See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/263291471

Fried-food consumption and risk of type 2 diabetes and coronary artery

disease: A prospective study in 2 cohorts of US women and men

Article in American Journal of Clinical Nutrition · June 2014

DOI: 10.3945/ajcn.114.084129 · Source: PubMed

CITATIONS READS

158 486

7 authors, including:

Leah Cahill An Pan

Dalhousie University Huazhong University of Science and Technology
77 PUBLICATIONS 11,539 CITATIONS 487 PUBLICATIONS 36,722 CITATIONS

SEE PROFILE SEE PROFILE

Stephanie E Chiuve Qi Sun

Brigham and Women's Hospital Harvard Medical School
159 PUBLICATIONS 27,746 CITATIONS 554 PUBLICATIONS 34,137 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by An Pan on 30 December 2015.

The user has requested enhancement of the downloaded file.

Fried-food consumption and risk of type 2 diabetes and coronary artery
disease: a prospective study in 2 cohorts of US women and men1–4
Leah E Cahill, An Pan, Stephanie E Chiuve, Qi Sun, Walter C Willett, Frank B Hu, and Eric B Rimm
ABSTRACT
Background: Through the processes of oxidation, polymerization,
and hydrogenation, the cooking method of frying modifies both
foods and their frying medium. However, it remains unknown whether
the frequent consumption of fried foods is related to long-term cardio-
metabolic health.
Objective: We examined fried-food consumption and risk of de-
veloping incident type 2 diabetes (T2D) or coronary artery disease
(CAD).
Design: Fried-food consumption was assessed by using a question-
naire in 70,842 women from the Nurses’ Health Study (1984–2010)
and 40,789 men from the Health Professionals Follow-Up Study
(1986–2010) who were free of diabetes, cardiovascular disease, and
cancer at baseline. Time-dependent Cox proportional hazards models
were used to estimate RRs and 95% CIs for T2D and CAD adjusted
for demographic, diet, lifestyle, and other cardiometabolic risk fac-
tors. Results were pooled by using an inverse variance–weighted
random-effects meta-analysis.
Results: We documented 10,323 incident T2D cases and 5778 in-
cident CAD cases. Multivariate-adjusted RRs (95% CIs) for indi-
viduals who consumed fried foods ,1, 1–3, 4–6, or $7 times/wk
were 1.00 (reference), 1.15 (0.97, 1.35), 1.39 (1.30, 1.49), and 1.55
(1.32, 1.83), respectively, for T2D and 1.00 (reference), 1.06 (0.98,
1.15), 1.23 (1.14, 1.33), and 1.21 (1.06, 1.39), respectively, for CAD.
Associations were largely attenuated when we further controlled for
biennially updated hypertension, hypercholesterolemia, and body
mass index.
Conclusions: Frequent fried-food consumption was significantly
associated with risk of incident T2D and moderately with incident
CAD, and these associations were largely mediated by body weight
and comorbid hypertension and hypercholesterolemia. Am J
Clin Nutr 2014;100:667–75.
INTRODUCTION
Frying is a common cooking method in Western countries,
especially outside of the home where French fries and fried-
chicken products make up a substantial percentage of the items
sold at fast-food restaurants. Through the processes of oxidation,
polymerization, and hydrogenation, frying modifies both the
composition of foods and their frying medium. With repeated use,
oils deteriorate, which leads to a change in the fatty acid com-
position and absorption of other oil-degradation products into
fried foods (1, 2). If proven to be detrimental to health, fried-food
consumption could cause a substantial health burden because 25–
36% of North American adults patronize fast-food restaurants
Am J Clin Nutr 2014;100:667–75. Printed in USA. Ó 2014 American Society for Nutrition
daily (3, 4). However, if foods are fried with previously unused
polyunsaturated and monounsaturated oils, recent evidence has
suggested that there may be a benefit from the oils’ n26 and
n23 fatty acids if they remain intact after frying (5). Therefore,
making dietary recommendations for an appropriate frequency
of fried-food consumption is currently complex.
Despite substantial research into the relation between diet and
Downloaded from ajcn.nutrition.org by guest on October 27, 2015
major chronic diseases, the influence of fried-food consumption
on long-term cardiometabolic health is unclear. In cross-sectional
studies, fried-food consumption has been positively associated
with several cardiometabolic risk factors including hypertension
(6), low serum HDL cholesterol (7), and obesity (7, 8). In pro-
spective studies, the Western-style dietary pattern includes fried
foods as a major component and is generally positively associated
with increased risk of type 2 diabetes (T2D)5 (9); however, to our
knowledge, no prospective research has specifically quantified
the association between fried-food consumption and T2D. One
prospective cohort study and 2 case-control studies have re-
ported inconsistent results of the association between fried-food
consumption and risk of coronary artery disease (CAD) (10–12).
Therefore, we aimed to examine prospectively whether the fre-
quency of fried-food consumption (both at home and away from
home) is associated with risk of incident T2D or CAD.
1
From the Departments of Nutrition (LEC, QS, WCW, FBH, and EBR)
and Epidemiology (WCW, FBH, and EBR), Harvard School of Public
Health, Boston, MA; the Channing Division of Network Medicine (QS,
WCW, FBH, and EBR) and Division of Preventive Medicine (SEC), De-
partment of Medicine, Brigham and Women’s Hospital and Harvard Medical
School, Boston, MA; and the Saw Swee Hock School of Public Health (AP)
and Yong Loo Lin School of Medicine (AP), National University of Singa-
pore and National University Health System, Singapore.
2
LEC and AP contributed equally to the article.
3
Supported by the NIH (grants P01CA087969, UM1CA167552, RO1HL35464,
RO1HL034594, U19CA055075, R01DK058845, P30DK046200, RO1HL60712,
and U54CA155626) and a Canadian Institutes of Health Research Postdoctoral
Fellowship (to LEC).
4
Address correspondence to LE Cahill, Department of Nutrition, Harvard
School of Public Health, 655 Huntington Avenue, Boston, MA 02115. E-mail:
lcahill@hsph.harvard.edu; or A Pan, Saw Swee Hock School of Public Health,
National University of Singapore, 16 Medical Drive, Singapore 117597. E-mail:
ephanp@nus.edu.sg.
5
Abbreviations used: AHEI, Alternative Healthy Eating Index; CAD, cor-
onary artery disease; FFQ, food-frequency questionnaire; HPFS, Health Pro-
fessionals Follow-Up Study; MI, myocardial infarction; NHS, Nurses’ Health
Study; T2D, type 2 diabetes.
Received January 17, 2014. Accepted for publication May 8, 2014.
First published online June 18, 2014; doi: 10.3945/ajcn.114.084129.
667
Supplemental Material can be found at:
http://ajcn.nutrition.org/content/suppl/2014/06/18/ajcn.114.0
84129.DCSupplemental.html
668 CAHILL ET AL
SUBJECTS AND METHODS
Study populations
The Nurses’ Health Study (NHS) is a prospective cohort of
121,700 female registered nurses aged 30–55 y at baseline in
1976. The Health Professionals Follow-Up Study (HPFS) is a
prospective study of 51,529 male health professionals aged
40–75 y at enrollment in 1986. Participants in both studies have
been followed through mailed biennial questionnaires that
ascertained medical histories, lifestyles, and health-related be-
haviors as previously described (13). More than 95% of par-
ticipants were of white European descent. Study protocols were
approved by the institutional review boards of the Brigham and
Women’s Hospital and the Harvard School of Public Health.
In the current analysis, we used the years when fried-food
consumption was first assessed in the cohorts as the baseline
(1984 for NHS: n = 97,476; 1986 for HPFS n = 51,529). Par-
ticipants were excluded from the current analysis at baseline if
they had self-reported cancer (except nonmelanoma skin can-
cer), diabetes, cardiovascular disease (angina, stroke, CAD, or
coronary artery bypass graft surgery) (n = 9890 women and 6554
men), reported unusual energy intakes (,800 or .4200 kcal/d
for men and ,500 or .3500 kcal/d for women) or did not answer
the food-frequency questionnaire (FFQ) or questions of fried-food
consumption (n = 16,026 women, and 1948 men) and subjects
who only replied to the baseline questionnaire or had an unknown
mortality or disease status during follow-up (n = 718 women and
2238 men). After exclusions, data from 70,842 women and 40,789
men were available for the analysis. The retention percentage of the
2 cohorts is .90% of potential person-times.
Fried-food consumption and other dietary assessment
On the FFQ, participants were asked “how often do you eat
fried food away from home (e.g. French fries, fried chicken,
fried fish)?” and “how often do you eat food that is fried at home
(Exclude the use of “Pam”-type spray)?” Both questions had 4
possible responses as follows: ,1, 1–3, 4–6 or $7 times/wk.
Questions were assessed in the NHS in 1984, 1986, and every
4 y thereafter and every 4 y starting from 1986 in the HPFS.
Fried-food consumption at home and fried-food consumption
away from home were analyzed separately and also added to-
gether to examine total fried-food consumption. To assess the
overall diet quality, a diet score for each participant was cal-
culated on the basis of the previously described 2010 Alternative
Healthy Eating Index (AHEI) (14), which was designed to target
food choices that have been associated with reduced chronic
disease risk.
Outcome assessment
Outcomes for the current analysis were incident T2D and CAD
[defined as nonfatal myocardial infarction (MI) or fatal CAD]. On
each biennial questionnaire, participants were asked to indicate
whether they had physician-diagnosed T2D or MI in the previous
2 y. For newly diagnosed T2D, we sent out a supplementary ques-
tionnaire to collect information regarding symptoms, diagnostic
tests, and hypoglycemic therapy. T2D was confirmed on the basis
of National Diabetes Data Group diagnostic criteria (15) before
1997 and American Diabetes Association criteria (16) after 1997
by using information from the supplementary questionnaire. Only
confirmed cases were included in the analysis. The validity of the
supplementary questionnaire for diabetes diagnosis has been
documented previously in both the HPFS (17) and NHS (18).
For newly reported MI, medical records and autopsy reports
were examined for confirmation by study physicians blinded to
the participant’s exposure status. Nonfatal MI was defined by
WHO criteria, which require clinical symptoms and either di-
agnostic changes on electrocardiogram or elevated cardiac en-
zymes (19). Deaths were identified from state vital records and
the National Death Index or reported by the participant’s next of
kin or the postal system. Fatal CAD was confirmed by hospital
records or autopsy.
Statistical analysis
We calculated each individual’s person-years from the date
of return of the baseline questionnaire to the date of diagnosis
of T2D or CAD, last returned questionnaire, death, or the end of
follow-up (30 June 2010 for the NHS and 31 January 2010 for
Downloaded from ajcn.nutrition.org by guest on October 27, 2015
the HPFS), whichever came first. To examine associations be-
tween fried-food consumption and cardiometabolic diseases, we
used time-dependent Cox proportional hazards models condi-
tioned on age and follow-up cycle to estimate RRs and 95% CIs
for both cohorts individually and then pooled together with the
use of an inverse-variance–weighted meta-analysis by a random-
effects model, which allowed for between-study heterogeneity
(20).
In multivariate models, we adjusted for the following potential
risk factors of cardiometabolic diseases: age (continuous), white
(yes or no), family history of diabetes (yes or no), smoking status
[never, past, or current (1–14, 15–24, or $25 cigarettes/d)], al-
cohol intake (0, 0.1–4.9, 5.0–14.9, or $15.0 g/d in women; 0,
0.1–4.9, 5.0–29.9, or $30.0 g/d in men), physical activity (,3.0,
3.0–8.9, 9.0–17.9, 18.0–26.9, or $27.0 metabolic equivalent
task hours per week (21, 22), total energy intake (quintiles), and
diet quality as represented by the AHEI (quintiles). Both total
polyunsaturated fats and trans fats were components of the
AHEI-2010. In sensitivity analyses, we further adjusted for specific
types of fat such as trans fat and polyunsaturated fat as well as
versions of the AHEI that were derived without trans and poly-
unsaturated fats. In women, we also adjusted for postmenopausal
status and menopausal hormone use [premenopausal or post-
menopausal (never, past, or current hormone use)]. We further
adjusted for biennially updated hypertension, hypercholesterolemia,
and BMI (in kg/m2) ,23.0, 23.0–24.9, 25.0–29.9, 30.0–34.9,
or $35.0) to assess whether risks for T2D or CAD could be
mediated by these comorbidities.
Most nondietary variables were updated every 2 y in the analysis,
and dietary variables were updated every 4 y. We stopped updating
dietary variables, including fried-food consumption, when a
participant reported a diagnosis of hypertension, hypercholes-
terolemia, or cancer (and also diabetes for the CAD analysis)
because these conditions might lead to changes in diet (23). We
also conducted a sensitivity analysis of continuing to update di-
etary variables after the diagnosis of these comorbidities alongside
adjustment for the comorbidity.
We conducted analyses stratified by cardiometabolic risk
factors including BMI, AHEI, hypertension, hypercholesterol-
emia, and dietary factors (including red meat, fish, chicken, and
 FRIED FOOD, DIABETES, AND CORONARY ARTERY DISEASE
potato consumption) to determine whether any interactions existed
between risk factors and fried-food consumption on risk of T2D
or CAD. The likelihood ratio test was used to test for interactions
of cross-product terms. We tested for proportional hazards as-
sumptions and examined differences over time by dividing the
study into 2 time periods to incorporate knowledge development
regarding trans fats (23). Analyses stratified by the type of fat
used for frying at home were also conducted. Data were ana-
lyzed with SAS software (version 9.2; SAS Institute). We con-
sidered 2-tailed P values #0.05 to be statistically significant.
RESULTS
In the NHS, 14.0% and 3.5% of women reported fried-food
consumption 4–6 and $7 times/wk, respectively (Table 1).
Corresponding proportions in men were 22.6% and 7.4%. In
both cohorts, frequent fried-food consumption was related to
younger age, lower physical activity, higher BMI, and a higher
prevalence of smoking. Participants with a higher frequency of
fried-food consumption also had a lower diet quality as mea-
sured by the AHEI. Specific to food items, participants with a
higher frequency of fried-food consumption consumed more red
meat, potatoes, and sugar-sweetened beverages and less fruit,
vegetables, whole grains, fish, and alcohol. The total energy
from trans fat was higher with the greater frequency of fried-
food consumption. We showed that participants who frequently
ate fried food at home were also more likely to consume fried
food away from home. However, people who frequently ate fried
food away from home ($4 times/wk) were younger, less likely
to be married, and drank more sugar-sweetened beverages than
did subjects who frequently ate fried food at home ($4 times/wk)
(see Supplemental Table 1 under “Supplemental data” in the online
issue).
During follow-up, we documented 10,323 incident T2D cases
(6974 women and 3349 men). In the multivariate-adjusted model,
compared with participants who consumed fried foods ,1 time/wk,
individuals who consumed fried foods 1–3, 4–6, or $7 times/wk
had significantly higher risk of T2D; pooled RRs (95% CIs) were
1.15 (0.97, 1.35), 1.39 (1.30, 1.49), and 1.55 (1.32, 1.83), re-
spectively (Table 2). The association was largely attenuated but
remained significant after additional adjustment for biennially
updated hypertension, hypercholesterolemia, and BMI with pooled
RRs (95% CIs) of 1.06 (0.97, 1.16), 1.13 (1.07, 1.20), and 1.19
(1.00, 1.40), respectively.
During follow-up, we documented 5778 incident CAD cases
(2687 women and 3091 men). In comparison with participants
who consumed fried foods ,1 time/wk, pooled multivariate-
adjusted RRs (95% CIs) for risk of CAD were 1.06 (0.98, 1.15),
1.23 (1.14, 1.33), and 1.21 (1.06, 1.39), for individuals who
consumed fried foods 1–3, 4–6, and $7 times/wk, respectively
(Table 3). The association was mostly attenuated with pooled
RRs (95% CIs) of 1.03 (0.97, 1.09), 1.13 (1.04, 1.22), and 1.08
(0.95, 1.24), respectively, after further adjustment for biennially
updated hypertension, hypercholesterolemia, and BMI. When
CAD analyses were conducted for fatal and nonfatal CAD sep-
arately from each other (data not shown), results were similar to
what we reported for total CAD.
The association with T2D was generally stronger for eating
fried food away from home [RR: 1.81 (95% CI: 1.58, 2.08) for
comparison of $4 times/ wk with ,1 time/wk] than eating fried
669
food at home (corresponding RR: 1.26; 95% CI: 1.09, 1.47) in
the multivariate model without hypertension, hypercholesterol-
emia, and BMI (Table 2). We also showed that eating fried food
away from home was more related to CAD than eating fried
food at home in women (Table 3) but not men.
Results for both T2D and CAD were essentially unchanged
when we adjusted for specific types of fat such as trans fat and
polyunsaturated fat or for other AHEI variables derived differ-
ently in regards to these specific types of fat (data not shown).
Results remained unchanged in the sensitivity analysis of con-
tinuing to update dietary variables after the diagnosis of a po-
tential intermediate health condition (data not shown). For both
T2D and CAD, we generally did not find significant interactions
between fried-food consumption and BMI, hypertension, hy-
percholesterolemia, dietary quality, or specific food items (red
meat, fish, chicken, or potato), or whether cases occurred before
or after 1998 (see Supplementary Tables 2 and 3 under “Sup-
plemental data” in the online issue). Neither stratified analyses
by these factors or by type of frying oil provided additional in-
formation on associations between fried-food consumption and
Downloaded from ajcn.nutrition.org by guest on October 27, 2015
cardiometabolic diseases (data not shown).
DISCUSSION
In 2 large, prospective cohorts, we observed that frequent
fried-food consumption was significantly associated with risk of
incident T2D and CAD. These associations remained significant
after extensive adjustment for demographic, diet, and lifestyle
factors.
To the best of our knowledge, this is the first prospective study
to examine the relation between the frequency of fried-food
consumption and risk of T2D. Previous studies have reported that
some of the major food items commonly used for frying, such as
potatoes and red meat, are positively associated with increased
risk of T2D (24–29). However, our results were not substantially
altered with adjustment for the overall diet quality and specific
food items, which suggested that the association was independent
of food items used for frying. The positive association between
fried-food consumption and T2D has been supported by findings
from other longitudinal studies that investigated the association
between fast- or restaurant-food consumption and T2D risk,
including the Black Women’s Health Study (30), the Singapore
Chinese Health Study (31), and the Coronary Artery Risk De-
velopment in Young Adults study (32).
We showed a stronger association with risk of T2D when
eating fried food away from home than eating at home, which
may have several explanations. First, oils deteriorate during frying,
especially when the oils are reused, which is a practice that may
be more common away from home than at home. Second, portion
sizes are often larger (33) and sodium contents higher (34) in
restaurant meals than in meals eaten at home. Furthermore, specific
food choices might differ at home compared with in a restaurant;
eg, our participants who frequently ate fried foods away from
home had higher sugar-sweetened–beverage consumption than
did those who did not (data not shown). However, the associations
remained after adjustment for diet quality, which included caloric
intake, sodium intake, and sugar-sweetened–beverage consumption.
In the current analysis, fried-food consumption was moder-
ately associated with incident CAD. Recently, the Spanish cohort
of the European Prospective Investigation into Cancer and Nutrition
670 CAHILL ET AL
TABLE 1
Baseline characteristics of participants according to the frequency of fried-food consumption in the NHS and the HPFS1
Frequency of fried-food consumption

Characteristic ,1 time/wk 1–3 times/wk 4–6 times/wk $7 times/wk

NHS (1984)
n 35,594 22,857 9910 2481
Age (y) 50.8 6 7.12 49.9 6 7.1 48.4 6 7.0 48.5 6 6.8
BMI (kg/m2) 24.6 6 4.4 24.9 6 4.6 25.5 6 5.1 26.1 6 5.6
Physical activity (MET-h/wk) 16.0 6 23.0 12.8 6 19.1 11.5 6 15.8 10.2 6 18.3
Race (white) (%) 98.2 97.6 97.4 96.3
Marital status (married) (%) 87.7 88.5 87.3 85.6
Current smoker (%) 23.0 25.1 25.5 25.9
Hypertension (%) 20.3 19.7 19.0 21.3
High cholesterol (%) 8.2 7.0 6.3 7.1
Family history of diabetes (%) 27.6 29.1 28.9 30.9
Parental MI before age 60 y (%) 19.2 19.1 18.8 19.7
Premenopausal (%) 48.9 53.5 61.2 60.7
Dietary intake
Total energy (kcal/d) 1618 6 494 1809 6 513 1961 6 540 2122 6 581
Alcohol (g/d) 7.2 6 11.2 7.0 6 11.5 6.5 6 11.1 5.9 6 11.0
AHEI score 50.8 6 10.8 45.5 6 9.8 42.9 6 9.2 40.9 6 9.1

Downloaded from ajcn.nutrition.org by guest on October 27, 2015

Fruit (servings/d) 2.3 6 1.4 2.0 6 1.3 1.9 6 1.2 1.8 6 1.3
Vegetables (servings/d) 3.2 6 1.7 2.9 6 1.5 2.8 6 1.5 2.8 6 1.5
Whole grain (g/d) 16.1 6 14.5 12.5 6 11.4 10.6 6 9.4 9.3 6 8.6
Nuts (servings/d) 0.1 6 0.2 0.1 6 0.2 0.1 6 0.2 0.1 6 0.2
Potato (servings/d) 0.4 6 0.3 0.6 6 0.4 0.7 6 0.4 0.8 6 0.5
Dairy products (servings/d) 2.0 6 1.3 2.0 6 1.3 2.0 6 1.3 2.0 6 1.4
SSB (servings/d) 0.7 6 1.0 0.8 6 1.1 1.0 6 1.1 1.1 6 1.3
Coffee (cups/d) 2.4 6 1.8 2.5 6 1.9 2.5 6 2.0 2.7 6 2.1
Red meat (servings/d) 0.9 6 0.6 1.2 6 0.7 1.5 6 0.7 1.7 6 0.8
Fish (servings/d) 0.3 6 0.3 0.2 6 0.2 0.2 6 0.2 0.2 6 0.2
Poultry (servings/d) 0.3 6 0.2 0.3 6 0.2 0.3 6 0.2 0.3 6 0.3
PS 0.5 6 0.2 0.6 6 0.2 0.6 6 0.2 0.5 6 0.2
trans Fat (% of energy) 1.7 6 0.6 2.1 6 0.6 2.2 6 0.6 2.3 6 0.6
HPFS (1986)
n 15,232 13,317 9218 3022
Age (y) 54.3 6 9.5 52.8 6 9.4 51.4 6 9.2 50.9 6 9.0
BMI (kg/m2) 25.2 6 3.1 25.5 6 3.3 25.8 6 3.4 25.8 6 3.3
Physical activity (MET-h/wk) 24.3 6 32.4 20.5 6 27.8 19.6 6 28.7 16.9 6 27.2
Race (white) 95.3 95.2 94.9 94.3
Marital status (married) 91.2 90.7 91.4 90.4
Current smoker 7.8 9.8 10.5 14.1
Hypertension 20.0 19.0 18.4 16.7
High cholesterol 11.4 10.0 9.3 8.0
Family history of diabetes 18.8 18.4 19.4 18.9
Parental MI before age 60 y 12.2 11.8 11.8 11.9
Dietary intake
Total energy (kcal/d) 1839 6 564 1980 6 591 2126 6 609 2356 6 643
Alcohol (g/d) 11.1 6 14.8 11.7 6 15.6 11.4 6 15.3 11.5 6 16.4
AHEI score 56.8 6 11.3 51.7 6 10.8 48.8 6 10.3 45.1 6 9.9
Fruit (servings/d) 2.7 6 1.7 2.3 6 1.5 2.1 6 1.4 1.8 6 1.3
Vegetables (servings/d) 3.3 6 1.8 3.0 6 1.6 2.8 6 1.5 2.8 6 1.6
Whole grain (g/d) 26.5 6 22.7 20.6 6 17.8 17.3 6 14.7 14.0 6 11.9
Nuts (servings/d) 0.2 6 0.4 0.2 6 0.4 0.3 6 0.4 0.3 6 0.4
Potato (servings/d) 0.4 6 0.3 0.6 6 0.4 0.7 6 0.4 0.9 6 0.5
Dairy products (servings/d) 1.8 6 1.3 2.0 6 1.4 2.0 6 1.4 2.1 6 1.5
SSB (servings/d) 0.6 6 1.0 0.8 6 1.0 0.9 6 1.0 1.0 6 1.1
Coffee (cups/d) 1.8 6 1.7 1.9 6 1.8 2.0 6 1.9 2.2 6 1.9
Red meat (servings/d) 0.8 6 0.7 1.2 6 0.7 1.4 6 0.8 1.9 6 0.9
Fish (servings/d) 0.4 6 0.3 0.3 6 0.3 0.3 6 0.2 0.3 6 0.2
Poultry (servings/d) 0.4 6 0.3 0.3 6 0.2 0.3 6 0.3 0.3 6 0.3
PS 0.6 6 0.2 0.6 6 0.2 0.5 6 0.2 0.51 6 0.2
trans Fat (% of energy) 1.1 6 0.5 1.3 6 0.5 1.4 6 0.5 1.6 6 0.5
1
AHEI, Alternative Healthy Eating Index; HPFS, Health Professionals Follow-Up Study; MET-h, metabolic equivalent task-hours; MI, myocardial
infarction; NHS, Nurses’ Health Study; PS, polyunsaturated fat:saturated fat; SSB, sugar-sweetened beverage.
2
Mean 6 SD (all such values).
 FRIED FOOD, DIABETES, AND CORONARY ARTERY DISEASE 671
TABLE 2
RRs (95% CIs) for type 2 diabetes in the NHS and the HPFS according to the frequency of fried-food consumption1
Frequency of fried-food consumption

,1 time/wk 1–3 times/wk 4–6 times/wk $7 times/wk P-trend

NHS (1984–2010)
Total fried food
Cases/person-years 3364/910,839 2336/522,931 1027/169,336 247/34,199 —
Age-adjusted model 1.00 1.21 (1.15, 1.27) 1.76 (1.64, 1.89) 2.12 (1.87, 2.42) ,0.001
Model 1 1.00 1.14 (1.08, 1.20) 1.55 (1.44, 1.66) 1.70 (1.50, 1.94) ,0.001
Model 2 1.00 1.06 (1.00, 1.12) 1.35 (1.26, 1.45) 1.43 (1.25, 1.64) ,0.001
Model 3 1.00 1.05 (1.00, 1.11) 1.22 (1.14, 1.32) 1.23 (1.08, 1.41) ,0.001
Model 4 1.00 1.02 (0.96, 1.08) 1.11 (1.04, 1.20) 1.09 (0.96, 1.25) 0.004
Fried food at home
Cases/person-years 4033/1,021,194 2531/549,827 410/66,283 —
Age-adjusted model 1.00 1.16 (1.10, 1.22) 1.66 (1.49, 1.83) ,0.001
Model 1 1.00 1.10 (1.05, 1.16) 1.40 (1.26, 1.55) ,0.001
Model 2 1.00 1.01 (0.96, 1.06) 1.18 (1.06, 1.31) 0.009
Model 3 1.00 1.00 (0.95, 1.05) 1.06 (0.95, 1.17) 0.36
Model 4 1.00 0.99 (0.94, 1.04) 1.01 (0.91, 1.12) 0.98
Fried food away from home

Downloaded from ajcn.nutrition.org by guest on October 27, 2015

Cases/person-years 5313/1,371,685 1563/25,4850 98/10,769 —
Age-adjusted model 1.00 1.72 (1.62, 1.82) 2.64 (2.16, 3.23) ,0.001
Model 1 1.00 1.54 (1.45, 1.63) 2.22 (1.82, 2.71) ,0.001
Model 2 1.00 1.41 (1.33, 1.49) 1.98 (1.62, 2.42) ,0.001
Model 3 1.00 1.28 (1.21, 1.36) 1.69 (1.38, 2.07) ,0.001
Model 4 1.00 1.15 (1.08, 1.22) 1.45 (1.19, 1.78) ,0.001
HPFS (1986–2010)
Total fried food
Cases/person-years 1253/372,858 1092/261,460 751/155,888 253/45,484 —
Age-adjusted model 1.00 1.30 (1.20, 1.41) 1.57 (1.43, 1.72) 1.88 (1.64, 2.15) ,0.001
Model 1 1.00 1.28 (1.18, 1.38) 1.49 (1.36, 1.64) 1.76 (1.53, 2.02) ,0.001
Model 2 1.00 1.25 (1.15, 1.36) 1.45 (1.32, 1.59) 1.69 (1.47, 1.95) ,0.001
Model 3 1.00 1.17 (1.08, 1.28) 1.28 (1.16, 1.41) 1.46 (1.27, 1.69) ,0.001
Model 4 1.00 1.12 (1.03, 1.21) 1.16 (1.06, 1.28) 1.29 (1.12, 1.49) ,0.001
Fried food at home
Cases/person-years 1830/501,598 1289/291,410 230/42,682 —
Age-adjusted model 1.00 1.25 (1.16, 1.34) 1.54 (1.34, 1.77) ,0.001
Model 1 1.00 1.22 (1.14, 1.31) 1.46 (1.27, 1.67) ,0.001
Model 2 1.00 1.18 (1.10, 1.27) 1.38 (1.19, 1.59) ,0.001
Model 3 1.00 1.12 (1.04, 1.20) 1.29 (1.12, 1.48) ,0.001
Model 4 1.00 1.07 (0.99, 1.15) 1.17 (1.02, 1.35) 0.02
Fried food away from home
Cases/person-years 1932/536,520 1207/263,463 210/35,707 —
Age-adjusted model 1.00 1.39 (1.30, 1.50) 1.93 (1.67, 2.23) ,0.001
Model 1 1.00 1.35 (1.26, 1.45) 1.78 (1.54, 2.06) ,0.001
Model 2 1.00 1.32 (1.22, 1.42) 1.71 (1.48, 1.98) ,0.001
Model 3 1.00 1.19 (1.10, 1.28) 1.47 (1.27, 1.70) ,0.001
Model 4 1.00 1.11 (1.03, 1.20) 1.33 (1.15, 1.54) ,0.001
Pooled results
Total fried food
Age-adjusted model 1.00 1.25 (1.16, 1.34) 1.67 (1.49, 1.87) 2.00 (1.77, 2.26) ,0.001
Model 1 1.00 1.20 (1.08, 1.34) 1.53 (1.44, 1.62) 1.73 (1.57, 1.90) ,0.001
Model 2 1.00 1.15 (0.97, 1.35) 1.39 (1.30, 1.49) 1.55 (1.32, 1.83) ,0.001
Model 3 1.00 1.11 (0.99, 1.23) 1.24 (1.17, 1.32) 1.34 (1.13, 1.59) ,0.001
Model 4 1.00 1.06 (0.97, 1.16) 1.13 (1.07, 1.20) 1.19 (1.00, 1.40) ,0.001
Fried food at home
Age-adjusted model 1.00 1.20 (1.11, 1.29) 1.61 (1.49, 1.75) ,0.001
Model 1 1.00 1.16 (1.05, 1.28) 1.42 (1.31, 1.54) ,0.001
Model 2 1.00 1.09 (0.93, 1.27) 1.26 (1.09, 1.47) 0.03
Model 3 1.00 1.05 (0.95, 1.17) 1.16 (0.96, 1.40) 0.17
Model 4 1.00 1.02 (0.94, 1.10) 1.08 (0.93, 1.25) 0.37
(Continued)
672 CAHILL ET AL

TABLE 2 (Continued )

Frequency of fried-food consumption

,1 time/wk 1–3 times/wk 4–6 times/wk $7 times/wk P-trend

Fried food away from home

Age-adjusted model 1.00 1.55 (1.26, 1.90) 2.24 (1.65, 3.04) ,0.001
Model 1 1.00 1.44 (1.27, 1.64) 1.97 (1.59, 2.44) ,0.001
Model 2 1.00 1.37 (1.28, 1.46) 1.81 (1.58, 2.08) ,0.001
Model 3 1.00 1.24 (1.15, 1.33) 1.55 (1.36, 1.76) ,0.001
Model 4 1.00 1.13 (1.08, 1.19) 1.37 (1.22, 1.54) ,0.001
1
Time-dependent Cox proportional hazards models conditioned on age and follow-up cycle were used for both cohorts individually and then pooled
together in an inverse-variance–weighted meta-analysis by using a random-effects model. Model 1 was adjusted for age (y), race (white or nonwhite), family
history of diabetes (yes or no), smoking status [never, past, current (1–14, 15–24, or $25 cigarettes/d)], alcohol intake (0, 0.1–4.9, 5.0–14.9, or $15.0 g/d in
women; 0, 0.1–4.9, 5.0–29.9, or $30.0 g/d in men), physical activity (,3.0, 3.0–8.9, 9.0–17.9, 18.0–26.9, or $27.0 metabolic equivalent task-hours per
week), postmenopausal status, and menopausal hormone use [premenopausal or postmenopausal (never, past, or current hormone use); in women only]. Model
2 was adjusted as for model 1 and for total energy intake and the Alternative Healthy Eating Index (quintiles). Model 3 was adjusted as for model 2 and for
hypertension and hypercholesterolemia. Model 4 was adjusted as for model 3 and for BMI (in kg/m2; ,23.0, 23.0–24.9, 25.0–29.9, 30.0–34.9, or $35.0).
HPFS, Health Professionals Follow-Up Study; NHS, Nurses’ Health Study.

Downloaded from ajcn.nutrition.org by guest on October 27, 2015

reported that fried-food consumption was not associated with
CAD in 40,757 adults during 8–12 y follow-up; the authors
suggested that this was because oils used for frying in Spain are
mainly olive oil and sunflower oil (11). In a case-control study
from Costa Rica, where frying food is a daily practice, Kabagambe
et al (10) reported no association between the consumption of
fried foods and risk of nonfatal acute MI, which was potentially
attributable to the absence of a large reference group of in-
dividuals who did not fry their foods regularly. The results of the
current analysis are consistent with the INTERHEART study
(5761 nonfatal MI cases and 10,646 controls from 52 countries),
which observed an OR of 1.13 (95% CI: 1.02, 1.25) for highest
compared with lowest quartiles of fried-food intake after mul-
tivariate adjustment (12).
In our analyses, relations of fried-food consumption with risks
of T2D and CAD were substantially attenuated with adjustment
for BMI, hypertension, and hypercholesterolemia. Several studies
have reported that fried-food consumption is associated with
weight gain, including a combined study of the NHS and HPFS
(35). Multiple cross-sectional studies have also linked the con-
sumption of fried foods to an increased likelihood of cardiometabolic
risk factors such as body weight and obesity (8), hypertension (6), and
low serum HDL cholesterol (7).
Although the process of frying food is complex and not well
understood, frying is known to alter the quality and energy
density of food (2) while also often improving the palatability of
food through changes to texture and color (such as making the
food crunchy, crisp, and golden brown). Through polymerization,
oxidation, and hydrogenation, frying modifies both foods and
frying mediums, which leads to an increase in the absorption of
oil-degradation products by foods being fried and also a loss of
unsaturated fatty acids such as linoleic and linolenic acids and an
increase in corresponding trans fatty acids (36). It has been well
documented that small changes in the proportion of essential
fatty acids in the diet can have important health impacts (37).
The amount that frying adulterates fats depends on the frying
technique (deep or pan frying), extent of oil degradation, type
and composition of the food (38), and type of oil used (39). For
example, olive oil is less prone to oxidation than other oils are
(39), and has been reported to impart cardiometabolic health
benefits when used for frying (5). However, in general, frying
increases amounts of cholesterol oxidation products (40) and
reduces the activity of paraoxonase (41), which is an enzyme
that inhibits the oxidation of LDL cholesterol. The consumption
of foods fried with reused oils has been associated with a higher
prevalence of arterial hypertension (6) and impaired arterial
endothelial function (42) compared with when the same oil was
previously unused.
There were several limitations in our analysis. First, we do not
know what specific fried foods our participants ate. We were also
limited in knowing the duration, temperature, and method (deep
or pan) used for frying and how often oils had been reused.
Certain less obviously fried foods, such as doughnuts, may not
have been considered fried foods by some participants. Further-
more, it is possible that persons with hypertension, high cho-
lesterol, or obesity may have been less likely to consume or report
the consumption of fried foods. Similarly, energy intakes reported
by the FFQ may have been underestimated (43), which would
have affected the adjustment for energy in our analysis. Our study
had several strengths, including a large sample size and the re-
peated comprehensive assessment of many lifestyle character-
istics that were gathered prospectively with a long duration of
follow-up. As such, we were able to assess the confounding effects
and modification of the association for fried-food consumption by
other dietary components or lifestyle factors. Nevertheless, it was
still possible that we were unable to account for all unhealthy
lifestyle characteristics associated with greater fried-food con-
sumption. The time period of our study (1984–2010) fit well with
changes to policy and subsequent practice regarding trans fats in
restaurants. The fast-food industry switched from beef tallow to
trans-fat rich vegetable oils for deep-fat frying after public de-
mand in 1985, and it was not until 2008 that many American
restaurants stopped the use of trans fats. Even though fried-food
questions were asked repeatedly and consistently every 4 y, we
could not exclude the possibility of unmeasured confounding.
However, the relatively homogenous study population could
have reduced residual confounding because of the unmeasured
socioeconomic variability. Because our participants were pre-
dominantly non-Hispanic white health professionals, the gen-
eralizability of observed associations may be limited to populations
 FRIED FOOD, DIABETES, AND CORONARY ARTERY DISEASE 673
TABLE 3
RRs (95% CIs) for coronary heart disease in the NHS and the HPFS according to the frequency of fried-food consumption1
Frequency of fried-food consumption

,1 time/wk 1–3 times/wk 4–6 times/wk $7 times/wk P-trend

NHS (1984–2010)
Total fried food
Cases/person-years 1430/942,425 861/545,997 316/180,120 80/37,259 —
Age-adjusted model 1.00 1.11 (1.02, 1.21) 1.40 (1.24, 1.58) 1.69 (1.35, 2.12) ,0.001
Model 1 1.00 1.07 (0.98, 1.16) 1.28 (1.13, 1.45) 1.48 (1.18, 1.86) ,0.001
Model 2 1.00 1.02 (0.93, 1.11) 1.17 (1.03, 1.33) 1.33 (1.06, 1.68) 0.02
Model 3 1.00 1.01 (0.93, 1.10) 1.10 (0.97, 1.25) 1.20 (0.95, 1.51) 0.05
Model 4 1.00 1.01 (0.92, 1.10) 1.09 (0.96, 1.24) 1.18 (0.93, 1.48) 0.08
Fried food at home
Cases/person-years 1621/1,059,906 918/575,006 148/70,889 —
Age-adjusted model 1.00 1.10 (1.02, 1.20) 1.44 (1.21, 1.70) ,0.001
Model 1 1.00 1.07 (0.98, 1.16) 1.30 (1.10, 1.54) 0.002
Model 2 1.00 1.01 (0.93, 1.10) 1.17 (0.98, 1.39) 0.14
Model 3 1.00 1.00 (0.92, 1.09) 1.08 (0.91, 1.29) 0.45
Model 4 1.00 1.00 (0.92, 1.09) 1.07 (0.90, 1.28) 0.52
Fried food away from home

Downloaded from ajcn.nutrition.org by guest on October 27, 2015

Cases/person-years 2210/1,422,207 452/271,536 25/12,058 —
Age-adjusted model 1.00 1.35 (1.22, 1.49) 1.71 (1.15, 2.54) ,0.001
Model 1 1.00 1.25 (1.13, 1.38) 1.44 (0.97, 2.14) ,0.001
Model 2 1.00 1.19 (1.07, 1.32) 1.34 (0.90, 1.99) 0.001
Model 3 1.00 1.12 (1.01, 1.24) 1.19 (0.80, 1.76) 0.04
Model 4 1.00 1.11 (1.00, 1.23) 1.16 (0.78, 1.72) 0.08
HPFS (1986-2010)
Total fried food
Cases/person-years 1287/380,284 988/268,648 643/161,080 173/47,439 —
Age-adjusted model 1.00 1.17 (1.08, 1.28) 1.40 (1.27, 1.54) 1.35 (1.15, 1.58) ,0.001
Model 1 1.00 1.14 (1.05, 1.24) 1.34 (1.21, 1.47) 1.25 (1.06, 1.46) ,0.001
Model 2 1.00 1.11 (1.02, 1.20) 1.27 (1.15, 1.40) 1.16 (0.98, 1.36) ,0.001
Model 3 1.00 1.06 (0.97, 1.16) 1.18 (1.06, 1.30) 1.06 (0.90, 1.25) 0.01
Model 4 1.00 1.05 (0.97, 1.15) 1.16 (1.05, 1.28) 1.04 (0.88, 1.22) 0.03
Fried food at home
Cases/person-years 1709/513,408 1180/299,716 202/44,327 —
Age-adjusted model 1.00 1.24 (1.15, 1.34) 1.38 (1.19, 1.60) ,0.001
Model 1 1.00 1.21 (1.12, 1.30) 1.30 (1.12, 1.51) ,0.001
Model 2 1.00 1.16 (1.08, 1.25) 1.21 (1.04, 1.41) ,0.001
Model 3 1.00 1.12 (1.04, 1.21) 1.15 (1.00, 1.35) 0.01
Model 4 1.00 1.11 (1.03, 1.20) 1.14 (0.98, 1.33) 0.02
Fried food away from home
Cases/person-years 2006/548,049 963/271,895 122/37,507 —
Age-adjusted model 1.00 1.16 (1.07, 1.25) 1.32 (1.10, 1.58) ,0.001
Model 1 1.00 1.12 (1.04, 1.21) 1.21 (1.01, 1.46) 0.002
Model 2 1.00 1.08 (1.00, 1.17) 1.15 (0.95, 1.38) 0.04
Model 3 1.00 1.02 (0.94, 1.10) 1.05 (0.87, 1.26) 0.60
Model 4 1.00 1.01 (0.93, 1.09) 1.03 (0.85, 1.24) 0.83
Pooled results
Total fried food
Age-adjusted model 1.00 1.14 (1.08, 1.21) 1.40 (1.30, 1.51) 1.49 (1.20, 1.85) ,0.001
Model 1 1.00 1.11 (1.03, 1.18) 1.31 (1.22, 1.42) 1.33 (1.13, 1.58) ,0.001
Model 2 1.00 1.06 (0.98, 1.15) 1.23 (1.14, 1.33) 1.21 (1.06, 1.39) ,0.001
Model 3 1.00 1.04 (0.97, 1.10) 1.15 (1.06, 1.24) 1.11 (0.97, 1.26) 0.001
Model 4 1.00 1.03 (0.97, 1.09) 1.13 (1.04, 1.22) 1.08 (0.95, 1.24) 0.006
Fried food at home
Age-adjusted model 1.00 1.17 (1.04, 1.32) 1.41 (1.26, 1.57) ,0.001
Model 1 1.00 1.14 (1.00, 1.29) 1.26 (1.13, 1.41) ,0.001
Model 2 1.00 1.08 (0.94, 1.24) 1.19 (1.07, 1.34) 0.002
Model 3 1.00 1.06 (0.95, 1.18) 1.13 (1.00, 1.26) 0.04
Model 4 1.00 1.06 (0.95, 1.17) 1.11 (0.99, 1.24) 0.04
(Continued)
674 CAHILL ET AL

TABLE 3 (Continued )

Frequency of fried-food consumption

,1 time/wk 1–3 times/wk 4–6 times/wk $7 times/wk P-trend

Fried food away from home

Age-adjusted model 1.00 1.25 (1.08, 1.44) 1.41 (1.12, 1.78) 0.005
Model 1 1.00 1.18 (1.06, 1.31) 1.25 (1.06, 1.48) 0.01
Model 2 1.00 1.13 (1.03, 1.23) 1.18 (0.99, 1.39) 0.03
Model 3 1.00 1.06 (0.97, 1.17) 1.07 (0.91, 1.27) 0.22
Model 4 1.00 1.05 (0.95, 1.15) 1.05 (0.89, 1.25) 0.33
1
Time-dependent Cox proportional hazards models conditioned on age and follow-up cycle were used for both cohorts individually and then pooled
together in an inverse-variance–weighted meta-analysis by using a random-effects model. Model 1 was adjusted for age (y), race (white or nonwhite), parental
myocardial infarction before the age of 60 y (yes or no), smoking status [never, past, or current (1–14, 15–24, or $25 cigarettes/d)], alcohol intake (0, 0.1–4.9,
5.0–14.9, or $15.0 g/d in women; 0, 0.1–4.9, 5.0–29.9, or $30.0 g/d in men), physical activity (,3, 3.0–8.9, 9.0–17.9, 18.0–26.9, or $27.0 metabolic
equivalent task hours per week), postmenopausal status, and menopausal hormone use [premenopausal or postmenopausal (never, past, or current hormone
use); in women only]. Model 2 was adjusted as for model 1 and for total energy intake and the Alternative Healthy Eating Index (quintiles). Model 3 was
adjusted as for model 2 and for hypertension and hypercholesterolemia. Model 4 was adjusted as for model 3 and for BMI (in kg/m2; ,23.0, 23.0–24.9, 25.0–
29.9, 30.0–34.9, or $35.0). HPFS, Health Professionals Follow-Up Study; NHS, Nurses’ Health Study.

of similar characteristics. Future studies with detailed information 4. Powell LM, Nguyen BT, Han E. Energy intake from restaurants: de-

Downloaded from ajcn.nutrition.org by guest on October 27, 2015

on amounts and portion sizes of fried foods as well as types of
oils used, times and temperatures used for frying, types of frying
(deep fried compared with pan fried), and the degree to which
oils are reused are warranted to confirm our findings in other
populations.
In conclusion, in 2 large, prospective cohorts, we observed that
frequent fried-food consumption was significantly associated
with risk of incident T2D and CAD. These associations were
mediated in part by BMI, hypertension, and hypercholesterol-
emia. The findings lend support to the large body of evidence
connecting the ubiquity of Western-style fast-food intake to the
global T2D and CAD epidemic. Additional studies are required
to confirm our findings and elucidate whether the strong asso-
ciations we observed between fried-food consumption and risk of
T2D and CAD are attributable to habitual fried-food consump-
tion having a causal role in the development of T2D and CAD or
rather being a marker of an unhealthy lifestyle. Regardless, from
a clinical and public health perspective, we have identified a risk
factor for T2D and CAD that may be readily modifiable by lifestyle
or cooking choices that lead to the consumption of less fried foods,
especially those foods consumed away from home.
We are indebted to participants in the NHS and HPFS for their continuing
outstanding support and the staff working in these studies for their valuable
help.
The authors’ responsibilities were as follows—LEC and AP: analyzed
data, wrote the manuscript, and had primary responsibility for the final content
of the manuscript; and all authors: provided critical revisions to the manuscript
for important intellectual content, satisfied the authorship criteria of the In-
ternational Committee of Medical Journal Editors, designed and conducted the
research, and read and approved the final manuscript. None of the authors
reported any conflicts of interest.
REFERENCES
1. Choe E, Min DB. Chemistry of deep-fat frying oils. J Food Sci 2007;
72:R77–86.
2. Fillion L, Henry CJ. Nutrient losses and gains during frying: a review.
Int J Food Sci Nutr 1998;49:157–68.
3. Morse KL, Driskell JA. Observed sex differences in fast-food con-
sumption and nutrition self-assessments and beliefs of college students.
Nutr Res 2009;29:173–9.
mographics and socioeconomics, 2003-2008. Am J Prev Med 2012;43:
498–504.
5. Farnetti S, Malandrino N, Luciani D, Gasbarrini G, Capristo E. Food
fried in extra-virgin olive oil improves postprandial insulin response in
obese, insulin-resistant women. J Med Food 2011;14:316–21.
6. Soriguer F, Rojo-Martinez G, Dobarganes MC, Garcia Almeida JM,
Esteva I, Beltran M, Ruiz De Adana MS, Tinahones F, Gomez-Zumaquero
JM, Garcia-Fuentes E, et al. Hypertension is related to the degradation
of dietary frying oils. Am J Clin Nutr 2003;78:1092–7.
7. Donfrancesco C, Lo Noce C, Brignoli O, Riccardi G, Ciccarelli P,
Dima F, Palmieri L, Giampaoli S. Italian network for obesity and cardio-
vascular disease surveillance: a pilot project. BMC Fam Pract 2008;
9:53.
8. Guallar-Castillón P, Rodriguez-Artalejo F, Fornes NS, Banegas JR,
Etxezarreta PA, Ardanaz E, Barricarte A, Chirlaque MD, Iraeta MD,
Larranaga NL, et al. Intake of fried foods is associated with obesity in
the cohort of Spanish adults from the European Prospective Investigation
into Cancer and Nutrition. Am J Clin Nutr 2007;86:198–205.
9. Alhazmi A, Stojanovski E, McEvoy M, Garg ML. The association
between dietary patterns and type 2 diabetes: a systematic review and
meta-analysis of cohort studies. J Hum Nutr Diet (Epub ahead of print
16 September 2013).
10. Kabagambe EK, Baylin A, Siles X, Campos H. Individual saturated
fatty acids and nonfatal acute myocardial infarction in Costa Rica. Eur
J Clin Nutr 2003;57:1447–57.
11. Guallar-Castillón P, Rodriguez-Artalejo F, Lopez-Garcia E, Leon-
Munoz LM, Amiano P, Ardanaz E, Arriola L, Barricarte A, Buckland
G, Chirlaque MD, et al. Consumption of fried foods and risk of cor-
onary heart disease: Spanish cohort of the European Prospective In-
vestigation into Cancer and Nutrition study. BMJ 2012;344:e363.
12. Iqbal R, Anand S, Ounpuu S, Islam S, Zhang X, Rangarajan S, Chifamba
J, Al-Hinai A, Keltai M, Yusuf S; INTERHEART Study Investigators.
Dietary patterns and the risk of acute myocardial infarction in 52
countries: results of the INTERHEART study. Circulation 2008;118:
1929–37.
13. Pan A, Sun Q, Bernstein AM, Schulze MB, Manson JE, Stampfer MJ,
Willett WC, Hu FB. Red meat consumption and mortality: results from
2 prospective cohort studies. Arch Intern Med 2012;172:555–63.
14. Chiuve SE, Fung TT, Rimm EB, Hu FB, McCullough ML, Wang M,
Stampfer MJ, Willett WC. Alternative dietary indices both strongly
predict risk of chronic disease. J Nutr 2012;142:1009–18.
15. Classification and diagnosis of diabetes mellitus and other categories of
glucose intolerance. National Diabetes Data Group. Diabetes 1979;28:
1039–57.
16. Report of the Expert Committee on the Diagnosis and Classification of
Diabetes Mellitus. Diabetes Care 1997;20:1183–97.
17. Hu FB, Leitzmann MF, Stampfer MJ, Colditz GA, Willett WC, Rimm
EB. Physical activity and television watching in relation to risk for type
2 diabetes mellitus in men. Arch Intern Med 2001;161:1542–8.
 FRIED FOOD, DIABETES, AND CORONARY ARTERY DISEASE
18. Manson JE, Rimm EB, Stampfer MJ, Colditz GA, Willett WC, Krolewski
AS, Rosner B, Hennekens CH, Speizer FE. Physical activity and in-
cidence of non-insulin-dependent diabetes mellitus in women. Lancet
1991;338:774–8.
19. Rose GA, Blackburn H. Cardiovascular survey methods. 2nd ed. Mono-
graph series no 58. Geneva, Switzerland; World Health Organization, 1982.
20. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin
Trials 1986;7:177–88.
21. Ainsworth BE, Haskell WL, Leon AS, Jacobs DR Jr, Montoye HJ,
Sallis JF, Paffenbarger RS Jr. Compendium of physical activities:
classification of energy costs of human physical activities. Med Sci
Sports Exerc 1993;25:71–80.
22. Ainsworth BE, Haskell WL, Whitt MC, Irwin ML, Swartz AM, Strath
SJ, O’Brien WL, Bassett DR Jr, Schmitz KH, Emplaincourt PO, et al.
Compendium of physical activities: an update of activity codes and
MET intensities. Med Sci Sports Exerc 2000;32:S498–504.
23. Hu FB, Stampfer MJ, Rimm E, Ascherio A, Rosner BA, Spiegelman D,
Willett WC. Dietary fat and coronary heart disease: a comparison of
approaches for adjusting for total energy intake and modeling repeated
dietary measurements. Am J Epidemiol 1999;149:531–40.
24. Liu S, Serdula M, Janket SJ, Cook NR, Sesso HD, Willett WC, Manson
JE, Buring JE. A prospective study of fruit and vegetable intake and the
risk of type 2 diabetes in women. Diabetes Care 2004;27:2993–6.
25. Ylönen SK, Virtanen SM, Groop L; Botnia Research Group. The intake
of potatoes and glucose metabolism in subjects at high risk for Type 2
diabetes. Diabet Med 2007;24:1049–50.
26. Khosravi-Boroujeni H, Mohammadifard N, Sarrafzadegan N, Sajjadi F,
Maghroun M, Khosravi A, Alikhasi H, Rafieian M, Azadbakht L. Potato
consumption and cardiovascular disease risk factors among Iranian pop-
ulation. Int J Food Sci Nutr 2012;63:913–20.
27. Pan A, Sun Q, Bernstein AM, Schulze MB, Manson JE, Willett WC,
Hu FB. Red meat consumption and risk of type 2 diabetes: 3 cohorts of US
adults and an updated meta-analysis. Am J Clin Nutr 2011;94:1088–96.
28. Fung TT, Schulze M, Manson JE, Willett WC, Hu FB. Dietary patterns,
meat intake, and the risk of type 2 diabetes in women. Arch Intern Med
2004;164:2235–40.
29. Halton TL, Willett WC, Liu S, Manson JE, Stampfer MJ, Hu FB.
Potato and french fry consumption and risk of type 2 diabetes in
women. Am J Clin Nutr 2006;83:284–90.
30. Krishnan S, Coogan PF, Boggs DA, Rosenberg L, Palmer JR. Con-
sumption of restaurant foods and incidence of type 2 diabetes in Af-
rican American women. Am J Clin Nutr 2010;91:465–71.
31. Odegaard AO, Koh WP, Yuan JM, Gross MD, Pereira MA. Western-
style fast food intake and cardiometabolic risk in an Eastern country.
Circulation 2012;126:182–8.
View publication stats
675
32. Pereira MA, Kartashov AI, Ebbeling CB, Van Horn L, Slattery ML,
Jacobs DR Jr, Ludwig DS. Fast-food habits, weight gain, and insulin
resistance (the CARDIA study): 15-year prospective analysis. Lancet
2005;365:36–42.
33. Young LR, Nestle M. Expanding portion sizes in the US marketplace:
implications for nutrition counseling. J Am Diet Assoc 2003;103:231–4.
34. Centers for Disease Control and Prevention (CDC). Vital signs: food
categories contributing the most to sodium consumption - United
States, 2007-2008. MMWR Morb Mortal Wkly Rep 2012;61:92–8.
35. 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;364:2392–404.
36. Li A, Ha Y, Wang F, Li W, Li Q. Determination of thermally induced
trans-fatty acids in soybean oil by attenuated total reflectance fourier
transform infrared spectroscopy and gas chromatography analysis.
J Agric Food Chem 2012;60:10709–13.
37. Panel on Macronutrients, Panel on the Definition of Dietary Fiber,
Subcommittee on Upper Reference Levels of Nutrients, Subcommittee
on Interpretation and Uses of Dietary Reference Intakes, and the
Standing Committee on the Scientific Evaluation of Dietary Reference
Intakes. Dietary Reference Intakes for energy, carbohydrate. fiber, fat,
fatty acids, cholesterol, protein, and amino acids (2002/2005). Washington,
DC; Food and Nutrition Board, Institute of Medicine of the National
Academies.
38. Ansorena D, Guembe A, Mendizabal T, Astiasaran I. Effect of fish and
Downloaded from ajcn.nutrition.org by guest on October 27, 2015
oil nature on frying process and nutritional product quality. J Food Sci
2010;75:H62–7.
39. Casal S, Malheiro R, Sendas A, Oliveira BP, Pereira JA. Olive oil
stability under deep-frying conditions. Food Chem Toxicol 2010;48:
2972–9.
40. Echarte M, Ansorena D, Astiasaran I. Fatty acid modifications and
cholesterol oxidation in pork loin during frying at different tempera-
tures. J Food Prot 2001;64:1062–6.
41. Sutherland WH, Walker RJ, de Jong SA, van Rij AM, Phillips V,
Walker HL. Reduced postprandial serum paraoxonase activity after
a meal rich in used cooking fat. Arterioscler Thromb Vasc Biol 1999;
19:1340–7.
42. Williams MJ, Sutherland WH, McCormick MP, de Jong SA, Walker
RJ, Wilkins GT. Impaired endothelial function following a meal rich in
used cooking fat. J Am Coll Cardiol 1999;33:1050–5.
43. Subar AF, Thompson FE, Kipnis V, Midthune D, Hurwitz P, McNutt
S, McIntosh A, Rosenfeld S. Comparative validation of the Block,
Willett, and National Cancer Institute food frequency questionnaires:
the Eating at America’s Table Study. Am J Epidemiol 2001;154:
1089–99.