Type of Vegetarian Diet, Body Weight, And Prevalence of Type 2 Diabetes

8/2/2019 Type of Vegetarian Diet, Body Weight, And Prevalence of Type 2 Diabetes

1/6

Type of Vegetarian Diet, Body Weight, andPrevalence of Type 2 DiabetesSERENA TONSTAD, MD, PHD1

TERRY BUTLER, DRPH2RU YAN, MSC3

GARY E. FRASER, MD, PHD4

OBJECTIVE We assessed the prevalence of type 2 diabetes in people following differenttypes of vegetarian diets compared with that in nonvegetarians.

RESEARCH DESIGN AND METHODS The study population comprised 22,434men and 38,469 women who participated in the Adventist Health Study-2 conducted in 20022006. We collected self-reported demographic, anthropometric, medical history, and lifestyledata from Seventh-Day Adventist church members acrossNorth America. The type of vegetariandiet was categorized based on a food-frequency questionnaire. We calculated odds ratios (ORs)and 95% CIs using multivariate-adjusted logistic regression.

RESULTS Mean BMI was lowest in vegans (23.6 kg/m2) and incrementally higher in lacto-ovo vegetarians (25.7 kg/m2), pesco-vegetarians (26.3 kg/m2), semi-vegetarians (27.3 kg/m2),

and nonvegetarians (28.8 kg/m2

). Prevalence of type 2 diabetes increased from 2.9% in vegansto 7.6% in nonvegetarians; the prevalence was intermediate in participants consuming lacto-ovo(3.2%), pesco (4.8%), or semi-vegetarian (6.1%) diets. After adjustment for age, sex, ethnicity,education, income, physical activity, television watching, sleep habits, alcohol use, and BMI,vegans (OR 0.51 [95% CI 0.400.66]), lacto-ovo vegetarians (0.54 [0.490.60]), pesco-vegetarians (0.70 [0.610.80]), andsemi-vegetarians (0.76 [0.650.90])had a lower risk of type2 diabetes than nonvegetarians.

CONCLUSIONS The 5-unitBMI difference between vegansand nonvegetarians indicatesa substantial potential of vegetarianism to protect against obesity. Increased conformity to veg-etarian diets protected against risk of type 2 diabetes after lifestyle characteristics and BMI weretaken into account. Pesco- and semi-vegetarian diets afforded intermediate protection.

Diabetes Care 32:791796, 2009

Vegetarian diets may play a beneficialrole in promoting health and pre-venting obesity (13). Vegetarian-

ism encompasses a spectrum of eatingpatterns: from diets that leave out all ani-mal meats and products (vegan) to dietsthat include eggs, milk, and milk prod-ucts (lacto-ovo vegetarian) or even fish inaddition to eggs, milk, and milk products(pesco-vegetarian). A previous study hasindicatedthat BMI increases when a widerspectrum of animal products are eaten.Specifically, the European Prospective In-vestigation found that BMI was highest in

meat eaters, lowest in vegans, and inter-mediate in fish eaters (4). The protectiveeffects of vegetarianism against over-weight may be due to avoidance of majorfood groups, displacement of calories to-ward food groups that are more satiating(5), or other factors.

Based on a review of experimentaldata, investigators have suggested that theportfolio of foods found in vegetarian di-ets may carry metabolic advantages forthe prevention of type 2 diabetes (6). Thisnotion has been confirmed in observa-tional studies. In the Nurses Health

Study, intakes of red meat and processedmeats were associated with increased risk

of diabetes (7). In a study of Seventh-DayAdventists, diabetes was less prevalent invegetarian than in nonvegetarian church-goers (8). Likewise, Fraser reported alower prevalence of diabetes in vegetari-ans than in semi- or nonvegetarians (1)and Vang et al. (9) found that processedmeat consumption was a risk factor fordiabetes. However, these church-basedcohorts were initiated in the 1960s1970s and included primarily non-Hispanic whites. Furthermore, the type ofvegetarianism or diabetes was notspecified.

A pertinent question is whether veg-etarian diets remain protective in currentobesity-promoting environments and indiverse populations. We studied a Sev-enth-Day Adventist cohort that includeda population of whom 25% of subjectswere black and that was characterized byvegetarian and nonvegetarian eating pat-terns. We hypothesized that more exclu-sively vegetarian diets, e.g., vegan, lacto-ovo, or pesco-vegetarian, are associatedwith lower prevalence of obesity and type2 diabetes compared with semi- or non-

vegetarian diets.

RESEARCH DESIGN AND

METHODS The Adventist HealthStudy-2 cohort, initiated in 20022006,longitudinally follows 97,000 Adventistchurch members in the U.S. and Canada(10). Participants were recruited throughtheir churches and were eligible if aged30 years and proficient in English. Thestudy was reviewed and approved by theinstitutional review board of Loma LindaUniversity, Loma Linda, California, and

informed consent was obtained.These analyses are based on cross-sectional data obtained at baseline.Data were collected from a 50-page self-administered questionnaire (11).Thequestionnaire included sections on ill-ness, diet, physical activity, demograph-ics, height, and weight. Cases of diabeteswere ascertained by asking whether aphysician had ever diagnosed type 1 ortype 2 diabetes and whether the respon-dent was treated for this in the last 12months. Race and ethnicity were dividedinto the following categories: black

From the 1Department of Health Promotion and Education, School of Public Health, and the Department ofPreventive Medicine, School of Medicine, Loma Linda University, Loma Linda, California; the 2Depart-ment of Biostatistics, School of Public Health, Loma Linda University, Loma Linda, California; 3LomaLinda University, Loma Linda, California; and the 4Department of Cardiology, School of Medicine, LomaLinda University, Loma Linda, California.

Corresponding author: Serena Tonstad, [email protected] 17 October 2008 and accepted 2 February 2009.DOI: 10.2337/dc08-1886 2009 by the American Diabetes Association. Readers may use this article as long as the work is properly

cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be herebymarked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

C l i n i c a l C a r e / E d u c a t i o n / N u t r i t i o n / P s y c h o s o c i a l R e s e a r c hO R I G I N A L A R T I C L E

DIABETES CARE, VOLUME 32, NUMBER 5, MAY 2009 791


2/6

(black/African American, West Indian/Caribbean, African, or other black) andnon-black (white non-Hispanic, His-panic, Middle Eastern, Asian, NativeHawaiian/other Pacific Islander, or

American Indian). Education was catego-rized to high school or less, some college,and college or higher based on eight op-

tions. Income was categorized into earn-ings of 10,000, 11,00030,000,31,00050,000, and51,000 USD.

Assessment of lifestyle exposuresThe food-frequency portion of the ques-tionnaire covered 130 hard-coded foodsor food groups that are commonly con-sumed and space for 50 write-ins andare assessed for diet during the past year.For different food items, there were 79frequency categories. A standard portion

size was given for each item, and subjectscould selectthator a smaller or larger por-tion. Previous validation of the question-naire pertained to nutrients includingvitamin, antioxidant, and fatty acid in-takes (12,13). Vegetarian status was cate-gorized by defining vegans as subjectswho reported consuming no animalproducts (red meat, poultry, fish, eggs,milk, and dairy products 1 time/month), lacto-ovo vegetarians as consum-ing dairy products and/or eggs 1 time/month but no fish or meat (red meat,poultry, and fish1 time/month), pesco-

vegetarians as consuming fish 1 time/month and dairy products and/or eggsbut no red meat or poultry (red meatand poultry 1 time/month), semi-vegetarians as consuming dairy productsand/or eggs and meat (red meat and poul-try 1 time/month and 1 time/week),and nonvegetarians as consuming animalproducts (red meat, poultry, fish, eggs,milk, and dairy products1 time/week).

Alcohol was defined as consumption ofany amount or none during the past 12months.

Physical-activity questions were pre-viously validated in non-black and blacksubjects (14,15). These were separatedinto six intensity levels, including nap-ping, lying down, and light, moderate,vigorous, and extremely vigorous activity.Participants reported the amount of timespent in each type of activity on a normalweekday and on Saturday and Sunday.Moderate, vigorous, and extremely vigor-ous activities were assigned scores of 4, 8,and 10, respectively, to represent the ap-proximate MET values expended andwere weighted according to theamount of

time spent in each activity to estimate av-erage daily energy expenditure (informa-tion available in the online appendix[http://care.diabetesjournals.org/cgi/content/full/dc08-1886]). Participants reported aver-age number of hours of sleep and hours perdayoftelevision watching.Responsesweredi-vided into three categories (6, 7, and8 hof sleep and1, 12, and3 h of televisionwatching).

Ascertainment of diseaseA representative subgroup of 1,007 studysubjects participated in a calibrationstudy and provided blood samples formeasurement of fasting serum glucoselevels (16). The calibration sample wasgenerated by a two-stage random-selectionmethod involving church size and sub-

jects within the church. Subjects who re-fused participation were replaced withindividuals randomly chosen from thesame church and matched by race, age,and sex.

We used a fasting glucose level126mg/dl to categorize subjects as probablydiabetic. We attempted telephone inter-views with subjects who, despite glucosemeasurements 126 mg/dl, reported aphysician-based diagnosis of type 2 dia-betes and subjects whose glucose mea-surement was 126 mg/dl but who didnot report type 2 diabetes. Of subjectswho had reported type 2 diabetes but had

a low glucose level (n 55), all 44 whowere contacted confirmed that they hadtype 2 diabetes, whereas 9 could not bereached and 2 were too deaf to under-stand the question. Of subjects who didnot report type 2 diabetes but had ahigh glucose level (n 53), 38 had noknowledge of diabetes, were informed ofdiabetes by their physician after thequestionnaire was administered, were in-formed only of a high or borderline glu-cose level, or had not been treated in thepast 12 months; 1 had diabetes but hadskipped that page of the questionnaire; 6

Table 1Distribution of participants by history of type 2 diabetes treated within 12 months in

conjunction with nondietary variables

Type 2 diabetes

reported Not reported P

N 3,430 57,473 Age (years) 62.5 11.8 56.1 13.7 0.0001

Female 61.7 63.3 0.0604Black 32.5 23.4 0.0001

BMI (kg/m2) 32.1 7.1 26.9 5.7 0.0001Physical activity: METs 0.0001

03.2 38.6 24.83.28.6 23.2 24.5

8.621.0 20.6 24.6

21.0 17.6 26.1

Education 0.0001High school or less 26.0 17.9

Some college 31.9 27.6College or higher 42.1 54.6

Income 0.0001

10,000 USD 25.0 19.5

11,00030,000 USD 42.9 36.331,00050,000 USD 19.0 23.6

51,000 USD 13.1 20.6Television watching 0.0001

None to1 h/day 11.8 27.012 h/day 42.9 47.4

3 h/day 45.3 25.6Sleep 0.0001

6 h/night 39.4 31.7

7 h/night 28.8 36.8

8 h/night 31.9 31.5

Alcohol use in last 12 months 7.1 10.4 0.0001

Data are means SD or percent unless otherwise indicated. Percentages might not total 100 because of

rounding.

Vegetarian diet type, body weight, and type 2 diabetes

792 DIABETES CARE, VOLUME 32, NUMBER 5, MAY 2009


3/6

were deceased; and 8 could not bereached.

Statistical analysesThe allocation of subjects into dietary cat-egories required responses to questionsregarding 27 variables including meat,fish and poultry, dairy, and egg consump-tion. Because on average 7% data weremissing for any particular variable, multi-ple imputation was used to fill missing

information. For about half of these vari-ables, we had a random subset of initiallymissing data filled in later by telephoneand used this to guide the imputation(17). For other variables, we assumed thatthe data were missing at random, which,even if not quite correct, will have littleinfluence when the missing data rate issmall (17). The imputation algorithm in-cluded age, sex, race, and food groups(meat, fish, dairy, and eggs). These foodgroups were subdivided into blocks ofvariables for which we had the randomsubset of filled-in data and those for

which we did not (e.g., fish variables thatwere included in those later filled in ver-sus fish variables not so included) for pur-poses of imputation. The imputationsoftware used was the Hmisc package forR, version 2.6.0 (18).

2 tests and one-way ANOVA were

used to analyze categorical and continu-ous variables, respectively. Multiple logis-tic regression analysis was used to obtainORs and 95% CIs of the relation between

characteristics, diet, and diabetes. All sta-tistical testing was two-tailed. Analyseswere performed using SPSS, version 13.0.

RESULTS There were 83,031 sub-jects whose questionnaire responses al-lowed categorization by vegetarian ornonvegetarian diet. Of these, 1,427 didnot respond to the diabetes ascertainmentquestion. A further 634 who reportedtreatment for type 1 diabetes were ex-cluded, and another 14,124, 4,638, 317,852, and 136 were missing data regardingphysical activity, income, education, tele-

vision watching, and sleep, respectively.This left 22,434 men and 38,469 women.Of these 60,903 subjects, 3,430 (5.6%)reported type 2 diabetes.

Table 1 shows that participantstreated for type 2 diabetes differed withregard to a variety of characteristics fromthose who did not report diabetes diagno-sis. Only 1,070 (1.8%) participants hadsmoked one or more cigarettes daily inthe past 12 months.

The consumption of major foodgroups differed among the dietary groups(data not shown). The prevalence of type2 diabetes increased incrementally amongvegans, lacto-ovo vegetarians, pesco-vegetarians, semi-vegetarians, and non-vegetarians (Table 2). This increase wasconcomitant to an incremental increase inmean BMI in the respective dietarygroups; additionally, demographic andlifestyle characteristics differed amongthe dietary groups (Table 2). For BMIs30 kg/m2, the prevalence of diabeteswas 8.0% in vegans, 9.4% in lacto-ovo

Table 2Unadjusted prevalence of type 2 diabetes and distribution of nondietary variables according to diet

Vegan

Lacto-ovo

vegetarian

Pesco-

vegetarian

Semi-

vegetarian Nonvegetarian P

N 2,731 20,408 5,617 3,386 28,761

Type 2 diabetes 2.9 3.2 4.8 6.1 7.6 0.0001 Age in years 58.1 13.3 58.1 14.1 57.2 13.8 57.7 13.6 54.9 13.2 0.0001

Female 60.1 62.3 65.9 65.7 63.2 0.0001Black 19.9 12.5 34.9 15.0 31.2 0.0001

BMI (kg/m2) 23.6 4.4 25.7 5.1 26.3 5.2 27.3 5.7 28.8 6.3 0.0001Physical activity: METS 0.0001

03.2 24.8 26.3 24.3 26.8 25.23.28.6 24.7 25.8 24.5 24.0 23.5

8.621.0 24.8 24.6 24.6 23.7 24.3

21.0 25.7 23.3 26.6 25.6 27.1

Education 0.0001High school or less 16.7 14.0 17.2 19.1 21.7

Some college 26.7 24.2 26.1 28.5 30.7College or higher 56.6 61.8 56.7 52.4 47.6

Income 0.0001

10,000 USD 27.8 21.1 18.0 20.2 18.6

11,00030,000 USD 38.6 35.8 34.4 38.0 37.431,00050,000 USD 18.3 24.2 24.0 23.4 23.1

51,000 USD 15.3 18.9 23.6 18.4 21.0Television watching 0.0001

None to 1 h/day 49.5 36.0 26.8 25.2 16.912 h/day 37.4 45.4 50.2 48.7 48.6

3 h/day 13.2 18.6 23.0 26.1 34.5Sleep 0.0001

6 h/night 25.8 25.3 34.9 29.8 37.3

7 h/night 38.3 39.8 36.3 36.9 33.7

8 h/night 35.9 34.9 28.9 33.4 29.0

Alcohol use in last 12 months 1.1 2.9 7.1 8.6 17.1 0.0001

Data are means SD or percent unless otherwise indicated.

Tonstad and Associates



4/6

vegetarians, 10.4% in pesco-vegetarians,

11.4% in semi-vegetarians, and 13.8% innonvegetarians, indicating the same trendas that in the entire population. For BMIs30 kg/m2, the prevalence was 2.0, 2.1,3 .3 , 3 .7 , a nd 4 .6 % in t he groups,respectively.

In multiple logistic regression analysis,vegan, lacto-ovo, and pesco- and semi-vegetarian diets were associated with alower prevalence of type 2 diabetes (Table3). The vegetarian diets were more stronglyassociated with less diabetes when BMI wasremoved from the analyses (Table 3).

CONCLUSIONS The main findingwas that vegan and lacto-ovo vegetarian di-ets were associated with a nearly one-halfreduction in risk of type 2 diabetes com-pared with the risk associated with nonveg-etarian diets after adjustment for a numberof socioeconomic and lifestyle factors, aswell as lowBMI,that aretypically associatedwith vegetarianism. Pesco- and semi-vegetarian diets were associated with inter-mediate risk reductions: between one-thirdand one-quarter. These data indicate thatvegetarian diets may in part counteract the

environmental forces leadingto obesity and

increased rates of type 2 diabetes, thoughonly vegandiets were associated with a BMIin the optimal range. Inclusion of meat,meat products, and fish in the diet, even ona less than weekly basis, seems to limit someof the protection associated with a vegan orlacto-ovo vegetarian diet. These findingsmay be explained by adverse effects of meatand fish,protective effectsof typicalconstit-uents of vegan and lacto-ovo vegetarian di-ets, other characteristics of people whochoose vegetariandiets, or a combinationofthese factors.

The notion that animal protein stim-ulates insulin secretion and possibly insu-lin resistance was proposed decades ago(19). However, a number of other dietaryconstituents are associated with protec-tion against diabetes in observationalstudies or influence insulin sensitivity infood trials (6). Vegetarian diets are rich invegetables and fruits, foods that reduceoxidative stress and chronic inflamma-tion. The vegan group consumed 650g/day of fruits and vegetables, which isabout one-third more than the amountconsumed by nonvegetarians (data not

shown). Observational evidence hasshown that these dietary constituents areassociated with a reduction in type 2 dia-betes of40% (6). Vegetarian diets con-tain substantially less saturated fat thannonvegetarian diets, and saturated fattyacids have been shown to reduce insulinsensitivity, though a recent review con-

cluded that some of the data supportingthis idea was flawed (20). The vegetariandiet typically includes foods that have alow glycemic index such as beans, le-gumes, and nuts. We did not calculate theglycemic load of the diets. Though low-glycemic-response diets are associatedwith less prevalence of type 2 diabetes,cohort studies have not consistentlyfound a relation between dietary glycemicindex or load and risk of diabetes (21,22);furthermore, whether the glycemic re-sponse causes diabetes is not established.

Protection against type 2 diabetes as-sociated with vegetarian diets is partlydue to the lower BMI of vegetarians (Table3), where the effects of diet when not ad-

justed for BMI were greater yet. Disentan-gling the effects of diet on insulinsensitivity independent of lower adiposityamong vegetarians may be difficult. Onlysparse data have investigated whethervegetarians matched to nonvegetarianswith regard to adiposity differ in insulinresistance or sensitivity. In a study thatmatched vegetarians and nonvegetarians,nonvegetarians had higher insulin, glu-

cose, and homeostasis model assessmentvalues than vegetarians (23). Whethervegetarians and nonvegetarians werematched with regard to abdominal girthwas not reported. The protective effect ofvegetarianism in thecurrent study was ev-ident in individuals with BMI below orabove 30 kg/m2, further strengtheningthe notion that independent effects of thediet are present.

Church attendees tend to have higherbody weight than nonattendees (24), andincreasing trends in BMI in the general

population have also been observedamong Adventists (data not shown). Veg-ans were the only church members whosemean BMI was25 kg/m2. Previous stud-ies have reported a difference of2 BMIunits between vegans and meat eaters (4).In the current study, the difference of 5BMI units may indicate greater protectionin current environments where a varietyof high-energy dense foods are available.Some evidence indicates a temporal rela-tionship between initiating plant-baseddiets and leanness (2,3), though a ran-domized study found that a vegetarian

Table 3Multiple logistic regression analysis of the relation between diet and type 2 diabetes

OR (95% CI)* OR (95% CI)

Age 1.04 (1.041.05) 1.03 (1.031.04)Female vs. male 0.67 (0.620.72) 0.78 (0.720.84)

Non-black vs. black 0.66 (0.610.72) 0.64 (0.590.69)BMI 1.11 (1.111.12)

Physical activity: METs3.28.6 vs. 03.2 0.85 (0.770.93) 0.76 (0.690.83)

8.621.0 vs. 03.2 0.77 (0.690.85) 0.65 (0.590.72)

21.0 vs. 03.2 0.65 (0.580.72) 0.52 (0.470.58)

EducationSome college vs. high school or less 1.00 (0.911.11) 1.04 (0.951.15)

College or higher vs. high school or less 1.00 (0.901.10) 0.95 (0.861.05)Income (USD)

11,00030,000 vs.10,000 0.87 (0.800.96) 0.82 (0.750.90)31,00050,000 vs.10,000 0.77 (0.680.86) 0.72 (0.650.81)

51,000 vs. 10,000 0.66 (0.580.76) 0.61 (0.530.70)Television watching (h/day)

12 1.31 (1.161.47) 1.54 (1.371.73)

3 1.62 (1.441.83) 2.26 (2.012.54)

Sleep (h/night)7 vs. 6 0.83 (0.760.91) 0.77 (0.710.85)

8 vs.6 0.94 (0.861.03) 0.86 (0.790.94) Alcohol use during last 12 months vs. none 0.69 (0.600.80) 0.64 (0.550.73)

Diet Vegan vs. nonvegetarian 0.51 (0.400.66) 0.32 (0.250.41)

Lacto-ovo vegetarian vs. nonvegetarian 0.54 (0.490.60) 0.43 (0.390.47)Pesco-vegetarian vs. nonvegetarian 0.70 (0.610.80) 0.56 (0.490.64)

Semi-vegetarian vs. nonvegetarian 0.76 (0.650.90) 0.69 (0.590.81)

*Adjusted for all factors. Adjusted for all factors except BMI. OR, odds ratio.




5/6

diet did not improve long-term weightloss (25). As with most dietary trials, theparticipants compliance to the diet de-clined substantially over time.

The present cohort is likely to bemore homogenous than general popula-tions regarding nondietary factors allow-ing comparisons between dietary groups

to be less affected by other differences.This may be true regarding smoking andalcohol use, which are practices stronglydiscouraged by the church. One of themajor confounders of diet and disease as-sociations in observational studies is cig-arette smoking. As the participants werealmost exclusively nonsmokers, the con-founding effects of smoking on bodyweight and risk of type 2 diabetes wereavoided. The cohort exhibited an unusu-ally wide range of dietary exposures andincluded one of the largest numbers of

vegans studied in any sample. The resultsare likely to be generalizable given thatwe found expected relationships be-tween diabetes and age, ethnicity, sex,BMI, physical activity, sleep, and televi-sion watching.

Study limitationsOur data are cross-sectional and do notallow causal inferences to be made. How-ever, reverse causation is unlikely in thatsubjects diagnosed with diabetes wouldbe less expected to differentially changetheir diet from vegetarian to omnivorous

than subjects without diabetes. We wereunableto assess physical activity for aboutone-sixth of the cohort because responsesto one or more of the questions requiredfor the calculation of MET units weremissing. Food-frequency questionnairesinvolve a certain degree of measurementerror; however, the ability to allocate sub-

jects into a broad dietary pattern is prob-ably very strong. All variables were self-reported; however, our calibration studyfound evidence for good validity for thediagnosis of diabetes. Diabetes may have

been underreported in the vegan andother vegetarians because of their lowerBMIs; however, this is unlikely to affectthe study conclusions substantially giventhe association we observed between dietand diabetes in individuals with BMI bothbelow and above 30 kg/m2.

The cohort was not representative ofthe general population; i.e., participantswere church attendees. Members whochoose vegetarianism are likely to bemore compliant with other church tenetsand to differ from nonvegetarians with re-gard to major determinants of type 2 dia-

betes. This was indeed the case withregard to some factors; e.g., nonvegetar-ian diets were more associated with blackethnicity, less education, more televisionwatching, and fewer hours of sleep thanwere vegetarian diets. On the other hand,nonvegetarians were younger and re-ported more physical activity and alcohol

consumption, which are all establishedprotective factors against type 2 diabetes.Nevertheless, the association betweendiet and type 2 diabetes remained strongafter adjustment for these factors.

In conclusion, this study showed thatall variants of vegetarian diets (vegan, lacto-ovo, and pesco- and semi-vegetarian) wereassociated with substantially lower risk oftype 2 diabetesand lower BMIthannonveg-etarian diets. The protection afforded byvegan and lacto-ovo vegetarian diets wasstrongest.

Acknowledgments This work was sup-ported by National Institutes of Health Grant1R01CA94594 and by the School of PublicHealth, Loma Linda University, Loma Linda,California.

No potential conflicts of interest relevant tothis article were reported.

References1. Fraser GE. Associations between diet and

cancer, ischemic heart disease, and all-cause mortality in non-Hispanic white

California Seventh-day Adventists. Am JClin Nutr 1999; 20(Suppl.):532S538S2. Rosell M, Appleby P, Spencer E, Key T.

Weight gain over 5 years in 21,966 meat-eating, fish-eating, vegetarian, and veganmen and women in EPIC-Oxford. Int JObes 2006;30:13891396

3. Phillips F, Hackett A, Billington D, Strat-ton G. Effectsof changing froma mixed toself-selected vegetarian diet on anthropo-metric measurements in UK adults. JHum Nutr Diet 2004;17:249255

4. Spencer EA, Appleby PN, Davey GK, KeyTJ. Diet and body mass index in 38,000EPICOxford meat-eaters,fish-eaters, veg-

etarians and vegans. Int J Obes 2003;27:728734

5. Appleby PN, Thorogood M, Mann JI, KeyTJ. Lowbody mass index in non-meat eat-ers: the possible roles of animal fat, di-etary fibre and alcohol. Int J Obes 1998;22:454460

6. Jenkins DJA, Kendall CWC, Marchie A, Jenkins AL, Augustin LSA, Ludwig DS,Barnard ND, Anderson JW. Type 2 diabe-tes and the vegetarian diet. AmJ ClinNutr2003; 78(Suppl.):610S616S

7. Fung TT, Schulze M, Manson JE, WillettWC, Hu FB. Dietary patterns, meat intake,and the risk of type 2 diabetes in women.

Arch Intern Med 2004;164:223522408. Snowdon DA, Phillips RL.Does a vegetar-

ian diet reduce the occurrence of diabe-tes? Am J Public Health1985;75:507512

9. Vang A, Singh PN, Lee JW, Haddad EH,Brinegar CH. Meats, processed meats,obesity, weight gain and occurrence of di-abetes among adults: findings from Ad-ventist Health Studies. Ann Nutr Metab2008;52:96104

10. Butler TL, Fraser GE, Beeson WL, KnutsenSF,Herring RP, Chan J, SabateJ, Montgom-ery S, Haddad E, Preston-Martin S, BennettH, Jaceldo-Siegl K. Cohort profile: the Ad-ventist Health Study-2 (AHS-2). Int J Epi-demiol 2007;37:260265

11. Montgomery S, Herring P, Yancey A, Bee-son L, Butler T, Knutsen S, Sabate J, ChanJ, Preston-Martin S, Fraser G. Comparingself-reported disease outcomes, diet andlifestyles in a national cohort of blackand white Seventh-day Adventists. PrevChronic Dis 2007;4:A62

12. Knutsen S, Fraser G, Beeson W, Lind-sted K, Shavlik D. Comparison of adi-pose tissue fatty acids with dietary fattyacids as measured by 24-hour recall andfood frequency questionnaire in blackand white Adventists: the AdventistHealth Study. Ann Epidemiol 2003;13:119127

13. Knutsen S, Fraser G, Lindsted K, BeesonW, Shavlik D. Validation of assessment ofnutrient intake: comparing biologicalmeasurements of vitamin C, folate, alpha-tocopherol and carotene with 24-hour di-etary recall information in non-Hispanicblacks and whites. Ann Epidemiol 2001;

11:40641614. Singh PN, Tonstad S, Abbey DE, Fraser

GE. Validity of selected physical activityquestions in white Seventh-day Ad-ventists and non-Adventists. Med SciSports Exer 1996;28:10261037

15. Singh PN, Fraser GE, Knutsen SF, Lind-sted DS Bennett H. Validity of a physicalactivity questionnaire among African-American Seventh-day Adventists. MedSci Sports Exer 2000;33:468 475

16. Jaceldo-Siegl K, Fraser GE, Chan J,Franke A, Sabate J. Validation of soy pro-tein estimates from a food-frequency

questionnaire with repeated 24-h recallsand isoflavonoid excretion in overnighturine in a Westernpopulationwith a widerange of soy intakes. Am J Clin Nutr 2008;87:14221427

17. Fraser GE, Yan R. Guided multiple im-putation of missing data: using a sub-sample to strengthen the missing-at-random assumption. Epidemiology 2007;18:246252

18. R Development Core Team. R: a lan-guage and environment for statisticalcomputing [article online], 2007. Vi-enna, Austria, R Foundation for Statis-tical Computing. Available from http://

Tonstad and Associates



6/6

biostat.mc.vanderbilt.edu/s/Hmisc/html/areglmpute.html. Accessed 10 January2008

19. ChaussainJL, Georges P, Gendrel D, Don-nadieu M, Job JC. Serum branched-chainamino acids in the diagnosis of hyperin-sulinism in infancy. J Pediatr 1980;97:923926

20. Galgani JE, Uauy RD, Aguirre CA, Diaz

EO. Effect of the dietary fat quality on in-sulin sensitivity. Br J Nutr 2008;100:471479

21. Salmeron J, Manson JE, Stampfer MJ,Colditz GA, Wing AL, Willett WC. Di-

etary fiber, glycemic load, andrisk of non-insulin dependent diabetes mellitus inwomen. JAMA 1997;277:472477

22. Sahyoun NR, Anderson AL, Tylavsky FA,Lee JS, Sellmeyer DE, Harris TB, theHealth, Aging, and Body CompositionStudy. Dietary glycemic index and glyce-mic load and the risk of type 2 diabetes inolder adults. Am J Clin Nutr 2008;

87:12613123. Valachovicova M, Kracovicova-Kud-

lackova M, Blazcek P, Babinska K. Noevidence of insulin resistance in normalweight vegetarians: a case control study.

Eur J Nutr 2006;45:525424. Gillum RF.Frequency of attendance at re-

ligious services, overweight, and obesityin American women and men: the ThirdNational Health and Nutrition Examina-tion Survey. Ann Epidemiol 2006;16:655660

25. BurkeLE, HudsonAG, Warziski MT,StynMA, Music E, Elci OU, Sereika SM. Effects

of a vegetarian diet and treatment prefer-ence on biochemical and dietary variablesin overweight and obese adults: a ran-domized clinical trial. Am J Clin Nutr2007;86:588596



Type of Vegetarian Diet, Body Weight, And Prevalence of Type 2 Diabetes

Documents

Transcript of Type of Vegetarian Diet, Body Weight, And Prevalence of Type 2 Diabetes