nutrients

Article
Changes in Food Cravings and Eating Behavior after a
Dietary Carbohydrate Restriction Intervention Trial
Katherene O.-B. Anguah 1, *, Majid M. Syed-Abdul 1 , Qiong Hu 1,2 , Miriam Jacome-Sosa 3 ,
Colette Heimowitz 4 , Vicki Cox 5 and Elizabeth J. Parks 1,2
1 Department of Nutrition and Exercise Physiology, University of Missouri, Columbia, MO 65211, USA;
ms9rf@mail.missouri.edu (M.M.S.-A.); huqiong313@gmail.com (Q.H.); parksej@missouri.edu (E.J.P.)
2 Department of Child Health, School of Medicine, University of Missouri, Columbia, MO 65212, USA
3 Department of Internal Medicine, Division of Nutritional Science, Washington University School of
Medicine, St. Louis, MO 63110, USA; mjacome@wustl.edu
4 Atkins Nutritionals, Inc., Denver, CO 80202, USA; Cheimowitz@atkins.com
5 Department of Nutrition, West Chester University, West Chester, PA 19383, USA; vickicuce@gmail.com
* Correspondence: anguahk@missouri.edu; Tel.: +1-573-882-8966




Received: 18 November 2019; Accepted: 19 December 2019; Published: 24 December 2019


Abstract: Compared to low-fat diets, low-carbohydrate (CHO) diets cause weight loss (WL) over
a faster time frame; however, it is unknown how changes in food cravings and eating behavior
contribute to this more rapid WL in the early phases of dieting. We hypothesized that reductions
in food cravings and improved eating behaviors would be evident even after a relatively short
(4-week) duration of CHO-restriction, and that these changes would be associated with WL. Adult
participants (n = 19, 53% males, mean ± SD: BMI = 34.1 ± 0.8 kg/m2 ; age 40.6 ± 1.9 years) consumed a
CHO-restricted diet (14% CHO, 58% fat, 28% protein) for 4 weeks. Before and after the intervention,
specific and total cravings were measured with the Food Craving Inventory (FCI) and eating behaviors
assessed with the Three-Factor Eating questionnaire. Food cravings were significantly reduced at
week 4, while women had significantly greater reductions in sweet cravings than men. Dietary
restraint was significantly increased by 102%, while disinhibiton and hunger scores were reduced
(17% and 22%, respectively, p < 0.05). Changes in cravings were unrelated to changes in body weight
except for the change in high-fat cravings where those who lost the most weight experienced the
least reductions in fat cravings (r = −0.458, p = 0.049). Changes in dietary restraint were inversely
related to several FCI subscales. A short-term, low-CHO diet was effective in reducing food cravings.
These data suggest that in subjects that have successfully lost weight on a low-CHO diet, those who
craved high-fat foods at the onset were able to satisfy their cravings—potentially due to the high-fat
nature of this restricted diet.

Keywords: carbohydrate restriction; food cravings; eating behaviors; weight loss

1. Introduction
About 11% of the variance in eating behavior and weight gain can be explained by the experience
of food cravings [1], which is generally defined as an intense desire to consume a particular kind of
food that is difficult to resist [2,3]. Cravings differ from hunger in the intensity and specificity for the
food craved for. Many overweight patients interested in weight loss (WL) may assume that dieting
may increase their food cravings. Although a few WL studies suggest cravings may not change [4]
or even increase [5,6], a large number of clinical trials, most from 12 weeks to 2 years in duration,
have demonstrated that cravings are reduced during energy restriction [5,7–18]. Some authors have
reported that greater WL is associated with greater reductions in food cravings [11–13], while others

Nutrients 2020, 12, 52; doi:10.3390/nu12010052 www.mdpi.com/journal/nutrients

Nutrients 2020, 12, 52 2 of 12

have found no relationships [5,19]. With regard to dietary macronutrient content, of the three studies
investigating low-CHO diet effects on food cravings, two reported reduced cravings [11,15], while
one [5] reported increased cravings. It is well recognized that low-CHO diets cause greater WL in the
initial four weeks after adoption [20,21] and the duration of most past studies ranged from 12 weeks to
two years. Only one study compared the effects of a low-fat and low-CHO diet on cravings for six
weeks and found increased cravings on the low-CHO diet [6]. The tool or scale used to measure food
cravings was not well defined in that paper and the sample size was small (n = 4). Thus, it is unknown
whether food cravings change in the first four weeks of CHO restriction, a time when the WL trajectory
is the steepest [20,21].
Very few studies have specifically investigated sex-related differences in cravings during WL.
Anton and colleagues [9] found that compared to women, men reported stronger cravings for high-fat
foods and fast-food fats at baseline and throughout the WL intervention, while women reported greater
cravings for sweets and fruits/vegetables at baseline and throughout the intervention. Dorling et al. [22]
also reported reductions in cravings for carbohydrates and fats at 24 months in males but not in females
in a calorie-restricted group compared to an ad libitum intake group. Thus the little data available
suggest sex-related differences in the types of foods craved for and this notion is supported by a 2016
review [23] that aimed to emphasize the relevance of investigating gender differences in food cravings.
Aside from food cravings, other eating behaviors affect WL success. The Three Factor Eating
questionnaire (TFEQ) is a commonly used tool to assess eating behaviors termed dietary restraint,
disinhibition, and hunger [24]. Higher baseline disinhibition was shown to be positively associated
with baseline BMI and less WL success [25], while greater increases in dietary restraint were associated
with greater WL [26–28] in obese subjects undergoing WL treatment. With regard to TFEQ hunger,
in a recent study where a low-energy diet was fed for eight weeks, a one-unit higher hunger score
at baseline and a one-unit reduction in scores of both disinhibition and hunger were associated with
larger WL [29].
Given the effect of low-CHO diets to lead to relatively fast WL (within the first weeks of energy
restriction), the purpose of the present analysis was to test the effect of a CHO-restricted diet (14% of
energy) on self-reported food cravings and eating behaviors over four weeks. Furthermore, we also
examined if there were any sex differences in food cravings and eating behavior. We hypothesized that
with the exception of fruits/vegetables, all the FCI subscales would be reduced and that reductions in
cravings would be correlated with WL, eating behaviors, and changes in blood glucose even during
the early stages of WL. Furthermore, we hypothesized that CHO-restriction would result in increased
restraint and decreased disinbition and hunger. The data on cravings and eating behaviors, analyzed
here, were collected as part of a larger study designed to test the effect of the WL intervention on
cardiovascular risk factors associated with metabolic syndrome. The results of that study have been
previously published [30].

2. Materials and Methods

2.1. Study Participants and Design

Detailed descriptions of the participants and the study design for the parent study have been
published previously [30]. The goal of the parent study was to determine whether four weeks of dietary
CHO-restriction would reduce vascular stiffness assessed by aortic pulse wave velocity. The present
paper focusses on results obtained from measures of cravings and eating behaviors assessed during
this trial. The overall study was approved by the University of Missouri Health Science Institutional
Review Board (Protocol #2004733) and written informed consent was obtained from all participants.
This trial was registered on ClinicalTrials.gov Identifier: NCT00427193. Men and premenopausal
women were recruited who were overweight or obese (BMI: 27–40 kg/m2 ), aged 30–55 years, and had
characteristics of insulin resistance [31] or the metabolic syndrome [30]. With regard to physical activity,
subjects had low to medium levels (5000 to 10,000 steps/day). Exclusion criteria included having
Nutrients 2020, 12, 52 3 of 12

diabetes, alcohol use of more than 5 standard drinks/week (70 g) for women and more than 10 standard
drinks/week (140 g) for men, use of tobacco products, taking prescription medication for clinically
significant
Nutrientsendocrine,
2020, 12, x FORgastrointestinal,
PEER REVIEW cardiovascular, hematological, hepatic, renal, respiratory, 3 of 12 or
genitourinary abnormalities or diseases, or being on any dietary regimen that would hinder adherence
to themedication
low-CHO for diet.clinically significant endocrine, gastrointestinal, cardiovascular, hematological,
hepatic, renal,
As shown in Figure respiratory, or genitourinary
1, 64 participants wereabnormalities
assessed for or diseases,and,
eligibility or being
as perontheany dietarystudy
original
regimen that would hinder adherence to the low-CHO diet.
design’s plan, 20 were assigned the CHO-restricted intervention [30]. To compare cravings with eating
As shown in Figure 1, 64 participants were assessed for eligibility and, as per the original study
behaviors, WL, and biochemical outcomes (insulin and glucose concentrations), complete data were
design’s plan, 20 were assigned the CHO-restricted intervention [30]. To compare cravings with
available for 19 of the original 20 subjects for food cravings and 18 out of the original 20 subjects for
eating behaviors, WL, and biochemical outcomes (insulin and glucose concentrations), complete data
eatingwere
behaviors,
availableTFEQ. The
for 19 of thelow-CHO
original 20diet provided
subjects for food1500 kcal/day
cravings and 18of out
energy
of theand included
original 20–25 g of
20 subjects
net CHO/day
for eating [30]. Subjects
behaviors, TFEQ.wereTheprovided
low-CHOall dietfood for the
provided first
1500 two weeks
kcal/day of the
of energy andstudy, during
included 20–25which
time they received
g of net CHO/day comprehensive
[30]. Subjects education
were providedto beallable
food tofor
cook
theand
first prepare
two weekstheir ownstudy,
of the low-CHO
duringmeals
during weeks
which time3 and
they 4. Thus, comprehensive
received for weeks 3 and 4, subjects
education consumed
to be able to cooka diet
andsimilar
preparetotheir
thatown
eaten during
low-
CHO
the first twomeals
weeksduringby weeks
following3 andthe
4. Thus, for weeks
guidelines and3 using
and 4, the
subjects consumedmaterials
educational a diet similar to that
given them by
studyeaten during the first
investigators. Studytwostaff
weeks
wereby in
following
contactthe guidelines
with researchand using the educational
participants materials
on a near-daily basis to
given them by study investigators. Study staff were in contact with research
monitor dietary intake and assess food acceptance. Dietary adherence was assessed by inspection of participants on a near-
daily basis to monitor dietary intake and assess food acceptance. Dietary adherence was assessed by
returned food containers (first two weeks), respiratory quotient, and measurement of plasma ketones
inspection of returned food containers (first two weeks), respiratory quotient, and measurement of
during week 4 [30]. Body weight was measured to the nearest 0.1 kg and fasting glucose and insulin
plasma ketones during week 4 [30]. Body weight was measured to the nearest 0.1 kg and fasting
concentrations
glucose andwere measured
insulin by enzymatic
concentrations assay and
were measured ELISA, respectively
by enzymatic [30]. respectively [30].
assay and ELISA,

Excluded through a phone screening

Assessed for eligibility with consent (n = 36)
Enrollment - Declined to participate (n = 0)
(n = 64)
- Not meeting inclusion criteria
(n =3 6) including:
age, n = 4
BMI, n = 9
time commitments, n = 10
tobacco use, n = 2
prescription medications that impact
body weight or energy balance, n = 2
diabetes, n = 2
recent or pending surgery, n = 3 or
pregnancies, n = 3
highly restricted dietary needs, n = 1

Excluded after in-person consenting and

measurement of anthropometrics and
biochemistries (n = 8)
- Declined to participate (n = 3)
- Not meeting inclusion criteria (n = 5)
including subject having normal levels
of glucose and insulin < 10 mU/L

Eligible (n = 20)

Allocated to intervention (n = 20)

¨ Received intervention (n = 20)
Intervention
¨ Did not receive allocated
intervention (n = 0)

Completed Parent Study (n = 20)

Follow-Up ¨ Lost to follow-up (n = 0)
¨ Discontinued (n = 0)

No follow up data for FCI

(n = 1)
No follow up data for TFEQ
(n = 2)
Analysed for cravings, FCI (n = 19)
Analysis Analysed for TFEQ (n = 18)

Figure
Figure 1. CONSORT
1. CONSORT diagram
diagram indicatingthe
indicating the sample
sample size sizeatateach
eachstage of of
stage thethe
study. ThisThis
study. diagram is
diagram is
modified
modified withwith permission
permission from
from Syed-Majidetetal.
Syed-Majid al. [30].
[30].
Nutrients 2020, 12, 52 4 of 12

2.2. Food Cravings

The Food Craving Inventory (FCI), originally developed by White et al. [32], has been previously
validated and demonstrates internal consistency (Cronbach’s α = 0.76–0.93). The FCI is a 33-item
questionnaire to assess cravings across five subscales: (1) high-fat foods (e.g., fried chicken, gravy,
sausages, hot dogs), (2) sweets (e.g., cake, cinnamon rolls, ice cream, cookies), (3) CHO/starches (e.g.,
sandwich bread, rice, biscuits, pasta, pancakes), (4) fast-food fats (pizza, french fries, hamburger,
chips), and (5) fruit/vegetable subscale (raw fruits, raw vegetables, canned fruit, fruit juice, and cooked
vegetables). The fruit/vegetable subscale was not included in the original FCI by White and colleagues
but has been previously used by others [9,13]. Items on the questionnaire are scored on a 1–5 scale with
1 = never and 5 = always/almost every day. A total score for each subscale is calculated by taking the
average of scores for individual foods within that subscale. A total craving score is calculated from the
aggregate or mean of all five subscales. For the present analysis, internal consistency (Cronbach’s α) was
found for baseline and post low-CHO diet, respectively, for the high-fat food subscale (0.85 and 0.83),
sweets (0.87 and 0.66), CHO/starches (0.88 and 0.80), fast-food fats (0.80 and 0.81), fruits/vegetables
(0.80 and 0.76), and overall cravings (0.94 and 0.92). These values represent acceptable to good internal
consistency of the scales used to measure food cravings.

2.3. Eating Behavior

The Three Factor Eating questionnaire (TFEQ), originally developed by Stunkard and Messick [24],
is a 51-item questionnaire with three subscales: a 21-item dietary restraint scale, a 16-item dietary
disinhibition scale, and a 14-item hunger scale. The dietary restraint scale measures cognitive control of
eating, i.e., the tendency to consciously restrict or control food intake as a means of controlling weight,
while dietary disinhibition scale measures loss of cognitive control of eating which is the tendency to
overeat in response to negative emotional states or the presence of highly palatable foods, and the
hunger scale measures susceptibility to feelings of hunger.

2.4. Statistical Analyses

All statistical analyses were performed by using the Statistical Package for Social Sciences (IBM
SPSS Statistics, version 25, SPSS Inc., Armonk, NY, USA). Participant characteristics are expressed as
mean ± SD, while mean cravings, and eating behavior scores (dietary restraint, dietary disinhibition,
and hunger) are expressed as mean ± SE. Statistical significance was accepted at p < 0.05. For the
food cravings data, analyses focused on each participant’s aggregated cravings ratings, which were
computed by averaging an outcome rating for foods in the same category (high fat, sweet, CHO/starches,
fast-food fats, fruits/vegetables) for baseline, and week 4 (post low-CHO diet). For each of the 33 items
measuring food cravings on the FCI, a total craving score was calculated as the average of the overall
score for high fat, sweet, CHO/starches, fast-food fats, and fruits/vegetables. Thus, six craving variables
(high fat, sweets, CHO/starches, fast-food fats, fruits/vegetables, and total cravings) were included as
outcomes. Similarly, for the eating behaviors, an aggregated score for each of three constructs (dietary
restraint, disinhibition, and hunger) was calculated by averaging the scores obtained by each subject
for each of the three different constructs. A paired-sample t-test was used to assess differences in
mean cravings scores, dietary restraint, dietary disinhibition, and hunger between baseline and post
low-CHO diet (week 4).
A repeated-measures ANOVA with sex as the between-subject factor and time (baseline, post
low-CHO diet) as the within-subject factor was carried out to test the main effect of sex on cravings
and eating behaviors. Pearson correlations were used to assess the relationships between change in
cravings and change in body weight, change in eating behavior, and change in insulin and glucose
concentrations. Change variables were calculated by subtracting the value at baseline from the
value at week 4. Thus, a positive value indicates increased cravings, restraint, disinhibition, hunger,
weight gain, glucose and insulin concentrations, while a negative value implies decreased cravings,
Nutrients 2020, 12, 52 5 of 12

restraint, disinhibition, hunger, weight loss, glucose and insulin concentrations. Percent changes were
calculated by dividing the difference between baseline and week 4 values by the baseline value and
then multiplying the answer by 100 for each subject. The average was then taken as the percent change
for the variable in question. This approach allowed us to capture variability in responses within
individual subjects.

3. Results
Nutrients 2020, 12, x FOR PEER REVIEW 5 of 12

3.1. Participants, Biochemical

then multiplying Indices by
the answer and Body
100 Weight
for each subject. The average was then taken as the percent
change for the variable in question. This approach allowed us to capture variability in responses
Baseline characteristics are presented in Table 1; subjects were middle-aged and had BMIs in the
within individual subjects.
overweight to obese range. Fasting glucose concentrations were normal at baseline and did not change
after WL (95.5 ± 2.2 vs. 90.3 ± 3.0 mg/dL, p = 0.139). Baseline fasting insulin concentrations tended
3. Results
to be elevated (12.2 ± 1.2 mU/L) and fell 22% with WL (8.8 ± 0.6 mU/L, p = 0.003). Participants lost
3.1. Participants, Biochemical Indices and Body Weight
6.0 ± 0.5 kg from baseline to week 4 representing ~6% of their baseline body weight. In men compared
Baseline characteristics are presented in Table 1; subjects were middle-aged and had BMIs in the
to women, a total of 7.2 ± 0.7 kg and 4.7 ± 0.5 kg was lost from baseline to week 4, representing 6.3%
overweight to obese range. Fasting glucose concentrations were normal at baseline and did not
and 5.0% of baseline
change after body
WL (95.5weight, respectively
± 2.2 vs. (treatment
90.3 ± 3.0 mg/dL, × sex
p = 0.139). effect,
Baseline p = 0.011).
fasting Men lost significantly
insulin concentrations
more weight than
tended women
to be elevated(p = 0.002,
(12.2 ANOVA
± 1.2 mU/L) and felleffect of sex).
22% with WL (8.8However,
± 0.6 mU/L,both menParticipants
p = 0.003). and women lost a
significantlost 6.0 ± 0.5ofkgweight
amount from baseline
from to week 4 representing
baseline. Weight loss ~6% trajectories
of their baseline
forbody weight. In and
all subjects, men men and
compared to women, a total of 7.2 ± 0.7 kg and 4.7 ± 0.5 kg was lost from baseline to week 4,
women are shown in Figure 2. All participants reported that they were either somewhat satisfied (45%)
representing 6.3% and 5.0% of baseline body weight, respectively (treatment × sex effect, p = 0.011).
or extremely
Mensatisfied (55%) with
lost significantly the diet
more weight post-intervention
than (data not
women (p = 0.002, ANOVA shown).
effect of sex). However, both
men and women lost a significant amount of weight from baseline. Weight loss trajectories for all
subjects, and men and 1. Baseline
women
Table characteristics
are shown of participants
in Figure 2. All reported 1that
study participants . they were either
somewhat satisfied (45%) or extremely satisfied (55%) with the diet post-intervention (data not
shown).Characteristic All (n = 19) Men (n = 10) Women (n = 9)
Age (year) 40.6 ± 8.4 41.8 ± 10.2
Table 1. Baseline characteristics of study participants 1.
39.2 ± 6.3
Body weight (kg) 103.4 ± 16.9 113.8 ± 15.4 91.9 ± 9.6
Height (m) Characteristic 1.74 ±All (n = 19) Men 1.82
0.10 (n = 10) Women (n = 9) 1.65 ± 0.03
± 0.06
Age (year)
BMI (kg/m2 ) 34.1 ±40.6
3.3± 8.4 41.8 ± 10.2
34.2 ± 3.0 39.2 ± 6.3 34.0 ± 3.8
Body weight (kg) 103.4 ± 16.9 113.8 ± 15.4 91.9 ± 9.6
Plasma insulin (mU/L)
Height (m)
12.2 ± 5.2
1.74 ± 0.10
11.6 ± 3.91.65 ± 0.03 12.8 ± 6.6
1.82 ± 0.06
BMI (kg/m2) 95.5 ±34.1
Plasma glucose (mg/dL) 9.6± 3.3 ± 3.0 ± 7.9 34.0 ± 3.8
34.2 93.5 97.8 ± 11.3
Physical activity (steps)
Plasma 7295 ±12.2
insulin (mU/L) 2549± 5.2 ± 3.9± 289712.8 ± 6.6
8586
11.6 5860 ± 887
Plasma glucose (mg/dL)
1 Data 95.5 ± 9.6 93.5 ± 7.9 97.8 ± 11.3
are mean ± standard deviation.
Physical activity (steps) 7295 ± 2549 8586 ± 2897 5860 ± 887
1 Data are mean ± standard deviation.

100% 99.1%
98.3%
99%
97.4%
98% 96.7%
98.1%
97% Women
97.3%
96% 96.7% 95.0%
% Starting
95%
body weight 95.6%
94% Men
93% 93.7%
92%
91%
90%
0 7 14 21 28
Days

Figure 2. Body
Figure weight loss (WL)
2. Body weight overover
loss (WL) 4 weeks
4 weeksof of aa low-carbohydrate (low-CHO),
low-carbohydrate (low-CHO), energy restricted
energy restricted
diet. Data diet.
are Data
mean are±mean ± SE reported
SE and and reported
forforNN== 19 19 subjects
subjects who hadhad
who complete data fordata
complete food for
cravings.
food cravings.
Analysis of body WL over time revealed significant effects for men and women combined (center line,
Analysis of body WL over time revealed significant effects for men and women combined (center line,
with diamond symbols) and also separately for the sexes. WL in men (dotted line with open circles)
with diamond symbols)greater
was significantly and alsothanseparately
in women (top forlinethe
withsexes.
closedWL in pmen
circles, (dotted
= 0.002, ANOVA line with
effect open circles)
of sex).
was significantly greater
This figure thanfrom
is modified in women (top
Syed-Majid line
et al. [30]with
withclosed circles, p = 0.002, ANOVA effect of sex).
permission.
This figure is modified from Syed-Majid et al. [30] with permission.
Nutrients 2020, 12, 52 6 of 12
Nutrients 2020, 12, x FOR PEER REVIEW 6 of 12

3.2. Food
Food Cravings
Cravings and Eating Behaviors
As shown
shown in inFigure
Figure3A, 3A,cravings
cravings were
weresignificantly
significantlyreduced
reduced from baseline
from to week
baseline 4 (post
to week low-
4 (post
CHO
low-CHOdiet)diet)
for most cravings
for most cravingssubscales:
subscales:high
highfatfat
(−0.3 ± 0.1,
(−0.3 p p= =0.028),
± 0.1, 0.028),sweets
sweets(−0.6 0.2,pp ==0.014),
(−0.6±±0.2, 0.014),
CHO/starches
CHO/starches (−0.4 0.1, ,pp==0.028),
(−0.4 ±± 0.1 0.028),fast-food
fast-food fats
fats (−0.6
(−0.6 ± 0.1, p =p 0.0003),
± 0.1, = 0.0003),
andand total
total cravings
cravings (−0.4(−0.4 ±
± 0.1,
p = 0.005).
0.1, p = 0.005). By contrast,
By contrast, cravings
cravings for fruits/vegetables
for fruits/vegetables only tended
only tended to be reduced
to be reduced 0.1, p±=0.1,
(−0.2 ±(−0.2 p=
0.094).
0.094).
Overall,Overall, for the majority
for the majority of reductions
of reductions in cravings,
in cravings, women exhibited
women exhibited greater
greater effects effects
than men,than men,
although
although only for
only for sweet sweet cravings—
cravings— this effectthis
significant (p = 0.013).
effect significant (p As
= 0.013).
shownAsinshown
Figurein 3B,Figure
sweet3B, sweet
cravings
cravings
dropped dropped
for all butfor
oneallwoman
but onewhereas
woman in whereas in men,in
men, changes changes in sweet were
sweet cravings cravings were variable.
variable.

A5 Baseline Low-CHO diet

4 p = 0.028 p = 0.014 p = 0.028 p = 0.0003 p = 0.094 p = 0.005

3
3.1
Craving score ±0.2

(1–5 scale) 2 2.5

2.3
2.4
2.4
2.3 ±0.2 ±0.2 2.2
±0.2 ±0.2
±0.2 1.9 2.0 ±0.2 2.0 2.0
1.9
±0.1 ±0.1 ±0.2 ±0.1
±0.1
1

0
High Fats Sweets Carbohydrates/ Fast-Food Fats Fruits/Vegetables Total Cravings
Starches

5 p = 0.013
B
p = 0.292 p = 0.026
4

3
Sweet cravings
(1–5 scale) 2

0
Men Women
Figure 3.
Figure Mean craving
3. Mean craving scores
scores over
over time
time (A),
(A), and
and sweet
sweet cravings
cravings by
by sex
sex (B).
(B). Data
Data are mean ±
are mean SE and
± SE and
reported for N =
reported for N = 19 subjects (10 men and 9 women) who had complete cravings data. A score of 1of
19 subjects (10 men and 9 women) who had complete cravings data. A score =
= never
1never and 5 = always/almost every day. All cravings subcategories except fruits/vegetables
and 5 = always/almost every day. All cravings subcategories except fruits/vegetables were were
significantly reduced
significantly reduced in the total
in the total sample.
sample. Only
Only sweet
sweet cravings
cravings showed
showed aa significant
significant main
main effect
effect of
of sex.
sex.
All significance tests were two-tailed.
All significance tests were two-tailed.

Dietary restraint was significantly increased (~102 ± 21%) from baseline to week 4 after the
Dietary restraint was significantly increased (~102 ± 21%) from baseline to week 4 after the low-
low-CHO diet, while both disinhibition (~17 ± 7%) and hunger scores were significantly reduced
CHO diet, while both disinhibition (~17 ± 7%) and hunger scores were significantly reduced (22 ± 9%)
from baseline to week 4 in the entire cohort (Figure 4). No significant main effect of sex was found
for any of the three TEFQ subscales (p > 0.05).
Nutrients 2020, 12, 52 7 of 12

(22 ± 9%) from baseline to week 4 in the entire cohort (Figure 4). No significant main effect of sex was
found for
Nutrients any
2020, 12,of the PEER
x FOR threeREVIEW (p > 0.05).
TEFQ subscales 7 of 12

16 Baseline Low-CHO diet

14 p < 0.001 p = 0.002 p = 0.012

Eating Behavior Score

12
11.8
10 ±1.1
9.8
8 ±0.9
8.0
6 ±0.8
6.6 6.4
±0.8 ±0.8
4 4.8
±0.7
2

Restraint Disinhibition Hunger

Figure 4. Mean eating behavior
behaviorscores
scoresininthe
thetotal
totalsample.
sample.Data
Dataare mean
are mean± SE
± SEand
andreported N =N18
forfor
reported =
subjects
18 (9 men
subjects and
(9 men 9 women
and with
9 women complete
with complete data
dataforfor
TFEQ). AllAll
TFEQ). significance tests
significance were
tests two-tailed.
were two-tailed.

3.3. Food
3.3. Food Cravings
Cravings Correlations
Correlations withwith Body
Body Weight,
Weight, Glucose
Glucose andand Eating
Eating Behavior
Behavior
At baseline,
At baseline, within
within the the FCI,
FCI, the
the higher
higher thethe baseline
baseline sweetsweet cravings
cravings scores,
scores, thethe greater
greater sweet
sweet
cravings fell during the intervention (r = −0.865, p < 0.0001, data not shown).
cravings fell during the intervention (r = −0.865, p < 0.0001, data not shown). The same was true for The same was true for
high fat
high fat cravings
cravings (r(r = −0.566,pp==0.012).
= −0.566, 0.012).At Atbaseline,
baseline,between
betweenfood foodcraving
cravingscores
scoresandandeating
eating behaviors
behaviors
scores, no significant relationships were found. With regard to changes in
scores, no significant relationships were found. With regard to changes in these variables over time, these variables over time,
Figure 5A–F shows the relationships between changes in food cravings
Figure 5A–F shows the relationships between changes in food cravings and eating behaviors, body and eating behaviors, body
weight, and
weight, and plasma
plasma glucose
glucose concentrations.
concentrations. Changes
Changes in in cravings
cravings for for high-fat
high-fat foods
foods from
from baseline
baseline to to
week 4 were significantly negatively related to change in body weight
week 4 were significantly negatively related to change in body weight (Figure 5A). In other words,(Figure 5A). In other words,
although all
although all subjects
subjects lost
lost weight
weight during
during the the WL
WL intervention,
intervention, subjects
subjects who
who lostlost the
the most
most weight
weight
experienced the
experienced the smallest
smallest reductions
reductions in in cravings
cravings forfor high-fat
high-fat foods.
foods. This
This relationship
relationship was was lost
lost when
when
controlled for by baseline high-fat craving levels, which suggest that the
controlled for by baseline high-fat craving levels, which suggest that the starting level of high-fat starting level of high-fat
craving governed
craving governed the the change.
change. No No other
other significant
significant correlations
correlations were were observed
observed forfor changes
changes in in the
the other
other
craving variables and changes in body weight. Changes in sweet cravings
craving variables and changes in body weight. Changes in sweet cravings were significantly were significantly positively
related to the
positively changes
related to thein blood
changes glucose
in bloodconcentrations (Figure 5B) such
glucose concentrations that5B)
(Figure those who
such exhibited
that those who the
greatest reductions
exhibited the greatestinreductions
blood glucose had glucose
in blood the greatest
had the reductions in cravingsinfor
greatest reductions sweets.for
cravings No other
sweets.
significant correlations were observed between changes in glucose and changes
No other significant correlations were observed between changes in glucose and changes in the other in the other craving
subscales.
craving ChangesChanges
subscales. in cognitive restraint were
in cognitive alsowere
restraint statistically negativelynegatively
also statistically correlated with changes
correlated within
sweet cravings
changes in sweet (Figure
cravings 5C).(Figure
A similar5C).significant
A similarnegative
significant trend was observed
negative trend was forobserved
changes forin cognitive
changes
restraint and three other cravings subscales: changes in CHO/starch cravings
in cognitive restraint and three other cravings subscales: changes in CHO/starch cravings (Figure (Figure 5D), changes in
5D),
fast-foodinfatfast-food
changes cravingsfat (Figure 5E) (Figure
cravings and changes5E) andin overall
changescravings
in overall (Figure 5F). (Figure
cravings Thus, those individuals
5F). Thus, those
whose cravings for sweets, CHO/starches, fast-food fats, and total cravings
individuals whose cravings for sweets, CHO/starches, fast-food fats, and total cravings decreased decreased the most alsothe
had the
most alsogreatest
had theincreases
greatest in cognitive
increases in restraint.
cognitive restraint.
Nutrients 2020, 12, 52 8 of 12
Nutrients 2020, 12, x FOR PEER REVIEW 8 of 12

2.0 B 3
A All subjects
1.5 Men All subjects
r = −0.458 2 r = 0.479
Women p = 0.038
Δ High Fats Craving

1.0 p = 0.049

Δ Sweets Craving
1
0.5
0.0 0
-0.5
-1
-1.0
-2
-1.5
-2.0 -3
-10 -9 -8 -7 -6 -5 -4 -3 -2 -40 -30 -20 -10 0 10 20 30 40
Δ Body Weight (kg) Δ Glucose Concentration (mg/dL)

3 D 2.0
C

Δ Cabohydrates/ Starches Craving

1.5 All subjects
2 All subjects Women only r = −0.572
r = −0.589 r = −0.801 1.0 p = 0.016
Δ Sweet Craving

1 p = 0.013 p = 0.017
0.5
0 0.0
-0.5
-1
-1.0
-2
-1.5
-3 -2.0
0 2 4 6 8 10 12 0 2 4 6 8 10 12
Δ Restraint Δ Restraint

E 1.5
F 1.0
All subjects All subjects
1.0
Δ Fast-Food Fats Craving

r = −0.572 0.5 r = −0.578

0.5 p = 0.016 p = 0.015
Δ Total Craving

0.0
0.0
-0.5 -0.5
-1.0
-1.0
-1.5
-1.5
-2.0
-2.5 -2.0
0 2 4 6 8 10 12 0 2 4 6 8 10 12
Δ Restraint Δ Restraint

Figure5.5. Relationships
Figure Relationships between
between craving
craving scores,
scores, body
body weight,
weight, glucose
glucoseconcentrations,
concentrations, and
andeating
eating
behaviors.
behaviors.Correlations
Correlationsbetween
betweenchange
changeininhigh-fat
high-fatcravings and
cravings andchange in body
change weight
in body (A),(A),
weight change in
change
sweet cravings and change in blood glucose concentration (B), and change in sweet, CHO/starches,
in sweet cravings and change in blood glucose concentration (B), and change in sweet, CHO/starches,
fast-food
fast-foodfats,
fats,total
totalcravings
cravingsand
anddietary
dietaryrestraint
restraint(C–F).
(C–F).Change
Changevalues
valueswere
werecalculated
calculatedby
bysubtracting
subtracting
the
thevalue
valueatatbaseline
baselinefrom
fromweek
week4.4.Data
Dataare
arereported
reportedforforNN==1919subjects
subjectsthat
thathad
hadcomplete
completedata
datafor
for
cravings,
cravings,body
bodyweight
weightandandglucose concentration
glucose concentration(A,B)
(A,B) N =N17
andand = subjects whowho
17 subjects hadhad
complete datadata
complete for
both cravings
for both and and
cravings TFEQ (C,D).
TFEQ All significance
(C,D). teststests
All significance werewere
two-tailed.
two-tailed.

4. Discussion
4. Discussion
In this study, we investigated changes in cravings and eating behaviors during a 4-week low-CHO,
In this study, we investigated changes in cravings and eating behaviors during a 4-week low-
WL intervention to understand how changes in cravings were related to eating behaviors and some
CHO, WL intervention to understand how changes in cravings were related to eating behaviors and
biochemical indices. We also explored sex differences in cravings and eating behaviors following the
some biochemical indices. We also explored sex differences in cravings and eating behaviors
intervention. We hypothesized that with the exception of fruits/vegetables, all the FCI subscales would
following the intervention. We hypothesized that with the exception of fruits/vegetables, all the FCI
be reduced and that reductions in cravings would be correlated with WL, eating behaviors, and blood
subscales would be reduced and that reductions in cravings would be correlated with WL, eating
glucose. Furthermore, we hypothesized that CHO-restriction would result in increased restraint and
behaviors, and blood glucose. Furthermore, we hypothesized that CHO-restriction would result in
decreased disinbition and hunger. First, consistent with previous studies of longer duration [5,7–17],
increased restraint and decreased disinbition and hunger. First, consistent with previous studies of
our data showed a reduction in specific and total cravings after only 4 weeks. In an elegant study
longer duration [5,7–17], our data showed a reduction in specific and total cravings after only 4
by Kahathuduwa et al. [18], a low-energy diet fed for 3 weeks resulted in similar WL (~4 kg) and
weeks. In an elegant study by Kahathuduwa et al. [18], a low-energy diet fed for 3 weeks resulted in
similar WL (~4 kg) and significant reductions in cravings for sweets and starchy foods. These lower
cravings were supported by fMRI data showing decreased activations in brain reward regions—
Nutrients 2020, 12, 52 9 of 12

significant reductions in cravings for sweets and starchy foods. These lower cravings were supported
by fMRI data showing decreased activations in brain reward regions—whether similar fMRI changes
would be observed with a low-CHO diet is not known. Our findings that most food cravings subscales
and total cravings were reduced support the classical conditioning model [33] for cravings and not the
deficiency model [2,34,35]. If the theory of deficiency were to hold, then we would expect cravings to
increase with the low-CHO, WL diet.
Second, no sex specific effects on cravings were observed, except for changes in sweet cravings
where women exhibited greater reductions in cravings than men. Although this line of research has
not been extensively studied, according to a 2016 review of cravings in obese individuals [23], sex
differences in cravings were found for (1) the types of food craved (women tended to crave sweets
while men craved savory foods), (2) the frequency and intensity of food craved (women reported
higher craving scores and frequency that men, and (3) the regulation of food craving (women reported
it was harder to regulate food cravings than men). Women are disproportionately burdened by
overweight/obesity and that previous data show that women may be less succesful at regulating
cravings and report higher cravings for sweets, compared to men [23]. Our finding that women were
more successful at reducing sweet cravings is encouraging and supports the idea that women may have
benefitted more from the short-term CHO-restriction than men. Our data suggest that sex differences
in sweet cravings should be taken into consideration when designing dietary regimens for weight loss.
Third, dietary restraint was increased in the total sample—a finding that is consistent with
previous reports [22,36], while disinhibition and hunger were reduced. Furthermore, our correlational
analyses demonstrated that changes in cravings were mostly unrelated to changes in body weight.
The one exception was that those who lost the most weight exhibited the smallest reductions in
cravings for high-fat foods; this relationship was lost after controlling for baseline high-fat. One
interpretaton of this result might be that those individuals who preferred high-fat foods at baseline,
had their food preference met by the low-CHO (high-fat) diet, which supported dietary adherence,
leading to more body WL. Thus, alignment of a subject’s cravings with the macronutrient content
of the diet could contribute to additional WL success. How their sustained cravings for high-fat
foods was related to this scenario is not known. Gilhooly et al. [4] found that over a 6-month WL
intervention of low-glycemic and high-glycemic load foods, subjects who lost a greater percentage of
weight also craved higher energy dense foods at month 6 but also reported giving in to food cravings
less frequently. This supports the notion that, it is the behavior that follows food cravings (resisting
food cravings or feeling in control of eating) rather than the intensity or frequency of cravings that
contributes to successful WL [37]. The fact that the significant relationship between change in high-fat
craving and change in body weight was lost after controlling for baseline fast-food craving suggests
that the initial level of high-fat cravings drove the change. Also, the greater the reductions in cravings
for sweets, CHO/starches, fast-food fats, and total cravings, the greater the dietary restraint during the
WL intervention. These observations suggest that cravings and dietary restraint are related to each
other. Lastly, those individuals whose cravings for sweets were reduced the most experienced the
greatest reductions in blood glucose concentrations, which suggests that the connection between these
two events may be mediated by reductions in dietary CHO intake.
Strengths of this study include the controlled dietary intake with the provision of some of the
foods. This may have made it easier for subjects to adhere to the diet and allowed the investigation
of changes in food cravings and eating behaviors under controlled conditions. The use of validated
questionnaires to assess food cravings and eating behaviors increased rigor, as did the use of biological
measures to test dietary adherence (blood ketones, indirect calorimetry [30]). Objectively measuring
compliance to WL diets is paramount to explaining weight changes as some have argued that it is the
adherence to the dietary regimen rather than changes in variables such as cravings and eating behavior
that moderates weight loss [16,38,39].
Study limitations included the lack of a control group not on a low-CHO diet; the current results
warrant a larger study of randomized design. The use of a standardized diet for all participants may
Nutrients 2020, 12, 52 10 of 12

have resulted in men losing more weight than women and the sample size was small, which is a
common limitation of studies where food intake is highly controlled. Nonetheless, the number of
subjects was sufficient to statistically detect within-subject-differences. It is possible that the study was
not sufficiently powered to be able to detect relationships between outcomes of interest by correlational
analyses and hence the inability to see significant relationships between changes in cravings and
changes in body weight for most of our craving subscales. Given an α of 0.05, power of 80%, and effect
size of Pearson’s r of 0.5 (large) as proposed by Cohen [40], the sample size required for statistically
significant correlations was 29 assuming a null hypothesis of zero correlation. In our correlational
analysis, we had complete data for 17–19 subjects depending on the outcome of interest.

5. Conclusions
In summary, a low-CHO diet promoted significant WL over a short time-frame and resulted
in significant reductions in specific and total cravings, reductions in disinhibition and hunger, and
increased dietary restraint in the total sample studied. Greater increases in dietary restraint resulted in
greater decreases in some specific and total cravings. Women exhibited greater reductions in sweet
cravings than men and thus, dietary strategies aimed at ameliorating sweet food cravings may need to
take sex into consideration. Though all participants lost weight, our data showed that a greater WL
induced by a low-CHO diet was associated with smaller reductions in high-fat cravings but was not
significantly related to any other specific or total craving. This relationship was lost when controlling
for baseline levels for high-fat food cravings which suggests that when high-fat cravings are elevated
before WL, a low-CHO diet may be efficacious. Due to the preliminary nature of our findings, future
studies should be conducted using a randomized, controlled design with a larger sample size and also
control for sex differences in energy requirements. These data support the conditioning model [33] for
cravings and the relationship between high fat cravings and body weight reductions suggests that
subjects that crave high-fat foods may be able to satisfy their cravings with a low-CHO, high-fat diet
and, thus, improve their WL success.

Author Contributions: E.J.P. designed the parent study’s diet intervention and C.H. provided advice on diet
preparation. K.O.-B.A. analyzed and interpreted the data and wrote the manuscript. M.M.S.-A., Q.H., M.J.-S.
acquired the data and V.C. and C.H. aided in data interpretation. E.J.P. aided in data interpretation and contributed
to the manuscript writing. All authors have read and agreed to the published version of the manuscript.
Funding: As described in the previous paper, funding for this project was provided by funds from the University
of Missouri and a grant from Atkins Nutritionals, Inc.
Acknowledgments: The authors would like to express appreciation to all the study participants in this research
for their dedication, and to Nathan Le for his expert technical support during data collection.
Conflicts of Interest: Parks is a member of the Scientific Advisory Board of Atkins Nutritionals Inc., who provided
some of the foods and funding for the parent study. The sponsors had no role in data collection nor in the conduct
of the study and the manuscript was written independently. As with all other co-authors, Heimowitz and Cox
(formerly an employee of Atkins Nutritionals) read and approved the manuscript. Anguah, Syed-Abdul, Hu, and
Jacome-Sosa have no conflicts to declare.

References
1. Boswell, R.G.; Kober, H. Food cue reactivity and craving predict eating and weight gain: A meta-analytic
review. Obes. Rev. 2016, 17, 159–177. [CrossRef]
2. Weingarten, H.P.; Elston, D. The phenomenology of food cravings. Appetite 1990, 15, 231–246. [CrossRef]
3. Weingarten, H.P.; Elston, D. Food cravings in a college population. Appetite 1991, 17, 167–175. [CrossRef]
4. Gilhooly, C.H.; Das, S.K.; Golden, J.K.; McCrory, M.A.; Dallal, G.E.; Saltzman, E.; Kramer, F.M.; Roberts, S.B.
Food cravings and energy regulation: The characteristics of craved foods and their relationship with eating
behaviors and weight change during 6 months of dietary energy restriction. Int. J. Obes. 2007, 31, 1849–1858.
[CrossRef] [PubMed]
Nutrients 2020, 12, 52 11 of 12

5. Jakubowicz, D.; Froy, O.; Wainstein, J.; Boaz, M. Meal timing and composition influence ghrelin levels,
appetite scores and weight loss maintenance in overweight and obese adults. Steroids 2012, 77, 323–331.
[CrossRef] [PubMed]
6. Segal-Isaacson, C.J.; Johnson, S.; Tomuta, V.; Cowell, B.; Stein, D.T. A randomized trial comparing low-fat
and low-carbohydrate diets matched for energy and protein. Obes. Res. 2004, 12 (Suppl. 2), 130S–140S.
[CrossRef]
7. Harvey, J.; Wing, R.R.; Mullen, M. Effects on food cravings of a very low calorie diet or a balanced, low
calorie diet. Appetite 1993, 21, 105–115. [CrossRef]
8. Martin, C.K.; O’Neil, P.M.; Pawlow, L. Changes in food cravings during low-calorie and very-low-calorie
diets. Obesity 2006, 14, 115–121. [CrossRef]
9. Anton, S.D.; Gallagher, J.; Carey, V.J.; Laranjo, N.; Cheng, J.; Champagne, C.M.; Ryan, D.H.; McManus, K.;
Loria, C.M.; Bray, G.A.; et al. Diet type and changes in food cravings following weight loss: Findings from
the POUNDS LOST Trial. Eat. Weight Disord.-Stud. Anorex. Bulim. Obes. 2012, 17, e101–e108. [CrossRef]
10. Batra, P.; Das, S.K.; Salinardi, T.; Robinson, L.; Saltzman, E.; Scott, T.; Pittas, A.G.; Roberts, S.B. Relationship
of cravings with weight loss and hunger. Results from a 6 month worksite weight loss intervention. Appetite
2013, 69, 1–7. [CrossRef]
11. Martin, C.K.; Rosenbaum, D.; Han, H.; Geiselman, P.J.; Wyatt, H.R.; Hill, J.O.; Brill, C.; Bailer, B.; Miller, B.V., 3rd;
Stein, R.; et al. Change in food cravings, food preferences, and appetite during a low-carbohydrate and
low-fat diet. Obesity 2011, 19, 1963–1970. [CrossRef] [PubMed]
12. Moldovan, C.P.; Weldon, A.J.; Daher, N.S.; Schneider, L.E.; Bellinger, D.L.; Berk, L.S.; Herme, A.C.;
Arechiga, A.L.; Davis, W.L.; Peters, W.R. Effects of a meal replacement system alone or in combination with
phentermine on weight loss and food cravings. Obesity 2016, 24, 2344–2350. [CrossRef] [PubMed]
13. Apolzan, J.W.; Myers, C.A.; Champagne, C.M.; Beyl, R.A.; Raynor, H.A.; Anton, S.A.; Williamson, D.A.;
Sacks, F.M.; Bray, G.A.; Martin, C.K. Frequency of consuming foods predicts changes in cravings for those
foods during weight Loss: The POUNDS Lost study. Obesity 2017, 25, 1343–1348. [CrossRef] [PubMed]
14. Bowen, J.; Brindal, E.; James-Martin, G.; Noakes, M. Randomized trial of a high protein, partial meal
replacement program with or without alternate day fasting: Similar effects on weight loss, retention status,
nutritional, metabolic, and behavioral outcomes. Nutrients 2018, 10, 1145. [CrossRef]
15. Castro, A.I.; Gomez-Arbelaez, D.; Crujeiras, A.B.; Granero, R.; Aguera, Z.; Jimenez-Murcia, S.; Sajoux, I.;
Lopez-Jaramillo, P.; Fernandez-Aranda, F.; Casanueva, F.F. Effect of a very low-calorie ketogenic diet on food
and alcohol cravings, physical and sexual activity, sleep disturbances, and quality of life in obese patients.
Nutrients 2018, 10, 1348. [CrossRef]
16. Sayer, R.D.; Peters, J.C.; Pan, Z.; Wyatt, H.R.; Hill, J.O. Hunger, food cravings, and diet satisfaction are related
to changes in body weight during a 6-month behavioral weight loss intervention: The Beef WISE study.
Nutrients 2018, 10, 700. [CrossRef]
17. Watson, N.A.; Dyer, K.A.; Buckley, J.D.; Brinkworth, G.D.; Coates, A.M.; Parfitt, G.; Howe, P.R.C.; Noakes, M.;
Murphy, K.J. Reductions in food cravings are similar with low-fat weight loss diets differing in protein and
carbohydrate in overweight and obese adults with type 2 diabetes: A randomized clinical trial. Nutr. Res.
2018, 57, 56–66. [CrossRef]
18. Kahathuduwa, C.N.; Davis, T.; O’Boyle, M.; Boyd, L.A.; Chin, S.H.; Paniukov, D.; Binks, M. Effects of 3-week
total meal replacement vs. typical food-based diet on human brain functional magnetic resonance imaging
food-cue reactivity and functional connectivity in people with obesity. Appetite 2018, 120, 431–441. [CrossRef]
19. Kahathuduwa, C.N.; Binks, M.; Martin, C.K.; Dawson, J.A. Extended calorie restriction suppresses overall
and specific food cravings: A systematic review and a meta-analysis. Obes. Rev. 2017, 18, 1122–1135.
[CrossRef]
20. Saslow, L.R.; Daubenmier, J.J.; Moskowitz, J.T.; Kim, S.; Murphy, E.J.; Phinney, S.D.; Ploutz-Snyder, R.;
Goldman, V.; Cox, R.M.; Mason, A.E.; et al. Twelve-month outcomes of a randomized trial of a
moderate-carbohydrate versus very low-carbohydrate diet in overweight adults with type 2 diabetes
mellitus or prediabetes. Nutr. Diabetes 2017, 7, 304. [CrossRef]
21. Dyson, P.A.; Beatty, S.; Matthews, D.R. An assessment of low-carbohydrate or low-fat diets for weight loss at
2 year’s follow-up. Diabet. Med. 2010, 27, 363–364. [CrossRef] [PubMed]
Nutrients 2020, 12, 52 12 of 12

22. Dorling, J.L.; Bhapkar, M.; Das, S.K.; Racette, S.B.; Apolzan, J.W.; Fearnbach, S.N.; Redman, L.M.; Myers, C.A.;
Stewart, T.M.; Martin, C.K. Change in self-efficacy, eating behaviors and food cravings during two years of
calorie restriction in humans without obesity. Appetite 2019, 143, 104397. [CrossRef] [PubMed]
23. Hallam, J.; Boswell, R.G.; DeVito, E.E.; Kober, H. Gender-related Differences in Food Craving and Obesity.
Yale J. Biol. Med. 2016, 89, 161–173. [PubMed]
24. Stunkard, A.J.; Messick, S. The three-factor eating questionnaire to measure dietary restraint, disinhibition
and hunger. J. Psychosom. Res. 1985, 29, 71–83. [CrossRef]
25. Savage, J.S.; Hoffman, L.; Birch, L.L. Dieting, restraint, and disinhibition predict women’s weight change
over 6 y. Am. J. Clin. Nutr. 2009, 90, 33–40. [CrossRef]
26. Urbanek, J.K.; Metzgar, C.J.; Hsiao, P.Y.; Piehowski, K.E.; Nickols-Richardson, S.M. Increase in cognitive
eating restraint predicts weight loss and change in other anthropometric measurements in overweight/obese
premenopausal women. Appetite 2015, 87, 244–250. [CrossRef]
27. Foster, G.D.; Wadden, T.A.; Swain, R.M.; Stunkard, A.J.; Platte, P.; Vogt, R.A. The Eating Inventory in obese
women: Clinical correlates and relationship to weight loss. Int. J. Obes. 1998, 22, 778–785. [CrossRef]
28. Spadaro, K.C.; Davis, K.K.; Sereika, S.M.; Gibbs, B.B.; Jakicic, J.M.; Cohen, S.M. Effect of mindfulness
meditation on short-term weight loss and eating behaviors in overweight and obese adults: A randomized
controlled trial. J. Complement. Integr Med. 2017, 15. [CrossRef]
29. Hansen, T.T.; Hjorth, M.F.; Sandby, K.; Andersen, S.V.; Astrup, A.; Ritz, C.; Bullo, M.; Camacho-Barcia, M.L.;
Garcia-Gavilan, J.F.; Salas-Salvado, J.; et al. Predictors of successful weight loss with relative maintenance of
fat-free mass in individuals with overweight and obesity on an 8-week low-energy diet. Br. J. Nutr. 2019,
122, 468–479. [CrossRef]
30. Syed-Abdul, M.M.; Hu, Q.; Jacome-Sosa, M.; Padilla, J.; Manrique-Acevedo, C.; Heimowitz, C.; Parks, E.J.
Effect of carbohydrate restriction-induced weight loss on aortic pulse wave velocity in overweight men and
women. Appl. Physiol. Nutr. Metab. 2018, 43, 1247–1256. [CrossRef]
31. Ahren, B.; Larsson, H. Quantification of insulin secretion in relation to insulin sensitivity in nondiabetic
postmenopausal women. Diabetes 2002, 51 (Suppl. 1), S202–S211. [CrossRef] [PubMed]
32. White, M.A.; Whisenhunt, B.L.; Williamson, D.A.; Greenway, F.L.; Netemeyer, R.G. Development and
validation of the food-craving inventory. Obes. Res. 2002, 10, 107–114. [CrossRef] [PubMed]
33. Gibson, E.L.; Desmond, E. Chocolate craving and hunger state: Implications for the acquisition and expression
of appetite and food choice. Appetite 1999, 32, 219–240. [CrossRef] [PubMed]
34. Polivy, J.; Coleman, J.; Herman, C.P. The effect of deprivation on food cravings and eating behavior in
restrained and unrestrained eaters. Int. J. Eat. Disord. 2005, 38, 301–309. [CrossRef]
35. Coelho, J.S.; Polivy, J.; Herman, C.P. Selective carbohydrate or protein restriction: Effects on subsequent food
intake and cravings. Appetite 2006, 47, 352–360. [CrossRef]
36. Williamson, D.A.; Martin, C.K.; Anton, S.D.; York-Crowe, E.; Han, H.; Redman, L.; Ravussin, E. Is caloric
restriction associated with development of eating-disorder symptoms? Results from the CALERIE trial.
Health Psychol. 2008, 27, S32–S42. [CrossRef]
37. Smithson, E.F.; Hill, A.J. It is not how much you crave but what you do with it that counts: Behavioural
responses to food craving during weight management. Eur. J. Clin. Nutr. 2016, 71, 625. [CrossRef]
38. MacLean, P.S.; Wing, R.R.; Davidson, T.; Epstein, L.; Goodpaster, B.; Hall, K.D.; Levin, B.E.; Perri, M.G.;
Rolls, B.J.; Rosenbaum, M.; et al. NIH working group report: Innovative research to improve maintenance of
weight loss. Obesity 2015, 23, 7–15. [CrossRef]
39. Sacks, F.M.; Bray, G.A.; Carey, V.J.; Smith, S.R.; Ryan, D.H.; Anton, S.D.; McManus, K.; Champagne, C.M.;
Bishop, L.M.; Laranjo, N.; et al. Comparison of weight-loss diets with different compositions of fat, protein,
and carbohydrates. N. Engl. J. Med. 2009, 360, 859–873. [CrossRef]
40. Cohen, J. A power primer. Psychol. Bull. 1992, 112, 155–159. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).