International Journal of Obesity (2012) 36, 1340 -- 1345
& 2012 Macmillan Publishers Limited All rights reserved 0307-0565/12
www.nature.com/ijo
ORIGINAL ARTICLE
The effects of calorie information on food selection and intake
L Girz1, J Polivy1, CP Herman1 and H Lee2
OBJECTIVES: To examine the effects of calorie labeling on food selection and intake in dieters and non-dieters, and to explore
whether expectations about food healthfulness moderate these effects.
METHODS: Participants were presented with a menu containing two items, a salad and a pasta dish. The menu had (a) no
calorie information, (b) information that the salad was low in calories and the pasta was high in calories, (c) information that
the salad was high in calories and the pasta was low in calories or (d) information that both were high in calories (study 2
only).
RESULTS: Calorie labels influenced food selection for dieters, but not for non-dieters. Dieters were more likely to order salad
when the salad was labeled as low in calories and more likely to order pasta, even high-calorie pasta, when the salad was
labeled as high in calories. Participants who chose high-calorie foods over low-calorie foods did not eat less in response to
calorie information, although non-dieters reduced their intake somewhat when calorie labels were put in the context of
recommended daily calories.
CONCLUSIONS: The results suggest that the rush to provide calorie information may not prove to be the best approach to
fighting the obesity epidemic.
International Journal of Obesity (2012) 36, 1340--1345; doi:10.1038/ijo.2011.135; published online 12 July 2011
Keywords: menu-labeling; calorie-labeling; food selection; food intake; restraint
INTRODUCTION
Although the percentage of overweight Americans has remained
relatively constant over the last 30 years, rates of obesity have
increased dramatically. Roughly one-third of Americans are
overweight and another third are obese.1 In contrast, between
1976 and 1980, roughly one-third of Americans were overweight,
but only 15% were obese. This increase in obesity is a publichealth concern owing to the link between obesity and serious
medical conditions, such as type 2 diabetes, heart disease and
certain types of cancer.2
One proposed contributor to the rise in obesity rates is the
increased frequency with which people eat at restaurants.3 The
number of meals eaten outside the home has increased
dramatically during the same time period in which obesity rates
have risen,4 and eating frequently in restaurants has been linked
to weight gain,5--8 possibly because people eat more calories
when they eat out than when they eat at home.9 There is a lack of
consistency, however, in the association between eating in
restaurants and weight gain, with some studies showing that
eating in restaurants is linked to higher weight in men but not
women,10,11 some studies showing that eating in fast-food
restaurants, but not full-service restaurants, is associated with
higher weight,6,7 and some showing no association between
eating out and weight.12
In an attempt to combat obesity, menu-labeling legislation has
now been passed as part of the health-care bill in the United
States.13 (A similar bill has recently been proposed in Ontario,
Canada). The United States legislation will require all chain
restaurants with 420 outlets to provide information about
calories next to each item on the menu. A succinct statement
concerning suggested daily caloric intake must also be posted.
The assumption behind this legislation appears to be that
providing individuals with calorie information will lead them to
reduce their caloric intake, and that weight loss will follow such
reduced intake. Reduced caloric intake, however, is only one of
several possible outcomes. Some people may not even notice the
caloric information; others may not care about the calorie
information and may ignore it. Among people who do care about
calorie information, various responses are possible if the food that
they plan to eat is high in calories: they could choose a lowercalorie meal or they could choose to eat less of the original highercalorie meal. Another possibility is that customers may discover
that the high-calorie items that they crave are not that much
higher in calories than are the lower-calorie items, and so, given
that the lower-calorie choice does not represent much of a
savings, they may decide to go for the higher-calorie, more
attractive option. Unfortunately, most menu-labeling studies have
examined the effects of calorie information on ordering behavior
(that is, effects of labeling on item selection), but have failed to
measure amount eaten. As it is entirely possible for people to
change their orders without altering the total calories that they
ingest, or to eat fewer calories without changing their orders, it is
not possible, without measuring the amount eaten, to determine
whether caloric labeling actually decreases caloric intake.
Only three studies to date have examined the effects of calorie
information on amount eaten, and just two of these examined
both calories ordered and amount eaten. Findings from Aaron
et al.14 and Harnack et al.15 suggest that calorie information may
have little effect on the amount that females eat, and may actually
increase the males’ intake. In contrast, findings from Roberto
et al.16 indicate that calorie information may reduce the amount
eaten by both males and females. Interestingly, the increased
intake of male participants in Harnack et al.15 was not
1
Department of Psychology, University of Toronto, Toronto, Ontario, Canada and 2Department of Psychology, University of Western Ontario, Toronto, Ontario, Canada.
Correspondence: Ms L Girz, Department of Psychology, University of Toronto, 100 Street George Street, Sidney Smith Hall, 4th Floor, Toronto, Ontario M5S 3G3, Canada.
E-mail: [email protected]
Received 18 February 2011; revised 16 May 2011; accepted 26 May 2011; published online 12 July 2011
The effects of calorie information
L Girz et al
accompanied by an increase in the number of calories ordered,
whereas the decreased intake in Roberto et al.16 was accompanied
by a decrease in calories ordered.
In light of these contradictory findings, further research
examining the impact of caloric labeling on food selection and
intake is clearly warranted. One aspect of caloric labeling that has
yet to be explored is whether expectations about the healthfulness of foods moderate the impact of caloric labels. It is
conceivable that calorie information affects food selection only
when the calorie information does not match expectations. For
example, restaurant patrons may be surprised to discover that a
salad is high in calories, and this information may dissuade them
from ordering the salad, whereas finding out that a plate of pasta
is high in calories may come as no surprise and thus fail to alter
ordering behavior. By the same token, patrons may be more likely
to order ‘unhealthful’ foods when these foods are lower in calories
than expected. Therefore, examining whether calorie information
affects ordering or intake of ‘healthy’ vs ‘unhealthy’ foods
differently might resolve some of the ambiguity in the previous
findings.
The present research examined the effects of calorie information on food selection and intake in restrained and unrestrained
eaters. The relation between calorie information and eating
behavior was explored both for foods that match expectations (for example, low-calorie ‘healthy’ foods) and for foods
that do not match expectations (for example, high-calorie ‘healthy’
foods).
STUDY 1
Study 1 was designed to examine whether providing calorie
information alters food selection of healthy and less healthy food
(salad and pasta), and if so, in whom (all eaters, or only in
restrained eaters/chronic dieters). Study 1 was also designed to
examine whether providing calorie information (that the menu
item contains 600 or 1200 calories) alters the amount of salad or
pasta eaten.
Participants and methods
Participants. Participants were 149 female students enrolled in the
introductory psychology subject pool. The mean age of participants
was 19.11 (s.d. ¼ 1.82). The Restraint Scale17 was used to categorize
participants as restrained eaters (scoring 15 or higher) or unrestrained eaters (scoring below 15).
Procedure. In order to standardize hunger levels, all participants
were told to refrain from eating for 3 h before participating in the
study. After consenting to participate in the study, participants
were told that they would be rating a potential new menu item for
a local restaurant. They were presented with a menu containing
two items, a salad and a pasta dish, each with a short description.
The salad consisted of cucumber, carrot, tomato, avocado and
cheddar cheese mixed with an oil-and-vinegar dressing. The pasta
dish consisted of rotini mixed with commercial pasta sauce and
mozzarella cheese. The two dishes contained the same number of
calories (approximately 1200 per serving) and the same energy
density (1.6 calories per gram), but the information provided to
participants about the content of each dish was varied across
conditions.
In the control condition (no calorie information), there was no
additional information on the menu beyond the following
descriptions: House salad---crunchy cucumbers, shredded carrots
and diced tomato tossed with fresh cheddar and avocado in our
signature house dressing and Pasta Marinara---rotini tossed with our
signature marinara sauce and topped with fresh mozzarella. In the
first experimental condition (low-calorie salad/high-calorie pasta),
the same ingredient information was presented as in the control
& 2012 Macmillan Publishers Limited
condition, but the salad was described as containing 600 calories
and the pasta was described as containing 1200 calories. In the
second experimental condition (high-calorie salad/low-calorie
pasta), the salad was described as containing 1200 calories,
whereas the pasta was described as containing 600 calories.
Participants were told that they should select either the salad or
the pasta for lunch. While waiting for the food to be prepared,
participants filled out several questionnaires. After the food was
served, participants were given 15 min to complete their meal and
to fill out a short taste-rating form. The experimenter instructed
participants to approach the meal as they would a meal in an
actual restaurant rather than as merely a taste test.
After eating, participants were asked to fill out a series of
questionnaires including the Restraint Scale.17 In order to probe
for suspicion or to determine whether the participant was able to
guess the actual purpose of the study, participants were asked
what they thought the true purpose of the study was before they
were debriefed.
Results
Means and s.d. for food chosen are presented in Table 1. To
examine whether calorie information altered food selection, we
conducted a binary logistic regression with Condition, Restraint
and Condition Restraint as the independent variables and food
selected (pasta vs salad) as the dependent variable. A main effect
was found for Restraint, X2(1) ¼ 4.39, P ¼ 0.036, with unrestrained
eaters being more likely to choose pasta than were restrained
eaters. No Condition, X2(2) ¼ 2.34, P ¼ 0.31, or Condition Restraint,
X2(2) ¼ 0.67, P ¼ 0.72, effects were found.
Notwithstanding these insignificant findings, Table 1 shows that
restrained eaters in the low-calorie salad/high-calorie pasta
condition chose salad more often than pasta whereas participants
in all other conditions chose pasta more often than salad. Indeed,
restrained eaters in the low-calorie salad/high-calorie pasta
condition were much more likely to choose salad instead of pasta
than was a pooled group of all other study participants, X2(1) ¼ 6.44,
P ¼ 0.011.
Means and s.d. for calories eaten are presented in Table 2 and
Figure 1. To examine whether calorie information affected the
amount of salad or pasta that participants ate, we conducted a
Condition Food Chosen Restraint analysis of variance with
calories eaten as the dependent variable. A Kolmogorov--Smirnov
test indicated that the calorie data fit a normal distribution,
Z ¼ 0.730, P ¼ 0.662, and Levene’s test indicated that variance was
equal across groups, F(11,135) ¼ 4.10, P ¼ 0.95. A main effect of Food
Chosen, F(1,135) ¼ 4.41, P ¼ 0.038, and a Condition-by-Food Chosen
interaction, F(2,135) ¼ 4.80, P ¼ 0.01, were found. Participants who
chose pasta ate more than those who chose salad in the no calorie
information condition (Po0.001), but amount eaten did not differ
between those who chose pasta and those who chose salad in the
other two conditions. In addition, among participants who chose
salad, those in the no calorie information condition ate less than
did those in the high-calorie salad/low-calorie pasta condition
(P ¼ 0.029).
Table 1.
Frequencies for food selected (pasta vs salad)
Unrestrained
Restrained
No cal information
600 cal salad, 1200
1200 cal salad, 600
No cal information
600 cal salad, 1200
1200 cal salad, 600
cal pasta
cal pasta
cal pasta
cal pasta
Pasta
Salad
21
18
22
11
10
11
10
9
7
8
15
7
International Journal of Obesity (2012) 1340 -- 1345
1341
The effects of calorie information
L Girz et al
1342
Table 2.
Means and s.d. for amount eaten (cal)
Chose pasta
No cal information
600 cal salad, 1200 cal pasta
1200 cal salad, 600 cal pasta
Unrestrained
Restrained
Unrestrained
Restrained
M (s.d.) N
M (s.d.) N
M (s.d.) N
M (s.d.) N
514.58 (227.02) 20
338.70 (162.15) 18
422.02 (175.81) 22
480.51 (251.80) 11
442.95 (192.28) 10
424.18 (264.63) 11
304.20 (143.88) 10
298.49 (217.85) 9
509.48 (203.18) 7
246.27 (180.12) 8
412.68 (165.17) 14
411.08 (152.71) 7
requires that a statement indicating suggested daily intake be
posted on the menu or menu board, no such statement was
included on our study menu. It is possible that calorie information
has a different effect on food selection and eating behavior when
such information is put in the context of a daily calorie allotment.
Second, men were not included in the study, so it is not possible
to determine whether men respond differently to calorie
information than do women.
600
Calories eaten
500
400
300
200
100
0
Unrestrained
Restrained
No calorie information
Restrained
Unrestrained
Chose pasta Chose pasta Chose salad
Chose salad
600 cal salad, 1200 cal pasta
1200 cal salad, 600 cal pasta
Figure 1.
Chose salad
Mean numbers of calories eaten (study 1).
Discussion
Calorie information influenced food selection among restrained
eaters, but not among unrestrained eaters. It appears that
participants generally preferred pasta to salad, but that restrained
eaters set aside this preference and tended to order salad when they
thought that the pasta was much higher in calories than the salad.
With regard to amount eaten, the provision of accurate calorie
information did not alter intake for participants who chose pasta.
Among participants who chose salad, however, those who
received no calorie information ate much less than those who
received accurate calorie information. Expectations about the
healthfulness of salad and pasta may account for these findings.
Pasta is not generally viewed as a low-calorie food, so it is likely
that participants who chose pasta expected it to be high in
calories regardless of whether or not they received calorie
information. As these participants were willing to choose a highcalorie option in the first place, they were probably not overly
concerned with restricting the amount they ate. On the other
hand, participants who chose salad after receiving no calorie
information may have done so because they were interested in
eating lightly and believed that the salad was a low-calorie option.
The same interest in eating lightly may also have also been the
reason that they ate a relatively small amount. In contrast,
participants who chose salad after being informed that it contained
1200 calories probably did not order salad in order to eat lightly,
but rather because they especially enjoy salad. Thus, these
participants probably ate the amount they desired rather than
attempting to restrict their intake.
There were several limitations of the study that may have
influenced the results. First, although menu-labeling legislation
International Journal of Obesity (2012) 1340 -- 1345
STUDY 2
Study 2 was designed to address these limitations and to extend the
findings of study 1. To better simulate menu labeling as it is
described in the current legislation, a menu condition was added in
which suggested daily calorie intake was included on the menu. In
addition, both males and females participated in study 2 so that
potential sex differences in response to caloric labeling could be
explored. Furthermore, the label for the lower calorie food was
changed from 600 to 400 calories to ensure that participants viewed
this as a truly low-calorie option. The analyses for amount eaten
were also altered; rather than examining the effects of food choice
(salad vs pasta) on amount eaten as was done in study 1, amount
eaten was examined as a function of whether participants received
no calorie information, calorie labels or calorie labels plus
information about recommended daily caloric intake to determine
the effects of labeling per se.
Participants and methods
Participants. Participants were 254 undergraduate students (138
females and 116 males) enrolled in the introductory psychology
subject pool. The mean age of female participants was 18.69
(s.d. ¼ 2.87) and the mean age of male participants was 18.71
(s.d. ¼ 1.79). As in study 1, the Restraint Scale17 was used to
categorize participants as either restrained eaters or unrestrained
eaters.
Procedure. The procedures used in study 2 mirrored those in used
in study 1. The descriptions for salad and pasta were also identical
to those from study 1, however, the calorie information provided
for each menu condition was altered for study 2.
In the control condition, no calorie information was included on
the menu. In the first experimental condition (low-calorie salad/
high-calorie pasta), the salad was described as containing 400
calories and the pasta dish was described as containing 1200
calories. In the second experimental condition (high-calorie salad/
low-calorie pasta), the salad was described as containing 1200
calories and the pasta dish was described as containing 400
calories. In the final experimental condition (high-calorie
salad/high-calorie pasta), both the salad and the pasta were
described as containing 1200 calories, and the recommended
caloric intake of the average male and female was presented
at the bottom of the menu (2000 calories for females and
2400 calories for males).
& 2012 Macmillan Publishers Limited
The effects of calorie information
L Girz et al
Results
Means and s.d. for food chosen are presented in Table 3. As in
study 1, female restrained eaters in the low-calorie salad/highcalorie pasta condition were more likely to choose salad/less likely
to choose pasta than a pooled group of all other study
participants (P ¼ 0.027, for two-sided Fisher’s exact test).
Predictors of food selection were examined using a binary
logistic regression with Condition, Restraint, Gender, Condition Restraint, Condition Gender, Restraint Gender and Condition Restraint Gender as the independent variables and food
selected (pasta vs salad) as the dependent variable. A main effect
was found for Gender, X2(1) ¼ 6.48, P ¼ 0.011, with males being
more likely to choose pasta than were females. A Condition Restraint interaction was also found, X2(3) ¼ 8.94, P ¼ 0.03.
Condition had no effect on food selected for unrestrained eaters,
X2(3) ¼ 1.92, P ¼ 0.59, but was a significant predictor of food
selected for restrained eaters, X2(3) ¼ 9.37, P ¼ 0.025. Restrained
eaters in the high-calorie salad/high-calorie pasta condition were
4.94 times more likely to choose pasta over salad compared with
restrained eaters in the no calorie information condition,
X2(1) ¼ 5.96, P ¼ 0.015, and 3.67 times more likely to choose pasta
over salad compared with restrained eaters in the low-calorie
salad/high-calorie pasta condition, X2(1) ¼ 3.98, P ¼ 0.046. Furthermore, restrained eaters in the high-calorie salad/low-calorie pasta
condition were 4.41 times more likely to choose pasta over salad
compared with restrained eaters in the no calorie information
condition, X2(1) ¼ 5.10, P ¼ 0.024.
Means and s.d. for calories eaten are presented in Table 4 and
Figure 2. A Condition Food Chosen Restraint analysis of
variance could not be conducted because too few participants
chose salad in some cells. Therefore, before conducting the
analyses for calories eaten, the low-calorie salad/high-calorie pasta
and high-calorie salad/low-calorie pasta conditions were combined to create a ‘calorie labels’ condition. We compared this
‘calorie labels’ condition with the ‘no calorie information’
Table 3.
condition and the ‘calorie labels plus framing information’
condition. To examine whether calorie information affected the
amount eaten, a Condition Restraint Gender analysis of
variance was then conducted with calories eaten as the
dependent variable. A Kolmogorov--Smirnov test indicated that
the calorie data fit a normal distribution, Z ¼ 0.765, P ¼ 0.602 and
Levene’s test indicated that variance was equal across groups,
F(11,242) ¼ 1.434, P ¼ 0.158. A main effect of Gender was found,
with men eating more than women, F(1,242) ¼ 88.96, Po0.001. No
other main effects or interactions were significant. All P-values
were above 0.2. In order to test the effects of menu-labeling
legislation more directly, a second analysis was conducted to
compare participants in the no calorie information condition with
those in the calorie labels plus framing information condition. A
main effect of Gender, F(1,118) ¼ 50.79, Po0.001, and a marginal
Condition Restraint effect, F(1,118) ¼ 3.07, P ¼ 0.082, were found.
Post hoc tests revealed that restrained eaters ate marginally less
than did unrestrained eaters in the no calorie information
condition, P ¼ 0.082, but not in the calorie labels plus framing
information condition, P ¼ 0.461. In addition, unrestrained eaters
ate marginally less in the calorie labels plus framing information
condition than in the no calorie information condition, P ¼ 0.083,
but the amount eaten by restrained eaters did not vary between
the two conditions, P ¼ 0.46. Eight participants reported that they
did not notice the calorie information. The analyses for food
selected and calories eaten were repeated with these participants
omitted. All results remained essentially the same.
Discussion
Study 2 provides evidence that restrained eaters are responsive to
calorie labels when selecting among menu options, although the
information just seems to move them away from healthier but less
preferred salads when the salad is seen accurately as highly caloric.
There was no difference in selection of salad between the salad
labeled (falsely) as low calorie and the unlabeled salad, so the effect
of caloric labels on salad ordering was only to decrease the extent to
Frequencies for food selected (pasta vs salad)
Pasta
Salad
12
11
13
11
14
14
7
13
11
15
17
8
7
11
10
15
2
5
1
3
3
0
6
1
8
3
6
6
8
4
6
3
1000
900
Unrestrained
Restrained
Women
Unrestrained
Restrained
Table 4.
400 cal salad, 1200 cal pasta
1200 cal salad, 400 cal pasta
No cal information
1200 cal salad, 1200 cal pasta
400 cal salad, 1200 cal pasta
1200 cal salad, 400 cal pasta
No cal information
1200 cal salad, 1200 cal pasta
400 cal salad, 1200 cal pasta
1200 cal salad, 400 cal pasta
No cal information
1200 cal salad, 1200 cal pasta
400 cal salad, 1200 cal pasta
1200 cal salad, 400 cal pasta
No cal information
1200 cal salad, 1200 cal pasta
800
700
Calories eaten
Men
600
No calorie information
500
Calorie labels
Calorie labels & framing
information
400
300
200
100
0
Unrestrained Restrained Unrestrained Restrained
men
men
women
women
Figure 2.
Mean numbers of calories eaten (study 2).
Means and s.d. for amount eaten (cal)
Men
No cal information
Cal labels
Cal labels plus framing information
& 2012 Macmillan Publishers Limited
Women
Unrestrained
Restrained
Unrestrained
Restrained
M (s.d.) N
M (s.d.) N
M (s.d.) N
M (s.d.) N
930.31 (304.47) 14
842.18 (321.20) 30
790.98 (375.83) 14
698.39 (356.56) 13
822.15 (303.50) 31
868.18 (317.41) 14
509.27 (257.65) 23
476.44 (258.20) 37
396.74 (218.60) 14
490.35 (209.79) 16
463.78 (286.16) 30
428.40 (226.17) 18
International Journal of Obesity (2012) 1340 -- 1345
1343
The effects of calorie information
L Girz et al
1344
which restrained eaters chose salad. More importantly, only unrestrained eaters responded to calorie labels with respect to actual
intake.
General discussion
The present research examined the effects of calorie information
on the selection and intake of salad and pasta in a restaurant-like
setting. Across both studies, calorie labels influenced food
selection for restrained eaters, but not for unrestrained eaters.
Overall, restrained eaters were more likely to order salad when the
salad was either unlabeled, or labeled as low in calories and more
likely to order pasta, even high-calorie pasta, when the salad was
labeled as high in calories. Presumably, then, many restrained
eaters order salad primarily to maintain their diets, and are no
longer interested in eating salad when they find out it is high in
calories.
Calorie information also influenced the amount that participants ate. In study 1, participants who were informed that the
salad contained 1200 calories ate more salad than participants
who received no calorie information (mainly due to the minimal
amount eaten by the restrained eaters who chose salad in the no
calorie information, thinking it was low in calories and thus
maintaining their diets), whereas participants who were informed
that the pasta contained 1200 calories ate the same amount as
those who received no calorie information, possibly because this
information fit their expectations about the caloric value of the
pasta dish. The results of study 2 showed that restrained
participants who received calorie information plus information
about daily caloric intake ate the same amount as those who
received no calorie information. However, unrestrained participants who received calorie information plus information about
daily caloric intake ate marginally less than those who received no
calorie information.
These findings differ somewhat from those of the three
previous studies that examined the effects of calorie information
on amount eaten. Harnack et al.15 found that calorie information
increased males’ intake, and Aaron et al.14 found that calorie
information increased both males’ and unrestrained eaters’ intake,
whereas Roberto et al.16 found that calorie information decreased
overall intake regardless of sex or restraint. These discrepant
findings may be due, at least in part, to the fact that, except for
study 1 in the present paper, none of the previous studies
specifically examined how expectations about food healthfulness
affect the amount eaten. Calorie information that matches
participants’ expectations may not alter amount eaten, whereas
calorie information that does not match expectations could affect
the amount eaten in either direction. When dieters expect a food
to be low calorie and find out that it is not, they may eat more
than when it is described as low in calories (encouraging them to
maintain their diets by eating less). Conversely, if they find out
that a food is less fattening than they believed, they may maintain
their diets and eat less of it. It is, therefore, entirely possible that
amount eaten differed across studies because of both the specific
types of foods served in a particular study and participant
expectations about the healthfulness of these foods. More
consistent patterns may emerge in future studies if amount eaten
is analyzed separately based on the type of food that participants
select.
With regard to food selection, these findings differ from those
of Harnack et al.15, and accord partially with those of Roberto
et al.16 Whereas Harnack et al.15 found no effect of calorie
information on calories ordered, Roberto et al.16 found that calorie
information decreased the number of calories ordered. Results
from study 2 may help to explain these mixed findings. Restrained
eaters in this study altered their ordering behavior in response to
calorie information only when the calorie labels violated expectations about the healthfulness of the foods. Thus, giving caloric
International Journal of Obesity (2012) 1340 -- 1345
information that indicates that a food usually regarded as low
calorie actually is low calorie does not change behavior, but
pointing out that such foods are actually not low calorie just leads
dieters to order the less healthy (and more preferred) high-calorie
option. Participant expectations about the foods served from
McDonald’s in Harnack et al.15 may have differed from expectations about foods served from Au Bon Pain in Roberto et al.16
Foods from McDonald’s are generally not regarded as healthful, so
calorie information for these menu items may not have surprised
participants. In contrast, some of the seemingly healthful items
from Au Bon Pain may have contained more calories than
participants expected. For example, many sandwiches from Au
Bon Pain contain more calories than a McDonald’s Big Mac.
Several limitations of this study should be noted. First, although
the study setting was designed to simulate a restaurant, it was
conducted in a laboratory and the menu was limited to two
options. Second, because participants ate alone, it was not
possible to explore whether the effects of calorie information are
different when people eat in the company of others. People eating
with others may be driven by self-presentation motives, for
example, to appear ‘healthy,’ which could lead them to alter both
the foods that they select and how much they eat.
In conclusion, results of this study indicate that the provision of
calorie information does not alter food selection among unrestrained eaters, but may lead restrained eaters to order ‘healthful’
items such as salad less frequently when these foods contain more
calories than expected, and to order typically higher-calorie foods
such as pasta more often when these foods are lower in calories
than expected. The effects of calorie information on amount eaten
are less clear. Although unrestrained eaters may eat somewhat
less when calorie labels are given along with information about
recommended daily caloric intake levels, there is no indication
that participants who choose high-calorie foods over low-calorie
food eat less in response to calorie information. The rush to
provide calorie information may not prove to be the best
approach to fighting the obesity epidemic.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
REFERENCES
1 Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity
among US adults, 1999-2008. JAMA 2010; 303: 235--241.
2 U.S. Department of Health and Human Services. Office of the Surgeon General.
Overweight and Obesity: Health Consequences 2010. Available at: http://www.
surgeongeneral.gov/topics/obesity/calltoaction/fact_consequences.html.
3 FDA/CFSAN (Food and Drug Administration/Center for Food Safety & Applied
Nutrition). Counting Calories: Report of the Working Group on Obesity 2004.
Available at: http://www.cfsan.fda.gov/Bdms/owg-rpthtml#v.
4 Lin B, Guthrie J, Frazao E. Away-from-Home Foods Increasingly Important to Quality of
American Diet. USDA, Economic Research Service, Agriculture Information Bulletin:
Washington, DC, 1999, p 749.
5 Binkley JK, Eales J, Jekanowski M. The relation between dietary change and rising
US obesity. Int J Obes 2000; 24: 1032--1039.
6 Duffey KJ, Gordon-Larsen P, Jacobs DR, Williams OD, Popkin BM. Differential
association of fast food and restaurant food consumption with 3-y change in
body mass index: the Coronary Artery Risk Development in Young Adults Study.
Am J Clin Nutr 2007; 85: 201--208.
7 Thompson OM, Ballew C, Resnicow K, Must A, Bandini LG, Cyr H et al. Food
purchased away from home as a predictor of change in BMI z-score among girls.
Int J Obes 2004; 28: 282--289.
8 Ma Y, Bertone ER, Stanek EJ, Reed GW, Hebert JR, Cohen NL et al. Association
between eating patterns and obesity in a free-living US adult population. Am J
Epidomiol 2003; 158: 85--92.
9 Guthrie JF, Lin B, Frazao E. Role of food prepared away from home in the
American diet, 1977-78 versus 1994-96: changes and consequences. J Nutr Educ
Behav 2002; 34: 140--150.
& 2012 Macmillan Publishers Limited
The effects of calorie information
L Girz et al
1345
10 Naska A, Orfanos P, Trichopoulou A, May AM, Overvad K, Jakobsen MU et al.
Eating out, weight, and weight gain. A Cross-sectional and prospective analysis in
the context of the EPIC-PANACEA study. Int J Obes 2011; 35: 416--426.
11 Bezerra IN, Sichieri R. Eating out of home and obesity: a Brazilian nationwide
survey. Public Health Nutr 2009; 12: 2037--2043.
12 Anderson M, Matsa DA. Are restaurants really supersizing America? Am Eco J 2011;
3: 152--188.
13 HR 3962, Section 2572. Affordable Health Care for America Act 2010. Available at:
http://www.govtrack.us/congress/bill.xpd?bill ¼ h111-3962&tab ¼ summary.
& 2012 Macmillan Publishers Limited
14 Aaron JI, Evans RE, Mela DJ. Paradoxical effects of a nutrition labelling scheme in a
student cafeteria. Nutr Res 1995; 15: 1251--1261.
15 Harnack LJ, French SA, Oakes JM, Story MT, Jeffrey RW, Rydell SA. Effects of calorie
labeling and value size pricing on fast food meal choices: Results from an
experimental trial. Int J Behav Nutr Phys Act 2008; 5: 63.
16 Roberto CA, Larsen PD, Agnew H, Baik J, Brownell KD. Evaluating the impact of
menu labeling on food choices and intake. Am J Public Health 2010; 100: 312--318.
17 Herman CP, Polivy J, Silver R. Effects of an observer on eating behavior: the
induction of 00 sensible00 eating. J Pers 1979; 47: 85--99.
International Journal of Obesity (2012) 1340 -- 1345