 CBVD deaths |
Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians |
CircD mort. CircD mort.e CVD mort.f IHD mort. IHD mort.e CBVD mort. CBVD mort.e |
0.93 (0.77–1.12) 1.01 (0.82–1.24)e 0.93 (0.75–1.14)f 0.86 (0.67–1.12) 0.98 (0.73–1.30)e 1.08 (0.75–1.54) 1.10 (0.75–1.63)e |
Age, sex, smoking |
| Appleby, 2002, UK [26] | Health Food Shoppers Study (HFSS) | 1973–1979 – 1997, 18.7 years follow-up |
10736 M/W, age 16–89 years: 1117 circulatory disease deaths 892 CVD deaths 562 IHD deaths 330 CBVD deaths |
Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians |
CircD mort. CircD mort.e CVD mort.f IHD mort. IHD mort.e CBVD mort. CBVD mort.e |
0.95 (0.84–1.07) 0.97 (0.85–1.12)e 0.90 (0.78–1.03)f 0.85 (0.71–1.01) 0.89 (0.73–1.09)e 0.99 (0.79–1.24) 0.97 (0.75–1.24)e |
Age, sex, smoking |
| Chang-Claude, 2005, Germany [27] | Heidelberg Study (HBS) | 1978–1999, 21 years follow-up |
1724 M/W, age ≥ 10 years: 219 circulatory disease deaths 60 IHD deaths |
Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians |
CircD mort. IHD mort. |
0.83 (0.62–1.12) 0.70 (0.41–1.18) |
Age, sex, smoking, physical activity, alcohol, BMI, education |
| Orlich, 2013, USA [21] | Adventist Health Study 2 (AHS-2) | 2002–2007 – 2009, 5.79 years follow-up |
73308 M/W, age ≥ 25 years: 987 CVD deaths 372 IHD deaths |
Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians, M Vegetarians vs. nonvegetarians, W Vegetarians vs. nonvegetarians, M Vegetarians vs. nonvegetarians, W Vegans vs. nonvegetarians Vegans vs. nonvegetarians |
CVD mort. IHD mort. CVD mort. CVD mort. IHD mort. IHD mort. CVD mort. IHD mort. |
0.87 (0.75–1.01) 0.81 (0.64–1.02) 0.71 (0.57–0.90) 0.99 (0.83–1.18) 0.71 (0.51–1.00) 0.88 (0.65–1.20) 0.91 (0.71–1.16) 0.90 (0.60–1.33) |
Age, sex (all), race, smoking, exercise, income, education, marital status, alcohol, region, sleep, and for women: menopause and hormone therapy |
| Appleby, 2016, UK pooled analysis [45] |
Oxford Vegetarian Study (OVS) + EPIC-Oxford |
1980–1984 – 2014 + 1993-1999 – 2014, 16.6 years follow-up |
60310 M/W, age 20–89 years: 433 circulatory disease deaths 175 IHD deaths 152 CBVD deaths |
Vegetarians vs. regular meat-eaters Vegetarians vs. regular meat-eaters Vegetarians vs. regular meat-eaters |
CircD mort.g IHD mort.g CBVD mort.g |
1.10 (0.95–1.27)g 0.99 (0.79–1.23)g 1.21 (0.94–1.56)g |
Age, smoking, alcohol, physical activity, married or cohabiting, nutritional supplementation, method of recruitment, sex, parity, oral contraceptive use, hormone replacement therapy use, prior diabetes, prior high blood pressure, receipt of long-term medical treatment |
| Tong, 2019, UK [23] | EPIC-Oxford | 1993–2001 – 2016, 18.1 years follow-up |
48188 M/W, age 20–90 years: 3892 CVD cases 2820 IHD cases 1072 stroke cases 519 ischemic stroke cases 300 hemorrhagic stroke cases |
Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters, M Vegetarians vs. meat-eaters, W Vegetarians vs. meat-eaters, M Vegetarians vs. meat-eaters, W Vegans vs. meat-eaters Vegans vs. meat-eaters Vegans vs. meat-eaters |
CVD inc.b CVD inc.b,e IHD inc. IHD inc.h IHD inc.e Total stroke inc. Total stroke inc.h Total stroke inc.e Ischemic stroke inc. Hemorr. stroke inc. IHD inc. IHD inc. Total stroke inc. Total stroke inc. CVD inc.b IHD inc. Total stroke inc. |
0.90 (0.82–0.98)b 0.90 (0.82–0.99)b,e 0.78 (0.70–0.87) 0.83 (0.75–0.92)h 0.80 (0.72–0.90)e 1.20 (1.02–1.40) 1.21 (1.03–1.42)h 1.19 (1.00–1.41)e 1.12 (0.90–1.41) 1.43 (1.08–1.90) 0.77 (0.66–0.91) 0.78 (0.68–0.90) 0.99 (0.73–1.34) 1.28 (1.07–1.53) 0.94 (0.75–1.17)b 0.82 (0.64–1.05) 1.35 (0.95–1.92) |
Age, sex (all), method of recruitment, region, year of recruitment, education, Townsend deprivation index, smoking, alcohol, physical activity, dietary supplements, plus oral contraceptive, and hormone replacement therapy use in women hBMI-adjusted model: same adjustments as M/W-model plus BMI |
| Chiu, 2020, Taiwan [25] |
Tzu Chi Health Study (TCHS, cohort 1) |
2007–2009 – 2014, 6.1 years follow-up |
5050 M/W, age ≥ 20 years: 54 stroke cases 31 ischemic stroke cases |
Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians |
Total stroke inc. Ischemic stroke inc. |
0.51 (0.25–1.06) 0.26 (0.08–0.88) |
Age, sex, smoking, alcohol, betel nut, physical activity, education, hypertension, diabetes mellitus, dyslipidaemia, IHD and BMI |
| Chiu, 2020, Taiwan [25] | Tzu Chi Vegetarian Study (TCVS, cohort 2) | 2005–2014, 9.25 years follow-up |
8302 M/W, age ≥ 20 years: 121 stroke cases 54 ischemic stroke cases 28 hemorrhagic stroke cases |
Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians Vegetarians vs. nonvegetarians |
Total stroke inc. Ischemic stroke inc. Hemorr. stroke inc. |
0.52 (0.33–0.82) 0.41 (0.19–0.88) 0.34 (0.12–1.00) |
Age, sex, smoking, alcohol, betel nut, physical activity, education, hypertension, diabetes mellitus, dyslipidaemia, IHD |
| Petermann-Rocha, 2021, UK [22] | UK Biobank (UKB) | 2006–2010 – 2020, 8.5 years follow-up |
422791 M/W, age 37–73 years: 106690 CVD cases 6580 CVD deaths 24794 IHD cases 2767 IHD deaths 5946 stroke cases 1088 stroke deaths |
Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters Vegetarians vs. meat-eaters, M Vegetarians vs. meat-eaters, W Vegetarians vs. meat-eaters, M Vegetarians vs. meat-eaters, W Vegetarians vs. meat-eaters, M Vegetarians vs. meat-eaters, W |
CVD inc. CVD inc.i CVD mort. IHD inc. IHD inc.i IHD mort. Total stroke inc. Total stroke inc.i Total stroke mort. CVD inc. CVD inc. IHD inc. IHD inc. Total stroke inc. Total stroke inc. |
0.91 (0.86–0.96) 0.84 (0.80–0.89)i 0.91 (0.71–1.17) 0.96 (0.85–1.07) 0.90 (0.80–1.01)i 0.88 (0.59–1.30) 0.84 (0.66–1.07) 0.82 (0.64–1.05)i 0.87 (0.48–1.58) 0.88 (0.81–0.96) 0.93 (0.87–0.99) 0.91 (0.78–1.07) 1.02 (0.87–1.20) 0.81 (0.56–1.18) 0.88 (0.63–1.22) |
Age, sex (all), deprivation, ethnicity, morbidity (43 diseases), smoking, sedentary time, alcohol, physical activity, BMI iModel with no BMI-adjustment: same adjustments as BMI-adjusted model except for BMI |
| Baden, 2021, USA [24] | Nurses’ Health Study 1 (NHS1) | 1984–2016, 28.3 years follow-up |
73890 W, age 30–55 years: 3604 stroke cases |
Vegetarians vs. nonvegetarians, W | Total stroke inc. | 1.03 (0.72–1.49) | Age, white race, smoking status, alcohol intake, physical activity, multivitamin use, aspirin use, margarine intake, total energy intake, BMI, postmenopausal hormone use |
| Baden, 2021, USA [24] | Nurses’ Health Study 2 (NHS2) | 1991–2017, 25.5 years follow-up |
92352 W, age 25–42 years: 740 stroke cases |
Vegetarians vs. nonvegetarians, W | Total stroke inc. | 0.71 (0.32–1.59) | Age, white race, smoking status, alcohol intake, physical activity, multivitamin use, aspirin use, margarine intake, total energy intake, BMI, postmenopausal hormone use |
| Baden, 2021, USA [24] | Health Professionals Follow-up Study (HPFS) | 1986–2012, 22.5 years follow-up |
43266 M, age 40–75 years: 1897 stroke cases |
Vegetarians vs. nonvegetarians, M | Total stroke inc. | 0.96 (0.59–1.58) | Age, white race, smoking status, alcohol intake, physical activity, multivitamin use, aspirin use, margarine intake, total energy intake, BMI, postmenopausal hormone use |
The relative risk estimates (RRs) compare either vegetarians to nonvegetarians or vegans to nonvegetarians and the main analysis uses incidence and mortality data on both sexes combined, but only incidence data if studies reported both
AHS-1 Adventist Health Study 1, AHS-2 Adventist Health Study 2, AMS Adventist Mortality Study, CBVD cerebrovascular disease, CI confidence interval, CircD circulatory disease, CVD cardiovascular disease, EPIC-Oxford European Prospective Investigation into Cancer and Nutrition – Oxford, HBS Heidelberg Study, hemorr. hemorrhagic, HFSS Health Food Shoppers Study, HPFS Health Professionals Follow-up Study, IHD ischemic heart disease, inc. incidence, M men, mort. mortality, NHS1 Nurses’ Health Study 1, NHS2 hemorr. hemorrhagic, Nurses’ Health Study 2, OVS Oxford Vegetarian Study, RR relative risk, TCHS Tzu Chi Health Study, TCVS Tzu Chi Vegetarian Study, UK United Kingdom, UKB UK Biobank Study, USA United States of America, W women
aThe data (RRs and 95% CIs) were calculated using the inverse method as original data compared nonvegetarians to vegetarians
bData on IHD and total stroke were pooled into RRs for CVD for these publications: Key 1999 (AMS, AHS-1) and Tong 2019 (EPIC-Oxford)
cThe risk estimate is a pooled RR and was only used in the analysis on vegan diets and the risk of CVD, IHD and total stroke
d−/d+The risk estimate is a pooled RR and was only used in a subgroup analysis on BMI-adjustment for vegetarian diets and IHD. Minus (-) indicates no BMI-adjustment. Pluss ( +) indicates BMI-adjustment
eModel with RR for exclusion of early follow-up: Appleby 2002 (HFSS, OVS) excluded the first 5 years of follow-up, Tong 2019 (EPIC-Oxford) excluded participants with less than 5 years of follow-up, and Petermann-Rocha 2021 (UKB) excluded the first two years of follow-up
fData on IHD and CBVD were pooled into RRs for CVD for one publication (Appleby 2002-HFSS, OVS)
gThe risk estimate is for EPIC-Oxford and OVS combined and was only used in subgroup analysis of vegetarian diets and the risk of mortality from CVD, IHD and total stroke
hModel with BMI adjustment
iModel without BMI-adjustment