Table 1.
Basic characteristics of the included prospective and case–cohort studies by exposures reported.
| First Author (year) |
Cohort Name | Country | Sample Size |
Follow-up (Median, y) | Dietary Assessment | Exposure Assessment | Outcome (n) | Comparison | Adjustments | NOS Score |
|---|---|---|---|---|---|---|---|---|---|---|
| Prospective cohort studies | ||||||||||
| Yamasaki (2015) [53] | JMS | Japan | 11,066 | 11.8 | Self-administered, FFQ, 30-item, and validated | Soy | CVDs deaths (198) | Almost daily vs. 1–2 times/week | a and b | 8 |
| Yamasaki (2015) [53] | JMS | Japan | 11,066 | 11.8 | Self-administered, FFQ, 30-item, and validated | Soy products | CVDs deaths (217) | Almost daily vs. 1–4 times per week | a and b | 8 |
| Nagata (2016) [48] | Takayama study | Japan | 29,079 | 16 | Self-administered, FFQ-169, and validated | Natto | CVDs deaths (1678), stroke (677), and CHD (308) | 7.3 vs. 0 g/d | a, b, and c | 9 |
| Zhang (2003) [57] | SWHS | China | 64,915 | 2.5 | Interview, FFQ, and validated | Total soy protein | CHD (62) | Q4 vs. Q1 ≥ 11.19 vs. <4.5 g/d | a, b, and c | 8 |
| Yan (2021) [31] | JACC Study | Japan | 21,925 | 5 | Self-administered, FFQ-40, and validated | Tofu | T2D (593) | Almost daily vs. <3 times/week | a, b, and c | 9 |
| Yan (2021) [31] | JACC Study | Japan | 21,925 | 5 | Self-administered, FFQ-40, and validated | Miso soup | T2D (593) | ≥3 bowls/day vs. <1 bowl/day | a, b, and c | 9 |
| Woo (2020) [35] | MR Cohort | Korea | 8269 | 6.18 | Interview, FFQ-106, and validated | Soy protein | T2D (531) | Q4 vs. Q1 | a, b, and c | 9 |
| Wang-a (2021) [12] | CKB study | China | 487,034 | 8.2 | Interview, FFQ, and validated | Soy | CVDs deaths (12582), CHD (3764), and stroke (5916) | ≥4 days per week vs. never or rarely | a, b, and c | 7 |
| Wang-b (2021) [12] | CKB study | China | 22,923 | 7.8 | Interview, FFQ, and validated | soy | CVDs deaths (2860), CHD (1123), and stroke (1473) | ≥4 days per week vs. never or rarely | a, b, and c | 7 |
| Villegas (2008) [30] | SWHS | China | 64,191 | 4.6 | Interview, FFQ-77, and validated | Soybeans | T2D (1605) | Q5 vs. Q1 32 vs. 2.8 g/d | a, b, and c | 9 |
| Tatsumi (2013) [34] | The Saku Study | Japan | 3039 | 4 | Self-administered, FFQ, and non-validated | soybean products | T2D (204) | ≥4 times/week vs. 0–1 time/week | a, b, and c | 7 |
| Talaei (2014) [52] | SCHS | China | 60,298 | 14.7 | Interview, FFQ-165, and validated | Tofu equivalents | CVDs deaths (4780), CHD (2697), and stroke (1298) | Q4 vs. Q1 197 vs. 42.8 g/d | a, b, and c | 9 |
| Nozue (2020) [51] | JPHC I, II | Japan | 79,648 | 13.3 | Self-administered, FFQ-138, and validated | Total soy products | CVDs events (4427), stroke (3743), and CHD (684) | Q4 vs. Q1 140.9 vs. 33.6 g/d | a, b, and c | 8 |
| Nouri (2021) [50] | ICS | Iran | 5432 | 13 | Interview, FFQ-48, and validated | Soybeans | CVDs events (751) | ≥1 times /week vs. <1 times/week | a and b | 7 |
| Ma (2020) [13] | NHS, NHSII, HPFS | USA | 210,700 | / | Self-administered, FFQ-130, and validated | Tofu | CHD events (8359) | ≥1 serving/week vs. <1 serving/month | a, b, and c | 6 |
| Ding (2016) [33] | NHS, NHSII, HPFS | USA | 163,457 | / | Self-administered, FFQ-116, and validated | Total soy food | T2D (9185) | ≥1 serving/week vs. non-consumer | a, b, and c | 6 |
| Nanri (2010) [38] | JPHC I, II | Japan | 59,791 | 10 | Self-administered, FFQ-147, and validated | Soy products | T2D (1114) | Q5 vs. Q1 ≥ 186 vs. 29 g/d | a, b, and c | 9 |
| Yu (2015) [54] | SWHS | China | 66,832 | 10 | Interview, FFQ-77, and validated | Total soy foods (dry weight) | IS (3110) | Q5 vs. Q1 33.5 vs. 6.6 g/d | a, b, and c | 9 |
| Im (2021) [42] | KGES | Korea | 4713 | 7.4 | Interview, FFQ, and validated | Total soy foods | CVDs events (282) | Q4 vs. Q1 18.03 vs. 5.27 servings/week | a, b, and c | 8 |
| Katagiri (2019) [43] | JPHC | Japan | 92,915 | 14.8 | Self-administered, FFQ-138, and validated | Total soy products | CVDs deaths (3326) and stroke (1326) | Q5 vs. Q1 178 vs. 37 g/d | a, b, and c | 9 |
| Kobayashi (2019) [44] | JPHC | Japan | 79,904 | 14.9 | Self-administered, FFQ-138, and validated | Soy dietary diversity | CVDs deaths (2942) | Q5 vs. Q1 3.5 vs. 1.1 of food items/day | a, b, and c | 9 |
| Kokubo (2007) [45] | JPHC I | Japan | 40,462 | 12.5 | Self-administered, FFQ-44, and NA | Soy | stroke (1230) and MI (308) | Highest vs. lowest ≥ 5 vs. 0–2 days/week | a, b, and c | 8 |
| Konishi (2019) [36] | Takayama study | Japan | 13,521 | 10 | Self-administered, FFQ-169, and validated | Total soy foods | T2D (438) | Q3 vs. Q1 141.2 vs. 50.7 g/d | a, b, and c | 8 |
| Nguyen (2018) [49] | NIP-PON DATA | Japan | 9244 | 24 | Weigh and record all foods and beverages on 3 consecutive days | Tofu | stroke (417) | Q4 vs. Q1 41.5 vs. 0.8 g/1000 kcal | a, b, and c | 8 |
| Mueller (2011) [37] | SCHS | China | 43,176 | 5.7 | Interview, FFQ-165, and validated | Unsweetened soy | T2D (2252) | ≥5/week vs. none | a, b, and c | 8 |
| Case-control studies | Baseline years a | |||||||||
| Guo (2013) [40] | ICS | China | 3547 | 1999.2–2003.3 | Self-reported, FFQ-19, and validated | Tofu | CHD events (1312) | >3 vs. <0.75 times/week | a and b | 6 |
| Liang (2009) [46] | / | China | 838 | 2007–2008 | Interview, FFQ-125, and validated | Total soy foods | IS (374) | ≥300 vs. <50 g | a, b, and c | 7 |
| Okamoto (2006) [56] | / | Japan | 411 | 1992.4–1997.3 | Interview, FFQ, and NA | Soy products | SAH (201) | Q4 vs. Q1 | a and b | 8 |
| Fang (2006) [39] | / | USA | 391 | 2000.9–2003.6 | Interview, FFQ-49, and validated | Soybean | stroke (187) | ≥3 vs. <3 times/week | a | 5 |
| Ho (2006) [41] | / | China | 32,462 | 1997.12–1999.01 | Proxy report, FFQ, and NA | Soy | stroke (2160) and fatal IHD (2016) | 4+/week vs. <1/mon | a, b, and c | 6 |
| Sasazuki (2001) [55] | FHS | Japan | 1846 | 1996.9–1998.9 | Interview, FFQ-23, and NA | Tofu | AMI (632) | 4+/week vs. <2/week | a and b | 8 |
| Sasazuki (2001) [55] | FHS | Japan | 1846 | 1996.9–1998.9 | Interview, FFQ-23, and NA | Miso soup | AMI (632) | 2+/week vs. <1/week | b and c | 8 |
| Nguyen (2017) [32] | / | Vietnam | 1198 | 2013.8–2015.10 | Interview, FFQ-128, and validated | Total soy foods | T2D (599) | >133.9 vs. = <44.1 g/d | a, b, and c | 7 |
Abbreviations: CVDs: cardiovascular diseases; CHD: coronary heart disease; IS: ischemic stroke; IHD: ischemic heart disease; SAH: subarachnoid hemorrhage; AMI: acute myocardial infarction; MI: myocardial infarction; T2D: type 2 diabetes; FFQ: Food Frequency Questionnaire; ICD: international classification of disease; JMS: Jichi Medical School Study; SWHS: Shanghai Women’s Health Study; JACC Study: Japan Collaborative Cohort Study for Evaluation of Cancer Risk; MR Cohort: Korean Multi-Rural Communities Cohort; CKB study: China Kadoorie Biobank study; SCHS: Singapore Chinese Health Study; JPHC: Japan Public Health Center-Based Study; JPHC I: Japan Public Health Center-Based Study Cohort I; JPHC I,II: Japan Public Health Center-Based Study Cohort I and Cohort II; ICS: Isfahan cohort study; NHS: Nurses’ Health Study; NHSII: Nurses’ Health Study II; HPFS: Health Professionals Follow-Up Study; KGES: Korean Genome and Epidemiology Study; NIP-PON DATA: National Integrated Project for Prospective Observation of Non-communicable Disease and its Trends in the Aged; ICS: INTERHEART China study; FHS: Fukuoka Heart Study; ADA: American Diabetes Association; WHO: World Health Organization; ECG: electrocardiogram; NA: not available; NOS: the Newcastle–Ottawa Scale; M: men; Mon: month; g: grams; g/d: grams/day; Q: quartile; y: years; Wang-a: individuals free of cardiovascular diseases at study baseline; and Wang-b: individuals with a history of cardiovascular diseases at study baseline. Degrees of adjustment for confounders were as follows: model a: sociodemographics, model b: risk factors for CVDs, and model c: dietary variables and other risk factors. a Baseline years were defined as included in the study analyzed.