Supplementary Material SUPPLEMENTARY TABLES Supplementary Table 1. Official Global, N America & Europe Recommendation on Supplements for CVD & Cancer (2004-2016) Supplementary Table 2. Search Strategy for vitamin and mineral supplements for cardiovascular diseases & all-cause mortality Supplementary Table 3. Characteristics of included RCT studies for CVD and all-cause mortality Supplementary Table 4. GRADE assessment for vitamin D Supplementary Table 5. GRADE assessment for vitamin A Supplementary Table 6. GRADE assessment for beta-carotene Supplementary Table 7. GRADE assessment for antioxidants Supplementary Table 8. GRADE assessment for vitamin E Supplementary Table 9. GRADE assessment for vitamin C Supplementary Table 10. GRADE assessment for selenium Supplementary Table 11. GRADE assessment for vitamin B-complex Supplementary Table 12. GRADE assessment for folic acid Supplementary Table 13. GRADE assessment for niacin (B3) Supplementary Table 14. GRADE assessment for vitamin B6 Supplementary Table 15. GRADE assessment for calcium Supplementary Table 16. GRADE assessment for iron Supplementary Table 17. GRADE assessment for multivitamin Supplementary Table 18. GRADE assessment for calcium and vitamin D Supplementary Table 19. Summary of the meta-analayses results for all-cause mortality, CVD mortality, total CVD risk and other significant assocations SUPPLEMENTARY FIGURES Supplementary Figure 1. Consort statement Supplementary Figure 2. Risk of bias graph for supplements and CVD and total mortality. 1
224
Embed
ars.els-cdn.com · Web viewFunnel plot of vitamin D supplementation and CVD and all-cause mortality risk.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Supplementary Material
SUPPLEMENTARY TABLES Supplementary Table 1. Official Global, N America & Europe Recommendation on Supplements for CVD & Cancer (2004-2016) Supplementary Table 2. Search Strategy for vitamin and mineral supplements for cardiovascular diseases & all-cause mortalitySupplementary Table 3. Characteristics of included RCT studies for CVD and all-cause mortalitySupplementary Table 4. GRADE assessment for vitamin DSupplementary Table 5. GRADE assessment for vitamin ASupplementary Table 6. GRADE assessment for beta-caroteneSupplementary Table 7. GRADE assessment for antioxidantsSupplementary Table 8. GRADE assessment for vitamin ESupplementary Table 9. GRADE assessment for vitamin CSupplementary Table 10. GRADE assessment for seleniumSupplementary Table 11. GRADE assessment for vitamin B-complexSupplementary Table 12. GRADE assessment for folic acidSupplementary Table 13. GRADE assessment for niacin (B3)Supplementary Table 14. GRADE assessment for vitamin B6Supplementary Table 15. GRADE assessment for calciumSupplementary Table 16. GRADE assessment for ironSupplementary Table 17. GRADE assessment for multivitaminSupplementary Table 18. GRADE assessment for calcium and vitamin DSupplementary Table 19. Summary of the meta-analayses results for all-cause mortality, CVD mortality, total CVD risk and other significant assocations
SUPPLEMENTARY FIGURES Supplementary Figure 1. Consort statementSupplementary Figure 2. Risk of bias graph for supplements and CVD and total mortality.Supplementary Figure 3. Summary of the pooled effect estimates of RCTs assessing the relationship between vitamin D supplementation and CVD and all cause mortality risk.Supplementary Figure 4. Forest plot of vitamin D supplementation and total CVD risk.Supplementary Figure 5. Forest plot of vitamin D supplementation and total CHD riskSupplementary Figure 6. Forest plot of vitamin D supplementation and MI risk.Supplementary Figure 7. Forest plot of vitamin D supplementation and stroke risk.Supplementary Figure 8. Forest plot of vitamin D supplementation and CVD mortality risk.Supplementary Figure 9. Forest plot of vitamin D supplementation and CHD mortality risk.Supplementary Figure 10. Forest plot of vitamin D supplementation and MI mortality risk. Supplementary Figure 11. Forest plot of vitamin D supplementation and stroke mortality risk.
1
Supplementary Figure 12. Forest plot of vitamin D supplementation and all-cause mortality risk.Supplementary Figure 13. Funnel plot of vitamin D supplementation and CVD and all-cause mortality risk.Supplementary Figure 14. Summary of the pooled effect estimates of RCTs assessing the relationship between vitamin A supplementation and CVD and all cause mortality risk.Supplementary Figure 15. Forest plot of vitamin A supplementation and all-cause mortality risk.Supplementary Figure 16. Summary of the pooled effect estimates of RCTs assessing the relationship between beta-carotene supplementation and CVD and all cause mortality risk.Supplementary Figure 17. Forest plot of beta-carotene supplementation and total CVD risk.Supplementary Figure 18. Forest plot of beta-carotene supplementation and total CHD risk.Supplementary Figure 19. Forest plot of beta-carotene supplementation and MI risk.Supplementary Figure 20. Forest plot of beta-carotene supplementation and stroke risk.Supplementary Figure 21. Forest plot of beta-carotene supplementation and CVD mortality risk.Supplementary Figure 22. Forest plot of beta-carotene supplementation and CHD mortality risk.Supplementary Figure 23. Forest plot of beta-carotene supplementation and MI mortality risk.Supplementary Figure 24. Forest plot of beta-carotene supplementation and stroke mortality risk.Supplementary Figure 25. Forest plot of beta-carotene supplementation and all-cause mortality risk.Supplementary Figure 26. Summary of the pooled effect estimates of RCTs assessing the relationship between antioxidants supplementation and CVD and all cause mortality risk.Supplementary Figure 27. Forest plot of antioxidants supplementation and total CVD risk.Supplementary Figure 28. Forest plot of antioxidants supplementation and total CHD risk.Supplementary Figure 29. Forest plot of antioxidants supplementation and MI risk.Supplementary Figure 30. Forest plot of antioxidants supplementation and stroke risk.Supplementary Figure 31. Forest plot of antioxidants supplementation and CVD mortality risk.Supplementary Figure 32. Forest plot of antioxidants supplementation and CHD mortality risk.Supplementary Figure 33. Forest plot of antioxidants supplementation and MI mortality risk. Supplementary Figure 34. Forest plot of antioxidants supplementation and stroke mortality risk.Supplementary Figure 35. Forest plot of antioxidants supplementation and all-cause mortality risk.Supplementary Figure 36. Funnel plot of antioxidants supplementation and all-cause mortality risk.Supplementary Figure 37. Summary of the pooled effect estimates of RCTs assessing the relationship between vitamin E supplementation and CVD and all cause mortality risk.Supplementary Figure 38. Forest plot of vitamin E supplementation and total CVD risk.
2
Supplementary Figure 39. Forest plot of vitamin E supplementation and total CHD risk.Supplementary Figure 40. Forest plot of vitamin E supplementation and MI risk.Supplementary Figure 41. Forest plot of vitamin E supplementation and stroke risk.Supplementary Figure 42. Forest plot of vitamin E supplementation and CVD mortality risk.Supplementary Figure 43. Forest plot of vitamin E supplementation and CHD mortality risk.Supplementary Figure 44. Forest plot of vitamin E supplementation and MI mortality risk.Supplementary Figure 45. Forest plot of vitamin E supplementation and stroke mortality risk.Supplementary Figure 46. Forest plot of vitamin E supplementation and all-cause mortality risk.Supplementary Figure 47. Funnel plot of vitamin E supplementation and CVD and all-cause mortality risk.Supplementary Figure 48. Summary of the pooled effect estimates of RCTs assessing the relationship between vitamin C supplementation and CVD and all cause mortality risk.Supplementary Figure 49. Forest plot of vitamin C supplementation and total CVD risk.Supplementary Figure 50. Forest plot of vitamin C supplementation and total CHD risk.Supplementary Figure 51. Forest plot of vitamin C supplementation and MI risk.Supplementary Figure 52. Forest plot of vitamin C supplementation and stroke risk.Supplementary Figure 53. Forest plot of vitamin C supplementation and CVD mortality risk.Supplementary Figure 54. Forest plot of vitamin C supplementation and MI mortality risk.Supplementary Figure 55. Forest plot of vitamin C supplementation and stroke mortality risk.Supplementary Figure 56. Forest plot of vitamin C supplementation and all-cause mortality risk.Supplementary Figure 57. Summary of the pooled effect estimates of RCTs assessing the relationship between selenium supplementation and CVD and all cause mortality risk.Supplementary Figure 58. Forest plot of selenium supplementation and total CVD risk.Supplementary Figure 59. Forest plot of selenium supplementation and total CHD risk.Supplementary Figure 60. Forest plot of selenium supplementation and MI risk.Supplementary Figure 61. Forest plot of selenium supplementation and stroke risk.Supplementary Figure 62. Forest plot of selenium supplementation and CVD mortality risk.Supplementary Figure 63. Forest plot of selenium supplementation and MI mortality risk.Supplementary Figure 64. Forest plot of selenium supplementation and stroke mortality risk.Supplementary Figure 65. Forest plot of selenium supplementation and all-cause mortality risk.Supplementary Figure 66. Summary of the pooled effect estimates of RCTs assessing the relationship between vitamin B complex supplementation and CVD and all cause mortality risk.Supplementary Figure 67. Forest plot of vitamin B complex supplementation and total CVD risk.Supplementary Figure 68. Forest plot of vitamin B complex supplementation and total CHD risk.Supplementary Figure 69. Forest plot of vitamin B complex supplementation and MI risk.Supplementary Figure 70. Forest plot of vitamin B complex supplementation and stroke risk.Supplementary Figure 71. Forest plot of vitamin B complex supplementation and CVD mortality risk.
3
Supplementary Figure 72. Forest plot of vitamin B complex supplementation and CHD mortality risk.Supplementary Figure 73. Forest plot of vitamin B complex supplementation and MI mortality risk.Supplementary Figure 74. Forest plot of vitamin B complex supplementation and stroke mortality risk.Supplementary Figure 75. Forest plot of vitamin B complex supplementation and all-cause mortality risk.Supplementary Figure 76. Funnel plot of vitamin B complex supplementation and CVD risk and all-cause mortality risk.Supplementary Figure 77. Summary of the pooled effect estimates of RCTs assessing the relationship between folic acid supplementation and CVD and all cause mortality risk.Supplementary Figure 78. Forest plot of folic acid supplementation and total CVD risk.Supplementary Figure 79. Forest plot of folic acid supplementation and total CHD risk.Supplementary Figure 80. Forest plot of folic acid supplementation and MI risk.Supplementary Figure 81. Forest plot of folic acid supplementation and stroke risk.Supplementary Figure 82. Forest plot of folic acid supplementation and CVD mortality risk.Supplementary Figure 83. Forest plot of folic acid supplementation and MI mortality risk.Supplementary Figure 84. Forest plot of folic acid supplementation and stroke mortality risk.Supplementary Figure 85. Forest plot of folic acid supplementation and all-cause mortality risk.Supplementary Figure 86. Funnel plot of folic acid supplementation and all-cause mortality risk.Supplementary Figure 87. Summary of the pooled effect estimates of RCTs assessing the relationship between niacin (B3) supplementation and CVD and all cause mortality risk.Supplementary Figure 88. Forest plot of niacin (B3) supplementation and total CVD risk.Supplementary Figure 89. Forest plot of niacin (B3) supplementation and total CHD risk.Supplementary Figure 90. Forest plot of niacin (B3) supplementation and MI risk.Supplementary Figure 91. Forest plot of niacin (B3) supplementation and stroke risk.Supplementary Figure 92. Forest plot of niacin (B3) supplementation and CVD mortality risk.Supplementary Figure 93. Forest plot of niacin (B3) supplementation and CHD mortality risk.Supplementary Figure 94. Forest plot of niacin (B3) supplementation and all-cause mortality risk.Supplementary Figure 95. Summary of the pooled effect estimates of RCTs assessing the relationship between vitamin B6 supplementation and CVD and all cause mortality risk.Supplementary Figure 96. Forest plot of vitamin B6 supplementation and total CVD risk.Supplementary Figure 97. Forest plot of vitamin B6 supplementation and MI risk.Supplementary Figure 98. Forest plot of vitamin B6 supplementation and stroke risk.Supplementary Figure 99. Forest plot of vitamin B6 supplementation and MI mortality risk.Supplementary Figure 100. Forest plot of vitamin B6 supplementation and all-cause mortality risk.Supplementary Figure 101. Summary of the pooled effect estimates of RCTs assessing the relationship between calcium supplementation and CVD and all cause mortality risk.Supplementary Figure 102. Forest plot of calcium supplementation and total CVD risk.Supplementary Figure 103. Forest plot of calcium supplementation and total CHD risk.
4
Supplementary Figure 104. Forest plot of calcium supplementation and MI risk.Supplementary Figure 105. Forest plot of calcium supplementation and stroke risk.Supplementary Figure 106. Forest plot of calcium supplementation and CVD mortality risk.Supplementary Figure 107. Forest plot of calcium supplementation and MI mortality risk.Supplementary Figure 108. Forest plot of calcium supplementation and stroke mortality risk.Supplementary Figure 109. Forest plot of calcium supplementation and all-cause mortality risk.Supplementary Figure 110. Summary of the pooled effect estimates of RCTs assessing the relationship between iron supplementation and CVD and all cause mortality risk.Supplementary Figure 111. Forest plot of iron supplementation and total CVD risk.Supplementary Figure 112. Forest plot of iron supplementation and MI risk.Supplementary Figure 113. Forest plot of iron supplementation and CVD mortality risk.Supplementary Figure 114. Forest plot of iron supplementation and all-cause mortality risk.Supplementary Figure 115. Summary of the pooled effect estimates of RCTs studies assessing the relationship between multivitamins supplementation and CVD and all cause mortality risk.Supplementary Figure 116. Forest plot of multivitamins supplementation and total CVD risk.Supplementary Figure 117. Forest plot of multivitamins supplementation and MI risk.Supplementary Figure 118. Forest plot of multivitamins supplementation and stroke risk.Supplementary Figure 119. Forest plot of multivitamins supplementation and CVD mortality risk.Supplementary Figure 120. Forest plot of multivitamins supplementation and MI mortality risk.Supplementary Figure 121. Forest plot of multivitamins supplementation and stroke mortality risk.Supplementary Figure 122. Forest plot of multivitamins supplementation and all-cause mortality risk.Supplementary Figure 123. Funnel plot of multivitamin supplementation and all-cause mortality risk.Supplementary Figure 124. Summary of the pooled effect estimates of RCTs assessing the relationship between calcium and vitamin D supplementation and CVD and all cause mortality risk.Supplementary Figure 125. Forest plot of calcium and vitamin D supplementation and total CVD risk.Supplementary Figure 126. Forest plot of calcium and vitamin D supplementation and total CHD risk.Supplementary Figure 127. Forest plot of calcium and vitamin D supplementation and MI risk.Supplementary Figure 128. Forest plot of calcium and vitamin D supplementation and stroke risk.Supplementary Figure 129. Forest plot of calcium and vitamin D supplementation and CVD mortality risk.Supplementary Figure 130. Forest plot of calcium and vitamin D supplementation and CHD mortality risk.
5
Supplementary Figure 131. Forest plot of calcium and vitamin D supplementation and MI mortality risk.Supplementary Figure 132. Forest plot of calcium and vitamin D supplementation and stroke mortality risk.Supplementary Figure 133. Forest plot of calcium and vitamin D supplementation and all-cause mortality risk.Supplementary Figure 134. Funnel plot of vitamin D & calcium supplementation and all-cause mortality risk.
6
Supplementary Table 1. Official Global, N America & Europe Recommendation on Supplements for CVD & Cancer (2004-2016)
Official Body (Ref) Cardiovascular Disease
Date Supplement Recommendation for Vitamins, Minerals & Multivitamins minerals
Standard Diet recommendations (fruit, vegetables, wholegrain cereals = (+))
World Heart Federation (1) 2016 no supplement recommendation (+) low saturated & trans fats, low sodium, fish, nuts, soy, alcohol in moderation
National Cholesterol Education Program ATP III (2,3)
2001 & Update 2004
no supplement recommendation as such possibly nicotinic acid for low HDL-C and high TG
American Heart Association’s/American College of Cardiology(4)
2013 no supplement recommendation (+) low fat dairy, poultry, fish, legumes non tropical vegetables oils & nuts. Limit sweets, sugar sweetened beverages & red meats.
Canadian Cardiovascular Society(5)
2012 no supplement recommendation but vitamins, minerals or supplements i.e., CoQ10, not recommended even for statin associated myalgia
(+)PUFA, MUFA, n-3 fats from fish avoid trains, limited saturated and total fat <7%, <30% total energy fiber>30g, cholesterol ≤ 200mg/day plus nuts, soy, viscous fibers, plant sterols
Canadian Cardiovascular Society (6)
2015 no supplement recommendation addition of Niacin to statins with lipids at target not recommended due to AIM-HIGH and HPS-2 THRIVE trial lack of benefit and possible harm. With high TG or low HDL-C or LDL-C above target-possible use of niacin
(+)Mediterranean diet. Do NOT recommend n-3 fatty acid supplements. High doses for marine, algal & yeast sources (2-4g/d) may lower raised TG. Replace sat fat with n-3/n-6
7
mixed sources (canola & soy bean) plus plant source of MUFA (olive, canola, nuts & seeds)
European Society of Cardiology/ European Atherosclerosis Society(7)
2011 no supplement recommendation but Nicotinic acid may be used with statin. But the results of HPS-2 THRIVE and AIM-HIGH trials are awaited
(+) legumes (inc. soy & soy protein), fish poultry without skin, skimmed milk. Moderation in refined cereals, processed fruit (dried, jams, canned etc) red meat, seafood, nuts. But “nutraceuticals” (phytosterols, soy protein, viscous fibers, n-3 fatty acids, polycosanols & red yeast rice) can be used as alternatives or in addition to lipid lowering drugs
European Atherosclerosis Society (Statin associated Muscle Symptoms -SAMS consensus statement)(8)
2015 Supplements Not Recommended. Niacin lowers LDL-C by 15-20% but recent large RCTs showed significant adverse events and no CVD benefit when added to statin.Co-Q10 and Vitamin D failed to reduce SAMS. Therefore not advised.Red Yeast Rice 20-30% LDL-C reduction. Need longer term studies
(+)(as in 2011 assumed since no diet details given) Low saturated fat & Avoid trans-fat plus incorporate a portfolio of plant sterol, soy protein , viscous fibers and nuts, appropriate for SAMS alone or with statin or non-statin drug therapy
British Heart Foundation- Health Eating(9)
N/A No supplement Recommendation (+) plus some dairy products, some meat, fish, eggs, beans and non-dairy product small amount of drinks high in fat or sugar, replace saturated fats mono & poly saturated fats
CVD & CancerUS Preventive Services Task Force (10)
2003 revised 2014
No Supplement Recommendation. Too little evidence on harms versus benefits. Except Recommendation AGAINST β-Carotene & vitamin E supplementation, increased cancer
(+) (Endorse dietary guidelines of 2010) plus Fat free and low fat dairy & seafood.
8
risks in smokers with β-Carotene and possible hemorrhagic stroke with vitamin E
CancerWorld Cancer Research Fund International(11)
N/A Supplements not recommended. Do not rely on supplements for Cancer prevention.Meet nutritional needs through diet alone (special situations, e.g., illness, supplements may be of value)
(+) (fruit and non-starchy (colored) vegetables 600g/d plus unprocessed cereals ≥ 25g fiber/d) pulses (legumes) with every meal
American Cancer Society (12)
20121/11/201220162/5/2016 revised
Supplements not recommended. Evidence not clear therefore not advised e.g.↑ Ca = ↓ colon but ↑ prostate; β-Carotene may ↑ lung cancer in smokers. Folate may increase risk of prostate, and colorectal cancer and possibly breast. Selenium not effective not recommended.Vitamin E may even increase prostate cancer risk
(+) plus less red & processed meat maintain healthy weight, limit alcohol
Canadian Cancer Society (13)
2016 No supplement recommendation except Vitamin D 1000 IU autumn oil winter – many benefit and not increase cancer risk
(+) (as in Canada’s food Guide) plus reduce portion size
European Code against cancer 4th Edition (14)
2015 No Supplement recommendation; Dietary supplements are not recommended for cancer prevention (World Cancer Research Fund 2007)
(+) plus reduce body fat avoid energy dense foods, sugary drinks, eat mostly plant foods, limit red meat, alcohol ≤ 2 drinks/d for men ≤ 1 drink /d for women
Cancer Research UK(15) 2014 (updated 24 Oct)
Supplements not recommended for prevention. Evidence for a healthy diet & cancer prevention.Supplements may interact with treatment therefore advice against antioxidant supplement use. But supplements may have a benefit in life prolongation of patients with advanced cancer
(+) plus a healthy diet provides micronutrients
9
Supplementary Table 2. Search Strategy for vitamin and mineral supplements for cardiovascular diseases & all-cause mortality
Databases Search TermsMEDLINE 1. exp Dietary Supplements/ Or supplement*.mp.
2. exp Vitamin D/ Or exp Cholecalciferol/ Or Vitamin D.mp. Or exp Vitamin A/ Or vitamin A.mp. Or retinol.mp. Or exp beta Carotene/ Or alpha carotene.mp. Or exp Antioxidants/ Or antioxidant.mp. Or Vitamin E.mp. Or exp alpha-Tocopherol/ Or tocopherol.mp. Or Vitamin C.mp. Or ascorbic acid.mp. Or ascorbic acid.mp. Or exp Selenium/ Or selenium.mp. Or exp Vitamin B Complex/ Or B complex.mp. Or exp Folic Acid/ Or folate.mp. Or exp Niacinamide/ Or Vitamin B3.mp. Or exp Niacin/ Or exp Vitamin B 6/ Or pyridoxine.mp. Or Vitamin B6.mp. Or exp Calcium/ Or exp Calcium Carbonate/ Or exp Iron/ Or Iron.mp. Or exp Magnesium/ Or magnesium.mp. Or exp Potassium/ Or potassium.mp. Or Zinc.mp. Or exp Zinc Or Multivitamin.mp. Or multi vitamin.mp. Or (exp Calcium/ Or calcium carbonate.mp. AND exp Vitamin D/ Or exp Cholecalciferol/)3. exp Cardiovascular Diseases/ Or exp Myocardial Infarction/ Or exp Stroke/ Or cardiovascular death.mp. Or exp Mortality/ Or all-cause mortality.mp. Or exp Death/ Or death.mp. Or cancer death.mp. Or cancer mortality.mp.4. 1 AND 2 AND 35. limit 4 to (yr="2012 -Current" and (meta analysis or observational study or randomized controlled trial))
PUB MED 1. ("Dietary Supplements" Or supplements) AND ("Vitamin D" Or Cholecalciferol Or "Vitamin A" Or retinol Or "beta-Carotene" Or "alpha-carotene" Or Antioxidants Or "Vitamin E" Or "alpha-Tocopherol" Or tocopherol Or "Vitamin C" Or "ascorbic acid" Or Selenium Or "Vitamin B-Complex" Or "Folic Acid" Or folate Or Niacinamide Or "Vitamin B3" Or Niacin Or "Vitamin B6" Or pyridoxine Or Calcium Or "Calcium Carbonate" Or Iron Or Magnesium Or Potassium Or Zinc Or Multivitamin Or "multi-vitamin" Or (Calcium Or "calcium carbonate" AND "Vitamin D" Or Cholecalciferol)) AND ("Cardiovascular Diseases" Or "Myocardial Infarction" Or Stroke Or "cardiovascular death" Or Mortality Or "all-cause mortality" Or Death Or "cancer death" Or "cancer mortality")2. limit 1 to (yr="2012/01/01 - 2017/12/31" and (meta analysis or observational study or randomized controlled trial))
10
COCHRANE 1. dietary supplements.mp. [mp=title, abstract, full text, keywords, caption text] Orsupplements.mp. [mp=title, abstract, full text, keywords, caption text] "2. vitamin D.mp. [mp=title, abstract, full text, keywords, caption text] Or cholecalciferol.mp. [mp=title, abstract, full text, keywords, caption text] Or Vitamin A.mp. [mp=title, abstract, full text, keywords, caption text] Or retinol.mp. [mp=title, abstract, full text, keywords, caption text] Or beta-carotene.mp. [mp=title, abstract, full text, keywords, caption text] Or alpha-carotene.mp. [mp=title, abstract, full text, keywords, caption text] Or Antioxidant*.mp. [mp=title, abstract, full text, keywords, caption text] Or Vitamin E.mp. [mp=title, abstract, full text, keywords, caption text] Or alpha-tocopherol.mp. [mp=title, abstract, full text, keywords, caption text] Or tocopherol.mp. [mp=title, abstract, full text, keywords, caption text] Or Vitamin C.mp. [mp=title, abstract, full text, keywords, caption text] Or ascorbic acid.mp. [mp=title, abstract, full text, keywords, caption text] Or Selenium.mp. [mp=title, abstract, full text, keywords, caption text] Or B-complex.mp. [mp=title, abstract, full text, keywords, caption text] Or Folate.mp. [mp=title, abstract, full text, keywords, caption text] Or Folic acid.mp. [mp=title, abstract, full text, keywords, caption text] Or niacinamide.mp. [mp=title, abstract, full text, keywords, caption text] Or vitamin B3.mp. [mp=title, abstract, full text, keywords, caption text] Or niacin.mp. [mp=title, abstract, full text, keywords, caption text] Or vitamin B6.mp. [mp=title, abstract, full text, keywords, caption text] Or pyridoxine.mp. [mp=title, abstract, full text, keywords, caption text] Or calcium.mp. [mp=title, abstract, full text, keywords, caption text] Or calcium carbonate.mp. [mp=title, abstract, full text, keywords, caption text] Or iron.mp. [mp=title, abstract, full text, keywords, caption text] Or magnesium.mp. [mp=title, abstract, full text, keywords, caption text] Or potassium.mp. [mp=title, abstract, full text, keywords, caption text] Or zinc.mp. [mp=title, abstract, full text, keywords, caption text] Or multivitamin.mp. [mp=title, abstract, full text, keywords, caption text] Or multi-vitamin.mp. [mp=title, abstract, full text, keywords, caption text]3. cardiovascular diseases.mp. [mp=title, abstract, full text, keywords, caption text] Or stroke.mp. [mp=title, abstract, full text, keywords, caption text] Or myocardial infarction.mp. [mp=title, abstract, full text, keywords, caption text] Or cardiovascular death.mp. [mp=title, abstract, full text, keywords, caption text] Or all-cause mortality.mp. [mp=title, abstract, full text, keywords, caption text] Or death.mp. [mp=title, abstract, full text, keywords, caption text] Or mortality.mp. [mp=title, abstract, full text, keywords, caption text] Or cancer mortality.mp. [mp=title, abstract, full text, keywords, caption text] Or cancer death.mp. [mp=title, abstract, full text, keywords, caption text]4. 1 AND 2 AND 3
11
5. limit 4 to ((meta analysis or randomized controlled trial) and yr="2012 -Current") [Limit not valid; records were retained]
* Original search date for all databases was Oct 21st 2016; update search date for all databases was October 27th 2017
Supplementary Table 3. Characteristics of included RCT studies for CVD and all-cause mortality
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
Cardiovascular disease (CVD) outcomes
Brohult et al., 1973 (16) Sweden Rheumatoid
arthritis
18-69Mean:
521 y N/A
Vitamin D:100000 IU
calciferol per day25/25 1/0 All-cause
mortality Agency
CDPRG 1975(17)
United States CHD 30-64
Mean:74
months
Follow up visit
consisting of an assessment
to drug adherence
Vitamin B3:3.0 gm/day 1119/2789
914/2333 Total CVD
Agency
114/386 MI95/311 Stroke
238/633 CVD mortality
203/535 CHD mortality
273/709 All-cause mortality
Gillilan et al., 1977 (18) USA
Angina pectoris with obstructive coronary disease
57 (mean)
Vitamin E: 189 ± 15.0 daysPlacebo:
192 ± 13.3 days
capsule count and urine
fluorescence test
Vitamin E: 1600IU daily 26/26 2/2 All-cause
mortality Industry
Inkovaara et al., 1983(19)
Finland Healthy 79.5 1 y Serumcalcium,
phosphate, creatinine,
transaminase and alkaline phosphatase
were measured
Vitamin D:1000 IU /d
45/42 0/1 Total CHD
N/A
12/5 Stroke
0/1 CHD mortality
4/2 Stroke mortality
7/5 All-cause mortality
12
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
Calcium+ Vitamin D:3g calcium
carbonate and 1000 IU vitamin D
46/42
¼ Total CHD
9/5 Stroke
1/3 CHD mortality
2/2 Stroke mortality
7/5 All-cause mortality
Corless et al., 1985 (20)
United Kingdom
Hospital for >4 weeks otherwise
healthy82 40 weeks
Plasma 25(OH)D measured
Vitamin D2 9000 IU /d 41/41 8/8 All-cause
mortality Agency
Aloia et al., 1988(21) USA Postmenopausal
with osteoporosis 50-80 2 y Capsule count
Vitamin D:0.50 µg calcitriol/d
with doseescalation if necessary.
Average: 0.8 µg calcitriol/d
12/15 F 1/1 MI Industry
Korpela et al., 1989(22) Finland Patients with
acute MIMean:
57 6 months
Serum selenium
concentration measured
Selenium(selenium-rich
yeast, 100 µg/day)40/41
1/2 MI
N/A
0/4 CVD mortality
McKeown-Eyssen et al.,
1988 (23)Canada
Patients believed to be free of polyps after
removal of at least
one colorectal polyp
58.0 2 yearsrandom urine
sample 400 mg each of vitamins C and E 96/89 4/3 All-cause
mortality Industry
13
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
Ott et al., 1989(24) USA
Postmenopausal with
>2 compression fractures
50-80 2 y N/A.
Vitamin D:0.25 g calcitriol
twice dailywith dose
escalation ifneeded
43/43 F
0/1 MI mortality
Agency
0/1 F All-cause mortality
Gallagher et al., 1990 (25) USA
postmenopausal women with
vertebral fractures
50-78Mean: 69.7
2 y N/A
Vitamin D:From 0.25µg
calcitriol twice daily until a
maximum dose of 1µg twice daily
25/25 F 1/0 F All-cause mortality Industry
Grady et al., 1991 (26) USA Healthy >69 6 months
Serum calcium was
measured and capsule count at each follow
up visit
0.25 g vitamin D3
orally, twice per day
50/48 1/0 All-cause mortality Industry
Chapuy et al., 1992 (27) France Healthy 84 3 y
Taken with a nurse
supervision
1.2g calcium and 20g (800 IU) of
vitamin D3/d1634/1636 F 258/274 All-cause
mortality Agency
Li et al., 1993 – NIT2(28) China
Under nourished population with a
previous cytologic
diagnosis of esophageal squamous dysplasia
40-69 6 y Pill count Multivitamins 1657/1661
22/35Stroke
mortality
Agency
157/167 All-cause mortality
Bogden et al., 1994 (29) USA Healthy 59-85 1 y n/a Multivitamins 33/32 0/1 MI
Agency &
Industry
14
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
0/1 All-cause mortality
de la Maza et al., 1995 (30) Chile
Decompensated ambulatoryalcoholic cirrhotics
48.7±8.3
(vitamin E)
50.9±9.9
(placebo)
1year Pill counts500 mg vitamin E
daily 37/37 5/4 All-cause mortality Industry
Hamdy et al., 1995 (31)
Europe (Belgium,
France, Netherlands,
UK)
Mild to moderate chronic renal
failure18-81 2 y
Serum calcium levels
were measured
0.25 g alfacalcidol
(vitamin D) every other daywith dose
escalation ifneeded to 1g to maintain serum
calcium concentrations at
upper limit.
89/87 4/1 All-cause mortality N/A
Ooms et al., 1995 (32) Netherlands Healthy >70 2 y
Questionnaire, by pill
counting, and by
measuring serum 250HD
levels
400 IU of vitamin D3/d
177/171 F 11/21 All-cause mortality Agency
15
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
Pike et al.,1995 (33) Canada
Healthy, non-institutionalized
elderly61-79
1 yearBottle checks
and interviews
Multivitamin 24/23 1/0 All-cause mortality Industry
Steiner et al., 1995 (34) USA
Transient ischemic attacks, minor strokes or
residual neurologic
deficits
71.1 2 y n/a α-tocopherol400 IU daily 52/48 3/6 Stroke Industry
Greenberg et al., 1996 – SCPS (35)
USA
Biopsy-proved basal cell or squamous
cell skin cancer patients
63.2 8.2 y
Patients filled questionnaires
about compliance in
taking the capsules
50 mg β-carotene/d 913/892
68/59 CVD mortality
Agency
146/139 All-cause mortality
Hennekens et al., 1996-PHS
(36)USA Healthy 40-84 12 y
Plasma beta carotene
concentrations in blood were
measuredin three
geographic areas.
50 mg β-carotene on alternate days
11036/11035 M
967/972 Total CVD
Agency &
Industry
468/489 MI
367/382 Stroke
338/313 CVD mortality
979/968 All-cause mortality
Lips et al., 1996 (37) Netherlands Healthy 80 4 y
Tablet count, questionnaire,
and serum levels.
400 IU vitamin D3/d
1291/1287 223/251 All-cause mortality
Agency &
Industry
Omenn et al., USA Smokers, former 45-74 4 y Weighing of 15mg beta- 9420/8894 544/424 All-cause Agency
16
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
1996 – CARET (38)
smokers and workers exposed
to asbestos
returned bottles/self-
report
carotene/25000IU retinol mortality
Stephens et al., 1996 – CHAOS
(39)
United Kingdom
Coronary atherosclerosis 62 510 d Timing of
refill request400 -800 IU vitamin E/d 1035/967
32/54 MI
Agency
27/23 CVD mortality
18/13 MI mortality
1/1 Stroke mortality
36/26 All-cause mortality
Dawson-Hughes et al.,
1997 (40)USA Healthy ≥65 3 y Pill counts
500 mg of calcium plus 700 IU of
vitamin D3
(cholecalciferol)per day
187/202 2/2 All-cause mortality
Agency &
Industry
Girodon et al., 1997 (41) France Elderly,
institutionalized 83.75 2 y Given by nurse
Antioxidants: One of 3
preparations:A: 20mg zinc
sulfate + 100µg selenite
B: 120mg ascorbic acid + 6mg β-
carotene + 15mg α-tocopherolC: A + B
21/20 7/7 All-cause mortality Industry
Moon et al., 1997 – SKICAP
AK (42)
United States
History of at least 10 actinic
keratosis & at most 2 squamous cell carcinoma or
basal cell carcinoma skin
cancers
Median: 63 3 y Capsule count 25, 000 IU Retinol 1157/1140 24/24
All-cause mortality
Agency &
Industry
Sano et al., 1997 – ADCS 1 USA Alzheimer’s
disease 73 2 y Serum levels 2000 IU of vitamin E/d 85/84 12/12 All-cause
mortality Agency
17
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
(43)Sato et al., 1997
(44) Japan Hemiplegia after stroke
Mean: 68 6 months N/A Vitamin D
1 μg/d 1α(OH)D 45/39 1/1 All-cause mortality N/A
Baeksgaard et al., 1998 (45) Denmark Healthy
(postmenopausal)
58-67Mean: 65.7
2 yNo formal
assessment of compliance
Calcium-Vitamin D:
1000 mg of calcium carbonate with 14 µg (560
IU) of cholecalciferol
80/80 F 0/1 F All-cause mortality Industry
Shoulson et al., 1998 –
DATATOP(46)
USA and Canada
Early Parkinson’s disease 62 8.2 y N/A
2000IU of vitamin E/d with or without 10mg/d of deprenyl
399/401 73/64 All-cause mortality
Agency &
Industry
Virtamo et al., 1998 – ATBC
(47)Finland Smokers 50-69 6.1 y Capsule count 50 mg (~75
IU)Vitamin E 6820/6849 M
519/534 M Total CHD
Agency307/296 MI
212/238 CHD mortality
Baron et al., 1999 – CPPS
(48)USA History of large
bowel adenomas 61 8 y Tablet count3g (1200mg of
elementary calcium)/d
464/466
50/46 Total CHD
Agency12/11 Stroke
25/22 All-cause mortality
Hoffman et al., 1999 (49) USA Coronary disease
patients
35-80Mean: 67.7
6 months Pill counts
Vitamin E:400 mg/day of D-
α-tocopherol acetate
27/12 M 1/0 M All-cause mortality
Agency &
Industry
Girodon et al., 1999 –
MIN.VIT.AOX (50)
France Institutionalized elderly patients
83.9 2 y Pill count and micronutrient
blood concentration
measured
Antioxidants:1) 20mg Zinc + 100µg Selenium
2) 120mg ascorbic acid + 6 mg β-
carotene+ 15mg α-tocopherol
3) 20mg Zinc + 100µg Selenium +
181/182 55/51 All-cause mortality
Industry
18
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
120mg ascorbic acid + 6 mg β-
carotene+ 15mg α-tocopherol4) Placebo
GISSI – Prevenzione Investigators
1999 (51)
ItalyRecent
MI(<3 months)
60 3.5 y Supplement refills
300mg vitamin E (synthetic) or 400
IU vitamin E (natural)
5660/5664
571/584 Total CVD
Agency &
industry
83/95 Stroke
310/329 CVD mortality
247/273 CHD mortality
228/251 MI mortality
488/529 All-cause mortality
Green et al., 1999 – NSCPT
(52)Australia
Healthy and patients who
previously had skin cancer
20-69 4.5 y
Tablet counts and
concentration of
betacarotene in the skin
30 mg/d of beta-carotene 801/820
6/12 CVD mortality
Agency and
Industry11/21 All-cause mortality
Komulainen et al., 1999 –
OSTPRE (53)Finland
Postmenopausal with 6–24
months elapsed since last
menstruation
47-56 5 y Participant reporting
300IU/d, and 100IU/d of vitamin D during the fifth
year with or without hormone
replacement therapy
112/115 F
1/0 MI
Industry
0/1 All-cause mortality
93mg Calcium and 300 IU vitamin D (cholecalciferol)/d
116/115 F
1/1 MI
2/1 Stroke
19
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
1/0 MI mortality
1/1 All-cause mortality
Krieg et al., 1999 (54) Switzerland Institutionalized
elderly 84.5 2 yTaken with a
nurse supervision
880 IU of vitamin D3 with 1000 mg of
calcium124/124 21/26 All-cause
mortality N/A
Sato et al.,1999 (55) Japan Parkinson’s
disease
65-88Mean: 70.6
18 months N/A
Vitamin D:1 µg 1α(OH)D3
daily43/43 1/0 All-cause
mortality N/A
Boaz et al., 2000 – SPACE
(56)Israel
Patients with stable
hemodialysis with documented history of CVD
64.6 1.4 y
Serum vitamin E
levels were monitored
800 IU of vitamin E daily (natural) 97/99
15/33 Total CVD
Agency
5/17 MI5/6 Stroke
9/15 CVD mortality
2/8 MI mortality
31/29 All-cause mortality
Correa et al., 2000 (57) Colombia
Subjects with confirmed histologic
diagnoses ofmultifocal
nonmetaplastic atrophy and/or
intestinal metaplasia
29-69 72 months Pill counts
2g Vitamin C 130/117 2/0
All-cause mortality Agency
30 mg Beta-Carotene 117/117 2/0
Antioxidants (Vit C and Beta-
carotene)121/117 2/0
Frazao et al., 2000 (58) USA
Hemodialysis patients with
hyperparathyroidism
52 24 weeksAdministered
with hemodialysis
Vitamin D:10µg 1α-
hydroxyvitamin D2
with dosage adjustments to
maintain plasma iPTH levels
71/67
1/2 CHD mortality Agency
1/2 All-cause mortality
20
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
Lee et al., 1999 – WHS (59) USA Healthy 45 Median
2.1 yFollow-up
questionnaires
50 mg of beta-carotene
given on alternatedays.
19939/19937 74/65 Total CVD
Agency &
Industry
Leppälä et al., 2000 – ATBC
(60)Finland Male smokers 50-69 6 y Tablet counts
Antioxidants:50mg/day Vitamin E and 20mg/day
beta-carotene
7118/7153 M
258/252 Stroke
Agency
46/34 Stroke mortality
20mg β-carotene14246/14273
M
554/503 Stroke
82/78 Stroke mortality
Jacobson et al., 2000 (61) USA Heavy smokers ≥18 6 months Bottle checks
Vitamin D:150,000 IU oral cholecalciferol every 3 months
353/333 F 2/0 F All-cause mortality
Agency &
Industry
44
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
Lehouck et al., 2012 (174) Belgium Hospitalized
COPD patients 68 1 y Oral syringe dispensers
100,000 IU vitamin D/4 weeks 91/91 9/6 All-cause
mortality Agency
Ma et al., 2012 – SIT (175) China
Seropositive for Helicobacte
pylori35-64 7.3 y
Tablet counts and serum
measurements
Antioxidants:500mg Vitamin C
+ 200IU Vitamin E + 15mg β-carotene + 75µg selenium
(β-carotene component was
discontinued after 6 months)
1706/1705
10/14 Stroke mortality
Agency
82/101 All-cause mortality
Punthakee et al., 2012 – TIDE
(176)33 countries
Type 2 Diabetes Mellitus and
other CVD risk factors
≥50 5.5 y N/A 1,000 IU vitamin D/day 607/614
2/3 Total CVD
Industry
1/1 MI1/1 Stroke
0/1 CVD mortality
0/2 All-cause mortality
Sesso et al., 2012 – PHS II
(177)USA Healthy 64.3 11.2 y
Annual mailed
questionnaireMultivitamin 7317/7324 M
876/856 Total CVD
Agency &
industry
317/335 MI
332/311 Stroke
408/421 CVD mortality
27/43 MI mortality
89/76 Stroke mortality
1345/1412 All-cause
45
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
mortality
Delanaye et al., 2013 (178) Belgium Hemodialysed
Patients ≥18 1 y
Supplements were taken
with a nurse ensuring
compliance
Vitamin D:Cholecalciferol
(25000 IU) therapy or placebo every 2
weeks
22/21 6/5 All-cause mortality N/A
Hewitt et al., 2013 (179) Australia
Patients on a thrice-weekly
hemodialysis for ≥ 3 months & with screening
levels of 25(OH)D ≤24ng/ml
(60nmol/L)
≥ 18 ≥6 months
Supplements provided by dialysis staff
to ensure compliance
10 ml of an oral solution of medium chain triglyceride containing 50000
IU of cholecalciferol or placebo. Taken
weekly for the first 8 weeks, followed by monthly doses
for 4 months.
30/30 1/1 All-cause mortality Industry
Lamas et al., 2013 – TACT
(180)
USA & Canada Post MI 65 3 y
Unused pills returned to
site
Multivitamins (28-component mixture)
853/855
94/115 Total CVD
Agency
58/61 MI8/15 Stroke
45/56 CVD mortality
87/93 All-cause mortality
Manning et al., 2013 (181)
New-Zealand
Metabolic syndrome 27-80 1 y Pill counts Vitamin E
(100 IU/day) 36/40 0/1 All-cause mortality
Agency &
industry
Prentice et al., 2013 –WHI CaD (182) USA Postmenopausal 50-79 Mean 7.2
y Semiannual pill counts
1000 mg of calcium + 400 IU Vitamin D3 (daily, given in two equal doses) or placebo.
Concurrent supplementation was permitted.
7718/7584 F(no personal supplement
users)
848/813 Total CVD
Agency
229/211 Total CHD
193/167 MI
184/162 Stroke
331/338 All-cause mortality
46
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
Witham et al., 2013 – VitDISH
(183)
United Kingdom
Isolated systolic hypertension &
baseline 25OHD level < 30 ng/ml
≥ 70 mean:
772 y
Patients were observed ingesting
supplement
Vitamin D:A total of 100 000
U of oral cholecalciferol or placebo every 3
months for 1 year.
80/79
2/2 Total CHD
Agency3/1 Stroke
0/1 All-cause mortality
Witham et al., 2013 (184) UK
History of myocardial infarction
66 (mean) 6 months n/a
Vitamin D3300,000
units39/36
5/4 Total CVD Agency
1/0 All-cause mortality
HPS2-THRIVE Collaborative Group 2014
(185)
United Kingdom,
China, Scandinavia
History of MI, CVD, peripheral arterial disease,
DM, & symptomatic
coronary disease
50-80 Median:3.9 y
Self-reported consumption & blood tests
2 g vitamin B3 (niacin)/d 12838/12835
1696/1758 Total CVD
Agency &
industry
668/694 Total CHD
402/431 MI498/499 Stroke
302/291 CHD mortality
798/732 All-cause mortality
Wang et al., 2014 – OPERA
(186)Hong Kong
Stage 3-5 CKD with LV
hypertrophy18-75 52 weeks
Recorded during visits at weeks 6,
12, 24, 36, 48, and 52.
Vitamin D: Oral paricalcitrol (1ug)
once daily30/30
0/6 Total CVD
Agency0/2 MI
0/2 Stroke
Van Wijngaarden et al., 2014 – B-PROOF (187)
Netherlands
Living independently with elevated homocysteine (12-50µmol)
≥65 2 yTablet counts,
periodic phone calls
B-complex (500µg vitamin B12 plus 400µg folic acid)
or placebo
1516/1511 37/42 All-cause mortality Agency
Wang et al., 2014 – PHS II
(188)USA Healthy
≥50(mean: 64.3)
Mean: 10.3 y
Self-reported (questionnaire
)
Vitamin E400 IU synthetic α-
tocopherol on alternate days
3659/3653M 397/406 M All-cause mortality
Agency &
industryVitamin C 3673/3653 M 414/406 MAntioxidants
(Vitamin E and C) 3656/3653 M 440/406 M
47
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
Huo et al., 2015 – CSPPT (189) China Hypertension 60 4.5 y Participant
reporting0.8 mg /d of folic
acid 10348/10354
324/405 Total CVD
Agency &
industry
25/24 MI282/355 Stroke
43/43 CVD mortality
5/4 MI mortality
18/10 Stroke mortality
302/320 All-cause mortality
Ponikowski et al., 2015 –
CONFIRM-HF (190)
Austria, Italy,
Poland, Portugal, Russia, Spain,
Sweden, UK, Ukraine
Heart failure 68 52 weeksIntervention delivered by
injection
i.v.iron, as ferric
carboxymaltose150/151
11/12CVD
mortality
Industry
12/14All-cause mortality
Wang et al., 2015 (191) China Healthy 60–74 1 y
Number of capsules
consumedwas recorded
to monitor compliance
400 µg folic acid, 2 mg vitamin B6, 10
µg vitamin B12195/195 4/6
All-cause mortality Agency
Van Dijk et al., 2015 –
B-PROOF (192)Netherlands
Living independently with elevated homocysteine (12-50µmol)
≥65 2 yTablet counts,
periodic phone calls
B-complex (500µg vitamin B12 plus 400µg folic acid)
or placebo
1516/1511181/170
Total CVD
Agency46/60 Stroke
45/43 MI
Gupta et al., 2016 (193) India
Ischaemic stroke survivors with
low serum vitamin D
(<75nmol/L)
60±11.3 6 months
Administered by healthcare
providers
Calcium + Vitamin D 25/28 4/11
All-cause mortality
None
Lappe et al., 2017 (194) USA Healthy
(postmenopausal) ≥556 y
(supplementation
Pill counts and returned
bottles
Calcium + Vitamin D (2000 IU/d
of vitamin D3 and 1156/1147 W 7/9 W All-cause
mortality
Agency &
industry
48
Study (reference) Country Health status
Age (years)
Duration (mean, median
or range)
Supplement intake
assessment
Supplement exposure (median or range, units)
Participants(intervention
/control)
Incident cases
(intervention/control)
Outcome Funding source
during 4 y) weighing 1500mg/d of
calcium)CHD, Coronary heart disease; CVD, cardiovascular disease; N/A, not available; T2DM, type two diabetes mellitus; CDPRG, The Coronary Drug Project Research Group; STOCKHOLM, Stockholm lschaemic Heart Disease Secondary Prevention Study; FATS, Familial Atherosclerosis Treatment Study;UCSF-SCOR, University of California, San Francisco, Arteriosclerosis Specialized Center of Research; CHAOS, Cambridge Heart Antioxidant Study; ADCS, Alzheimer’s Disease Cooperative Study; DATATOP, Deprenyl And Tocopherol AntioxidativeTherapy Of Parkinsonism; ATBC , The Alpha- Tocopherol, Beta-Carotene Cancer Prevention Study; CPPS, Calcium Polyp Prevention Study;OSTPRE, Kuopio Osteoporosis Study; SPACE, Secondary Prevention with Antioxidants of Cardiovascular disease in Endstage renal disease; HOPE, Heart Outcomes Prevention Evaluation; AREDS, Age-Related Eye Disease Study; HATS, The HDL-Atherosclerosis Treatment Study; PPP, Primary Prevention Project; VEAPS, The Vitamin E Atherosclerosis Prevention Study; The Swiss Heart Study, The Swiss Heart Study: A Randomized Control Trial; REACT, The Roche European American Cataract Trial; WAVE, Women’s Angiographic Vitamin and Estrogen; FITNESS, The Frailty Interventions Trial iN Elderly SubjectS; NoNOF, The Notthinghan Neck Of Femur Study; VECAT, Vitamin E, Cataract and Age-related Maculopathy Trial;ARBITER2, Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol; RECORD, Randomized Evaluation of Calcium Or vitamin D; VIP, Visual ImPairment Trial; WHI, Women’s Health Study; NORVIT, The NORwegian Viamin Trial; HOPE, Heart Outcomes Prevention Evaluation; ASFAST, Atherosclerosis and Folic Acid Supplementation Trial; MAVET, The Melbourne Atherosclerosis Vitamin E Trial; FACIT, Folic Acid and Carotid Intima-media Thickness; HOST, Homocysteinemia in Kidney and End Stage Renal Disease; ASCENT, AIPC (Androgenindependent Prostate Cancer) Study of Calcitriol Enhancing Taxotere; WAFACS, Women’s Antioxidant and Folic Acid Cardiovascular Study; CAIFOS, Calcium Intake Fracture Outcome Study UK CAP, The United Kingdom Colorectal Adenoma Prevention Trial; FERRIC- HF, Ferric Iron Sucrose in Heart Failure; PHS, Physicians Health Study; FAIR-HF, Ferinject Assessment in Patients with Iron Deficiency and Chronic Heart Failure Ttrial; BVAIT, The B-Vitamin Atherosclerosis Intervention Trial; WACS, The Women’s Antioxidant Cardiovascular Study; NHS/HPFS, Nurse’s Health Study/ Health Professionals Follow-up Study; SELECT, The Selenium and Vitamin E Cancer Prevention Trial; ARBITER, Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol; SEARCH, Study of the Effectiveness of Additional Reductions in Cholesterol and Homocysteine; SU.VI.MAX, The Supplémentation en Vitamines et Minéraux AntioXydants study; VITAL, VITamins And Lifestyle cohort CAIFOS, Calcium Intake Fracture Outcome Study; WAFACS, Women’s Antioxidant and Folic Acid Cardiovascular Study; UK CAP, The United Kingdom Colorectal Adenoma Prevention Trial; SELECT, The Selenium and Vitamin E Cancer Prevention
.
49
Supplementary Table 4. GRADE assessment for vitamin D
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Vitamin D and total CVD risk (random effects model) - RCTs
6 randomised trials
not serious not serious not serious a serious b none c RR 0.95(0.86 to 1.05)
8 fewer per 1,000
(from 8 more to 23
fewer)
⨁⨁⨁◯MODERATE
Vitamin D and total CHD risk (random effects model) - RCTs
3 randomised trials
not serious not serious not serious d very serious e none c RR 0.97(0.22 to 4.22)
0 fewer per 1,000
(from 11 fewer to 45
more)
⨁⨁◯◯LOW
Vitamin D and MI risk (random effects model) - RCTs
12 randomised trials
not serious not serious not serious f serious g none RR 0.95(0.83 to 1.10)
3 fewer per 1,000
(from 6 more to 10
fewer)
⨁⨁⨁◯MODERATE
Vitamin D and stroke risk (random effects model) - RCTs
11 randomised trials
not serious not serious not serious h serious i none RR 1.12(0.94 to 1.34)
5 more per 1,000
(from 2 fewer to 14
more)
⨁⨁⨁◯MODERATE
Vitamin D and CVD mortality risk - (random effects model) - RCTs
2 randomised trials
not serious not serious not serious j serious k none c RR 0.86(0.66 to 1.10)
8 fewer per 1,000
(from 6 more to 21
fewer)
⨁⨁⨁◯MODERATE
Vitamin D and CHD mortality risk - (random effects model) - RCTs
2 randomised trials
not serious l not serious serious m very serious n none c RR 0.41(0.06 to 2.72)
16 fewer per 1,000(from 26
fewer to 47 more)
⨁◯◯◯VERY LOW
Vitamin D and stroke mortality risk - (random effects model) - RCTs
2 randomised trials
not serious o not serious not serious p serious q none c RR 1.13(0.68 to 1.87)
3 more per 1,000
(from 6 fewer to 18
more)
⨁⨁⨁◯MODERATE
Vitamin D and all-cause mortality risk (random effects model) - RCTs
43 randomised trials
not serious not serious not serious r not serious none RR 0.99(0.95 to 1.03)
2 fewer per 1,000
(from 5 more to 8
fewer)
⨁⨁⨁⨁HIGH
50
CI: Confidence interval; RR: Risk ratio
Explanationsa. No serious indirectness for vitamin D supplementation and total CVD risk (RCTs). Five of six studies were primary prevention and one study was secondary prevention. b. Serious imprecision for vitamin D supplementation and total CVD risk (RCTs), as the 95% CI (RR, 0.86-1.05) overlaps with the minimally important difference for clinical benefit (RR<0.95). c. Publication bias was not assessed since there were <10 studies. d. No serious indirectness for vitamin D supplementation and total CHD risk (RCTs). All 3 studies were primary prevention. e. Very serious imprecision for vitamin D supplementation and total CHD risk (RCTs) as the 95% CI (RR, 0.22-4.22) include both clinically important benefit (RR<0.95) and harm (RR>1.05). f. No serious indirectness for vitamin D supplementation and MI risk (RCTs). All 12 studies were primary prevention. g. Serious imprecision for vitamin D supplementation and MI risk (RCTs) as the 95% CI (RR, 0.83-1.10) include both clinically important benefit (RR<0.95) and harm (RR>1.05). h. No serious indirectness for vitamin D supplementation and stroke risk (RCTs). All 11 included studies were primary prevention. i. Serious imprecision for vitamin D supplementation and stroke risk (RCTs) as the 95% CI (RR, 0.94-1.34) include both clinically important benefit (RR<0.95) and harm (RR>1.05). j. No serious indirectness for vitamin D supplementation and CVD mortality risk (RCTs). Both studies were primary prevention. k. Serious imprecision for vitamin D supplementation and CVD mortality risk (RCTs) as the 95% CI (RR, 0.66-1.10) include both clinically important benefit (RR<0.95) and harm (RR>1.05). l. No serious risk of bias for vitamin D supplementation and total CHD mortality risk (RCTs) despite one study being high risk and the other being unclear for random sequence allocation. m. Serious indirectness for vitamin D supplementation and CHD mortality risk (RCTs) as 61% of the study population were hemodialysis patients. n. Very serious imprecision for vitamin D supplementation and CHD mortality risk (RCTs) as the 95% CI (RR, 0.06-2.72) include both clinically important benefit (RR<0.95) and harm (RR>1.05). o. No serious risk of bias for vitamin D supplementation and stroke mortality risk (RCTs) despite one study being high risk and the other being unclear for random sequence allocation. p. No serious indirectness for vitamin D supplementation and stroke mortality risk (RCTs). Both studies were primary prevention. q. Serious imprecision for vitamin D supplementation and stroke mortality risk (RCTs) as the 95% CI (RR, 0.68-1.87) include both clinically important benefit (RR<0.95) and harm (RR>1.05). r. No serious indirectness for vitamin D supplementation and all-cause mortality risk (RCTs). All 44 studies were primary prevention.
Supplementary Table 5. GRADE assessment for vitamin A
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Vitamin A and all-cause mortality (random effects model) - RCTs
1 randomised trials
not serious not serious a serious c serious d none b RR 0.99(0.56 to 1.72)
0 fewer per 1,000
(from 9 fewer to 15
more)
⨁⨁◯◯LOW
CI: Confidence interval; RR: Risk ratio
Explanationsa. Unable to assess inconsistency as only one study was included. b. Publication bias was not assessed as there were < 10 studies. c. Serious indirectness for vitamin A supplementation and all-cause mortality risk (RCTs), as the only included study was conducted among a population with actinic keratoses and squamous cell carcinoma or basal cell carcinoma skin cancers. d. Serious imprecision for vitamin A and all-cause mortality risk (RCTs), as the 95% CI (RR, 0.56-1.72) include both clinically important benefit (RR<0.95) and harm (RR>1.05).
Supplementary Table 6. GRADE assessment for beta-carotene
Certainty assessment Effect Certainty
№ of studies
Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Beta-carotene and total CVD risk (random effects model) - RCTs
3 randomised trials
not serious not serious not serious a serious b none c RR 1.03(0.96 to 1.10)
1 more per 1,000
(from 2 fewer to 4
more)
⨁⨁⨁◯MODERATE
Beta-carotene and total CHD risk (random effects model) - RCTs
51
Certainty assessment Effect Certainty
№ of studies
Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
2 randomised trials
not serious not serious serious e serious f none c RR 1.02(0.94 to 1.10)
2 more per 1,000
(from 6 fewer to 9
more)
⨁⨁◯◯LOW
Beta-carotene and MI risk (random effects model) - RCTs
3 randomised trials
not serious not serious not serious g serious h none c RR 0.99(0.91 to 1.09)
0 fewer per 1,000
(from 4 fewer to 4
more)
⨁⨁⨁◯MODERATE
Beta-carotene and stroke risk (random effects model) - RCTs
3 randomised trials
not serious not serious not serious i serious j none c RR 1.06(0.95 to 1.19)
2 more per 1,000
(from 2 fewer to 7
more)
⨁⨁⨁◯MODERATE
Beta-carotene and CVD mortality risk (random effects model) - RCTs
4 randomised trials
not serious not serious not serious k serious l none c RR 1.10(0.98 to 1.23)
3 more per 1,000
(from 1 fewer to 8
more)
⨁⨁⨁◯MODERATE
Beta-carotene and CHD mortality risk (random effects model) - RCTs
1 randomised trials
not serious not serious d serious m serious n none c RR 0.99(0.83 to 1.18)
0 fewer per 1,000
(from 6 fewer to 6
more)
⨁⨁◯◯LOW
Beta-carotene and MI mortality risk (random effects model) - RCTs
1 randomised trials
not serious not serious d serious o not serious none c RR 0.42(0.20 to 0.87)
3 fewer per 1,000
(from 1 fewer to 5
fewer)
⨁⨁⨁◯MODERATE
Beta-carotene and stroke mortality risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious p serious q none c RR 1.33(0.73 to 2.44)
2 more per 1,000
(from 1 fewer to 7
more)
⨁⨁◯◯LOW
Beta-carotene and all-cause mortality risk (random effects model) - RCTs
6 randomised trials
not serious not serious not serious r serious s none c RR 1.03(0.96 to 1.11)
3 more per 1,000
(from 4 fewer to 11
more)
⨁⨁⨁◯MODERATE
CI: Confidence interval; RR: Risk ratio
Explanationsa. No serious indirectness for beta-carotene supplementation and total CVD risk (RCTs). Majority of studies were primary prevention with the exception of one study (Cook et al. 2007) which included individuals with a history of CVD or at least three cardiac risk factors. b. Serious imprecision for beta-carotene supplementation and total CVD risk (RCTs), as the 95% CI (RR, 0.96-1.10) overlaps with the minimally important difference for clinical harm (RR>1.05). c. Publication bias was not assessed since there were <10 studies.
52
d. Not able to assess consistency as only one study was included. e. Serious indirectness for beta-carotene supplementation and CHD risk (RCTs), as the 2 included studies were conducted either in middle-aged, Finnish male smokers or middle-aged females with a history of CVD, or at least three cardiac risk factors. f. Serious imprecision for beta-carotene supplementation and total CHD risk (RCTs), as the 95% CI (RR, 0.94, 1.10) include both clinically important benefit (RR<0.95) and harm (RR>1.05). g. No serious indirectness for beta-carotene supplementation and MI risk (RCTs). Majority of studies were primary prevention with the exception of one study (Cook et al. 2007) which included individuals with history of CVD or at least three cardiac risk factors. h. Serious imprecision for beta-carotene supplementation and MI risk (RCTs), as the 95% CI (RR, 0.91, 1.09) include both clinically important benefit (RR<0.95) and harm (RR>1.05). i. No serious indirectness for beta-carotene supplementation and stroke risk (RCTs). Majority of studies were primary prevention with the exception of one study (Cook et al., 2007) which included individuals with history of CVD or at least three cardiac risk factors. j. Serious imprecision for beta-carotene supplementation and stroke risk (RCTs), as the 95% CI (RR, 0.95-1.19) overlaps with the minimally important difference for clinical harm (RR>1.05). k. No serious indirectness for beta-carotene and CVD mortality risk (RCTs). Majority of the studies were primary prevention, with the exception of one study (Cook et al., 2007) which included individuals with history of CVD, or at least three cardiac risk factors. l. Serious imprecision for beta-carotene and CVD mortality risk (RCTs), as the 95% CI (RR, 0.98-1.23) overlaps with the minimally important difference for clinical harm (RR>1.05). m. Serious indirectness for beta-carotene supplementation and CHD mortality risk (RCTs), as the 1 included study was conducted in middle-aged, Finnish male smokers. n. Serious imprecision for beta-carotene supplementation and CHD mortality risk (RCTs), as the 95% CI (RR, 0.83, 1.18) include both clinically important benefit (RR<0.95) and harm (RR>1.05). o. Serious indirectness for beta-carotene supplementation and MI mortality risk (RCTs), as the one included study was conducted in a female population at high risk for CVD. This was a secondary prevention study. p. Serious indirectness for beta-carotene supplementation and stroke mortality risk (RCTs), as the 2 included studies were conducted in middle-aged, Finnish male smokers or middle-aged females with a history of CVD or at least 3 cardiac risk factors. q. Serious imprecision for beta-carotene supplementation and stroke mortality risk (RCTs), as the 95% CI (RR 0.73, 2.44), include both clinically important benefit (RR<0.95) and harm (RR>1.05). r. No serious indirectness for beta-carotene supplementation and all-cause mortality risk (RCTs). Majority of the studies were primary prevention, with the exception of one study (Cook et al., 2007) which included individuals with of CVD, or at least three cardiac risk factors. s. Serious imprecision for beta-carotene supplementation and all-cause mortality risk (RCTs), as the 95% CI (RR, 0.96-1.11) overlaps with the minimally important difference for clinical harm (RR>1.05).
Supplementary Table 7. GRADE assessment for antioxidants
Certainty assessment Effect Certainty
№ of studies
Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Antioxiants and total CVD (random effects model) - RCTs
7 randomised trials
not serious not serious not serious a not serious none b RR 0.99(0.95 to 1.04)
1 fewer per 1,000
(from 5 more to 6
fewer)
⨁⨁⨁⨁HIGH
Antioxidants and total CHD (random effects model) - RCTs
1 randomised trials
not serious not serious c serious d serious e none b RR 0.97(0.86 to 1.09)
2 fewer per 1,000
(from 7 more to 11
fewer)
⨁⨁◯◯LOW
Antioxidants and MI (random effects model) - RCTs
6 randomised trials
not serious not serious not serious f serious g none b RR 0.98(0.90 to 1.08)
1 fewer per 1,000
(from 3 more to 4
fewer)
⨁⨁⨁◯MODERATE
Antioxidants and stroke (random effects model) - RCTs
7 randomised trials
not serious not serious not serious h serious i none b RR 1.00(0.92 to 1.09)
0 fewer per 1,000
(from 3 fewer to 3
more)
⨁⨁⨁◯MODERATE
Antioxidants and CVD mortality (random effects model) - RCTs
7 randomised trials
not serious not serious not serious j serious k none b RR 1.02(0.94 to 1.10)
1 more per 1,000
(from 3 fewer to 5
more)
⨁⨁⨁◯MODERATE
Antioxidants and CHD mortality (random effects model) - RCTs
53
Certainty assessment Effect Certainty
№ of studies
Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
2 randomised trials
not serious not serious serious l serious m none b RR 1.02(0.93 to 1.13)
1 more per 1,000
(from 4 fewer to 7
more)
⨁⨁◯◯LOW
Antioxidants and MI mortality (random effects model) - RCTs
3 randomised trials
not serious not serious serious n very serious o none b RR 1.51(0.39 to 5.93)
3 more per 1,000
(from 4 fewer to 31
more)
⨁◯◯◯VERY LOW
Antioxidants and stroke mortality (random effects model) - RCTs
5 randomised trials
not serious not serious not serious p serious q none b RR 1.10(0.87 to 1.39)
1 more per 1,000
(from 1 fewer to 2
more)
⨁⨁⨁◯MODERATE
Antioxidants and all-cause mortality (random effects model) - RCTs
21 randomised trials
not serious not serious not serious r serious s none RR 1.06(1.00 to 1.12)
5 more per 1,000
(from 0 fewer to 9
more)
⨁⨁⨁◯MODERATE
CI: Confidence interval; RR: Risk ratio
Explanationsa. No serious indirectness for antioxidant supplementation and total CVD risk (RCTs). Two studies were secondary prevention, four studies were primary prevention and one study was a combination of both. b. Publication bias was not assessed since there were < 10 studies. c. Unable to assess inconsistency as only one study was included. d. Serious indirectness for antioxidants supplementation and total CHD risk (RCTs) as there was only one study included (Tornwall et al., 2004) and it was conducted among Finnish male smokers. e. Serious imprecision for antioxidant supplementation and total CHD risk (RCTs), as the 95% CI (RR, 0.86-1.09) include both clinically important benefit (RR<0.95) and harm (RR>1.05). f. No serious indirectness for antioxidant supplementation and MI risk (RCTs). Three studies were for secondary prevention, two studies were primary prevention and one study was a combination of both. g. Serious imprecision for antioxidant supplementation and MI risk (RCTs), as the 95% CI (RR, 0.90-1.08) include both clinically important benefit (RR<0.95) and harm (RR>1.05). h. No serious indirectness for antioxidant supplementation and stroke risk (RCTs). Three studies were secondary prevention, three studies were primary prevention and one study was a combination of both. i. Serious imprecision for antioxidant supplementation and stroke risk (RCTs), as the 95% CI (RR, 0.92-1.09) include both clinically important benefit (RR<0.95) and harm (RR>1.05). j. No serious indirectness for antioxidant supplementation and CVD mortality risk (RCTs). Three studies were for secondary prevention, three studies were primary prevention and one study was a combination of both. k. Serious imprecision for antioxidants supplementation and CVD mortality risk (RCTs), as the 95% CI (RR, 0.94-1.10) include both clinically important benefit( (RR<0.95) harm (RR>1.05). l. Serious indirectness for antioxidants supplementation and CHD mortality risk (RCTs) as there was only two studies included and one of them (Tornwall et al., 2004) was conducted among Finnish male smokers only, which accounted for ~30% of the population. m. Serious imprecision for antioxidant supplementation and CHD mortality risk (RCTs), as the 95% CI (RR, 0.93-1.13) include both clinically important benefit (RR<0.95) and harm (RR>1.05). n. Serious indirectness for antioxidant supplementation MI mortality risk (RCTs), as the included studies were in specific populations: early age-related cataract (Chylack et al., 2002), smokers >10 cigarettes/day (Mooney et al., 2005) and peripheral artery disease (CLIPS group 2007). o. Very serious imprecision for antioxidant supplementation and MI mortality risk (RCTs), as the 95% CI (RR, 0.39-5.93) include both appreciable clinically important benefit (RR<0.95) and harm (RR>1.05). p. No serious indirectness for antioxidant supplementation and stroke mortality risk (RCTs). One study was secondary prevention, 3 studies were primary prevention and 1 study was a combination of both. q. Serious imprecision for antioxidant supplementation and stroke mortality risk (RCTs), as the 95% CI (RR, 0.87-1.39) include both clinically important benefit (RR<0.95) and harm (RR>1.05). r. No serious indirectness for antioxidant supplementation and all-cause mortality (RCTs). Sixteen studies were primary prevention, three were secondary prevention and two were a combination of both. s. Serious imprecision for antioxidants supplementation and all-cause mortality risk (RCTs)-random effects model, as the 95% CI (RR, 1.00-1.12) overlaps with the minimally important difference for clinical harm (RR>1.05).
Supplementary Table 8. GRADE assessment for vitamin E
54
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Vitamin E and total CVD risk (random effects model) - RCTs
10 randomised trials
not serious not serious a not serious b serious c publication bias strongly suspected d
RR 0.96(0.89 to 1.03)
3 fewer per 1,000
(from 2 more to 9
fewer)
⨁⨁◯◯LOW
Vitamin E and total CHD risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious e serious f none g RR 0.97(0.90 to 1.06)
3 fewer per 1,000
(from 6 more to 10
fewer)
⨁⨁◯◯LOW
Vitamin E and MI risk (random effects model) - RCTs
10 randomised trials
not serious serious h not serious i serious j publication bias strongly suspected d
RR 0.87(0.75 to 1.01)
5 fewer per 1,000
(from 0 fewer to 9
fewer)
⨁◯◯◯VERY LOW
Vitamin E and stroke risk (random effects model) - RCTs
11 randomised trials
not serious not serious not serious k serious l none g RR 0.98(0.88 to 1.08)
0 fewer per 1,000
(from 2 fewer to 2
more)
⨁⨁◯◯MODERATE
Vitamin E and CVD mortality risk (random effects model) - RCTs
11 randomised trials
not serious not serious not serious m serious n none g RR 0.95(0.89 to 1.02)
1 fewer per 1,000
(from 1 more to 3
fewer)
⨁⨁⨁◯MODERATE
Vitamin E and CHD Mortality risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious o serious p none g RR 0.90(0.80 to 1.02)
4 fewer per 1,000
(from 1 more to 8
fewer)
⨁⨁◯◯LOW
Vitamin E and MI mortality risk (random effects model) - RCTs
7 randomised trials
not serious not serious not serious q serious r none g RR 0.91(0.78 to 1.07)
1 fewer per 1,000
(from 1 more to 2
fewer)
⨁⨁⨁◯MODERATE
Vitamin E and stroke mortality risk (random effects model) - RCTs
5 randomised trials
not serious not serious not serious s serious t none g RR 0.97(0.66 to 1.43)
0 fewer per 1,000
(from 0 fewer to 1
more)
⨁⨁⨁◯MODERATE
Vitamin E and all-cause mortality risk (random effects model) - RCTs
32 randomised trials
not serious not serious not serious u not serious none RR 1.00(0.97 to 1.04)
0 fewer per 1,000
(from 2 fewer to 3
more)
⨁⨁⨁⨁HIGH
55
CI: Confidence interval; RR: Risk ratio
Explanations
a. Although there was evidence of serious inconsistency for vitamin E and total CVD risk (RCTs) (I2=50%, p=0.04), independent removal of one trial (Boaz et al.) explained half of the heterogeneity (I2=24%, p=0.24). b. No serious indirectness for vitamin E and total CVD risk (RCTs). Two studies were secondary prevention, five studies were primary prevention and three were a combination of both. c. Serious imprecision for vitamin E supplementation and total CVD risk (RCTs), as the 95% CI (RR 0.89, 1.03) overlaps with the minimally important difference for clinical benefit (RR<0.95). d. Publication bias strongly detected due to funnel plot asymmetry and Egger’s and Begg’s tests (p<0.05) e. Serious indirectness for vitamin E supplementation and total CHD risk (RCTs), as the two included studies were conducted either in Finnish, middle-aged male smokers or middle-aged women with history of CVD, or at least three cardiac risk factors. f. Serious imprecision for vitamin E supplementation and total CHD risk (RCTs), as the 95% CI (RR, 0.90, 1.06) include both clinically important benefit (RR<0.95) and harm (RR>1.05). g. Publication bias was not assessed since there were <10 studies. h. Serious inconsistency for vitamin E supplementation and MI risk (RCTs), as I2=59%, p=0.01. i. No serious indirectness for vitamin E and total MI risk (RCTs). Two studies were secondary prevention, five studies were primary prevention and three studies were a combination of both. j. Serious imprecision for vitamin E supplementation and MI risk (RCTs), as the 95% CI (RR, 0.75-1.01) overlaps with the minimally important difference for clinical benefit (RR<0.95) k. No serious indirectness for vitamin E and stroke risk (RCTs). Three studies were secondary prevention, five studies were primary prevention and three studies were a combination of both. l. Serious imprecision for vitamin E supplementation and stroke risk (RCTs), as the 95% CI (RR 0.88, 1.08) include both clinically important benefit (RR<0.95) and harm (RR>1.05). m. No serious indirectness for vitamin E supplementation and CVD mortality risk (RCTs). Three studies were secondary prevention, five studies were primary prevention and three studies were a combination of both. n. Serious imprecision for vitamin E supplementation and CVD mortality risk (RCTs), as the 95% CI (RR, 0.89-1.02) overlaps with the minimally important difference for clinical benefit (RR<0.95) o. Serious indirectness for vitamin E supplementation and CHD mortality risk (RCTs), as the two included studies were conducted in Italian, middle-aged individuals with a recent MI or Finnish, middle-aged male smokers. p. Serious imprecision for vitamin E supplementation and CHD mortality risk (RCTs), as the 95% CI (RR, 0.80-1.02) overlaps with the minimally important difference for clinical benefit (RR<0.95) q. No serious indirectness for vitamin E and MI mortality risk (RCTs). Three studies were secondary prevention, three studies were primary prevention and two studies were a combination of both. r. Serious imprecision for vitamin E supplementation and MI mortality risk (RCTs), as the 95% CI (RR, 0.78-1.07) include both clinically important benefit (RR <0.95) and harm (RR>1.05). s. No serious indirectness for vitamin E supplementation and stroke mortality risk (RCTs). One study was secondary prevention, three studies were primary prevention and two studies were a combination of both. t. Serious imprecision for vitamin E supplementation and stroke mortality risk (RCTs), as the 95% CI (RR, 0.66-1.43) include both clinically important benefit (RR <0.95) and harm (RR>1.05). u. No serious indirectness for vitamin E supplementation and all-cause mortality risk (RCTs). Nine studies were secondary prevention and 23 studies were primary prevention.
Supplementary Table 9. GRADE assessment for vitamin C
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Vitamin C and total CVD risk (random effects model) - RCTs
2 randomised trials
not serious not serious not serious a serious b none c RR 0.99(0.90 to 1.10)
1 fewer per 1,000
(from 9 fewer to 9
more)
⨁⨁⨁◯MODERATE
Vitamin C and total CHD risk (random effects model) - RCTs
1 randomised trials
not serious not serious d serious e serious f none c RR 1.04(0.93 to 1.17)
5 more per 1,000
(from 8 fewer to 20
more)
⨁⨁◯◯LOW
Vitamin C and MI risk (random effects model) - RCTs
2 randomised trials
not serious not serious not serious g serious h none c RR 0.96(0.81 to 1.14)
1 fewer per 1,000
(from 5 more to 7
fewer)
⨁⨁⨁◯MODERATE
Vitamin C and stroke risk (random effects model) - RCTs
56
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
2 randomised trials
not serious not serious not serious i serious j none c RR 0.92(0.78 to 1.09)
3 fewer per 1,000
(from 3 more to 8
fewer)
⨁⨁⨁◯MODERATE
Vitamin C and CVD mortality risk (random effects model) - RCTs
2 randomised trials
not serious not serious not serious k serious l none c RR 1.07(0.92 to 1.25)
3 more per 1,000
(from 3 fewer to 10
more)
⨁⨁⨁◯MODERATE
Vitamin C and MI mortality risk (random effects model) - RCTs
1 randomised trials
not serious not serious not serious m serious n none c RR 0.79(0.40 to 1.55)
1 fewer per 1,000
(from 3 fewer to 3
more)
⨁⨁⨁◯MODERATE
VItamin C and stroke mortality risk (random effects model) - RCTs
1 randomised trials
not serious not serious not serious o serious p none c RR 0.83(0.42 to 1.65)
1 fewer per 1,000
(from 3 fewer to 3
more)
⨁⨁⨁◯MODERATE
Vitamin C and all-cause mortality risk (random effects model) - RCTs
4 randomised trials
not serious not serious not serious q serious r none c RR 1.02(0.94 to 1.11)
2 more per 1,000
(from 7 fewer to 12
more)
⨁⨁⨁◯MODERATE
CI: Confidence interval; RR: Risk ratio
Explanationsa. No serious indirectness for vitamin C supplementation and total CVD risk (RCTs). Both studies were a combination of primary and secondary prevention and had large populations, one of men and the other of women. b. Serious imprecision for vitamin C supplementation and total CVD risk (RCTs), as the 95% CI (RR, 0.90-1.10) include both clinically important benefit (RR<0.95) and harm (RR>1.05). c. Publication bias was not assessed as there were < 10 studies. d. Unable to assess inconsistency as only one study was included. e. Serious indirectness for vitamin C supplementation and total CHD risk (RCTs), as the one included study was conducted in women only. f. Serious imprecision for vitamin C supplementation and total CHD risk (RCTs), as the 95% CI (RR, 0.93-1.17) include both clinically important benefit (RR<0.95) and harm (RR>1.05). g. No serious indirectness for vitamin C supplementation and MI risk (RCTs). Both studies were a combination of primary and secondary prevention and had large populations, one of men and the other of women. h. Serious imprecision for vitamin C supplementation and MI risk (RCTs), as the 95% CI (RR, 0.81-1.14) include both clinically important benefit (RR<0.95) and harm (RR>1.05). i. No serious indirectness for vitamin C supplementation and total CVD risk (RCTs). Both studies were a combination of primary and secondary prevention and had large populations, one of men and the other of women. j. Serious imprecision for vitamin C supplementation and stroke risk (RCTs), as the 95% CI (RR, 0.78-1.09) include both clinically important benefit (RR<0.95) and harm (RR>1.05). k. No serious indirectness for vitamin C supplementation and CVD mortality risk (RCTs). Both studies were a combination of primary and secondary prevention and had large populations, one of men and the other of women. l. Serious imprecision for vitamin C supplementation and CVD mortality risk (RCTs), as the 95% CI (RR, 0.92-1.25) include both clinically important benefit (RR<0.95) and harm (RR>1.05). m. No serious indirectness for vitamin C supplementation and MI mortality risk (RCTs). Both studies were a combination of primary and secondary prevention and had large populations, one of men and the other of women. n. Serious imprecision for vitamin C supplementation and MI mortality risk (RCTs), as the 95% CI (RR, 0.40-1.55) includes both clinically important benefit (RR<0.95) and harm (RR>1.05). o. No serious indirectness for vitamin C supplementation and stroke mortality risk (RCTs). Both studies were a combination of primary and secondary prevention and had large populations, one of men and the other of women. p. Serious imprecision for vitamin C supplementation and stroke mortality risk (RCTs), as the 95% CI (RR, 0.42-1.65) include both clinically important benefit (RR<0.95) and harm (RR>1.05). q. No serious indirectness for vitamin C supplementation and all-cause mortality risk (RCTs). Three studies were primary prevention and one was a combination of both primary and secondary prevention. r. Serious imprecision for vitamin C supplementation and all-cause mortality risk (RCTs), as the 95% CI (RR, 0.94-1.11) include both clinically important benefit (RR<0.95) and harm (RR>1.05).
57
Supplementary Table 10. GRADE assessment for selenium
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Selenium and total CVD risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious a serious b none c RR 1.04(0.96 to 1.12)
5 more per 1,000
(from 5 fewer to 15
more)
⨁⨁◯◯LOW
Selenium and total CHD risk (random effects model) - RCTs
1 randomised trials
not serious not serious d serious e serious f none c RR 1.06(0.76 to 1.48)
7 more per 1,000
(from 28 fewer to 57
more)
⨁⨁◯◯LOW
Selenium and MI risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious g serious h none c RR 0.93(0.62 to 1.39)
6 fewer per 1,000
(from 32 fewer to 32
more)
⨁⨁◯◯LOW
Selenium and stroke risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious i serious j none c RR 0.89(0.70 to 1.14)
2 fewer per 1,000
(from 2 more to 4
fewer)
⨁⨁◯◯LOW
Selenium and CVD mortality risk (random effects model) - RCTs
3 randomised trials
not serious not serious serious k serious l none c RR 1.00(0.66 to 1.51)
0 fewer per 1,000
(from 6 fewer to 10
more)
⨁⨁◯◯LOW
Selenium and MI mortality risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious m very serious n none c RR 1.21(0.49 to 2.99)
3 more per 1,000
(from 7 fewer to 27
more)
⨁◯◯◯VERY LOW
Selenium and stroke mortality risk (random effects model) - RCTs
1 randomised trials
not serious not serious d serious o very serious p none c RR 1.13(0.44 to 2.93)
0 fewer per 1,000
(from 1 fewer to 2
more)
⨁◯◯◯VERY LOW
Selenium and all-cause mortality risk (random effects model) - RCTs
4 randomised trials
not serious not serious serious q serious r none c RR 0.99(0.88 to 1.12)
1 fewer per 1,000
(from 6 fewer to 6
more)
⨁⨁◯◯LOW
CI: Confidence interval; RR: Risk ratio
58
Explanationsa. Serious indirectness for selenium supplementation and total CVD risk (RCTs) since both included studies were conducted in specific populations, that is, Lippman et al., 2009 included males only (at higher CVD risk) and Stranges et al., 2006 was conducted in patients with history of non-melanoma skin cancers. b. Serious imprecision for selenium supplementation and total CVD risk (RCTs), as the 95% CI (RR, 0.96-1.12) overlaps with the minimally important difference for clinical harm (RR>1.05). c. Publication bias was not assessed as there were < 10 studies. d. Unable to assess inconsistency as only one study was conducted. e. Serious indirectness for selenium supplementation and total CHD risk (RCTs) as there was only one study which included individuals with a history of non-melanoma skin cancer. f. Serious imprecision for selenium supplementation and total CHD risk (RCTs) as the 95% CI (RR, 0.76-1.48) include both clinically important benefit (RR<0.95) and harm (RR>1.05). g. Serious indirectness for selenium supplementation and MI risk (RCTs) as one of the included studies, Stranges et al., 2006 - NPC, which represented ~ 96% of the population, was conducted in patients with history of non-melanoma skin cancers. h. Serious imprecision for selenium supplementation and MI risk (RCTs) as the 95% CI (RR, 0.62-1.39) include both clinically important benefit (RR<0.95) and harm (RR>1.05). i. Serious indirectness for selenium supplementation and stroke risk (RCTs), since both included studies were conducted in specific populations, that is, Lippman et al., 2009 included males only (at higher CVD risk) and Stranges et al., 2006 was conducted in patients with history of non-melanoma skin cancers. j. Serious imprecision for selenium supplementation and stroke risk (RCTs) as the 95% CI (RR, 0.70-1.14) include both clinically important benefit (RR<0.95) and harm (RR>1.05). k. Serious indirectness for selenium supplementation and total CVD mortality risk (RCTs) since one of the included studies, Lippman et al., 2009-SELECT, which represented ~94% of the population was conducted in males. l. Serious imprecision for selenium supplementation and CVD mortality risk (RCTs) as the 95% CI (RR, 0.66-1.51) include both clinically important benefit (RR<0.95) and harm (RR>1.05). m. Serious indirectness for selenium supplementation and MI mortality risk (RCTs) as one of the included studies, Lippman et al., 2009-SELECT, which represented ~95% of the population was conducted in males. n. Very serious imprecision for selenium supplementation and MI mortality risk (RCTs) as the 95% CI (RR, 0.49-2.99) include both appreciable clinically important benefit (RR<0.95) and harm (RR>1.05). o. Serious indirectness for selenium supplementation and stroke mortality risk (RCTs) as the one included study was conducted in males with higher CVD risks. p. Very serious imprecision for selenium supplementation and stroke mortality risk (RCTs) as the 95% CI (RR, 0.44-2.93) include both appreciable clinically important benefit (RR<0.95) and harm (RR>1.05). q. Serious indirectness for selenium supplementation and all-cause mortality risk (RCTs), as one of the included studies was conducted in males and accounted for 94% of the population. r. Serious imprecision for selenium supplementation and all-cause mortality risk (RCTs) as the 95% CI (RR, 0.88-1.12) include both clinically important benefit (RR<0.95) and harm (RR>1.05).
Supplementary Table 11. GRADE assessment for vitamin B-complex
Certainty assessment Effect Certainty
№ of studies
Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
B-Complex and total CVD risk (random effects model) - RCTs
9 randomised trials
not serious not serious not serious a serious b none c RR 0.98(0.93 to 1.04)
3 fewer per 1,000
(from 7 more to 12
fewer)
⨁⨁⨁◯MODERATE
B-complex and total CHD risk (random effects model) - RCTs
5 randomised trials
not serious not serious not serious d serious e none c RR 1.04(0.96 to 1.14)
6 more per 1,000
(from 6 fewer to 21
more)
⨁⨁⨁◯MODERATE
B-Complex and MI risk (random effects model) - RCTs
13 randomised trials
not serious not serious not serious f serious g none RR 1.00(0.93 to 1.07)
0 fewer per 1,000
(from 5 fewer to 5
more)
⨁⨁⨁◯MODERATE
B-complex and stroke risk (random effects model) - RCTs
12 randomised trials
not serious not serious not serious h serious i none RR 0.90(0.81 to 1.00)
5 fewer per 1,000
(from 0 fewer to 9
fewer)
⨁⨁⨁◯MODERATE
B-Complex and CVD mortality risk (random effects model) - RCTs
5 randomised trials
not serious not serious not serious j serious k none c RR 0.98(0.87 to 1.11)
2 fewer per 1,000
(from 9 more to 10
fewer)
⨁⨁⨁◯MODERATE
59
Certainty assessment Effect Certainty
№ of studies
Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
B-complex and CHD mortality risk (random effects model) - RCTs
3 randomised trials
not serious not serious not serious l serious m none c RR 1.09(0.97 to 1.23)
6 more per 1,000
(from 2 fewer to 16
more)
⨁⨁⨁◯MODERATE
B-complex and MI mortality risk (random effects model) - RCTs
2 randomised trials
not serious not serious not serious n serious o none c RR 1.11(0.93 to 1.32)
4 more per 1,000
(from 2 fewer to 11
more)
⨁⨁⨁◯MODERATE
B-complex and stroke mortality risk (random effects model) - RCTs
2 randomised trials
not serious not serious not serious p serious q none c RR 0.91(0.68 to 1.21)
1 fewer per 1,000
(from 2 more to 3
fewer)
⨁⨁⨁◯MODERATE
B-Complex and all-cause mortality risk (random effects model) - RCTs
16 randomised trials
not serious not serious not serious r serious s none RR 1.02(0.97 to 1.06)
3 more per 1,000
(from 4 fewer to 8
more)
⨁⨁⨁◯MODERATE
CI: Confidence interval; RR: Risk ratio
Explanationsa. No serious indirectness for vitamin B-complex supplementation and total CVD risk (RCTs). Three studies were primary prevention and six studies were secondary prevention. b. Serious imprecision for vitamin B-complex supplementation and total CVD risk (RCTs) as the 95% CI (RR, 0.93-1.04) overlaps with the minimally important difference for clinical benefit (RR<0.95). c. Publication bias was not assessed since there were < 10 studies. d. No serious indirectness for vitamin B-complex supplementation and total CHD risk (RCTs). All studies were secondary prevention. e. Serious imprecision for vitamin B-complex supplementation and total CHD risk (RCTs) as the 95% CI (RR, 0.96-1.14) overlaps with the minimally important difference for clinical harm (RR>1.05). f. No serious indirectness for vitamin B-complex supplementation and MI risk (RCTs). Four studies were primary prevention, seven studies were secondary prevention and two studies were a combination of both. g. Serious imprecision for vitamin B-complex supplementation and MI risk (RCTs) as the 95% CI (RR, 0.93-1.07) include both clinically important benefit (RR <0.95) and harm (RR>1.05). h. No serious indirectness in vitamin B-complex supplementation and stroke risk (RCTs). Four studies were primary prevention and eight studies were secondary prevention. i. Serious imprecision for vitamin B-complex supplementation and stroke risk (RCTs) as the 95% CI (RR, 0.81-1.00) overlaps with the minimally important difference for clinical benefit (RR<0.95). j. No serious indirectness for vitamin B-complex supplementation and total CVD mortality risk (RCTs). Three studies were secondary prevention and two studies were a combination of primary and secondary prevention. k. Serious imprecision for vitamin B-complex supplementation and CVD mortality risk (RCTs) as the 95% CI (RR, 0.87, 1.11) include both clinically important benefit (RR<0.95) and harm (RR>1.05). l. No serious indirectness for vitamin B-complex supplementation and CHD mortality risk (RCTs). One study was primary prevention and two studies were secondary prevention. m. Serious imprecision for vitamin B-complex supplementation and CHD mortality risk (RCTs) as the 95% CI (RR, 0.97-1.23) overlaps with the minimally important difference for clinical harm (RR>1.05). n. No serious indirectness for vitamin B-complex supplementation and MI mortality risk (RCTs). Both studies were secondary prevention. o. Serious imprecision for vitamin B-complex supplementation and MI mortality risk (RCTs) as the 95% CI (RR, 0.93-1.32) include both clinically important benefit (RR<0.95) and harm (RR>1.05). p. No serious indirectness for vitamin B-complex supplementation and stroke mortality risk (RCTs), although both studies were secondary prevention. q. Serious imprecision for vitamin B-complex supplementation and stroke mortality risk (RCTs) as the 95% CI (RR, 0.68-1.21) include both clinically important benefit (RR<0.95) and harm (RR>1.05). r. No serious indirectness for vitamin B-complex supplementation and all-cause mortality risk (RCTs). Seven studies were primary prevention, seven studies were secondary prevention and two studies were a combination of both. s. Serious imprecision for vitamin B-complex supplementation and all-cause mortality risk (RCTs) as the 95% CI (RR, 0.97-1.06) overlaps with the minimally important difference for clinical harm (RR>1.05).
Supplementary Table 12. GRADE assessment for folic acid
60
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Folic acid and total CVD risk (random effects model) - RCTs
5 randomised trials
not serious not serious serious a not serious none b RR 0.83(0.73 to 0.93)
8 fewer per 1,000
(from 3 fewer to 13
fewer)
⨁⨁⨁◯MODERATE
Folic acid and total CHD risk (random effects model) - RCTs
2 randomised trials
not serious not serious not serious c serious d none b RR 1.47(0.95 to 2.28)
13 more per 1,000
(from 1 fewer to 36
more)
⨁⨁⨁◯MODERATE
Folic acid and MI risk (random effects model) - RCTs
6 randomised trials
not serious not serious serious e serious f none b RR 1.21(0.78 to 1.88)
1 more per 1,000
(from 1 fewer to 3
more)
⨁⨁◯◯LOW
Folic acid and stroke risk (random effects model) - RCTs
7 randomised trials
not serious not serious serious g not serious none b RR 0.80(0.69 to 0.93)
not serious not serious serious n serious o none RR 0.87(0.72 to 1.05)
5 fewer per 1,000
(from 2 more to 10
fewer)
⨁⨁◯◯LOW
CI: Confidence interval; RR: Risk ratio
Explanationsa. Serious indirectness for folic acid supplementation and CVD risk (RCTs), as the benefit appears to be driven by one study (Huo et al. 2015-CSPPT) conducted in a Chinese population-removal of this one study changes the pooled effect estimate from a beneficial lowering to no lowering in CVD risk (RR=0.89; 95%CI: 0.72, 1.10, p=0.28). One study was secondary prevention and four studies were primary prevention.
61
b. Publication bias was not assessed since there were <10 studies. c. No serious indirectness for folic acid supplementation and total CHD risk (RCTs). One study was primary prevention, while the other one was secondary prevention. d. Serious imprecision for folic acid supplementation and total CHD risk (RCTs), as the 95% CI (RR, 0.95-2.28) overlaps with the minimally important difference for clinical harm (RR>1.05). e. Serious indirectness for folic acid supplementation and MI risk (RCTs). The majority of the population came from one study (Huo et al., 2015-CSPPT) that was conducted in a chinese population only. Two studies were secondary prevention and four studies were primary prevention. f. Serious imprecision for folic acid supplementation and MI risk (RCTs) as the 95% CI (RR, 0.78-1.88) include both clinically important benefit (RR<0.95) and harm (RR>1.05). g. Serious indirectness for folic acid supplementation and stroke risk (RCTs), as the benefit appears to be driven by one study (Huo et al. 2015-CSPPT) conducted in a Chinese population -removal of this one study changes the pooled effect estimate from a beneficial lowering to no lowering in stroke risk (RR=0.92; 95%CI: 0.52, 1.63, p=0.78). Five studies were primary prevention and two were secondary prevention. h. Serious indirectness for folic acid supplementation and CVD mortality risk (RCTs). The majority of the population came from one study (Huo et al., 2015-CSPPT) that was conducted in a chinese population only. One study was secondary prevention and four studies were primary prevention. i. Serious imprecision for folic acid supplementation and CVD mortality risk (RCTs) as the 95% CI (RR, 0.68-1.17) include both clinically important benefit (RR<0.95) and harm (RR>1.05). j. Serious indirectness for folic acid supplementation and MI mortality risk (RCTs). The majority of the population came from one study (Huo et al., 2015-CSPPT) that was conducted in a chinese population only. One study was secondary prevention and one study was primary prevention. k. Very serious imprecision for folic acid supplementation and MI mortality risk (RCTs) as the 95% CI (RR, 0.39-3.49) include both appreciable clinically important benefit (RR<0.95) and harm (RR>1.05). l. Serious indirectness for folic acid supplementation and stroke mortality risk (RCTs). The majority of the population came from one study (Huo et al., 2015-CSPPT) that was conducted in a chinese population only. One study was secondary prevention and one study was primary prevention. m. Serious imprecision for folic acid supplementation and stroke mortality risk (RCTs) as the 95% CI (RR, 0.88-3.93) include both clinically important benefit (RR<0.95) and harm (RR>1.05). n. Serious indirectness for folic acid supplementation and all-cause mortality risk (RCTs). The majority of the population came from one study (Huo et al., 2015-CSPPT) that was conducted in a chinese population only. Two studies were secondary prevention, seven studies were primary prevention and one study was conducted in heart transplantation recipients. o. Serious imprecision for folic acid supplementation and all-cause mortality risk (RCTs), as the 95% CI (RR, 0.72-1.05) overlaps with the minimally important difference for clinical benefit (RR<0.95).
Supplementary Table 13. GRADE assessment for niacin (B3)
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Vitamin B3 (Niacin) and total CVD risk (random effects model) - RCTs
4 randomised trials
not serious not serious not serious a serious b none c RR 0.97(0.93 to 1.03)
7 fewer per 1,000
(from 7 more to 17
fewer)
⨁⨁⨁◯MODERATE
Vitamin B3 (Niacin) and total CVD risk - No statins (random effects model) - RCTs
1 randomised trials
not serious not serious d serious e not serious none c RR 0.98(0.95 to 1.01)
17 fewer per 1,000
(from 8 more to 42
fewer)
⨁⨁⨁◯MODERATE
Vitamin B3 (Niacin) and total CVD risk - Background statin treatment (random effects model) - RCTs
3 randomised trials
not serious not serious not serious f serious g none c RR 0.97(0.81 to 1.17)
4 fewer per 1,000
(from 22 more to 25
fewer)
⨁⨁⨁◯MODERATE
Vitamin B3 (Niacin) and total CHD risk (random effects model) - RCTs
2 randomised trials
not serious not serious not serious serious h none c RR 0.96(0.87 to 1.07)
2 fewer per 1,000
(from 4 more to 7
fewer)
⨁⨁⨁◯MODERATE
Vitamin B3 (Niacin) and MI risk (random effects model) - RCTs
5 randomised trials
not serious not serious not serious i serious j none c RR 0.90(0.75 to 1.06)
5 fewer per 1,000
(from 3 more to 13
fewer)
⨁⨁⨁◯MODERATE
62
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Vitamin B3 (Niacin) and MI risk - No statins (random effects model) - RCTs
1 randomised trials
not serious not serious d serious k not serious none c RR 0.74(0.60 to 0.90)
36 fewer per 1,000(from 14
fewer to 55 fewer)
⨁⨁⨁◯MODERATE
Vitamin B3 (Niacin) and MI risk- Background statin treatment (random effects model) - RCTs
4 randomised trials
not serious not serious not serious l serious m none c RR 0.96(0.85 to 1.08)
not serious not serious not serious aj serious ak none c RR 1.1(1.0 to 1.2)
6 more per 1,000
(from 0 fewer to 11
more)
⨁⨁⨁◯MODERATE
CI: Confidence interval; RR: Risk ratio
Explanationsa. No serious indirectness for vitamin B3 (niacin) supplementation and total CVD risk (RCTs), despite all studies being in populations with a history of CVD or CHD; the total population was also high (29,254). b. Serious imprecision for vitamin B3 (niacin) supplementation and total CVD risk (RCTs), as the 95% CI (RR, 0.93-1.03) overlaps with the minimally important difference for clinical benefit (RR<0.95). c. Publication bias was not assessed since there were <10 studies. d. Not able to assess inconsistency as only one study was included. e. Serious indirectness for vitamin B3 (niacin) supplementation and total CVD risk (RCTs-no statins) as the one included study was conducted in a population with CHD. f. No serious indirectness for vitamin B3 (niacin) supplementation and total CVD risk (RCTs-background statin). All three included studies were secondary prevention. g. Serious imprecision for vitamin B3 (niacin) supplementation and total CVD risk (RCTs-background statin), as the 95% CI (RR, 0.81-1.17) include both clinically important benefit (RR<0.95) and harm (RR>1.05). h. Serious imprecision for vitamin B3 (niacin) supplementation and total CHD risk (RCTs), as the 95% CI (RR, 0.87-1.07) include both clinically important benefit (RR<0.95) and harm (RR>1.05). i. No serious indirectness for vitamin B3 (niacin) supplementation and MI risk (RCTs). Four studies were secondary prevention and one study was primary prevention. j. Serious imprecision for vitamin B3 (niacin) supplementation and MI risk (RCTs-random effects model), as the 95% CI (RR, 0.75-1.06) include both clinically important benefit (RR<0.95) and harm (RR>1.05).
64
k. Serious indirectness for vitamin B3 (niacin) supplementation and MI risk (RCTs-no statins) as the one included study was conducted in a population with CHD. l. No serious indirectness for vitamin B3 (niacin) supplementation and MI risk (RCTs-background statin). All four included studies were secondary prevention. m. Serious imprecision for vitamin B3 (niacin) supplementation and MI risk (RCTs-background statin), as the 95% CI (RR 0.85-1.08) include both clinically important benefit (RR<0.95) and harm (RR>1.05). n. Although there was evidence of serious inconsistency for vitamin B3 (niacin) supplementation and stroke risk (RCTs) (I²=60%, p=0.04), independent removal of one trial (CDPRG 1975) explained almost half of the heterogeneity (I²=37%, P=0.19). o. No serious indirectness for vitamin B3 (niacin) supplementation and stroke risk (RCTs). Four studies were secondary prevention and one study was primary prevention. p. Serious imprecision for vitamin B3 (niacin) supplementation and stroke risk (RCTs-random effects model), as the 95% CI (RR, 0.72-1.26) include both clinically important benefit (RR<0.95) and harm (RR>1.05). q. No serious indirectness for vitamin B3 (niacin) supplementation and stroke risk (RCTs-no statins). The included study was secondary prevention. r. No serious indirectness for vitamin B3 (niacin) supplementation and stroke risk (RCTs-backgroud statin). Three studies were secondary prevention and one study was primary prevention. s. Serious imprecision for vitamin B3 (niacin) supplementation and stroke risk (RCTs-background statin) (random effects model) as the 95% CI (RR, 0.70-1.72) include both clinically important benefit (RR<0.95) and harm (RR>1.05). t. No serious indirectness for vitamin B3 (niacin) supplementation and CVD mortality risk (RCTs), despite all studies being in populations with a history of CVD or CHD; the total population was also high (7,489). u. Serious imprecision for vitamin B3 (niacin) supplementation and CVD mortality risk (RCTs), as the 95% CI (RR, 0.84-1.08) include both clinically important benefit (RR<0.95) and harm (RR> 1.05). v. Serious indirectness for niacin (B3) supplementation and CVD mortality risk (RCTs-no statins) as the one included study was conducted in a population with CHD. w. Serious imprecision for vitamin B3 (niacin) supplementation and CVD mortality risk (RCTs-no statins), as the 95% CI (RR, 0.82-1.07) include both clinically important benefit (RR<0.95) and harm (RR>1.05). x. No serious indirectness for vitamin B3 (niacin) supplementation and CVD mortality risk (RCTs-background statin). Both included studies were secondary prevention. y. Serious imprecision for vitamin B3 (niacin) supplementation and CVD mortality risk (RCTs-background statin), as the 95% CI (RR 0.75, 1.73) include both clinically important benefit (RR<0.95) and harm (RR>1.05). z. No serious indirectness for vitamin B3 (niacin) supplementation and CHD mortality risk (RCTs). All three studies were secondary prevention. aa. Serious imprecision for vitamin B3 (niacin) supplementation and CHD mortality risk (RCTs), as the 95% CI (RR, 0.89-1.10) include both clinically important benefit (RR<0.95) and harm (RR>1.05). ab. No serious indirectness for vitamin B3 (niacin) supplementation and CHD mortality risk (RCTs-no statin). The one included study was secondary prevention. ac. Serious imprecision for vitamin B3 (niacin) supplementation and CHD mortality risk (RCTs-no statins), as the 95% CI (RR, 0.82-1.09) includes both clinically important benefit (RR<0.95) and harm (RR>1.05). ad. No serious indirectness for vitamin B3 (niacin) supplementation and CHD mortality risk (RCTs-background statin). Both studies were secondary prevention. ae. Serious imprecision for vitamin B3 (niacin) supplementation and CHD mortality risk (RCTs-background statin), as the 95% CI (RR, 0.90-1.21) include both clinically important benefit (RR<0.95) and harm (RR>1.05). af. No serious indirectness for vitamin B3 (niacin) supplementation and all-cause mortality risk (RCTs), despite all studies being in populations with a history of CVD or CHD; the total population was also high (33,103). ag. Serious imprecision for vitamin B3 (niacin) supplementation and all-cause mortality risk (RCTs), as the 95% CI (RR, 0.95-1.14) overlaps with the minimally important difference for clinical harm (RR>1.05). ah. Serious indirectness for vitamin B3 (niacin) supplementation and all-cause mortality risk (RCTs-no statin) as the one included study was conducted in a population with CHD. ai. Serious imprecision for vitamin B3 (niacin) supplementation and all-cause mortality risk (RCTs-no statin), as the 95% CI (RR, 0.85-1.08) include both clinically benefit (RR<0.95) and harm (RR>1.05) aj. No serious indirectness for vitamin B3 (niacin) supplementation and all-cause mortality risk (RCTs-background statin). All 3 studies were secondary prevention. ak. Serious imprecision for vitamin B3 (niacin) supplementation and all-cause mortality risk (RCTs-background statin) as the 95% CI (RR, 1.0-1.2) overlaps with the minimally important difference for clinical harm (RR>1.05)
Supplementary Table 14. GRADE assessment for vitamin B6
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Vitamin B6 (Pyridoxine) and total CVD risk (random effects model) - RCTs
1 randomised trials
not serious not serious a not serious b serious c none d RR 1.03(0.85 to 1.24)
5 more per 1,000
(from 27 fewer to 44
more)
⨁⨁⨁◯MODERATE
Vitamin B6 (Pyridoxine) and MI risk (random effects model) - RCTs
2 randomised trials
not serious not serious not serious e serious f none d RR 1.04(0.87 to 1.23)
not serious not serious not serious k serious l none d RR 1.02(0.80 to 1.30)
1 more per 1,000
(from 14 fewer to 21
more)
⨁⨁⨁◯MODERATE
CI: Confidence interval; RR: Risk ratio
Explanationsa. Unable to assess inconsistency as only one study was included. b. No serious indirectness for vitamin B6 (pyridoxine) supplementation and total CVD risk (RCTs), although the single study was secondary prevention. c. Serious imprecision for vitamin B6 (pyridoxine) supplementation and total CVD risk (RCTs), as the 95% CI (RR, 0.85-1.24) include both clinically important benefit (RR<0.95) and harm (RR>1.05). d. Publication bias was not assessed since there were <10 studies. e. No serious indirectness for vitamin B6 (pyridoxine) supplementation and MI risk (RCTs), although both included studies were secondary prevention. f. Serious imprecision for vitamin B6 (pyridoxine) supplementation and MI risk (RCTs), as the 95% CI (RR, 0.87- 1.23) include both clinically important benefit (RR<0.95) and harm (RR>1.05). g. No serious indirectness for vitamin B6 (pyridoxine) supplementation and stroke risk (RCTs), although both included studies were secondary prevention. h. Serious imprecision for vitamin B6 (pyridoxine) supplementation and stroke risk (RCTs), as the 95% CI (RR, 0.61-1.40) include both clinically important benefit (RR<0.95) and harm (RR>1.05). i. No serious indirectness for vitamin B6 (pyridoxine) supplementation and MI mortality risk (RCTs), although the one included study was conducted in a population with MI. It was a secondary prevention study. j. Serious imprecision for vitamin B6 (pyridoxine) supplementation and MI mortality risk (RCTs), as the 95% CI (RR, 0.74-1.48) include both clinically important benefit (RR<0.95) and harm (RR>1.05). k. No serious indirectness for vitamin B6 (pyridoxine) supplementation and all-cause mortality risk (RCTs), although both included studies were conducted in a population with MI and suspected CAD. Both were secondary prevention studies. l. Serious imprecision for vitamin B6 (pyridoxine) supplementation and all-cause mortality risk (RCTs), as the 95% CI (RR 0.80, 1.30) include both clinically important benefit (RR<0.95) and harm (RR>1.05).
Supplementary Table 15. GRADE assessment for calcium
Certainty assessment Effect Certainty
№ of studies
Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Calcium and total CVD risk (random effects model) - RCTs
3 randomised trials
not serious serious a serious b serious c none d RR 1.43(0.79 to 2.59)
41 more per 1,000
(from 20 fewer to 150
more)
⨁◯◯◯VERY LOW
Calcium and total CHD risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious e serious f none d RR 1.16(0.87 to 1.56)
8 more per 1,000
(from 6 fewer to 27
more)
⨁⨁◯◯LOW
Calcium and MI risk (random effects model) - RCTs
4 randomised trials
not serious serious g not serious h serious i none d RR 1.69(0.94 to 3.04)
24 more per 1,000
(from 2 fewer to 72
more)
⨁⨁◯◯LOW
66
Certainty assessment Effect Certainty
№ of studies
Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Calcium and stroke risk (random effects model) - RCTs
3 randomised trials
not serious not serious serious j serious k none d RR 1.29(0.96 to 1.72)
12 more per 1,000
(from 2 fewer to 29
more)
⨁⨁◯◯LOW
Calcium and CVD mortality risk (random effects model) - RCTs
2 randomised trials
not serious not serious l serious m very serious n none d RR 1.24(0.27 to 5.65)
4 more per 1,000
(from 12 fewer to 79
more)
⨁◯◯◯VERY LOW
Calcium and MI mortality risk (random effects model) - RCTs
1 randomised trials
not serious not serious o serious p serious q none d RR 1.44(0.62 to 3.36)
5 more per 1,000
(from 5 fewer to 29
more)
⨁⨁◯◯LOW
Calcium and stroke mortality risk (random effects model) - RCTs
1 randomised trials
not serious not serious o serious r very serious s none d RR 0.75(0.26 to 2.15)
3 fewer per 1,000
(from 8 fewer to 13
more)
⨁◯◯◯VERY LOW
Calcium and all-cause mortality risk (random effects model) - RCTs
6 randomised trials
not serious not serious not serious t serious u none d RR 1.08(0.97 to 1.21)
9 more per 1,000
(from 3 fewer to 22
more)
⨁⨁⨁◯MODERATE
CI: Confidence interval; RR: Risk ratio
Explanationsa. Serious inconsistency for calcium supplementation and total CVD risk (RCTs), as I²=80% and p=0.007. b. Serious indirectness for calcium supplementation and total CVD risk (RCTs), as all of the included studies were conducted in post-menopausal women. c. Serious imprecision for calcium supplementation and total CVD risk (RCTs), as the 95% CI (RR, 0.79-2.59) include both clinically important benefit (RR<0.95) and harm (RR>1.05). d. Publication bias was not assessed as there were <10 studies. e. Serious indirectness for calcium supplementation and total CHD risk (RCTs), as 29% of the population have a history of large bowel adenomas and the remaining 71% have low trauma fracture. f. Serious imprecision for calcium supplementation and total CHD risk (RCTs), as the 95% CI (RR, 0.87-1.56) include both clinically important benefit (RR<0.95) and harm (RR>1.05). g. Serious inconsistency for calcium supplementation and MI risk (RCTs), as I²=69% and p=0.02. h. No serious indirectness for calcium supplementation and MI risk (RCTs). All studies were primary prevention. i. Serious imprecision for calcium supplementation and MI risk (RCTs), as the 95% CI (RR, 0.94-3.04) include both clinically important benefit (RR<0.95) and harm (RR>1.05). j. Serious indirectness for calcium supplementation and stroke risk (RCTs), as 24% of total participants had a history of large bowel adenomas and 76% of total participants are post-menopausal women. k. Serious imprecision for calcium supplementation and stroke risk (RCTs), as the 95% CI (RR, 0.96-1.72) overlaps with the minimally important difference for clinical harm (RR>1.05). l. No serious inconsistency for calcium supplementation and CVD mortality risk (RCTs), although I²=53% and p=0.14. m. Serious indirectness for calcium supplementation and total CVD mortality risk (RCTs), as both included studies were conducted in post-menopausal women. n. Very serious imprecision for calcium supplementation and CVD mortality risk (RCTs), as the 95% CI (RR, 0.27-5.65) include both appreciable clinically important benefit (RR<0.95) and harm (RR>1.05). o. Unable to assess inconsistency as only one study was included. p. Serious indirectness for calcium supplementation and MI mortality risk (RCTs), as the only study was conducted in post-menopausal women. q. Serious imprecision for calcium supplementation and MI mortality risk (RCTs), as the 95% CI (RR, 0.62-3.36) include both clinically important benefit (RR<0.95) and harm (RR>1.05). r. Serious indirectness for calcium supplementation and stroke mortality risk (RCTs), as the only included study was conducted in post-menopausal women. s. Very serious imprecision for calcium supplementation and stroke mortality risk (RCTs), as the 95% CI (RR, 0.26-2.15) include both appreciable clinically important benefit (RR<0.95) and harm (RR>1.05). t. No serious indirectness for calcium supplementation and all-cause mortality risk (RCTs). All studies were primary prevention. u. Serious imprecision for calcium supplementation and all-cause mortality risk (RCTs), as the 95% CI (RR, 0.97-1.21) overlaps with the minimally important difference for clinical harm (RR>1.05).
67
Supplementary Table 16. GRADE assessment for iron
Certainty assessment Effect Certainty
№ of studies
Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Iron and total CVD risk (IV iron for iron deficiency in heart failure) (random effects model) - RCTs
1 randomised trials
not serious not serious a serious b not serious none c RR 0.49(0.28 to 0.86)
72 fewer per 1,000(from 20
fewer to 102 fewer)
⨁⨁⨁◯MODERATE
Iron and MI risk (IV iron for iron deficiency in heart failure) (random effects model) - RCTs
1 randomised trials
not serious not serious a serious d very serious e none c RR 0.34(0.06 to 2.01)
13 fewer per 1,000(from 18
fewer to 20 more)
⨁◯◯◯VERY LOW
Iron and CVD mortality risk (IV iron for iron deficiency in heart failure) (random effects model) - RCTs
2 randomised trials
not serious not serious not serious f serious g none c RR 0.80(0.40 to 1.58)
10 fewer per 1,000(from 30
more to 31 fewer)
⨁⨁⨁◯MODERATE
Iron and all-cause mortality risk (Iron for iron deficiency in heart failure) (random effects model) - RCTs
2 randomised trials
not serious not serious serious h serious i none c RR 0.79(0.42 to 1.51)
12 fewer per 1,000(from 30
more to 34 fewer)
⨁⨁◯◯LOW
CI: Confidence interval; RR: Risk ratio
Explanationsa. Unable to assess inconsistency as only one study was included. b. Serious indirectness for iron supplementation and total CVD risk (RCTs), as the one included study was conducted in patients with chronic heart failure. c. Publication bias was not assessed as there were <10 studies. d. Serious indirectness for iron supplementation and MI risk (RCTs), as the one included study was conducted in patients with chronic heart failure. e. Very serious imprecision for iron supplementation and MI risk (RCTs), as the 95% CI (RR, 0.06-2.01) include both appreciable clinically important benefit (RR<0.95) and harm (RR>1.05). f. No serious indirectness for iron supplementation and total CVD mortality risk (RCTs), even though there were only two studies, one study was a multi-center study and included several countries and both studies were conducted in high risk populations for CVD mortality. g. Serious imprecision for iron supplementation and CVD mortality risk (RCTs), as the 95% CI (RR, 0.40-1.58) include both clinically important benefit (RR<0.95) and harm (RR>1.05). h. Serious indirectness for iron supplementation and all-cause mortality risk (RCTs), as both studies were conducted in patients with heart failure. i. Serious imprecision for iron supplementation and all-cause mortality risk (RCTs), as the 95% CI (RR, 0.42-1.51) include both clinically important benefit (RR<0.95) and harm (RR>1.05).
Supplementary Table 17. GRADE assessment for multivitamin
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
Multivitamin and total CVD risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious a serious b none c RR 0.95(0.77 to 1.17)
6 fewer per 1,000
(from 20 more to 27
fewer)
⨁⨁◯◯LOW
Multivitamin and MI risk (random effects model) - RCTs
68
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
3 randomised trials
not serious not serious not serious e serious f none c RR 0.95(0.82 to 1.09)
2 fewer per 1,000
(from 4 more to 9
fewer)
⨁⨁⨁◯MODERATE
Multivitamin and stroke (random effects model) - RCTs
2 randomised trials
not serious serious g serious h serious i none c RR 0.86(0.46 to 1.62)
6 fewer per 1,000
(from 22 fewer to 25
more)
⨁◯◯◯VERY LOW
Multivitamin and CVD mortality risk (random effects model) - RCTs
3 randomised trials
not serious not serious serious j serious k none c RR 0.94(0.83 to 1.06)
4 fewer per 1,000
(from 4 more to 10
fewer)
⨁⨁◯◯LOW
Multivitamin and MI mortality risk (random effects model) - RCTs
1 randomised trials
not serious not serious d serious l serious m none c RR 0.63(0.39 to 1.02)
2 fewer per 1,000
(from 0 fewer to 4
fewer)
⨁⨁◯◯LOW
Multivitamins and stroke mortality risk (random effects model) - RCTs
2 randomised trials
not serious not serious not serious n serious o none c RR 0.88(0.51 to 1.51)
1 fewer per 1,000
(from 6 fewer to 6
more)
⨁⨁⨁◯MODERATE
Multivitamin and all-cause mortality risk (random effects model) - RCTs
10 randomised trials
not serious not serious not serious p serious q none RR 0.95(0.90 to 1.01)
8 fewer per 1,000
(from 2 more to 16
fewer)
⨁⨁⨁◯MODERATE
CI: Confidence interval; RR: Risk ratio
Explanationsa. Serious indirectness for multivitamin supplementation and total CVD risk (RCTs) as one of the included studies (Sesso et al., 2012- PHS II), which represented ~90% of the population, was conducted in males only. One study was secondary prevention and the other one was primary prevention. b. Serious imprecision for multivitamin supplementation and total CVD risk (RCTs), as the 95% CI (RR, 0.77-1.17) include both clinically important benefit (RR<0.95) and harm (RR>1.05). c. Publication bias was not assessed since there were <10 studies. d. Inconsistency was not assessed as only one study was included. e. No serious indirectness for multivitamin supplementation and MI risk (RCTs). Majority of the studies were primary prevention, with the exception of one (Lamas et al., 2013) which included individuals post MI. f. Serious imprecision for multivitamin supplementation and MI risk (RCTs), as the 95% CI (RR, 0.82-1.09) include both clinically important benefit (RR<0.95) and harm (RR>1.05). g. Serious inconsistency for multivitamin supplementation and stroke risk (RCTs), as I²=59% and p=0.12 (effect estimates for the two included studies go in opposite directions). h. Serious indirectness for multivitamin supplementation and stroke risk (RCTs), as one of the included studies (Sesso et al., 2012- PHS II), which represented ~83% of the population, was conducted in males only. One included study was primary prevention and the other one was for secondary prevention. i. Serious imprecision for multivitamin supplementation and stroke risk (RCTs), as the 95% CI (RR, 0.46-1.62) include both clinically important benefit (RR<0.95) and harm (RR>1.05). j. Serious indirectness for multivitamin supplementation and CVD mortality risk (RCTs), as one of the included studies (Sesso et al., 2012- PHS II), which represented ~83% of the population, was conducted in males only. k. Serious imprecision for multivitamin supplementation and CVD mortality risk (RCTs), as the 95% CI (RR, 0.83-1.06) include both clinically important benefit (RR<0.95) and harm (RR>1.05). l. Serious indirectness for multivitamin supplementation and MI mortality risk (RCTs), as the one included study (Sesso et al., 2012- PHS II), was conducted in males only. m. Serious imprecision for multivitamin supplementation and MI mortality risk (RCTs), as the 95% CI (RR, 0.39-1.02) overlaps with the minimally important difference for clinical benefit (RR<0.95). n. No serious indirectness for multivitamin supplementation and stroke mortality risk (RCTs). One study was for primary prevention and the second one, Sesso et al., 2012 was a combination of both.
69
o. Serious imprecision for multivitamin supplementation and stroke mortality risk (RCTs), as the 95% CI (RR, 0.49-1.63) include both clinically important benefit (RR<0.95) and harm (RR>1.05). p. No serious indirectness for multivitamin supplementation and all-cause mortality (RCTs). Majority of the studies were for primary prevention with the exception for one study (Lamas et al., 2013) which included individuals post MI. q. Serious imprecision for multivitamin supplementation and all-cause mortality risk (RCTs), as the 95% CI (RR, 0.90-1.01) overlaps with the minimally important difference for clinical benefit (RR<0.95).
Supplementary Table 18. GRADE assessment for calcium and vitamin D
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
VitD+Ca and total CVD risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious a serious b none c RR 1.03(0.94 to 1.12)
3 more per 1,000
(from 6 fewer to 13
more)
⨁⨁◯◯LOW
VitD+Ca and total CHD risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious e very serious f none c RR 0.73(0.20 to 2.68)
8 fewer per 1,000
(from 23 fewer to 47
more)
⨁◯◯◯VERY LOW
VitD+Ca and MI risk (random effects model) - RCTs
5 randomised trials
not serious not serious serious g serious h none c RR 1.14(0.95 to 1.37)
3 more per 1,000
(from 1 fewer to 8
more)
⨁⨁◯◯LOW
VitD+Ca and stroke risk (random effects model) - RCTs
7 randomised trials
not serious not serious not serious i serious j none c RR 1.17(0.98 to 1.39)
4 more per 1,000
(from 0 fewer to 9
more)
⨁⨁⨁◯MODERATE
Vit D+Ca and CVD mortality risk (random effects model) - RCTs
1 randomised trials
not serious not serious d serious k serious l none c RR 0.92(0.77 to 1.10)
1 fewer per 1,000
(from 1 more to 3
fewer)
⨁⨁◯◯LOW
VitD+Ca and CHD mortality risk (random effects model) - RCTs
2 randomised trials
not serious not serious serious m serious n none c RR 0.94(0.54 to 1.64)
0 fewer per 1,000
(from 3 fewer to 5
more)
⨁⨁◯◯LOW
Vit D+Ca and MI mortality risk (random effects model) - RCTs
1 randomised trials
not serious not serious d serious o very serious p none c RR 2.97(0.12 to 72.26)
0 fewer per 1,000
(from 0 fewer to 0
fewer)
⨁◯◯◯VERY LOW
VitD+Ca and stroke mortality risk (random effects model) - RCTs
70
Certainty assessment Effect
Certainty№ of
studiesStudy design Risk of bias Inconsistency Indirectness Imprecision Other considerations Relative
(95% CI)Absolute(95% CI)
2 randomised trials
not serious not serious serious q serious r none c RR 0.90(0.63 to 1.29)
0 fewer per 1,000
(from 1 fewer to 1
more)
⨁⨁◯◯LOW
VitD+Ca and all-cause mortality (random effects model) - RCTs
20 randomised trials
not serious not serious not serious s serious t none RR 0.95(0.89 to 1.01)
4 fewer per 1,000
(from 1 more to 10
fewer)
⨁⨁⨁◯MODERATE
CI: Confidence interval; RR: Risk ratio
Explanationsa. Serious indirectness for calcium and vitamin D supplementation and total CVD risk (RCTs), as both studies were in women only. b. Serious imprecision for calcium and vitamin D supplementation and total CVD risk (RCTs), as the 95% CI (RR, 0.94-1.12) include both clinically important benefit (RR<0.95) and harm (RR>1.05). c. Publication bias was not assessed as there were < 10 studies. d. Unable to assess inconsistency as there was only 1 study included. e. Serious indirectness for calcium and vitamin D supplementation and total CHD risk (RCTs), as nearly 100% of the population were women. f. Very serious imprecision for calcium and vitamin D supplementation and total CHD risk (RCTs), as the 95% CI (RR, 0.20-2.68) include both appreciable clinically important benefit (RR<0.95) and harm (RR>1.05). g. Serious indirectness for calcium and vitamin D supplementation and MI risk (RCTs), as 95% of the population were women. All studies were primary prevention. h. Serious imprecision for calcium and vitamin D supplementation and MI risk (RCTs), as the 95% CI (RR, 0.95-1.37) overlaps with minimally important difference for clinical harm (RR>1.05). i. Serious indirectness for calcium and vitamin D supplementation and stroke risk (RCTs). All studies were primary prevention. j. Serious imprecision for calcium and vitamin D supplementation and stroke (RCTs), as the 95% CI (RR, 0.98-1.39) overlaps with minimally important difference for clinical harm (RR>1.05). k. Serious indirectness for calcium and vitamin D supplementation and CVD mortality risk (RCTs), as the only included study was conducted in women only. l. Serious imprecision for calcium and vitamin D supplementation and CVD mortality (RCTs), as the 95% CI (RR, 0.77-1.10) include both clinically important benefit (RR<0.95) and harm (RR>1.05). m. Serious indirectness for calcium and vitamin D supplementation and CHD mortality risk (RCTs), as >99% of the population were women. n. Serious imprecision for calcium and vitamin D supplementation and CHD mortality risk (RCTs), as the 95% CI (RR, 0.54-1.64) include both clinically important benefit (RR<0.95) and harm (RR>1.05). o. Serious indirectness for calcium and vitamin D supplementation and MI mortality risk (RCTs), as the only study was conducted in early menopausal women only. p. Very serious imprecision for calcium and vitamin D supplementation and MI mortality risk (RCTs), as the 95% CI (RR, 0.12-72.26) include both appreciable clinically important benefit (RR<0.95) and harm (RR>1.05). q. Serious indirectness for calcium and vitamin D supplementation and stroke mortality risk (RCTs), as >99% of the population were women. r. Serious imprecision for calcium and vitamin D supplementation and stroke mortality risk (RCTs), as the 95% CI (RR, 0.63-1.29) include both clinically important benefit (RR<0.95) and harm (RR>1.05). s. No serious indirectness for calcium and vitamin D supplementation and all-cause mortality risk (RCTs). All but one study were primary prevention, the sole secondary prevention study was conducted in ischemic stroke survivors (Gupta et al., 2016). t. Serious imprecision for calcium and vitamin D supplementation and all-cause mortality risk (RCTs), as the 95% CI (RR, 0.89-1.01) overlaps with minimally important difference for clinical benefit (RR<0.95).
71
Supplementary Table 19. Summary of the meta-analayses results for all-cause mortality, CVD mortality, total CVD risk and other significant assocations
MI risk 0 4 RCTs 6m-3.9y 0.96 0.85, 1.08 0.53 509/15,313 527/14,883 0 Random ModerateStroke risk 0 4 RCTs 6m-3.9y 1.10 0.70, 1.72 0.68 528/15,313 519/14,883 37 Random Moderate
Study lengths I2 (%)
Model (R/F)
Grade95% CI P-valueSupplements EffectNumber of
RCTs/CohortsRisk
Ratio Participant number
Beta carotene
Vitamin A
Vitamin D
Vitamin C
Vitamin B3 (Niacin) - overall effect
Folic acid
Selenium
B-Complex
Vitamin E
Antioxidants
Vitamin B3 (Niacin) - no statin
Vitamin B3 (Niacin)- with background statin
72
Supplementary Table 20 (continued). Summary of the meta-analayses results for all-cause mortality, CVD mortality, total CVD risk and other significant assocations
1383 Excluded 479 Duplicates 2 Foreign Language 96 Less than 6 months 158 No outcome of interest 100 No supplement of interest 8 Non-human 59 Observational studies 9 SRMA of cohorts 181 Non-supplemental 74 Protocol 8 Review 209 Wrong population
113 PAPERS 55 SRMA (952 RCTs, 2 RCTs/cohorts)
54 single RCTs 4 single RCT/cohorts
833 Excluded (Duplicates, not the nutrient of interest, no control arm, foreign language, no supplement use, less than 6 months)
Studies identified through systematic search.
All individual RCT and cohort studies.
179 INCLUDED RCTs Outcome Vit D Vit A β-carotene Antioxidants Vit E Vit C Selenium B-complex Folic acid Total CVD 6 0 3 7 10 2 2 9 5 Total CHD 3 0 2 1 2 1 1 5 2 MI 12 0 3 6 10 2 2 13 6 Stroke 11 0 3 7 11 2 2 12 7 CVD mortality 2 0 4 7 11 2 3 5 5 CHD mortality 2 0 1 2 2 0 0 3 0 MI mortality 4 0 1 3 8 2 2 2 2 Stroke mortality 2 0 2 5 6 2 1 2 2 All-cause mortality
Supplementary Figure 2. Risk of bias summary for supplements and CVD and total mortality. Review authors' judgments about each risk of bias item for each included study.
75
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Albert et al., 2008 - WAFACS ?
Aloia et al., 1988 ?
Aloia et al., 2005 +
Alvarez et al., 2012 ?
Anker et al., 2009 - FAIR-HF +
AREDS Research Group 2001 +
Armitage et al., 2010 - SEARCH +
Avenell et al., 2005 - MAVIS +
Baeksgaard et al., 1998 ?
Bairati et al., 2006 +
Baker et al., 2002 ?
Baron et al., 1999 - CPPS +
Bjorkman et al., 2008 ?
Boaz et al., 2000 - SPACE +
Boden et al., 2011 - AIM HIGH ?
Bogden et al., 1994 +
Bolland et al., 2008 +
Bolton-Smith et al., 2007 +
Bonaa et al., 2006 - NORVIT ?
Brazier et al., 2005 ?
Brohult et al., 1973 ?
Brown et al., 2001 - HATS ?
CDPRG 1975 ?
Chailurkit et al., 2010 ?
Chandra et al., 1992 +
Chapuy et al., 1992 ?
Chapuy et al., 2002 - Decalyos II ?
Cherniack et al., 2011 ?
Chylack et al.., 2002 - REACT +
CLIPS Group 2007 ?
Coburn et al., 2004 ?
Cole et al., 2007 - AFPPS +
Collins et al., 2003 +
Cook et al., 2007 - WACS +
Cooper et al., 2003 ?
Corless et al., 1985 +
Correa et al., 2000 +
Coyne et al., 2006 +
CTNS Study Group 2008 ?
Daly et al., 2008 +
Dawson-Hughes et al., 1997 ?
de Gaetano et al., 2001 - PPP +
de la Maza et al., 1995 +
Delanaye et al., 2013 ?
Desnuelle et al., 2001 - ALSRT ?
de Waart et al., 2001 ?
de Zeeuw et al., 2010 - VITAL +
Dukas et al., 2004 ?
Durga et al., 2007 - FACIT ?
Ebbing et al., 2008 - WENBIT +
Flicker et al., 2005 +
Frazao et al., 2000 ?
Galan et al., 2010 - SU.FOL.OM3 +
Gallagher et al., 1990 +
Gallagher et al., 2001 - STOP IT –
Gallagher et al., 2012 - VIDOS +
Gillilan et al., 1997 ?
Girodon et al., 1997 ?
Girodon et al., 1999 - MIN.VIT.AOX ?
GISSI-Prevenzione Investigators 1999 +
Glendenning et al., 2012 +
Graat et al., 2002 +
Grady et al., 1991 ?
Graf et al., 2005 ?
Grant et al., 2005 - RECORD +
Greenberg et al., 1996 - SCPS +
Green et al., 1999 - NSCPT +
Grimnes et al. 2011 ?
Gupta et al., 2016 +
Guyton et al., 2008 ?
Hamdy et al., 1995 ?
Harwood et al., 2004 - NoNOF +
Heinz et al., 2010 ?
Hennekens et al., 1996 - PHS +
Hercberg et al., 2010 - SU.VI.MAX ?
Hewitt et al., 2013 +
Hodis et al., 2002 - VEAPS +
Hodis et al., 2009 - BVAIT +
Hoffman et al., 1999 ?
House et al., 2010- DIVINe +
HPS2-THRIVE Collaborative Group 2014 +
HPS Collaborative Group 2002 +
Huo et al., 2015 - CSPPT +
Imasa et al., 2009 ?
Inkovaara et al., 1983 –
Jacobson et al., 2000 ?
Jamison et al., 2007 - HOST +
Janssen et al.,2010 ?
Komulainen et al., 1999 - OSTPRE ?
Korpela et al., 1989 ?
Krieg et al., 1999 ?
Lacroix et al., 2009 - WHI +
Lamas et al., 2013 - TACT +
Lange et al., 2004 ?
Lappe et al., 2007 +
Lappe et al., 2017 +
Larsen et al., 2004 +
Law et al., 2006 +
Lee et al., 1999 - WHS +
Lee et al., 2005 - WHS +
Lehouck et al., 2012 +
Leppälä et al., 2000 - ATBC ?
Lewis et al., 2011 - CAIFOS +
Liem et al., 2003 +
Liem et al., 2004 ?
Li et al., 1993 - NIT 2 ?
Limburg et al., 2005 +
Lippman et al., 2009 - SELECT ?
Lips et al., 1996 +
Liu et al., 2007 +
Logan et al., 2008 - UKCAP +
Lonn et al., 2005 - HOPE & HOPE TOO ?
Lonn et al., 2006 - HOPE TOO ?
Lyons et al., 2007 +
Ma et al., 2012 - SIT ?
Magliano et al., 2006 - MAVET ?
Manning et al., 2013 +
Manuel-Y-Kennoy et al., 2004 - DATOR ?
Marshall et al., 2011 ?
McKeown-Eyssen et al., 1988 ?
McNeil et al., 2004 - VECAT ?
Meier et al., 2004 ?
Meydani et al., 2004 +
Meyer et al., 2002 –
Milman et al., 2008 - ICARE +
Moon et al., 1997 - SKICAP AK ?
Mooney et al., 2005 +
Omenn et al., 1996 - CARET ?
Ooms et al., 1995 ?
Ott et al., 1989 ?
Petersen et al., 2005 - ADCS 2 +
Pike et al., 1995 +
Plummer et al., 2007 +
Ponikowski et al. 2015 - CONFIRM-HF +
Porthouse et al., 2005 +
Potena et al., 2005 +
Prentice et al., 2013 - WHI CaD +
Prince et al., 2008 +
Punthakee et al., 2012 - TIDE +
Reid et al., 2008 +
Righetti et al., 2003 ?
Salonen et al., 2000 - ASAP ?
Salovaara et al., 2010 - OSTRE-FPS +
Sanders et al., 2010 - Vital D +
Sang et al., 2009 ?
Sano et al., 1997 - ADCS 1 ?
Sanyal et al., 2010 - PIVENS ?
Saposnik et al., 2009 - HOPE 2 +
Sato et al., 1997 ?
Sato et al., 1999 +
Sato et al., 2005 +
Schleitoff et al., 2006 +
Schnyder et al., 2002 - The Swiss Heart Study ?
Sesso et al., 2008 - PHS II +
Sesso et al., 2012 - PHS II +
Shoulson et al., 1998 - DATATOP ?
Smith et al., 2007 ?
Steiner et al., 1995 ?
Stephens et al., 1996 - CHAOS +
Stranges et al., 2006 - NPC ?
Taylor et al., 2004 - ARBITER 2 +
Tornwall et al., 2004 - ATBC ?
Trivedi et al., 2003 ?
Van Dijk et al., 2015 - B-PROOF +
Van Wijngaarden et al., 2014 - B-PROOF +
Vianna et al., 2007 ?
Virtamo et al., 1998 - ATBC ?
Virtamo et al., 2003 - ATBC ?
VITATOPS Trial Study Group 2010 +
Wang et al., 2014 - OPERA +
Wang et al., 2014 - PHS II +
Wang et al., 2015 +
Waters et al., 2002 - WAVE +
Witham et al., 2013 +
Witham et al., 2013 - VitDISH +
Wluka et al., 2002 +
Wu et al., 2009 - NHS/HPFS +
You et al., 2001 - SIT ?
Zhu et al., 2008 - CAIFOS +
Zoungas et al., 2006 - ASFAST ?
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
+
?
?
+
+
+
+
?
?
?
?
?
+
?
+
?
+
+
+
?
?
?
?
+
?
?
?
?
+
+
+
+
+
+
?
?
?
+
?
+
?
+
?
?
?
?
+
+
+
+
+
+
+
+
+
+
?
?
?
+
+
+
?
?
+
?
+
+
–
+
?
+
+
?
?
?
?
?
?
+
+
+
+
+
+
?
+
+
?
?
?
?
+
?
–
–
–
?
?
?
+
?
+
?
?
?
+
+
+
+
+
+
+
+
?
+
+
?
?
?
?
?
+
?
+
?
?
+
+
?
+
?
+
+
+
+
?
+
+
+
+
+
?
+
+
?
+
+
?
?
+
?
?
?
?
?
+
?
?
+
+
?
?
+
+
?
?
?
+
?
?
+
+
?
+
?
+
+
+
?
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(per
form
ance
bia
s)
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
–
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Inco
mpl
ete
outc
ome
data
(attr
ition
bia
s)
+
+
+
+
+
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
?
+
+
?
+
+
?
+
+
+
+
+
?
+
+
+
–
+
+
+
+
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
+
?
+
+
+
+
+
?
+
?
?
+
+
+
+
?
?
+
+
+
+
+
?
+
?
+
?
?
?
–
+
?
+
?
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
+
?
+
+
?
?
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
?
+
?
+
?
+
?
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
+
+
+
Sel
ectiv
e re
porti
ng (r
epor
ting
bias
)
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Albert et al., 2008 - WAFACS ?
Aloia et al., 1988 ?
Aloia et al., 2005 +
Alvarez et al., 2012 ?
Anker et al., 2009 - FAIR-HF +
AREDS Research Group 2001 +
Armitage et al., 2010 - SEARCH +
Avenell et al., 2005 - MAVIS +
Baeksgaard et al., 1998 ?
Bairati et al., 2006 +
Baker et al., 2002 ?
Baron et al., 1999 - CPPS +
Bjorkman et al., 2008 ?
Boaz et al., 2000 - SPACE +
Boden et al., 2011 - AIM HIGH ?
Bogden et al., 1994 +
Bolland et al., 2008 +
Bolton-Smith et al., 2007 +
Bonaa et al., 2006 - NORVIT ?
Brazier et al., 2005 ?
Brohult et al., 1973 ?
Brown et al., 2001 - HATS ?
CDPRG 1975 ?
Chailurkit et al., 2010 ?
Chandra et al., 1992 +
Chapuy et al., 1992 ?
Chapuy et al., 2002 - Decalyos II ?
Cherniack et al., 2011 ?
Chylack et al.., 2002 - REACT +
CLIPS Group 2007 ?
Coburn et al., 2004 ?
Cole et al., 2007 - AFPPS +
Collins et al., 2003 +
Cook et al., 2007 - WACS +
Cooper et al., 2003 ?
Corless et al., 1985 +
Correa et al., 2000 +
Coyne et al., 2006 +
CTNS 2008 ?
Daly et al., 2006 +
Dawson-Hughes et al., 1997 ?
de Gaetano et al., 2001 - PPP +
de la Maza et al., 1995 +
Delanaye et al., 2013 ?
Desnuelle et al., 2001 - ALSRT ?
de Waart et al., 2001 ?
de Zeeuw et al., 2010 - VITAL +
Dukas et al., 2004 ?
Durga et al., 2007 - FACIT ?
Ebbing et al., 2008 - WENBIT +
Flicker et al., 2005 +
Frazao et al., 2000 ?
Galan et al., 2010 - SU.FOL.OM3 +
Gallagher et al., 1990 +
Gallagher et al., 2001 - STOP IT –
Gallagher et al., 2012 - VIDOS +
Gillilan et al., 1997 ?
Girodon et al., 1997 ?
Girodon et al., 1999 - MIN.VIT.AOX ?
GISSI-Prevenzione Investigators 1999 +
Glendenning et al., 2012 +
Graat et al., 2002 +
Grady et al., 1991 ?
Graf et al., 2005 ?
Grant et al., 2005 - RECORD +
Greenberg et al., 1996 - SCPS +
Green et al., 1999 - NSCPT +
Grimnes et al. 2011 ?
Gupta et al., 2016 +
Guyton et al., 2008 ?
Hamdy et al., 1995 ?
Harwood et al., 2004 - NoNOF +
Heinz et al., 2010 ?
Hennekens et al., 1996 - PHS +
Hercberg et al., 2010 - SU.VI.MAX ?
Hewitt et al., 2013 +
Hodis et al., 2002 - VEAPS +
Hodis et al., 2009 - BVAIT +
Hoffman et al., 1999 ?
House et al., 2010- DIVINe +
HPS2-THRIVE Collaborative Group 2014 +
HPS Collaborative Group 2002 +
Huo et al., 2015 - CSPPT +
Imasa et al., 2009 ?
Inkovaara et al., 1983 –
Jacobson et al., 2000 ?
Jamison et al., 2007 - HOST +
Janssen et al.,2010 ?
Komulainen et al., 1999 - OSTPRE ?
Korpela et al., 1989 ?
Krieg et al., 1999 ?
Lacroix et al., 2009 - WHI +
Lamas et al., 2013 - TACT +
Lange et al., 2004 ?
Lappe et al., 2007 +
Lappe et al., 2017 +
Larsen et al., 2004 +
Law et al., 2006 +
Lee et al., 1999 - WHS +
Lee et al., 2005 - WHS +
Lehouck et al., 2012 +
Leppälä et al., 2000 - ATBC ?
Lewis et al., 2011 - CAIFOS +
Liem et al., 2003 +
Liem et al., 2004 ?
Li et al., 1993 - NIT 2 ?
Limburg et al., 2005 +
Lippman et al., 2009 - SELECT ?
Lips et al., 1996 +
Liu et al., 2007 +
Logan et al., 2008 - UKCAP +
Lonn et al., 2005 - HOPE & HOPE TOO ?
Lonn et al., 2006 - HOPE TOO ?
Lyons et al., 2007 +
Ma et al., 2012 - SIT ?
Magliano et al., 2006 - MAVET ?
Manning et al., 2013 +
Manuel-Y-Kennoy et al., 2004 - DATOR ?
Marshall et al., 2011 ?
McKeown-Eyssen et al., 1988 ?
McNeil et al., 2004 - VECAT ?
Meier et al., 2004 ?
Meydani et al., 2004 +
Meyer et al., 2002 –
Milman et al., 2008 - ICARE +
Moon et al., 1997 - SKICAP AK ?
Mooney et al., 2005 +
Omenn et al., 1996 - CARET ?
Ooms et al., 1995 ?
Ott et al., 1989 ?
Petersen et al., 2005 - ADCS 2 +
Pike et al., 1995 +
Plummer et al., 2007 +
Ponikowski et al. 2015 - CONFIRM-HF +
Porthouse et al., 2005 +
Potena et al., 2005 +
Prentice et al., 2013 - WHI CaD +
Prince et al., 2008 +
Punthakee et al., 2012 - TIDE +
Reid et al., 2008 +
Righetti et al., 2003 ?
Salonen et al., 2000 - ASAP ?
Salovaara et al., 2010 - OSTRE-FPS +
Sanders et al., 2010 - Vital D +
Sang et al., 2009 ?
Sano et al., 1997 - ADCS 1 ?
Sanyal et al., 2010 - PIVENS ?
Saposnik et al., 2009 - HOPE 2 +
Sato et al., 1997 ?
Sato et al., 1999 +
Sato et al., 2005 +
Schleitoff et al., 2006 +
Schnyder et al., 2002 - The Swiss Heart Study ?
Sesso et al., 2008 - PHS II +
Sesso et al., 2012 - PHS II +
Shoulson et al., 1998 - DATATOP ?
Smith et al., 2007 ?
Steiner et al., 1995 ?
Stephens et al., 1996 - CHAOS +
Stranges et al., 2006 - NPC ?
Taylor et al., 2004 - ARBITER 2 +
Tornwall et al., 2004 - ATBC ?
Trivedi et al., 2003 ?
Van Dijk et al., 2015 - B-PROOF +
Van Wijngaarden et al., 2014 - B-PROOF +
Vianna et al., 2007 ?
Virtamo et al., 1998 - ATBC ?
Virtamo et al., 2003 - ATBC ?
VITATOPS Trial Study Group 2010 +
Wang et al., 2014 - OPERA +
Wang et al., 2014 - PHS II +
Wang et al., 2015 +
Waters et al., 2002 - WAVE +
Witham et al., 2013 +
Witham et al., 2013 - VitDISH +
Wluka et al., 2002 +
Wu et al., 2009 - NHS/HPFS +
You et al., 2001 - SIT ?
Zhu et al., 2008 - CAIFOS +
Zoungas et al., 2006 - ASFAST ?
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
+
?
?
+
+
+
+
?
?
?
?
?
+
?
+
?
+
+
+
?
?
?
?
+
?
?
?
?
+
+
+
+
+
+
?
?
?
+
?
+
?
+
?
?
?
?
+
+
+
+
+
+
+
+
+
+
?
?
?
+
+
+
?
?
+
?
+
+
–
+
?
+
+
?
?
?
?
?
?
+
+
+
+
+
+
?
+
+
?
?
?
?
+
?
–
–
–
?
?
?
+
?
+
?
?
?
+
+
+
+
+
+
+
+
?
+
+
?
?
?
?
?
+
?
+
?
?
+
+
?
+
?
+
+
+
+
?
+
+
+
+
+
?
+
+
?
+
+
?
?
+
?
?
?
?
?
+
?
?
+
+
?
?
+
+
?
?
?
+
?
?
+
+
?
+
?
+
+
+
?
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(per
form
ance
bia
s)
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
–
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Inco
mpl
ete
outc
ome
data
(attr
ition
bia
s)
+
+
+
+
+
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
?
+
+
?
+
+
?
+
+
+
+
+
?
+
+
+
–
+
+
+
+
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
+
?
+
+
+
+
+
?
+
?
?
+
+
+
+
?
?
+
+
+
+
+
?
+
?
+
?
?
?
–
+
?
+
?
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
+
?
+
+
?
?
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
?
+
?
+
?
+
?
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
+
+
+
Sel
ectiv
e re
porti
ng (r
epor
ting
bias
)
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Supplementary Figure 2 (Continued). Risk of bias summary for supplements and CVD and total mortality. Review authors' judgments about each risk of bias item for each included study.
76
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Albert et al., 2008 - WAFACS ?
Aloia et al., 1988 ?
Aloia et al., 2005 +
Alvarez et al., 2012 ?
Anker et al., 2009 - FAIR-HF +
AREDS Research Group 2001 +
Armitage et al., 2010 - SEARCH +
Avenell et al., 2005 - MAVIS +
Baeksgaard et al., 1998 ?
Bairati et al., 2006 +
Baker et al., 2002 ?
Baron et al., 1999 - CPPS +
Bjorkman et al., 2008 ?
Boaz et al., 2000 - SPACE +
Boden et al., 2011 - AIM HIGH ?
Bogden et al., 1994 +
Bolland et al., 2008 +
Bolton-Smith et al., 2007 +
Bonaa et al., 2006 - NORVIT ?
Brazier et al., 2005 ?
Brohult et al., 1973 ?
Brown et al., 2001 - HATS ?
CDPRG 1975 ?
Chailurkit et al., 2010 ?
Chandra et al., 1992 +
Chapuy et al., 1992 ?
Chapuy et al., 2002 - Decalyos II ?
Cherniack et al., 2011 ?
Chylack et al.., 2002 - REACT +
CLIPS Group 2007 ?
Coburn et al., 2004 ?
Cole et al., 2007 - AFPPS +
Collins et al., 2003 +
Cook et al., 2007 - WACS +
Cooper et al., 2003 ?
Corless et al., 1985 +
Correa et al., 2000 +
Coyne et al., 2006 +
CTNS Study Group 2008 ?
Daly et al., 2008 +
Dawson-Hughes et al., 1997 ?
de Gaetano et al., 2001 - PPP +
de la Maza et al., 1995 +
Delanaye et al., 2013 ?
Desnuelle et al., 2001 - ALSRT ?
de Waart et al., 2001 ?
de Zeeuw et al., 2010 - VITAL +
Dukas et al., 2004 ?
Durga et al., 2007 - FACIT ?
Ebbing et al., 2008 - WENBIT +
Flicker et al., 2005 +
Frazao et al., 2000 ?
Galan et al., 2010 - SU.FOL.OM3 +
Gallagher et al., 1990 +
Gallagher et al., 2001 - STOP IT –
Gallagher et al., 2012 - VIDOS +
Gillilan et al., 1997 ?
Girodon et al., 1997 ?
Girodon et al., 1999 - MIN.VIT.AOX ?
GISSI-Prevenzione Investigators 1999 +
Glendenning et al., 2012 +
Graat et al., 2002 +
Grady et al., 1991 ?
Graf et al., 2005 ?
Grant et al., 2005 - RECORD +
Greenberg et al., 1996 - SCPS +
Green et al., 1999 - NSCPT +
Grimnes et al. 2011 ?
Gupta et al., 2016 +
Guyton et al., 2008 ?
Hamdy et al., 1995 ?
Harwood et al., 2004 - NoNOF +
Heinz et al., 2010 ?
Hennekens et al., 1996 - PHS +
Hercberg et al., 2010 - SU.VI.MAX ?
Hewitt et al., 2013 +
Hodis et al., 2002 - VEAPS +
Hodis et al., 2009 - BVAIT +
Hoffman et al., 1999 ?
House et al., 2010- DIVINe +
HPS2-THRIVE Collaborative Group 2014 +
HPS Collaborative Group 2002 +
Huo et al., 2015 - CSPPT +
Imasa et al., 2009 ?
Inkovaara et al., 1983 –
Jacobson et al., 2000 ?
Jamison et al., 2007 - HOST +
Janssen et al.,2010 ?
Komulainen et al., 1999 - OSTPRE ?
Korpela et al., 1989 ?
Krieg et al., 1999 ?
Lacroix et al., 2009 - WHI +
Lamas et al., 2013 - TACT +
Lange et al., 2004 ?
Lappe et al., 2007 +
Lappe et al., 2017 +
Larsen et al., 2004 +
Law et al., 2006 +
Lee et al., 1999 - WHS +
Lee et al., 2005 - WHS +
Lehouck et al., 2012 +
Leppälä et al., 2000 - ATBC ?
Lewis et al., 2011 - CAIFOS +
Liem et al., 2003 +
Liem et al., 2004 ?
Li et al., 1993 - NIT 2 ?
Limburg et al., 2005 +
Lippman et al., 2009 - SELECT ?
Lips et al., 1996 +
Liu et al., 2007 +
Logan et al., 2008 - UKCAP +
Lonn et al., 2005 - HOPE & HOPE TOO ?
Lonn et al., 2006 - HOPE TOO ?
Lyons et al., 2007 +
Ma et al., 2012 - SIT ?
Magliano et al., 2006 - MAVET ?
Manning et al., 2013 +
Manuel-Y-Kennoy et al., 2004 - DATOR ?
Marshall et al., 2011 ?
McKeown-Eyssen et al., 1988 ?
McNeil et al., 2004 - VECAT ?
Meier et al., 2004 ?
Meydani et al., 2004 +
Meyer et al., 2002 –
Milman et al., 2008 - ICARE +
Moon et al., 1997 - SKICAP AK ?
Mooney et al., 2005 +
Omenn et al., 1996 - CARET ?
Ooms et al., 1995 ?
Ott et al., 1989 ?
Petersen et al., 2005 - ADCS 2 +
Pike et al., 1995 +
Plummer et al., 2007 +
Ponikowski et al. 2015 - CONFIRM-HF +
Porthouse et al., 2005 +
Potena et al., 2005 +
Prentice et al., 2013 - WHI CaD +
Prince et al., 2008 +
Punthakee et al., 2012 - TIDE +
Reid et al., 2008 +
Righetti et al., 2003 ?
Salonen et al., 2000 - ASAP ?
Salovaara et al., 2010 - OSTRE-FPS +
Sanders et al., 2010 - Vital D +
Sang et al., 2009 ?
Sano et al., 1997 - ADCS 1 ?
Sanyal et al., 2010 - PIVENS ?
Saposnik et al., 2009 - HOPE 2 +
Sato et al., 1997 ?
Sato et al., 1999 +
Sato et al., 2005 +
Schleitoff et al., 2006 +
Schnyder et al., 2002 - The Swiss Heart Study ?
Sesso et al., 2008 - PHS II +
Sesso et al., 2012 - PHS II +
Shoulson et al., 1998 - DATATOP ?
Smith et al., 2007 ?
Steiner et al., 1995 ?
Stephens et al., 1996 - CHAOS +
Stranges et al., 2006 - NPC ?
Taylor et al., 2004 - ARBITER 2 +
Tornwall et al., 2004 - ATBC ?
Trivedi et al., 2003 ?
Van Dijk et al., 2015 - B-PROOF +
Van Wijngaarden et al., 2014 - B-PROOF +
Vianna et al., 2007 ?
Virtamo et al., 1998 - ATBC ?
Virtamo et al., 2003 - ATBC ?
VITATOPS Trial Study Group 2010 +
Wang et al., 2014 - OPERA +
Wang et al., 2014 - PHS II +
Wang et al., 2015 +
Waters et al., 2002 - WAVE +
Witham et al., 2013 +
Witham et al., 2013 - VitDISH +
Wluka et al., 2002 +
Wu et al., 2009 - NHS/HPFS +
You et al., 2001 - SIT ?
Zhu et al., 2008 - CAIFOS +
Zoungas et al., 2006 - ASFAST ?
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
+
?
?
+
+
+
+
?
?
?
?
?
+
?
+
?
+
+
+
?
?
?
?
+
?
?
?
?
+
+
+
+
+
+
?
?
?
+
?
+
?
+
?
?
?
?
+
+
+
+
+
+
+
+
+
+
?
?
?
+
+
+
?
?
+
?
+
+
–
+
?
+
+
?
?
?
?
?
?
+
+
+
+
+
+
?
+
+
?
?
?
?
+
?
–
–
–
?
?
?
+
?
+
?
?
?
+
+
+
+
+
+
+
+
?
+
+
?
?
?
?
?
+
?
+
?
?
+
+
?
+
?
+
+
+
+
?
+
+
+
+
+
?
+
+
?
+
+
?
?
+
?
?
?
?
?
+
?
?
+
+
?
?
+
+
?
?
?
+
?
?
+
+
?
+
?
+
+
+
?
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(per
form
ance
bia
s)
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
–
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Inco
mpl
ete
outc
ome
data
(attr
ition
bia
s)
+
+
+
+
+
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
?
+
+
?
+
+
?
+
+
+
+
+
?
+
+
+
–
+
+
+
+
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
+
?
+
+
+
+
+
?
+
?
?
+
+
+
+
?
?
+
+
+
+
+
?
+
?
+
?
?
?
–
+
?
+
?
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
+
?
+
+
?
?
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
?
+
?
+
?
+
?
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
+
+
+
Sel
ectiv
e re
porti
ng (r
epor
ting
bias
)
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Albert et al., 2008 - WAFACS ?
Aloia et al., 1988 ?
Aloia et al., 2005 +
Alvarez et al., 2012 ?
Anker et al., 2009 - FAIR-HF +
AREDS Research Group 2001 +
Armitage et al., 2010 - SEARCH +
Avenell et al., 2005 - MAVIS +
Baeksgaard et al., 1998 ?
Bairati et al., 2006 +
Baker et al., 2002 ?
Baron et al., 1999 - CPPS +
Bjorkman et al., 2008 ?
Boaz et al., 2000 - SPACE +
Boden et al., 2011 - AIM HIGH ?
Bogden et al., 1994 +
Bolland et al., 2008 +
Bolton-Smith et al., 2007 +
Bonaa et al., 2006 - NORVIT ?
Brazier et al., 2005 ?
Brohult et al., 1973 ?
Brown et al., 2001 - HATS ?
CDPRG 1975 ?
Chailurkit et al., 2010 ?
Chandra et al., 1992 +
Chapuy et al., 1992 ?
Chapuy et al., 2002 - Decalyos II ?
Cherniack et al., 2011 ?
Chylack et al.., 2002 - REACT +
CLIPS Group 2007 ?
Coburn et al., 2004 ?
Cole et al., 2007 - AFPPS +
Collins et al., 2003 +
Cook et al., 2007 - WACS +
Cooper et al., 2003 ?
Corless et al., 1985 +
Correa et al., 2000 +
Coyne et al., 2006 +
CTNS Study Group 2008 ?
Daly et al., 2008 +
Dawson-Hughes et al., 1997 ?
de Gaetano et al., 2001 - PPP +
de la Maza et al., 1995 +
Delanaye et al., 2013 ?
Desnuelle et al., 2001 - ALSRT ?
de Waart et al., 2001 ?
de Zeeuw et al., 2010 - VITAL +
Dukas et al., 2004 ?
Durga et al., 2007 - FACIT ?
Ebbing et al., 2008 - WENBIT +
Flicker et al., 2005 +
Frazao et al., 2000 ?
Galan et al., 2010 - SU.FOL.OM3 +
Gallagher et al., 1990 +
Gallagher et al., 2001 - STOP IT –
Gallagher et al., 2012 - VIDOS +
Gillilan et al., 1997 ?
Girodon et al., 1997 ?
Girodon et al., 1999 - MIN.VIT.AOX ?
GISSI-Prevenzione Investigators 1999 +
Glendenning et al., 2012 +
Graat et al., 2002 +
Grady et al., 1991 ?
Graf et al., 2005 ?
Grant et al., 2005 - RECORD +
Greenberg et al., 1996 - SCPS +
Green et al., 1999 - NSCPT +
Grimnes et al. 2011 ?
Gupta et al., 2016 +
Guyton et al., 2008 ?
Hamdy et al., 1995 ?
Harwood et al., 2004 - NoNOF +
Heinz et al., 2010 ?
Hennekens et al., 1996 - PHS +
Hercberg et al., 2010 - SU.VI.MAX ?
Hewitt et al., 2013 +
Hodis et al., 2002 - VEAPS +
Hodis et al., 2009 - BVAIT +
Hoffman et al., 1999 ?
House et al., 2010- DIVINe +
HPS2-THRIVE Collaborative Group 2014 +
HPS Collaborative Group 2002 +
Huo et al., 2015 - CSPPT +
Imasa et al., 2009 ?
Inkovaara et al., 1983 –
Jacobson et al., 2000 ?
Jamison et al., 2007 - HOST +
Janssen et al.,2010 ?
Komulainen et al., 1999 - OSTPRE ?
Korpela et al., 1989 ?
Krieg et al., 1999 ?
Lacroix et al., 2009 - WHI +
Lamas et al., 2013 - TACT +
Lange et al., 2004 ?
Lappe et al., 2007 +
Lappe et al., 2017 +
Larsen et al., 2004 +
Law et al., 2006 +
Lee et al., 1999 - WHS +
Lee et al., 2005 - WHS +
Lehouck et al., 2012 +
Leppälä et al., 2000 - ATBC ?
Lewis et al., 2011 - CAIFOS +
Liem et al., 2003 +
Liem et al., 2004 ?
Li et al., 1993 - NIT 2 ?
Limburg et al., 2005 +
Lippman et al., 2009 - SELECT ?
Lips et al., 1996 +
Liu et al., 2007 +
Logan et al., 2008 - UKCAP +
Lonn et al., 2005 - HOPE & HOPE TOO ?
Lonn et al., 2006 - HOPE TOO ?
Lyons et al., 2007 +
Ma et al., 2012 - SIT ?
Magliano et al., 2006 - MAVET ?
Manning et al., 2013 +
Manuel-Y-Kennoy et al., 2004 - DATOR ?
Marshall et al., 2011 ?
McKeown-Eyssen et al., 1988 ?
McNeil et al., 2004 - VECAT ?
Meier et al., 2004 ?
Meydani et al., 2004 +
Meyer et al., 2002 –
Milman et al., 2008 - ICARE +
Moon et al., 1997 - SKICAP AK ?
Mooney et al., 2005 +
Omenn et al., 1996 - CARET ?
Ooms et al., 1995 ?
Ott et al., 1989 ?
Petersen et al., 2005 - ADCS 2 +
Pike et al., 1995 +
Plummer et al., 2007 +
Ponikowski et al. 2015 - CONFIRM-HF +
Porthouse et al., 2005 +
Potena et al., 2005 +
Prentice et al., 2013 - WHI CaD +
Prince et al., 2008 +
Punthakee et al., 2012 - TIDE +
Reid et al., 2008 +
Righetti et al., 2003 ?
Salonen et al., 2000 - ASAP ?
Salovaara et al., 2010 - OSTRE-FPS +
Sanders et al., 2010 - Vital D +
Sang et al., 2009 ?
Sano et al., 1997 - ADCS 1 ?
Sanyal et al., 2010 - PIVENS ?
Saposnik et al., 2009 - HOPE 2 +
Sato et al., 1997 ?
Sato et al., 1999 +
Sato et al., 2005 +
Schleitoff et al., 2006 +
Schnyder et al., 2002 - The Swiss Heart Study ?
Sesso et al., 2008 - PHS II +
Sesso et al., 2012 - PHS II +
Shoulson et al., 1998 - DATATOP ?
Smith et al., 2007 ?
Steiner et al., 1995 ?
Stephens et al., 1996 - CHAOS +
Stranges et al., 2006 - NPC ?
Taylor et al., 2004 - ARBITER 2 +
Tornwall et al., 2004 - ATBC ?
Trivedi et al., 2003 ?
Van Dijk et al., 2015 - B-PROOF +
Van Wijngaarden et al., 2014 - B-PROOF +
Vianna et al., 2007 ?
Virtamo et al., 1998 - ATBC ?
Virtamo et al., 2003 - ATBC ?
VITATOPS Trial Study Group 2010 +
Wang et al., 2014 - OPERA +
Wang et al., 2014 - PHS II +
Wang et al., 2015 +
Waters et al., 2002 - WAVE +
Witham et al., 2013 +
Witham et al., 2013 - VitDISH +
Wluka et al., 2002 +
Wu et al., 2009 - NHS/HPFS +
You et al., 2001 - SIT ?
Zhu et al., 2008 - CAIFOS +
Zoungas et al., 2006 - ASFAST ?
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
+
?
?
+
+
+
+
?
?
?
?
?
+
?
+
?
+
+
+
?
?
?
?
+
?
?
?
?
+
+
+
+
+
+
?
?
?
+
?
+
?
+
?
?
?
?
+
+
+
+
+
+
+
+
+
+
?
?
?
+
+
+
?
?
+
?
+
+
–
+
?
+
+
?
?
?
?
?
?
+
+
+
+
+
+
?
+
+
?
?
?
?
+
?
–
–
–
?
?
?
+
?
+
?
?
?
+
+
+
+
+
+
+
+
?
+
+
?
?
?
?
?
+
?
+
?
?
+
+
?
+
?
+
+
+
+
?
+
+
+
+
+
?
+
+
?
+
+
?
?
+
?
?
?
?
?
+
?
?
+
+
?
?
+
+
?
?
?
+
?
?
+
+
?
+
?
+
+
+
?
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(per
form
ance
bia
s)
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
–
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Inco
mpl
ete
outc
ome
data
(attr
ition
bia
s)
+
+
+
+
+
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
?
+
+
?
+
+
?
+
+
+
+
+
?
+
+
+
–
+
+
+
+
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
+
?
+
+
+
+
+
?
+
?
?
+
+
+
+
?
?
+
+
+
+
+
?
+
?
+
?
?
?
–
+
?
+
?
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
+
?
+
+
?
?
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
?
+
?
+
?
+
?
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
+
+
+
Sel
ectiv
e re
porti
ng (r
epor
ting
bias
)
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Albert et al., 2008 - WAFACS ?
Aloia et al., 1988 ?
Aloia et al., 2005 +
Alvarez et al., 2012 ?
Anker et al., 2009 - FAIR-HF +
AREDS Research Group 2001 +
Armitage et al., 2010 - SEARCH +
Avenell et al., 2005 - MAVIS +
Baeksgaard et al., 1998 ?
Bairati et al., 2006 +
Baker et al., 2002 ?
Baron et al., 1999 - CPPS +
Bjorkman et al., 2008 ?
Boaz et al., 2000 - SPACE +
Boden et al., 2011 - AIM HIGH ?
Bogden et al., 1994 +
Bolland et al., 2008 +
Bolton-Smith et al., 2007 +
Bonaa et al., 2006 - NORVIT ?
Brazier et al., 2005 ?
Brohult et al., 1973 ?
Brown et al., 2001 - HATS ?
CDPRG 1975 ?
Chailurkit et al., 2010 ?
Chandra et al., 1992 +
Chapuy et al., 1992 ?
Chapuy et al., 2002 - Decalyos II ?
Cherniack et al., 2011 ?
Chylack et al.., 2002 - REACT +
CLIPS Group 2007 ?
Coburn et al., 2004 ?
Cole et al., 2007 - AFPPS +
Collins et al., 2003 +
Cook et al., 2007 - WACS +
Cooper et al., 2003 ?
Corless et al., 1985 +
Correa et al., 2000 +
Coyne et al., 2006 +
CTNS Study Group 2008 ?
Daly et al., 2008 +
Dawson-Hughes et al., 1997 ?
de Gaetano et al., 2001 - PPP +
de la Maza et al., 1995 +
Delanaye et al., 2013 ?
Desnuelle et al., 2001 - ALSRT ?
de Waart et al., 2001 ?
de Zeeuw et al., 2010 - VITAL +
Dukas et al., 2004 ?
Durga et al., 2007 - FACIT ?
Ebbing et al., 2008 - WENBIT +
Flicker et al., 2005 +
Frazao et al., 2000 ?
Galan et al., 2010 - SU.FOL.OM3 +
Gallagher et al., 1990 +
Gallagher et al., 2001 - STOP IT –
Gallagher et al., 2012 - VIDOS +
Gillilan et al., 1997 ?
Girodon et al., 1997 ?
Girodon et al., 1999 - MIN.VIT.AOX ?
GISSI-Prevenzione Investigators 1999 +
Glendenning et al., 2012 +
Graat et al., 2002 +
Grady et al., 1991 ?
Graf et al., 2005 ?
Grant et al., 2005 - RECORD +
Greenberg et al., 1996 - SCPS +
Green et al., 1999 - NSCPT +
Grimnes et al. 2011 ?
Gupta et al., 2016 +
Guyton et al., 2008 ?
Hamdy et al., 1995 ?
Harwood et al., 2004 - NoNOF +
Heinz et al., 2010 ?
Hennekens et al., 1996 - PHS +
Hercberg et al., 2010 - SU.VI.MAX ?
Hewitt et al., 2013 +
Hodis et al., 2002 - VEAPS +
Hodis et al., 2009 - BVAIT +
Hoffman et al., 1999 ?
House et al., 2010- DIVINe +
HPS2-THRIVE Collaborative Group 2014 +
HPS Collaborative Group 2002 +
Huo et al., 2015 - CSPPT +
Imasa et al., 2009 ?
Inkovaara et al., 1983 –
Jacobson et al., 2000 ?
Jamison et al., 2007 - HOST +
Janssen et al.,2010 ?
Komulainen et al., 1999 - OSTPRE ?
Korpela et al., 1989 ?
Krieg et al., 1999 ?
Lacroix et al., 2009 - WHI +
Lamas et al., 2013 - TACT +
Lange et al., 2004 ?
Lappe et al., 2007 +
Lappe et al., 2017 +
Larsen et al., 2004 +
Law et al., 2006 +
Lee et al., 1999 - WHS +
Lee et al., 2005 - WHS +
Lehouck et al., 2012 +
Leppälä et al., 2000 - ATBC ?
Lewis et al., 2011 - CAIFOS +
Liem et al., 2003 +
Liem et al., 2004 ?
Li et al., 1993 - NIT 2 ?
Limburg et al., 2005 +
Lippman et al., 2009 - SELECT ?
Lips et al., 1996 +
Liu et al., 2007 +
Logan et al., 2008 - UKCAP +
Lonn et al., 2005 - HOPE & HOPE TOO ?
Lonn et al., 2006 - HOPE TOO ?
Lyons et al., 2007 +
Ma et al., 2012 - SIT ?
Magliano et al., 2006 - MAVET ?
Manning et al., 2013 +
Manuel-Y-Kennoy et al., 2004 - DATOR ?
Marshall et al., 2011 ?
McKeown-Eyssen et al., 1988 ?
McNeil et al., 2004 - VECAT ?
Meier et al., 2004 ?
Meydani et al., 2004 +
Meyer et al., 2002 –
Milman et al., 2008 - ICARE +
Moon et al., 1997 - SKICAP AK ?
Mooney et al., 2005 +
Omenn et al., 1996 - CARET ?
Ooms et al., 1995 ?
Ott et al., 1989 ?
Petersen et al., 2005 - ADCS 2 +
Pike et al., 1995 +
Plummer et al., 2007 +
Ponikowski et al. 2015 - CONFIRM-HF +
Porthouse et al., 2005 +
Potena et al., 2005 +
Prentice et al., 2013 - WHI CaD +
Prince et al., 2008 +
Punthakee et al., 2012 - TIDE +
Reid et al., 2008 +
Righetti et al., 2003 ?
Salonen et al., 2000 - ASAP ?
Salovaara et al., 2010 - OSTRE-FPS +
Sanders et al., 2010 - Vital D +
Sang et al., 2009 ?
Sano et al., 1997 - ADCS 1 ?
Sanyal et al., 2010 - PIVENS ?
Saposnik et al., 2009 - HOPE 2 +
Sato et al., 1997 ?
Sato et al., 1999 +
Sato et al., 2005 +
Schleitoff et al., 2006 +
Schnyder et al., 2002 - The Swiss Heart Study ?
Sesso et al., 2008 - PHS II +
Sesso et al., 2012 - PHS II +
Shoulson et al., 1998 - DATATOP ?
Smith et al., 2007 ?
Steiner et al., 1995 ?
Stephens et al., 1996 - CHAOS +
Stranges et al., 2006 - NPC ?
Taylor et al., 2004 - ARBITER 2 +
Tornwall et al., 2004 - ATBC ?
Trivedi et al., 2003 ?
Van Dijk et al., 2015 - B-PROOF +
Van Wijngaarden et al., 2014 - B-PROOF +
Vianna et al., 2007 ?
Virtamo et al., 1998 - ATBC ?
Virtamo et al., 2003 - ATBC ?
VITATOPS Trial Study Group 2010 +
Wang et al., 2014 - OPERA +
Wang et al., 2014 - PHS II +
Wang et al., 2015 +
Waters et al., 2002 - WAVE +
Witham et al., 2013 +
Witham et al., 2013 - VitDISH +
Wluka et al., 2002 +
Wu et al., 2009 - NHS/HPFS +
You et al., 2001 - SIT ?
Zhu et al., 2008 - CAIFOS +
Zoungas et al., 2006 - ASFAST ?
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
+
?
?
+
+
+
+
?
?
?
?
?
+
?
+
?
+
+
+
?
?
?
?
+
?
?
?
?
+
+
+
+
+
+
?
?
?
+
?
+
?
+
?
?
?
?
+
+
+
+
+
+
+
+
+
+
?
?
?
+
+
+
?
?
+
?
+
+
–
+
?
+
+
?
?
?
?
?
?
+
+
+
+
+
+
?
+
+
?
?
?
?
+
?
–
–
–
?
?
?
+
?
+
?
?
?
+
+
+
+
+
+
+
+
?
+
+
?
?
?
?
?
+
?
+
?
?
+
+
?
+
?
+
+
+
+
?
+
+
+
+
+
?
+
+
?
+
+
?
?
+
?
?
?
?
?
+
?
?
+
+
?
?
+
+
?
?
?
+
?
?
+
+
?
+
?
+
+
+
?
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(per
form
ance
bia
s)
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
–
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Inco
mpl
ete
outc
ome
data
(attr
ition
bia
s)
+
+
+
+
+
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
?
+
+
?
+
+
?
+
+
+
+
+
?
+
+
+
–
+
+
+
+
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
+
?
+
+
+
+
+
?
+
?
?
+
+
+
+
?
?
+
+
+
+
+
?
+
?
+
?
?
?
–
+
?
+
?
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
+
?
+
+
?
?
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
?
+
?
+
?
+
?
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
+
+
+
Sel
ectiv
e re
porti
ng (r
epor
ting
bias
)
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Albert et al., 2008 - WAFACS ?
Aloia et al., 1988 ?
Aloia et al., 2005 +
Alvarez et al., 2012 ?
Anker et al., 2009 - FAIR-HF +
AREDS Research Group 2001 +
Armitage et al., 2010 - SEARCH +
Avenell et al., 2005 - MAVIS +
Baeksgaard et al., 1998 ?
Bairati et al., 2006 +
Baker et al., 2002 ?
Baron et al., 1999 - CPPS +
Bjorkman et al., 2008 ?
Boaz et al., 2000 - SPACE +
Boden et al., 2011 - AIM HIGH ?
Bogden et al., 1994 +
Bolland et al., 2008 +
Bolton-Smith et al., 2007 +
Bonaa et al., 2006 - NORVIT ?
Brazier et al., 2005 ?
Brohult et al., 1973 ?
Brown et al., 2001 - HATS ?
CDPRG 1975 ?
Chailurkit et al., 2010 ?
Chandra et al., 1992 +
Chapuy et al., 1992 ?
Chapuy et al., 2002 - Decalyos II ?
Cherniack et al., 2011 ?
Chylack et al.., 2002 - REACT +
CLIPS Group 2007 ?
Coburn et al., 2004 ?
Cole et al., 2007 - AFPPS +
Collins et al., 2003 +
Cook et al., 2007 - WACS +
Cooper et al., 2003 ?
Corless et al., 1985 +
Correa et al., 2000 +
Coyne et al., 2006 +
CTNS Study Group 2008 ?
Daly et al., 2008 +
Dawson-Hughes et al., 1997 ?
de Gaetano et al., 2001 - PPP +
de la Maza et al., 1995 +
Delanaye et al., 2013 ?
Desnuelle et al., 2001 - ALSRT ?
de Waart et al., 2001 ?
de Zeeuw et al., 2010 - VITAL +
Dukas et al., 2004 ?
Durga et al., 2007 - FACIT ?
Ebbing et al., 2008 - WENBIT +
Flicker et al., 2005 +
Frazao et al., 2000 ?
Galan et al., 2010 - SU.FOL.OM3 +
Gallagher et al., 1990 +
Gallagher et al., 2001 - STOP IT –
Gallagher et al., 2012 - VIDOS +
Gillilan et al., 1997 ?
Girodon et al., 1997 ?
Girodon et al., 1999 - MIN.VIT.AOX ?
GISSI-Prevenzione Investigators 1999 +
Glendenning et al., 2012 +
Graat et al., 2002 +
Grady et al., 1991 ?
Graf et al., 2005 ?
Grant et al., 2005 - RECORD +
Greenberg et al., 1996 - SCPS +
Green et al., 1999 - NSCPT +
Grimnes et al. 2011 ?
Gupta et al., 2016 +
Guyton et al., 2008 ?
Hamdy et al., 1995 ?
Harwood et al., 2004 - NoNOF +
Heinz et al., 2010 ?
Hennekens et al., 1996 - PHS +
Hercberg et al., 2010 - SU.VI.MAX ?
Hewitt et al., 2013 +
Hodis et al., 2002 - VEAPS +
Hodis et al., 2009 - BVAIT +
Hoffman et al., 1999 ?
House et al., 2010- DIVINe +
HPS2-THRIVE Collaborative Group 2014 +
HPS Collaborative Group 2002 +
Huo et al., 2015 - CSPPT +
Imasa et al., 2009 ?
Inkovaara et al., 1983 –
Jacobson et al., 2000 ?
Jamison et al., 2007 - HOST +
Janssen et al.,2010 ?
Komulainen et al., 1999 - OSTPRE ?
Korpela et al., 1989 ?
Krieg et al., 1999 ?
Lacroix et al., 2009 - WHI +
Lamas et al., 2013 - TACT +
Lange et al., 2004 ?
Lappe et al., 2007 +
Lappe et al., 2017 +
Larsen et al., 2004 +
Law et al., 2006 +
Lee et al., 1999 - WHS +
Lee et al., 2005 - WHS +
Lehouck et al., 2012 +
Leppälä et al., 2000 - ATBC ?
Lewis et al., 2011 - CAIFOS +
Liem et al., 2003 +
Liem et al., 2004 ?
Li et al., 1993 - NIT 2 ?
Limburg et al., 2005 +
Lippman et al., 2009 - SELECT ?
Lips et al., 1996 +
Liu et al., 2007 +
Logan et al., 2008 - UKCAP +
Lonn et al., 2005 - HOPE & HOPE TOO ?
Lonn et al., 2006 - HOPE TOO ?
Lyons et al., 2007 +
Ma et al., 2012 - SIT ?
Magliano et al., 2006 - MAVET ?
Manning et al., 2013 +
Manuel-Y-Kennoy et al., 2004 - DATOR ?
Marshall et al., 2011 ?
McKeown-Eyssen et al., 1988 ?
McNeil et al., 2004 - VECAT ?
Meier et al., 2004 ?
Meydani et al., 2004 +
Meyer et al., 2002 –
Milman et al., 2008 - ICARE +
Moon et al., 1997 - SKICAP AK ?
Mooney et al., 2005 +
Omenn et al., 1996 - CARET ?
Ooms et al., 1995 ?
Ott et al., 1989 ?
Petersen et al., 2005 - ADCS 2 +
Pike et al., 1995 +
Plummer et al., 2007 +
Ponikowski et al. 2015 - CONFIRM-HF +
Porthouse et al., 2005 +
Potena et al., 2005 +
Prentice et al., 2013 - WHI CaD +
Prince et al., 2008 +
Punthakee et al., 2012 - TIDE +
Reid et al., 2008 +
Righetti et al., 2003 ?
Salonen et al., 2000 - ASAP ?
Salovaara et al., 2010 - OSTRE-FPS +
Sanders et al., 2010 - Vital D +
Sang et al., 2009 ?
Sano et al., 1997 - ADCS 1 ?
Sanyal et al., 2010 - PIVENS ?
Saposnik et al., 2009 - HOPE 2 +
Sato et al., 1997 ?
Sato et al., 1999 +
Sato et al., 2005 +
Schleitoff et al., 2006 +
Schnyder et al., 2002 - The Swiss Heart Study ?
Sesso et al., 2008 - PHS II +
Sesso et al., 2012 - PHS II +
Shoulson et al., 1998 - DATATOP ?
Smith et al., 2007 ?
Steiner et al., 1995 ?
Stephens et al., 1996 - CHAOS +
Stranges et al., 2006 - NPC ?
Taylor et al., 2004 - ARBITER 2 +
Tornwall et al., 2004 - ATBC ?
Trivedi et al., 2003 ?
Van Dijk et al., 2015 - B-PROOF +
Van Wijngaarden et al., 2014 - B-PROOF +
Vianna et al., 2007 ?
Virtamo et al., 1998 - ATBC ?
Virtamo et al., 2003 - ATBC ?
VITATOPS Trial Study Group 2010 +
Wang et al., 2014 - OPERA +
Wang et al., 2014 - PHS II +
Wang et al., 2015 +
Waters et al., 2002 - WAVE +
Witham et al., 2013 +
Witham et al., 2013 - VitDISH +
Wluka et al., 2002 +
Wu et al., 2009 - NHS/HPFS +
You et al., 2001 - SIT ?
Zhu et al., 2008 - CAIFOS +
Zoungas et al., 2006 - ASFAST ?
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
+
?
?
+
+
+
+
?
?
?
?
?
+
?
+
?
+
+
+
?
?
?
?
+
?
?
?
?
+
+
+
+
+
+
?
?
?
+
?
+
?
+
?
?
?
?
+
+
+
+
+
+
+
+
+
+
?
?
?
+
+
+
?
?
+
?
+
+
–
+
?
+
+
?
?
?
?
?
?
+
+
+
+
+
+
?
+
+
?
?
?
?
+
?
–
–
–
?
?
?
+
?
+
?
?
?
+
+
+
+
+
+
+
+
?
+
+
?
?
?
?
?
+
?
+
?
?
+
+
?
+
?
+
+
+
+
?
+
+
+
+
+
?
+
+
?
+
+
?
?
+
?
?
?
?
?
+
?
?
+
+
?
?
+
+
?
?
?
+
?
?
+
+
?
+
?
+
+
+
?
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(per
form
ance
bia
s)
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
–
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Inco
mpl
ete
outc
ome
data
(attr
ition
bia
s)
+
+
+
+
+
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
?
+
+
?
+
+
?
+
+
+
+
+
?
+
+
+
–
+
+
+
+
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
+
?
+
+
+
+
+
?
+
?
?
+
+
+
+
?
?
+
+
+
+
+
?
+
?
+
?
?
?
–
+
?
+
?
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
+
?
+
+
?
?
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
?
+
?
+
?
+
?
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
+
+
+
Sel
ectiv
e re
porti
ng (r
epor
ting
bias
)
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Albert et al., 2008 - WAFACS ?
Aloia et al., 1988 ?
Aloia et al., 2005 +
Alvarez et al., 2012 ?
Anker et al., 2009 - FAIR-HF +
AREDS Research Group 2001 +
Armitage et al., 2010 - SEARCH +
Avenell et al., 2005 - MAVIS +
Baeksgaard et al., 1998 ?
Bairati et al., 2006 +
Baker et al., 2002 ?
Baron et al., 1999 - CPPS +
Bjorkman et al., 2008 ?
Boaz et al., 2000 - SPACE +
Boden et al., 2011 - AIM HIGH ?
Bogden et al., 1994 +
Bolland et al., 2008 +
Bolton-Smith et al., 2007 +
Bonaa et al., 2006 - NORVIT ?
Brazier et al., 2005 ?
Brohult et al., 1973 ?
Brown et al., 2001 - HATS ?
CDPRG 1975 ?
Chailurkit et al., 2010 ?
Chandra et al., 1992 +
Chapuy et al., 1992 ?
Chapuy et al., 2002 - Decalyos II ?
Cherniack et al., 2011 ?
Chylack et al.., 2002 - REACT +
CLIPS Group 2007 ?
Coburn et al., 2004 ?
Cole et al., 2007 - AFPPS +
Collins et al., 2003 +
Cook et al., 2007 - WACS +
Cooper et al., 2003 ?
Corless et al., 1985 +
Correa et al., 2000 +
Coyne et al., 2006 +
CTNS Study Group 2008 ?
Daly et al., 2008 +
Dawson-Hughes et al., 1997 ?
de Gaetano et al., 2001 - PPP +
de la Maza et al., 1995 +
Delanaye et al., 2013 ?
Desnuelle et al., 2001 - ALSRT ?
de Waart et al., 2001 ?
de Zeeuw et al., 2010 - VITAL +
Dukas et al., 2004 ?
Durga et al., 2007 - FACIT ?
Ebbing et al., 2008 - WENBIT +
Flicker et al., 2005 +
Frazao et al., 2000 ?
Galan et al., 2010 - SU.FOL.OM3 +
Gallagher et al., 1990 +
Gallagher et al., 2001 - STOP IT –
Gallagher et al., 2012 - VIDOS +
Gillilan et al., 1997 ?
Girodon et al., 1997 ?
Girodon et al., 1999 - MIN.VIT.AOX ?
GISSI-Prevenzione Investigators 1999 +
Glendenning et al., 2012 +
Graat et al., 2002 +
Grady et al., 1991 ?
Graf et al., 2005 ?
Grant et al., 2005 - RECORD +
Greenberg et al., 1996 - SCPS +
Green et al., 1999 - NSCPT +
Grimnes et al. 2011 ?
Gupta et al., 2016 +
Guyton et al., 2008 ?
Hamdy et al., 1995 ?
Harwood et al., 2004 - NoNOF +
Heinz et al., 2010 ?
Hennekens et al., 1996 - PHS +
Hercberg et al., 2010 - SU.VI.MAX ?
Hewitt et al., 2013 +
Hodis et al., 2002 - VEAPS +
Hodis et al., 2009 - BVAIT +
Hoffman et al., 1999 ?
House et al., 2010- DIVINe +
HPS2-THRIVE Collaborative Group 2014 +
HPS Collaborative Group 2002 +
Huo et al., 2015 - CSPPT +
Imasa et al., 2009 ?
Inkovaara et al., 1983 –
Jacobson et al., 2000 ?
Jamison et al., 2007 - HOST +
Janssen et al.,2010 ?
Komulainen et al., 1999 - OSTPRE ?
Korpela et al., 1989 ?
Krieg et al., 1999 ?
Lacroix et al., 2009 - WHI +
Lamas et al., 2013 - TACT +
Lange et al., 2004 ?
Lappe et al., 2007 +
Lappe et al., 2017 +
Larsen et al., 2004 +
Law et al., 2006 +
Lee et al., 1999 - WHS +
Lee et al., 2005 - WHS +
Lehouck et al., 2012 +
Leppälä et al., 2000 - ATBC ?
Lewis et al., 2011 - CAIFOS +
Liem et al., 2003 +
Liem et al., 2004 ?
Li et al., 1993 - NIT 2 ?
Limburg et al., 2005 +
Lippman et al., 2009 - SELECT ?
Lips et al., 1996 +
Liu et al., 2007 +
Logan et al., 2008 - UKCAP +
Lonn et al., 2005 - HOPE & HOPE TOO ?
Lonn et al., 2006 - HOPE TOO ?
Lyons et al., 2007 +
Ma et al., 2012 - SIT ?
Magliano et al., 2006 - MAVET ?
Manning et al., 2013 +
Manuel-Y-Kennoy et al., 2004 - DATOR ?
Marshall et al., 2011 ?
McKeown-Eyssen et al., 1988 ?
McNeil et al., 2004 - VECAT ?
Meier et al., 2004 ?
Meydani et al., 2004 +
Meyer et al., 2002 –
Milman et al., 2008 - ICARE +
Moon et al., 1997 - SKICAP AK ?
Mooney et al., 2005 +
Omenn et al., 1996 - CARET ?
Ooms et al., 1995 ?
Ott et al., 1989 ?
Petersen et al., 2005 - ADCS 2 +
Pike et al., 1995 +
Plummer et al., 2007 +
Ponikowski et al. 2015 - CONFIRM-HF +
Porthouse et al., 2005 +
Potena et al., 2005 +
Prentice et al., 2013 - WHI CaD +
Prince et al., 2008 +
Punthakee et al., 2012 - TIDE +
Reid et al., 2008 +
Righetti et al., 2003 ?
Salonen et al., 2000 - ASAP ?
Salovaara et al., 2010 - OSTRE-FPS +
Sanders et al., 2010 - Vital D +
Sang et al., 2009 ?
Sano et al., 1997 - ADCS 1 ?
Sanyal et al., 2010 - PIVENS ?
Saposnik et al., 2009 - HOPE 2 +
Sato et al., 1997 ?
Sato et al., 1999 +
Sato et al., 2005 +
Schleitoff et al., 2006 +
Schnyder et al., 2002 - The Swiss Heart Study ?
Sesso et al., 2008 - PHS II +
Sesso et al., 2012 - PHS II +
Shoulson et al., 1998 - DATATOP ?
Smith et al., 2007 ?
Steiner et al., 1995 ?
Stephens et al., 1996 - CHAOS +
Stranges et al., 2006 - NPC ?
Taylor et al., 2004 - ARBITER 2 +
Tornwall et al., 2004 - ATBC ?
Trivedi et al., 2003 ?
Van Dijk et al., 2015 - B-PROOF +
Van Wijngaarden et al., 2014 - B-PROOF +
Vianna et al., 2007 ?
Virtamo et al., 1998 - ATBC ?
Virtamo et al., 2003 - ATBC ?
VITATOPS Trial Study Group 2010 +
Wang et al., 2014 - OPERA +
Wang et al., 2014 - PHS II +
Wang et al., 2015 +
Waters et al., 2002 - WAVE +
Witham et al., 2013 +
Witham et al., 2013 - VitDISH +
Wluka et al., 2002 +
Wu et al., 2009 - NHS/HPFS +
You et al., 2001 - SIT ?
Zhu et al., 2008 - CAIFOS +
Zoungas et al., 2006 - ASFAST ?
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
+
?
?
+
+
+
+
?
?
?
?
?
+
?
+
?
+
+
+
?
?
?
?
+
?
?
?
?
+
+
+
+
+
+
?
?
?
+
?
+
?
+
?
?
?
?
+
+
+
+
+
+
+
+
+
+
?
?
?
+
+
+
?
?
+
?
+
+
–
+
?
+
+
?
?
?
?
?
?
+
+
+
+
+
+
?
+
+
?
?
?
?
+
?
–
–
–
?
?
?
+
?
+
?
?
?
+
+
+
+
+
+
+
+
?
+
+
?
?
?
?
?
+
?
+
?
?
+
+
?
+
?
+
+
+
+
?
+
+
+
+
+
?
+
+
?
+
+
?
?
+
?
?
?
?
?
+
?
?
+
+
?
?
+
+
?
?
?
+
?
?
+
+
?
+
?
+
+
+
?
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(per
form
ance
bia
s)
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
–
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Inco
mpl
ete
outc
ome
data
(attr
ition
bia
s)
+
+
+
+
+
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
?
+
+
?
+
+
?
+
+
+
+
+
?
+
+
+
–
+
+
+
+
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
+
?
+
+
+
+
+
?
+
?
?
+
+
+
+
?
?
+
+
+
+
+
?
+
?
+
?
?
?
–
+
?
+
?
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
+
?
+
+
?
?
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
?
+
?
+
?
+
?
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
+
+
+
Sel
ectiv
e re
porti
ng (r
epor
ting
bias
)
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Supplementary Figure 2 (Continued). Risk of bias summary for supplements and CVD and total mortality. Review authors' judgments about each risk of bias item for each included study
77
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Albert et al., 2008 - WAFACS ?
Aloia et al., 1988 ?
Aloia et al., 2005 +
Alvarez et al., 2012 ?
Anker et al., 2009 - FAIR-HF +
AREDS Research Group 2001 +
Armitage et al., 2010 - SEARCH +
Avenell et al., 2005 - MAVIS +
Baeksgaard et al., 1998 ?
Bairati et al., 2006 +
Baker et al., 2002 ?
Baron et al., 1999 - CPPS +
Bjorkman et al., 2008 ?
Boaz et al., 2000 - SPACE +
Boden et al., 2011 - AIM HIGH ?
Bogden et al., 1994 +
Bolland et al., 2008 +
Bolton-Smith et al., 2007 +
Bonaa et al., 2006 - NORVIT ?
Brazier et al., 2005 ?
Brohult et al., 1973 ?
Brown et al., 2001 - HATS ?
CDPRG 1975 ?
Chailurkit et al., 2010 ?
Chandra et al., 1992 +
Chapuy et al., 1992 ?
Chapuy et al., 2002 - Decalyos II ?
Cherniack et al., 2011 ?
Chylack et al.., 2002 - REACT +
CLIPS Group 2007 ?
Coburn et al., 2004 ?
Cole et al., 2007 - AFPPS +
Collins et al., 2003 +
Cook et al., 2007 - WACS +
Cooper et al., 2003 ?
Corless et al., 1985 +
Correa et al., 2000 +
Coyne et al., 2006 +
CTNS Study Group 2008 ?
Daly et al., 2008 +
Dawson-Hughes et al., 1997 ?
de Gaetano et al., 2001 - PPP +
de la Maza et al., 1995 +
Delanaye et al., 2013 ?
Desnuelle et al., 2001 - ALSRT ?
de Waart et al., 2001 ?
de Zeeuw et al., 2010 - VITAL +
Dukas et al., 2004 ?
Durga et al., 2007 - FACIT ?
Ebbing et al., 2008 - WENBIT +
Flicker et al., 2005 +
Frazao et al., 2000 ?
Galan et al., 2010 - SU.FOL.OM3 +
Gallagher et al., 1990 +
Gallagher et al., 2001 - STOP IT –
Gallagher et al., 2012 - VIDOS +
Gillilan et al., 1997 ?
Girodon et al., 1997 ?
Girodon et al., 1999 - MIN.VIT.AOX ?
GISSI-Prevenzione Investigators 1999 +
Glendenning et al., 2012 +
Graat et al., 2002 +
Grady et al., 1991 ?
Graf et al., 2005 ?
Grant et al., 2005 - RECORD +
Greenberg et al., 1996 - SCPS +
Green et al., 1999 - NSCPT +
Grimnes et al. 2011 ?
Gupta et al., 2016 +
Guyton et al., 2008 ?
Hamdy et al., 1995 ?
Harwood et al., 2004 - NoNOF +
Heinz et al., 2010 ?
Hennekens et al., 1996 - PHS +
Hercberg et al., 2010 - SU.VI.MAX ?
Hewitt et al., 2013 +
Hodis et al., 2002 - VEAPS +
Hodis et al., 2009 - BVAIT +
Hoffman et al., 1999 ?
House et al., 2010- DIVINe +
HPS2-THRIVE Collaborative Group 2014 +
HPS Collaborative Group 2002 +
Huo et al., 2015 - CSPPT +
Imasa et al., 2009 ?
Inkovaara et al., 1983 –
Jacobson et al., 2000 ?
Jamison et al., 2007 - HOST +
Janssen et al.,2010 ?
Komulainen et al., 1999 - OSTPRE ?
Korpela et al., 1989 ?
Krieg et al., 1999 ?
Lacroix et al., 2009 - WHI +
Lamas et al., 2013 - TACT +
Lange et al., 2004 ?
Lappe et al., 2007 +
Lappe et al., 2017 +
Larsen et al., 2004 +
Law et al., 2006 +
Lee et al., 1999 - WHS +
Lee et al., 2005 - WHS +
Lehouck et al., 2012 +
Leppälä et al., 2000 - ATBC ?
Lewis et al., 2011 - CAIFOS +
Liem et al., 2003 +
Liem et al., 2004 ?
Li et al., 1993 - NIT 2 ?
Limburg et al., 2005 +
Lippman et al., 2009 - SELECT ?
Lips et al., 1996 +
Liu et al., 2007 +
Logan et al., 2008 - UKCAP +
Lonn et al., 2005 - HOPE & HOPE TOO ?
Lonn et al., 2006 - HOPE TOO ?
Lyons et al., 2007 +
Ma et al., 2012 - SIT ?
Magliano et al., 2006 - MAVET ?
Manning et al., 2013 +
Manuel-Y-Kennoy et al., 2004 - DATOR ?
Marshall et al., 2011 ?
McKeown-Eyssen et al., 1988 ?
McNeil et al., 2004 - VECAT ?
Meier et al., 2004 ?
Meydani et al., 2004 +
Meyer et al., 2002 –
Milman et al., 2008 - ICARE +
Moon et al., 1997 - SKICAP AK ?
Mooney et al., 2005 +
Omenn et al., 1996 - CARET ?
Ooms et al., 1995 ?
Ott et al., 1989 ?
Petersen et al., 2005 - ADCS 2 +
Pike et al., 1995 +
Plummer et al., 2007 +
Ponikowski et al. 2015 - CONFIRM-HF +
Porthouse et al., 2005 +
Potena et al., 2005 +
Prentice et al., 2013 - WHI CaD +
Prince et al., 2008 +
Punthakee et al., 2012 - TIDE +
Reid et al., 2008 +
Righetti et al., 2003 ?
Salonen et al., 2000 - ASAP ?
Salovaara et al., 2010 - OSTRE-FPS +
Sanders et al., 2010 - Vital D +
Sang et al., 2009 ?
Sano et al., 1997 - ADCS 1 ?
Sanyal et al., 2010 - PIVENS ?
Saposnik et al., 2009 - HOPE 2 +
Sato et al., 1997 ?
Sato et al., 1999 +
Sato et al., 2005 +
Schleitoff et al., 2006 +
Schnyder et al., 2002 - The Swiss Heart Study ?
Sesso et al., 2008 - PHS II +
Sesso et al., 2012 - PHS II +
Shoulson et al., 1998 - DATATOP ?
Smith et al., 2007 ?
Steiner et al., 1995 ?
Stephens et al., 1996 - CHAOS +
Stranges et al., 2006 - NPC ?
Taylor et al., 2004 - ARBITER 2 +
Tornwall et al., 2004 - ATBC ?
Trivedi et al., 2003 ?
Van Dijk et al., 2015 - B-PROOF +
Van Wijngaarden et al., 2014 - B-PROOF +
Vianna et al., 2007 ?
Virtamo et al., 1998 - ATBC ?
Virtamo et al., 2003 - ATBC ?
VITATOPS Trial Study Group 2010 +
Wang et al., 2014 - OPERA +
Wang et al., 2014 - PHS II +
Wang et al., 2015 +
Waters et al., 2002 - WAVE +
Witham et al., 2013 +
Witham et al., 2013 - VitDISH +
Wluka et al., 2002 +
Wu et al., 2009 - NHS/HPFS +
You et al., 2001 - SIT ?
Zhu et al., 2008 - CAIFOS +
Zoungas et al., 2006 - ASFAST ?
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
+
?
?
+
+
+
+
?
?
?
?
?
+
?
+
?
+
+
+
?
?
?
?
+
?
?
?
?
+
+
+
+
+
+
?
?
?
+
?
+
?
+
?
?
?
?
+
+
+
+
+
+
+
+
+
+
?
?
?
+
+
+
?
?
+
?
+
+
–
+
?
+
+
?
?
?
?
?
?
+
+
+
+
+
+
?
+
+
?
?
?
?
+
?
–
–
–
?
?
?
+
?
+
?
?
?
+
+
+
+
+
+
+
+
?
+
+
?
?
?
?
?
+
?
+
?
?
+
+
?
+
?
+
+
+
+
?
+
+
+
+
+
?
+
+
?
+
+
?
?
+
?
?
?
?
?
+
?
?
+
+
?
?
+
+
?
?
?
+
?
?
+
+
?
+
?
+
+
+
?
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(per
form
ance
bia
s)
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
–
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Inco
mpl
ete
outc
ome
data
(attr
ition
bia
s)
+
+
+
+
+
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
?
+
+
?
+
+
?
+
+
+
+
+
?
+
+
+
–
+
+
+
+
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
+
?
+
+
+
+
+
?
+
?
?
+
+
+
+
?
?
+
+
+
+
+
?
+
?
+
?
?
?
–
+
?
+
?
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
+
?
+
+
?
?
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
?
+
?
+
?
+
?
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
+
+
+
Sel
ectiv
e re
porti
ng (r
epor
ting
bias
)
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Albert et al., 2008 - WAFACS ?
Aloia et al., 1988 ?
Aloia et al., 2005 +
Alvarez et al., 2012 ?
Anker et al., 2009 - FAIR-HF +
AREDS Research Group 2001 +
Armitage et al., 2010 - SEARCH +
Avenell et al., 2005 - MAVIS +
Baeksgaard et al., 1998 ?
Bairati et al., 2006 +
Baker et al., 2002 ?
Baron et al., 1999 - CPPS +
Bjorkman et al., 2008 ?
Boaz et al., 2000 - SPACE +
Boden et al., 2011 - AIM HIGH ?
Bogden et al., 1994 +
Bolland et al., 2008 +
Bolton-Smith et al., 2007 +
Bonaa et al., 2006 - NORVIT ?
Brazier et al., 2005 ?
Brohult et al., 1973 ?
Brown et al., 2001 - HATS ?
CDPRG 1975 ?
Chailurkit et al., 2010 ?
Chandra et al., 1992 +
Chapuy et al., 1992 ?
Chapuy et al., 2002 - Decalyos II ?
Cherniack et al., 2011 ?
Chylack et al.., 2002 - REACT +
CLIPS Group 2007 ?
Coburn et al., 2004 ?
Cole et al., 2007 - AFPPS +
Collins et al., 2003 +
Cook et al., 2007 - WACS +
Cooper et al., 2003 ?
Corless et al., 1985 +
Correa et al., 2000 +
Coyne et al., 2006 +
CTNS Study Group 2008 ?
Daly et al., 2008 +
Dawson-Hughes et al., 1997 ?
de Gaetano et al., 2001 - PPP +
de la Maza et al., 1995 +
Delanaye et al., 2013 ?
Desnuelle et al., 2001 - ALSRT ?
de Waart et al., 2001 ?
de Zeeuw et al., 2010 - VITAL +
Dukas et al., 2004 ?
Durga et al., 2007 - FACIT ?
Ebbing et al., 2008 - WENBIT +
Flicker et al., 2005 +
Frazao et al., 2000 ?
Galan et al., 2010 - SU.FOL.OM3 +
Gallagher et al., 1990 +
Gallagher et al., 2001 - STOP IT –
Gallagher et al., 2012 - VIDOS +
Gillilan et al., 1997 ?
Girodon et al., 1997 ?
Girodon et al., 1999 - MIN.VIT.AOX ?
GISSI-Prevenzione Investigators 1999 +
Glendenning et al., 2012 +
Graat et al., 2002 +
Grady et al., 1991 ?
Graf et al., 2005 ?
Grant et al., 2005 - RECORD +
Greenberg et al., 1996 - SCPS +
Green et al., 1999 - NSCPT +
Grimnes et al. 2011 ?
Gupta et al., 2016 +
Guyton et al., 2008 ?
Hamdy et al., 1995 ?
Harwood et al., 2004 - NoNOF +
Heinz et al., 2010 ?
Hennekens et al., 1996 - PHS +
Hercberg et al., 2010 - SU.VI.MAX ?
Hewitt et al., 2013 +
Hodis et al., 2002 - VEAPS +
Hodis et al., 2009 - BVAIT +
Hoffman et al., 1999 ?
House et al., 2010- DIVINe +
HPS2-THRIVE Collaborative Group 2014 +
HPS Collaborative Group 2002 +
Huo et al., 2015 - CSPPT +
Imasa et al., 2009 ?
Inkovaara et al., 1983 –
Jacobson et al., 2000 ?
Jamison et al., 2007 - HOST +
Janssen et al.,2010 ?
Komulainen et al., 1999 - OSTPRE ?
Korpela et al., 1989 ?
Krieg et al., 1999 ?
Lacroix et al., 2009 - WHI +
Lamas et al., 2013 - TACT +
Lange et al., 2004 ?
Lappe et al., 2007 +
Lappe et al., 2017 +
Larsen et al., 2004 +
Law et al., 2006 +
Lee et al., 1999 - WHS +
Lee et al., 2005 - WHS +
Lehouck et al., 2012 +
Leppälä et al., 2000 - ATBC ?
Lewis et al., 2011 - CAIFOS +
Liem et al., 2003 +
Liem et al., 2004 ?
Li et al., 1993 - NIT 2 ?
Limburg et al., 2005 +
Lippman et al., 2009 - SELECT ?
Lips et al., 1996 +
Liu et al., 2007 +
Logan et al., 2008 - UKCAP +
Lonn et al., 2005 - HOPE & HOPE TOO ?
Lonn et al., 2006 - HOPE TOO ?
Lyons et al., 2007 +
Ma et al., 2012 - SIT ?
Magliano et al., 2006 - MAVET ?
Manning et al., 2013 +
Manuel-Y-Kennoy et al., 2004 - DATOR ?
Marshall et al., 2011 ?
McKeown-Eyssen et al., 1988 ?
McNeil et al., 2004 - VECAT ?
Meier et al., 2004 ?
Meydani et al., 2004 +
Meyer et al., 2002 –
Milman et al., 2008 - ICARE +
Moon et al., 1997 - SKICAP AK ?
Mooney et al., 2005 +
Omenn et al., 1996 - CARET ?
Ooms et al., 1995 ?
Ott et al., 1989 ?
Petersen et al., 2005 - ADCS 2 +
Pike et al., 1995 +
Plummer et al., 2007 +
Ponikowski et al. 2015 - CONFIRM-HF +
Porthouse et al., 2005 +
Potena et al., 2005 +
Prentice et al., 2013 - WHI CaD +
Prince et al., 2008 +
Punthakee et al., 2012 - TIDE +
Reid et al., 2008 +
Righetti et al., 2003 ?
Salonen et al., 2000 - ASAP ?
Salovaara et al., 2010 - OSTRE-FPS +
Sanders et al., 2010 - Vital D +
Sang et al., 2009 ?
Sano et al., 1997 - ADCS 1 ?
Sanyal et al., 2010 - PIVENS ?
Saposnik et al., 2009 - HOPE 2 +
Sato et al., 1997 ?
Sato et al., 1999 +
Sato et al., 2005 +
Schleitoff et al., 2006 +
Schnyder et al., 2002 - The Swiss Heart Study ?
Sesso et al., 2008 - PHS II +
Sesso et al., 2012 - PHS II +
Shoulson et al., 1998 - DATATOP ?
Smith et al., 2007 ?
Steiner et al., 1995 ?
Stephens et al., 1996 - CHAOS +
Stranges et al., 2006 - NPC ?
Taylor et al., 2004 - ARBITER 2 +
Tornwall et al., 2004 - ATBC ?
Trivedi et al., 2003 ?
Van Dijk et al., 2015 - B-PROOF +
Van Wijngaarden et al., 2014 - B-PROOF +
Vianna et al., 2007 ?
Virtamo et al., 1998 - ATBC ?
Virtamo et al., 2003 - ATBC ?
VITATOPS Trial Study Group 2010 +
Wang et al., 2014 - OPERA +
Wang et al., 2014 - PHS II +
Wang et al., 2015 +
Waters et al., 2002 - WAVE +
Witham et al., 2013 +
Witham et al., 2013 - VitDISH +
Wluka et al., 2002 +
Wu et al., 2009 - NHS/HPFS +
You et al., 2001 - SIT ?
Zhu et al., 2008 - CAIFOS +
Zoungas et al., 2006 - ASFAST ?
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
+
?
?
+
+
+
+
?
?
?
?
?
+
?
+
?
+
+
+
?
?
?
?
+
?
?
?
?
+
+
+
+
+
+
?
?
?
+
?
+
?
+
?
?
?
?
+
+
+
+
+
+
+
+
+
+
?
?
?
+
+
+
?
?
+
?
+
+
–
+
?
+
+
?
?
?
?
?
?
+
+
+
+
+
+
?
+
+
?
?
?
?
+
?
–
–
–
?
?
?
+
?
+
?
?
?
+
+
+
+
+
+
+
+
?
+
+
?
?
?
?
?
+
?
+
?
?
+
+
?
+
?
+
+
+
+
?
+
+
+
+
+
?
+
+
?
+
+
?
?
+
?
?
?
?
?
+
?
?
+
+
?
?
+
+
?
?
?
+
?
?
+
+
?
+
?
+
+
+
?
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(per
form
ance
bia
s)
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
–
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Inco
mpl
ete
outc
ome
data
(attr
ition
bia
s)
+
+
+
+
+
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
?
+
+
?
+
+
?
+
+
+
+
+
?
+
+
+
–
+
+
+
+
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
+
?
+
+
+
+
+
?
+
?
?
+
+
+
+
?
?
+
+
+
+
+
?
+
?
+
?
?
?
–
+
?
+
?
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
+
?
+
+
?
?
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
?
+
+
?
+
?
+
?
+
?
+
+
+
+
?
+
?
+
+
+
?
+
+
+
+
?
?
+
+
?
?
?
+
+
?
?
+
+
+
Sel
ectiv
e re
porti
ng (r
epor
ting
bias
)
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Random Effects
Comparison RCTs N Events RR (95% CIs) RR (95% CIs) I2 p -value
Favours vitamin D/ Favours control/ Negative association Positive association
Pooled Effect Estimates
0.0 0.5 1.0 1.5 2.0
Supplementary Figure 3. Summary of the pooled effect estimates of RCTs assessing the relationship between vitamin D supplementation and CVD and all-cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsGallagher et al., 2001 - STOP IT 4 123 3 123 0.4% 1.33 [0.30, 5.83]Trivedi et al., 2003 477 1,345 503 1,341 96.6% 0.95 [0.86, 1.04]Sanders et al., 2010 - Vital D 17 1,131 13 1,127 1.9% 1.30 [0.64, 2.67]Punthakee et al., 2012 - TIDE 2 607 3 614 0.3% 0.67 [0.11, 4.02]Witham et al., 2013 5 39 4 36 0.6% 1.15 [0.34, 3.96]Wang et al., 2014 - OPERA 0 30 6 30 0.1% 0.08 [0.00, 1.31]
Study or SubgroupGallagher et al., 2001 - STOP ITTrivedi et al., 2003Sanders et al., 2010 - Vital DPunthakee et al., 2012 - TIDEWitham et al., 2013Wang et al., 2014 - OPERA
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 4.23, df = 5 (P = 0.52); I² = 0%Test for overall effect: Z = 1.02 (P = 0.31)
Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours vitamin D Favours control
Supplementary Figure 4. Forest plot of vitamin D supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
78
Subgroup and Study, Year Events Total Events TotalRCTsInkovaara et al., 1983 0 45 1 42 21.4% 0.31 [0.01, 7.44]de Zeeuw et al., 2010 - VITAL 1 95 0 93 21.2% 2.94 [0.12, 71.20]Witham et al., 2013 - VitDISH 2 80 2 79 57.5% 0.99 [0.14, 6.84]
Study or SubgroupInkovaara et al., 1983de Zeeuw et al., 2010 - VITALWitham et al., 2013 - VitDISH
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 0.96, df = 2 (P = 0.62); I² = 0%Test for overall effect: Z = 0.04 (P = 0.97)
Events012
3
Total459580
220
Events102
3
Total429379
214
Weight21.4%21.2%57.5%
100.0%
M-H, Random, 95% CI0.31 [0.01, 7.44]
2.94 [0.12, 71.20]0.99 [0.14, 6.84]
0.97 [0.22, 4.22]
Year198320102013
Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.001 0.1 1 10 1000Favours Vitamin D Favours Control
Supplementary Figure 5. Forest plot of vitamin D supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsAloia et al., 1988 1 12 1 15 0.3% 1.25 [0.09, 17.98]Ott et al., 1989 1 43 0 43 0.2% 3.00 [0.13, 71.65]Komulainen et al., 1999 - OSTPRE 1 112 0 115 0.2% 3.08 [0.13, 74.81]Trivedi et al., 2003 224 1,345 233 1,341 74.2% 0.96 [0.81, 1.13]Coburn et al., 2004 0 27 2 28 0.2% 0.21 [0.01, 4.13]Grant et al., 2005 - RECORD* 78 2649 84 2643 22.5% 0.93 [0.68, 1.25]Lappe et al., 2007† 3 445 2 446 0.7% 1.50 [0.25, 8.95]Prince et al., 2008‡ 2 151 3 151 0.7% 0.67 [0.11, 3.93]de Zeeuw et al., 2010 - VITAL 2 95 0 93 0.2% 4.90 [0.24, 100.62]Cherniack et al., 2011 1 23 1 23 0.3% 1.00 [0.07, 15.04]Punthakee et al., 2012 - TIDE 1 607 1 614 0.3% 1.01 [0.06, 16.14]Wang et al., 2014 - OPERA 0 30 2 30 0.2% 0.20 [0.01, 4.00]
Study or SubgroupAloia et al., 1988Ott et al., 1989Komulainen et al., 1999 - OSTPRETrivedi et al., 2003Coburn et al., 2004Grant et al., 2005 - RECORDLappe et al., 2007Prince et al., 2008de Zeeuw et al., 2010 - VITALCherniack et al., 2011Punthakee et al., 2012 - TIDEWang et al., 2014 - OPERA
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 4.68, df = 11 (P = 0.95); I² = 0%Test for overall effect: Z = 0.64 (P = 0.52)
Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours vitamin D Favours control
Supplementary Figure 6. Forest plot of vitamin D supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. *Lappe et al., 2007 - data taken from meta-analysis; Bolland et al., 2014; †Prince et al., 2008 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
79
Subgroup and Study, Year Events Total Events TotalRCTsInkovaara et al., 1983 12 45 5 42 3.3% 2.24 [0.86, 5.82]Gallagher et al., 2001 - STOP IT 4 123 3 123 1.4% 1.33 [0.30, 5.83]Trivedi et al., 2003 105 1,345 101 1,341 44.2% 1.04 [0.80, 1.35]Grant et al., 2005 - RECORD* 118 2,649 104 2,643 45.7% 1.13 [0.87, 1.47]Lappe et al., 2007† 6 446 5 445 2.2% 1.20 [0.37, 3.89]Prince et al., 2008‡ 3 151 3 151 1.2% 1.00 [0.21, 4.88]de Zeeuw et al., 2010 - VITAL 1 95 0 93 0.3% 2.94 [0.12, 71.20]Gallagher et al., 2012 - VIDOS 1 20 0 21 0.3% 3.14 [0.14, 72.92]Punthakee et al., 2012 - TIDE 1 607 1 614 0.4% 1.01 [0.06, 16.14]Witham et al., 2013 - VitDISH 3 80 1 79 0.6% 2.96 [0.31, 27.88]Wang et al., 2014 - OPERA 0 30 2 30 0.3% 0.20 [0.01, 4.00]
Study or SubgroupInkovaara et al., 1983Gallagher et al., 2001 - STOP ITTrivedi et al., 2003Grant et al., 2005 - RECORDLappe et al., 2007Prince et al., 2008de Zeeuw et al., 2010 - VITALGallagher et al., 2012 - VIDOSPunthakee et al., 2012 - TIDEWitham et al., 2013 - VitDISHWang et al., 2014 - OPERA
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 5.22, df = 10 (P = 0.88); I² = 0%Test for overall effect: Z = 1.30 (P = 0.20)
Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours vitamin D Favours control
Supplementary Figure 7. Forest plot of vitamin D supplementation and stroke risk. M-H, Manthel-Haenszel. *Grant et al., 2005 - data taken from meta-analysis Bolland et al., 2014 †Lappe et al., 2007 - data taken from meta-analysis Bolland et al., 2014 ‡Prince et al., 2008 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsTrivedi et al., 2003 101 1,345 117 1,341 99.4% 0.86 [0.67, 1.11]Punthakee et al., 2012 - TIDE 0 607 1 614 0.6% 0.34 [0.01, 8.26]
Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours Vitamin D Favours Control
Supplementary Figure 8. Forest plot of vitamin D supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
80
Subgroup and Study, Year Events Total Events TotalRCTsInkovaara et al., 1983 0 45 1 42 35.9% 0.31 [0.01, 7.44]Frazao et al., 2000 1 71 2 67 64.1% 0.47 [0.04, 5.08]
Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.005 0.1 1 10 200Favours Vitamin D Favours Control
Supplementary Figure 9. Forest plot of vitamin D supplementation and CHD mortality risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsOtt et al., 1989 0 43 1 43 1.6% 0.33 [0.01, 7.96]Trivedi et al., 2003 42 1,345 49 1,341 95.3% 0.85 [0.57, 1.28]Coburn et al., 2004 0 27 1 28 1.6% 0.35 [0.01, 8.12]Cherniack et al., 2011 1 23 0 23 1.6% 3.00 [0.13, 70.02]
Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours Vitamin D Favours Control
Supplementary Figure 10. Forest plot of vitamin D supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
81
Subgroup and Study, Year Events Total Events TotalRCTsInkovaara et al., 1983 4 45 2 42 9.4% 1.87 [0.36, 9.67]Trivedi et al., 2003 28 1,345 26 1,341 90.6% 1.07 [0.63, 1.82]
Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours Vitamin D Favours Control
Supplementary Figure 11. Forest plot of vitamin D supplementation and stroke mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Risk RatioM-H, Random, 95% CI in all-cause mortality risk
Study or SubgroupBrohult et al., 1973Inkovaara et al., 1983Corless et al., 1985Ott et al., 1989Gallagher et al., 1990Grady et al., 1991Hamdy et al., 1995Ooms et al., 1995Lips et al., 1996Sato et al., 1997Sato et al., 1999Komulainen et al., 1999 - OSTPREFrazao et al., 2000Gallagher et al., 2001 - STOP ITMeyer et al., 2002Trivedi et al., 2003Cooper et al., 2003Harwood et al., 2004 - NONOFDukas et al., 2004Sato et al., 2005Grant et al., 2005 - RECORDAloia et al., 2005Flicker et al., 2005Coyne et al., 2006Law et al., 2006Schleithoff et al., 2006Smith et al., 2007Lyons et al., 2007Prince et al., 2008Bjorkman et al., 2008Janssen et al.,2010Sanders et al., 2010 - Vital Dde Zeeuw et al., 2010 - VITALCherniack et al., 2011Grimnes et al. 2011Punthakee et al., 2012 - TIDEAlvarez et al., 2012Lehouck et al., 2012Glendenning et al., 2012Witham et al., 2013Hewitt et al., 2013Delanaye et al., 2013Witham et al., 2013 - VitDISH
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 28.51, df = 42 (P = 0.94); I² = 0%Test for overall effect: Z = 0.55 (P = 0.58)
Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours Vitamin D Favours Control
Supplementary Figure 12. Forest plot of vitamin D supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. *Brohult et al., 1973 - data taken from meta-analysis Bjelakovic et al., 2014 †Gallagher et al., 1990 - data taken from meta-analysis Avenell et al., 2014 ‡Grady et al., 1991 - data taken from meta-analysis Chowdhury et al., 2014 §Ooms et al., 1995 - data taken from meta-analysis Zheng et al., 2013 ǁSato et al., 1999 - data taken from meta-analysis Bjelakovic et al., 2014 ¶Cooper et al., 2003 - data taken from meta-analysis Zheng et al., 2013**Sato et al., 2005 - data taken from meta-analysis Chowdhury et al., 2014, Bjelakovic et al., 2014 ††Aloia et al., 2005 - data taken from meta-analysis Zheng et al., 2013 ‡‡Smith et al., 2007 - data taken from meta-analysis Bjelakovic et al., 2014 §§Jansen et al., 2010 - data taken from meta-analysis Avenell et al., 2014. ǁǁAvenell et al., 2012 - Data provided included results from the RCT and follow-up period combined. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P <
83
0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Supplementary Figure 13. Funnel plot of vitamin D supplementation and CVD and all-cause mortality outcomes: (A) MI risk, (B) stroke risk, and (C) all-cause mortality risk. The vertical line represents the pooled effect estimate expressed as a RR. Dashed lines represent pseudo-95% confidence intervals (CI). The circles represent risk estimates for each study, and the horizontal lines represent standard errors of the RR. We were unable to test for funnel plot asymmetry for other CVD outcomes (<10 RCTs).
Random Effects
Comparison RCTs N Events RR (95% CIs) RR (95% CIs) I2 p -value
Favours vitamin A / Favours control/ Negative association Positive association
Pooled Effect Estimates
0.5 1.0 1.5
84
Supplementary Figure 14. Summary of the pooled effect estimates of RCTsassessing the relationship vitamin A supplementation and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable or not available. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTs Moon et al., 1997- SKICAP-AK 24 1,157 24 1,140 100.0% 0.99 [0.56, 1.72]
Total (95% CI) 1,157 1,140 100% 0.99 [0.56, 1.72]Total events 24 24 Heterogeneity: Not applicable 0.02 0.1 1 10 50Test for overall effect: Z = 0.05 (P = 0.96) Favours vitamin A Favours control
Risk RatioM-H, Random, 95% CI in all-cause mortality
Vitamin A ControlWeight
Risk RatioM-H, Random, 95% CI
Study or SubgroupMoon et al., 1997- SKICAP-AK
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.05 (P = 0.96)
Events24
24
Total1157
1157
Events24
24
Total1140
1140
Weight100.0%
100.0%
M-H, Random, 95% CI0.99 [0.56, 1.72]
0.99 [0.56, 1.72]
Year1997
Vitamin A Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours Vitamin A Favours Control
Supplementary Figure 15. Forest plot of vitamin A supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
85
Random Effects
Comparison RCTs N Events RR (95% CIs) RR (95% CIs) I2 p -value
Favours beta-carotene/ Favours control/ Negative association Positive association
0.0 0.5 1.0 1.5 2.0
Supplementary Figure 16. Summary of the pooled effect estimates of RCT assessing the relationship between beta-carotene supplementation and CVD and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable or not available. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsHennekens et al., 1996 - PHS 967 11,036 972 11,035 66.6% 0.99 [0.91, 1.08]Lee et al., 1999 - WHS 74 19,939 65 19,937 4.4% 1.14 [0.82, 1.59]Cook et al., 2007 - WACS 435 4,084 399 4,087 29.0% 1.09 [0.96, 1.24]
Experimental Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.7 0.85 1 1.2 1.5Favours beta-carotene Favours control
Supplementary Figure 17. Forest plot of beta-carotene supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
86
Subgroup and Study, Year Events Total Events TotalRCTsTornwall et al., 2004 - ATBC 548 6,821 534 6,849 50.8% 1.03 [0.92, 1.16]Cook et al., 2007 - WACS 500 4,084 499 4,087 49.2% 1.00 [0.89, 1.13]
Experimental Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.7 0.85 1 1.2 1.5Favours beta-carotene Favours control
Supplementary Figure 18. Forest plot of beta-carotene supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsHennekens et al., 1996 - PHS 468 11,036 489 11,035 52.0% 0.96 [0.85, 1.08]Tornwall et al., 2004 - ATBC 314 6,821 296 6,849 33.2% 1.07 [0.91, 1.24]Cook et al., 2007 - WACS 135 4,084 139 4,087 14.8% 0.97 [0.77, 1.23]
Experimental Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.85 0.9 1 1.1 1.2Favours beta-carotene Favours control
Supplementary Figure 19. Forest plot of beta-carotene supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
87
Subgroup and Study, Year Events Total Events TotalRCTsHennekens et al., 1996 - PHS 367 11,036 382 11,035 36.6% 0.96 [0.83, 1.11]Leppala et al., 2000 - ATBC 554 14,246 503 14,273 44.3% 1.10 [0.98, 1.24]Cook et al., 2007 - WACS 161 4,084 137 4,087 19.2% 1.18 [0.94, 1.47]
Experimental Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.7 0.85 1 1.2 1.5Favours beta-carotene Favours control
Supplementary Figure 20. Forest plot of beta-carotene supplementation and stroke risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsHennekens et al., 1996 - PHS 338 11,036 313 11,035 54.2% 1.08 [0.93, 1.26]Greenberg et al., 1996 - SCPS 68 913 59 892 11.0% 1.13 [0.80, 1.58]Green et al., 1999 - NSCPT 6 801 12 820 1.3% 0.51 [0.19, 1.36]Cook et al., 2007 - WACS 211 4,084 184 4,087 33.4% 1.15 [0.95, 1.39]
Experimental Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours beta-carotene Favours control
Supplementary Figure 21. Forest plot of beta-carotene supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
88
Subgroup and Study, Year Events Total Events TotalRCTsTornwall et al., 2004 - ATBC 234 6,821 238 6,849 100% 0.99 [0.83, 1.18]
Total (95% CI) 6,821 6,849 100% 0.99 [0.83, 1.18]Total events 234 238Heterogeneity: Not applicableTest for overall effect: Z = 0.14 (P = 0.89) 0.2 0.5 1 1.5 2.0
Favours beta-carotene Favours control
Beta-carotene ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in CHD mortality risk
Study or SubgroupTornwall et al., 2004 - ATBC
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.14 (P = 0.89)
Events234
234
Total6821
6821
Events238
238
Total6849
6849
Weight100.0%
100.0%
M-H, Random, 95% CI0.99 [0.83, 1.18]
0.99 [0.83, 1.18]
Year2004
Experimental Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours beta-carotene Favours control
Supplementary Figure 22. Forest plot of beta-carotene supplementation and CHD mortality risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsCook et al., 2007 - WACS 10 4,084 24 4,087 100.0% 0.42 [0.20, 0.87]
Total (95% CI) 4,084 4,087 100% 0.42 [0.20, 0.87]Total events 10 24Heterogeneity: Not applicable 0.05 0.2 1 5 20 Test for overall effect: Z = 2.33 (P = 0.02) Favours beta-carotene Favours control
Beta-carotene ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in MI mortality risk
Study or SubgroupCook et al., 2007 - WACS
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 2.33 (P = 0.02)
Events10
10
Total4084
4084
Events24
24
Total4087
4087
Weight100.0%
100.0%
M-H, Random, 95% CI0.42 [0.20, 0.87]
0.42 [0.20, 0.87]
Year2007
Experimental Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours beta-carotene Favours control
Supplementary Figure 23. Forest plot of beta-carotene supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLeppala et al., 2000 - ATBC 82 14,246 78 14,273 63.5% 1.05 [0.77, 1.43]Cook et al., 2007 - WACS 22 4,084 11 4,087 36.5% 2.00 [0.97, 4.12]
Total (95% CI) 18,330 18,360 100% 1.33 [0.73, 2.44]Total events 104 89Heterogeneity: Tau² = 0.13; Chi² = 2.57, df = 1 (P = 0.11); I² = 61% 0.2 0.5 1 2 5 Test for overall effect: Z = 0.93 (P = 0.35) Favours beta-carotene Favours control
Beta-carotene ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in stroke mortality risk
Study or SubgroupLeppala et al., 2000 - ATBCCook et al., 2007 - WACS
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.13; Chi² = 2.57, df = 1 (P = 0.11); I² = 61%Test for overall effect: Z = 0.93 (P = 0.35)
Experimental Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours beta-carotene Favours control
Supplementary Figure 24. Forest plot of beta-carotene supplementation and stroke mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
89
Subgroup and Study, Year Events Total Events TotalRCTsGreenberg et al., 1996 - Skin Cancer Prevention 146 913 139 892 10.0% 1.03 [0.83, 1.27]Hennekens et al., 1996 - PHS 979 11,036 968 11,035 41.0% 1.01 [0.93, 1.10]Green et al., 1999 - NSCPT 11 820 21 801 1.0% 0.51 [0.25, 1.05]Correa et al., 2000 2 117 0 117 0.1% 5.00 [0.24, 103.03]Virtamo et al., 2003 - ATBC 919 7,282 851 7,287 39.5% 1.08 [0.99, 1.18]Cook 2007 - WACS 124 1,020 124 1,022 8.5% 1.00 [0.79, 1.27]
Risk RatioM-H, Random, 95% CI in all-cause mortality risk
Study or SubgroupGreenberg et al., 1996 - Skin Cancer PreventionHennekens et al., 1996 - PHSGreen et al., 1999 - NSCPTCorrea et al., 2000Virtamo et al., 2003 - ATBCCook 2007 - WACS
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 5.99, df = 5 (P = 0.31); I² = 17%Test for overall effect: Z = 0.90 (P = 0.37)
Supplementary Figure 25. Forest plot of beta-carotene supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Random Effects
Comparison RCTs N Events RR (95% CIs) RR (95% CIs) I2 p -value
Supplementary Figure 26. Summary of the pooled effect estimates of RCTs assessing the relationship between antioxidants supplementation and CVD and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
90
Subgroup and Study, Year Events Total Events TotalRCTsBrown et al., 2001 - HATS* 11 42 12 38 0.3% 0.83 [0.42, 1.65]HPS Collaborative Group 2002 2,306 10,269 2,312 10,267 63.7% 1.00 [0.95, 1.05]CLIPS Group 2007 16 185 11 181 0.3% 1.42 [0.68, 2.98]Plummer et al., 2007 1 990 - 990 0.0% 3.00 [0.12, 73.55]Sesso et al., 2008 - PHS II 310 3,656 316 3,653 7.3% 0.98 [0.84, 1.14]Lippman et al., 2009 - SELECT 1,041 8,904 1,050 8,910 25.3% 0.99 [0.92, 1.08]Hercberg et al., 2010 - SU.VI.MAX 137 6,377 143 6,364 3.1% 0.96 [0.76, 1.21]
Risk RatioM-H, Random, 95%CI in total CVD riskWeight
Study or SubgroupBrown et al., 2001 - HATSHPS Collaborative Group 2002CLIPS Group 2007Plummer et al., 2007Sesso et al., 2008 - PHS IILippman et al., 2009 - SELECTHercberg et al., 2010 - SU.VI.MAX
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 1.79, df = 6 (P = 0.94); I² = 0%Test for overall effect: Z = 0.29 (P = 0.77)
Antioxidants Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours antioxidants Favours control
Supplementary Figure 27. Forest plot of antioxidants supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. *Brown et al., 2001 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsTornwall et al., 2004 - ATBC 511 6,781 534 6,849 100.0% 0.97 [0.86, 1.09]
Total (95% CI) 6,781 6,849 100% 0.97 [0.86, 1.09]Total events 511 534 Heterogeneity: Not applicable 0.5 0.7 1 1.5 2Test for overall effect: Z = 0.57 (P = 0.57) Favours antioxidant Favours control
Antioxidant ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95%CI in total CHD risk
Study or SubgroupTornwall et al., 2004 - ATBC
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.57 (P = 0.57)
Events511
511
Total6781
6781
Events534
534
Total6849
6849
Weight100.0%
100.0%
M-H, Random, 95% CI0.97 [0.86, 1.09]
0.97 [0.86, 1.09]
Year2004
Antioxidants Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours antioxidants Favours control
Supplementary Figure 28. Forest plot of antioxidants supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence intervals, using the Manthel-Haenszel method with random effects model.
91
Subgroup and Study, Year Events Total Events TotalRCTsBrown et al., 2001 - HATS* 1 42 4 38 0.2% 0.23 [0.03, 1.94]HPS Collaborative Group 2002 464 10,269 467 10,267 51.6% 0.99 [0.88, 1.13]Waters et al., 2002 - WAVE 3 105 1 108 0.2% 3.09 [0.33, 29.19]Tornwall et al., 2004 - ATBC 289 6,781 296 6,849 32.3% 0.99 [0.84, 1.16]CLIPS Group 2007 9 185 4 181 0.6% 2.20 [0.69, 7.02]Sesso et al., 2008 - PHS II 133 3,656 144 3,653 15.2% 0.92 [0.73, 1.16]
Antioxidants Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours antioxidants Favours control
Supplementary Figure 29. Forest plot of antioxidants supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. *Brown et al., 2001 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLeppala et al., 2000 - ATBC 258 7,118 252 7,153 25.6% 1.03 [0.87, 1.22]Brown et al., 2001 - HATS* 2 42 2 38 0.2% 0.90 [0.13, 6.11]Waters et al., 2002 - WAVE 1 105 3 108 0.1% 0.34 [0.04, 3.24]HPS Collaborative Group 2002 511 10,269 518 10,267 52.4% 0.99 [0.88, 1.11]CLIPS Group 2007 6 185 5 181 0.5% 1.17 [0.36, 3.78]Sesso et al., 2008 - PHS II 104 3,656 113 3,653 10.8% 0.92 [0.71, 1.20]Lippman et al., 2009 - SELECT 111 8,904 100 8,910 10.3% 1.11 [0.85, 1.45]
Study or SubgroupLeppala et al., 2000 - ATBCBrown et al., 2001 - HATSWaters et al., 2002 - WAVEHPS Collaborative Group 2002CLIPS Group 2007Sesso et al., 2008 - PHS IILippman et al., 2009 - SELECT
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 2.09, df = 6 (P = 0.91); I² = 0%Test for overall effect: Z = 0.02 (P = 0.98)
Antioxidants Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours antioxidants Favours control
Supplementary Figure 30. Forest plot of antioxidants supplementation and stroke risk. M-H, Manthel-Haenszel. *Brown et al., 2001 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
92
Subgroup and Study, Year Events Total Events TotalRCTsBrown et al., 2001 - HATS - 42 1 38 0.1% 0.30 [0.01, 7.21]You et al., 2001 9 1,706 12 1,705 0.9% 0.75 [0.32, 1.77]Waters et al., 2002 - WAVE 4 105 2 108 0.2% 2.06 [0.38, 10.99]HPS Collaborative Group 2002 878 10,269 840 10,267 77.2% 1.05 [0.95, 1.14]CLIPS Group 2007 6 185 3 181 0.3% 1.96 [0.50, 7.71]Sesso et al., 2008 - PHS II 127 3,656 122 3,653 10.6% 1.04 [0.81, 1.33]Lippman et al., 2009 - SELECT 117 8,904 142 8,910 10.7% 0.82 [0.65, 1.05]
Risk RatioM-H, Random, 95%CI in CVD mortality risk
Study or SubgroupBrown et al., 2001 - HATSYou et al., 2001Waters et al., 2002 - WAVEHPS Collaborative Group 2002CLIPS Group 2007Sesso et al., 2008 - PHS IILippman et al., 2009 - SELECT
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 5.84, df = 6 (P = 0.44); I² = 0%Test for overall effect: Z = 0.45 (P = 0.65)
Antioxidants Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours antioxidants Favours control
Supplementary Figure 31. Forest plot of antioxidants supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsHPS Collaborative Group 2002 664 10,269 630 10,267 72.0% 1.05 [0.95, 1.17]Tornwall et al., 2004 - ATBC 222 6,781 238 6,849 28.0% 0.94 [0.79, 1.13]
Total (95% CI) 17,050 17,116 100% 1.02 [0.93, 1.13]Total events 886 868 Heterogeneity: Tau² = 0.00; Chi² = 1.11, df = 1 (P = 0.29); I² = 10% 0.5 0.7 1 1.5 2 Test for overall effect: Z = 0.42 (P = 0.68) Favours antioxidant Favours control
Antioxidant ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95%CI in CHD mortality risk
Study or SubgroupHPS 2002Tornwall et al., 2004 - ATBC
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 1.11, df = 1 (P = 0.29); I² = 10%Test for overall effect: Z = 0.42 (P = 0.68)
Antioxidants Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours antioxidants Favours control
Supplementary Figure 32. Forest plot of antioxidants supplementation and CHD mortality risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
93
Subgroup and Study, Year Events Total Events TotalRCTsChylack et al.., 2002 - REACT 2 149 1 148 32.6% 1.99 [0.18, 21.67]Mooney et al., 2005 1 142 0 142 18.3% 3.00 [0.12, 73.03]CLIPS Group 2007 2 185 2 181 49.1% 0.98 [0.14, 6.87]
Antioxidants Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours antioxidants Favours control
Supplementary Figure 33. Forest plot of antioxidants supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLeppala et al., 2000 - ATBC 46 7,118 34 7,153 25.4% 1.36 [0.87, 2.12]HPS Collaborative Group 2002 108 10,269 107 10,267 58.9% 1.01 [0.77, 1.32]CLIPS Group 2007 3 185 0 181 0.6% 6.85 [0.36, 131.67]Lippman et al., 2009 - SELECT 12 8,904 8 8,910 6.8% 1.50 [0.61, 3.67]Ma et al., 2012 10 1,677 14 1,688 8% 0.72 [0.32, 1.61]
Antioxidants Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours antioxidants Favours control
Supplementary Figure 34. Forest plot of antioxidants supplementation and stroke mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
94
Subgroup and Study, Year Events Total Events TotalRCTsMcKeown-Eyssen et al., 1988 4 96 3 89 0.10% 1.24 [0.28, 5.37]Omenn et al., 1996 - CARET 544 9,420 424 8,894 12.50% 1.21 [1.07, 1.37]Girodon et al., 1997* 7 21 7 20 0.40% 0.95 [0.41, 2.23]Girodon et al., 1999 - MIN.VIT.AOX† 55 181 51 182 2.6% 1.08 [0.79, 1.49]Salonen et al., 2000 - ASAP 1 130 1 130 0.0% 1.00 [0.06, 15.82]Correa et al., 2000 2 121 0 117 0.0% 4.84 [0.23, 99.67]Jacobson et al., 2000 0 57 1 55 0.0% 0.32 [0.01, 7.74]Brown et al., 2001 - HATS 0 42 1 38 0.0% 0.30 [0.01, 7.21]AREDS Research Group 2001 251 2,304 240 2,325 8.1% 1.06 [0.89, 1.25]HPS Collaborative Group 2002 1,446 10,269 1,389 10,267 22.8% 1.04 [0.97, 1.11]Waters et al., 2002 - WAVE 6 105 2 108 0.1% 3.09 [0.64, 14.95]Chylack et al.., 2002 - REACT 9 149 3 148 0.2% 2.98 [0.82, 10.79]Virtamo et al., 2003 - ATBC 932 7,278 851 7,287 18.6% 1.10 [1.01, 1.20]Mooney et al., 2005 1 142 0 142 0.0% 3.00 [0.12, 73.03]CLIPS Group 2007 7 185 4 181 0.2% 1.71 [0.51, 5.75]Plummer et al., 2007 16 990 11 990 0.5% 1.45 [0.68, 3.12]Cook et al., 2007 - WAC 133 1,020 124 1,022 4.8% 1.07 [0.85, 1.35]Lippman et al., 2009 - SELECT 359 8,904 382 8,910 10.4% 0.94 [0.82, 1.08]Hercberg et al., 2010 - SU.VI.MAX 77 6,377 99 6,364 3.1% 0.78 [0.58, 1.04]Ma et al., 2012 - SIT 82 1,706 101 1,705 3.3% 0.81 [0.61, 1.08]Wang et al., 2014 - PHS II 440 3,656 406 3,653 12.1% 1.08 [0.95, 1.23]
Risk RatioM-H, Random, 95%CI in all-cause mortality risk
Study or SubgroupMcKeown-Eyssen et al., 1988Omenn et al., 1996 - CARETGirodon et al., 1997Girodon et al., 1999 - MIN.VIT.AOXSalonen et al., 2000 - ASAPCorrea et al., 2000Jacobson et al., 2000Brown et al., 2001 - HATSAREDS Research Group 2001HPS Collaborative Group 2002Waters et al., 2002 - WAVEChylack et al.., 2002 - REACTVirtamo et al., 2003 - ATBCMooney et al., 2005CLIPS Group 2007Plummer et al., 2007Cook et al., 2007 - WACLippman et al., 2009 - SELECTHercberg et al., 2010 - SU.VI.MAXMa et al., 2012 - SITWang et al., 2014 - PHS II
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 24.05, df = 20 (P = 0.24); I² = 17%Test for overall effect: Z = 2.00 (P = 0.05)
Antioxidants Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours antioxidants Favours control
Supplementary Figure 35. Forest plot of antioxidants supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. *Girodon et al., 1997 - vitamin C, vitamin E, β-carotene, selenium and zinc vs placebo; †Girodon et al., 1999 - vitamin C, vitamin E, β-carotene, selenium and zinc vs placebo. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
95
Supplementary Figure 36. Funnel plot of antioxidants supplementation and all-cause mortality risk. The vertical line represents the pooled effect estimate expressed as a RR. Dashed lines represent pseudo-95% confidence intervals (CI). The circles represent risk estimates for each study, and the horizontal lines represent standard errors of the RR. We were unable to test for other CVD outcomes (<10 RCTs).
Favours vitamin E/ Favours control/ Negative association Positive association
Pooled Effect Estimates
0.0 0.5 1.0 1.5 2.0
Supplementary Figure 37. Summary of the pooled effect estimates of RCTs assessing the relationship between vitamin E supplementation and CVD and all-cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
96
Subgroup and Study, Year Events Total Events TotalRCTsGISSI-Prevenzione Investigators 1999 571 5,660 584 5,664 15.6% 0.98 [0.88, 1.09]Boaz et al., 2000 - SPACE 15 97 33 99 1.5% 0.46 [0.27, 0.80]de Gaetano et al., 2001 - PPP 56 2,231 53 2,264 3.1% 1.07 [0.74, 1.55]Hodis et al., 2002 - VEAPS 11 177 14 176 0.8% 0.78 [0.36, 1.67]Lonn et al., 2005 - HOPE & HOPE-TOO 1,022 4,761 985 4,780 19.3% 1.04 [0.96, 1.13]Lee et al., 2005 - WHS 482 19,937 517 19,939 14.2% 0.93 [0.82, 1.05]Cook et al., 2007 - WACS 399 4,083 435 4,088 13.6% 0.92 [0.81, 1.04]Sesso et al., 2008 - PHS II 310 3,659 316 3,653 11.6% 0.98 [0.84, 1.14]Milman et al., 2008 16 726 33 708 1.3% 0.47 [0.26, 0.85]Lippman et al., 2009 - SELECT 1,034 8,863 1,050 8,910 19% 0.99 [0.91, 1.07]
Study or SubgroupGISSI-Prevenzione Investigators 1999Boaz et al., 2000 - SPACEde Gaetano et al., 2001 - PPPHodis et al., 2002 - VEAPSLonn et al., 2005 - HOPE & HOPE-TOOLee et al., 2005 - WHSCook et al., 2007 - WACSMilman et al., 2008Sesso et al., 2008 - PHS IILippman et al., 2009 - SELECT
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 17.84, df = 9 (P = 0.04); I² = 50%Test for overall effect: Z = 1.21 (P = 0.23)
Vitamin E Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours vitamin E Favours control
Supplementary Figure 38. Forest plot of vitamin E supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsVirtamo et al., 1998 - ATBC 519 6,820 534 6,849 50.1% 0.98 [0.87, 1.10]Cook et al., 2007 - WACS 491 4,083 508 4,088 49.9% 0.97 [0.86, 1.09]
Vitamin E Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.7 0.85 1 1.2 1.5Favours Vitamin E Favours Control
Supplementary Figure 39. Forest plot of vitamin E supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
97
Subgroup and Study, Year Events Total Events TotalRCTsStephens et al., 1996 - CHAOS 32 1,035 54 967 7.8% 0.55 [0.36, 0.85]Virtamo et al., 1998 - ATBC 307 6,820 296 6,849 17.4% 1.04 [0.89, 1.22]Boaz et al., 2000 - SPACE 5 97 17 99 2.2% 0.30 [0.12, 0.78]de Gaetano et al., 2001 - PPP 22 2,231 25 2,264 5.3% 0.89 [0.51, 1.58]Hodis et al., 2002 - VEAPS 5 177 6 176 1.5% 0.83 [0.26, 2.67]Lee et al., 2005 - WHS 196 19,937 195 19,939 15.7% 1.01 [0.83, 1.22]Lonn et al., 2005 - HOPE & HOPE-TOO 724 4,761 686 4,780 19.7% 1.06 [0.96, 1.17]Cook et al., 2007 - WACS 131 4,083 143 4,088 14.2% 0.92 [0.73, 1.16]Milman et al., 2008 7 726 17 708 2.6% 0.40 [0.17, 0.96]Sesso et al., 2008 - PHS II 107 3,659 144 3,653 13.6% 0.74 [0.58, 0.95]
Study or SubgroupStephens et al., 1996 - CHAOSVirtamo et al., 1998 - ATBCBoaz et al., 2000 - SPACEde Gaetano et al., 2001 - PPPHodis et al., 2002 - VEAPSLee et al., 2005 - WHSLonn et al., 2005 - HOPE & HOPE-TOOCook et al., 2007 - WACSMilman et al., 2008Sesso et al., 2008 - PHS II
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.03; Chi² = 25.34, df = 9 (P = 0.003); I² = 64%Test for overall effect: Z = 1.81 (P = 0.07)
Vitamin E Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours Vitamin E Favours Control
Supplementary Figure 40. Forest plot of vitamin E supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsSteiner et al., 1995 3 52 6 48 0.6% 0.46 [0.12, 1.74]GISSI-Prevenzione Investigators 1999 83 5,660 95 5,664 10.4% 0.87 [0.65, 1.17]Boaz et al., 2000 - SPACE 5 97 6 99 0.8% 0.85 [0.27, 2.70]de Gaetano et al., 2001 - PPP 22 2,231 18 2,264 2.6% 1.24 [0.67, 2.31]Hodis et al., 2002 - VEAPS 0 177 2 176 0.1% 0.20 [0.01, 4.11]Lee et al., 2005 - WHS 241 19,937 246 19,939 22.1% 0.98 [0.82, 1.17]Lonn et al., 2005 - HOPE & HOPE-TOO 270 4,761 246 4,780 23.6% 1.10 [0.93, 1.30]Cook et al., 2007 - WACS 137 4,083 151 4,088 15.5% 0.91 [0.72, 1.14]Milman et al., 2008 6 726 11 708 1.1% 0.53 [0.20, 1.43]Sesso et al., 2008 - PHS II 133 3,659 113 3,653 13.70% 1.18 [0.92, 1.50]Lippman et al., 2009 - SELECT 70 8,863 92 8,910 9.40% 0.76 [0.56, 1.04]
Study or SubgroupSteiner et al., 1995GISSI-Prevenzione Investigators 1999Boaz et al., 2000 - SPACEde Gaetano et al., 2001 - PPPHodis et al., 2002 - VEAPSLee et al., 2005 - WHSLonn et al., 2005 - HOPE & HOPE-TOOCook et al., 2007 - WACSMilman et al., 2008Sesso et al., 2008 - PHS IILippman et al., 2009 - SELECT
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 11.78, df = 10 (P = 0.30); I² = 15%Test for overall effect: Z = 0.43 (P = 0.67)
Vitamin E Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours vitamin E Favours control
Supplementary Figure 41. Forest plot of vitamin E supplementation and stroke risk. * M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
98
Subgroup and Study, Year Events Total Events TotalRCTsStephens et al., 1996 - CHAOS 27 1,035 23 967 1.7% 1.10 [0.63, 1.90]GISSI-Prevenzione Investigators 1999 310 5,660 329 5,664 22.1% 0.94 [0.81, 1.10]Boaz et al., 2000 - SPACE 9 97 15 99 0.8% 0.61 [0.28, 1.33]de Gaetano et al., 2001 - PPP 22 2,231 26 2,264 1.6% 0.86 [0.49, 1.51]Lee et al., 2005 - WHS 106 19,937 140 19,939 7.9% 0.76 [0.59, 0.97]Lonn et al., 2005 - HOPE & HOPE-TOO 482 4,761 475 4,780 34.8% 1.02 [0.90, 1.15]Magliano et al., 2006 - MAVET 2 205 4 204 0.2% 0.50 [0.09, 2.69]Cook et al., 2007 - WACS 193 4,083 202 4,088 13.6% 0.96 [0.79, 1.16]Sesso et al., 2008 - PHS II 131 3,659 122 3,653 8.6% 1.07 [0.84, 1.37]Milman et al., 2008 3 726 5 708 0.2% 0.59 [0.14, 2.44]Lippman et al., 2009 - SELECT 119 8,863 142 8,910 8.60% 0.84 [0.66, 1.07]
Risk RatioM-H, Random, 95%CI in CVD mortality risk
Study or SubgroupStephens et al., 1996 - CHAOSGISSI-Prevenzione Investigators 1999Boaz et al., 2000 - SPACEde Gaetano et al., 2001 - PPPLee et al., 2005 - WHSLonn et al., 2005 - HOPE & HOPE-TOOMagliano et al., 2006 - MAVETCook et al., 2007 - WACSSesso et al., 2008 - PHS IIMilman et al., 2008Lippman et al., 2009 - SELECT
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 8.96, df = 10 (P = 0.54); I² = 0%Test for overall effect: Z = 1.41 (P = 0.16)
Vitamin E Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours vitamin E Favours control
Supplementary Figure 42. Forest plot of vitamin E supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Risk Ratio Risk RatioSubgroup and Study, Year Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI in CHD mortality riskRCTsVirtamo et al., 1998 - ATBC 212 6,820 238 6,849 46.0% 0.89 [0.75, 1.07]GISSI-Prevenzione Investigators 1999 247 5,660 273 5,664 54.0% 0.91 [0.77, 1.07]
Vitamin E Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours Vitamin E Favours Control
Supplementary Figure 43. Forest plot of vitamin E supplementation and CHD mortality risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
99
Subgroup and Study, Year Events Total Events TotalRCTsStephens et al., 1996 - CHAOS 18 1,035 13 967 5.1% 1.29 [0.64, 2.63]GISSI-Prevenzione Investigators 1999 228 5,660 251 5,664 82.3% 0.91 [0.76, 1.08]Boaz et al., 2000 - SPACE 2 97 8 99 1.1% 0.26 [0.06, 1.17]de Gaetano et al., 2001 - PPP 3 2,231 7 2,264 1.4% 0.43 [0.11, 1.68]Hodis et al., 2002 - VEAPS 1 177 1 176 0.3% 0.99 [0.06, 15.77]Lee et al., 2005 - WHS 12 19,937 14 19,939 4.3% 0.86 [0.40, 1.85]Cook et al., 2007 - WACS 18 4,083 16 4,088 5.6% 1.13 [0.58, 2.21]
Study or SubgroupStephens et al., 1996 - CHAOSGISSI-Prevenzione Investigators 1999Boaz et al., 2000 - SPACEde Gaetano et al., 2001 - PPPHodis et al., 2002 - VEAPSLee et al., 2005 - WHSCook et al., 2007 - WACS
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 5.19, df = 6 (P = 0.52); I² = 0%Test for overall effect: Z = 1.13 (P = 0.26)
Vitamin E Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours Vitamin E Favours Control
Supplementary Figure 44. Forest plot of vitamin E supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsStephens et al., 1996 - CHAOS 1 1,035 1 967 2.0% 0.93 [0.06, 14.92]de Gaetano et al., 2001 - PPP 2 2,231 5 2,264 5.6% 0.41 [0.08, 2.09]Lee et al., 2005 - WHS 21 19,937 24 19,939 43.8% 0.88 [0.49, 1.57]Cook et al., 2007 - WACS 18 4,083 15 4,088 32.1% 1.20 [0.61, 2.38]Lippman et al., 2009 - SELECT 9 8,863 8 8,910 16.6% 1.13 [0.44, 2.93]
Vitamin E Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours vitamin E Favours control
Supplementary Figure 45. Forest plot of vitamin E supplementation and stroke mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Risk RatioM-H, Random, 95%CI in all-cause mortality risk
Study or Subgroupde la Maza et al., 1995Stephens et al., 1996 - CHAOSSano et al., 1997 - ADCSGillilan et al., 1997Shoulson et al., 1998 - DATATOPGISSI-Prevenzione Investigators 1999Hoffman et al., 1999Salonen et al., 2000 - ASAPBoaz et al., 2000 - SPACEde Gaetano et al., 2001 - PPPDesnuelle et al., 2001 - ALSRTde Waart et al., 2001Wluka et al., 2002Graat et al., 2002Hodis et al., 2002 - VEAPSVirtamo et al., 2003 - ATBCCollins et al., 2003McNeil et al., 2004 - VECATManuel-Y-Keenoy et al., 2004 - DATORMeydani et al., 2004Lonn et al., 2005 - HOPE & HOPE-TOOGraf et al., 2005Peterson et al., 2005 - ADCS 2Lee et al., 2005 - WHSBairati et al., 2006Magliano et al., 2006 - MAVETCook et al., 2007 - WACSMilman et al., 2008Lippman et al., 2009 - SELECTSanyal et al., 2010 - PIVENSManning et al., 2013Wang et al., 2014 - PHS II
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 17.22, df = 31 (P = 0.98); I² = 0%Test for overall effect: Z = 0.17 (P = 0.87)
Vitamin E Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours vitamin E Favours control
Supplementary Figure 46. Forest plot of vitamin E supplementation and all-cause mortality. M-H, Manthel-Haenszel. *Hoffman et al., 1999 - data taken from Curtis et al., 2014; †Collins et al., 2003 - data taken from Curtis et al., 2014. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
101
Supplementary Figure 47. Funnel plot of vitamin E supplementation and and CVD and all-cause mortality outcomes: (A) total CVD risk, (B) MI risk, (C) stroke risk, (D) CVD mortality risk and (E) all-cause mortality risk. The vertical line represents the pooled effect estimate expressed as a RR. Dashed lines represent pseudo-95% confidence intervals (CI). The circles represent risk estimates for each study, and the horizontal lines represent standard errors of the RR. We were unable to test for funnel plot asymmetry for other CVD outcomes (<10 RCTs).
Favours vitamin C/ Favours control/ Negative association Positive association
Pooled Effect Estimates
0.0 0.5 1.0 1.5 2.0
Supplementary Figure 48. Summary of the pooled effect estimates of RCTs assessing the relationship between vitamin C supplementation and CVD and all-cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsCook et al., 2007 - WACS 419 4,087 415 4,084 57.6% 1.01 [0.89, 1.15]Sesso et al., 2008 - PHS II* 309 3,673 316 3,653 42.4% 0.97 [0.84, 1.13]
Vitamin C Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours vitamin C Favours control
Supplementary Figure 49. Forest plot of vitamin C supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. *Sesso et al., 2008 - Data taken from most updated trial data from Wang et al., 2014. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
103
Subgroup and Study, Year Events Total Events TotalRCTsCook et al., 2007 - WACS 510 4,087 489 4,084 100% 1.04 [0.93, 1.17]
Total (95% CI) 4,087 4,084 100% 1.04 [0.93, 1.17]Total events 510 489 Heterogeneity: Not applicable 0.5 0.7 1 1.5 2Test for overall effect: Z = 0.70 (P = 0.49) Favours vitamin C Favours control
Vitamin C ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in CHD risk
Study or SubgroupCook et al., 2007 - WACS
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.70 (P = 0.49)
Events510
510
Total4087
4087
Events489
489
Total4084
4084
Weight100.0%
100.0%
M-H, Random, 95% CI1.04 [0.93, 1.17]
1.04 [0.93, 1.17]
Year2007
Vitamin C Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours vitamin C Favours control
Supplementary Figure 50. Forest plot of vitamin C supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsCook et al., 2007 - WACS 140 4,087 134 4,084 50.3% 1.04 [0.83, 1.32]Sesso et al., 2008 - PHS II* 127 3,673 144 3,653 49.7% 0.88 [0.69, 1.11]
Vitamin C Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours vitamin C Favours control
Supplementary Figure 51. Forest plot of vitamin C supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. *Sesso et al., 2008 - Data taken from most updated trial data from Wang et al., 2014. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsCook et al., 2007 - WACS 138 4,087 160 4,084 56.8% 0.86 [0.69, 1.08]Sesso et al., 2008 - PHS II* 114 3,673 113 3,653 43.2% 1.00 [0.78, 1.30]
Vitamin C Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours vitamin C Favours control
Supplementary Figure 52. Forest plot of vitamin C supplementation and stroke risk. M-H, Manthel-Haenszel. *Sesso et al., 2008 - Data taken from most updated trial data from Wang et al., 2014. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
104
Subgroup and Study, Year Events Total Events TotalRCTsCook et al., 2007 - WACS 206 4,087 189 4,084 61.5% 1.09 [0.90, 1.32]Sesso et al., 2008 - PHS II* 129 3,673 122 3,653 38.5% 1.05 [0.82, 1.34]
Vitamin C Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours vitamin C Favours control
Supplementary Figure 53. Forest plot of vitamin C supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. *Sesso et al., 2008 - Data taken from most updated trial data from Wang et al., 2014. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsCook et al., 2007 - WACS 15 4,087 19 4,084 100.0% 0.79 [0.40, 1.55]
Total (95% CI) 4,087 4,084 100% 0.79 [0.40, 1.55]Total events 15 19 Heterogeneity: Not applicable 0.02 0.1 1 10 50Test for overall effect: Z = 0.69 (P = 0.49) Favours vitamin C Favours control
Vitamin C ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in MI mortality risk
Study or SubgroupCook et al., 2007 - WACS
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.69 (P = 0.49)
Events15
15
Total4087
4087
Events19
19
Total4084
4084
Weight100.0%
100.0%
M-H, Random, 95% CI0.79 [0.40, 1.55]
0.79 [0.40, 1.55]
Year2007
Vitamin C Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours vitamin C Favours control
Supplementary Figure 54. Forest plot of vitamin C supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. *Sesso et al., 2008 - Data taken from most updated trial data from Wang et al., 2014. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
105
Subgroup and Study, Year Events Total Events TotalRCTsCook et al., 2007 - WACS 15 4,087 18 4,084 100.0% 0.83 [0.42, 1.65]
Total (95% CI) 4,087 4,084 100% 0.83 [0.42, 1.65]Total events 15 18 Heterogeneity: Not applicableTest for overall effect: Z = 0.52 (P = 0.60) 0.02 0.1 1 10 50
Favours vitamin C Favours control
Vitamin C ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in stroke mortality risk
Study or SubgroupCook et al., 2007 - WACS
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.52 (P = 0.60)
Events15
15
Total4087
4087
Events18
18
Total4084
4084
Weight100.0%
100.0%
M-H, Random, 95% CI0.83 [0.42, 1.65]
0.83 [0.42, 1.65]
Year2007
Vitamin C Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours vitamin C Favours control
Supplementary Figure 55. Forest plot of vitamin C supplementation and stroke mortality risk. M-H, Manthel-Haenszel. *Sesso et al., 2008 - Data taken from most updated trial data from Wang et al., 2014. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsSalonen et al., 2000- ASAP 1 130 1 130 0.1% 1.00 [0.06, 15.82]Correa et al., 2000 2 130 0 117 0.1% 4.50 [0.22, 92.86]Cook et al., 2007 - WACS 504 4,087 491 4,084 55.0% 1.03 [0.91, 1.15]Wang et al., 2014 - PHS II 414 3,673 406 3,653 44.8% 1.01 [0.89, 1.15]
Vitamin C Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours vitamin C Favours control
Supplementary Figure 56. Forest plot of vitamin C supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
106
Random Effects
Comparison RCTs N Events RR (95% CIs) RR (95% CIs) I2 p -value
Favours selenium/ Favours control/ Negative association Positive association
0.0 0.5 1.0 1.5 2.0
Supplementary Figure 57. Summary of the pooled effect estimates of RCTs assessing the relationship between selenium supplementation and CVD and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects.
Subgroup and Study, Year Events Total Events TotalRCTsStranges et al., 2006 - NPC 103 504 96 500 9.3% 1.06 [0.83, 1.37]Lippman et al., 2009 - SELECT 1,080 8,856 1,050 8,910 90.7% 1.03 [0.96, 1.12]
Selenium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.7 0.85 1 1.2 1.5Favours selenium Favours control
Supplementary Figure 58. Forest plot of selenium supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
107
Subgroup and Study, Year Events Total Events TotalRCTsStranges et al., 2006 - NPC 63 504 59 500 100.0% 1.06 [0.76, 1.48]
Total (95% CI) 504 500 100% 1.06 [0.76, 1.48]Total events 63 59Heterogeneity: Not applicableTest for overall effect: Z = 0.34 (P = 0.73) 0.2 0.5 1 2 5
Favours selenium Favours control
Selenium ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in total CHD risk
Study or SubgroupStranges et al., 2006 - NPC
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.34 (P = 0.73)
Events63
63
Total504
504
Events59
59
Total500
500
Weight100.0%
100.0%
M-H, Random, 95% CI1.06 [0.76, 1.48]
1.06 [0.76, 1.48]
Year2006
Selenium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours selenium Favours control
Supplementary Figure 59. Forest plot of selenium supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsKorpela et al., 1989 1 40 2 41 2.9% 0.51 [0.05, 5.43]Stranges et al., 2006 - NPC 41 504 43 500 97.1% 0.95 [0.63, 1.42]
Selenium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours selenium Favours control
Supplementary Figure 60. Forest plot of selenium supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsStranges et al., 2006 - NPC 35 504 32 500 28.2% 1.09 [0.68, 1.72]Lippman et al., 2009 - SELECT 82 8,856 100 8,910 71.8% 0.82 [0.62, 1.10]
Selenium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours selenium Favours control
Supplementary Figure 61. Forest plot of selenium supplementation and stroke risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
108
Subgroup and Study, Year Events Total Events TotalRCTsKorpela et al., 1989 0 40 4 41 2.0% 0.11 [0.01, 2.05]Stranges et al., 2006 - NPC 40 504 31 500 38.9% 1.28 [0.81, 2.01]Lippman et al., 2009 - SELECT 129 8,856 142 8,910 59.1% 0.91 [0.72, 1.16]
Selenium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.005 0.1 1 10 200Favours Selenium Favours control
Supplementary Figure 62. Forest plot of selenium supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLimburg et al., 2005 1 90 0 90 8.1% 3.00 [0.12, 72.68]Stranges et al., 2006 - NPC 9 504 8 500 91.9% 1.12 [0.43, 2.87]
Selenium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours selenium Favours control
Supplementary Figure 63. Forest plot of selenium supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLippman et al., 2009 - SELECT 9 8,856 8 8,910 100.0% 1.13 [0.44, 2.93]
Total (95% CI) 8,856 8,910 100% 1.13 [0.44, 2.93]Total events 9 8Heterogeneity: Not applicableTest for overall effect: Z = 0.26 (P = 0.80) 0.01 0.1 1 10 100
Favours selenium Favours control
Selenium ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in stroke mortality risk
Study or SubgroupLippman et al., 2009 - SELECT
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.26 (P = 0.80)
Events9
9
Total8856
8856
Events8
8
Total8910
8910
Weight100.0%
100.0%
M-H, Random, 95% CI1.13 [0.44, 2.93]
1.13 [0.44, 2.93]
Year2009
Selenium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours selenium Favours control
Supplementary Figure 64. Forest plot of selenium supplementation and stroke mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
109
Subgroup and Study, Year Events Total Events TotalRCTsLimburg et al., 2005 1 90 0 90 0.1% 3.00 [0.12, 72.68]Stranges et al., 2006 - NPC 110 504 111 500 26.0% 0.98 [0.78, 1.24]Lippman et al., 2009 - SELECT 378 8856 382 8910 72.9% 1.00 [0.87, 1.14]Marshall et al., 2011 4 212 6 211 0.9% 0.66 [0.19, 2.32]
Selenium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours Selenium Favours control
Supplementary Figure 65. Forest plot of selenium supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Favours B-complex/ Favours control/ Negative association Positive association
Pooled Effect Estimates
0.5 1.0 1.5
Supplementary Figure 66. Summary of the pooled effect estimates of RCTs assessing the relationship between vitamin B complex supplementation and CVD and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
110
Subgroup and Study, Year Events Total Events TotalRCTLonn et al., 2006 - HOPE-TOO 519 2,758 547 2,764 17.8% 0.95 [0.85, 1.06]Bonaa et al., 2006 - NORVIT* 201 937 172 943 8.5% 1.18 [0.98, 1.41]Albert et al., 2008 - WAFACS 205 2,721 211 2,721 8.3% 0.97 [0.81, 1.17]Hodis et al., 2009 - BVAIT† 9 254 11 252 0.5% 0.81 [0.34, 1.93]Heinz et al., 2010 83 327 98 323 5.0% 0.84 [0.65, 1.07]VITATOPS Trial Study Group 2010 616 4,089 678 4,075 19.4% 0.91 [0.82, 1.00]Galan et al., 2010 - SU.FOL.OM3 75 1,242 82 1,259 3.5% 0.93 [0.68, 1.26]Armitage et al., 2010 - SEARCH 1,537 6,033 1,493 6,031 29.7% 1.03 [0.97, 1.09]Van Dijk et al., 2015 - B-PROOF 181 1,516 170 1,511 7.50% 1.06 [0.87, 1.29]
Study or SubgroupLonn et al., 2006 - HOPE 2Bonaa et al., 2006 - NORVITAlbert et al., 2008 - WAFACSHodis et al., 2009 - BVAITHeinz et al., 2010VITATOPS Trial Study Group 2010Galan et al., 2010 - SU.FOL.OM3Armitage et al., 2010 - SEARCHVan Dijk et al., 2015 - B-PROOF
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 11.20, df = 8 (P = 0.19); I² = 29%Test for overall effect: Z = 0.55 (P = 0.58)
B-Complex Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours B-Complex Favours Control
Supplementary Figure 67. Forest plot of vitamin B complex supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. *Bonaa et al., 2006 - folic acid, B6 and B12 vs placebo. †Hodis et al., 2009 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTLange et al., 2004 53 316 35 320 4.4% 1.53 [1.03, 2.28]Albert et al., 2008 - WAFACS 283 2,721 280 2,721 23.7% 1.01 [0.86, 1.18]Imasa et al., 2009 15 118 13 125 1.5% 1.22 [0.61, 2.46]Galan et al., 2010 - SU.FOL.OM3 49 1,242 55 1,259 4.9% 0.90 [0.62, 1.32]Armitage et al., 2010 - SEARCH 1,229 6,033 1,185 6,031 65.6% 1.04 [0.97, 1.11]
B-Complex Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours B-complex Favours control
Supplementary Figure 68. Forest plot of vitamin B complex supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
111
Subgroup and Study, Year Events Total Events TotalRCTsSchnyder et al., 2002 - The Swiss Heart Study 7 272 12 281 0.6% 0.60 [0.24, 1.51]Lange et al., 2004 3 316 2 320 0.2% 1.52 [0.26, 9.03]Bonaa et al., 2006 - NORVIT* 182 937 153 943 12.7% 1.20 [0.99, 1.45]Lonn et al., 2006 - HOPE-TOO 341 2,758 349 2,764 24.8% 0.98 [0.85, 1.13]Jamison et al., 2007 - HOST 129 1,032 150 1,024 10.1% 0.85 [0.69, 1.06]Albert et al., 2008 - WAFACS 65 2,721 74 2,721 4.5% 0.88 [0.63, 1.22]Ebbing et al., 2008 - WENBIT† 59 772 58 780 4.0% 1.03 [0.73, 1.46]VITATOPS Trial Study Group 2010 118 4,089 114 4,075 7.5% 1.03 [0.80, 1.33]Galan et al., 2010 - SU.FOL.OM3 28 1,242 32 1,259 1.9% 0.89 [0.54, 1.46]Armitage et al., 2010 - SEARCH 431 6,033 429 6,031 29.2% 1.00 [0.88, 1.14]House et al., 2010 - DIVINe 8 119 4 119 0.4% 2.00 [0.62, 6.46]Heinz et al., 2010 20 327 19 323 1.3% 1.04 [0.57, 1.91]Van Dijk et al., 2015 - B-PROOF 45 1,516 43 1,511 2.9% 1.04 [0.69, 1.57]
Study or SubgroupSchnyder et al., 2002 - The Swiss Heart StudyLange et al., 2004Bonaa et al., 2006 - NORVITLonn et al., 2006 - HOPE 2Jamison et al., 2007 - HOSTAlbert et al., 2008 - WAFACSEbbing et al., 2008 - WENBITVITATOPS Trial Study Group 2010Galan et al., 2010 - SU.FOL.OM3Armitage et al., 2010 - SEARCHHouse et al., 2010 - DIVINeHeinz et al., 2010Van Dijk et al., 2015 - B-PROOF
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 9.06, df = 12 (P = 0.70); I² = 0%Test for overall effect: Z = 0.00 (P = 1.00)
B-Complex Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours B-Complex Favours Control
Supplementary Figure 69. Forest plot of vitamin B complex supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. *Bonaa et al., 2006 - folic acid, B6 and B12 vs placebo. †Ebbing et al., 2008 - folic acid, B6 and B12 vs placebo. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Study or SubgroupBonaa et al., 2006 - NORVITJamison et al., 2007 - HOSTEbbing et al., 2008 - WENBITAlbert et al., 2008 - WAFACSSaposnik et al., 2009 - HOPE 2Imasa et al., 2009VITATOPS Trial Study Group 2010Heinz et al., 2010Galan et al., 2010 - SU.FOL.OM3Armitage et al., 2010 - SEARCHHouse et al., 2010 - DIVINeVan Dijk et al., 2015 - B-PROOF
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.01; Chi² = 13.13, df = 11 (P = 0.28); I² = 16%Test for overall effect: Z = 2.01 (P = 0.04)
B-Complex Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours B-Complex Favours Control
Supplementary Figure 70. Forest plot of vitamin B complex supplementation and stroke risk. M-H, Manthel-Haenszel. *Bonaa et al., 2006 - folic acid, B6 and B12 vs placebo. †Ebbing et al., 2008 - folic acid, B6 and B12 vs placebo. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
112
Subgroup and Study, Year Events Total Events TotalRCTsLonn et al., 2006 - HOPE-TOO 276 2,758 291 2,764 25.0% 0.95 [0.81, 1.11]Albert et al., 2008 - WAFACS 96 2,721 94 2,721 12.9% 1.02 [0.77, 1.35]VITATOPS Trial Study Group 2010 328 4,089 380 4,075 27.1% 0.86 [0.75, 0.99]Galan et al., 2010 - SU.FOL.OM3 26 1,242 14 1,259 3.2% 1.88 [0.99, 3.59]Armitage et al., 2010 - SEARCH 578 6,033 559 6,031 31.7% 1.03 [0.93, 1.15]
B-Complex Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours B-Complex Favours Control
Supplementary Figure 71. Forest plot of vitamin B complex supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsSchnyder et al., 2002 - The Swiss Heart Study 3 272 6 281 0.8% 0.52 [0.13, 2.04]Heinz et al., 2010 37 327 32 323 7.4% 1.14 [0.73, 1.79]Armitage et al., 2010 - SEARCH 463 6,033 423 6,031 91.8% 1.09 [0.96, 1.24]
B-Complex Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours B-complex Favours control
Supplementary Figure 72. Forest plot of vitamin B complex supplementation and CHD mortality risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsBonaa et al., 2006 - NORVIT* 68 937 59 943 27.1% 1.16 [0.83, 1.62]Armitage et al., 2010 - SEARCH 185 6,033 170 6,031 72.9% 1.09 [0.89, 1.34]
B-Complex Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours B-complex Favours control
Supplementary Figure 73. Forest plot of vitamin B complex supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. *Bonaa et al., 2006 - folic acid, B6 and B12 vs placebo. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
113
Subgroup and Study, Year Events Total Events TotalRCTsSaposnik et al., 2009 - HOPE 2 27 2,758 30 2,764 31.5% 0.90 [0.54, 1.51]Armitage et al., 2010 - SEARCH 59 6,033 65 6,031 68.5% 0.91 [0.64, 1.29]
B-Complex Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours B-complex Favours control
Supplementary Figure 74. Forest plot of vitamin B complex supplementation and stroke mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsSchnyder et al., 2002 - The Swiss Heart Study 4 272 8 281 0.1% 0.52 [0.16, 1.70]Lange et al., 2004 1 316 1 320 0.0% 1.01 [0.06, 16.12]Lonn et al., 2006 - HOPE-TOO 470 2,758 475 2,764 14.5% 0.99 [0.88, 1.11]Bonaa et al., 2006 - NORVIT* 104 937 89 943 2.7% 1.18 [0.90, 1.54]Jamison et al., 2007 - HOST 448 1,032 436 1,024 19.7% 1.02 [0.92, 1.13]Ebbing et al., 2008 - WENBIT† 35 772 30 780 0.9% 1.18 [0.73, 1.90]Albert et al., 2008 - WAFACS 250 2,721 256 2,721 7.1% 0.98 [0.83, 1.15]Hodis et al., 2009 - BVAIT 0 254 2 252 0.0% 0.20 [0.01, 4.11]Imasa et al., 2009 22 118 20 125 0.6% 1.17 [0.67, 2.02]Heinz et al., 2010 102 327 92 323 3.5% 1.10 [0.86, 1.39]Armitage et al., 2010 - SEARCH 983 6,033 951 6,031 29.3% 1.03 [0.95, 1.12]Galan et al., 2010 - SU.FOL.OM3 72 1,242 45 1,259 1.5% 1.62 [1.13, 2.33]VITATOPS Trial Study Group 2010 614 4,089 633 4,075 18.7% 0.97 [0.87, 1.07]House et al., 2010 - DIVINe 7 119 6 119 0.2% 1.17 [0.40, 3.37]Van Wijngaarden et al., 2014 - B-PROOF 37 1,516 42 1,511 1.00% 0.88 [0.57, 1.36]Wang et al., 2015 4 195 6 195 0% 0.67 [0.19, 2.33]
Risk RatioM-H, Random, 95%CI in all-cause mortality risk
Study or SubgroupSchnyder et al., 2002 - The Swiss Heart StudyLange et al., 2004Lonn et al., 2006 - HOPE 2Bonaa et al., 2006 - NORVITJamison et al., 2007 - HOSTEbbing et al., 2008 - WENBITAlbert et al., 2008 - WAFACSHodis et al., 2009Imasa et al., 2009Heinz et al., 2010Armitage et al., 2010 - SEARCHGalan et al., 2010 - SU.FOL.OM3VITATOPS Trial Study Group 2010House et al., 2010 - DIVINeVan Wijngaarden et al., 2014 - B-PROOFWang et al., 2015
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 13.22, df = 15 (P = 0.58); I² = 0%Test for overall effect: Z = 0.82 (P = 0.41)
B-Complex Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours B-complex Favours control
Supplementary Figure 75. Forest plot of vitamin B complex supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. *Bonaa et al., 2006 - folic acid, B6 and B12 vs placebo. †Ebbing et al., 2008 - folic acid, B6 and B12 vs placebo. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
114
Supplementary Figure 76. Funnel plot of vitamin B complex supplementation and CVD and all-cause mortality outcomes: (A) MI risk, (B) stroke risk, and (C) all-cause mortality risk. The vertical line represents the pooled effect estimate expressed as a RR. Dashed lines represent pseudo-95% confidence intervals (CI). The circles represent risk estimates for each study, and the horizontal lines represent standard errors of the RR. We were unable to test for funnel plot asymmetry for other CVD outcomes (<10 RCTs).
115
Random Effects
Comparison RCTs N Events RR (95% CIs) RR (95% CIs) I2 p -valueCVD
Favours folic acid/ Favours control/ Negative association Positive association
0.0 0.5 1.0 1.5 2.0
Supplementary Figure 77. Summary of the pooled effect estimates of RCTs assessing the relationship between folic acid supplementation and CVD and all-cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
Folic acid Control Weight Risk RatioSubgroup and Study, Year Events Total Events TotalRCTsRighetti et al., 2003 13 51 11 30 3.2% 0.70 [0.36, 1.35]Liem et al., 2004 43 140 45 143 11.9% 0.98 [0.69, 1.38]Zoungas et al., 2006 - ASFAST 46 156 55 159 13.7% 0.85 [0.62, 1.18]Vianna et al., 2007 9 93 9 93 1.9% 1.00 [0.42, 2.41]Huo et al., 2015 - CSPPT 324 10,348 405 10,354 69.4% 0.80 [0.69, 0.92]
Folate Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours folate Favours control
Supplementary Figure 78. Forest plot of folic acid supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
116
Subgroup and Study, Year Events Total Events TotalRCTsBaker et al., 2002 23 942 12 940 40.1% 1.91 [0.96, 3.82]Zoungas et al., 2006 - ASFAST 23 156 19 159 59.9% 1.23 [0.70, 2.17]
Folate Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours folate Favours control
Supplementary Figure 79. Forest plot of folic acid supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLiem et al., 2003 3 300 4 293 8.3% 0.73 [0.17, 3.24]Liem et al., 2004 8 140 10 143 20.5% 0.82 [0.33, 2.01]Cole et al., 2007 - AFPPS 14 516 8 505 22.1% 1.71 [0.72, 4.05]Logan et al., 2008 - ukCAP 3 470 0 469 2.2% 6.99 [0.36, 134.86]Wu et al., 2009 - NHS/HPFS 6 338 1 334 4.3% 5.93 [0.72, 48.98]Huo et al., 2015 - CSPPT 25 10,348 24 10,354 42.6% 1.04 [0.60, 1.82]
Folate Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours folate Favours control
Supplementary Figure 80. Forest plot of folic acid supplementation and MI risk. Manthel-Haenszel M-H, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
117
Subgroup and Study, Year Events Total Events TotalRCTsLiem et al., 2003 4 300 3 293 1.0% 1.30 [0.29, 5.77]Liem et al., 2004 1 140 0 143 0.2% 3.06 [0.13, 74.58]Zoungas et al., 2006 - ASFAST 8 156 18 159 3.4% 0.45 [0.20, 1.01]Cole et al., 2007 - AFPPS 9 516 5 505 1.9% 1.76 [0.59, 5.22]Logan et al., 2008 - ukCAP 1 470 1 469 0.3% 1.00 [0.06, 15.91]Wu et al., 2009 - NHS/HPFS 4 338 3 334 1.0% 1.32 [0.30, 5.84]Huo et al., 2015 - CSPPT 282 10,348 355 10,354 92.3% 0.79 [0.68, 0.93]
Study or SubgroupLiem et al., 2003Liem et al., 2004Zoungas et al., 2006 - ASFASTCole et al., 2007 - AFPPSLogan et al., 2008Wu et al., 2009 - NHS/HPFSHuo et al., 2015 - CSPPT
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 5.51, df = 6 (P = 0.48); I² = 0%Test for overall effect: Z = 2.92 (P = 0.003)
Folate Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours folate Favours control
Supplementary Figure 81. Forest plot of folic acid supplementation and stroke risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLiem et al., 2003 7 300 9 293 7.9% 0.76 [0.29, 2.01]Zoungas et al., 2006 - ASFAST 21 156 24 159 25.7% 0.89 [0.52, 1.53]Vianna et al., 2007 17 93 21 93 23.1% 0.81 [0.46, 1.43]Wu et al., 2009 - NHS/HPFS 0 338 3 334 0.9% 0.14 [0.01, 2.72]Huo et al., 2015 - CSPPT 43 10,348 43 10,354 42.4% 1.00 [0.66, 1.53]
Folate Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours folate Favours control
Supplementary Figure 82. Forest plot of folic acid supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with effects model.
118
Subgroup and Study, Year Events Total Events TotalRCTsLiem et al., 2004 2 140 2 143 31.3% 1.02 [0.15, 7.15]Huo et al., 2015 - CSPPT 5 10,348 4 10,354 68.7% 1.25 [0.34, 4.66]
Folate Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.002 0.1 1 10 500Favours folate Favours control
Supplementary Figure 83. Forest plot of folic acid supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLiem et al., 2004 1 140 0 143 5.5% 3.06 [0.13, 74.58]Huo et al., 2015 - CSPPT 18 10,348 10 10,354 94.5% 1.80 [0.83, 3.90]
Risk RatioM-H, Random, 95%CI in stroke mortality risk
Study or SubgroupLiem et al., 2004Huo et al., 2015 - CSPPT
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 0.10, df = 1 (P = 0.75); I² = 0%Test for overall effect: Z = 1.61 (P = 0.11)
Events1
18
19
Total140
10348
10488
Events0
10
10
Total143
10354
10497
Weight5.5%
94.5%
100.0%
M-H, Random, 95% CI3.06 [0.13, 74.58]
1.80 [0.83, 3.90]
1.85 [0.88, 3.93]
Year20042015
Folate Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours folate Favours control
Supplementary Figure 84. Forest plot of folic acid supplementation and stroke mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
119
Subgroup and Study, Year Events Total Events TotalRCTsLiem et al., 2003 12 300 14 293 5.7% 0.84 [0.39, 1.78]Liem et al., 2004 6 140 7 143 3.0% 0.88 [0.30, 2.54]Potena et al., 2005 1 26 0 25 0.4% 2.89 [0.12, 67.75]Zoungas et al., 2006 - ASFAST 45 156 46 159 20.2% 1.00 [0.71, 1.41]Vianna et al., 2007 23 93 30 93 13.3% 0.77 [0.48, 1.22]Durga et al., 2007 - FACIT 8 405 4 413 2.4% 2.04 [0.62, 6.72]Cole et al., 2007 - AFPPS 10 516 19 505 5.7% 0.52 [0.24, 1.10]Logan et al., 2008 - UKCAP 1 470 7 469 0.8% 0.14 [0.02, 1.15]Wu et al., 2009 - NHS/HPFS 7 338 15 334 4.3% 0.46 [0.19, 1.12]Huo et al., 2015 - CSPPT 302 10,348 320 10,354 44.1% 0.94 [0.81, 1.10]
Risk RatioM-H, Random, 95%CI in all-cause mortality risk
Study or SubgroupLiem et al., 2003Liem et al., 2004Potena et al., 2005Zoungas et al., 2006 - ASFASTVianna et al., 2007Durga et al., 2007 - FACITCole et al., 2007 - AFPPSLogan et al., 2008 - UKCAPWu et al., 2009 - NHS/HPFSHuo et al., 2015 - CSPPT
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.02; Chi² = 10.82, df = 9 (P = 0.29); I² = 17%Test for overall effect: Z = 1.46 (P = 0.14)
Folate Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours folate Favours control
Supplementary Figure 85. Forest plot of folic acid supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Supplementary Figure 86. Funnel plot of folic acid supplementation and all-cause mortality risk. The vertical line represents the pooled effect estimate expressed as a RR. Dashed lines represent pseudo-95% confidence intervals (CI). The circles represent risk estimates for each study, and the horizontal lines represent standard errors of the RR. We were unable to test for other CVD outcomes (<10 RCTs).
Supplementary Figure 87. Summary of the pooled effect estimates of RCTs assessing the relationship between niacin (B3) supplementation and CVD and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
121
Subgroup and Study, Year Events Total Events TotalRCTsNo statinsCDPRG 1975 914 1,119 2,333 2,789 58.3% 0.98 [0.95, 1.01]Subtotal (95% CI) 1,119 2,789 58.3% 0.98 [0.95, 1.01]Total events 914 2,333 Heterogeneity: Not applicableTest for overall effect: Z = 1.45 (P = 0.15)
Vitamin B3 (Niacin) Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours B3 (Niacin) Favours Control
Supplementary Figure 88. Forest plot of niacin (B3) supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. *Boden et al., 2011 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsGuyton et al., 2008 1 670 0 272 0.1% 1.22 [0.05, 29.87]HPS2-THRIVE Collaborative Group 2014 668 12,838 694 12,835 99.9% 0.96 [0.87, 1.07]
Study or SubgroupGuyton et al., 2008HPS 2 THRIVE 2014
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 0.02, df = 1 (P = 0.88); I² = 0%Test for overall effect: Z = 0.72 (P = 0.47)
Events1
668
669
Total670
12838
13508
Events0
694
694
Total272
12835
13107
Weight0.1%
99.9%
100.0%
M-H, Random, 95% CI1.22 [0.05, 29.87]
0.96 [0.87, 1.07]
0.96 [0.87, 1.07]
Year20082014
Vitamin B3 (Niacin) Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours Vitamin B3 Favours Control
Supplementary Figure 89. Forest plot of niacin (B3) supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
122
Subgroup and Study, Year Events Total Events TotalRCTsNo statinsCDPRG 1975 114 1,119 386 2,789 32.7% 0.74 [0.60, 0.90]Subtotal (95% CI) 1,119 2,789 32.7% 0.74 [0.60, 0.90]Total events 114 386 Heterogeneity: Not applicableTest for overall effect: Z = 3.05 (P = 0.002)
Vitamin B3 (Niacin) Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours B3 (Niacin) Favours Control
Supplementary Figure 90. Forest plot of niacin (B3) supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. *Boden et al., 2011 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
123
Subgroup and Study, Year Events Total Events TotalRCTsNo statinsCDPRG 1975 95 1,119 311 2,789 37.8% 0.76 [0.61, 0.95]Subtotal (95% CI) 1,119 2,789 37.8% 0.76 [0.61, 0.95]Total events 95 311 Heterogeneity: Not applicableTest for overall effect: Z = 2.44 (P = 0.01)
Vitamin B3 (Niacin) Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours B3 (Niacin) Favours Control
Supplementary Figure 91. Forest plot of niacin (B3) supplementation and stroke risk. M-H, Manthel-Haenszel. *Boden et al., 2011 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsNo statinsCDPRG 1975 238 1,119 633 2,789 91.0% 0.94 [0.82, 1.07]Subtotal (95% CI) 1,119 2,789 91.0% 0.94 [0.82, 1.07]Total events 238 633 Heterogeneity: Not applicableTest for overall effect: Z = 0.97 (P = 0.33)
Vitamin B3 (Niacin) Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours B3 (Niacin) Favours Control
Supplementary Figure 92. Forest plot of niacin (B3) supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
124
Subgroup and Study, Year Events Total Events TotalRCTsNo statinsCDPRG 1975 203 1,119 535 2,789 51.5% 0.95 [0.82, 1.09]Subtotal (95% CI) 1,119 2,789 51.5% 0.95 [0.82, 1.09]Total events 203 535 Heterogeneity: Not applicableTest for overall effect: Z = 0.75 (P = 0.45)
Vitamin B3 (Niacin) Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.7 0.85 1 1.2 1.5Favours B3 (Niacin) Favours Control
Supplementary Figure 93. Forest plot of niacin (B3) supplementation and CHD mortality risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalNo statinsCDPRG 1975 273 1,119 709 2,789 38.8% 0.96 [0.85, 1.08]Subtotal (95% CI) 1,119 2,789 38.8% 0.96 [0.85, 1.08]Total events 273 709 Heterogeneity: Not applicableTest for overall effect: Z = 0.67 (P = 0.51)
Vitamin B3 (Niacin) Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours B3 (Niacin) Favours Control
Supplementary Figure 94. Forest plot of niacin (B3) supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. *Sang et al., 2009 - Data taken from meta-analysis Keene et al., 2014. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Favours vitamin B6/ Favours control/ Negative association Positive association
0.5 1.0 1.5
Supplementary Figure 95. Summary of the pooled effect estimates of RCTs studies assessing the relationship between vitamin B6 supplementation and CVD and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsBonaa et al., 2006 - NORVIT 175 934 172 943 100.0% 1.03 [0.85, 1.24]
Total (95% CI) 934 943 100% 1.03 [0.85, 1.24]Total events 175 172 Heterogeneity: Not applicableTest for overall effect: Z = 0.28 (P = 0.78) 0.2 0.5 1 2 5
Favours vitamin B6 Favours control
Vitamin B6 ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in total CVD risk
Study or SubgroupBonaa et al., 2006 - NORVIT
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.28 (P = 0.78)
Events175
175
Total934
934
Events172
172
Total943
943
Weight100.0%
100.0%
M-H, Random, 95% CI1.03 [0.85, 1.24]
1.03 [0.85, 1.24]
Year2006
Vitamin B6 Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours vitamin B6 Favours control
Supplementary Figure 96. Forest plot of vitamin B6 supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
126
Subgroup and Study, Year Events Total Events TotalRCTsBonaa et al., 2006 - NORVIT 161 934 153 943 75.6% 1.06 [0.87, 1.30]Ebbing et al., 2008 - WENBIT 55 772 58 780 24.4% 0.96 [0.67, 1.37]
Vitamin B6 Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.7 0.85 1 1.2 1.5Favours vitamin B6 Favours control
Supplementary Figure 97. Forest plot of vitamin B6 supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsBonaa et al., 2006 - NORVIT 22 934 27 943 55.5% 0.82 [0.47, 1.43]Ebbing et al., 2008 - WENBIT 20 772 19 780 44.5% 1.06 [0.57, 1.98]
Vitamin B6 Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours vitamin B6 Favours control
Supplementary Figure 98. Forest plot of vitamin B6 supplementation and stroke risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsBonaa et al., 2006 - NORVIT 61 934 59 943 100% 1.04 [0.74, 1.48]
Total (95% CI) 934 943 100% 1.04 [0.74, 1.48]Total events 61 59 Heterogeneity: Not applicableTest for overall effect: Z = 0.24 (P = 0.81) 0.5 0.7 1 1.5 2
Favours vitamin B6 Favours control
Vitamin B6 ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in MI mortality risk
Study or SubgroupBonaa et al., 2006 - NORVIT
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.24 (P = 0.81)
Events61
61
Total934
934
Events59
59
Total943
943
Weight100.0%
100.0%
M-H, Random, 95% CI1.04 [0.74, 1.48]
1.04 [0.74, 1.48]
Year2006
Vitamin B6 Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours vitamin B6 Favours control
Supplementary Figure 99. Forest plot of vitamin B6 supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
127
Subgroup and Study, Year Events Total Events TotalRCTsBonaa et al., 2006 - NORVIT 92 934 89 943 76.9% 1.04 [0.79, 1.38]Ebbing et al., 2008 - WENBIT 28 772 30 780 23.1% 0.94 [0.57, 1.56]
Vitamin B6 Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours vitamin B6 Favours control
Supplementary Figure 100. Forest plot of vitamin B6 supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Random Effects
Comparison RCTs N Events RR (95% CIs) RR (95% CIs) I2 p -value
Favours calcium/ Favours control/ Negative association Positive association
Pooled Effect Estimates
0.0 0.5 1.0 1.5 2.0
Supplementary Figure 101. Summary of the pooled effect estimates of RCTs studies assessing the relationship between calcium supplementation and CVD and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
128
Subgroup and Study, Year Events Total Events TotalRCTsBolland et al., 2008 101 732 54 739 47.1% 1.89 [1.38, 2.59]Chailurkit et al., 2010 2 201 0 196 3.6% 4.88 [0.24, 100.93]Lewis et al., 2011 - CAIFOS* 104 730 103 730 49.4% 1.01 [0.78, 1.30]
Study or SubgroupBolland et al., 2008Chailurkit et al., 2010Lewis et al., 2011 - CAIFOS
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.17; Chi² = 10.04, df = 2 (P = 0.007); I² = 80%Test for overall effect: Z = 1.19 (P = 0.23)
Events101
2104
207
Total732201730
1663
Events540
103
157
Total739196730
1665
Weight47.1%3.6%
49.4%
100.0%
M-H, Random, 95% CI1.89 [1.38, 2.59]
4.88 [0.24, 100.93]1.01 [0.78, 1.30]
1.43 [0.79, 2.59]
Year200820102011
Calcium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours calcium Favours control
Supplementary Figure 102. Forest plot of calcium supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. *Lewis et al., 2011 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsBaron et al., 1999 50 464 46 466 60.0% 1.09 [0.75, 1.60]Grant et al., 2005 - RECORD* 39 1,113 31 1,128 40.0% 1.28 [0.80, 2.03]
Calcium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours calcium Favours control
Supplementary Figure 103. Forest plot of calcium supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. *Grant et al., 2005 - data taken from Lewis et al, 2015 (meta), however, data include women only; The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
129
Subgroup and Study, Year Events Total Events TotalRCTsGrant et al., 2005 - RECORD* 26 1,113 13 1,128 27.6% 2.03 [1.05, 3.92]Reid et al., 2008† 2 108 - 107 3.5% 4.95 [0.24, 101.99]Bolland et al., 2008 45 732 19 739 31.6% 2.39 [1.41, 4.05]Lewis et al., 2011 - CAIFOS‡ 63 730 63 730 37.3% 1.00 [0.72, 1.40]
Calcium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours calcium Favours control
Supplementary Figure 104. Forest plot of calcium supplementation and MI risk. M-H, Manthel-Haenszel, MI, Myocardial infarction. *Grant et al., 2005 - data taken from Lewis et al, 2015 (meta), however, data include women only; †Reid et al., 2008 - highest dose of calcium (1200mg/day) used as intervention group; ‡Lewis et al., 2011 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsBaron et al., 1999 12 464 11 466 12.9% 1.10 [0.49, 2.46]Bolland et al., 2008 52 732 34 739 47.6% 1.54 [1.01, 2.35]Lewis et al., 2011 - CAIFOS* 36 730 33 730 39.5% 1.09 [0.69, 1.73]
Calcium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours calcium Favours control
Supplementary Figure 105. Forest plot of calcium supplementation and stroke risk. M-H, Manthel-Haenszel. *Lewis et al., 2011 - reported number of participants with at least one event; The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
130
Subgroup and Study, Year Events Total Events TotalRCTsBolland et al., 2008 4 732 1 739 29.9% 4.04 [0.45, 36.04]Lewis et al., 2011 - CAIFOS* 18 730 24 730 70.1% 0.75 [0.41, 1.37]
Risk RatioM-H, Random, 95% CI in CVD mortality risk
Study or SubgroupBolland et al., 2008Lewis et al., 2011 - CAIFOS
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.76; Chi² = 2.13, df = 1 (P = 0.14); I² = 53%Test for overall effect: Z = 0.28 (P = 0.78)
Events4
18
22
Total732730
1462
Events1
24
25
Total739730
1469
Weight29.9%70.1%
100.0%
M-H, Random, 95% CI4.04 [0.45, 36.04]
0.75 [0.41, 1.37]
1.24 [0.27, 5.65]
Year20082011
Calcium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.002 0.1 1 10 500Favours calcium Favours control
Supplementary Figure 106. Forest plot of calcium supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. *Lewis et al., 2011 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLewis et al., 2011 - CAIFOS* 13 730 9 730 100% 1.44 [0.62, 3.36]
Total (95% CI) 730 730 100% 1.44 [0.62, 3.36]Total events 13 9 Heterogeneity: Not applicableTest for overall effect: Z = 0.85 (P = 0.39) 0.01 0.1 1 10 100
Favours calcium Favours control
Calcium ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in MI mortality risk
Study or SubgroupLewis et al., 2011 - CAIFOS
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.85 (P = 0.39)
Events13
13
Total730
730
Events9
9
Total730
730
Weight100.0%
100.0%
M-H, Random, 95% CI1.44 [0.62, 3.36]
1.44 [0.62, 3.36]
Year2011
Calcium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours calcium Favours control
Supplementary Figure 107. Forest plot of calcium supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. *Lewis et al., 2011 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
131
Subgroup and Study, Year Events Total Events TotalRCTsLewis et al., 2011 - CAIFOS* 6 730 8 730 100.0% 0.75 [0.26, 2.15]
Total (95% CI) 730 730 100% 0.75 [0.26, 2.15]Total events 6 8 Heterogeneity: Not applicableTest for overall effect: Z = 0.54 (P = 0.59) 0.002 0.1 1 10 500
Favours calcium Favours control
Calcium ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in stroke mortality risk
Study or SubgroupLewis et al., 2011 - CAIFOS
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.54 (P = 0.59)
Events6
6
Total730
730
Events8
8
Total730
730
Weight100.0%
100.0%
M-H, Random, 95% CI0.75 [0.26, 2.15]
0.75 [0.26, 2.15]
Year2011
Calclium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.002 0.1 1 10 500Favours calcium Favours control
Supplementary Figure 108. Forest plot of calcium supplementation and stroke mortality risk. M-H, Manthel-Haenszel. *Lewis et al., 2011 - reported number of participants with at least one event. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsBaron et al., 1999 25 464 22 466 3.9% 1.14 [0.65, 1.99]Grant et al., 2005 - RECORD 464 2,617 434 2,675 84.9% 1.09 [0.97, 1.23]Reid et al., 2008* 1 108 1 107 0.2% 0.99 [0.06, 15.64]Bolland et al., 2008 38 732 30 739 5.5% 1.28 [0.80, 2.04]Chailurkit et al., 2010 1 201 1 196 0.2% 0.98 [0.06, 15.48]Lewis et al., 2011 - CAIFOS 29 730 38 730 5.4% 0.76 [0.48, 1.22]
Calcium Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours calcium Favours control
Supplementary Figure 109. Forest plot of calcium supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. *Reid et al., 2008 - highest dose of calcium (1200mg/day) used as intervention group. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
132
Random Effects
Comparison RCTs N Events RR (95% CIs) RR (95% CIs) I2 p -value
Supplementary Figure 110. Summary of the pooled effect estimates of RCTs and studies assessing the relationship between iron supplementation and CVD and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsAnker et al., 2009 - FAIR-HF* 21 304 22 155 100% 0.49 [0.28, 0.86]
Total (95% CI) 304 155 100% 0.49 [0.28, 0.86]Total events 21 22Heterogeneity: Not applicableTest for overall effect: Z = 2.49 (P = 0.01) 0.01 0.1 1 10 100
Favours iron Favours control
Iron ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95%CI in total CVD risk
Study or SubgroupAnker et al., 2009 - FAIR-HF
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 2.49 (P = 0.01)
Events21
21
Total304
304
Events22
22
Total155
155
Weight100.0%
100.0%
M-H, Random, 95% CI0.49 [0.28, 0.86]
0.49 [0.28, 0.86]
Year2009
Iron Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.01 0.1 1 10 100Favours iron Favours control
Supplementary Figure 111. Forest plot of iron supplementation and total CVD risk. M-H, Manthel-Haenszel. CVD, cardiovascular disease. *Anker et al., 2009 - To treat iron deficiency in heart failure patients. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
133
Subgroup and Study, Year Events Total Events TotalRCTsAnker et al., 2009 - FAIR-HF* 2 304 3 155 100% 0.34 [0.06, 2.01]
Total (95% CI) 304 155 100% 0.34 [0.06, 2.01]Total events 2 3 Heterogeneity: Not applicableTest for overall effect: Z = 1.19 (P = 0.23) 0.002 0.1 1 10 500
Favours iron Favours control
Iron ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95%CI in MI risk
Study or SubgroupAnker et al., 2009 - FAIR-HF
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 1.19 (P = 0.23)
Events2
2
Total304
304
Events3
3
Total155
155
Weight100.0%
100.0%
M-H, Random, 95% CI0.34 [0.06, 2.01]
0.34 [0.06, 2.01]
Year2009
Iron Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.002 0.1 1 10 500Favours iron Favours control
Supplementary Figure 112. Forest plot of iron supplementation and MI risk. M-H, Manthel-Haenszel. MI, myocardial infarction. *Anker et al., 2009 - To treat iron deficiency in heart failure patients. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsAnker et al., 2009 - FAIR-HF* 4 304 4 155 25% 0.51 [0.13, 2.01]Ponikowski et al., 2015 - CONFIRM HF 11 150 12 151 75% 0.92 [0.42, 2.03]
Iron Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours iron Favours control
Supplementary Figure 113. Forest plot of iron supplementation and CVD mortality risk. M-H, Manthel-Haenszel. CVD, cardiovascular disease. *Anker et al., 2009 - To treat iron deficiency in heart failure patients. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsAnker et al., 2009 - FAIR-HF* 5 304 4 155 24.3% 0.64 [0.17, 2.34]Ponikowski et al. 2015 - CONFIRM-HF* 12 152 14 151 75.7% 0.85 [0.41, 1.78]
Iron Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours iron Favours control
Supplementary Figure 114. Forest plot of iron supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. *Anker et al., 2009 - To treat iron deficiency in heart failure patients; *Ponikowski et al., 2015 - To treat iron deficiency in heart failure patients. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
134
Random Effects
Comparison RCTs N Events RR (95% CIs) RR (95% CIs) I2 p -value
Favours multivitamins/ Favours control/ Negative association Positive association
Pooled Effect Estimates
0.0 0.5 1.0 1.5 2.0
Supplementary Figure 115. Summary of the pooled effect estimates of RCTs assessing the relationship between multivitamins supplementation and CVD and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsSesso et al., 2012 - PHS II 876 7,317 856 7,324 65.0% 1.02 [0.94, 1.12]Lamas et al., 2013 - TACT 94 853 115 855 35.0% 0.82 [0.63, 1.06]
Multivitamins Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours multivitamins Favours control
Supplementary Figure 116. Forest plot of multivitamins supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
135
Subgroup and Study, Year Events Total Events TotalRCTsBogden et al., 1994 0 33 1 32 0.2% 0.32 [0.01, 7.66]Sesso et al., 2012 - PHS II 317 7,317 335 7,324 84.1% 0.95 [0.82, 1.10]Lamas et al., 2013 - TACT 58 853 61 855 15.8% 0.95 [0.67, 1.35]
Multivitamins Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours multivitamins Favours control
Supplementary Figure 117. Forest plot of multivitamins supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsSesso et al., 2012 - PHS II 332 7,317 311 7,324 69.1% 1.07 [0.92, 1.24]Lamas et al., 2013 - TACT 8 853 15 855 30.9% 0.53 [0.23, 1.25]
Multivitamins Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours multivitamins Favours control
Supplementary Figure 118. Forest plot of multivitamins supplementation and stroke risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsCTNS 2008 23 501 31 501 5.4% 0.74 [0.44, 1.25]Sesso et al., 2012 - PHS II 408 7,317 421 7,324 84.5% 0.97 [0.85, 1.11]Lamas et al., 2013 - TACT 45 853 56 855 10.2% 0.81 [0.55, 1.18]
Multivitamins Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.5 0.7 1 1.5 2Favours multivitamins Favours control
Supplementary Figure 119. Forest plot of multivitamins supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered
136
to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsSesso et al., 2012 - PHS II 27 7,317 43 7,324 100% 0.63 [0.39, 1.02]
Total (95% CI) 7,317 7,324 100% 0.63 [0.39, 1.02]Total events 27 43 Heterogeneity: Not applicableTest for overall effect: Z = 1.90 (P = 0.06) 0.05 0.2 1 5 20
Favours multivitamin Favours control
Multivitamins ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95%CI in MI mortality risk
Study or SubgroupSesso et al., 2012 - PHS II
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 1.90 (P = 0.06)
Events27
27
Total7317
7317
Events43
43
Total7324
7324
Weight100.0%
100.0%
M-H, Random, 95% CI0.63 [0.39, 1.02]
0.63 [0.39, 1.02]
Year2012
Multivitamins Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours multivitamins Favours control
Supplementary Figure 120. Forest plot of multivitamins supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLi et al., 1993 - NIT2 22 1,657 35 1,661 43.7% 0.63 [0.37, 1.07]Sesso et al., 2012 - PHS II 89 7,317 76 7,324 56.3% 1.17 [0.86, 1.59]
Multivitamins Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours multivitamins Favours control
Supplementary Figure 121. Forest plot of multivitamins supplementation and stroke mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsChandra et al., 1992 0 48 2 48 0.0% 0.20 [0.01, 4.06]Li et al., 1993 - NIT2 157 1,657 167 1,661 8.1% 0.94 [0.77, 1.16]Bogden et al., 1994 0 33 1 32 0.0% 0.32 [0.01, 7.66]Pike et al., 1995 1 24 0 23 0.0% 2.88 [0.12, 67.29]Graat et al., 2002 0 163 5 153 0.0% 0.09 [0.00, 1.53]Avenell et al., 2005 - MAVIS 8 456 4 454 0% 1.99 [0.60, 6.57]Liu et al., 2007 96 375 97 373 5.90% 0.98 [0.77, 1.26]CTNS 2008 77 510 81 510 4% 0.95 [0.71, 1.27]Sesso et al., 2012 - PHS II 1,345 7,317 1,412 7,324 77% 0.95 [0.89, 1.02]Lamas et al., 2013 - TACT 87 853 93 855 5% 0.94 [0.71, 1.24]
Risk RatioM-H, Random, 95%CI in all-cause mortality risk
Study or SubgroupChandra et al., 1992Li et al., 1993 - NIT2Bogden et al., 1994Pike et al., 1995Graat et al., 2002Avenell et al., 2005 - MAVISLiu et al., 2007CTNS 2008Sesso et al., 2012 - PHS IILamas et al., 2013 - TACT
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 6.20, df = 9 (P = 0.72); I² = 0%Test for overall effect: Z = 1.58 (P = 0.12)
Multivitamins Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours multivitamins Favours control
Supplementary Figure 122. Forest plot of multivitamins supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-
137
study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects effects model.
Supplementary Figure 123. Funnel plot of multivitamin supplementation and all-cause mortality risk. The vertical line represents the pooled effect estimate expressed as a RR. Dashed lines represent pseudo-95% confidence intervals (CI). The circles represent risk estimates for each study, and the horizontal lines represent standard errors of the RR. We were unable to test for funnel plot asymmetry for other CVD outcomes (<10 RCTs).
Supplementary Figure 124. Summary of the pooled effect estimates of RCTs and assessing the relationship between calcium and vitamin D supplementation and CVD and all cause mortality risk. RR, risk ratio; CIs, confidence intervals; RTCs, randomized clinical trials; CVD, cardiovascular disease; CHD, coronary heart disease; MI, myocardial infarction; N/A, not applicable. The diamonds represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% CIs, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsBrazier et al., 2005 6 95 5 97 0.6% 1.23 [0.39, 3.88]Prentice et al., 2013 - WHI CaD* 848 7,718 813 7,584 99.4% 1.02 [0.94, 1.12]
Calcium & Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours Ca/VitD Favours control
Supplementary Figure 125. Forest plot of calcium and vitamin D supplementation and total CVD risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. *Prentice et al., 2013 - removed users of personal calcium or vitamin D supplement use. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
139
Subgroup and Study, Year Events Total Events TotalRCTsInkovaara et al., 1983 1 46 4 42 24.9% 0.23 [0.03, 1.96]Prentice et al., 2013 - WHI CaD* 229 7,718 211 7,584 75.1% 1.07 [0.89, 1.28]
Calcium and Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours Ca/VitD Favours control
Supplementary Figure 126. Forest plot of calcium and vitamin D supplementation and total CHD risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. *Prentice et al., 2013 - removed users of personal calcium or vitamin D supplement use. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsKomulainen et al., 1999 - OSTPRE 1 116 1 115 0.4% 0.99 [0.06, 15.66]Grant et al., 2005 - RECORD 44 1,306 39 1,332 18.5% 1.15 [0.75, 1.76]Brazier et al., 2005 3 95 0 97 0.4% 7.15 [0.37, 136.50]Lappe et al., 2007 3 446 2 288 1.0% 0.97 [0.16, 5.76]Prentice et al., 2013 - WHI CaD* 193 7,718 167 7,584 79.6% 1.14 [0.93, 1.39]
Calcium & Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.1 0.2 0.5 1 2 5 10Favours Ca/VitD Favours control
Supplementary Figure 127. Forest plot of calcium and vitamin D supplementation and MI risk. M-H, Manthel-Haenszel, MI, myocardial infarction. *Prentice et al., 2013 - removed users of personal calcium or vitamin D supplement use. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
140
Subgroup and Study, Year Events Total Events TotalRCTsInkovaara et al., 1983 9 46 5 42 3.0% 1.64 [0.60, 4.51]Komulainen et al., 1999 - OSTPRE 2 116 1 115 0.5% 1.98 [0.18, 21.56]Brazier et al., 2005 1 95 1 97 0.4% 1.02 [0.06, 16.09]Grant et al., 2005 - RECORD 60 1,306 48 1,332 22.5% 1.27 [0.88, 1.85]Lappe et al., 2007 6 446 4 288 2.0% 0.97 [0.28, 3.40]Gallagher et al., 2012 1 21 0 21 0.3% 3.00 [0.13, 69.70]Prentice et al., 2013 - WHI CaD* 184 7,718 162 7,584 71.3% 1.12 [0.91, 1.38]
Study or SubgroupInkovaara et al., 1983Komulainen et al., 1999 - OSTREBrazier et al., 2005Grant et al., 2005 - RECORDLappe et al., 2007Gallagher et al., 2012Prentice et al., 2013 - WHI CaD
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 1.46, df = 6 (P = 0.96); I² = 0%Test for overall effect: Z = 1.72 (P = 0.09)
Calcium & Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.05 0.2 1 5 20Favours Ca/VitD Favours control
Supplementary Figure 128. Forest plot of calcium and vitamin D supplementation and stroke risk. M-H, Manthel-Haenszel. *Prentice et al., 2013 - removed users of personal calcium or vitamin D supplement use. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsLaCroix et al., 2009 - WHI CaD 226 18,176 244 18,106 100.0% 0.92 [0.77, 1.10]
Total (95% CI) 18,176 18,106 100% 0.92 [0.77, 1.10]Total events 226 244 Heterogeneity: Not applicable 0.2 0.5 1 2 5Test for overall effect: Z = 0.88 (P = 0.38) Favours Ca+VitD Favours control
Calcium & Vitamin D ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in CVD mortality risk
Study or SubgroupLaCroix et al., 2009 - WHI CaD
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.88 (P = 0.38)
Events226
226
Total18176
18176
Events244
244
Total18106
18106
Weight100.0%
100.0%
M-H, Random, 95% CI0.92 [0.77, 1.10]
0.92 [0.77, 1.10]
Year2009
Vit D+Ca Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours Vit D+Ca Favours Control
Supplementary Figure 129. Forest plot of calcium and vitamin D supplementation and CVD mortality risk. M-H, Manthel-Haenszel, CVD, cardiovascular disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
141
Subgroup and Study, Year Events Total Events TotalRCTsInkovaara et al., 1983 1 46 3 42 5.9% 0.30 [0.03, 2.81]
Calcium & Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours Ca/VitD Favours control
Supplementary Figure 130. Forest plot of calcium and vitamin D supplementation and CHD mortality risk. M-H, Manthel-Haenszel, CHD, coronary heart disease. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsKomulainen et al., 1999 - OSTPRE 1 116 0 115 100.0% 2.97 [0.12, 72.26]
Total (95% CI) 116 115 100% 2.97 [0.12, 72.26]
Total events 1 0
Heterogeneity: Not applicable 0.001 0.1 1 10 1000Test for overall effect: Z = 0.67 (P = 0.50) Favours Ca+VitD Favours control
Calcium & Vitamin D ControlWeight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI in MI mortality risk
Study or SubgroupKomulainen et al., 1999 - OSTRE
Total (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.67 (P = 0.50)
Events1
1
Total116
116
Events0
0
Total115
115
Weight100.0%
100.0%
M-H, Random, 95% CI2.97 [0.12, 72.26]
2.97 [0.12, 72.26]
Year1999
Experimental Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.001 0.1 1 10 1000Favours Vit D and calcium Favours Control
Supplementary Figure 131. Forest plot of calcium and vitamin D supplementation and MI mortality risk. M-H, Manthel-Haenszel, MI, myocardial infarction. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
Subgroup and Study, Year Events Total Events TotalRCTsInkovaara et al., 1983 2 46 2 42 3.5% 0.91 [0.13, 6.20]LaCroix et al., 2009 - WHI CaD 54 18,176 60 18,106 96.5% 0.90 [0.62, 1.29]
Calcium & Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.2 0.5 1 2 5Favours Ca/VitD Favours control
Supplementary Figure 132. Forest plot of calcium and vitamin D supplementation and stroke mortality risk. M-H, Manthel-Haenszel. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
142
Subgroup and Study, Year Events Total Events TotalRCTsInkovaara et al., 1983 7 46 5 42 0.3% 1.28 [0.44, 3.72]Chapuy et al., 1992 258 1,634 274 1,636 14.8% 0.94 [0.81, 1.10]Dawson-Hughes et al., 1997* 2 187 2 202 0.1% 1.08 [0.15, 7.59]Baeksgaard et al., 1998 0 80 1 80 0.0% 0.33 [0.01, 8.06]Komulainen et al., 1999 1 116 1 115 0.0% 0.99 [0.06, 15.66]Krieg et al., 1999 21 124 26 124 1.3% 0.81 [0.48, 1.36]Chapuy et al., 2002 - Decalyos II 67 393 43 190 3.1% 0.75 [0.54, 1.06]Harwood et al., 2004 - NoNOF 6 39 5 37 0.3% 1.14 [0.38, 3.41]Meier et al., 2004 0 30 1 25 0.0% 0.28 [0.01, 6.58]Larsen et al., 2004† 832 4,957 839 4,648 47.0% 0.93 [0.85, 1.01]Grant et al., 2005 - RECORD 221 1306 217 1332 12.2% 1.04 [0.88, 1.23]Porthouse et al., 2005 57 1,321 68 1,993 3.0% 1.26 [0.90, 1.79]Brazier et al., 2005 3 95 1 97 0.1% 3.06 [0.32, 28.93]Daly et al., 2006 1 85 0 82 0.0% 2.90 [0.12, 70.07]Bolton-Smith et al., 2007‡ 0 62 1 61 0.0% 0.33 [0.01, 7.90]Zhu et al., 2008 - CAIFOS 0 39 2 41 0.0% 0.21 [0.01, 4.24]Salovaara et al., 2010 - OSTRE-FPS 15 1,586 13 1,609 0.7% 1.17 [0.56, 2.45]Prentice et al., 2013 - WHI CaD§ 331 7,718 338 7,584 16.3% 0.96 [0.83, 1.12]Gupta et al., 2016 4 25 11 28 0.4% 0.41 [0.15, 1.12]Lappe et al., 2017 7 1,156 9 1,147 0.4% 0.77 [0.29, 2.07]
M-H, Random, 95% CI in all-cause mortality risk Calcium & Vitamin D
Study or SubgroupInkovaara et al., 1983Chapuy et al., 1992Dawson-Hughes et al., 1997Baeksgaard et al., 1998Komulainen et al., 1999Krieg et al., 1999Chapuy et al., 2002 - Decalyos IIHarwood et al., 2004 - NoNOFMeier et al., 2004Larsen et al., 2004Grant et al., 2005 - RECORDPorthouse et al., 2005Brazier et al., 2005Daly et al., 2006Bolton-Smith et al., 2007Zhu et al., 2008 - CAIFOSSalovaara et al., 2010 - OSTRE-FPSPrentice et al., 2013 - WHI CaDGupta et al., 2016Lappe et al., 2017
Total (95% CI)Total eventsHeterogeneity: Tau² = 0.00; Chi² = 13.61, df = 19 (P = 0.81); I² = 0%Test for overall effect: Z = 1.69 (P = 0.09)
Calcium and Vitamin D Control Risk Ratio Risk RatioM-H, Random, 95% CI
0.02 0.1 1 10 50Favours Ca/VitD Favours control
Supplementary Figure 133. Forest plot of calcium and vitamin D supplementation and all-cause mortality risk. M-H, Manthel-Haenszel. *Dawson-Hughes et al., 1997 - Data taken from meta-analysis Bolland et al., 2014, Bjelakovic et al., 2014; †Larsen et al., 2004 - Data taken from meta-analysis Bjelakovic et al., 2014; ‡Bolton-Smith et al., 2007 - Data taken from meta-analysis Bjelakovic et al., 2014. The diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q statistic (Chi2) at a significance level of P < 0.10, and quantified by the I2 statistic. An I2 value ≥ 50% is considered to indicate substantial heterogeneity. All results are presented as risk ratios with 95% Confidence Intervals, using the Manthel-Haenszel method with random effects model.
143
Supplementary Figure 134. Funnel plot of vitamin D & calcium supplementation and all-cause mortality risk. The vertical line represents the pooled effect estimate expressed as a RR. Dashed lines represent pseudo-95% confidence intervals (CI). The circles represent risk estimates for each study, and the horizontal lines represent standard errors of the RR. We were unable to test for funnel plot asymmetry for other CVD outcomes (<10 RCTs).
144
References
1. Anand SS, Hawkes C, de Souza RJ et al. Food Consumption and its Impact on Cardiovascular Disease: Importance of Solutions Focused on the Globalized Food System: A Report From the Workshop Convened by the World Heart Federation. Journal of the American College of Cardiology 2015;66:1590-614.
2. Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). Jama 2001;285:2486-97.
3. Grundy SM, Cleeman JI, Merz CN et al. Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation 2004;110:227-39.
4. Stone NJ, Robinson JG, Lichtenstein AH et al. 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2014;129:S1-45.
5. Anderson TJ, Gregoire J, Hegele RA et al. 2012 update of the Canadian Cardiovascular Society guidelines for the diagnosis and treatment of dyslipidemia for the prevention of cardiovascular disease in the adult. The Canadian journal of cardiology 2013;29:151-67.
6. Anderson TJ, Gregoire J, Pearson GJ et al. 2016 Canadian Cardiovascular Society Guidelines for the Management of Dyslipidemia for the Prevention of Cardiovascular Disease in the Adult. The Canadian journal of cardiology 2016;32:1263-1282.
7. Reiner Z, Catapano AL, De Backer G et al. ESC/EAS Guidelines for the management of dyslipidaemias: the Task Force for the management of dyslipidaemias of the European Society of Cardiology (ESC) and the European Atherosclerosis Society (EAS). European heart journal 2011;32:1769-818.
8. Stroes ES, Thompson PD, Corsini A et al. Statin-associated muscle symptoms: impact on statin therapy-European Atherosclerosis Society Consensus Panel Statement on Assessment, Aetiology and Management. European heart journal 2015;36:1012-22.
9. British Heart Foundation. Healthy Eating. https://www.bhf.org.uk/heart-health/preventing-heart-disease/healthy-eating. Accessed December 6, 2016.
10. Moyer VA. Vitamin, mineral, and multivitamin supplements for the primary prevention of cardiovascular disease and cancer: U.S. Preventive services Task Force recommendation statement. Annals of internal medicine 2014;160:558-64.
11. World Cancer Research Fund International. Our cancer prevention recommendations. http://www.wcrf.org/int/research-we-fund/our-cancer-prevention-recommendations. Accessed December 6, 2016.
12. Kushi LH, Doyle C, McCullough M et al. American Cancer Society Guidelines on nutrition and physical activity for cancer prevention: reducing the risk of cancer with healthy food choices and physical activity. CA: a cancer journal for clinicians 2012;62:30-67.
13. Canadian Cancer Society. Nutrition and Fitness (Healthy Eating VDhwccep-a-sl-wn-a-froAD.
14. Norat T, Scoccianti C, Boutron-Ruault MC et al. European Code against Cancer 4th Edition: Diet and cancer. Cancer epidemiology 2015;39 Suppl 1:S56-66.
15. Cancer Research UK. Diet and Cancer. http://www.cancerresearchuk.org/about-cancer/causes-of-cancer/diet-and-cancer. Accessed December 7, 2016.
16. Brohult J, Jonson B. Effects of large doses of calciferol on patients with rheumatoid arthritis. A double-blind clinical trial. Scandinavian journal of rheumatology 1973;2:173-6.
17. Clofibrate and niacin in coronary heart disease. Jama 1975;231:360-81.
18. Gillilan RE, Mondell B, Warbasse JR. Quantitative evaluation of vitamin E in the treatment of angina pectoris. Am Heart J 1977;93:444-9.
19. Inkovaara J, Gothoni G, Halttula R, Heikinheimo R, Tokola O. Calcium, vitamin D and anabolic steroid in treatment of aged bones: double-blind placebo-controlled long-term clinical trial. Age and ageing 1983;12:124-30.
20. Corless D, Dawson E, Fraser F et al. Do vitamin D supplements improve the physical capabilities of elderly hospital patients? Age and ageing 1985;14:76-84.
21. Aloia JF, Vaswani A, Yeh JK, Ellis K, Yasumura S, Cohn SH. Calcitriol in the treatment of postmenopausal osteoporosis. The American journal of medicine 1988;84:401-8.
22. Korpela H, Kumpulainen J, Jussila E et al. Effect of selenium supplementation after acute myocardial infarction. Research communications in chemical pathology and pharmacology 1989;65:249-52.
23. McKeown-Eyssen G, Holloway C, Jazmaji V, Bright-See E, Dion P, Bruce WR. A randomized trial of vitamins C and E in the prevention of recurrence of colorectal polyps. Cancer Res 1988;48:4701-5.
24. Ott SM, Chesnut CH, 3rd. Calcitriol treatment is not effective in postmenopausal osteoporosis. Annals of internal medicine 1989;110:267-74.
25. Gallagher JC, Goldgar D. Treatment of postmenopausal osteoporosis with high doses of synthetic calcitriol. A randomized controlled study. Annals of internal medicine 1990;113:649-55.
26. Grady D, Halloran B, Cummings S et al. 1,25-Dihydroxyvitamin D3 and muscle strength in the elderly: a randomized controlled trial. The Journal of clinical endocrinology and metabolism 1991;73:1111-7.
27. Chapuy MC, Arlot ME, Duboeuf F et al. Vitamin D3 and calcium to prevent hip fractures in the elderly women. The New England journal of medicine 1992;327:1637-42.
28. Li JY, Taylor PR, Li B et al. Nutrition intervention trials in Linxian, China: multiple vitamin/mineral supplementation, cancer incidence, and disease-specific mortality among adults with esophageal dysplasia. Journal of the National Cancer Institute 1993;85:1492-8.
29. Bogden JD, Bendich A, Kemp FW et al. Daily micronutrient supplements enhance delayed-hypersensitivity skin test responses in older people. The American journal of clinical nutrition 1994;60:437-47.
30. de la Maza MP, Petermann M, Bunout D, Hirsch S. Effects of long-term vitamin E supplementation in alcoholic cirrhotics. J Am Coll Nutr 1995;14:192-6.
31. Hamdy NA, Kanis JA, Beneton MN et al. Effect of alfacalcidol on natural course of renal bone disease in mild to moderate renal failure. BMJ (Clinical research ed) 1995;310:358-63.
32. Ooms ME, Roos JC, Bezemer PD, van der Vijgh WJ, Bouter LM, Lips P. Prevention of bone loss by vitamin D supplementation in elderly women: a randomized double-blind trial. The Journal of clinical endocrinology and metabolism 1995;80:1052-8.
33. Pike J, Chandra RK. Effect of vitamin and trace element supplementation on immune indices in healthy elderly. International journal for vitamin and nutrition research Internationale Zeitschrift fur Vitamin- und Ernahrungsforschung Journal international de vitaminologie et de nutrition 1995;65:117-21.
34. Steiner M, Glantz M, Lekos A. Vitamin E plus aspirin compared with aspirin alone in patients with transient ischemic attacks. Am J Clin Nutr 1995;62:1381s-1384s.
35. Greenberg ER, Baron JA, Karagas MR et al. Mortality associated with low plasma concentration of beta carotene and the effect of oral supplementation. Jama 1996;275:699-703.
146
36. Hennekens CH, Buring JE, Manson JE et al. Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. The New England journal of medicine 1996;334:1145-9.
37. Lips P, Graafmans WC, Ooms ME, Bezemer PD, Bouter LM. Vitamin D supplementation and fracture incidence in elderly persons. A randomized, placebo-controlled clinical trial. Annals of internal medicine 1996;124:400-6.
38. Omenn GS, Goodman GE, Thornquist MD et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. The New England journal of medicine 1996;334:1150-5.
39. Stephens NG, Parsons A, Schofield PM, Kelly F, Cheeseman K, Mitchinson MJ. Randomised controlled trial of vitamin E in patients with coronary disease: Cambridge Heart Antioxidant Study (CHAOS). Lancet (London, England) 1996;347:781-6.
40. Dawson-Hughes B, Harris SS, Krall EA, Dallal GE. Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age or older. The New England journal of medicine 1997;337:670-6.
41. Girodon F, Lombard M, Galan P et al. Effect of micronutrient supplementation on infection in institutionalized elderly subjects: a controlled trial. Annals of nutrition & metabolism 1997;41:98-107.
42. Moon TE, Levine N, Cartmel B et al. Effect of retinol in preventing squamous cell skin cancer in moderate-risk subjects: a randomized, double-blind, controlled trial. Southwest Skin Cancer Prevention Study Group. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology 1997;6:949-56.
43. Sano M, Ernesto C, Thomas RG et al. A controlled trial of selegiline, alpha-tocopherol, or both as treatment for Alzheimer's disease. The Alzheimer's Disease Cooperative Study. The New England journal of medicine 1997;336:1216-22.
44. Sato Y, Maruoka H, Oizumi K. Amelioration of hemiplegia-associated osteopenia more than 4 years after stroke by 1 alpha-hydroxyvitamin D3 and calcium supplementation. Stroke 1997;28:736-9.
45. Baeksgaard L, Andersen KP, Hyldstrup L. Calcium and vitamin D supplementation increases spinal BMD in healthy, postmenopausal women. Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 1998;8:255-60.
46. Mortality in DATATOP: a multicenter trial in early Parkinson's disease. Parkinson Study Group. Annals of neurology 1998;43:318-25.
47. Virtamo J, Rapola JM, Ripatti S et al. Effect of vitamin E and beta carotene on the incidence of primary nonfatal myocardial infarction and fatal coronary heart disease. Archives of internal medicine 1998;158:668-75.
48. Baron JA, Beach M, Mandel JS et al. Calcium supplements for the prevention of colorectal adenomas. Calcium Polyp Prevention Study Group. The New England journal of medicine 1999;340:101-7.
49. Hoffman RM, Garewal HS. Alpha-tocopherol supplementation for men with existing coronary artery disease: a feasibility study. Preventive medicine 1999;29:112-8.
50. Girodon F, Galan P, Monget AL et al. Impact of trace elements and vitamin supplementation on immunity and infections in institutionalized elderly patients: a randomized controlled trial. MIN. VIT. AOX. geriatric network. Archives of internal medicine 1999;159:748-54.
51. Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E after myocardial infarction: results of the GISSI-Prevenzione trial. Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto miocardico. Lancet (London, England) 1999;354:447-55.
147
52. Green A, Williams G, Neale R et al. Daily sunscreen application and betacarotene supplementation in prevention of basal-cell and squamous-cell carcinomas of the skin: a randomised controlled trial. Lancet (London, England) 1999;354:723-9.
53. Komulainen M, Kroger H, Tuppurainen MT et al. Prevention of femoral and lumbar bone loss with hormone replacement therapy and vitamin D3 in early postmenopausal women: a population-based 5-year randomized trial. The Journal of clinical endocrinology and metabolism 1999;84:546-52.
54. Krieg MA, Jacquet AF, Bremgartner M, Cuttelod S, Thiebaud D, Burckhardt P. Effect of supplementation with vitamin D3 and calcium on quantitative ultrasound of bone in elderly institutionalized women: a longitudinal study. Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 1999;9:483-8.
55. Sato Y, Manabe S, Kuno H, Oizumi K. Amelioration of osteopenia and hypovitaminosis D by 1alpha-hydroxyvitamin D3 in elderly patients with Parkinson's disease. Journal of neurology, neurosurgery, and psychiatry 1999;66:64-8.
56. Boaz M, Smetana S, Weinstein T et al. Secondary prevention with antioxidants of cardiovascular disease in endstage renal disease (SPACE): randomised placebo-controlled trial. Lancet (London, England) 2000;356:1213-8.
57. Correa P, Fontham ET, Bravo JC et al. Chemoprevention of gastric dysplasia: randomized trial of antioxidant supplements and anti-helicobacter pylori therapy. Journal of the National Cancer Institute 2000;92:1881-8.
58. Frazao JM, Elangovan L, Maung HM et al. Intermittent doxercalciferol (1alpha-hydroxyvitamin D(2)) therapy for secondary hyperparathyroidism. American journal of kidney diseases : the official journal of the National Kidney Foundation 2000;36:550-61.
59. Lee IM, Cook NR, Manson JE, Buring JE, Hennekens CH. beta-Carotene Supplementation and Incidence of Cancer and Cardiovascular Disease: the Women's Health Study. Journal of the National Cancer Institute 1999;91:2102-2106.
60. Leppala JM, Virtamo J, Fogelholm R et al. Controlled trial of alpha-tocopherol and beta-carotene supplements on stroke incidence and mortality in male smokers. Arteriosclerosis, thrombosis, and vascular biology 2000;20:230-5.
61. Jacobson JS, Begg MD, Wang LW et al. Effects of a 6-month vitamin intervention on DNA damage in heavy smokers. Cancer Epidemiol Biomarkers Prev 2000;9:1303-11.
62. Salonen JT, Nyyssonen K, Salonen R et al. Antioxidant Supplementation in Atherosclerosis Prevention (ASAP) study: a randomized trial of the effect of vitamins E and C on 3-year progression of carotid atherosclerosis. Journal of internal medicine 2000;248:377-86.
63. A randomized, placebo-controlled, clinical trial of high-dose supplementation with vitamins C and E and beta carotene for age-related cataract and vision loss: AREDS report no. 9. Archives of ophthalmology (Chicago, Ill : 1960) 2001;119:1439-52.
64. Brown BG, Zhao XQ, Chait A et al. Simvastatin and niacin, antioxidant vitamins, or the combination for the prevention of coronary disease. The New England journal of medicine 2001;345:1583-92.
65. de Gaetano G. Low-dose aspirin and vitamin E in people at cardiovascular risk: a randomised trial in general practice. Collaborative Group of the Primary Prevention Project. Lancet (London, England) 2001;357:89-95.
66. Desnuelle C, Dib M, Garrel C, Favier A. A double-blind, placebo-controlled randomized clinical trial of alpha-tocopherol (vitamin E) in the treatment of amyotrophic lateral sclerosis. ALS riluzole-tocopherol Study Group. Amyotrophic lateral sclerosis and other motor neuron disorders : official publication of the World Federation of Neurology, Research Group on Motor Neuron Diseases 2001;2:9-18.
148
67. de Waart FG, Kok FJ, Smilde TJ, Hijmans A, Wollersheim H, Stalenhoef AF. Effect of glutathione S-transferase M1 genotype on progression of atherosclerosis in lifelong male smokers. Atherosclerosis 2001;158:227-31.
68. Gallagher JC, Fowler SE, Detter JR, Sherman SS. Combination treatment with estrogen and calcitriol in the prevention of age-related bone loss. The Journal of clinical endocrinology and metabolism 2001;86:3618-28.
69. You WC, Chang YS, Heinrich J et al. An intervention trial to inhibit the progression of precancerous gastric lesions: compliance, serum micronutrients and S-allyl cysteine levels, and toxicity. European journal of cancer prevention : the official journal of the European Cancer Prevention Organisation (ECP) 2001;10:257-63.
70. Baker F PD, Blackwood S, Hunt J, Erskine M, Dyas M, Ashby M, Siva A, Brown MJ. . Blinded comparison of folic acid and placebo in patients with ischaemic heart disease: an outcome trial. . Circulation 2002;106.
71. Chapuy MC, Pamphile R, Paris E et al. Combined calcium and vitamin D3 supplementation in elderly women: confirmation of reversal of secondary hyperparathyroidism and hip fracture risk: the Decalyos II study. Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 2002;13:257-64.
72. Chylack LT, Jr., Brown NP, Bron A et al. The Roche European American Cataract Trial (REACT): a randomized clinical trial to investigate the efficacy of an oral antioxidant micronutrient mixture to slow progression of age-related cataract. Ophthalmic epidemiology 2002;9:49-80.
73. Graat JM, Schouten EG, Kok FJ. Effect of daily vitamin E and multivitamin-mineral supplementation on acute respiratory tract infections in elderly persons: a randomized controlled trial. Jama 2002;288:715-21.
74. Hodis HN, Mack WJ, LaBree L et al. Alpha-tocopherol supplementation in healthy individuals reduces low-density lipoprotein oxidation but not atherosclerosis: the Vitamin E Atherosclerosis Prevention Study (VEAPS). Circulation 2002;106:1453-9.
75. MRC/BHF Heart Protection Study of antioxidant vitamin supplementation in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet (London, England) 2002;360:23-33.
76. Meyer HE, Smedshaug GB, Kvaavik E, Falch JA, Tverdal A, Pedersen JI. Can vitamin D supplementation reduce the risk of fracture in the elderly? A randomized controlled trial. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 2002;17:709-15.
77. Schnyder G, Roffi M, Flammer Y, Pin R, Hess OM. Effect of homocysteine-lowering therapy with folic acid, vitamin B12, and vitamin B6 on clinical outcome after percutaneous coronary intervention: the Swiss Heart study: a randomized controlled trial. Jama 2002;288:973-9.
78. Waters DD, Alderman EL, Hsia J et al. Effects of hormone replacement therapy and antioxidant vitamin supplements on coronary atherosclerosis in postmenopausal women: a randomized controlled trial. Jama 2002;288:2432-40.
79. Wluka AE, Stuckey S, Brand C, Cicuttini FM. Supplementary vitamin E does not affect the loss of cartilage volume in knee osteoarthritis: a 2 year double blind randomized placebo controlled study. The Journal of rheumatology 2002;29:2585-91.
80. Collins EG, Edwin Langbein W, Orebaugh C et al. PoleStriding exercise and vitamin E for management of peripheral vascular disease. Medicine and science in sports and exercise 2003;35:384-93.
149
81. Cooper L, Clifton-Bligh PB, Nery ML et al. Vitamin D supplementation and bone mineral density in early postmenopausal women. The American journal of clinical nutrition 2003;77:1324-9.
82. Liem A, Reynierse-Buitenwerf GH, Zwinderman AH, Jukema JW, van Veldhuisen DJ. Secondary prevention with folic acid: effects on clinical outcomes. Journal of the American College of Cardiology 2003;41:2105-13.
83. Righetti M, Ferrario GM, Milani S et al. Effects of folic acid treatment on homocysteine levels and vascular disease in hemodialysis patients. Medical science monitor : international medical journal of experimental and clinical research 2003;9:Pi19-24.
84. Trivedi DP, Doll R, Khaw KT. Effect of four monthly oral vitamin D3 (cholecalciferol) supplementation on fractures and mortality in men and women living in the community: randomised double blind controlled trial. BMJ (Clinical research ed) 2003;326:469.
85. Virtamo J, Pietinen P, Huttunen JK et al. Incidence of cancer and mortality following alpha-tocopherol and beta-carotene supplementation: a postintervention follow-up. Jama 2003;290:476-85.
86. Dukas L, Bischoff HA, Lindpaintner LS et al. Alfacalcidol reduces the number of fallers in a community-dwelling elderly population with a minimum calcium intake of more than 500 mg daily. Journal of the American Geriatrics Society 2004;52:230-6.
87. Harwood RH, Sahota O, Gaynor K, Masud T, Hosking DJ. A randomised, controlled comparison of different calcium and vitamin D supplementation regimens in elderly women after hip fracture: The Nottingham Neck of Femur (NONOF) Study. Age and ageing 2004;33:45-51.
88. Coburn JW, Maung HM, Elangovan L et al. Doxercalciferol safely suppresses PTH levels in patients with secondary hyperparathyroidism associated with chronic kidney disease stages 3 and 4. American journal of kidney diseases : the official journal of the National Kidney Foundation 2004;43:877-90.
89. Lange H, Suryapranata H, De Luca G et al. Folate therapy and in-stent restenosis after coronary stenting. The New England journal of medicine 2004;350:2673-81.
90. Larsen ER, Mosekilde L, Foldspang A. Vitamin D and calcium supplementation prevents osteoporotic fractures in elderly community dwelling residents: a pragmatic population-based 3-year intervention study. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 2004;19:370-8.
91. Liem AH, van Boven AJ, Veeger NJ et al. Efficacy of folic acid when added to statin therapy in patients with hypercholesterolemia following acute myocardial infarction: a randomised pilot trial. International journal of cardiology 2004;93:175-9.
92. Taylor AJ, Sullenberger LE, Lee HJ, Lee JK, Grace KA. Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol (ARBITER) 2: a double-blind, placebo-controlled study of extended-release niacin on atherosclerosis progression in secondary prevention patients treated with statins. Circulation 2004;110:3512-7.
93. Tornwall ME, Virtamo J, Korhonen PA et al. Effect of alpha-tocopherol and beta-carotene supplementation on coronary heart disease during the 6-year post-trial follow-up in the ATBC study. European heart journal 2004;25:1171-8.
94. Manuel-y-Keenoy B, Vinckx M, Vertommen J, van Gaal L, de Leeuw I. Impact of vitamin E supplementation on lipoprotein peroxidation and composition in type 1 diabetic patients treated with atorvastatin. Atherosclerosis 2004;175:369-376.
95. McNeil JJ, Robman L, Tikellis G, Sinclair MI, McCarty CA, Taylor HR. Vitamin E supplementation and cataract: randomized controlled trial. Ophthalmology 2004;111:75-84.
96. Meier C, Woitge HW, Witte K, Lemmer B, Seibel MJ. Supplementation with oral vitamin D3 and calcium during winter prevents seasonal bone loss: a randomized controlled
150
open-label prospective trial. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 2004;19:1221-30.
97. Meydani SN, Leka LS, Fine BC et al. Vitamin E and respiratory tract infections in elderly nursing home residents: a randomized controlled trial. Jama 2004;292:828-36.
98. Aloia JF, Talwar SA, Pollack S, Yeh J. A randomized controlled trial of vitamin D3 supplementation in African American women. Archives of internal medicine 2005;165:1618-23.
99. Avenell A, Campbell MK, Cook JA et al. Effect of multivitamin and multimineral supplements on morbidity from infections in older people (MAVIS trial): pragmatic, randomised, double blind, placebo controlled trial. BMJ (Clinical research ed) 2005;331:324-9.
100. Brazier M, Grados F, Kamel S et al. Clinical and laboratory safety of one year's use of a combination calcium + vitamin D tablet in ambulatory elderly women with vitamin D insufficiency: results of a multicenter, randomized, double-blind, placebo-controlled study. Clinical therapeutics 2005;27:1885-93.
101. Flicker L, MacInnis RJ, Stein MS et al. Should older people in residential care receive vitamin D to prevent falls? Results of a randomized trial. Journal of the American Geriatrics Society 2005;53:1881-8.
102. Graf M, Ecker D, Horowski R et al. High dose vitamin E therapy in amyotrophic lateral sclerosis as add-on therapy to riluzole: results of a placebo-controlled double-blind study. Journal of neural transmission (Vienna, Austria : 1996) 2005;112:649-60.
103. Grant AM, Avenell A, Campbell MK et al. Oral vitamin D3 and calcium for secondary prevention of low-trauma fractures in elderly people (Randomised Evaluation of Calcium Or vitamin D, RECORD): a randomised placebo-controlled trial. Lancet (London, England) 2005;365:1621-8.
104. Lee IM, Cook NR, Gaziano JM et al. Vitamin E in the primary prevention of cardiovascular disease and cancer: the Women's Health Study: a randomized controlled trial. Jama 2005;294:56-65.
105. Limburg PJ, Wei W, Ahnen DJ et al. Randomized, placebo-controlled, esophageal squamous cell cancer chemoprevention trial of selenomethionine and celecoxib. Gastroenterology 2005;129:863-73.
106. Lonn E, Bosch J, Yusuf S et al. Effects of long-term vitamin E supplementation on cardiovascular events and cancer: a randomized controlled trial. Jama 2005;293:1338-47.
107. Mooney LA, Madsen AM, Tang D et al. Antioxidant vitamin supplementation reduces benzo(a)pyrene-DNA adducts and potential cancer risk in female smokers. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology 2005;14:237-42.
108. Petersen RC, Thomas RG, Grundman M et al. Vitamin E and donepezil for the treatment of mild cognitive impairment. The New England journal of medicine 2005;352:2379-88.
109. Porthouse J, Cockayne S, King C et al. Randomised controlled trial of calcium and supplementation with cholecalciferol (vitamin D3) for prevention of fractures in primary care. BMJ (Clinical research ed) 2005;330:1003.
110. Potena L, Grigioni F, Magnani G et al. Homocysteine-lowering therapy and early progression of transplant vasculopathy: a prospective, randomized, IVUS-based study. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2005;5:2258-64.
111. Sato Y, Iwamoto J, Kanoko T, Satoh K. Low-dose vitamin D prevents muscular atrophy and reduces falls and hip fractures in women after stroke: a randomized controlled trial. Cerebrovascular diseases (Basel, Switzerland) 2005;20:187-92.
151
112. Bairati I, Meyer F, Jobin E et al. Antioxidant vitamins supplementation and mortality: a randomized trial in head and neck cancer patients. International journal of cancer 2006;119:2221-4.
113. Bonaa KH, Njolstad I, Ueland PM et al. Homocysteine lowering and cardiovascular events after acute myocardial infarction. The New England journal of medicine 2006;354:1578-88.
114. Coyne D, Acharya M, Qiu P et al. Paricalcitol capsule for the treatment of secondary hyperparathyroidism in stages 3 and 4 CKD. American journal of kidney diseases : the official journal of the National Kidney Foundation 2006;47:263-76.
115. Daly RM, Brown M, Bass S, Kukuljan S, Nowson C. Calcium- and vitamin D3-fortified milk reduces bone loss at clinically relevant skeletal sites in older men: a 2-year randomized controlled trial. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 2006;21:397-405.
116. Law M, Withers H, Morris J, Anderson F. Vitamin D supplementation and the prevention of fractures and falls: results of a randomised trial in elderly people in residential accommodation. Age and ageing 2006;35:482-6.
117. Lonn E, Yusuf S, Arnold MJ et al. Homocysteine lowering with folic acid and B vitamins in vascular disease. The New England journal of medicine 2006;354:1567-77.
118. Magliano D, McNeil J, Branley P et al. The Melbourne Atherosclerosis Vitamin E Trial (MAVET): a study of high dose vitamin E in smokers. European journal of cardiovascular prevention and rehabilitation : official journal of the European Society of Cardiology, Working Groups on Epidemiology & Prevention and Cardiac Rehabilitation and Exercise Physiology 2006;13:341-7.
119. Schleithoff SS, Zittermann A, Tenderich G, Berthold HK, Stehle P, Koerfer R. Vitamin D supplementation improves cytokine profiles in patients with congestive heart failure: a double-blind, randomized, placebo-controlled trial. The American journal of clinical nutrition 2006;83:754-9.
120. Stranges S, Marshall JR, Trevisan M et al. Effects of selenium supplementation on cardiovascular disease incidence and mortality: secondary analyses in a randomized clinical trial. American journal of epidemiology 2006;163:694-9.
121. Zoungas S, McGrath BP, Branley P et al. Cardiovascular morbidity and mortality in the Atherosclerosis and Folic Acid Supplementation Trial (ASFAST) in chronic renal failure: a multicenter, randomized, controlled trial. Journal of the American College of Cardiology 2006;47:1108-16.
122. Bolton-Smith C, McMurdo ME, Paterson CR et al. Two-year randomized controlled trial of vitamin K1 (phylloquinone) and vitamin D3 plus calcium on the bone health of older women. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 2007;22:509-19.
123. Catalano M, Born G, Peto R. Prevention of serious vascular events by aspirin amongst patients with peripheral arterial disease: randomized, double-blind trial. Journal of internal medicine 2007;261:276-84.
124. Cole BF, Baron JA, Sandler RS et al. Folic acid for the prevention of colorectal adenomas: a randomized clinical trial. Jama 2007;297:2351-9.
125. Cook NR, Albert CM, Gaziano JM et al. A randomized factorial trial of vitamins C and E and beta carotene in the secondary prevention of cardiovascular events in women: results from the Women's Antioxidant Cardiovascular Study. Archives of internal medicine 2007;167:1610-8.
126. Durga J, van Boxtel MP, Schouten EG et al. Effect of 3-year folic acid supplementation on cognitive function in older adults in the FACIT trial: a randomised, double blind, controlled trial. Lancet (London, England) 2007;369:208-16.
152
127. Jamison RL, Hartigan P, Kaufman JS et al. Effect of homocysteine lowering on mortality and vascular disease in advanced chronic kidney disease and end-stage renal disease: a randomized controlled trial. Jama 2007;298:1163-70.
128. Lappe JM, Travers-Gustafson D, Davies KM, Recker RR, Heaney RP. Vitamin D and calcium supplementation reduces cancer risk: results of a randomized trial. The American journal of clinical nutrition 2007;85:1586-91.
129. Liu BA, McGeer A, McArthur MA et al. Effect of multivitamin and mineral supplementation on episodes of infection in nursing home residents: a randomized, placebo-controlled study. Journal of the American Geriatrics Society 2007;55:35-42.
130. Lyons RA, Johansen A, Brophy S et al. Preventing fractures among older people living in institutional care: a pragmatic randomised double blind placebo controlled trial of vitamin D supplementation. Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 2007;18:811-8.
131. Plummer M, Vivas J, Lopez G et al. Chemoprevention of precancerous gastric lesions with antioxidant vitamin supplementation: a randomized trial in a high-risk population. Journal of the National Cancer Institute 2007;99:137-46.
132. Smith H, Anderson F, Raphael H, Maslin P, Crozier S, Cooper C. Effect of annual intramuscular vitamin D on fracture risk in elderly men and women--a population-based, randomized, double-blind, placebo-controlled trial. Rheumatology (Oxford, England) 2007;46:1852-7.
133. Vianna AC, Mocelin AJ, Matsuo T et al. Uremic hyperhomocysteinemia: a randomized trial of folate treatment for the prevention of cardiovascular events. Hemodialysis international International Symposium on Home Hemodialysis 2007;11:210-6.
134. Albert CM, Cook NR, Gaziano JM et al. Effect of folic acid and B vitamins on risk of cardiovascular events and total mortality among women at high risk for cardiovascular disease: a randomized trial. Jama 2008;299:2027-36.
135. Bjorkman M, Sorva A, Risteli J, Tilvis R. Vitamin D supplementation has minor effects on parathyroid hormone and bone turnover markers in vitamin D-deficient bedridden older patients. Age and ageing 2008;37:25-31.
136. Bolland MJ, Barber PA, Doughty RN et al. Vascular events in healthy older women receiving calcium supplementation: randomised controlled trial. BMJ (Clinical research ed) 2008;336:262-6.
137. Maraini G, Williams SL, Sperduto RD et al. A randomized, double-masked, placebo-controlled clinical trial of multivitamin supplementation for age-related lens opacities. Clinical trial of nutritional supplements and age-related cataract report no. 3. Ophthalmology 2008;115:599-607.e1.
138. Ebbing M, Bleie O, Ueland PM et al. Mortality and cardiovascular events in patients treated with homocysteine-lowering B vitamins after coronary angiography: a randomized controlled trial. Jama 2008;300:795-804.
139. Guyton JR, Brown BG, Fazio S, Polis A, Tomassini JE, Tershakovec AM. Lipid-altering efficacy and safety of ezetimibe/simvastatin coadministered with extended-release niacin in patients with type IIa or type IIb hyperlipidemia. Journal of the American College of Cardiology 2008;51:1564-72.
140. Logan RF, Grainge MJ, Shepherd VC, Armitage NC, Muir KR. Aspirin and folic acid for the prevention of recurrent colorectal adenomas. Gastroenterology 2008;134:29-38.
141. Milman U, Blum S, Shapira C et al. Vitamin E supplementation reduces cardiovascular events in a subgroup of middle-aged individuals with both type 2 diabetes mellitus and the haptoglobin 2-2 genotype: a prospective double-blinded clinical trial. Arteriosclerosis, thrombosis, and vascular biology 2008;28:341-7.
153
142. Prince RL, Austin N, Devine A, Dick IM, Bruce D, Zhu K. Effects of ergocalciferol added to calcium on the risk of falls in elderly high-risk women. Archives of internal medicine 2008;168:103-8.
143. Reid IR, Ames R, Mason B et al. Randomized controlled trial of calcium supplementation in healthy, nonosteoporotic, older men. Archives of internal medicine 2008;168:2276-2282.
144. Sesso HD, Buring JE, Christen WG et al. Vitamins E and C in the prevention of cardiovascular disease in men: the Physicians' Health Study II randomized controlled trial. Jama 2008;300:2123-33.
145. Zhu K, Devine A, Dick IM, Wilson SG, Prince RL. Effects of calcium and vitamin D supplementation on hip bone mineral density and calcium-related analytes in elderly ambulatory Australian women: a five-year randomized controlled trial. The Journal of clinical endocrinology and metabolism 2008;93:743-9.
146. Anker SD, Comin Colet J, Filippatos G et al. Ferric carboxymaltose in patients with heart failure and iron deficiency. The New England journal of medicine 2009;361:2436-48.
147. Hodis HN, Mack WJ, Dustin L et al. High-dose B vitamin supplementation and progression of subclinical atherosclerosis: a randomized controlled trial. Stroke 2009;40:730-6.
148. Imasa MS, Gomez NT, Nevado JB, Jr. Folic acid-based intervention in non-ST elevation acute coronary syndromes. Asian cardiovascular & thoracic annals 2009;17:13-21.
149. LaCroix AZ, Kotchen J, Anderson G et al. Calcium plus vitamin D supplementation and mortality in postmenopausal women: the Women's Health Initiative calcium-vitamin D randomized controlled trial. The journals of gerontology Series A, Biological sciences and medical sciences 2009;64:559-67.
150. Lippman SM, Klein EA, Goodman PJ et al. Effect of selenium and vitamin E on risk of prostate cancer and other cancers: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). Jama 2009;301:39-51.
151. Saposnik G, Ray JG, Sheridan P, McQueen M, Lonn E. Homocysteine-lowering therapy and stroke risk, severity, and disability: additional findings from the HOPE 2 trial. Stroke 2009;40:1365-72.
152. Sang ZC, Wang F, Zhou Q et al. Combined use of extended-release niacin and atorvastatin: safety and effects on lipid modification. Chinese medical journal 2009;122:1615-20.
153. Wu K, Platz EA, Willett WC et al. A randomized trial on folic acid supplementation and risk of recurrent colorectal adenoma. The American journal of clinical nutrition 2009;90:1623-31.
154. Armitage JM, Bowman L, Clarke RJ et al. Effects of homocysteine-lowering with folic acid plus vitamin B12 vs placebo on mortality and major morbidity in myocardial infarction survivors: a randomized trial. Jama 2010;303:2486-94.
155. Chailurkit LO, Saetung S, Thakkinstian A, Ongphiphadhanakul B, Rajatanavin R. Discrepant influence of vitamin D status on parathyroid hormone and bone mass after two years of calcium supplementation. Clinical endocrinology 2010;73:167-72.
156. de Zeeuw D, Agarwal R, Amdahl M et al. Selective vitamin D receptor activation with paricalcitol for reduction of albuminuria in patients with type 2 diabetes (VITAL study): a randomised controlled trial. Lancet (London, England) 2010;376:1543-51.
157. Galan P, Kesse-Guyot E, Czernichow S, Briancon S, Blacher J, Hercberg S. Effects of B vitamins and omega 3 fatty acids on cardiovascular diseases: a randomised placebo controlled trial. BMJ (Clinical research ed) 2010;341:c6273.
158. Heinz J, Kropf S, Domrose U et al. B vitamins and the risk of total mortality and cardiovascular disease in end-stage renal disease: results of a randomized controlled trial. Circulation 2010;121:1432-8.
154
159. Hercberg S, Kesse-Guyot E, Druesne-Pecollo N et al. Incidence of cancers, ischemic cardiovascular diseases and mortality during 5-year follow-up after stopping antioxidant vitamins and minerals supplements: a postintervention follow-up in the SU.VI.MAX Study. International journal of cancer 2010;127:1875-81.
160. House AA, Eliasziw M, Cattran DC et al. Effect of B-vitamin therapy on progression of diabetic nephropathy: a randomized controlled trial. Jama 2010;303:1603-9.
161. Janssen HC, Samson MM, Verhaar HJ. Muscle strength and mobility in vitamin D-insufficient female geriatric patients: a randomized controlled trial on vitamin D and calcium supplementation. Aging clinical and experimental research 2010;22:78-84.
162. Salovaara K, Tuppurainen M, Karkkainen M et al. Effect of vitamin D(3) and calcium on fracture risk in 65- to 71-year-old women: a population-based 3-year randomized, controlled trial--the OSTPRE-FPS. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 2010;25:1487-95.
163. Sanders KM, Stuart AL, Williamson EJ et al. Annual high-dose oral vitamin D and falls and fractures in older women: a randomized controlled trial. Jama 2010;303:1815-22.
164. Sanyal AJ, Chalasani N, Kowdley KV et al. Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis. The New England journal of medicine 2010;362:1675-85.
165. B vitamins in patients with recent transient ischaemic attack or stroke in the VITAmins TO Prevent Stroke (VITATOPS) trial: a randomised, double-blind, parallel, placebo-controlled trial. The Lancet Neurology 2010;9:855-65.
166. Boden WE, Probstfield JL, Anderson T et al. Niacin in patients with low HDL cholesterol levels receiving intensive statin therapy. The New England journal of medicine 2011;365:2255-67.
167. Cherniack EP, Florez HJ, Hollis BW, Roos BA, Troen BR, Levis S. The response of elderly veterans to daily vitamin D3 supplementation of 2,000 IU: a pilot efficacy study. Journal of the American Geriatrics Society 2011;59:286-90.
168. Grimnes G, Figenschau Y, Almas B, Jorde R. Vitamin D, insulin secretion, sensitivity, and lipids: results from a case-control study and a randomized controlled trial using hyperglycemic clamp technique. Diabetes 2011;60:2748-57.
169. Lewis JR, Calver J, Zhu K, Flicker L, Prince RL. Calcium supplementation and the risks of atherosclerotic vascular disease in older women: results of a 5-year RCT and a 4.5-year follow-up. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 2011;26:35-41.
170. Gallagher JC, Sai A, Templin T, 2nd, Smith L. Dose response to vitamin D supplementation in postmenopausal women: a randomized trial. Annals of internal medicine 2012;156:425-37.
171. Marshall JR, Tangen CM, Sakr WA et al. Phase III trial of selenium to prevent prostate cancer in men with high-grade prostatic intraepithelial neoplasia: SWOG S9917. Cancer prevention research (Philadelphia, Pa) 2011;4:1761-9.
172. Alvarez JA, Law J, Coakley KE et al. High-dose cholecalciferol reduces parathyroid hormone in patients with early chronic kidney disease: a pilot, randomized, double-blind, placebo-controlled trial. The American journal of clinical nutrition 2012;96:672-9.
173. Glendenning P, Zhu K, Inderjeeth C, Howat P, Lewis JR, Prince RL. Effects of three-monthly oral 150,000 IU cholecalciferol supplementation on falls, mobility, and muscle strength in older postmenopausal women: a randomized controlled trial. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 2012;27:170-6.
174. Lehouck A, Mathieu C, Carremans C et al. High doses of vitamin D to reduce exacerbations in chronic obstructive pulmonary disease: a randomized trial. Annals of internal medicine 2012;156:105-14.
155
175. Ma JL, Zhang L, Brown LM et al. Fifteen-year effects of Helicobacter pylori, garlic, and vitamin treatments on gastric cancer incidence and mortality. Journal of the National Cancer Institute 2012;104:488-92.
176. Punthakee Z, Bosch J, Dagenais G et al. Design, history and results of the Thiazolidinedione Intervention with vitamin D Evaluation (TIDE) randomised controlled trial. Diabetologia 2012;55:36-45.
177. Sesso HD, Christen WG, Bubes V et al. Multivitamins in the prevention of cardiovascular disease in men: the Physicians' Health Study II randomized controlled trial. Jama 2012;308:1751-60.
178. Delanaye P, Weekers L, Warling X et al. Cholecalciferol in haemodialysis patients: a randomized, double-blind, proof-of-concept and safety study. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association 2013;28:1779-86.
179. Hewitt NA, O'Connor AA, O'Shaughnessy DV, Elder GJ. Effects of cholecalciferol on functional, biochemical, vascular, and quality of life outcomes in hemodialysis patients. Clinical journal of the American Society of Nephrology : CJASN 2013;8:1143-9.
180. Lamas GA, Boineau R, Goertz C et al. Oral high-dose multivitamins and minerals after myocardial infarction: a randomized trial. Annals of internal medicine 2013;159:797-805.
181. Manning PJ, Sutherland WH, Williams SM et al. The effect of lipoic acid and vitamin E therapies in individuals with the metabolic syndrome. Nutrition, metabolism, and cardiovascular diseases : NMCD 2013;23:543-9.
182. Prentice RL, Pettinger MB, Jackson RD et al. Health risks and benefits from calcium and vitamin D supplementation: Women's Health Initiative clinical trial and cohort study. Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 2013;24:567-80.
183. Witham MD, Price RJ, Struthers AD et al. Cholecalciferol treatment to reduce blood pressure in older patients with isolated systolic hypertension: the VitDISH randomized controlled trial. JAMA internal medicine 2013;173:1672-9.
184. Witham MD, Dove FJ, Khan F, Lang CC, Belch JJ, Struthers AD. Effects of vitamin D supplementation on markers of vascular function after myocardial infarction--a randomised controlled trial. Int J Cardiol 2013;167:745-9.
185. Landray MJ, Haynes R, Hopewell JC et al. Effects of extended-release niacin with laropiprant in high-risk patients. The New England journal of medicine 2014;371:203-12.
186. Wang AY, Fang F, Chan J et al. Effect of paricalcitol on left ventricular mass and function in CKD--the OPERA trial. Journal of the American Society of Nephrology : JASN 2014;25:175-86.
187. van Wijngaarden JP, Swart KM, Enneman AW et al. Effect of daily vitamin B-12 and folic acid supplementation on fracture incidence in elderly individuals with an elevated plasma homocysteine concentration: B-PROOF, a randomized controlled trial. The American journal of clinical nutrition 2014;100:1578-86.
188. Wang L, Sesso HD, Glynn RJ et al. Vitamin E and C supplementation and risk of cancer in men: posttrial follow-up in the Physicians' Health Study II randomized trial. The American journal of clinical nutrition 2014;100:915-23.
189. Huo Y, Li J, Qin X et al. Efficacy of folic acid therapy in primary prevention of stroke among adults with hypertension in China: the CSPPT randomized clinical trial. Jama 2015;313:1325-35.
190. Ponikowski P, van Veldhuisen DJ, Comin-Colet J et al. Beneficial effects of long-term intravenous iron therapy with ferric carboxymaltose in patients with symptomatic heart failure and iron deficiencydagger. European heart journal 2015;36:657-68.
156
191. Wang LL, H. Zhou, Y. Jin, L. Liu, J. Low-dose B vitamins supplementation ameliorates cardiovascular risk: a double-blind randomized controlled trial in healthy Chinese elderly. Eur J Nutr 2015;54:455-64.
192. van Dijk SC, Enneman AW, Swart KM et al. Effects of 2-year vitamin B12 and folic acid supplementation in hyperhomocysteinemic elderly on arterial stiffness and cardiovascular outcomes within the B-PROOF trial. Journal of hypertension 2015;33:1897-906; discussion 1906.
193. Gupta A, Prabhakar S, Modi M et al. Effect of Vitamin D and calcium supplementation on ischaemic stroke outcome: a randomised controlled open-label trial. International journal of clinical practice 2016;70:764-70.
194. Lappe J, Watson P, Travers-Gustafson D et al. Effect of Vitamin D and Calcium Supplementation on Cancer Incidence in Older Women: A Randomized Clinical Trial. Jama 2017;317:1234-1243.