Vegan For Life
by Jack Norris, RD &
Ginny Messina, MPH, RD

For Updates:

Disease Markers of Vegetarians

Cholesterol & Blood Lipids | Blood Pressure | Body Weight

Last updated: December 2009

Numerous studies have measured cholesterol levels, blood pressure, obesity, and other markers of disease in vegans. Most of these studies included information on lacto-ovo vegetarians (Lacto-Ovo), fish-eaters (Pesco), and non-vegetarians (Non-Veg). This article surveys those published since 1980. Not much was published on vegans before that time.

Contents

Blood Lipids
Blood Pressure
- Why Do Vegans Have Lower Blood Pressure?
Body Mass Index
Body Fat
Homocysteine
Conclusion
References

Blood Lipids

Lipids are fat-soluble substances, including cholesterol and fatty acids. Blood lipid measurements generally include total cholesterol, LDL cholesterol, HDL cholesterol, and triglycerides.

Table 1. Cholesterol in USA Vegans^{2, 4, 5, 11, 13}
	Vegan	Lacto-Ovo	Non-Veg
Cholesterol (mg/dl)	146.4 (135)^a	177.3 (174)^a	194.2 (117)^a
LDL (mg/dl)	67.1 (110)^a	103.0 (149)^a	117.0 (10)^a
HDL (mg/dl)	46.0 (135)^a	50.2 (174)^a	52.7 (81)^a
Cholesterol : HDL	3.2	3.5	3.7
^aNumber of people measured

Total cholesterol is a measure of all the various types of cholesterol in the blood. Cholesterol can be separated according to the lipoprotein that carries it in the blood. Cholesterol carried by low-density lipoproteins (LDL) is considered "bad" because it tends to be deposited on the artery walls, causing heart disease. Cholesterol carried by high-density lipoproteins (HDL) is considered "good" because it tends to be taken to the liver where it is then broken down or excreted into the digestive tract in the form of bile. Dietary fiber (especially soluble) can then bind to some and it will be excreted in the stool.

There are also other lipoproteins, such as very low density lipoproteins (VLDL). They will not be examined here as they have not been measured in many vegans.

Total Cholesterol in Vegans in the United States

There have been five studies measuring the cholesterol levels of vegans in the USA since 1980. Of them, the lowest average finding was 135 mg/dl. The data from all 5 studies is compiled in Table 1. The total cholesterol of the 135 vegans averaged out to 146 mg/dl.

Table 2. Cholesterol in Western Vegans^{1- 17}
	Vegan	Lacto-Ovo	Pesco	Non-Veg
Cholesterol (mg/dl)	160.0 (749)^a	185.3 (2,135)^a	196.2 (467)^a	201.5 (1,857)^a
LDL (mg/dl)	90.3 (398)^a	106.0 (1,859)^a	113.7 (467)^a	120.9 (1,516)^a
HDL (mg/dl)	51.9 (472)^a	56.7 (1,898)^a	61.0 (467)^a	55.0 (1,587)^a
Cholesterol : HDL	3.1	3.3	3.2	3.7
^aNumber of people measured

Total Cholesterol in Western Vegans

The cholesterol levels of vegans living in Western countries have been measured in 17 studies since 1980. The results (including the 5 studies of U.S. vegans in Table 1) are listed in Table 2. The average cholesterol level of vegans was 160 compared to 202 mg/dl for non-vegetarians (Non-Veg).

In 13 studies, a statistically significant (SS) difference was found between the vegans and Non-Veg (1, 2, 4-8, 10-12, 14, 16, 17). Two studies did not find a SS difference (3,13). One study found a SS difference for the men only (9), and the remaining study did not include any non-vegans (15).

Total Cholesterol: Do Vegan Men and Women Differ?

Of the studies measuring total cholesterol levels of vegans, nine reported totals for men separately (2-4, 7-9, 11, 12, 16), and eight reported totals for women separately (2-4, 7-9, 11, 12). The details are in Table 3. There is practically no difference between male and female adult vegans.

Table 3. Cholesterol in Male & Female Vegans
	Vegan	Lacto-Ovo	Pesco	Non-Veg
Men (mg/dl)	157.3 (340)^a	172.3 (368)^a	215.8 (26)^a	189.2 (398)^a
Women (mg/dl)	156.6 (128)^a	188.1 (138)^a	220.5 (26)^a	196.0 (98)^a
^aNumber of people measured

Why Are Vegans' Cholesterol Levels Lower?

The largest study looking at vegans' cholesterol levels (of 233 men) (16) did not attempt to determine why vegans had lower levels. The second largest study, the Oxford Vegetarian study, was summarized in a 1999 paper (18). They found that total cholesterol decreased as fiber intake increased, and that total cholesterol increased as the Keys Dietary Score increased. The Keys Dietary Score predicts that saturated fat and cholesterol intake will increase, while polyunsaturated fat will decrease, total cholesterol levels (it does not account for monounsaturated fat). Thus it appears that higher fiber and polyunsaturated fat intake, and lower cholesterol and saturated fat intake, is at least partially responsible for lower cholesterol levels in vegans.

LDL & HDL Cholesterol of Western Vegans

Table 2 shows the numbers for eleven studies reporting LDL (bad) cholesterol in vegans (2, 3, 5, 6, 8-12, 14, 15). The differences in cholesterol levels between vegans and other groups can largely be explained by differences in LDL cholesterol.

There is some concern that vegan diets may lower HDL (good) cholesterol levels. Table 2 includes 14 studies reporting HDL cholesterol levels in vegans (2-6, 8-15, 17), and shows this concern to be largely unfounded. Fish-eaters appear to have higher HDL levels than other diet groups (a statistically significant finding in one large study of 415 fish-eaters) (6). However, when the ratio of total cholesterol to HDL is calculated, vegans have the lowest ratio, as noted in Table 2. This indicates that vegans' somewhat lower HDL is more of a function of their low overall cholesterol rather than an unhealthy lack of HDL cholesterol.

Table 4. Triglycerides in Western Vegans ^{2-5, 9-11, 13-15, 17}
	Vegan	Lacto-Ovo	Non-Veg
Triglycerides (mg/dl)	86.5 (310)^a	107.6 (253)^a	107.5 (281)^a
^aNumber of people measured

Triglycerides

Elevated triglycerides are generally thought to increase the risk for heart disease. However, there is a debate as to whether moderately high triglycerides are merely associated with other risk factors for heart disease, while not being a cause in themselves. Normal triglycerides for adults is 40-160 mg/dl for men and 35-135 mg/dl for women (20). Triglyceride levels above 250 mg/dl are more of a concern (20).

Some people are concerned that, although a vegan diet can lower cholesterol levels, it may increase triglyceride levels. As can be seen from Table 4, in the 11 studies that measured triglycerides, vegans were shown to have lower levels than Lacto-Ovo and Non-Veg.

Summary

The total cholesterol of Western vegans averages out to 160 mg/dl. This is 40 points lower than the non-vegetarians in these studies and well below the "desirable" level of less than 200 mg/dl according to the National Heart Lung and Blood Institute.

It is possible to eat a vegan diet that is high in fat and hydrogentaed oils and is highly processed with little fiber. This sort of diet might not provide the benefits seen in the studies compiled above. Additionally, some people have a strong genetic predisposition to high cholesterol. The College of American Pathologists recommends that people over the age of 20 have their cholesterol levels checked every 5 years (18).

Blood Pressure

Table 11. Relative Rates of High Blood Pressure in AHS-2 (2009)³⁹
Diet	Risk
Non-Veg	1.00
Semi-Veg	.77 (.72-.82)
Pesco	.62 (.59-.66)
Lacto-Ovo	.45 (.44-.47)
Vegan	.25 (.22-.28)
Adjusted for age, gender, and race.

In 2009, preliminary cross-sectional date from Adventist Health Study-2 was reported. Relative rates for having high blood pressure are show in Table 11. Vegans had a considerably lower rate of high blood pressure. Results were not adjusted for smoking.

In 2012, a more thorough, cross-sectional report was published from Adventist Health Study-2. It included only white people, and results didn't appear to be adjusted for anything. Rates for having high blood pressure are shown in Table 12; vegans had considerably lower rates of blood pressure.

Table 12. Relative Rates of High Blood Pressure in AHS-2 (2012)⁴⁰
Diet	Risk
Non-Veg	1.00
Semi-Veg & Pesco	.92 (.50-.79)
Lacto-Ovo	.57 (.36-.92)
Vegan	.37 (.19-.74)

In 2002, a study was published in which 11,004 participants of the EPIC-Oxford study were asked if they had high blood pressure (22). Results are shown in Table 5.

Table 5. High Blood Pressure in EPIC-Oxford (2002)²²
High BP	Vegan	Lacto-Ovo	Pesco	Non-Veg
Men	6% (272)^a	10% (786)^a	10% (297)^a	15% (996)^a
Women	8% (467)^a	9% (3,014)^a	10% (1,431)^a	12% (3,741)^a
^aNumber of people measured

The lower percentage of vegans with high blood pressure was statistically significant. This is the only study that has compared the percentage of vegans with high blood pressure to other diet groups.

Blood pressure was then measured in 8,663 participants who did not have high blood pressure. Those results are in Table 6. Results from the 4 other studies measuring blood pressure in vegans since 1980 are also in Table 6. Finally, the combined results of all 5 studies are listed in Table 6.

Table 6. Blood Pressure in Vegans (mm Hg)
	Vegan	Lacto-Ovo	Pesco	Non-Veg
EPIC-Oxford²²
Systole / Diastole	119 / 73.4 (612)^a	121 / 74.5 (3,233)^a	120 / 73.5 (1,404)^a	121 / 74.8 (3,524)^a
All Studies Except EPIC-Oxford^{7, 11, 12, 21}
Systole / Diastole	121 / 76.9 (104)^a	127 / 78.0 (192)^a		123 / 78.7 (100)^a
All Studies^{7, 11, 12, 21, 22}
Systole / Diastole	120 / 73.9 (716)^a	121 / 74.7 (3,425)^a	120 / 73.5 (1,404)^a	121 / 74.9 (3,624)^a
^aNumber of people measured

The results show that vegans have slightly lower blood pressures than those in other diet groups. If the data from Epic-Oxford had included all the participants, rather than only those without high blood pressure, the differences between the vegans and Non-Veg in Table 6 would have been larger. The difference also might have been larger if the participants in one study (7) had randomly chosen non-vegetarians to participate, rather than choosing non-vegetarians with a similar BMI to the vegans in the study.

Why Do Vegans Have Lower Blood Pressure?

EPIC-Oxford (22) and Adventist Health Study-2 (40) found lower Body Mass Index to explain most of the differences in blood pressure among the diet groups. Other contributory factors could be higher consumption of potassium, lower consumption of sodium, modulation of baroreceptor sensitivity, direct vasodilatory effects, changes in catecholamine and renin–angiotensin–aldosterone metabolism, improvement of glucose tolerance with lower insulin levels, and lower blood viscosity in vegetarians (40).

Body Mass Index

Body mass index (BMI) is measured by taking one's weight in kilograms and dividing it by their height in meters squared (i.e., kg/m²). It is a way of measuring weight while taking into consideration differences in height. A healthy BMI is considered to be between 20 and 25. Generally, a BMI of 30 or higher is considered obese (32).

Recent research has shown that a BMI of 22.5 to 25.0 is associated with the lowest mortality rate. It has been known for some time that a lower BMI has been associated with an increased risk of death, but that was thought to be due mostly to smoking-related diseases. A 2009 meta-analysis of 900,000 people found that even in those who never smoked, there is a slight increase in mortality below a BMI of 22.5 (37). The excess mortality below 22.5 has not been explained. One theory is that the excess mortality might be due to lower fat-free mass, which would most likely be lower muscle mass (though could also technically be bones, or even some organs) (37, 38). Studies on BMI and mortality to date have not differentiated between fat and fat-free body mass.

2009 Report From Adventist Health Study-2

In 2009, cross-sectional data on BMI were released from the Adventist health Study-2 (36). Vegans had a lower BMI than all other diet groups, and the finding was statistically significant.

Table 7. Body Mass Index in Adventist Health Study-2 (2009)³⁶
	Vegan	Lacto-Ovo	Pesco	Semi-Veg	Non-veg
BMI (kg/m²)	23.6^b (2,731)^a	25.7 (20,408)^a	26.3 (5,617)^a	27.3 (3,386)^a	28.8 (28,761)^a
BMI - Body Mass Index \| Semi-Veg - eat red meat and poultry ≥ 1 time per month and < 1 time per week \| ^aNumber of people measured \| ^bStatistically significant across diet groups \| Results were not adjusted

2003 Report From EPIC-Oxford

A report on BMI from EPIC-Oxford was published in 2003. Results are in Table 8.

Table 8. Body Mass Index in EPIC-Oxford (2003)³⁴
	Vegan	Lacto-Ovo	Pesco	Non-Veg
Male	22.5 (570)^a	23.4 (2,888)^a	23.4 (1,095)^a	24.4 (4,318)^a
Female	22.0 (983)^a	22.7 (9,419)^a	22.7 (5,096)^a	23.5 (13,506)^a
Body Mass Index = kg/m² \| Adjusted for age, smoking, education level, physical activity, marital status, ethnicity, parity. \| Differences between the vegans and other groups were statistically significant. \| ^aNumber of people measured

The differences between the vegans and meat-eaters was accounted for mostly by differences in protein, polyunsaturated fat, and fiber intake. The authors note that protein intake's influence on weight has not been reported often in the literature, but there is some mention of it altering hormones in a way that increases abdominal fat. They also note that low fiber intakes have been previously associated with higher body weight and this is thought to be via making people feel full on less calories, insulin control, and reducing fat absorption.

Table 9. Body Mass Index in Western Vegans Thru 2003
	Vegan	Lacto-Ovo	Pesco	Non-Veg
Height and Weight Measured by Researchers^{7-12, 14, 15, 17, 23-26, 28, 29}
Male	22.2 (84)^a	24.3 (112)^a)	22.0 (4)^a	23.4 (33)^a
Female	22.1 (78)^a	25.9 (109)^a	22.0 (5)^a	22.1 (37)^a
M & F	22.1 (279)^a	24.2 (334)^a	22.0 (9)^a	24.2 (212)^a
Height and Weight Reported by Participants Using a Questionnaire^{8, 27, 30}
Male	22.5 (974)^a	23.5 (3,994)^a	23.6 (1,682)^a	24.9 (7,544)^a
Female	21.9 (1,617)^a	22.7 (14,017)^a	22.9 (8,060)^a	24.3 (25,159)^a
M & F	22.1 (2,591)^a	22.9 (18,011)^a	23.0 (9,742)^a	24.4 (32,703)^a
Totals of All 17 Studies Above
Male	22.5 (1,058)^a	23.5 (4,106)^a	23.6 (1,686)^a	24.9 (7,577)^a
Female	21.9 (1,695)^a	22.7 (14,126)^a	22.9 (8,065)^a	24.3 (25,196)^a
M & F	22.1 (2,870)^a	22.9 (18,345)^a	23.0 (9,751)^a	24.4 (32,915)^a
Body Mass Index = kg/m² \| Note: You cannot obtain the "M & F" totals by combining the Male total with the Female total as some studies did not break down BMI according to gender. \| ^aNumber of people measured

BMI of Western Vegans in Studies up to 2003

Table 9 shows the combined results of 17 studies measuring the BMI of Western vegans up until 2003, not including the EPIC-Oxford report described above. The results for Non-Veg in two of these studies (a total of 40 Non-Veg) were not included because the researchers specifically chose Non-Veg who weighed the same as the vegans (7, 25). Additionally, a study (13) with 25 vegans was not included because the participants had to be within 120% of their ideal body weight, possibly biasing the BMI results.

Because the BMIs of many of the people measured were calculated using weights and heights reported by participants by way of questionnaires, rather than being directly measured by the researchers, the results in Table 9 are divided into two groups accordingly.

The results in Table 9 show vegans to have the lowest BMI in all scenarios. BMI for vegans were practically the same regardless of whether reported by questionnaire or measured by researchers. The BMIs for participants using a questionnaire are the same as the totals of all 17 studies.

The largest study of 2,488 vegans and 32,594 Non-Veg found a statistically significant difference between the vegans and Non-Veg (30, 33).

Because all of the studies above on BMI were cross-sectional, it is possible that the differences could be explained by leaner individuals being more likely to adopt a vegan diet, rather than a vegan diet causing people to be more lean.

Body Mass Index Over Time as a Vegan

A 1996 letter to the editor of the British Medical Journal from the authors of the EPIC-Oxford study (31) reported BMI according to the time on current diet (less than or greater than 5 years). The number in each group were:

1,652 Vegan
8,827 Lacto-Ovo
3,776 Pesco
6,850 Non-Veg

The actual BMIs were not given, but a graph was provided (which can be viewed at bmj.com/cgi/content/full/313 /7060/816/F1). The graph shows that those on a vegan diet for more than 5 years had the lowest BMI, followed by those on a vegan diet for less than 5 years, for both men and women. This is impressive, as most weight loss is not sustained for more than one year. Of course, weight loss can sometimes be difficult even for vegans, and some people actually gain weight after becoming vegan. But, on average, the evidence supports the notion that becoming vegan is conducive to permanent weight loss.

In 2006, a report from EPIC-Oxford (35) showed that over a 5 year period, vegans had the lowest weight-gain compared to meat-eaters, fish-eaters, and lacto-ovo vegetarians. The group who had switched to a diet of eating less animal products had the lowest weight gain of all. The group of people who reverted to a diet of more animal products had the most weight-gain, but this was not statistically significant. All groups had some weight gain over the 5 year period.

Body Fat

What does it matter if vegans weigh less if they simply have less muscle mass? Above (Table 9) we saw that vegans have an average BMI of about 22.2 to 22.5, which is right in the middle of the healthy range of 20 to 25. So, vegans are not too thin. But what if a lower percentage of their body weight is muscle (which would mean that a higher percentage of their body weight is fat)?

Table 10 lists the studies that measured percentage of body fat or skinfold thickness (an indicator of body fat) in vegans. Determining the percentage of body fat can vary greatly from method to method, so averaging the results would not be appropriate. Instead, we should look at the general trend. In the 5 comparisons made, the vegans had lower body fat in all five. In three of those comparisons, the differences were statistically significant.

Table 10. Percentage of Body Fat in Vegans
Year	Country	Vegan	Lacto-Ovo	Non-Veg	Adjustments	SS
Body Fat
1996¹⁰	UK	21.9% 21 M & 17 F	22.5% 6 M & 5 F	24.0% 20 M & 19 F	None. Vegans had more men.	Not SS^a
1995²⁸	Canada	23.7% 8 F	24.1% 15 F	27.4% 22 F	Vegans slightly older.	P < .05^a
1990²⁶	USA	8.0% 15 M	8.6% 20 M	14.0% 18 M	Vegans were older.	P < .05^a
Tricep Skinfold Thickness
1987⁷	UK	8.8 mm 11 M 13.5 mm 11 F		10.7 mm 11 M 17.3 mm 11 F	Matched for age, body build. Energy intake did not differ.	NR
Sum of skinfold measurements
1978¹	UK	43 mm 12 M & 10 F		76 mm 12M & 10 F	Matched for age, height, ethnic, socio economic status.	P < .01^a
F - female \| M - male \| NR - not reported \| SS - statistical significance \| P - the % chance that the finding was due to random chance \| ^aStatistically significant between vegans and non-veg

So, we now know that vegans have lower BMIs and they also tend to have lower body fat percentage (though the numbers measured are small).

Homocysteine

There is one marker of disease that has recently garnered much inerest: blood homocysteine levels. Elevated homocysteine levels are associated with heart disease, stroke, and early death. Numerous studies have looked at homocysteine in vegans and indicated that if vegans are not taking vitamin B12, they probably have high homocysteine levels. For more information, please read the chapter B12 and Chronic Disease: Homocysteine in Vitamin B12: Are You Getting It?

Conclusion

In summary, the evidence shows:

Vegans have lower total cholesterol, LDL cholesterol, and triglycerides, while having about the same HDL cholesterol as Lacto-Ovo and Non-Veg.
Vegans have lower rates of high blood pressure than Lacto-Ovo and Non-Veg.
Vegans have a lower BMI and body fat percentage than Lacto-Ovo and Non-Veg. People who have been vegan for more than 5 years have the lowest BMI of all diet groups studied here.

References

1. Sanders TA, Ellis FR, Dickerson JW. Am J Clin Nutr 1978 May;31(5):805-13.

2. Lock DR, et al. 1982 Sep;31(9):917-21.

3. Roshanai F, Sanders TA. Hum Nutr Appl Nutr. 1984 Oct;38(5):345-54.

4. Kritchevsky D, Tepper SA, Goodman G. Am J Clin Nutr. 1984 Oct;40(4 Suppl):921-6.

5. Fisher M, et al. Arch Intern Med. 1986 Jun;146(6):1193-7.

6. Thorogood M, et al. Britain.Br Med J (Clin Res Ed). 1987 Aug 8;295(6594):351-3.

7. Sanders TA, Key TJ. Hum Nutr Appl Nutr. 1987 Jun;41(3):204-11.

8. Thorogood M, et al. BMJ. 1990 May 19;300(6735):1297-301.

9. Sanders TA, Roshanai F. Eur J Clin Nutr. 1992 Nov;46(11):823-31.

10. Thomas EL, Frost G, Barnard ML, et al. Lipids. 1996 Feb;31(2):145-51.

11. Toohey ML, et al. J Am Coll Nutr. 1998 Oct;17(5):425-34.

12. Li D, et al. Eur J Clin Nutr. 1999 Aug;53(8):612-9.

13. Haddad EH, et al. Am J Clin Nutr. 1999;70(suppl):586S-93S.

14. Krajcovicova-Kudlackova M, et al. Scand J Clin Lab Invest. 2000 Dec;60(8):657-64.

15. Fokkema MR, et al. Prostaglandins Leukot Essent Fatty Acids. 2000 Nov;63(5):287-92.

16. Allen NE, et al. Br J Cancer 2000 Jul;83(1):95-7.

17. Bissoli L, et al. Ann Nutr Metab. 2002;46(2):73-9.

18. CAP. College of American Pathologists. Cholesterol Testing Information. http://www.cap.org/health_and_wellness/Cholesterol_CAP.html. Accessed February 7, 2003.

19. Appleby PN, et al. Am J Clin Nutr. 1999 Sep;70(3 Suppl):525S-531S.

20. LAB. Fischbach F. A Manual of Laboratory & Diagnostic Tests, 6th Ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2000.

21. Sacks FM, Wood PG, Kass EH. Hypertension. 1984 Mar-Apr;6(2 Pt 1):199-201.

22. Appleby PN, Davey GK, Key TJ. Hypertension and blood pressure among meat eaters, fish eaters, vegetarians and vegans in EPIC-Oxford. Public Health Nutr. 2002 Oct;5(5):645-54.

23. Abdulla M, et al. Am J Clin Nutr 1981 Nov;34(11):2464-77.

24. Carlson E, et al. J Plant Foods. 1985;6:89-100.

25. Rana SK, Sanders TA. Br J Nutr. 1986 Jul;56(1):17-27.

26. Ross JK, Pusateri DJ, Shultz TD. Am J Clin Nutr. 1990 Mar;51(3):365-70.23.

27. Key TJ, et al. Br J Nutr. 1990 Jul;64(1):111-9.

28. Janelle KC, Barr SI. J Am Diet Assoc. 1995 Feb;95(2):180-6.

29. Herrmann W, et al. Clin Chem. 2001 Jun;47(6):1094-101.

30. Davey GK, et al. Public Health Nutrition. 2003. (In Press)

31. Key T, Davey G. BMJ. 1996 Sep 28;313(7060):816-7.

32. MA. Mahan LK, Escott-Stump S. Krause's Food, Nutrition, & Diet Therapy, 10th Ed. Philadelphia, PA: W.B. Saunders, Co. 2000.

33. Personal communication with Paul Appleby. February 17, 2003.

34. Spencer EA, Appleby PN, Davey GK, Key TJ. Diet and body mass index in 38000 EPIC-Oxford meat-eaters, fish-eaters, vegetarians and vegans. Int J Obes Relat Metab Disord. 2003 Jun;27(6):728-34.

35. Rosell M, Appleby P, Spencer E, Key T. Weight gain over 5 years in 21,966 meat-eating, fish-eating, vegetarian, and vegan men and women in EPIC-Oxford. Int J Obes (Lond). 2006 Sep;30(9):1389-96. Epub 2006 Mar 14.

36. Tonstad S, Butler T, Yan R, Fraser GE. Type of vegetarian diet, body weight, and prevalence of type 2 diabetes. Diabetes Care. 2009 May;32(5):791-6. Epub 2009 Apr 7.

37. Prospective Studies Collaboration, Whitlock G, Lewington S, Sherliker P, Clarke R, Emberson J, Halsey J, Qizilbash N, Collins R, Peto R. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies. Lancet. 2009 Mar 28;373(9669):1083-96.

38. Wandell PE, Carlsson AC, Theobald H. The association between BMI value and long-term mortality. Int J Obes (Lond). 2009 May;33(5):577-82.

39. Fraser GE. Vegetarian diets: what do we know of their effects on common chronic diseases? Am J Clin Nutr. 2009 May;89(5):1607S-1612S. Epub 2009 Mar 25. Review. Erratum in: Am J Clin Nutr. 2009 Jul;90(1):248. Link

40. Pettersen BJ, Anousheh R, Fan J, Jaceldo-Siegl K, Fraser GE. Vegetarian diets and blood pressure among white subjects: results from the Adventist Health Study-2 (AHS-2). Public Health Nutr. 2012 Jan 10:1-8. [Epub ahead of print] Link