Content uploaded by Wei Bao
Author content
All content in this area was uploaded by Wei Bao
Content may be subject to copyright.
Available via license: CC BY-NC-ND 3.0
Content may be subject to copyright.
Prepregnancy Dietary Protein Intake,
Major Dietary Protein Sources, and the
Risk of Gestational Diabetes Mellitus
A prospective cohort study
WEI BAO,MD,PHD
1
KATHERINE BOWERS,PHD
1
DEIRDRE K. TOBIAS,SCD
2
FRANK B. HU,MD,PHD
2,3
CUILIN ZHANG,MD,PHD
1
OBJECTIVEdDietary protein is an important modulator of glucose metabolism. However,
studies regarding the association between dietary protein intake and gestational diabetes mellitus
(GDM) risk are sparse. This study was to examine the association.
RESEARCH DESIGN AND METHODSdOur study included 21,457 singleton preg-
nancies reported among 15,294 participants of the Nurses’Health Study II cohort between 1991
and 2001. Included pregnancies were free of chronic diseases before pregnancy or previous
GDM. Generalized estimating equations were used to estimate the relative risks (RRs) and
95% CIs.
RESULTSdAfter adjustment for age, parity, nondietary and dietary factors, and BMI, multi-
variable RRs (95% CIs) comparing the highest with lowest quintiles were 1.49 (1.03–2.17) for
animal protein intake and 0.69 (0.50–0.97) for vegetable protein intake. The substitution of 5%
energy from vegetable protein for animal protein was associated with a 51% lower risk of GDM
(RR (95% CI), 0.49 (0.29–0.84)). For major dietary protein sources, multiva riable RRs (95% CIs)
comparing the highest with the lowest quintiles were 2.05 (1.55–2.73) for total red meat and
0.73 (0.56–0.95) for nuts, respectively. The substitution of red meat with poultry, fish, nuts, or
legumes showed a significantly lower risk of GDM.
CONCLUSIONSdHigher intake of animal protein, in particular red meat, was significantly
associated with a greater risk of GDM. By contrast, higher intake of vegetable protein, specifically
nuts, was associated with a significantly lower risk. Substitution of vegetable protein for animal
protein, as well as substitution of some healthy protein sources for red meat, was associated
with a lower risk of GDM.
Gestational diabetes mellitus (GDM),
defined as glucose intolerance with
onset or first recognition during
pregnancy, is a growing health concern
(1). Approximately 7% (ranging from 1 to
14%) of all pregnancies in the U.S. are
complicated by GDM, resulting in more
than 200,000 cases annually (2). GDM is
associated with an increased risk of
adverse pregnancy and perinatal out-
comes (3) and long-term adverse health
consequences for both mothers and their
children, including a predisposition to
obesity, metabolic syndrome, and type 2
diabetes mellitus (T2DM) (1,2,4); thus,
the identification of modifiable risk fac-
tors that may contribute to the prevention
of GDM is important.
Recently, several dietary and lifestyle
factors have been associated with GDM
risk, although precise underlying mecha-
nisms have yet to be established (5). Mac-
ronutrients including carbohydrates (6)
and fats (7) have previously been evalu-
ated for their association with GDM risk.
The association with protein, however,
remains unclear. Dietary proteins and
amino acids are important modulators
of glucose metabolism, and a diet high
in protein may impact glucose homeosta-
sis by promoting insulin resistance and
increasing gluconeogenesis (8). More-
over, emerging data suggest that protein
actions may vary by the amino acid types
and food sources. For instance, a prospec-
tive cohort study in Europeans showed
that long-term high intake of animal pro-
tein but not vegetable protein was associ-
ated with an increased risk of T2DM (9).
Additionally, a study of metabolomics re-
cently demonstrated that plasma concen-
trations of several kinds of amino acids,
including branched-chain amino acids
(BCAAs) and aromatic amino acids,
were strongly and significantly associated
with incident T2DM risk (10).
Several major food sources of animal
protein, such as red meat, were positively
associated with the risk of both T2DM
(11) and GDM (12). Conversely, higher
intakesofnuts(13)andlegumes(14)
were associated with a lower risk of
T2DM, but their associations with GDM
have not yet been evaluated. In addition,
the associations between other major
sources of animal protein (e.g., poultry,
fish, and dairy products) and GDM risk
have not been reported.
In this prospective cohort study, we
aimed to examine the associations of
prepregnancy dietary protein intake (to-
tal, animal, and vegetable protein) as well
as major dietary protein sources with the
risk of GDM. We also estimated the effect
of substituting prepregnancy protein for
carbohydrates, substituting vegetable
protein for animal protein, and substitut-
ing other major dietary protein sources
for red meat on the risk of GDM.
ccccccccccccccccccccccccccccccccccccccccccccccccc
From the
1
Epidemiology Branch, Division of Epidemiology, Statistics and Prevention Research, Eunice
Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health,
Rockville, Maryland; the
2
Departments of Nutrition and Epidemiology, Harvard School of Public Health,
Boston, Massachusetts; and the
3
Channing Division of Network Medicine, Department of Medicine,
Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts.
Corresponding author: Cuilin Zhang, zhangcu@mail.nih.gov.
Received 3 October 2012 and accepted 19 December 2012.
DOI: 10.2337/dc12-2018
This article contains Supplementary Data online at http://care.diabetesjournals.org/lookup/suppl/doi:10
.2337/dc12-2018/-/DC1.
© 2013 by the American Diabetes Association. Readers may use this article as long as the work is properly
cited, the use is educa tional and not for profit, and the work is not alte red. See http://creativecommons.org/
licenses/by-nc-nd/3.0/ for details.
care.diabetesjournals.org DIABETES CAR E 1
Epidemiology/Health Services Research
ORIGINAL ARTICLE
Diabetes Care Publish Ahead of Print, published online February 12, 2013
RESEARCH DESIGN AND
METHODSdThe Nurses’Health
Study II (NHS II) is an ongoing prospec-
tive cohort study of 116,678 female
nurses aged 25–44 years at study incep-
tion in 1989 (15). The participants are
sent a biennial questionnaire regarding
disease outcomes and lifestyle character-
istics, such as smoking status, medication
use, and physical activity. Follow-up for
each questionnaire cycle was .90%. This
study has been approved by the institu-
tional review board of the Partners Health
Care System (Boston, MA), with partici-
pants’consent implied by the return of
the questionnaires.
NHS II participants were included in
this analysis if they reported at least one
singleton pregnancy lasting .6months
between 1991 and 2001. GDM was last
captured on the 2001 questionnaire, as
the majority of NHS II participants passed
reproductive age by then. Pregnancies
were excluded if the participant reported
GDM in a previous pregnancy, a diagnosis
of T2DM, cancer, or a cardiovascular
event prior to an otherwise eligible preg-
nancy. Pregnancies reported after GDM
were not included because women with
GDM in a previous pregnancy may
change their diet and lifestyle during the
next pregnancy to prevent recurrent
GDM. Pregnancies were also excluded if
the participant did not return a prepreg-
nancy food frequency questionnaire
(FFQ), left .70 FFQ items blank, or re-
ported unrealistic total energy intake
(,600 or .3,500 kcal/day).
Exposure assessment
Beginning in 1991 and every 4 years
thereafter, participants were asked to re-
port their food intakes using a semiquan-
titative FFQ. Answers were provided in
nine possible categories ranging from
“never”to “6 or more times/day,”with a
standard portion size specified for each
food. Major protein sources included
the following (16,17): unprocessed red
meat (beef or lamb as main dish, pork as
main dish, hamburger, and beef, pork, or
lamb as a sandwich or mixed dish), pro-
cessed red meat (bacon, beef hot dogs,
and sausage, salami, bologna, and other
processed meats), poultry (chicken with
and without skin, chicken sandwich, and
chicken/turkey hot dog), fish (canned
tuna, dark- and light-fleshed fish, and
breaded fish), dairy products (whole
milk, ice cream, hard cheese, full-fat
cheese, cream, sour cream, cream cheese,
butter, skim/low-fat milk, 1% and 2%
milk, yogurt, cottage and ricotta cheeses,
low-fat cheese, and sherbet), eggs, nuts
(peanuts, peanut butter, walnuts, and
other nuts), and legumes (tofu or soy-
beans, string beans, peas or lima beans,
and beans or lentils). Total red meat in-
take was calculated as the sum of unpro-
cessed and processed red meat intakes.
Intakes of individual nutrients in-
cluding protein were computed by mul-
tiplying the frequency of consumption of
each unit of food by the nutrient content
of the specified portions based on food
composition data from U.S. Department
of Agriculture sources (18). The repro-
ducibility and validity of the FFQ have
been extensively documented elsewhere
(19–21). Pearson correlation coefficient
between energy-adjusted protein intakes
assessed by the food-frequency question-
naire compared with four 1-week diet re-
cords was 0.52 in a similar cohort of U.S.
women (20).
Covariate assessment
Participants reported their current weight
on each biennial questionnaire. Self-reported
weight was highly correlated with mea-
sured weight (r= 0.97) in a previous val-
idation study (22). BMI was computed as
weight in kilograms divided by the
square of height in meters. Total physical
activity was ascertained by frequency of
engaging in common recreational activi-
ties, from which MET hours per week
were derived. The questionnaire-based
estimates correlated well with detailed ac-
tivity diaries in a prior validation study
(r= 0.56) (23).
Outcome ascertainment
Incident GDM was ascertained by self-
report on each biennial questionnaire
through 2001. In the case of more than
one pregnancy lasting .6monthsreported
within a 2-year questionnaire period, GDM
status was attributed to the first pregnancy.
In a prior validation study among a sub-
group of the NHS II cohort, 94% of GDM
self-reports were confirmed by medical
records (15). In a random sample of par-
ous women without GDM, 83% reported
a glucose screening test during preg-
nancy and 100% reported frequent pre-
natal urine screening, suggesting a high
level of GDM surveillance in this cohort
(15).
Statistical analysis
Exposure was computed as the percent-
age of total energy intake from protein
using the nutrient-density method (24).
Prepregnancy dietary exposure measures
were used to calculate the updated cumu-
lative average intake for each individual at
each time period to reduce within-subject
variation and represent long-term habit-
ual prepregnancy diet (25).
Participants were divided into quin-
tiles according to the cumulative average
intakes of dietary protein (% of energy) or
major protein sources (servings/day) in
their diet. Relative risks (RRs) and 95%
CIs were estimated through multivariate
logistic regression with generalized esti-
mating equations (GEE), specifying an
exchangeable correlation structure. Gen-
eralized estimating equations allowed us
to account for correlations among re-
peated observations (pregnancies) con-
tributed by a single participant. To
compute the test for a significant trend
across quintiles, we modeled median
values of each quintile as a continuous
variable.
Covariates in the multivariable mod-
els included age; parity; race/ethnicity;
family history of diabetes; cigarette smok-
ing; alcohol intake; physical activity; total
energy intake; intakes of saturated fat,
monounsaturated fat, polyunsaturated
fat, trans fat, dietary cholesterol, glycemic
load, and dietary fiber; and updated BMI
when total protein intake was modeled as
the exposure of interest. Animal protein
and vegetable protein were mutually ad-
justed for one another. For the intake of
major dietary protein sources, we ad-
justed for age; parity; race/ethnicity; fam-
ily history of diabetes; cigarette smoking;
alcohol intake; physical activity; total en-
ergy intake; dietary intakes of fruits,
sugar-sweetened beverages, whole grains,
and other major dietary protein sources
(for mutual adjustment); and updated
BMI.
To simulate the substitution of di-
etary protein for carbohydrates, we fit
isocaloric models (24) by simultaneously
including total energy intake and the per-
centages of energy (continuous) derived
from total fat and protein, as well as the
potential confounders listed above. The
b-coefficient for total protein from these
models estimated the effect of substitut-
ing 1% of energy from carbohydrates with
1% of energy from protein (26). For the
estimation of substituting animal protein
with vegetable protein, we simulta-
neously included total energy intake and
the percentages of energy derived from
vegetable protein, as well as the potential
confounders listed above in the model.
Similarly, we estimated the effect of
2DIABETES CARE care.diabetesjournals.org
Protein intake and GDM risk
substituting one major protein source for
another by simultaneously modeling all
major protein sources (servings/day)
with total energy and other potential con-
founders listed above. RRs and 95% CIs
were estimated by computing the differ-
ence in the coefficients for two protein
sources and their own variances and co-
variance (16,17,27).
Advanced maternal age is an estab-
lished risk factor for GDM (28). We eval-
uated effect modification by age (,35 vs.
$35 years), parity (nulliparous vs. par-
ous), family history of diabetes (yes vs.
no), and physical activity (highest two
quintiles vs. lowest three quintiles) by
stratified analyses. Since BMI is a possible
intermediate between dietary protein and
GDM, we estimated the proportion of
the association between dietary protein
intake and GDM risk that is explained
by prepregnancy BMI (modeled con-
tinuously) (29) using an SAS macro de-
veloped by Dr. D. Spiegelman and
colleagues at the Harvard School of Public
Health (http://www.hsph.harvard.edu/
faculty/donna-spiegelman/software/
mediate/).
All statistical analyses were per-
formed with SAS software (version 9.1;
SAS Institute). P,0.05 was considered
statistically significant.
RESULTS
Baseline characteristics
Among the 21,457 eligible singleton
pregnancies from the 15,294 women,
during the 10 years of follow-up we
documented 870 incident GDM pregnan-
cies. Compared with the participants with
lower total protein intake, those with
higher protein intake were more likely
to be nonsmokers and consumed more
cholesterol, dietary fiber, magnesium,
hemeiron,potassium,calcium,meat,
vegetables, and dairy products but less
alcohol, carbohydrate, trans fat, and
sugar-sweetened beverages during the
prepregnancy time period (Table 1).
Women who consumed more animal
protein were likely to consume more total
fat, saturated fat, cholesterol, heme iron,
calcium, red meat, poultry, and dairy
products. Women who consumed more
vegetable proteins were likely to consume
less of these food and nutrients.
Prepregnancy dietary protein intake
and the risk of GDM
The median intakes of prepregnancy total
calories from protein in this population
were 15.2 and 23.3% of energy in the
lowest and the highest quintile, respec-
tively. Animal protein accounted for the
majority of total protein intake. After
adjustment for age, parity, nondietary
and dietary factors, and BMI, animal pro-
tein intake was significantly and posi-
tively associated with GDM risk while
vegetable protein intake was significantly
and inversely associated with the risk;
multivariable RRs (95% CIs), comparing
the highest with lowest quintiles were
1.28 (0.90–1.83) for total protein intake,
1.49 (1.03–2.17) for animal protein in-
take, and 0.69 (0.50–0.97) for vegetable
protein intake (Table 2).
Substituting 5% of energy from car-
bohydrates with animal protein was as-
sociated with a significant 29% greater
risk of GDM (multivariable RR [95% CI],
1.08–1.54; P= 0.006). Substituting 5% of
energy from vegetable protein for animal
protein was associated with a 51% lower
risk (0.49 [0.29–0.84]; P= 0.009) (Table 3).
The associations between prepreg-
nancy dietary protein intake and GDM
risk were not significantly modified by
age, parity, family history of diabetes, or
physical activity. Mediation analyses esti-
mated that prepregnancy BMI explained
35.7% (95% CI 10.6–60.8; P= 0.005)
and 31.1% (10.7–51.6; P= 0.003) of the
total effects of total protein and animal
protein on GDM risk, respectively. The
effect of vegetable protein intake on
GDM risk was not significantly mediated
by BMI (12.2% [95% CI 221.1 to 45.4];
P= 0.47).
Major prepregnancy dietary protein
sources and the risk of GDM
Prepregnancy red meat consumption was
significantly and positively associated
with the risk of GDM. Multivariable RRs
(95% CIs) for GDM among participants
with the highest compared with the lowest
quintiles of intakes were 2.46 (1.86–3.25),
1.89 (1.43–2.48), and 1.48 (1.13–1.95)
for total red meat, unprocessed red meat,
and processed red meat, respectively.
These associations were attenuated but
remained significant after additional ad-
justment for BMI, with RRs of 2.05
(1.55–2.73), 1.60 (1.21–2.12), and 1.36
(1.03–1.80), respectively. By contrast,
greater prepregnancy nut consumption
was significantly associated with a lower
risk of GDM; the fully adjusted RR com-
paring the highest with lowest quintiles of
intake was 0.73 (0.56–0.95) (Table 4).
Substituting one serving per day of
total red meat with some healthy protein
sources was significantly associated with a
lower risk of GDM: 29% lower risk for
poultry (RR (95% CI), 0.71 [0.54–0.94]),
33% for fish (0.67 [0.46–0.98]), 51% for
nuts (0.49 [0.36–0.66]), and 33% for le-
gumes (0.67 [0.51–0.88]). Similar substi-
tution estimates were observed for the
replacement of unprocessed red meat
and processed red meat (Supplementary
Fig. 1).
CONCLUSIONSdIn this prospective
cohort study, we observed that a prepreg-
nancy intake of animal protein, in partic-
ular red meat, was significantly and
positively associated with GDM risk,
while vegetable protein intake, specifi-
cally nuts, was significantly and inversely
associated with GDM risk. Substituting
5% of energy from vegetable protein for
animal protein and substitution of poul-
try, fish, nuts, or legumes for red meat
were associated with a lower GDM risk.
Although protein may have beneficial
effects on energy homeostasis by pro-
moting thermogenesis, inducing satiety,
and possibly increasing energy expendi-
ture, it may also have detrimental effects
on glucose homeostasis (8). Consump-
tion of a high-protein diet for 6 months
in healthy individuals induced a higher
glucose-stimulated insulin secretion due
to reduced glucose threshold of the endo-
crine b-cells, increased endogenous glu-
cose output and plasma glucagon level,
and enhanced gluconeogenesis (30). Re-
cent studies examining different types
and sources of dietary protein suggest
that animal protein and vegetable protein
may have divergent effects on diabetes. A
prospective cohort study with 10 years of
follow-up showed that the risk of T2DM
increased with higher intakes of total pro-
tein and animal protein, but vegetable
protein intake was not related to T2DM
risk (9). It has been proposed that ele-
vated incidence of diabetes in relation to
high protein intake, in particular animal
protein, might result from an accelerated
“fatigue”or “failure”of pancreas islets (8).
Meat consumption has continued to
rise over the past century in the U.S., with
the largest proportion from red meat
(58%) (31). Red meat consumption has
been found to be positively associated
with long-term weight gain (32) and risk
of T2DM (11), coronary heart disease
(16), stroke (17), and all-cause mortality
(27). In this study, red meat consumption
was significantlyassociatedwithanin-
creased risk of GDM, which is consistent
with our previous findings with shorter
care.diabetesjournals.org DIABETES CAR E 3
Bao and Associates
Table 1dBaseline characteristics according to quintiles of prepregnancy dietary protein intake among 15,294 participants in the Nurses’Health Study II
Total protein intake Animal protein intake Vegetable protein intake
Q1 Q3 Q5 Q1 Q3 Q5 Q1 Q3 Q5
Participants (n) 3,006 3,173 2,793 3,015 3,140 2,862 3,288 3,101 2,871
Age in 1991 (years) 31.8 (3.3) 32.0 (3.2) 32.2 (3.3) 32.0 (3.3) 32.0 (3.3) 32.0 (3.3) 31.4 (3.1) 32.0 (3.2) 32.7 (3.4)
White (%) 92 94 93 92 94 93 90 94 93
Family history of diabetes (%) 11 11 13 10 12 12 11 11 11
Nulliparous (%) 41 33 37 43 34 36 34 34 44
Current smoking (%) 12 8 8 10 8 9 13 9 7
Alcohol consumption (g/day) 3.9 (7.1) 3.0 (4.9) 2.2 (3.8) 3.6 (6.1) 2.9 (4.5) 2.4 (4.2) 3.2 (6.6) 3.0 (4.8) 2.8 (4.3)
BMI (kg/m
2
) 22.7 (4.0) 23.3 (4.1) 24.2 (4.6) 22.6 (3.9) 23.3 (4.1) 24.3 (4.7) 24.0 (4.8) 23.4 (4.2) 22.7 (3.7)
Physical activity (h/week) 23.7 (30.7) 23.1 (29.2) 24.3 (29.7) 25.7 (32.8) 21.9 (27.9) 23.3 (28.5) 19.1 (25.2) 22.6 (27.5) 30.1 (35.8)
Dietary factors (per day)
Total calories (kcal) 1,928 (601) 1,860 (534) 1,658 (492) 1,899 (599) 1,878 (532) 1,673 (506) 1,864 (570) 1,844 (535) 1,760 (544)
Carbohydrate (%E) 56.3 (7.5) 50.2 (5.8) 45.5 (6.2) 56.9 (7.2) 50.5 (5.5) 44.7 (6.0) 48.1 (8.5) 50.0 (5.9) 55.0 (6.7)
Total fat (%E) 29.9 (6.1) 31.4 (5.2) 30.9 (5.3) 29.3 (6.1) 31.3 (5.1) 31.7 (5.3) 32.8 (5.8) 31.3 (4.8) 27.8 (5.3)
Saturated fat (%E) 10.7 (2.7) 11.4 (2.3) 11.2 (2.3) 10.3 (2.6) 11.3 (2.2) 11.7 (2.3) 12.4 (2.5) 11.3 (2.0) 9.4 (2.1)
Monounsaturated fat (%E) 11.4 (2.6) 11.9 (2.2) 11.4 (2.4) 11.1 (2.6) 11.8 (2.2) 11.7 (2.4) 12.4 (2.5) 11.8 (2.1) 10.4 (2.3)
Polyunsaturated fat (%E) 5.4 (1.6) 5.5 (1.2) 5.4 (1.2) 5.5 (1.6) 5.4 (1.2) 5.4 (1.2) 5.2 (1.4) 5.5 (1.2) 5.6 (1.4)
Trans fat (%E) 1.7 (0.7) 1.6 (0.6) 1.4 (0.5) 1.6 (0.7) 1.6 (0.6) 1.5 (0.5) 1.7 (0.6) 1.6 (0.6) 1.3 (0.5)
Cholesterol (mg) * 186 (54) 240 (51) 291 (62) 178 (50) 238 (47) 296 (61) 262 (68) 241 (56) 199 (63)
Glycemic index* 55.4 (3.2) 53.9 (2.9) 52.5 (3.3) 55.2 (3.1) 54.1 (3.0) 52.4 (3.4) 53.8 (3.9) 54.0 (3.0) 54.2 (3.1)
Glycemic load* 1.4 (0.2) 1.2 (0.2) 1.1 (0.2) 1.4 (0.2) 1.2 (0.2) 1.1 (0.2) 1.2 (0.3) 1.2 (0.2) 1.3 (0.2)
Total fiber (g)* 17.6 (6.2) 18.1 (4.9) 18.4 (5.5) 19.6 (7.1) 17.8 (4.7) 17.0 (5.1) 13.2 (3.0) 18.0 (3.4) 24.0 (6.3)
Magnesium (mg)* 289 (75) 320 (67) 346 (74) 308 (84) 315 (68) 335 (72) 280 (66) 316 (63) 369 (77)
Heme iron (mg)* 0.7 (0.3) 1.1 (0.3) 1.5 (0.4) 0.7 (0.3) 1.1 (0.3) 1.5 (0.4) 1.2 (0.4) 1.1 (0.3) 0.9 (0.4)
Potassium (mg)* 2,627 (537) 2,903 (457) 3,124 (482) 2,728 (581) 2,877 (469) 3,067 (480) 2,672 (489) 2,888 (449) 3,110 (558)
Calcium (mg)* 890 (357) 1,090 (392) 1,199 (473) 907 (367) 1,078 (390) 1,214 (484) 1,110 (482) 1,065 (400) 1,042 (399)
Vitamin C (mg)* 263 (296) 239 (253) 252 (291) 281 (317) 229 (230) 243 (286) 213 (238) 237 (243) 304 (334)
Vitamin E (mg)* 37.0 (110.1) 32.4 (93.4) 34.9 (107.9) 39.4 (114.7) 29.9 (83.9) 33.7 (104.8) 27.2 (72.4) 33.1 (91.5) 45.6 (127.4)
Fruits (servings) 1.2 (1.1) 1.2 (0.9) 1.1 (0.8) 1.3 (1.2) 1.2 (0.9) 1.1 (0.8) 0.9 (0.8) 1.3 (0.9) 1.5 (1.1)
Vegetables (servings) 2.9 (2.0) 3.2 (1.9) 3.5 (2.2) 3.3 (2.3) 3.2 (1.9) 3.2 (1.9) 2.3 (1.3) 3.1 (1.6) 4.3 (2.7)
Unprocessed red meat (servings) 0.4 (0.3) 0.6 (0.4) 0.6 (0.5) 0.4 (0.3) 0.6 (0.4) 0.6 (0.5) 0.7 (0.5) 0.5 (0.3) 0.3 (0.3)
Processed red meat (servings) 1.2 (1.2) 1.2 (1.2) 0.9 (1.1) 1.1 (1.2) 1.2 (1.2) 1.0 (1.1) 1.4 (1.3) 1.2 (1.1) 0.7 (1.0)
Poultry (servings) 0.3 (0.2) 0.4 (0.2) 0.7 (0.4) 0.3 (0.2) 0.5 (0.2) 0.7 (0.4) 0.5 (0.3) 0.5 (0.3) 0.4 (0.3)
Fish (servings) 0.1 (0.1) 0.2 (0.2) 0.3 (0.3) 0.2 (0.2) 0.2 (0.2) 0.3 (0.3) 0.2 (0.2) 0.2 (0.2) 0.3 (0.2)
Dairy products (servings) 2.2 (1.5) 2.7 (1.5) 2.6 (1.6) 2.1 (1.4) 2.7 (1.5) 2.8 (1.7) 3.0 (1.8) 2.6 (1.4) 2.0 (1.3)
Eggs (servings) 0.2 (0.2) 0.2 (0.2) 0.2 (0.2) 0.2 (0.2) 0.2 (0.2) 0.2 (0.2) 0.2 (0.2) 0.2 (0.2) 0.1 (0.2)
Nuts (servings) 0.3 (0.4) 0.3 (0.3) 0.2 (0.2) 0.3 (0.5) 0.3 (0.3) 0.2 (0.2) 0.1 (0.2) 0.2 (0.3) 0.4 (0.5)
Legumes (servings) 0.3 (0.3) 0.4 (0.3) 0.4 (0.4) 0.4 (0.4) 0.4 (0.3) 0.3 (0.3) 0.2 (0.2) 0.3 (0.3) 0.5 (0.5)
SSBs (servings) 1.2 (1.4) 0.4 (0.6) 0.2 (0.3) 0.9 (1.3) 0.5 (0.7) 0.2 (0.4) 1.0 (1.3) 0.4 (0.6) 0.2 (0.4)
Whole grains (servings) 1.1 (1.1) 1.1 (1.0) 1.0 (0.9) 1.3 (1.2) 1.1 (1.0) 0.9 (0.8) 0.6 (0.6) 1.1 (0.9) 1.8 (1.3)
Values are means (SD) unless otherwise specified. All comparisons across quintiles of dietary protein intake are significant except the following: total protein, physical activity, and vitamin E intake; animal protein, age,
and vegetable intake; and vegetable protein, family history of diabetes, and alcohol consumption. , energy; SSB, sugar-sweetened beverage. *Value is energy adjusted.
4DIABETES CARE care.diabetesjournals.org
Protein intake and GDM risk
follow-up (12). Although no significant
association was observed for other animal
foods, the substitution of fish and poultry
for red meat was associated with a lower
risk of GDM. By contrast, a significant and
inverse association was found between
nut consumption and risk of GDM.
An inverse association between nut
consumption and risk of T2DM was
previously observed in the NHS partici-
pants (13). Nuts have a healthful nutri-
tional profile; in addition to being a good
source of vegetable protein, nuts are rich
in monounsaturated fatty acids, polyun-
saturated fatty acids, fiber, and magne-
sium and have a relatively low glycemic
index (13,33). These factors, either indi-
vidually or in combination, have been
associated with improved insulin sensitiv-
ity and lower diabetes risk (33).
The distinct effects of animal protein
and vegetable protein on the incidence of
GDM could be attributable to other nu-
trients coexistent in foods rich in protein,
e.g., the co-occurrence of cholesterol and
saturated fat in foods rich in animal pro-
tein. However, in the current study, the
association of animal protein and GDM
risk remained significant even after the
adjustment of dietary cholesterol and
saturated fat intakes. The distinct effects
could be also due to variations of amino
acid composition in these foods. Several
in vitro and in vivo studies support an
important role of amino acids in glucose
homeostasis through regulation of glu-
cose uptake and glycogen synthesis in
skeletal muscle, hepatic glucose produc-
tion, and insulin secretion (8). It has long
been recognized that the plasma concen-
trations of BCAAs are dramatically raised
to ~200% of fasting values after the inges-
tion of an animal protein-rich meal (34).
A recent study showed that dietary pro-
teins isolated from beef and pork meat
resulted in significantly higher plasma
concentrations of BCAAs than soy protein
Table 2dRRs (95% CIs) for GDM according to quintiles of prepregnancy dietary protein intake
Q1 Q2 Q3 Q4 Q5 P
Total protein
Median protein intake
(% energy per day) 15.23 17.53 19.14 20.78 23.30 d
Model 1 1.00 1.03 (0.83–1.29) 1.02 (0.82–1.28) 1.31 (1.06–1.62) 1.22 (0.98–1.53) 0.011
Model 2 1.00 1.10 (0.88–1.37) 1.07 (0.85–1.34) 1.38 (1.11–1.71) 1.26 (1.01–1.58) 0.007
Model 3 1.00 1.17 (0.92–1.48) 1.19 (0.91–1.55) 1.58 (1.19–2.10) 1.46 (1.03–2.07) 0.012
Model 4 1.00 1.13 (0.89–1.44) 1.13 (0.86–1.48) 1.43 (1.08–1.91) 1.28 (0.90–1.83) 0.086
Animal protein
Median intake
(% energy per day) 10.00 12.41 14.10 15.86 18.58 d
Model 1 1.00 1.02 (0.81–1.28) 1.01 (0.80–1.27) 1.37 (1.10–1.69) 1.45 (1.17–1.81) ,0.001
Model 2 1.00 1.04 (0.83–1.30) 1.00 (0.79–1.26) 1.38 (1.11–1.72) 1.44 (1.15–1.80) ,0.001
Model 3 1.00 1.07 (0.84–1.37) 1.09 (0.83–1.45) 1.57 (1.16–2.13) 1.65 (1.14–2.38) 0.002
Model 4 1.00 1.05 (0.82–1.34) 1.06 (0.80–1.40) 1.46 (1.08–1.99) 1.49 (1.03–2.17) 0.013
Vegetable protein
Median intake
(% energy per day) 3.78 4.46 4.96 5.47 6.36 d
Model 1 1.00 0.89 (0.73–1.08) 0.76 (0.62–0.94) 0.71 (0.58–0.88) 0.53 (0.42–0.67) ,0.001
Model 2 1.00 0.94 (0.77–1.15) 0.82 (0.67–1.01) 0.79 (0.63–0.97) 0.59 (0.47–0.75) ,0.001
Model 3 1.00 0.93 (0.75–1.16) 0.83 (0.65–1.06) 0.82 (0.63–1.09) 0.70 (0.50–0.98) 0.038
Model 4 1.00 0.92 (0.74–1.15) 0.84 (0.65–1.08) 0.83 (0.62–1.09) 0.69 (0.50–0.97) 0.034
Model 1: age (5-year category) and parity (0, 1, 2, and $3). Model 2: model 1 adjustments plus race/ethnicity (Caucasian, African American, Hispanic, Asian, and
others), family history of diabetes (yes and no), cigarette smoking (never, past, and current), alcohol intake (0, 0.1–5.0, 5.1–10.0, or .10 g/day), physical activity
(quintiles), and total energy intake (quintiles). Model 3: model 2 adjustments plus saturated fat (quintiles), monounsaturated fat (quintiles), polyunsaturated fat
(quintiles), trans fat (quintiles), dietary cholesterol (quintiles), glycemic load (quintiles), dietary fiber (quintiles), and mutual adjustment for animal protein and
vegetable protein. Model 4: model 3 adjustments plus BMI (,21, 21–22.9, 23.0–24.9, 25.0–26.9, 27.0–28.9, 29.0–30.9, 31.0–32.9, 33.0–34.9, and $35.0 kg/m
2
).
Q, quintile.
Table 3dMultivariate RRs (95% CIs) of GDM associated with increases in 5% of energy from
types of protein
RRs (95% CIs)
in the multivariate
model P
RRs (95% CIs)
in the multivariate
model plus BMI P
Substitution for carbohydrate intake
Total protein 1.35 (1.14–1.61) 0.001 1.28 (1.07–1.53) 0.007
Animal protein 1.36 (1.14–1.62) 0.001 1.29 (1.08–1.54) 0.006
Vegetable protein 0.61 (0.35–1.08) 0.089 0.58 (0.33–1.03) 0.064
Substitution for animal protein intake
Vegetable protein 0.49 (0.29–0.83) 0.008 0.49 (0.29–0.84) 0.009
Covariates adjusted in the multivariate model were age (5-year category), parity (0, 1, 2, and $3), race/
ethnicity (Caucasian, African American, Hispanic, Asian, and others), family history of diabetes (yes and no),
cigarette smoking (never, past, and current), alcohol intake (0, 0.1–5.0, 5.1–10.0, or .10 g/day), physical
activity (quintiles), total energy intake (quintiles), dietary cholesterol (quintiles), glycemic load (quintiles),
and dietary fiber (quintiles). BMI was categorized as ,21, 21–22.9, 23.0–24.9, 25.0–26.9, 27.0–28.9, 29.0–
30.9, 31.0–32.9, 33.0–34.9, and $35.0 kg/m
2
.
care.diabetesjournals.org DIABETES CAR E 5
Bao and Associates
(35). Although the BCAAs constitute only
around 20% of the total amino acid con-
tent in red meat, they represent the ma-
jority of the amino acids entering the
systemic circulation after a red meat
meal (34). Recently, a BCAAs-related
metabolic signature has been implicated
in the development of insulin resistance
among both obese (36) and nonobese
(37) individuals, and elevated plasma lev-
els of BCAAs, tyrosine, and phenylalanine
have been linked to incident diabetes in a
metabolomics study (10). The associa-
tions of BCAAs and other amino acid in-
takes with GDM risk needs to be
elucidated in future studies.
Strengths of this study include the
large sample size, the high response rates
Table 4dRRs (95% CIs) for GDM according to quintiles of prepregnancy intake (servings per day) of major sources of dietary protein
Q1 Q2 Q3 Q4 Q5 PRR for 1 serving/day
Total red meat
Median intake 0.20 0.48 0.70 0.99 1.50
Model 1 1.00 1.43 (1.14–1.81) 1.69 (1.34–2.14) 2.02 (1.60–2.56) 2.72 (2.16–3.43) ,0.001 2.05 (1.76–2.39)
Model 2 1.00 1.37 (1.08–1.75) 1.61 (1.25–2.08) 1.90 (1.45–2.47) 2.46 (1.86–3.25) ,0.001 1.88 (1.55–2.27)
Model 3 1.00 1.27 (0.99–1.62) 1.43 (1.10–1.85) 1.63 (1.24–2.12) 2.05 (1.55–2.73) ,0.001 1.66 (1.36–2.02)
Unprocessed red meat
Median intake 0.13 0.31 0.44 0.70 1.07
Model 1 1.00 1.39 (1.10–1.75) 1.52 (1.20–1.92) 1.97 (1.57–2.46) 2.48 (1.97–3.12) ,0.001 2.50 (2.03–3.08)
Model 2 1.00 1.24 (0.97–1.58) 1.33 (1.02–1.71) 1.61 (1.25–2.07) 1.89 (1.43–2.48) ,0.001 1.95 (1.50–2.52)
Model 3 1.00 1.17 (0.91–1.49) 1.23 (0.95–1.59) 1.44 (1.12–1.87) 1.60 (1.21–2.12) ,0.001 1.65 (1.26–2.14)
Processed red meat
Median intake 0.00 0.07 0.14 0.28 0.56
Model 1 1.00 1.28 (1.00–1.63) 1.61 (1.29–2.02) 1.72 (1.36–2.17) 2.04 (1.61–2.58) ,0.001 2.85 (2.04–3.98)
Model 2 1.00 1.16 (0.90–1.49) 1.36 (1.07–1.74) 1.38 (1.06–1.79) 1.48 (1.13–1.95) 0.012 1.66 (1.12–2.47)
Model 3 1.00 1.13 (0.88–1.46) 1.28 (1.01–1.64) 1.29 (0.99–1.68) 1.36 (1.03–1.80) 0.062 1.47 (0.98–2.20)
Poultry
Median intake 0.14 0.28 0.43 0.57 0.86
Model 1 1.00 1.12 (0.89–1.39) 1.12 (0.90–1.39) 1.04 (0.82–1.31) 1.15 (0.92–1.45) 0.422 1.12 (0.85–1.49)
Model 2 1.00 1.12 (0.89–1.41) 1.11 (0.88–1.39) 1.04 (0.81–1.33) 1.18 (0.92–1.51) 0.363 1.16 (0.85–1.58)
Model 3 1.00 1.10 (0.88–1.39) 1.08 (0.86–1.36) 0.98 (0.76–1.26) 1.04 (0.81–1.34) 0.874 0.97 (0.71–1.33)
Fish
Median intake 0.07 0.13 0.17 0.24 0.50
Model 1 1.00 0.92 (0.75–1.14) 0.80 (0.63–1.00) 0.96 (0.79–1.16) 0.84 (0.68–1.03) 0.177 0.73 (0.46–1.15)
Model 2 1.00 0.96 (0.77–1.19) 0.83 (0.66–1.05) 1.05 (0.86–1.29) 0.96 (0.77–1.20) 0.993 1.00 (0.62–1.63)
Model 3 1.00 0.95 (0.76–1.18) 0.84 (0.66–1.06) 1.07 (0.88–1.32) 0.95 (0.76–1.18) 0.898 0.97 (0.60–1.57)
Eggs
Median intake 0.00 0.07 0.10 0.14 0.43
Model 1 1.00 0.96 (0.78–1.18) 0.94 (0.66–1.32) 1.00 (0.80–1.23) 1.18 (0.95–1.48) 0.031 1.66 (1.05–2.61)
Model 2 1.00 0.88 (0.71–1.10) 0.86 (0.60–1.23) 0.84 (0.67–1.05) 0.95 (0.75–1.21) 0.587 1.15 (0.69–1.90)
Model 3 1.00 0.88 (0.71–1.10) 0.79 (0.55–1.13) 0.83 (0.66–1.05) 0.93 (0.73–1.19) 0.748 1.09 (0.65–1.81)
Dairy products
Median intake 0.76 1.36 1.97 2.91 4.20
Model 1 1.00 0.96 (0.75–1.21) 0.97 (0.77–1.21) 1.05 (0.84–1.32) 0.82 (0.66–1.03) 0.123 0.96 (0.91–1.01)
Model 2 1.00 1.01 (0.79–1.29) 1.05 (0.82–1.35) 1.11 (0.86–1.43) 0.88 (0.67–1.15) 0.297 0.96 (0.90–1.03)
Model 3 1.00 0.97 (0.76–1.25) 1.04 (0.81–1.34) 1.07 (0.83–1.39) 0.83 (0.63–1.09) 0.144 0.95 (0.89–1.02)
Nuts
Median intake 0.00 0.07 0.14 0.28 0.60
Model 1 1.00 0.86 (0.70–1.07) 0.94 (0.76–1.16) 0.88 (0.72–1.08) 0.69 (0.55–0.87) 0.004 0.59 (0.42–0.84)
Model 2 1.00 0.90 (0.72–1.11) 0.95 (0.77–1.19) 0.92 (0.74–1.14) 0.72 (0.55–0.93) 0.015 0.62 (0.42–0.91)
Model 3 1.00 0.88 (0.71–1.10) 0.95 (0.76–1.19) 0.93 (0.75–1.17) 0.73 (0.56–0.95) 0.028 0.64 (0.44–0.95)
Legumes
Median intake 0.07 0.20 0.29 0.43 0.79
Model 1 1.00 1.02 (0.84–1.24) 0.98 (0.79–1.22) 0.94 (0.77–1.16) 1.07 (0.87–1.33) 0.648 1.07 (0.80–1.43)
Model 2 1.00 1.03 (0.84–1.26) 0.97 (0.77–1.22) 0.91 (0.74–1.13) 1.02 (0.81–1.29) 0.909 0.98 (0.72–1.34)
Model 3 1.00 1.04 (0.84–1.27) 0.98 (0.78–1.22) 0.93 (0.75–1.16) 1.06 (0.84–1.33) 0.854 1.03 (0.75–1.41)
Model 1: age (5-year category) and parity (0, 1, 2, and $3). Model 2: model 1 adjustments plus race/ethnicity (Caucasian, African American, Hispanic, Asian, and
others), family history of diabetes (yes and no), cigarette smoking (never, past, and current), alcohol intake (0, 0.1–5.0, 5.1–10.0, and .10 g/day), physical activity
(quintiles), and total energy intake (quintiles), fruits intake (quintiles), sugar-sweetened beverage intake (quintiles), whole grain intake (quintiles), and mutual
adjustment for other major di etary protein sources. Model 3: model 2 adjustmen ts plus BMI (,21, 21–22.9, 23.0–24.9, 25.0–26.9, 27.0–28.9, 29.0–30.9, 31.0–32.9,
33.0–34.9, and $35.0 kg/m
2
).
6DIABETES CARE care.diabetesjournals.org
Protein intake and GDM risk
of long-term follow-up, and the detailed
prospective dietary assessments with
FFQs that have been extensively validated
against multiple weeks of food records in
previous studies (19–21). We acknowl-
edge that there are several limitations.
First, misclassification of dietary protein
intake is possible. However, the random
within-personerrorwouldbenondif-
ferential, given that the prepregnancy
dietary information was captured pro-
spectively; therefore, our observed associ-
ations may underestimate the true relative
risks. Furthermore, the use of cumulative
averages of dietary intakes for participants
with more than one prepregnancy FFQ
reduces random error. Second, the NHS
II cohort did not assess diet during preg-
nancy. Therefore, we are unable to assess
the association of prepregnancy protein
intake with GDM risk, independent of
diet during pregnancy. However, evi-
dence suggests that the overall dietary pat-
terns and dietary intakes of major protein
sources change little from prepregnancy
to during pregnancy (38,39). In addition,
although GDM is a pregnancy complica-
tion (usually diagnosed in 24–28 weeks
of gestation), increasing evidence sug-
gests that most women with GDM seem
to have a chronic b-cell defect before
pregnancy (40). Women who develop
GDM are thought to have a compromised
capacity to adapt to the metabolic chal-
lenges of pregnancy, which serves to
unmask a predisposition to glucose met-
abolic disorders in these women (40,41).
Therefore, prepregnancy dietary factors
implicated in glucose homeostasis are
also physiologically relevant to the devel-
opment of GDM. Third, our study pop-
ulation consisted mostly of Caucasian
American women; thus, the generaliza-
tion of our findings to other races and
ethnic groups may be limited. However,
the relative homogeneity of our study
population reduces potential confound-
ing. Fourth, plasma amino acid data were
not available thus far in this population.
The inclusion of plasma amino acids data
may help better understand the underly-
ing mechanisms for the divergent effects
of animal and vegetable protein intakes
on GDM risk. Finally, although major
potential confounders have been ad-
justed in the current study, we cannot
completely rule out the possibility of re-
sidual confounding from unmeasured
factors. In addition, because of the high
correlations among nutrients coexisting
with protein in common food sources,
we could not exclude the possibility of
overadjustment, which may lead to an
underestimate of the real associations of
animal and vegetable protein intake with
GDM risk.
In summary, our findings indicate
that prepregnancy intake of animal pro-
tein, in particular red meat, is significantly
and positively associated with GDM risk,
whereas consumption of vegetable pro-
tein, specifically nuts, is inversely associ-
ated with the risk. Moreover, our findings
suggest that among women of reproduc-
tive age, substitution of vegetable protein
for animal protein, as well as substitution
of some healthy protein sources (e.g.,
nuts, legumes, poultry, and fish) for red
meat may potentially lower GDM risk.
Along with our previous findings on
associations of GDM risk with carbohy-
drates (6) and fats (7), the joint effects of
different amounts and types of these mac-
ronutrients on the risk of GDM warrant
further investigation in future studies.
AcknowledgmentsdThis study was funded
by the Intramural Research Program of the
Eunice Kennedy Shriver National Institute of
Child Health and Human Development, Na-
tional Institutes of Health (contract No.
HHSN275201000020C). The Nurses’Health
Study II was funded by research grants
DK58845, CA50385, and P30 DK46200 from
the National Institutes of Health.
No potential conflicts of interest relevant to
this article were reported.
W.B. contributed to the design and analysis
of the study and wrote the manuscript. K.B.
contributed to the data analysis, and reviewed
and edited the manuscript. D.K.T. conducted
the technique review and reviewed and edited
the manuscript. F.B.H. interpreted the results
and reviewed and edited the manuscript. C.Z.
contributed to the design and analysis of the
study, and reviewed and edited the manu-
script. W.B. and C.Z. are the guarantors of this
work and, as such, had full access to all the
data in the study an d take responsibility for the
integrity of the data and the accuracy of the
data analysis.
Parts of this study were presented in ab-
stract form at the 45th Society for Epidemio-
logic Research Annual Meeting, Minneapolis,
Minnesota, 27–30 June 2012.
References
1. Reece EA, Leguizamón G, Wiznitzer A.
Gestational diabetes: the need for a com-
mon ground. Lancet 2009;373:1789–1797
2. American Diabetes Association. Gesta-
tional diabetes mellitus. Diabetes Care
2004;27(Suppl 1):S88–S90
3. Metzger BE, Lowe LP, Dyer AR, et al.;
HAPO Study Cooperative Research Group.
Hyperglycemia and adverse pregnancy out-
comes. N Engl J Med 2008;358:1991–
2002
4. Bellamy L, Casas JP, Hingorani AD,
Williams D. Type 2 diabetes mellitus after
gestational diabetes: a systematic review
and meta-analysis. Lancet 2009;373:
1773–1779
5. Zhang C, Ning Y. Effect of dietary and
lifestyle factors on the risk of gestational
diabetes: review of epidemiologic evi-
dence. Am J Clin Nutr 2011;94(Suppl.):
1975S–1979S
6. Zhang C, Liu S, Solomon CG, Hu FB.
Dietary fiber intake, dietary glycemic load,
and the risk for gestational diabetes mel-
litus. Diabetes Care 2006;29:2223–2230
7. Bowers K, Tobias DK, Yeung E, Hu FB,
Zhang C. A prospective study of pre-
pregnancy dietary fat intake and risk of
gestational diabetes. Am J Clin Nutr 2012;
95:446–453
8. Tremblay F, Lavigne C, Jacques H,
Marette A. Role of dietary proteins and
amino acids in the pathogenesis of insulin
resistance. Annu Rev Nutr 2007;27:293–
310
9. Sluijs I, Beulens JW, van der A DL,
Spijkerman AM, Grobbee DE, van der
Schouw YT. Dietary intake of total, ani-
mal, and vegetable protein and risk of type
2 diabetes in the European Prospective
Investigation into Cancer and Nutrition
(EPIC)-NL study. Diabetes Care 2010;33:
43–48
10. Wang TJ, Larson MG, Vasan RS, et al.
Metabolite profiles and the risk of de-
veloping diabete s. Nat Med 2011;17:448–
453
11. Pan A, Sun Q, Bernstein AM, et al. Red
meat consumption and risk of type 2 di-
abetes: 3 cohorts of US adults and an
updated meta-analysis. Am J Clin Nutr
2011;94:1088–1096
12. Zhang C, Schulze MB, Solomon CG, Hu
FB. A prospective study of dietary pat-
terns, meat intake and the risk of gesta-
tional diabetes mellitus. Diabetologia
2006;49:2604–2613
13. Jiang R, Manson JE, Stampfer MJ, Liu S,
Willett WC, Hu FB. Nut and peanut but-
ter consumption and risk of type 2 di-
abetes in women. JAMA 2002;288:2554–
2560
14. Villegas R, Gao YT, Yang G, et al. Legume
and soy food intake and the incidence of
type 2 diabetes in the Shanghai Women’s
Health Study. Am J Clin Nutr 2008;87:
162–167
15. Solomon CG, Willett WC, Carey VJ, et al.
A prospective study of pregravid deter-
minants of gestational diabetes mellitus.
JAMA 1997;278:1078–1083
16. Bernstein AM, Sun Q, Hu FB, Stampfer
MJ, Manson JE, Willett WC. Major dietary
protein sources and risk of coronary heart
disease in women. Circulation 2010;122:
876–883
care.diabetesjournals.org DIABETES CAR E 7
Bao and Associates
17. Bernstein AM, Pan A, Rexrode KM, et al.
Dietary protein sources and the risk of
stroke in men and women. Stroke 2012;
43:637–644
18. Hu FB, Stampfer MJ, Manson JE, et al.
Dietary protein and risk of ischemic heart
disease in women. Am J Clin Nutr 1999;
70:221–227
19. Willett WC, Sampson L, Stampfer MJ,
et al. Reproducibility and validity of a
semiquantitative food frequency question-
naire. Am J Epidemiol 1985;122:51–65
20. Will ett WC, Sampson L, Browne ML, et al.
The use of a self-administered question-
naire to assess diet four years in the past.
Am J Epidemiol 1988;127:188–199
21. Salvini S, Hunter DJ, Sampson L, et al.
Food-based validation of a dietary ques-
tionnaire: the effects of week-to-week
variation in food consumption. Int J Epi-
demiol 1989;18:858–867
22. Rimm EB, Stampfer MJ, Colditz GA,
Chute CG, Litin LB, Willett WC. Validity
of self-reported waist and hip circum-
ferences in men and women. Epidemiol-
ogy 1990;1:466–473
23. Wolf AM, Hunter DJ, Colditz GA, et al.
Reproducibility and validity of a self-
administered physical activity question-
naire. Int J Epidemiol 1994;23:991–999
24. Willett WC. Nutritional Epidemiology.2nd
ed. New York, Oxford University Press,
1998
25. Hu FB, Stampfer MJ, Rimm E, et al. Di-
etary fat and coronary heart disease:
a comparison of approaches for adjusting
for total energy intake and modeling
repeated dietary measurements. Am J
Epidemiol 1999;149:531–540
26. Hu FB, Stampfer MJ, Manson JE, et al.
Dietary fat intake and the risk of coronary
heart disease in women. N Engl J Med
1997;337:1491–1499
27. Pan A, Sun Q, Bernstein AM, et al. Red
meat consumption and mortality: results
from 2 prospective cohort studies. Arch
Intern Med 2012;172:555–563
28. Lao TT, Ho LF, Chan BC, Leung WC.
Maternal age and prevalence of gestational
diabetes mellitus. Diabetes Care 2006;29:
948–949
29. Lin DY, Fleming TR, De Gruttola V. Esti-
mating the proportion of treatment effect
explained by a surrogate marker. Stat Med
1997;16:1515–1527
30. Linn T, Santosa B, Grönemeyer D, et al.
Effect of long-term dietary protein intake
on glucose metabolism in humans. Dia-
betologia 2000;43:1257–1265
31. Daniel CR, Cross AJ, Koebnick C, Sinha R.
Trends in meat consumption in the USA.
Public Health Nutr 2011;14:575–583
32. Mozaffarian D, Hao T, Rimm EB, Willett
WC, Hu FB. Changes in diet and lifestyle
and long-term weight gain in women and
men. N Engl J Med 2011;364:2392–2404
33. Kendall CW, Josse AR, Esfahani A,
Jenkins DJ. Nuts, metabolic syndrome
and diabetes. Br J Nutr 2010;104:465–
473
34. Adeva MM, Calviño J, Souto G, Donapetry
C. Insulin resistance and the metabolism
of branched-chain amino acids in hu-
mans. Amino Acids 2012;43:171–181
35. Brandsch C, Shukla A, Hirche F, Stangl
GI, Eder K. Effect of proteins from beef,
pork, and turkey me at on plasma and liver
lipids of rats compared with casein and
soy protein. Nutrition 2006;22:1162–
1170
36. Newgard CB, An J, Bain JR, et al. A
branched-chain amino acid-related met-
abolic signature that differentiates obese
and lean humans and contributes to in-
sulin resistance. Cell Metab 2009;9:311–
326
37. Tai ES, Tan ML, Stevens RD, et al. Insulin
resistance is associated with a metabolic
profile of altered protein metabolism in
Chinese and Asian-Indian men. Dia-
betologia 2010;53:757–767
38. Cucó G, Fern
andez-Ballart J, Sala J, et al.
Dietary patterns and associated lifestyles
in preconception, pregnancy and post-
partum. Eur J Clin Nutr 2006;60:364–
371
39. Crozier SR, Robinson SM, Godfrey KM,
Cooper C, Inskip HM. Women’s dietary
patterns change little from before to dur-
ing pregnancy. J Nutr 2009;139:1956–
1963
40. Buchanan TA, Xiang AH, Page KA. Ges-
tational diabetes mellitus: risks and man-
agement during and after pregnancy. Nat
Rev Endocrinol 2012;8:639–649
41. Zhang C, Solomon CG, Manson JE, Hu
FB. A prospective study of pregravid
physical activity and sedentary behaviors
in relation to the risk for gestational di-
abetes mellitus. Arch Intern Med 2006;
166:543–548
8DIABETES CARE care.diabetesjournals.org
Protein intake and GDM risk
