Appetite 121 (2018) 285e293

Contents lists available at ScienceDirect

Appetite
journal homepage: www.elsevier.com/locate/appet

Consumers’ health-related perceptions of bread e Implications for

labeling and health communication
Pernilla Sandvik*, Margaretha Nydahl, Iwona Kihlberg, Ingela Marklinder
Department of Food, Nutrition and Dietetics, Uppsala University, Box 560, SE-75122 Uppsala, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: There is a wide variety of commercial bread types and the present study identifies potential pitfalls in
Received 3 February 2017 consumer evaluations of bread from a health perspective. The aim is to describe consumers’ health-
Received in revised form related perceptions of bread by exploring which health-related quality attributes consumers associate
27 September 2017
with bread and whether there are differences with regard to age, gender and education level. A postal
Accepted 10 November 2017
and web-based sequential mixed-mode survey (n ¼ 1134, 62% responded online and 38% by paper) with
open-ended questions and an elicitation task with pictures of commercial breads were used. Responses
were content analyzed and inductively categorized. Three fourths (n ¼ 844) knew of breads they
Keywords:
Quality perception
considered healthy; these were most commonly described using terms such as “coarse,” “whole grain,”
Cues “fiber rich,” “sourdough,” “crisp,” “less sugar,” “dark,” “rye,” “seeds,” “a commercial brand,” “homemade”
Bread and “kernels.” The breads were perceived as healthy mainly because they “contain fiber,” are “good for
Whole grain the stomach,” have good “satiation” and beneficial “glycemic properties.” The frequency of several eli-
Age cited attributes and health effects differed as a function of age group (18e44 vs. 45e80 years), gender
Education and education level group (up to secondary education vs. university). Difficulties identifying healthy
bread were perceived as a barrier for consumption especially among consumers with a lower education
level. Several of the health effects important to consumers cannot be communicated on food packages
and consumers must therefore use their own cues to identify these properties. This may lead to con-
sumers being misled especially if a bread is labeled e.g., as a sourdough bread or a rye bread, despite a
low content.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction nonetheless an important source of whole grain and fiber in the

Health is an important dimension in consumers' overall quality
perception of food products (Grunert, 2002). The healthiness of
food is described as a credence quality dimension, as the consumer
can evaluate it only to a limited degree after consumption
(Steenkamp, 1990). This makes external sources of information
important in forming consumers’ health-related quality expecta-
tions about food products (Grunert, 2002). In Sweden, proponents
of low carbohydrate diets have questioned the healthiness of
carbohydrate-rich foods such as bread, which may have affected
consumer perceptions (Gunnarsson & Elam, 2012). Exposure to
contradictory nutritional health information has been associated
with nutrition confusion and may undermine the success of healthy
eating campaigns and interventions (Nagler, 2014). Bread is
* Corresponding author.
E-mail address: pernilla.sandvik@ikv.uu.se (P. Sandvik).
Swedish diet (Kyrø et al., 2012). However, like in many other
countries, intake of both whole grain and fiber in Sweden do not
meet the national recommendations (Amcoff et al., 2012). Lower
intake of whole-grain bread among Swedish adults has been
associated with a lower education level and younger age (18e45
years), and younger consumers have reported eating less bread
overall (Sandvik, Kihlberg, Lindroos, Marklinder, & Nydahl, 2014).
There is limited knowledge of whether health-related perceptions
of bread vary among younger and older consumers as well as
consumers with different educational levels. Gaining insight into
consumer perceptions of healthy eating is valuable in assessing
how health-promotion messages are interpreted and put into
practice. But also, on how consumers may be misled when
attempting to identify healthy food products.
In Sweden, bread is the individual food group contributing the
most to the total energy intake, with a mean of 11 percent (Amcoff
et al., 2012). From a nutritional perspective, there is a wide range of

https://doi.org/10.1016/j.appet.2017.11.092
0195-6663/© 2017 Elsevier Ltd. All rights reserved.
286 P. Sandvik et al. / Appetite 121 (2018) 285e293
commercial breads due to the great variation in ingredients as well
as rising and baking methods. Attributes suggested to be important
for cereal products promoted as being healthy from a nutritional
perspective are a high whole grain and dietary fiber content and a
low salt, added sugar and saturated fat content. With regard to
physiological functionality, cereal products should contribute pre-
biotic potential, increased gastrointestinal transit rate and fecal
weight as well as promote satiation and a slow glycemic profile
(Poutanen, Sozer, & Della Valle, 2014). Regarding bread, these
physiological functionalities can be achieved in different ways.
High fiber content is related to beneficial gastrointestinal effects
(Anderson, Baird, & Davis, 2009) and can be achieved either by
using fiber-rich cereals such as rye instead of wheat or by supple-
menting with bran or other fiber sources (Poutanen et al., 2014).
The structure of the bread is important for satiety and is e.g.
affected by the dietary fiber present in the grain (Isaksson,
Fredriksson, Andersson, Olsson, & Åman, 2009). Moreover, sour-
dough fermentation, whole cereal kernels, a high rye content and
soluble dietary fiber are examples of attributes that may positively
affect the glycemic profile (Fardet, Leenhardt, Lioger, Scalbert, &
Remesy, 2006; Liljeberg, Lo € nner, & Bjo €rck, 1995; Rose n et al.,
2009). For consumers without a scientific background, the above-
mentioned attributes may be difficult to understand, making it
challenging for such consumers to identify healthy bread in a
buying situation. Just as in other countries (Slavin, Tucker,
Harriman, & Jonnalagadda, 2013), official national Swedish rec-
ommendations regarding bread consumption are to choose whole
grain as well as Keyhole labeled products (NFA, 2015a). The Keyhole
is a Nordic evaluative signpost label which indicate that the bread
contains more whole grain and fiber and less sugar and salt (NFA,
2005, 2015b).
When forming health-related expectations about a food product
prior to purchase, consumers may use intrinsic cues, i.e. the phys-
ical characteristics of the product, and extrinsic cues, i.e. packaging,
price, brand and/or labels (Grunert, 2002). Although research has
shown widespread interest among consumers in nutrition infor-
mation on food packaging, responses to nutrition labeling, health
symbols and nutrition and health claims have been studied largely
in isolation from the rest of the package information (L€ ahteenma €ki,
2015). The food packages “speak” to the consumer about the
health-related and other quality attributes of the product. This is
done through the semiotic cocktail that includes brand elements
and signpost labels, food names, text, facts and figures, illustrations
and other non-verbal elements (Smith, Clement, Møgelvang-
Hansen, & Sørensen, 2011). Consumers encounter the packaging
with their different backgrounds and knowledge, which are deci-
sive in making sense of the visual cues, and cues other than those
regulated by authorities may be associated with healthfulness in a
given consumer's mind.
Several qualitative research studies have focused on consumer
perceptions of healthy eating from an overall perspective (Bisogni,
Jastran, Seligson, & Thompson, 2012). Overall perceived healthiness
of selected meals has for example also been investigated by
studying consumers' choices in a fake food buffet (Bucher, Müller, &
Siegrist, 2015; Mo € tteli, Keller, Siegrist, Barbey, & Bucher, 2016).
Gellynck, Kühne, Bockstaele, Van de Walle & Dewettnick (2009)
investigated Belgian consumers’ overall quality perceptions of
bread. The consumers were segmented based on their perceptions
but in all groups the majority of the respondents agreed to the
statement that “little is known about the nutritional value of
bread”. This study did not differentiate between different bread
types and one health-related statement that the respondents
answered were e.g., “Bread is good for the digestion.” In previous
research consumers have rated overall perceived healthiness of
different breads without packaging (Bucher et al., 2015; Sandvik,
Nydahl, Marklinder, Naes, & Kihlberg, 2017). Consumers were
shown to differentiate between bread types and when associating
their ratings to sensory descriptions of the bread samples, dark
brown color, compact texture and sour flavor was shown to be
associated with perceived overall healthiness (Sandvik et al., 2017).
In the present study, consumer perceptions of health-related bread
quality are further investigated, giving insight into the pre-existing
knowledge that consumers use when making sense of different
breads and packages, for example in the shopping situation. This
knowledge can help in identifying the potential pitfalls in con-
sumer evaluations of bread from a health perspective.
The aim of the present study is to describe consumers’ health-
related perceptions of bread by exploring which health-related
quality attributes consumers associate with bread and whether
consumer perceptions differ as a function of age, gender and edu-
cation level.
2. Materials and method
A total of 3000 subjects aged 18e80 years, chosen by random
selection from the national population address register, received a
postal invitation to participate in a web-based survey. Non re-
spondents thereafter received an identical postal questionnaire and
data were thereby collected in a web-based and a postal sequential
mixed-mode survey (De Leeuw, 2005). The ambition was for most
of the questionnaires to be completed online, enabling both
financial and environmental savings. Every postal invitation
included a unique code that was registered upon initiation of the
survey response session. Fig. 1 shows the data collection procedure.
The survey was part of a larger study investigating consumer
perceptions of bread and the present study focuses on the
perceived healthiness of bread. First and foremost, open-ended
questions were used to explore consumers’ health-related per-
ceptions of bread; it is these perceptions that are analyzed in the
present paper. The goal of this approach is to try to understand how
consumers respond without the assisted recall provided by a set of
predetermined response alternatives (Perreault & Leigh, 1989). The
approach has previously been used to study consumer perceptions
of different foods (e.g., Hough & Ferraris, 2010; Pe neau, Linke,
Escher, & Nuessli, 2009; Symoneaux, Galmarini, & Mehinagic,
2012). The following questions were included to explore whether
the respondents differentiate between bread types from a health
perspective, which bread types are perceived as healthy and which
health effects are most well-known: Are you aware of bread that, in
your opinion, is good for health? If yes, which type of bread are you
thinking of? And in what way is it, in your opinion, good for health?
Attribute elicitation (Bech-Larsen & Nielsen, 1999), based on
pictures of commercial breads, was used to further explore which
product properties were important in evoking expectations of
health-related bread quality. Pictures of six commercial breads
were shown, both with and without their packaging. The back-of-
package information (ingredient list in descending order and nu-
trients; energy, protein, carbohydrates, sugars, fat, saturated fat,
fiber and sodium) was also available to the respondents by clicking
a separate link in the online survey or by looking on the back of the
paper version. In store, consumers must pick up the specific
product to read the back-of-package information. Thus, to increase
the validity, respondents were required to take an extra step to
access this information. All six breads were produced by major
national bakeries and available throughout the country. The breads
(A-F) were selected to achieve variation in the factors whole grain,
cereal type, sugar content and fermentation method. Sample A
(Fro €
€kusar, Fazer AB) and F (Antligen Toast, Pågen AB) could be
described as modern bread types, A: a roll made with spelt and
sunflower seeds and F: a white wheat toast bread with added
 P. Sandvik et al. / Appetite 121 (2018) 285e293 287

Address unknown n = 26
Postal invitation Unable to participate n = 14
National random sample No reply n = 2344
Total responses
18-80 yrs.
n = 1141 38%
n = 3000 Responses (after a first Web 62 % (n=710)
postcard reminder) n = 642 Paper 38 % (n= 431)

Second reminder Responses n = 499 Used in analysis

Postal questionnaire n=1134
n = 2344
Address unknown n = 42
No reply n = 1803

Fig. 1. Data collection procedure.

whole grain. Sample D (Lingongrova, Pågen AB) was the most sold
bread on the market, a sliced loaf with wheat and rye, some whole
meal flour and added syrup and sugar. Sample B (Skogaholmslimpa,
€d Skovmand,
Fazer AB), C (Rågkusar, Fazer AB) and E (Danskt rågbro
Fazer AB) were more traditional breads, B: a sliced loaf with sifted
wheat and rye with syrup, C: a Finnish style whole-grain rye
sourdough roll and E: a Danish style rye bread with whole rye
kernels (Table 1). As shown in Table 1, sample C and E have the
highest content of whole grain and lowest content of total sugars
while sample B contains almost 10% total sugars and only refined
flour. A Keyhole label were displayed on sample A, C, E and F. For
elicitation, a simple sorting task was used. The respondents chose
which of the six breads they expected to be most and least healthy
(multiple choice). After choosing, they were asked to briefly moti-
vate their choices. This approach minimized the workload for the
respondents while consumer-relevant properties were still
captured.
Background questions included gender, age, education level and
bread-eating habits. The respondents were also asked whether
they in general found it difficult to know which bread was the
healthier choice when buying bread (yes/no). The questionnaire
was pilot tested (n ¼ 15) before being distributed; minor changes
were made following the pilot test.
The open-ended answers were content analyzed and, based on
the gathered data, categories were formed inductively (Spiggle,
1994). The data were primarily coded by the first author. The
open-ended answers were written in one or two words or in a short
sentence. In order to check reliability, a randomly sample of 10
percent was selected for all four open-ended questions and coded
by the third and last author using the inductively formed

Table 1
Ingredients and characteristics of commercial breads shown in pictures with and without packaging and respondents choice of samples as healthiest or least healthy.

Bread products and ingredients Whole Fibera Total Key- Ranking sales Consumers % (n)
grain % sugarsa holeb statistics 2011 who sorted bread
as most or least
healthy
(multiple choice)

Healthiest Least
healthy

A) Roll with spelt and sunflower seeds 25 5.5 4.5 Yes 3 39 (441) 2 (27)
Whole-grain wheat flour, wheat flour, spelt (10%), water, sunflower seeds (9%), flaxseed (5%), syrup,
wheat gluten, yeast, rye fiber, scalded rye whole grain rye flour, wort, pumpkin seeds (2%), salt,
thickener (guar gum flour), rye malt, preservative (sorbic acid).
B) Loaf with sifted wheat and rye and syrup 0 5.0 9.5 No 2 2 (18) 54
Flour (Wheat flour and sifted rye flour), water, syrup, yeast, scalded rye, vegetable fat, wheat (605)
sourdough powder, salt, wheat gluten.
C) Whole grain rye sourdough roll 100 11.5 2.0 Yes 32 51 (571) 1 (38)
Whole-grain rye flour, water, whole-grain rye sourdough, whole-grain wheat, rye fiber, salt yeast.
83% of the flour is rye flour
D) Sweet wheat-rye partly whole grain loaf 25 6.5 9.0 No 1 13 (148) 10
Wheat flour, water, rye flour, rye and wheat whole-grain flour, syrup, cut whole-grain rye, wheat and (114)
whole-grain rye sourdough (4.2 %), wort, wheat gluten, sugar, wheat bran, yeast, vegetable oil
from rapeseeds, lingon berries (1,5%), redcurrants, prefermented wheat flour, salt
E) Danish style whole kernel rye bread 87 8.0 2.0 Yes >50 61 (681) 2 (32)
Whole-grain rye sourdough, rye kernels (19%), whole-grain rye flour, water, sifted rye flour, bread
crumbs, malt syrup from barley, salt, yeast, vinegar, rye sourdough powder, rapeseed oil, barley
malt flour, wheat flour.
F) White toast with whole grain 25 7.0 3.0 Yes 30 3 (32) 71
Wheat flour, water, whole-grain wheat flour, wheat gluten, corn starch, yeast, vegetable oil, syrup, (794)
salt, wheat sourdough.
a
Grams per 100 g of bread.
b
According to LIVSFS 2009:6, the keyhole symbol may be used for bread with 25% whole grain (on dry weight) and per 100 g  7 g fat, 5 g sugars, 0.5 g Na and 5 g
dietary fiber (NFA, 2005). Updated regulations in LIVSFS 2015:1 were not yet published at the time of this survey.
288 P. Sandvik et al. / Appetite 121 (2018) 285e293
categories. Inter-rater reliability was checked using Krippendorff's
Alpha (Hayes & Krippendorff, 2007). Krippendorff's Alpha was
a ¼ 0.89 for descriptions of healthy bread i.e. “Which type of bread
are you thinking of and a ¼ 0.79 for “In what way is it, in your opinion,
good for health.” And a ¼ 0.85 and a ¼ 0.92 respectively for the
answers motivating why breads shown on pictures were perceived
as more or less healthy. Although there is no single value to control
for reliability, an a-value of at least 0.67 has been suggested as the
lowest conceivable limit and an a-value over 0.80 as a more certain
level for establishing reliability (Krippendorff, 2004). The frequency
of each category was counted, and categories with a frequency of
20 were considered for all open-ended questions. The frequency
of elicitation can be related to the importance of a concept in the
consumers' minds (Guerrero, Colomer, Gua rdia, Xicola, & Clotet,
2000) and was analyzed as a function of age, gender and educa-
tional level. The answers to the two questions e Which type of bread
are you thinking of? And: In what way is it, in your opinion, good for
health? e were combined in a cross-tabulation illustrating the
percentage of respondents assigning a specific health property to
their description of healthy bread, and correspondence analysis
was used to provide a graphical illustration of the results (Hoffman
& Franke, 1986). For a clearer visualization, only categories with a
frequency of 40 elicitations were included in this step. Cross-
tabulations of frequency of elicited attributes for each bread sam-
ple (A-F) in the elicitation task were also constructed together with
a correspondence analysis. Descriptive statistics, Mann-Whitney U
test and Chi-square with correction for continuity for 2  2 tables
were used for the statistical analyses. The level of significance was
set at p¼ < 0.01 to minimize the risk of type 1 errors due to multiple
tests. The analyses were performed in the Statistical Package for
Social Sciences version 22.
3. Results
A total of 1134 questionnaires were included in the analysis
(response rate 38%). Among these, approximately 75% answered
each of the open-ended questions rendering written answers from
more than 800 respondents for each question. Of the total re-
spondents, 52% (n ¼ 586) were females and the mean age was 53
years (SD 17). When the respondents were divided into two groups
by age, 33% (n ¼ 371) were “younger” (18e44 years) and 67%
(n ¼ 763) were “older” (45e80 years). Regarding education level,
49% (n ¼ 554) had “a lower education level,” i.e. up to upper sec-
ondary school, and 51% (n ¼ 567) a university or university college
degree. More than half, 59% (n ¼ 665) of respondents reported
eating bread every week up to two slices a day, 38% (n ¼ 432) ate
more than two slices of bread a day and 3% (n ¼ 29) ate no bread or
less than one slice of bread a week. Half of the respondents (53%,
n ¼ 600) reported that they had, during the past two years, changed
their consumption of bread. No differences were found as a func-
tion of age, gender or education level. About 11% reported to have
reduced their total intake during the last two years, 18% had
changed the type of bread consumed and 22% had both reduced
their total intake and changed the type of bread consumed. Few
(3%) claimed to have increased their bread consumption.
The majority (75%, n ¼ 844) answered yes to the question “Are
you aware of bread that, in your opinion, is good for health?” Among
the 75%, significantly more respondents were in the higher edu-
cation level group (high education level 79% vs. low education level
71%, p ¼ 0.002). However, 33% of the respondents who knew of
healthy bread, reported finding it challenging to identify the
healthier alternative. Among these 33%, a significantly larger pro-
portion of respondents had a low education level (38%) as
compared to a higher education level (28%) (p ¼ 0.004). No dif-
ferences were found between gender or age-groups.
The most frequent open-ended answers used to describe
healthy bread in response to the question “Which type of bread are
you thinking of?” e which was answered by 99% (n ¼ 831) of the
respondents who had previously stated that they knew bread they
perceived as healthy e were “coarse,” “fiber rich,” “whole grain,”
“sourdough” and “crisp bread” (Table 2). The mean number of eli-
cited attributes among the respondents were 1.7. The number were
significantly higher among females (1.8) compared to males (1.6) as
well as for respondents with a high educational level (1.7)
compared to those with a low education level (1.6). Respondents
with a university degree stated “no/less sugar” more often
compared to the group with a lower education level (high educa-
tion level 12% vs. low education level 7%, p ¼ 0.009). Younger
consumers more often described healthy bread as containing
“seeds” (younger group 11% vs. older group 3%, p¼<0.001). Females
more often than males described healthy bread using the attribute
“sourdough” (18% females vs. 11% males, p ¼ 0.005).
The most common answers to the question “In what way is it, in
your opinion, good for health” e which was answered by 93%
(n ¼ 782) of the respondents who stated that they knew bread they
considered healthy e were “good for the stomach,” “contains fiber,”
“glycemic properties,” “good for the bowel” and “satiating”
(Table 2). Glycemic properties included attributes such as “blood
glucose control,” “low GI” and “slow carbohydrates.” The mean
number of elicited attributes among the respondents were 1.6. The
number were significantly higher for females (1.7) compared to
males (1.5) and for respondents with a high educational level (1.7)
compared to those with a low education level (1.5). Younger re-
spondents stated “satiation” (younger group 21% vs. older group
8%) (p¼<0.001) and ”glycemic properties” (younger group 22% vs.
older group 14%) (p ¼ 0.007) more often than older respondents
did, while older consumers more often stated “good for the stom-
ach” (older group 33% vs. younger group 21%) (p¼<0.001). Re-
spondents with a university degree, in contrast to respondents
without one, more often answered “glycemic properties” (high
education level 21% vs. low education level 12%) (p ¼ 0.001).
Significantly more females compared to males stated “satiation”
(18% females vs. 7% males, p¼<0.001) when describing in what way
the bread was perceived as being good for health.
The correspondence analysis of the two questions “Which type of
bread are you thinking of?” and “In what way is it, in your opinion,
good for health?” (Fig. 2) included all answers with a frequency of
40 mentions. Homemade was thereby excluded but it should be
mentioned that home-made bread was in almost half of the cases
described as healthy due to being “natural/knowing content.” The
analysis shows that “crisp bread,” a “commercial brand,” and “rye”
were often described as healthy because they contain fiber. “Fiber
rich,” in turn, was described as “good for the stomach” and “good
for the bowels.” “Satiating” was used to describe why “seeds,”
“whole grain” and “fiber rich” were perceived as healthy in bread.
“Glycemic properties” was often used when explaining why
“sourdough” and “no/less sugar” were healthy in bread. The pattern
was not as clear for “coarse,” “whole grain,” “dark color,” and
“seeds,” indicating that they were attributed a variety of health
effects.
The respondents’ choices regarding which of the six bread
samples they classified as most/least healthy are shown in Table 1.
The majority of the consumers selected sample C and E, the sam-
ples that also contained the highest content of whole grain and the
lowest content of sugar as being the healthiest. The majority of
respondents (71%) chose sample F, the white toast bread with a
Keyhole symbol, as the least healthy bread. The respondents dis-
agreed on the healthiness of sample D, the wheat-rye loaf with
some whole-meal flour and a high content of syrup; more older
respondents (older group 16% vs. younger group 10%) (p ¼ 0.001),
 P. Sandvik et al. / Appetite 121 (2018) 285e293 289

Table 2
Answers elicited from consumers on open-ended descriptions of healthy bread (n ¼ 831) and in what way it is good for health (n ¼ 782).

Which (healthy) bread In what way is it, in your

are you thinking of? opinion, good for health?

Elicited descriptions Mentioned Elicited descriptions Mentioned

% (n) % (n)

Sensory descriptions Health beneficial content

Coarse 28 (236) Contains fibers 24 (186)
Dark color 12 (98) Unsweetened/less sugar 9 (71)
Should contain Contains nutrients 5 (41)
Fiber rich 23 (187) Healthy content 4 (35)
Whole grain 23 (190) Natural/know content 4 (34)
Sourdough 14 (119) Less carbohydrates 3 (27)
Rye 9 (74) Health effect
Seeds 6 (46) Good for stomach 29 (223)
Kernels 4 (30) Glycemic properties 17 (129)
Should not contain Good for bowel 16 (125)
No/less sugar 10 (81) Satiating 13 (98)
Type of bread Good for digestion 8 (59)
Crisp bread 12 (101) Weight control 3 (20)
Commercial brand 6 (49) Overall wellbeing 3 (24)
Homemade 5 (38)

Total number of elicited words: 1387 Total number of elicited words: 1257
Average mentioned: 1.7 Average mentioned: 1.6

Words with a frequency of 20 mentions.

Crisp bread

Good for
Fiber-rich the stomach
Good for
bread digestion Contains fiber

Coarse
Satiating Whole grain
Commercial
Good for the Dark color Rye brand
bowels Seeds
Contains
nutrients
Bread with
no/less sugar
Contains no/less sugar

Glycemic
properties Sourdough

Fig. 2. Correspondence analysis of contingency table crossing the two open-ended questions “Which (healthy) bread are you thinking of?” (C ¼ bread type) and “In what way is it,
in your opinion, good for health?” (, ¼ health effect) with a frequency of 40 mentions. “Kernels,” “Homemade bread,” “Natural/knowing content,” “less carbohydrates” and
“overall wellbeing” were thereby excluded.

less educated respondents (low education level 17% vs. high edu-
cation level 9%) (p¼<0.001) and more males (16%) vs. females (10%)
chose D as one of the healthiest (p ¼ 0.004).
Fig. 3 shows consumer-relevant attributes in the classification of
healthier or less healthy bread. In total, 75% of the respondents
open-endedly described why the samples were perceived as more
or less healthy. An analysis of non-respondents showed that these
were more often older compared to younger consumers (29% older
group vs. 18% younger group, p¼<0.001), males compared to fe-
males (31% males vs. 21% females, p¼<0.001) and respondents with
290 P. Sandvik et al. / Appetite 121 (2018) 285e293
Fig. 3. Correspondence analysis of contingency table crossing bread samples (A-F) and elicited properties (with a frequency of 20) for perceiving bread as most (816 respondents)
or least healthy (808 respondents). - ¼ bread and B ¼ elicited attributes. Underlined attributes are perceived as negative from a health perspective. Numbers in parenthesis for
attributes: Percentage of respondents mentioning the specific attribute.
a lower educational level compared to respondents with a higher
educational level (low education level 28% vs. high education level
22%, p ¼ 0.31).
A total of 816 respondents answered the open-ended question,
describing why the samples were perceived as healthier. Healthier
bread was primarily explained using attributes such as “coarse”
(19%), “fiber” (28%), “no/less sugar/syrup” (16%), “sourdough” (11%)
and “whole grain” (11%), but also e.g. “dark color” (5%) and less
carbohydrates (4%). A total of 808 respondents answered the open-
ended questions, describing why the samples were perceived as
less healthy. The most common answers were “Contains sugar/
syrup” (29%), “white bread” (24%) and “no/less fiber” (19%) but also
e.g. “wheat flour” (6%) and “fast carbohydrates” (6%). Sample B, the
sifted wheat-rye loaf with syrup, was perceived as less healthy,
primarily due to “sweet/syrup/sugar” and sample F, the white toast
bread, as being “white bread” and “airy/fluffy texture”(Fig. 3).
4. Discussion
The majority of respondents reported knowing of bread that
they considered healthy. One out of four was nonetheless not aware
of any healthy bread, and the proportion responding in this way
was significantly higher among respondents with a lower educa-
tion level. One third also reported having decreased their bread
intake. One reason for these results may be that the healthiness of
carbohydrate-rich and gluten containing food such as bread has
been questioned, mainly in the media and blogosphere over the
course of the past decade (Gunnarsson & Elam, 2012; Pollard et al.,
2017). Consequently, increasing these consumers’ awareness of
different types of bread and of their contribution to a healthy diet is
a challenge. Among respondents who did know of bread they
considered healthy, more with a lower education level found it
challenging to identify which bread is healthier e something that
could indicate lower nutrition literacy (Velardo, 2015). Together
with the higher number of respondents in this group claiming not
to know of any healthy bread, this may also contribute to the less
healthy bread consumption pattern previously seen in this group
(Sandvik et al., 2014).
Food can be described as affecting our health and wellbeing
from different perspectives, such as nutritional, esthetic, social,
cultural and symbolic meaning perspectives, as well as indirectly
through ecological, social and other consequences (Siipi, 2013). In
the present study, the respondents' definition of healthy bread was
based on their own interpretations, and they primarily proposed
nutritional qualities. Although using a lay language, the re-
spondents’ answers were generally well in line with current sci-
entific descriptions and recommendations of bread from a health
perspective (NFA, 2015a, 2015b; Poutanen et al., 2014). Gastroin-
testinal benefits were frequently mentioned, especially by older
consumers, and high fiber content was perceived as the main
contributor to these benefits. High fiber content can therefore be
interpreted as what consumers see as the most important health-
related attribute of bread. Fiber has previously been shown to be
perceived as contributing to the healthiness of bread (Berg, Jonsson,
Conner, & Lissner, 2003; Lyly, Soini, Rauramo, & La €hteenma €ki,
2004). Whole grain was frequently mentioned, which is in accor-
dance with previous research showing that consumers are gener-
ally aware that whole-grain products are healthy, although
differences were identified across countries (Arvola et al., 2007).
Younger respondents and respondents with a higher education
 P. Sandvik et al. / Appetite 121 (2018) 285e293
level more frequently mentioned glycemic properties. This might
be partly explained by the more recent media focus on the glycemic
index (GI) (Borra & Bouchoux, 2009). Consumers associated espe-
cially sourdough and having less/no sugar with beneficial glycemic
properties. Sourdough fermentation has the potential to lower the
GI if a critical amount of acids is formed (Liljeberg et al., 1995). Yet,
in another study, conducted in Sweden, a sourdough label on the
package was shown not to be a reliable indication of beneficial
glycemic properties, because the amount of sourdough included
was sometimes negligible (Sandvik, Marklinder, Nydahl, Næs, &
Kihlberg, 2016). It has been shown, that the GI of baked goods
with added sugar was not elevated relative to the unsweetened
variety (Wolever & Miller, 1995). Sugar was often mentioned as
something that healthy bread should contain less of. This may be
because many bread types in Sweden are quite sweet compared to
bread in other countries, containing up to around 10% added sugar
(e.g., sample B and D). Rye was the only specific cereal mentioned,
and it was described as healthy primarily due to its high fiber
content, which is accurate (Rakha, Åman, & Andersson, 2010). The
negative perception of wheat from a health perspective has also
been seen in other international studies (Golley, Corsini, Topping,
Morell, & Mohr, 2015). Although gluten was not specifically
mentioned by the consumers in the present study, the negative
perception of wheat may be related to the trend among some
consumers to exclude gluten in the diet despite a lack of a celiac
diagnosis (Aziz et al., 2014). That consumers did not mention
gluten-free as an important attribute of healthy bread could also
indicate that the gluten-free trend not is as widespread in Sweden
as for example in the U.S (Kim et al., 2016). It should however also
be noted that the consumers more often responded what they
perceived a healthy bread should contain, rather than what it
should not. Crisp bread was perceived as healthy due to its high
fiber content. Crisp bread is a traditional bread type in Sweden that
is often high in fiber. Its health potential, however, varies depend-
ing on ingredients, and crisp bread made from rye has been shown,
e.g., to be more satiating than wheat crisp bread (Johansson, Lee,
Riserus, Langton, & Landberg, 2015). Home-made bread was
mainly associated with “being natural” or “knowing the content.”
Natural and authentic food is thought to be pure and essentially
normal in that it does not contain ingredients that do not belong in
food (Siipi, 2013). Familiarity of food has been described as
contributing to its perceived safety and healthiness, owing to the
knowledge associated with familiarity (Siipi, 2013). Building con-
sumers' trust is therefore a challenge for the baking industry. This
also indicates that consumers may perceive as healthy bread with a
more home-made appearance or a package indicating small-scale
or artisan baking.
European Union laws on food information strive to enable
consumers to identify and make appropriate use of a food and to
make choices that suit their individual dietary needs (European
Commission, 2011). But in relation to consumers’ perceptions of
healthy bread, what are their actual possibilities to make informed
choices? Including an ingredient list on the package as well as (as of
December 13, 2016) a nutrition declaration is required. Nutrition
and health claims that have been authorized by the European Food
Safety Administration (EFSA) may also be used, but few of the eli-
cited health effects in the present study can be communicated.
Bread with a high content of specific fibers may be labeled as
beneficial for bowel function or as increasing fecal bulk (Poutanen
et al., 2014). This is rarely done on bread packaging in Sweden,
although there are several products on the market that would be
eligible for such labeling. Given the strong association consumers
made between fiber and gut health, this may be considered
redundant. Conversely, there are no approved health claims related
to, for example, the satiation or glycemic properties of bread, nor
291
are there any recommendations for considering glycemic proper-
ties in the Nordic Nutrition Recommendations (Nordic Council of
Ministers, 2014). Moreover, although the authorities recommend
whole grain consumption, there is no approved health claim for
whole grain, because as a constituent whole grain is not considered
to have been sufficiently characterized (EFSA, 2010).
Still, the above-mentioned attributes are on consumers' minds,
and consumers must therefore rely on their own cues when trying
to identify bread with these health benefits. Sütterlin and Siegrist
(2015) showed that adding the word “fruit” on the package
caused consumers to perceive sugar as something healthier. Sym-
bolic information may have a misleading effect on the perceived
healthiness of food, and this was described as a “halo effect” or
nutri-cleaning. The present study shows that particularly attributes
such as whole grain, sourdough, kernels and rye, which are not
regulated, can increase consumers' perception of the healthiness of
bread if the words are used on the packaging. A quantitative indi-
cation of ingredients, must be declared on the package, for in-
gredients that for example is emphasized on the labeling in words,
pictures or graphics (European Commission, 2011). Consumers,
however, do not always study this often back-of-package infor-
mation in detail, and if such attributes are accentuated on the
packaging, despite being a minor ingredient, this may be
misleading (Sørensen, Clement, & Gabrielsen, 2012). In a study
among older Americans (65 years), 46% of the respondents also
misidentified bread as being whole-grain bread (Violette et al.,
2016). One positive development, however, is a sector-wide
agreement among bakeries in Sweden that bread labeled “whole
grain” should contain at least 50% whole grain. Moreover, in the
most recent update of the Keyhole symbol, and in addition to other
labeling requirements, a “rye bread” must contain at least 30% rye
to be labeled with the Keyhole (NFA, 2015b). Interestingly, however,
the white wheat whole-grain bread (F) was classified as the least
healthy by 71% of respondents despite having a Keyhole symbol.
Moreover, the Keyhole symbol was not mentioned in the re-
spondents’ descriptions of healthy bread. Prior research also has
shown limited or no effect of the Keyhole label on consumer
judgments of food healthfulness (Grunert, Ferna ndez-Celemín,
Wills, Bonsmann, & Nureeva, 2010; Orquin, 2014). Although the
label is well known to consumers, these have been shown to
question the credibility of the symbol and to have a low awareness
of the criteria for using the symbol in different food groups
(Enghardt-Barbieri, 2013).
Respondents who classified the brown wheat-rye bread as the
healthiest (more often older respondents, males and those with a
lower education level) e despite the bread's lack of a Keyhole
symbol and its high sugar content e may have been focusing more
on the bread's dark color. The appearance of the bread and espe-
cially the color has previously been shown to be an important cue
for perceived healthiness (Bucher et al., 2015; Sandvik et al., 2017).
The color has also shown to have a relatively greater impact on
perception of nutritional value than do price and nutrition infor-
mation combined (Peterson, 1977). Some of the elicited attributes
were intrinsic, and these were mainly related to texture (“coarse,”
“airy/fluffy texture”) and color (“dark,” “white”). A compact and
chewy texture has been shown to improve the glycemic properties
of bread (Eelderink et al., 2015), and the outer part of the cereal
kernel induces a darker color and may therefore be an indication of
whole grain content (Heinio € , Liukkonen, Katina, Myllyma €ki, &
Poutanen, 2003). However, a dark color may also be achieved by
using, for example, dark syrup, and white bread made with whole
grain is available. Although color may not be a reliable cue, some
sensory attributes e foremost texture (e.g., chewiness), but also
flavor (e.g., sourness) e have been shown to be potential aids to
consumers, in that they can serve as quality indicators of the
292 P. Sandvik et al. / Appetite 121 (2018) 285e293
glycemic properties of commercial rye bread (Sandvik et al., 2016).
Several previous studies have explored consumer's health-
related perceptions of food from an overall perspective, but the
present study also shows the value in focusing on a specific food
group. In recent years, bread has often been described as being
perceived as unhealthy by consumers (Gellynck et al., 2009; Serra-
Majem & Bautista-Castano, 2015; O'connor, 2012). The present
study however shows that consumers, from a health perspective,
differentiate between bread types, that a majority know of bread
which they consider healthy and can describe attributes they
perceived as more or less healthy in a bread. Results from the
present study indicate that questions such as “bread is good for
digestion” which was used by Gellynck et al. (2009) would be
difficult to answer since the consumer perceive some breads to be
good for digestion and other breads not to be.
The respondents could not touch the bread packages or taste the
bread, and their answers therefore reflect quality perceptions based
on overall attributes, prior to purchase. The choices of the most and
least healthy breads among the included samples represent the
combined effect of cues such as brand, color, packaging and labels
together with the respondents' previous experiences. In a real
world shopping situation, the number of breads available would
most often be many more than six and consumers often have
limited time to evaluate them all. The open-ended approach used in
the present study allowed an elicitation of attributes and cues on
top of consumers' minds when evaluating the healthiness of a
bread. These attributes would probably be of importance also in a
real life shopping situation. Some differences in mentioned attri-
butes could be identified in relation to age, gender and educational
level. For example, females more often than males described
“satiation” to be an important attribute of a healthy bread. A po-
tential drawback with this method however would be that some
consumers may be reluctant to answer open-ended comments. In
the present study, females, younger consumers and respondents
with a higher education level more often answered the open-ended
questions. Females and respondents with a higher educational level
also used somewhat more attributes in their descriptions. The
response rate should be considered low and limits the generaliz-
ability of the results, but is in line with other studies from the same
period (Amcoff et al., 2012; Dickson-Spillmann & Siegrist, 2011).
The participating respondents may have had a special interest in
bread or health, which may have biased the results. Even so, a large
sample was collected, both females and males were well repre-
sented in relation to the general Swedish population although the
response rate was lower in the younger consumers group and
among consumers with a lower educational level.
The present study contributes insight into the perceived
healthiness of bread among consumers by exploring which attri-
butes of bread consumers associate with different health effects
and, thereby, identifying potential pitfalls in consumer evaluations
of bread from a health perspective. To conclude, since one out of
four did not know any bread with they considered healthy there is a
need to increase consumer knowledge and awareness of how
different types of bread contribute to a healthy diet and how to
identify these types in the shopping situation; this need is espe-
cially marked among consumers with a lower education level.
Coarse, whole grain, fiber, sourdough and rye were attributes that
were perceived to be good for the stomach and bowel as well as to
have good satiation and glycemic properties. Several of the health
effects important to consumers cannot be communicated on the
packages. There are no approved health claims regarding e.g., the
satiation or glycemic properties of bread, two aspects mentioned
especially by younger consumers. Consumers have to use their own
cues to evaluate these health effects, and cues other than nutri-
tional labeling, for example color or the attributes rye and
sourdough, were often used. These are often not regulated and
consumers may therefore be misled. A “sourdough bread” could e.g.
contain a very low content of sourdough which not would affect the
glycemic properties of the bread. The findings are relevant to health
professionals working in dietary counseling when recommending
which type of bread should be chosen and how this can be done.
The results emphasize the importance of e.g. not only recom-
mending “rye bread,” whole-grain bread” or “sourdough bread” but
also to support the patients in studying the ingredient list to ensure
that the content actually is high. With regards to the intrinsic
sensory properties of the bread, it should be highlighted not only to
look at the color of the bread but to find bread with a dense, chewy
texture and a sour flavor. This has been shown to be useful in-
dicators for choosing bread with more beneficial glycemic proper-
ties (Sandvik et al., 2016). According to EU food law, information on
food packages should not be misleading. When designing packages
and labels, food companies should consider how consumes may
perceive the attributes communicated and if this could be
misleading from a health perspective. Attributes such as rye,
sourdough, whole grain, fiber seeds and kernels as well as sensory
attributes such as coarse and dark as well as a more home-made
look signals healthiness to the consumers. To ensure that con-
sumers are not misled when trying to identify healthier products,
this should be considered by authorities when drawing up regu-
lations on food labeling. The results also indicate a need to increase
consumers’ knowledge and trust in the Keyhole label. Further
research should expand the knowledge on how identification of
bread from a health perspective may be simplified for consumers
through labeling or other means especially focusing on health-
attributes that consumers finds important such as satiety and gly-
cemic properties.
References
Amcoff, E., Edberg, A., Enghardt Barbieri, H., Lindroos, A. K., Na€lsen, C., Pearson, M.,
et al. (2012). Riksmaten 2010-11, Food and nutrient intake among adults in Sweden
(in Swedish). Uppsala: National Food Administration. Retrieved from: http://
www.slv.se/upload/dokument/rapporter/mat_naring/2012/riksmaten_2010_
2011.pdf.
Anderson, J. W., Baird, P., & Davis, R. H., Jr. (2009). Health benefits of dietary fiber.
Nutrition Reviews, 67(4), 188e205.
Arvola, A., L€ ahteenma €ki, L., Dean, M., Vassallo, M., Winkelmann, M.,
Claupein, E., … Shepherd, R. (2007). Consumers' beliefs about whole and refined
grain products in the UK, Italy and Finland. Journal of Cereal Science, 46(3),
197e206. https://doi.org/10.1016/j.jcs.2007.06.001.
Aziz, I., Lewis, N. R., Hadjivassiliou, M., Winfield, S. N., Rugg, N.,
Kelsall, A., … Sanders, D. S. (2014). A UK study assessing the population prev-
alence of self-reported gluten sensitivity and referral characteristics to sec-
ondary care. European Journal of Gastroenterology & Hepatology, 26(1), 33e39.
Bech-Larsen, T., & Nielsen, N. A. (1999). A comparison of five elicitation techniques
for elicitation of attributes of low involvement products. Journal of Economic
Psychology, 20(3), 315e341. https://doi.org/10.1016/s0167-4870(99)00011-2.
Berg, C., Jonsson, I., Conner, M., & Lissner, L. (2003). Perceptions and reasons for
choice of fat- and fibre-containing foods by Swedish schoolchildren. Appetite,
40(1), 61e67. https://doi.org/10.1016/s0195-6663(02)00142-3.
Bisogni, C. A., Jastran, M., Seligson, M., & Thompson, A. (2012). How people interpret
healthy Eating: Contributions of qualitative research. Journal of Nutrition Edu-
cation and Behavior, 44(4), 282e301. https://doi.org/10.1016/j.jneb.2011.11.009.
Borra, S. T., & Bouchoux, A. (2009). Effects of science and the media on consumer
perceptions about dietary sugars. Journal of Nutrition, 139(6), 1214Se1218S.
https://doi.org/10.3945/jn.108.097915.
Bucher, T., Müller, B., & Siegrist, M. (2015). What is healthy food? Objective nutrient
profile scores and subjective lay evaluations in comparison. Appetite, 95,
408e414. https://doi.org/10.1016/j.appet.2015.08.005.
De Leeuw, E. D. (2005). To mix or not to mix data collection modes in surveys.
Journal of Official Statistics, 21(2), 233e255.
Dickson-Spillmann, M., & Siegrist, M. (2011). Consumers' knowledge of healthy
diets and its correlation with dietary behaviour. Journal of Human Nutrition and
Dietetics, 24(1), 54e60. https://doi.org/10.1111/j.1365-277X.2010.01124.x.
Eelderink, C., Noort, M. W., Sozer, N., Koehorst, M., Holst, J. J.,
Deacon, C. F., … Oudhuis, L. (2015). The structure of wheat bread influences the
postprandial metabolic response in healthy men. Food & Function, 6(10),
3236e3248.
EFSA. (2010). Scientific Opinion on the substantiation of health claims related to
 P. Sandvik et al. / Appetite 121 (2018) 285e293
whole grain (ID 831, 832, 833, 1126, 1268, 1269, 1270, 1271, 1431) pursuant to
Article 13(1) of Regulation (EC) No 1924/2006. EFSA Journal, 8(10). https://
doi.org/10.2903/j.efsa.2010.1766. 1766-n/a.
Enghardt-Barbieri, H. (2013). Views on healthy eating habits and dietary advice e
an evaluation of the National Food Agency's advice. (in Swedish) Synen på bra
matvanor och kostråd - en utva €rdering av Livsmedelsverkets råd. https://www.
livsmedelsverket.se/globalassets/rapporter/2013/2013_livsmedelsverket_22_u
tvardering_kostrad.pdf, 2017.09.06.
European Commission. (2011). Regulation (EU) No 1169/2011 of the European
Parliament and of the Council of 25 October 2011 on the provision of food in-
formation to consumers. Official Journal, L 304/18, 22/11/2011.
Fardet, A., Leenhardt, F., Lioger, D., Scalbert, A., & Remesy, C. (2006). Parameters
controlling the glycaemic response to breads. Nutrition Research Reviews, 19(1),
18e25. https://doi.org/10.1079/nrr2006118.
Gellynck, X., Kühne, B., Van Bockstaele, F., Van de Walle, D., & Dewettinck, K. (2009).
Consumer perception of bread quality. Appetite, 53(1), 16e23. https://doi.org/
10.1016/j.appet.2009.04.002.
Golley, S., Corsini, N., Topping, D., Morell, M., & Mohr, P. (2015). Motivations for
avoiding wheat consumption in Australia: Results from a population survey.
Public Health Nutrition, 18(3), 490e499. https://doi.org/10.1017/
S1368980014000652.
Grunert, K. G. (2002). Current issues in the understanding of consumer food choice.
Trends in Food Science and Technology, 13(8), 275e285.
Grunert, K., Ferna ndez-Celemín, L., Wills, J. M., Bonsmann, S. S., & Nureeva, L. (2010).
Use and understanding of nutrition information on food labels in six European
countries. Journal of Public Health, 18(3), 261e277.
Guerrero, L., Colomer, Y., Gu ardia, M. D., Xicola, J., & Clotet, R. (2000). Consumer
attitude towards store brands. Food Quality and Preference, 11(5), 387e395.
https://doi.org/10.1016/S0950-3293(00)00012-4.
Gunnarsson, A., & Elam, M. (2012). Food fight! The Swedish low-carb/high fat
(LCHF) movement and the turning of science popularisation against the sci-
entists. Science as Culture, 21(3), 315e334. https://doi.org/10.1080/
09505431.2011.632000.
Hayes, A. F., & Krippendorff, K. (2007). Answering the call for a standard reliability
measure for coding data. Communication Methods and Measures, 1(1), 77e89.
Heinio €, R.-L., Liukkonen, K.-H., Katina, K., Myllyma €ki, O., & Poutanen, K. (2003).
Milling fractionation of rye produces different sensory profiles of both flour and
bread. LWT - Food Science and Technology, 36(6), 577e583. https://doi.org/
10.1016/S0023-6438(03)00063-X.
Hoffman, D. L., & Franke, G. R. (1986). Correspondence analysis: Graphical repre-
sentation of categorical data in marketing research. Journal of Marketing
Research, 213e227.
Hough, G., & Ferraris, D. (2010). Free listing: A method to gain initial insight of a
food category. Food Quality and Preference, 21(3), 295e301. https://doi.org/
10.1016/j.foodqual.2009.04.001.
Isaksson, H., Fredriksson, H., Andersson, R., Olsson, J., & Åman, P. (2009). Effect of
rye bread breakfasts on subjective hunger and satiety: A randomized controlled
trial. Nutrition Journal, 8(1), 39. https://doi.org/10.1186/1475-2891-8-39.
Johansson, D. P., Lee, I., Rise rus, U., Langton, M., & Landberg, R. (2015). Effects of
unfermented and fermented whole grain rye crisp breads served as part of a
standardized breakfast, on appetite and postprandial glucose and insulin re-
sponses: A randomized cross-over trial. PLoS One, 10(3), e0122241. https://
doi.org/10.1371/journal.pone.0122241.
Kim, H., Patel, K., Orosz, E., Kothari, N., Demyen, M., Pyrsopoulos, N., et al. (2016).
Time trends in the prevalence of celiac disease and gluten-free diet in the us
population: Results from the national health and nutrition examination surveys
2009-2014. JAMA Internal Medicine, 176(11), 1716e1717. https://doi.org/10.1001/
jamainternmed.2016.5254.
Krippendorff, K. (2004). Reliability in content analysis. Human Communication
Research, 30(3), 411e433. https://doi.org/10.1111/j.1468-2958.2004.tb00738.x.
Kyrø, C., Skeie, G., Dragsted, L. O., Christensen, J., Overvad, K.,
Hallmans, G., … Olsen, A. (2012). Intake of whole grain in Scandinavia: Intake,
sources and compliance with new national recommendations. Scandinavian
Journal of Public Health, 40(1), 76e84. https://doi.org/10.1177/
1403494811421057.
€hteenma
La €ki, L. (2015). Consumer interpretation of nutrition and other information
on food and beverage labels. In P. Berryman (Ed.), Advances in food and beverage
labelling (pp. 133e148). Oxford: Woodhead Publishing.
Liljeberg, H., Lo € nner, C., & Bjo €rck, I. (1995). Sourdough fermentation or addition of
organic acids or corresponding salts to bread improves nutritional properties of
starch in healthy humans. The Journal of Nutrition, 125(6), 1503e1511.
Lyly, M., Soini, E., Rauramo, U., & La €hteenma €ki, L. (2004). Perceived role of fibre in a
healthy diet among Finnish consumers. Journal of Human Nutrition and Dietetics,
17(3), 231e239.
Mo €tteli, S., Keller, C., Siegrist, M., Barbey, J., & Bucher, T. (2016). Consumers' practical
understanding of healthy food choices: A fake food experiment. British Journal
of Nutrition, 116(3), 559e566. https://doi.org/10.1017/S0007114516002130.
Nagler, R. H. (2014). Adverse outcomes associated with media exposure to con-
tradictory nutrition messages. Journal of Health Communication, 19(1), 24e40.
https://doi.org/10.1080/10810730.2013.798384.
NFA. (2005). Regulations amending the National Food Agency's regulations (SLVFS
2005:9) on the use of a particular symbol (SLVFS 2005:9). Retrieved from: http://
www.livsmedelsverket.se/globalassets/om-oss/lagstiftning/livsmedelsinfo-till-
konsume-markning/livsfs-2015-1-particular-symbol-eng.pdf.
NFA. (2015a). Wholemeal - advice. Retrieved from: http://www.livsmedelsverket.
293
se/en/food-habits-health-and-environment/dietary-guidelines/vuxna/whole
meal/.
NFA. (2015b). Regulations amending the National Food Agency's regulations (SLVFS
2005:9) on the use of a particular symbol (SLVFS 2015:1).
Nordic Council of Ministers. (2014). Nordic nutrition recommendations 2012: Inte-
grating nutrition and physical activity. S.l.: Nordic Council of Ministers.
O'connor, A. (2012). Bread consumption in the UK: What are the main attitudinal
factors affecting current intake and its place in a healthy diet? Nutrition Bulletin,
37(4), 368e379. https://doi.org/10.1111/j.1467-3010.2012.01989.x.
Orquin, J. L. (2014). A Brunswik lens model of consumer health judgments of
packaged foods. Journal of Consumer Behaviour, 13(4), 270e281.
neau, S., Linke, A., Escher, F., & Nuessli, J. (2009). Freshness of fruits and vege-
Pe
tables: Consumer language and perception. British Food Journal, 111(3),
243e256. https://doi.org/10.1108/00070700910941453.
Perreault, W. D., Jr., & Leigh, L. E. (1989). Reliability of nominal data based on
qualitative judgments. Journal of Marketing Research, 26(2), 135e148.
Peterson, R. A. (1977). Consumer perceptions as a function of product color, price
and nutrition labeling. In William D. Perreault, Jr. (Ed.), NA - advances in con-
sumer research (Vol. 04, pp. 61e63). Atlanta, GA: Association for Consumer
Research.
Pollard, C. M., Pulker, C. E., Meng, X., Scott, J. A., Denham, F. C., Solah, V. A., et al.
(2017). Consumer attitudes and misperceptions associated with trends in self-
reported cereal foods consumption: Cross-sectional study of Western Austra-
lian adults, 1995 to 2012. BMC Public Health, 17(597). https://doi.org/10.1186/
s12889-017-4511-5.
Poutanen, K., Sozer, N., & Della Valle, G. (2014). How can technology help to deliver
more of grain in cereal foods for a healthy diet? Journal of Cereal Science, 59(3),
327e336. https://doi.org/10.1016/j.jcs.2014.01.009.
Rakha, A., Åman, P., & Andersson, R. (2010). Characterisation of dietary fibre com-
ponents in rye products. Food Chemistry, 119(3), 859e867. https://doi.org/
10.1016/j.foodchem.2009.09.090.
Rose €
n, L., Silva, L., Andersson, U., Holm, C., Ostman, E., & Bjo€rck, I. (2009). Endo-
sperm and whole grain rye breads are characterized by low post-prandial in-
sulin response and a beneficial blood glucose profile. Nutrition Journal, 8(1), 42.
https://doi.org/10.1186/1475-2891-8-42.
Sandvik, P., Kihlberg, I., Lindroos, A. K., Marklinder, I., & Nydahl, M. (2014). Bread
consumption patterns in a Swedish national dietary survey focusing particu-
larly on whole-grain and rye bread. Food & Nutrition Research, 58. https://
doi.org/10.3402/fnr.v58.24024.
Sandvik, P., Marklinder, I., Nydahl, M., Næs, T., & Kihlberg, I. (2016). Characterization
of commercial rye bread based on sensory properties, fluidity index and
chemical acidity. Journal of Sensory Studies., 31(4), 283e295. https://doi.org/
10.1111/joss.12211.
Sandvik, P., Nydahl, M., Marklinder, I., Naes, T., & Kihlberg, I. (2017). Different liking
but similar healthiness perceptions of rye bread among younger and older
consumers in Sweden. Food Quality and Preference, 61, 26e37. https://doi.org/
10.1016/j.foodqual.2017.04.016.
Serra-Majem, L., & Bautista-Castano, I. (2015). Relationship between bread and
obesity. British Journal of Nutrition, 113(S2), S29eS35. https://doi.org/10.1017/
S0007114514003249.
Siipi, H. (2013). Is natural food healthy? Journal of Agricultural and Environmental
Ethics, 26(4), 797e812. https://doi.org/10.1007/s10806-012-9406-y.
Slavin, J., Tucker, M., Harriman, C., & Jonnalagadda, S. S. (2013). Whole grains:
Definition, dietary recommendations, and health benefits. Cereal Foods World,
58(4), 191e198.
Smith, V., Clement, J., Møgelvang-Hansen, P., & Sørensen, H. S. (2011). Assessing in-
store food-to-consumer communication from a fairness perspective: An inte-
grated approach. Fachspracheeinternational Journal of Specialized Communica-
tion, 33(1e2), 84e106.
Sørensen, H. S., Clement, J., & Gabrielsen, G. (2012). Food labels e an exploratory
study into label information and what consumers see and understand. The
International Review of Retail, Distribution and Consumer Research, 22(1),
101e114. https://doi.org/10.1080/09593969.2011.634072.
Spiggle, S. (1994). Analysis and interpretation of qualitative data in consumer
research. Journal of Consumer Research, 21(3), 491e503. https://doi.org/10.2307/
2489688.
Steenkamp, J. B. E. M. (1990). Conceptual model of the quality perception process.
Journal of Business Research, 21(4), 309e333.
Sütterlin, B., & Siegrist, M. (2015). Simply adding the word “fruit” makes sugar
healthier: The misleading effect of symbolic information on the perceived
healthiness of food. Appetite, 95, 252e261. https://doi.org/10.1016/
j.appet.2015.07.011.
Symoneaux, R., Galmarini, M. V., & Mehinagic, E. (2012). Comment analysis of
consumer's likes and dislikes as an alternative tool to preference mapping. A
case study on apples. Food Quality and Preference, 24(1), 59e66. https://doi.org/
10.1016/j.foodqual.2011.08.013.
Velardo, S. (2015). The nuances of health literacy, nutrition literacy, and food lit-
eracy. Journal of Nutrition Education and Behavior., 47(4), 385e389. https://
doi.org/10.1016/j.jneb.2015.04.328.
Violette, C., Kantor, M. A., Ferguson, K., Reicks, M., Marquart, L., Laus, M. J., et al.
(2016). Package information used by older adults to identify whole grain foods.
Journal of Nutrition in Gerontology and Geriatrics, 35(2), 146e160. https://
doi.org/10.1080/21551197.2016.1168759.
Wolever, T., & Miller, J. B. (1995). Sugars and blood glucose control. The American
Journal of Clinical Nutrition, 62(1), 212Se221S.