International Journal of Obesity (2007) 31, 1688–1695
& 2007 Nature Publishing Group All rights reserved 0307-0565/07 $30.00
www.nature.com/ijo
ORIGINAL ARTICLE
Effects of food form on appetite and energy intake in
lean and obese young adults
DM Mourao1, J Bressan2, WW Campbell1 and RD Mattes1
1
Department of Foods and Nutrition, Purdue University, West Lafayette, IN, USA and 2Universidade Federal de Vicosa,
Vicosa, Brazil
Objective: To investigate the independent effect of food form on appetite and energy intake in lean and obese adults using high
carbohydrate, fat or protein food stimuli.
Design: Crossover dietary challenge with matched beverage and solid food forms: high carbohydrate (watermelon and
watermelon juice); high protein (cheese and milk); high fat (coconut meat and coconut milk).
Subjects: A total of 120 lean (18–23 kg/m2; N ¼ 60) and obese (30–35 kg/m2; N ¼ 60) adults (18–50 years old) with stable body
weight. Forty different participants (N ¼ 20 lean and 20 obese) were tested with each of the food systems.
Measurements: Appetitive sensations, food palatability and dietary intake.
Results: Regardless of the predominant energy source, the beverage food form elicited a weaker compensatory dietary response
than the matched solid food form. Thus, total daily energy intake was significantly higher by 12.4, 19 and 15% on days the
beverage forms of the high-carbohydrate, -fat and -protein foods were ingested, respectively. This was due more to a weak
effect on satiety than satiation. The obese participants had higher energy intake at the lunch, including the beverage highprotein load, but overall differences between lean and obese participants were small and not systematic.
Conclusion: Food rheology exerts an independent effect on energy intake. Dietary compensation for beverages is weaker than
for solid food forms of comparable nutrient content. Thus, they pose a greater risk for promoting positive energy balance.
International Journal of Obesity (2007) 31, 1688–1695; doi:10.1038/sj.ijo.0803667; published online 19 June 2007
Keywords: viscosity; beverage; energy intake; appetite; feeding; human
Introduction
Accumulating evidence indicates that energy-yielding beverages evoke weaker appetitive and compensatory dietary
responses than energy-matched food challenges in solid
form.1–7 Further, there is increasing documentation of a
positive association between beverage consumption and
body weight or body mass index (BMI).4,8–13 The mechanisms by which beverages and solid food forms elicit
differential appetitive and dietary responses are not known.
Studies comparing responses to different beverage and
solid foods (for example, fruit juice vs cheese and crackers,14
or soda vs cookies15) have yielded mixed findings. To isolate
the independent effect of food rheology on appetitive and
dietary responses, it is essential to hold other attributes
constant. One study applied this approach to contrast
Correspondence: Dr RD Mattes, Professor of Foods and Nutrition, Department
of Foods and Nutrition, Purdue University, 212 Stone Hall, 700 W State Street,
West Lafayette, IN 47907-2059, USA. E-mail: [email protected]
Received 11 December 2006; revised 7 April 2007; accepted 8 May 2007;
published online 19 June 2007
responses to a solid casserole and soup,16 but responses to
soups and beverages differ markedly – soup has higher satiety
value.17,18 An aim of this study was to contrast appetitive
and acute compensatory dietary responses to solid foods and
beverages closely matched on all but rheological properties.
Macronutrient-specific effects on satiety and satiation are
well documented.19–21 Most, although not all, studies of
these effects employed solid food vehicles. Less is known
about the existence or magnitude of differential influences
in beverages. Some findings suggest that they may not
hold.22,23 While there may be differences in nutrient
composition and energy density among beverages that
mitigate viewing all beverages similarly, from an energy
balance perspective, distinctions may not be valid. The
principal source of energy from beverages in the US diet over
the past two decades has derived from carbohydrate,24 but
there is increasing popularity of beverages with different
primary energy sources (for example, specialty coffees with
high fat content and performance enhancing beverages
containing high protein concentrations). Whether, and to
what extent, these items pose a particular challenge to
energy balance requires assessment. A second aim of this
Effects of food form on appetite and energy intake
DM Mourao et al
1689
study was to compare solid and beverage forms of foods
comprised primarily of carbohydrate, fat or protein. The
intent was to document the independent effect of food form
in foods with different dominant macronutrient sources, not
to contrast the relative magnitude of responses across the
foods. Isolation of macronutrient effects would have
required matching the items on all other relevant attributes
(for example, energy density, osmolality) and this was not
possible in the present context.
The obese may be at particular risk for positive energy
balance due to beverage consumption. Reduction of beverage intake leads to a more robust loss of body weight in this
group.25 However, the role of responsiveness to a beverage
medium has not been explored as a mechanism. Responses
of lean and obese people to comparable beverage and solid
foods were contrasted in this trial.
Methods
Participants were instructed to fast for 10 h overnight, eat
their typical breakfast (consisting of the same meal for each
of the 3 test days) and fast again for at least 3 h before
reporting to the lunch appointment. A finger stick glucose
measurement was taken to verify the pre-test fast was
maintained (values o110 mg/dl). Visits were scheduled
during their customary lunchtime. On arrival at the
laboratory, participants completed a motor skills test and
an appetite questionnaire on a personal digital assistant
(PDA). Participants were then provided a meal, consisting of
chicken sandwiches and water for the control session or the
sandwiches and a solid food or beverage sample on the other
2 test days. Meals were consumed in their entirety in
individual booths within 20 min. Participants were provided
two sandwich sections to begin the trial and were permitted
to reach behind a partition to obtain as many additional
sandwich sections as they desired to reach a comfortable
level of fullness. They were unaware of how many sandwich
sections were behind the partition.26 Minimizing the risk of
participants passing a level of comfortable fullness, the test
food was divided into thirds. On the days that samples were
provided, participants were instructed to eat the first twothirds of the test foods, to evaluate their level of fullness, and
then to decide if eating the sandwiches would interfere with
finishing the sample and going beyond a comfortable level
of fullness. Samples were rated for palatability. After finishing their lunch, participants again completed the motor
skills test and the appetite questionnaire. Before leaving the
lab, participants were instructed to keep a food record of
each eating or drinking occasion until going to sleep for the
night. At the initial session, they were instructed in portion
size estimation using NASCO food models (Fort Atkinson,
WI, USA) and true-size pictures with a PowerPoint presentation. Additionally, the participants were requested to
complete the appetite questionnaire on the PDA each hour
and draw a single geometric form in the PDA every other
hour until going to sleep for the night. Participants were
informed of the researcher’s interest in the association
between appetite and fine motor control as a means to deemphasize the focus on ingestive behavior and thereby
minimize atypical feeding behavior or biased dietary reporting. All participants completed and signed an informed
consent form approved by the Purdue University Institutional Review Board and received monetary compensation.
Participants
Study eligibility was established through completion of a
screening questionnaire eliciting health and demographic
information as well as anthropometric measurements.
Participants meeting eligibility criteria, consisting of a BMI
between 18–23 or 30–35 kg/m2, 18–50 years of age, stable
body weight (p3 kg change within the past 3 months), not
taking medication known to influence appetite, self-reported
regular consumer of breakfast and lunch, non-restrained
eater (score o14 on the three factor eating questionnaire27),
and nonsmoker were recruited by public advertisement.
They were divided into two groups, lean and obese, for three
dietary interventions based on preloads of foods providing
energy primarily as protein (N ¼ 40, 20 lean and 20 obese),
carbohydrate (N ¼ 40, 20 lean and 20 obese) and fat (N ¼ 40,
20 lean and 20 obese). Characteristics of these groups are
presented in Table 1.
Meals/loads
The participants were presented a meal consisting of chicken
sandwiches that were provided in excess and were ingested
ad libitum. However, only two sandwich sections were visible
at any time.
Table 1 Means7s.d. age and anthropometric characteristics of the
participants
Height (m)
Weight (kg)
BMI (kg/m2)
Age (years)
1.6270.04A
1.6870.06A
1.7770.05B
1.7770.07B
54.4572.88A
89.1077.00B
66.3175.93C
97.2378.30D
20.7970.65A
31.6971.16B
21.0971.15A
31.1571.19B
23.1073.38A
24.6074.64A
20.0071.40B
25.2073.88A
Carbohydrate (n ¼ 40)
Female/lean 1.6070.04A 50.4673.55A
Female/obese 1.6270.04A 84.5775.25B
Male/lean
1.7670.02B 68.2072.16C
Male/obese
1.8070.02B 101.2374.69D
19.7970.89A
32.0971.90B
21.9270.62A
31.2671.04B
24.0073.09B
30.9078.64C
21.8071.78A
24.7074.95B
Fat (n ¼ 40)
Female/lean
Female/obese
Male/lean
Male/obese
21.5070.81A
32.2171.17B
22.4370.75A
32.4171.62B
21.6071.43A
32.0271.61C
20.6071.12A
25.1076.56B
Protein (n ¼ 40)
Female/lean
Female/obese
Male/lean
Male/obese
1.6270.06A 57.8073.51A
1.5971.63A 84.0171.61B
1.7970.08B 72.5478.36C
1.8070.04B 103.7677.42D
Different letters within a column are significantly different, for the same
macronutrient, F-test (Po0.05).
International Journal of Obesity
Effects of food form on appetite and energy intake
DM Mourao et al
1690
Table 2
Composition of the standard sandwich consumed with every test meal and the liquid and solid test foods
Chicken
Milk
Cheese
Sandwichesa
Energy load (kcal)
Weight (g)
Volume (ml)
Carbohydrate (g)
Fat (g)
Protein (g)
Energy content (kcal)
Fiber content (g)
Viscosity (cps)
Hardness (g)
22
F
4.5
2.6
3.05
54
0.59
F
F
125
432
430
5.4
0
21.4
107
0
26.27
F
225
772
770
9.6
0
38.5
192
0
F
125
225
70
118
360b
650b
4.7
10.3
0
0
24.5
42.7
117
213
1.1
2
263.9
F
199.9
Watermelon
Watermelon
Coconut
Coconut
Juice
Fruit
Milk
Meat
125
400
390
28.7
1.7
2.5
140
o1.6
171.35
F
225
720
710
51.7
32
4.5
254
o2.9
F
125
400
390
28.7
1.7
2.5
140
1.6
225
720
710
51.7
3.2
4.5
254
2.9
107.5
125
79
83c
2.1
11.4
1.04
115
0
225
148
150c
3.75
20.6
1.9
208
0
F
125
35
150b
5
11.7
1.16
130
3.1
225
63
240b
9.1
21.1
2.1
235
5.6
1125.6
The 125 kcal version was provided to the lean participants and the 225 version was provided to the obese participants. aTyson premium chunk chicken breast (12 g)
with mayonnaise and 10 g of Iron Kids Crustless bread, a total of 22 g per sandwich. bVolume of water consumed with the solid treatment in order to match the
volume of the liquid. cCoconut milk was diluted with water (1:1), being the final volume 166 and 300 ml respectively for the lean and obese.
The test food samples consisted of solid and beverage
forms with one predominant macronutrient (Table 2). For
the protein samples, a fat-free, low carbohydrate milk (Carb
Countdown Fat Free Dairy Beverage, Hood, Chelsea, MA,
USA) and cheese fortified with whey to a concentration
comparable to the milk were consumed (fat free Mozzarella,
Kraft Foods North America Inc., Rye Brook, NY, USA). For the
carbohydrate samples watermelon fruit and watermelon
juice were consumed. The fat sample consisted of coconut
milk (A Taste of Thai Andre Prost Inc, Old Saybrook, CT, USA)
and coconut meat (fresh). Water (1:1) and a sweetener (1 tsp
of Equals Sweetener, Chicago, IL, Merisant Co.) were added
to the coconut milk to match the sweetness of the meat (this
formulation was developed and pilot tested before the
study). To match the beverage and solid sample volumes,
participants were required to drink additional water when
consuming the samples of coconut meat and cheese.
Participants in the BMI range of 18–23 kg/m2 received a
125 kcal load and those between 30–35 kg/m2 received a
225 kcal load. These energy levels were selected to provide
the two groups roughly comparable metabolic challenges
(that is, 5–10% of estimated energy requirement) and
volumes.
Anthropometric assessment
BMI was assessed by measurement of height and weight.
Height was measured (70.1 cm) on a wall-mounted stadiometer. Body weight was measured (70.1 kg) using calibrated scales with participants wearing no shoes and a light
gown. Participants were categorized as normal weight
(BMI ¼ 18–23 kg/m2) or obese (BMI ¼ 30–35 kg/m2).
Dietary analyses and appetite/sensory assessments
Food records were kept each testing day to permit determination of energy intake and dietary compensation. They
International Journal of Obesity
were analyzed using version 7.6 of the Food Processor
nutrient database (ESHA, Research, Salem, OR). Satiation
was estimated according to the discretionary energy consumed at lunch. Satiety was estimated in two ways; by energy
consumed at the first meal after lunch, and by time (minutes
between completion of lunch and onset of next eating or
drinking occasion greater than or equal to 150 kcal).
Percentage dietary energy compensation was calculated as
[((energy intake in the absence of a preload)(actual intake
on test daysexperimental load)/energy intake in the
absence of a preload) 100)]. The baseline energy values
were for the midday meal or daily intake depending on the
time frame of interest. Appetite ratings were recorded by
participants on visual analog scales on PDA’s, before and
after consuming lunch, and each hour after leaving the lab.
Questions included ‘How hungry do you feel right now?
How full are you right now? How strong is your desire to eat
right now? How strong is your feeling of thirst right now?
How strong is your desire to eat something salty right now?
How strong is your desire to eat something sweet right now?
How strong is your desire to eat something fatty right
now? ‘The scales were anchored with ‘not at ally/very weak’
and ‘as yI’ve ever felt/very strong’. The validity of this
approach has been established.28,29 Food sample palatability
was rated on a visual analog scale with end anchors of ‘dislike
extremely’ and ‘like extremely’ after sampling each item.
Statistical analysis
Repeated measures analyses of variance with one withinsubject factor (meal form – beverage and solid) and one
between-subject factor (lean vs obese) were used to assess
responses to the intervention foods in each macronutrient
pair. The study was not powered for comparisons between
food products, nor would they reveal effects unequivocally
attributable to the dominant macronutrient, because the
foods varied on numerous additional attributes. The criter-
Effects of food form on appetite and energy intake
DM Mourao et al
1691
ion for statistical significance was Po0.05, two-tailed. The
dependent variables were discretionary energy consumed at
lunch, energy consumed after lunch and the sum of these
estimates, referred to as post-breakfast energy intake, satiety
(that is, energy content of the first post-lunch meal and
interval between lunch and first eating occasion of
4150 kcal) and each appetitive index. Power calculations
indicated a sample size of 20 participants per group (lean vs
obese), would permit detection of between-group treatment
effects equal to a standardized difference of 1 with 80%
power. The Statistical Package for the Social Science (SPSS)
version 12.0 was used for all analyses.
Results
Appetite and hedonic ratings
With the exception of an isolated marginal effect of the
beverage protein load being associated with elevated desire
to eat something salty relative to the solid load, no
significant effects of any test food form (Table 3) or BMI
were observed for the mean daily appetitive ratings.
Mean hedonic ratings of the samples are presented in
Table 4. All samples were rated as neutral or better (that is,
43.0); however, the solid high CHO (Po0.01) and fat
(Po0.05) foods were rated higher than their beverage forms.
There was no significant difference between the cheese and
milk.
Table 3
Energy intake
High-protein foods. Post-breakfast energy intake and grams
of fat consumed after lunch were higher with the beverage
load compared to the solid (Table 5). Dietary compensation
was significantly higher with the solid load compared to the
beverage version for the post-breakfast energy intake (Table 5). The score was negative for the beverage load
indicating slight reverse compensation (that is, failing to
compensate for any portion of the beverage energy and
actually eating more of the customary diet than baseline).
The score was over 100% for the solid load, indicating the
reduction of energy intake over the day exceeded the energy
content of the load itself. There was a significant treatment
by BMI interaction for energy consumed at lunch (Table 5).
Obese participants consumed more energy at lunch than the
lean (577 vs 436 kcal) with the beverage load (P ¼ 0.011), but
their consumption did not differ from the lean participants
with the solid load (470 vs 408 kcal). The obese ate more with
the beverage load compared to the solid (577 vs 471 kcal)
(Po0.0001). The latter was not true for the lean participants
(436 vs 408 kcal) (P ¼ 0.318). Energy consumed after lunch
and the intermeal interval (IMI; an index of satiety) were not
significantly different between food loads or between the
lean and obese groups.
High-carbohydrate foods. Post-breakfast energy intake and
energy consumed after lunch were significantly higher with
the beverage load compared to the solid (Table 6). Dietary
compensation over the day was also higher with the solid
Daily appetite rating during the protein, carbohydrate and fat trials
Table 4 Hedonic ratings for the beverage and solid protein, carbohydrate
and fat beverage and solid foods
Beverage
Solid
P
Protein
Hunger
Fullness
Desire to eat
Desire for SALT food
Desire for SWEET food
Desire for High FAT food
Thirst
23.571.4
42.671.7
22.171.5
13.971.7
17.271.8
12.171.7
27.771.6
23.171.6
40.671.8
21.971.7
12.471.5
17.871.8
12.071.5
26.371.6
0.739
0.205
0.862
0.046
0.655
0.949
0.273
Carbohydrate
Hunger
Fullness
Desire to eat
Desire for SALT food
Desire for SWEET food
Desire for FAT food
Thirst
21.971.4
43.972.1
21.671.6
16.471.8
17.671.5
13.571.4
31.972.1
22.871.5
41.671.8
22.871.6
17.171.5
17.171.6
14.871.5
33.172.1
0.521
0.215
0.441
0.620
0.770
0.362
0.455
Table 5 Dietary energy consumed, percent compensation and satiety
(intermeal interval) with the high-protein beverage and solid loads (N ¼ 20
obese and 20 lean)
Fat
Hunger
Fullness
Desire to eat
Desire for SALT food
Desire for SWEET food
Desire for FAT food
Thirst
24.071.3
42.771.9
24.471.6
15.971.5
20.972.1
15.571.8
34.272.2
24.371.7
40.772.0
24.371.7
15.171.5
23.572.6
15.671.8
35.172.3
0.882
0.306
0.939
0.375
0.133
0.973
0.584
Energy at lunch (kcal)
After lunch energy (kcal)
Post-breakfast energy intake (kcal)
Intermeal interval (minutes)
Protein after lunch (g)
Carbohydrate after lunch (g)
Fat after lunch (g)
Dietary compensation (lunch) (%)
Dietary compensation (total day) (%)
Values are percent difference from the low anchor of a visual analog scale.
Means7s.e.
Loads
Beverage
Solid
P
Protein (milk and cheese)
Carbohydrate (watermelon)
Fat (coconut)
5.370.4
5.670.4
4.470.6
4.570.4
7.770.3
5.870.6
0.16
o0.001
o0.05
Ratings were obtained on a 9-point category scale (1-like extremely; 9-dislike
extremely). Differences between beverage and solid forms were compared by
t-test. Means7s.e.
Variables
Beverage
Solid
TRT BMI*
14087122
1225794
19157129
16657101
229718
249726
4973
4574
171717
166714
5575
4374
579
2278
547102
127758
P
0.035
0.107
0.030
0.483
0.396
0.763
0.042
0.003
0.029
Means7s.e., Significance level applies to values in a row.
International Journal of Obesity
Effects of food form on appetite and energy intake
DM Mourao et al
1692
Table 6 Means and standard errors of energy consumed, percent compensation and satiety (intermeal interval) with the high carbohydrate beverage
and solid loads (N ¼ 20 obese and 20 lean)
Table 7 Means and standard errors of energy consumed, percent compensation and satiety (intermeal interval) with the high fat beverage and solid
loads (N ¼ 20 obese and 20 lean)
Variables
Variables
Energy at lunch (kcal)
After lunch energy (kcal)
Post-breakfast energy intake (kcal)
Intermeal interval (minutes)
Protein after lunch (g)
Carbohydrate after lunch (g)
Fat after lunch (g)
Dietary compensation (lunch) (%)
Dietary compensation (total days) (%)
Beverage
Solid
TRT BMI*
1498783
1305793
1970798
1752798
259720
226717
5674
4773
183713
168714
5975
4974
8777
152752
P
0.027
0.044
0.026
0.217
0.084
0.245
0.152
0.010
0.006
Energy at lunch (kcal)
After lunch energy (kcal)
Post-breakfast energy intake (kcal)
Intermeal interval (minutes)
Protein after lunch (g)
Carbohydrate after lunch (g)
Fat after lunch (g)
Dietary compensation (lunch) (%)
Dietary compensation (total day)
Beverage
Solid
P
565731
19507169
25177180
223718
8178
243726
6977
0.975
179797
530732
1585799
21167110
226721
6275
190714
6276
977
42748
0.100
0.026
0.016
0.901
0.068
0.026
0.371
0.147
0.016
Means7s.e., Significance level applies to values in a row.
Means7s.e., Significance level applies to values in a row.
load. It reflected a small reverse daily compensation with the
beverage load and overcompensation with the solid load.
There was a significant treatment by BMI interaction for
percentage dietary energy compensation at lunch. Obese
participants compensated more (P ¼ 0.016) with the beverage load (46 vs 6%) compared to lean, but this did not hold
for the solid treatment (P ¼ 0.284). The lean participants
compensated more completely with the solid load
(P ¼ 0.001) compared to the beverage load (24 vs 6%). There
was a significant treatment by BMI interaction for energy
consumed at lunch. The lean participants consumed more
energy at lunch with the beverage load, compared with solid
(519 vs 460 kcal) (P ¼ 0.008), but this did not hold for the
obese participants (P ¼ 0.643). The obese and lean participants did not differ in energy consumed at lunch with the
beverage (P ¼ 0.321) or solid (P ¼ 0.794) loads. No difference
in post-lunch eating interval was detected.
obesity.7 This has prompted recommendations to moderate
their level of ingestion.30,31 Some have targeted these
recommendations to particular beverages, but from an
energy balance perspective, the rationale for this approach
has not been substantiated. Others have challenged the data
specifically linking beverage ingestion to BMI, primarily on
the grounds that there is a lack of a plausible biological basis
for such an association.32 There is a need to identify and
characterize the properties of beverages that may contribute
to the reported differential responses they elicit relative to
solid foods to substantiate this view and to develop
appropriate guidelines for beverage consumption. To this
end, the primary hypothesis addressed in this work was that
energy-yielding beverages exert weaker appetitive and compensatory dietary responses than the same foods ingested in
solid form. Further, these effects were explored in foods
varying in the predominant energy source to determine the
relative importance of the energy source and medium in
these outcomes. Finally, the responses of lean and obese
individuals to challenges with beverages and solid foods
were contrasted to gain insights on whether the obese are at
greater risk for positive energy balance due to beverage
consumption. The key feature of this work was the isolation
of the rheological property of the test foods as the
independent variable.
No evidence was obtained that the beverage and solid
forms of the same foods elicited different appetitive
responses. This is consistent with findings from several
trials,15,33 but that work used foods with different nutritional
composition, thereby hampering conclusions about the role
of rheology. Others have noted weaker appetitive responses
for fruit juices relative the whole fruit,34,35 but here again,
there are compositional differences, such as higher fiber in
the solid forms. Even a small addition of fiber to a beverage
that increases its viscosity augments reported hunger reduction.6 Some work indicates soups, a beverage food form, are
actually more satiating than solids, but this appears to be a
unique property of this food (likely the cognitive expectations it elicits36), so of limited relevance to a consideration of
beverage effects.18 Clearly, there is some structural difference
High-fat foods. Significantly, more energy was consumed
after lunch (P ¼ 0.026) and post-breakfast (P ¼ 0.016) with
the beverage load compared to the solid (Table 7). There was
no difference of energy consumed at lunch between treatments. Post-lunch carbohydrate energy consumed was higher after the beverage load compared to solid, but there were
no differences for protein and fat consumption. Dietary
compensation was significantly higher with solid load
compared to the beverage load over the day (P ¼ 0.016).
The beverage load led to marked reverse compensation,
while the solid load resulted in partial compensation. No
difference of compensation was observed at lunch. The
intermeal interval was not significantly different between
loads or between the lean and obese groups.
Discussion
A strong body of evidence indicates that energy-yielding
beverages are contributing to the positive energy balance
and increasing incidence and prevalence of overweight/
International Journal of Obesity
Effects of food form on appetite and energy intake
DM Mourao et al
1693
between beverages and solid foods. The present results
suggest that disruption of that structure (for example, by
blending), without changing composition, does not exert a
marked effect on appetite, as measured by questionnaire.
Although it is commonly assumed that appetitive sensations serve to link energy need with energy intake, this is not
reliably observed.17 The present data are a case in point
because, despite the lack of a differential appetitive response
to the beverage and solid food forms, there was a strong and
consistent difference in the compensatory dietary response
they elicited. This finding is not surprising given the
multitude of factors that independently influence appetite
and feeding.
With all three foods, daily energy intake was significantly
greater when the beverage form was ingested compared to
the solid. Indeed, the compensation scores were negative for
all three beverages, indicating they resulted in reverse
compensation. That is, daily energy intake was actually
greater than baseline by an amount that exceeded the energy
contributed by the beverage. In contrast, the compensation
scores for the solid forms were all positive. They varied from
partial to over compensation (that is, intake lower than
baseline). The compensation score for the high-fat food was
the lowest, consistent with other reports;37 however, given
that only three foods were tested, the generality of this
observation is uncertain. The energy intake differences and
compensation scores at lunch were less consistent and
smaller in magnitude. This suggests the effects of rheology
are more pronounced on satiety than satiation. This was not
reflected in the interval between lunch and the next eating
occasion, as this was not different for the beverage and solid
food forms. Thus, the difference may lie in the energy
content of subsequent eating events. Only small and
inconsistent differences in post-lunch macronutrient intake
were noted between the two food forms, suggesting the
compensation was attributable to altered intake of the
general diet. Others have also reported a lack of macronutrient compensation.38
Differential macronutrient effects on appetite and dietary
compensation have been reported such that protein4carbohydrate4fat.39 Generally, this stems from trials involving
solid foods. There are suggestions that these discrepancies
are less reliable in beverages. In particular, the reported highsatiety value of protein and its suppressive effect on energy
intake often is not observed when delivered in beverage
form.22,23 The three foods tested here derived most of their
energy from a different macronutrient and they were not
closely matched on other nutrients. Thus, contrasting
responses across foods would be of limited value. However,
in the beverages, none of the macronutrients prompted any
offsetting adjustment in subsequent energy intake. Postbreakfast energy intake was significantly higher than baseline in each case. This raises questions about the likely
benefit of substituting one form of beverage for another (for
example, 100% fruit juice or milk for soda) with respect to
energy balance. There may be some differences in nutrient
density between beverages that are nutritionally relevant
when making public health recommendations, but if the
focus is on energy balance, it is not clear that distinctions are
appropriate.
There is a long history of research on potential differences
in behavioral40 and metabolic41 responses to foods in the
lean and obese that may account for their phenotypes.
Evidence that the increasing incidence of obesity stems from
a disproportionate increase in BMI by heavier individuals
than the lean42 suggests that there may be BMI-related
differences in responses to food cues. This study provides
new data on the appetitive and dietary responses of lean and
obese individuals to energy provided in different rheological
forms. The loads differed in energy content so that comparable metabolic challenges were present to the two groups. No
BMI differences were observed for appetitive ratings. Differences in intake were small and inconsistent. The obese
consumed more energy at the lunch that included the
beverage high-protein load, while there was no difference
with the solid load. However, there was no difference in
energy consumed at lunch between the BMI groups when
the high-carbohydrate or high-fat loads were presented.
Others43 have also reported little BMI-related difference in
dietary responses to beverage and solid loads, although the
stimuli used previously (flavored milk and an angel food
cake with jam) differed on many dimensions. Thus, these
data do not support a view that beverage consumption is
more likely to lead to positive energy balance in the obese.
The present findings should not be over-interpreted. This
study involved acute testing of a relatively small sample of
participants. The sample size limits extrapolation of the
findings, but not the study power, because significant effects
were clearly documented. Data are presented at a meal and
over a single day. The latter observations are emphasized as
they provide greater opportunity for compensatory responses to manifest, but even daily intake is highly variable
and additional compensation may occur over days and
weeks.44 However, there are data indicating a reduction of
energy from beverages results in weight loss45,10,25 and
addition of energy-yielding beverages leads to weight
gain.8,46,12 Both of these sets of observations are consistent
with the present acute, but more tightly controlled tests.
Another limitation is that only three foods were tested. They
were selected on the basis of their primary energy source and
potential for modification of form. They do not represent the
universe of foods and many facets of products, ranging from
the cognitive impression they impart17 to the bioaccessibility of the energy they contain,47 may modulate the
responses to and consequences of ingesting foods of varying
rheology. Still, accumulating evidence indicates this food
attribute holds nutritional consequence. While efforts were
made to equate the foods on palatability, this was not
accomplished. The solid forms of the high-carbohydrate and
high-fat foods were rated higher than their beverage version.
This was probably attributable to the novelty of these
beverage forms; the solid and beverage forms of the highInternational Journal of Obesity
Effects of food form on appetite and energy intake
DM Mourao et al
1694
protein food are commonly consumed. As palatability is
generally assumed to be directly related to intake,48,49 it
would be predicted that these solid forms would lead to
greater intake than the beverage versions. However, the
opposite was observed. Thus, it would not appear that this
discrepancy accounts for the study findings.
With documentation of this effect of food form, the
question turns to possible mechanisms. Feeding is guided by
environmental and physiological (for example, cognitive,
orosensory, digestive, metabolic, endocrine, neural) influences. Differential responses to beverages and solid food
forms may be posited at each level. Environmentally, portion
sizes of beverages have increased markedly,50,51 they are
among the least expensive sources of energy and are a meal
component that is provided in unlimited quantity in most
commercial restaurants. Beverages have lower expected
satiety value, lower demand for oral processing, shorter
gastrointestinal transit times and the energy they contain
has greater bioaccessibility and bioavailability. Each of these
attributes has been associated with weaker effects on appetite
and dietary compensation.52–59 The absolute and relative
importance of these properties, and others, has not been
established, but warrant exploration.
In summary, the present trial supports an independent
effect of food rheology on energy intake. The inclusion of a
caloric beverage in a lunch meal led to greater daily energy
intake compared to customary intake or days, where a solid
version of the same food was ingested. This occurred
regardless of the primary energy source. There was no clear
indication that the lean and obese differ in this regard.
Efforts to moderate energy intake should consider the
contribution of beverages of all types.
Acknowledgements
This work was supported by PHS grant no. 1 R01 DK 063185
and CNPq, Coordenac- ão de Aperfeic- oamento de Pessoal de
Nı́vel Superior (CAPES), Brası́lia/Brazil.
References
1 Crapo PA, Henry RR. Postprandial metabolic responses to the
influence of food form. Am J Clin Nutr 1998; 48: 560–564.
2 Tournier A, Louis-Sylvestre J. Effect of the physical state of a food
on subsequent intake in human subjects. Appetite 1991; 16:
17–24.
3 Hulshof T, De Graaf C, Weststrate JA. The effects of preloads
varying in physical state and fat content on satiety and energy
intake. Appetite 1993; 21: 273–286.
4 DiMeglio DP, Mattes RD. Liquid versus solid carbohydrate: effects
on food intake and body weight. Int J Obes Relat Metab Disord
2000; 24: 794–800.
5 Peracchi M, Santangelo A, Conte D, Fraquelli M, Tagliabue R,
Gebbia C et al. The physical state of a meal affects hormone
release and postprandial thermogenesis. Br J Nutr 2000; 83:
623–628.
International Journal of Obesity
6 Mattes RD, Rothacker D. Beverage viscosity is inversely related to
postprandial hunger in humans. Physiol Behav 2001; 74: 551–557.
7 Mattes RD. Beverages and positive energy balance: the menace is
the medium. Intl J Obes 2006a; 30: S60–S65.
8 Ludwig DS, Peterson KE, Gortmaker SL. Relation between
consumption of sugar-sweetened drinks and childhood obesity:
a prospective, observational analysis. Lancet 2001; 357: 505–508.
9 WHO Diet, nutrition and the prevention of chronic diseases.
World Health Organ Tech Rep Ser 2006; 91, i–viii, 1–149.
10 James J, Thomas P, Cavan D, Kerr D. Preventing childhood
obesity by reducing consumption of carbonated drinks: cluster
randomized controlled trial. Br Med J 2004; 328: 1237.
11 McConahy KL, Smiciklas-Wright H, Mitchell DC, Picciano MF.
Portion size of common foods predicts energy intake among
preschool-aged children. J Am Diet Assoc 2004; 104: 975–979.
12 Striegel-Moore RH, Thompson D, Affenito SG, Franko DL,
Obarzanek E, Barton BA et al. Correlates of beverage intake in
adolescent girls: The National Heart, Lung, and Blood Institute
Growth and Health Study. J Pediatr 2006; 148: 183–187.
13 Malik VS, Schulze MB, Hu FB. Intake of sugar-sweetened
beverages and weight gain: a systematic review. Am J Clin Nutr
2006; 84: 274–288.
14 Kissileff HR. Effects of physical state (liquid-solid) of foods on
food intake: procedural and substantive contributions. Am J Clin
Nutr 1985; 42: 956–965.
15 Almiron-Roig E, Drewnowski A. Hunger, thirst, and energy
intakes following consumption of caloric beverages. Physiol Behav
2003; 79: 767–773.
16 Rolls BJ, Bell EA, Thorwart ML. Water incorporated into a food
but not served with a food decreases energy intake in lean
women. Am J Clin Nutr 1999; 70: 448–455.
17 Mattes RD, Hollis J, Hayes D, Stunkard AJ. Appetite: measurement
and manipulation misgivings. J Am Diet Assoc 2005; 105:
S87–S97.
18 Mattes RD. Fluid calories and energy balance: the good, the bad,
and the uncertain. Physiol Behav 2006b; 89: 66–70.
19 Stubbs RJ, van Wyk MCW, Johnstone AM, Harbron CG. Breakfasts
high in protein, fat or carbohydrate: effect on within-day appetite
and energy balance. Eur J Clin Nutr 1995; 50: 409–417.
20 Westerterp-Plantenga MS. Eating behavior in humans, characterized by cumulative food intake curves – a review. Neurosci
Biobehav Rev 2000; 24: 239–248.
21 Fischer K, Colombani PC, Wenk C. Metabolic and cognitive
coefficients in the development of hunger sensations after
pure macronutrient ingestion in the morning. Appetite 2004;
42: 49–61.
22 Hill AJ. The effect of dietary composition on hunger, satiation
and satiety. In: Guy-Grand and Ailhaud G (eds). Progress in Obesity
Research: 8. 1999/8th International Congress on Obesity. John Libbey
& Co.: London., 1999. pp 209–213.
23 Rumpler WV, Kramer M, Rhodes DG, Paul DR. The impact of the
covert manipulation of macronutrient intake on energy intake
and the variability in daily food intake in nonobese men. Intl J
Obes 2006; 30: 774–781.
24 Johnson RK, Frary C. Choose beverages and foods to moderate
your intake of sugars: the 2000 dietary guidelines for Americans –
what’s all the fuss about? J Nutr 2001; 131: 2766S–2771S.
25 Ebbeling CB, Feldman HA, Osganian SK, Chomitz VR, Ellenbogen
SJ, Ludwig DS. Effects of decreasing sugar-sweetened beverage
consumption on body weight in adolescents: a randomized,
controlled pilot study. Pediatrics 2006; 17: 673–680.
26 Rolls BJ, Morris EL, Roe LS. Portion size of food affects energy
intake in normal-weight and overweight men and women. Am J
Clin Nutr 2002; 76: 1207–1213.
27 Stunkard AJ, Messick S. The three-factor eating questionnaire to
measure dietary restraint, disinhibition and hunger. J Psychosom
Res 1985; 29: 71–83.
28 Flint A, Raben A, Blundell JE, Astrup A. Reproducibility,
power and validity of visual analogue scales in assessment of
Effects of food form on appetite and energy intake
DM Mourao et al
1695
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
appetite sensations in single test meal studies. Int J Obes 2000; 24:
38–48.
Jamison RN, Gracely RH, Raymond SA, Levine JG, Marino B,
Herrmann TJ et al. Comparative study of electronic vs paper VAS
ratings: a randomized, crossover trial using healthy volunteers.
Pain 2002; 99: 341–347.
Bray GA, Nielsen SJ, Popkin BM. Consumption of high-fructose
corn syrup in beverages may play a role in the epidemic of
obesity. Am J Clin Nutr 2004; 79: 537–543.
Popkin BM, Armstrong LE, Bray GM, Caballero B, Frei B, Willett
WC. A new proposed guidance system for beverage consumption
in the United States. Am J Clin Nutr 2006; 83: 529–542.
Anderson GH. Sugars Containing Beverages and Post-Prandial
Satiety and Food Intake. Intl J Obes 2006; 30: S52–S59.
Mustad VA, Jonnalagadda SS, Smutko SA, Pelkman CL, Rolls BJ,
Behr SR et al. Comparative lipid and lipoprotein responses to
solid-food diets and defined liquid-formula diets. Am J Clin Nutr
1999; 70: 839–846.
Haber GB, Heaton KW, Murphy D, Burroughs LF. Depletion and
disruption of dietary fibre. Effects on satiety, plasma-glucose, and
serum-insulin. Lancet 1977; 2: 679–682.
Bolton RP, Heaton KW, Burroughs LF. The role of dietary fiber in
satiety, glucose, and insulin: studies with fruit and fruit juice. Am
J Clin Nutr 1981; 34: 211–217.
Mattes RD. Soup and Satiety. Physiol Behav 2005; 83: 739–747.
Blundell JE, Macdiarmid JI. Fat as a risk factor for overconsumption: Satiation, satiety, and patterns of eating. J Am Dietet Assoc
1997; 97 (Suppl): S63–S69.
Gatenby SJ, Aaron JI, Jack VA, Mela DJ. Extended use of foods
modified in fat and sugar content: nutritional implications
in a free-living female population. Am J Clin Nutr 1997; 65:
1867–1873.
Prentice A, Poppitt S. Importance of energy density and
macronutrients in the regulation of energy intake. Intl J Obes
1996; 20: S18–S23.
Rodin J. Effects of obesity and set point on taste responsiveness
and ingestion in humans. J Compr Physiol Psychol 1975; 89:
1003–1009.
Glasbrenner B, Pieramico O, Brecht-Krauss D, Baur M, Malfertheiner P. Gastric emptying of solids and liquids in obesity.
Clin Investig 1993; 71: 542–546.
Flegal KM, Troiano RP. Changes in the distribution of body mass
index of adults and children in the US population. Intl J Obes
2000; 24: 807–818.
Pliner PL. Effect of liquid and solid preloads on eating behavior of
obese and normal persons. Physiol Behav 1973; 11: 285–290.
Tarasuk V, Beaton GH. The nature and individuality of withinsubject variation in energy intake. Am J Clin Nutr 1991; 54: 464–470.
45 Gillis LJ, Bar-Or O. Food away from home, sugar-sweetened
drink consumption and juvenile obesity. J Am Coll Nutr 2003; 22:
539–545.
46 Raben A, Vasilaras TH, Moller AC, Astrup A. Sucrose compared
with artificial sweeteners: different effects on ad libitum food
intake and body weight after 10 wk of supplementation in
overweight subjects. Am J Clin Nutr 2002; 76: 721–729.
47 Ellis PR, Kendall CWC, Ren Y, Parker C, Pacy JF, Waldron KW et al.
Role of cell walls in the bioaccessibility of lipids in almond seeds.
Am J Clin Nutr 2004; 80: 604–613.
48 Yeomans MR, Gray RW, Mitchell CJ, True S. Independent effects
of palatability and within-meal pauses on intake and appetite
ratings in human volunteers. Appetite 1997; 29: 61–76.
49 Pliner P, Mann N. Influence of social norms and palatability
on amount consumed and food choice. Appetite 2004; 42:
227–237.
50 Smiciklas-Wright H, Mitchell DC, Mickle SJ, Goldman JD, Cook
A. Foods commonly eaten in the United States, 1989–1991 and
1994–1996: are the portion sizes changing? J Am Diet Assoc 2003;
103: 41–47.
51 Nielsen SJ, Popkin BM. Patterns and trends in food portion sizes,
1977–1998. JAMA 2003; 289: 450–453.
52 Sakata T, Yoshimatsu H. Homestatic maintenance regulated by
hypothalamic neuronal histamine. Meth Find Exp Clin Pharmacol
1996; 17: 51–56.
53 Fujise T, Yoshimatsu H, Kurokawa M, Oohara A, Kang M, Nakata
M et al. Satiation and masticatory function modulated by brain
histamine in rats. PSEBM 1998; 217: 228–234.
54 Delgado-Aros S, Cremonini F, Castillo JE, Chial HJ, Burton DD,
Ferber I et al. Independent influences of body mass and gastric
volumes on satiation in humans. Gastroenterology 2004; 126:
432–440.
55 Geliebter A, Westreich S, Gage D. Gastric distention by balloon
and test-meal intake in obese and lean subjects. Am J Clin Nutr
1988; 48: 592–594.
56 Hveem K, Jones KL, Chatterton BE, Horowitz M. Scintigraphic
measurement of gastric emptying and ultrasonographic
assessment of antral area: relation to appetite. Gut 1996; 38:
816–821.
57 Ellis P, Kendall CW, Ren Y, Parker C, Pacy J, Waldron K et al. Role
of cell walls in the bioaccessability of lipids in almond seeds. Am J
Clin Nutr 2004; 80: 604–613.
58 Miles CW, Kelsay JL, Wong NP. Effect of dietary fiber on
the metabolizable energy of human diets. J Nutr 1988; 118:
1075–1081.
59 Wisker E, Feldheim W. Metabolizable energy of diets low or high
in dietary fiber from fruits and vegetables when consumed by
humans. J Nutr 1990; 120: 1331–1337.
International Journal of Obesity