Clinical Nutrition ESPEN xxx (2018) e1ee7

Contents lists available at ScienceDirect

Clinical Nutrition ESPEN

journal homepage: http://www.clinicalnutritionespen.com

Original article

Acceptance of texture-modified in-between-meals among old adults

with dysphagia
S.L. Okkels a, b, *, M. Saxosen a, S. Bügel b, A. Olsen c, T.W. Klausen d, A.M. Beck a, e
a
Dietetic and Nutritional Research Unit, Gentofte University Hospital, Gentofte, Denmark
b
Department of Nutrition, Exercise and Sports, Faculty of Science, University of Copenhagen, Frederiksberg C., Denmark
c
Department of Food Science, Faculty of Science, University of Copenhagen, Frederiksberg C., Denmark
d
Department of Hematology, Herlev University Hospital, Herlev, Denmark
e
Metropolitan University, Faculty of Health and Technology, Department of Nutrition and Midwifery, Copenhagen, Denmark

a r t i c l e i n f o s u m m a r y

Article history: Background and aims: Old adults suffering from dysphagia have difficulties swallowing, chewing and/or
Received 18 December 2017 eating, and are therefore at high risk of undernutrition. In-between-meals that are texture modified are
Accepted 17 March 2018 of particular importance for people suffering from dysphagia. To meet their adequate daily amount of
food intake they are recommended to eat 3e5 in-between-meals daily. The aim of the current pilot study
Keywords: was to identify the most liked in-between-meals for old adults based on flavour and describe the basic
Minimal Eating Observational Form (MEOF)
sensory properties of these in-between-meals. Following, the equality between flavour and appearance-
Eating Validation Scheme (EVS)
based preferences was investigated.
Liking
Nursing home residents
Methods: From three nursing homes 30 old adults aged 70 years or older suffering from dysphagia were
Protein enriched products recruited. They were assessing 20 texture modified in-between-meals based on their flavour and
Undernutrition appearance on a 3 point hedonic scale.
Results: When participants were asked to assign liking based on flavour, the most liked in-between-
meals were frozen, cold and sweet (vanilla ice cream, strawberry parfait and panna cotta). These
meals were among the in-between-meals richest in fat and energy. Liking based on flavour and
appearance was equal in 18 out of 20 samples. Furthermore, nutritional and sensory characteristics of the
preferred meals were described.
Conclusion: Flavour and sensory-based ranking of in-between-meals opens the possibility to design new
in-between-meals to old adults with dysphagia, by choosing the most liked in-between-meals to offer
the target group.
© 2018 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights
reserved.

1. Introduction neurological conditions, trauma, disease, surgery or other addi-

tional diagnoses. The simple fact of aging is another reason for
Dysphagia is defined as a condition related to eating or drink- dysphagia, because the muscle mass and strength related to
ing, impairing a person's ability to swallow or chew [1]. It is swallowing, changes naturally with time [5].
estimated that approximately 15% of community dwelling, old Dysphagia is a significant factor among the causes of under-
adults are affected by dysphagia [2]. Dysphagia is also very nutrition in old adult individuals [6] and has been associated with
common among patients [3] and nursing home residents [4]. increased mortality, partly due to the increased risk of pneumonia
The group of people having dysphagia is often cognitively [7]. The European society for clinical nutrition and metabolism
impaired due to dementia and the wide variety of medical con- (ESPEN) has made guidelines on nutrition for people with de-
ditions causing the dysphagia, including acute or progressive mentia [8] and elderly in risk of undernutrition (Beck, AM per-
sonal communication, October 23, 2017). Both recommend that
in-between-meals should be provided to meet an adequate food
* Corresponding author. Dietetic and Nutritional Research Unit, Gentofte Uni- intake of the individual. In Denmark the Danish Ministry of Food
versity Hospital, Gentofte, Denmark. has made this recommendation very clear and states that old
E-mail address: [email protected] (S.L. Okkels).

https://doi.org/10.1016/j.clnesp.2018.03.119
2405-4577/© 2018 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Okkels SL, et al., Acceptance of texture-modified in-between-meals among old adults with dysphagia, Clinical
Nutrition ESPEN (2018), https://doi.org/10.1016/j.clnesp.2018.03.119
e2 S.L. Okkels et al. / Clinical Nutrition ESPEN xxx (2018) e1ee7

adults with eating problems (including people with dysphagia), difficulties and nutritional status together with a research assis-
should eat at least three high in protein and energy in-between- tant (MS) by observations and interviews of the participants with
meals a day [9]. In order to ensure safety and proper nutrition the use of Minimal Eating Observational Form (MEOFII) [12] and
for people suffering from dysphagia, texture modified food (TMF) Eating Validation Scheme (EVS) [13].
with a special consistency has been created. TMF is one among
different nutritional interventions to be used for people with 2.2.1. Minimal Eating Observational Form
dysphagia [10]. In this pilot study the texture types of the in-be- The state of eating difficulties was identified through the
tween-meals were either minced & moist or pureed. All the screening tool Minimal Eating Observational Form (MEOF) II. This is
included participants were able to eat these texture types. In order a categorical scale tested for reliability and validity in nursing home
to have puree texture the liquids were adjusted with a thickening residents with dysphagia. Besides problems with chewing and
agent (liquids should be taken with a spoon). swallowing the scale includes difficulties with ingestion and energy
Sensory evaluations of different Texture Modified Foods have in eating the meal. The scale ranges from 1 to 9, where to a higher
been done with meat products, carrot-based foods, soups, pa ^te
s number corresponds a worse disability [12].
and timbales [11,12]. Despite the fact that people with dysphagia is
one of the most vulnerable groups in regards to undernutrition [5], 2.2.2. Eating Validation Scheme
none of these studies have evaluated which TMF in-between-meals Nutritional assessments were performed using the Eating Vali-
this group prefer. Therefore, these people were the focus of this dation Scheme (EVS), which in Denmark is recommended for old
study. adults receiving home care and nursing home residents [9]. The
The aim of the current pilot study was to identify the most liked EVS assessment involves questions of eating habits, nutritional risk
in-between-meals for old adults based on flavour and describe the factors and weight loss. The evaluation can result in three different
basic sensory properties of these in-between-meals. Following, the scores: 0 for participants not at risk of being undernourished; 1
equality between flavour and appearance-based preferences was means that there is a risk of undernourishment; 2 means that the
investigated. participant will benefit from a nutritional intervention [13]. Since
chewing and swallowing problems are among the included nutri-
2. Materials and methods tional risk factors dysphagia leads to a risk of undernourishment. A
participant suffering from this condition will have an EVS score of at
2.1. Participants least 1.

To be included in the pilot study participants should be: 2.3. Study design

(1) older than 70, 2.3.1. Selection of in-between-meals

(2) having the symptom dysphagia, assessed by an occupational In order to create the in-between-meals to be tested, 20 of the
therapist or the nursing staff at the nursing home, most frequently ordered in-between-meals for old adults with
(3) able to understand Danish, small appetite were chosen from two hospitals and one municipal
(4) able to make assessments of appearance and taste, kitchen menus. A variety of in-between-meals of different flavours,
(5) able to eat Texture Modified Foods. colours and food types were chosen, all with high protein and
energy contents, ranging from mousses, porridges and soups. Many
On the other hand participants were excluded from the study if of the in-between-meals were traditional, Danish in-between-
they were: meals. For example, rye bread soup is a popular breakfast among
old, Danish adults.
(1) in terminal condition,
(2) having food allergies or intolerances, 2.3.2. Texture adaptation
(3) exclusively receiving enteral or parenteral nutrition. Since the in-between-meals were targeting old adults with
dysphagia the consistency chosen was similar to TMF, having
Dementia was not considered an exclusion criterion, since the minced & moist or puree texture. Texture descriptors were done
tests of the pilot study did not require high functions of memory, according to Danish standardizations [9], very similar to the
concentration or language abilities. framework from the International Dysphagia Diet Standardization
Initiative (IDDSI) [14].
2.2. Data collection To meet this requirement, when needed some meals were pu-
reed or blended. Furthermore, the vanilla ice cream and the
The pilot study took place in the spring 2016 at three nursing strawberry parfait used in the pilot study were in a special texture
homes in three different municipality areas in Denmark. A that remained pureed after melting.
designated contact person in each nursing home identified par-
ticipants that fulfilled the inclusion criteria for the pilot study. 2.3.3. Enrichment
Two research assistants were connected to the pilot project (MS Based on the original recipe, some meals were protein enriched
and SLO). These researchers approached the participants in the with Adosan® protein powder (100% protein) from veal or Arla®
nursing homes, and introduced them to the pilot study by oral and Protino (80% protein) from whey origin. Each in-between-meal
written information. Informed consent was obtained from all contained protein and energy varying between 1.9 and 9.7 g and
participants. The consent form was signed by next of kin in cases 165e1409 kJ per portion. The soups and the mashed potato dish
where participants were not able to sign themselves. At the time were chosen to be a part of the in-between-meal offer, to include a
of inclusion in the pilot study, a contact person of the nursing variety of flavours and temperatures in the menu tested. The in-
home collected the following data from the participants' medical between-meals of liquid consistency and water used for mouth
record: Age, diagnosis, registration of dementia and mode of rinse between tasting were adjusted with a thickening agent to
feeding and drinking. The procedure was done in collaboration have the desired consistency (Ressource, ThickenUp Clear, Nestle 
with SLO. Furthermore SLO performed assessments of eating Institute of Health Sciences, Switzerland).

Please cite this article in press as: Okkels SL, et al., Acceptance of texture-modified in-between-meals among old adults with dysphagia, Clinical
Nutrition ESPEN (2018), https://doi.org/10.1016/j.clnesp.2018.03.119
 S.L. Okkels et al. / Clinical Nutrition ESPEN xxx (2018) e1ee7 e3

2.3.4. Storage the nursing homes where breakfast at 8.30 AM, lunch at 12 AM and
For practical reasons, the in-between-meals were stored dinner at 18 PM. The normal assortment of in-between-meals were
frozen in plastic cups (ca. 100 g each) and thawed in refrigerator served in the afternoon or late evening and included cake, sand-
the day before serving. Before serving soups, mashed potato, rice wiches and protein drinks.
pudding and rye bread soup were warmed approximately 2 min To control for carry-over effect, participants were served the
in a microwave oven at 400 Watt until it reached a temperature samples of in-between-meals in randomized order. The program
of approximately 45
C. Strawberry parfait and vanilla ice cream “research randomizer” was used to find the exact order of the 20 in-
were served frozen after being lightly tempered at room between-meals for each participant. Two research assistants (MS
temperature. The rest of the meals were served cold at approx- and SLO) collected the data by filling in the answers on the ques-
imately 15
C. Full recipes are available from the first author upon tionnaire. As far as practically possible, the same research assistant
request. interviewed participants on all four days to make the participant
feel as comfortable as possible.
2.3.5. Sprinkles The participant was instructed to clear their mouth with a
Six of the meals were garnished to make them look more spoonful of water (thickened if needed) between every taste.
appealing: The milkshake was sprinkled with raspberry dust, the
rum mousse with buckthorn dust, the panna cotta with blueberry
dust and the mashed potato and pumpkin soup with parsley dust. 2.6. Statistical analysis
Furthermore, the rice porridge was served with cinnamon sugar.
Statistical analysis was performed using IBM SPSS Statistics
2.3.6. Layers version 22.3, and Graphpad Prism version 6.
The following seven in-between-meals were made in layers: the Data was tested for normal distribution and T-test was used as a
rye bread soup with whipped cream, the prune trifle, raspberry robust and valid tool when analysing these data [18]; [19]. A paired
jelly, strawberry porridge, apple porridge with vanilla cream and T-test was performed to investigate significant differences between
the mashed potato with bacon and onion puree. appearance and flavour ratings of the 20 in-between-meals. These
statistical tests and P-values were two-sided, with P < 0.05
2.3.7. Basic sensory characterization considered statistically significant. Characteristics of the old adults
Two people with culinary and sensory experience (MS and SLO) were presented as means ± standard deviation (SD) or numbers
evaluated each in-between-meal during a tasting session, and and percentages (%). When dividing ranked flavour liking against
selected the sensory descriptors to be used in the pilot study ac- nutritional content (protein, fat and energy respectively) the data
cording to predominant basic taste and serving temperature. The was non-parametric distributed and correlations were calculated
predominant basic tastes in the 20 in-between-meals were divided using Spearman's rank correlation coefficient. When dividing
in three: salty, sour and sweet, as well as the serving temperatures flavour liking according to sensory characteristics (predominant
warm, cold or frozen. basic taste, serving temperature, layer and sprinkles) Kruskal Walis'
test was adopted when testing the difference between these (one-
2.4. Questionnaire design way ANOVA on rank). A simple generalized estimation equation
was done to test whether the degree of appetite (independent
The questionnaire for the in-between-meal evaluation was variable) influenced the rating of flavour (dependent variable).
developed in cooperation with health care professionals and the
authors, pilot tested with different sizes of scales, and constructed
in Danish. Since appetite could be a confounder influencing the 2.7. Ethical committee
flavor preference [15], this aspect was assessed in order to be able
to make correlation analysis on how it influenced the results. The After reviewing the study protocol, the Danish Ethical Com-
questionnaire consisted of items regarding level of appetite, mittee concluded that formal approval of the pilot study was not
appearance and flavour of each sample. A 3-point Likert scale with required (ref number H-16019673).
symbols representing facial expressions and the response cate-
gories: “Bad”, “Neither good nor bad”, and “Good” were chosen to
make the questionnaire simple and understandable for the target 3. Results
group [16].
Using a simple category scale, the participant was asked at the 3.1. Participants
beginning of the interview: “how hungry do you feel right now?”.
Following, before tasting the sample, the participant was asked: Fiftyfive old adults were informed about the pilot study and
“how much do you like the appearance of the sample?”. And after screened for inclusion and exclusion criteria. Of these twelve did
tasting a teaspoon of the sample “how much do you like the flavour not wish to participate and thirteen did not meet the inclusion
of the sample?” Participants were instructed on how to differen- criteria. Subsequently, thirty old adults with dysphagia from three
tiate their answers. different nursing homes accepted to participate. The primary cau-
ses of the dysphagia in these participants were neurological dis-
2.5. Tasting session eases like stroke or Parkinson's disease. Three of the participants
only completed a partial amount of tasting sessions because of
Each participant tasted and rated 5 different in-between-meals unwillingness (N ¼ 1) or tiredness (N ¼ 2).
once per day for four consecutive days, for a total of 20 in-between- A high degree of the participants suffered from dementia (83%),
meals. The session was carried out like an interview and lasted 63% of them would benefit from a nutritional intervention (EVS
approximately 15 min per day. Since it was previously shown that score 2), and with regard to the extent of eating difficulties (MEOF
appetite for in-between-meals is higher in the morning than the II) participants fell into three almost evenly split groups. Average
afternoon, the time point of the tasting session was chosen to be age of the participants (years ±SD) was 84 ± 7.8, and average body
between 9.30 and 11.30 AM [17]. Other time points of main meals at mass indexes 23.7 ± 4.8 (BMI ±SD) (Table 1).

Please cite this article in press as: Okkels SL, et al., Acceptance of texture-modified in-between-meals among old adults with dysphagia, Clinical
Nutrition ESPEN (2018), https://doi.org/10.1016/j.clnesp.2018.03.119
e4 S.L. Okkels et al. / Clinical Nutrition ESPEN xxx (2018) e1ee7

Table 1 classified according to its protein, fat and energy content pr. 100 g.
Characteristics of the 30 old adults regarding Age, Body Mass Indexes (BMI), Nursing In order to separate the meals predominant basic taste was defined
home, Gender, Dementia, Usage of thickening agents, Level of eating difficulties
(MEOF II) and Nutritional status (EVS). Data are presented as N (%) and average
for each meal according to its salty, sour or sweet taste. Serving
(±SD). temperature was divided in warm, cold or frozen, and appearance
was defined depending on whether the food included layers and/or
Characteristics Average (±SD)
sprinkles (Table 2).
Age 84 (±7.8) Ranked means of flavour liking are seen in Fig. 1A. The in-
Body Mass Indexes (BMI) 23.7 (±4.8)
between-meals that scored highest were vanilla ice cream, straw-
Number of participants N (%)
Nursing home 1 4 (13) berry parfait and panna cotta. At the lower end of the spectrum
Nursing home 2 16 (54) were the pumpkin soup, carrot soup and clear soup. This ranking
Nursing home 3 10 (33) was used to test the correlation of mean liking of flavour against
Females 19 (63)
nutritional factors of the in-between-meals. A significant spear-
Males 11 (37)
Dementia 25 (83)
man's rank correlation coefficient was identified for fat content and
Using thickening agent in fluids 5 (17) flavour liking (r ¼ 0.5706, p-value 0.0086, Fig. 1B) as well as energy
Minimal Eating Observational Form (MEOF)II content and flavour liking (r ¼ 0.7414, p-value 0.0002, Fig. 1C), both
Number 1e3 8 (27) resulting in a higher flavour liking.
Number 4e6 10 (33)
No correlation between protein content and flavour liking could
Number 7e9 12 (40)
Eating Validation Scheme (EVS) be seen (r ¼ 0.0339, p-value not significant (0.8871), Fig. 1D).
Score 1 (at risk of undernutrition) 11 (37)
Score 2 (benefit from intervention) 19 (63)

3.3.2. Flavour liking according to sensory characteristics

Spearman's rank correlation coefficient of the mean flavour
3.2. The influence of appetite liking was tested according to the sensory characteristics (serving
temperature, layer and sprinkles, Table 1). Significant differences
A high liking on flavour concomitant with a high degree of were found with serving temperature (frozen, cold, warm). The
appetite could indicate a bias in the data, therefore a simple lower temperature the higher liking, r ¼ 0.822, p < 0.0001. There
generalized estimation equation was adopted to test whether the were no significant differences with respect to in-between-meals
degree of appetite (independent variable with 20X30 observations) with/without layers or with/without sprinkles (data not shown).
influenced the flavour rating (dependent variable with 20X30 ob-
servations). The P-value of the model was 0.9928, indicating that
the two parameters were not associated. 3.3.3. Predominant basic taste
Fig. 2 shows flavour liking average groupings according to pre-
3.3. Preferred in-between-meals and sensory classifications dominant basic taste (sour, salty or sweet). In-between-meals of
sour and sweet taste scored significantly higher than in-between-
3.3.1. Flavour liking meals of the salty taste. No significant difference was found when
In Table 2 each in-between-meal was ranked by mean flavour comparing flavour ratings between sweet and sour in-between-
liking from lowest to highest (assessed by the participants) and meals.

Table 2
Classification of 20 in-between-meals including: In-between-meal name, protein, fat and energy content pr. 100 g, predominant basic taste, serving temperature, layers and
sprinkles. Ranked by mean flavour preference.

In-between-meal name Protein pr. Fat pr. Energy pr. Predominant Serving tempb Layersc Sprinklesd
100 grams (g) 100 grams (g) 100 gram. (kJ) basic tastea

Pumpkin soup 4.3 3.9 258 1 1 1 2

Carrot soup 6.4 9.4 557 1 1 1 1
Clear soup 7.6 0.7 165 1 1 1 1
Mashed potato with bacon and onion 9.5 20.0 995 1 1 2 2
Rice porridge with cinnamon sugar 3.6 4.1 429 1 1 1 2
Soup of asparagus with chicken 6.4 6.3 390 1 1 1 1
Milkshake 9.2 0.9 363 2 2 1 2
Rum mousse 5.5 19.6 1226 3 2 1 2
Apple porridge with vanilla cream 1.9 2.4 304 2 2 2 1
Protein drink 5.9 5.8 630 3 2 1 1
Rye bread soup with whipped cream 4.4 5.7 575 1 1 2 1
Raspberry jelly with vanilla cream 4.4 6.2 564 2 2 2 1
Prune trifle 3.8 5.7 630 2 2 2 1
Yoghurt/strawberry drink 9.7 11.4 846 2 2 1 1
Strawberry porridge with vanilla cream 4.6 7.7 649 2 2 2 1
Chocolate mousse 4.3 18.4 1409 3 2 1 1
Lemon Mousse 5.4 17.8 1112 2 2 1 1
Panna cotta 8.0 16.2 967 3 2 2 2
Strawberry parfait 4.6 19.8 1043 3 3 1 1
Vanilla ice cream 8.8 24.5 1292 3 3 1 1
a
Predominant basic taste: Salt ¼ 1, Sour ¼ 2, Sweet ¼ 3.
b
Serving temperature: Warm ¼ 1, Cold ¼ 2, Frozen ¼ 3.
c
Layers: 1 ¼ no layer, 2 ¼ layer.
d
Sprinkles: 1 ¼ no sprinkles, 2 ¼ sprinkles.

Please cite this article in press as: Okkels SL, et al., Acceptance of texture-modified in-between-meals among old adults with dysphagia, Clinical
Nutrition ESPEN (2018), https://doi.org/10.1016/j.clnesp.2018.03.119
 S.L. Okkels et al. / Clinical Nutrition ESPEN xxx (2018) e1ee7 e5

Fig. 1. Ranked mean flavour liking (A), Correlation of ranked in-between-meal flavour liking average and fat pr. 100 g (B), energy pr. 100 g (C), and protein pr. 100 g (D). Correlations
were operated using Spearman's rank correlation coefficient. Significant correlations indicated with asterisks ***P < 0.001, **P < 0.01.

3.3.4. Flavour vs. appearance liking 4. Discussion

A paired T-test comparing the flavour liking with appearance
liking showed no significant difference in 18 of 20 meals. Only two 4.1. Most liked in-between-meals
meals (clear soup and rum mousse) showed a significantly higher
liking of appearance with P-value 0.047 and 0.003 respectively The aim of the current pilot study was to identify the most liked
(data not shown). in-between-meals for old adults based on flavour and describe the

Please cite this article in press as: Okkels SL, et al., Acceptance of texture-modified in-between-meals among old adults with dysphagia, Clinical
Nutrition ESPEN (2018), https://doi.org/10.1016/j.clnesp.2018.03.119
e6 S.L. Okkels et al. / Clinical Nutrition ESPEN xxx (2018) e1ee7

definition of an appealing in-between-meal appearance differs

from the old adults', where other factors such as transparency
and familiarity may play more important roles for this age group.
One of the reasons for this could be due to a worsening in the
eye-sight of these old adults. Finally, layers and sprinkles are
cosmetic factors that simply may not influence the overall
appearance to any measurable degree in this specific group of
participants.

4.3. Flavour vs. appearance

The equality between flavour and appearance-based prefer-

ences was investigated. Flavour means the synthesis of taste
(sweet, sour, salty, bitter and umami), the odour and temperature of
a food, and flavour is the most desired aspect of foods that con-
sumers with dysphagia have [29]. It is of crucial importance,
however, in order for food to be entirely appreciated that the
flavour of the food is not different from what is expected from its
appearance [31]. The food should be presented in a way that
Fig. 2. Flavour liking average groupings divided in sour, salty and sweet in-between-
meals. Comparison between groups operated using Kruskal Wallis test. Significant stimulates the appetite, especially in this group of people. For all
differences indicated with asterisks ***P < 0.001. stimuli, except rum mousse and clear soup the evaluations of flavor
and appearance were in accordance indicating that they delivered
the flavour expected from their appearance. This means that there
basic sensory properties of these in-between-meals. The most liked is good reason to believe that these in-between-meals would serve
in-between-meals (vanilla ice cream, strawberry parfait and panna well in overall acceptance.
cotta) were all dessert like in the sense that they were cold, frozen
and predominantly sweet. Although this is the first study to eval- 4.4. Nutritional content
uate hedonic measures of in-between-meals in this particular
group of old people with dysphagia, the findings are in line with Undernourishment is a common problem in old adults suffering
studies of in-between-meal preferences in patients and nursing from dysphagia, because the foods offered are not meeting their
home residents who had low appetite and maybe dysphagia nutritional and sensory needs [32]. Creating nutritious in-between-
[17,20e22] where ice cream, fresh fruit and in-between-meals of meals that fit the target groups' preferences may be one way to
cold temperature, sour and sweet taste were also preferred. In the improve their nutritional intake. When looking at the nutritional
context of this some studies have shown that elderly compared to content of the most liked in-between-meals there was a significant
younger people prefer sweeter foods [23,24] and this was sup- correlation between mean flavour preference and fat and energy
ported in a recent review [25] which showed that in dependent old content. This is promising since these nutrients are of high
adults, sweet taste enhancement had positive effect on general food importance for old adults with dysphagia, who are often under-
acceptance. nourished, and in need of these nutrients. On the other hand, no
However, the previous studies in patients also observed pref- correlation was seen with regards to the flavour liking and the
erences for savory in-between-meals [17,21], which was in protein content of the in-between-meals. The amount of protein
contrast to the findings in our pilot study where meals of pre- intake for old adults is crucial because of their increased level of
dominant sweet and sour taste were liked significantly more than muscle degradation [5], which creates room for further improve-
those with predominant salty taste. This is supported by studies of ment of the flavour of the in-between-meals with high protein
institutionalized, community dwelling and independent old content.
adults [26e28] where the savory and salty food products were
ranked lowest on liking. The inconsistencies of the findings might 4.5. Strengths and limitations
be explained by different preferences of the healthier participants
in the studies by Beelen et al., 2017 [26], Kremer et al., 2007 [27] Because there may be special food related issues connected to
and Mojet et al., 2004 [28] compared to the old adults in our pilot each population it is important to use the same participants in the
study and the patients in the studies by Price et al., 2006 [23] and meal study as the population for whom the meals are intended. The
Olin et al., 1996 [21]. Furthermore, differences in time periods of target group of this pilot study was old adults with dysphagia, who
the studies might also be the reason why the results are not could have sensory impairments [33]. Therefore, it is a strength that
similar. old adults with dysphagia are the participants of the study. This is,
to our knowledge the first time, the liking of in-between-meals
4.2. Appearance liking have been tested in old adults with dysphagia, although it has
previously been suggested [26,32].
The appearance of a meal (colour and the way it is served) Normally, hedonic scales for young and healthy adults are
plays a significant role in evaluating foods for old adults with differentiated into 9 response categories, but the less capable the
dysphagia [29,30]. Therefore, the in-between-meals chosen for participants are at differentiating, the fewer categories are recom-
this pilot study were colorful, sprinkled with green herb, purple mended [34]. Due to the weakness of the participants, the method
or red berry dust, or arranged in layers with the intention to and questionnaire had to be designed in a very simple way, to make
make the appearance more appealing. When comparing the most the answers reliable. Thus, in order for most of the participants to
liked in-between-meals taking into consideration sprinkles and be able to answer, the method designed for this study involved a
layer differentiation as key factors, however, no correlation was hedonic scale consisting of 3 response categories. Other studies on
found. It is possible that the researchers' interpretations on the the issue of TMF have so far used participants that are healthier, or

Please cite this article in press as: Okkels SL, et al., Acceptance of texture-modified in-between-meals among old adults with dysphagia, Clinical
Nutrition ESPEN (2018), https://doi.org/10.1016/j.clnesp.2018.03.119
 S.L. Okkels et al. / Clinical Nutrition ESPEN xxx (2018) e1ee7 e7

they are using the opinions from nursing staff or sensory focus [4] Steele CM, Greenwood C, Ens I, Robertson C, Seidman-Carlson R. Mealtime
difficulties in a home for the aged: not just dysphagia. Dysphagia 1997;12(1):
groups of which in-between-meals the old adults or patients prefer
43e50.
[11,30,32]. A recent study found that nursing home residents with [5] Serra-Prat M, Palomera M, Gomez C, Sar-Shalom D, Saiz A, Montoya JG, et al.
dysphagia liked different purees significantly better than healthy Oropharyngeal dysphagia as a risk factor for malnutrition and lower respi-
old adults did [29]. This underlines the need to use a precise target ratory tract infection in independently living older persons: a population-
based prospective study. Age Ageing 2012;41(3):376e81.
group to test liking and food preferences of texture-modified foods, [6] Cook IJ. Oropharyngeal dysphagia. Gastroenterol Clin N Am 2009;38(3):
which is a strength of the current study. The study took place at 411e31.
three different nursing homes, with 30 residents suffering from [7] Sura L, Madhavan A, Carnaby G, Crary MA. Dysphagia in the elderly: man-
agement and nutritional considerations. Clin Interv Aging 2012;7:287e98.
dysphagia selected for inclusion. It was not possible to include [8] Volkert D, Chourdakis M, Faxen-Irving G, Fruhwald T, Landi F, Suominen MH,
more nursing homes in the study due to agreements and time et al. ESPEN guidelines on nutrition in dementia. Clin Nutr 2015;34(6):
constraints of the pilot project. It is recommended to repeat the 1052e73.
[9] Pedersen AN, Ovesen L. Recommendations of the Danish Institution Diet.
pilot study with a larger group of participants and for a longer time Minist Food Agric Fish Dan Vet Food Adm 2015;5(1).
period, in order to ensure the validity of the results. [10] Flynn EP, Smith CH, Walsh CD, Walshe M, Flynn EP. Modifying the consistency
of food and fluids for swallowing difficulties in dementia. In: Cochrane
Database of Systematic Reviews; 2014. p. 1e19.
4.6. Future perspectives [11] Wendin K, Ekman S, Bulow M, Ekberg O, Johansson D, Rothenberg E, et al.
Objective and quantitative definitions of modified food textures based on
sensory and rheological methodology. Food Nutr Res 2010;54:1e12.
A tailored in-between-meal menu would serve as a way of
[12] Westergren A, Lindholm C, Mattsson A, Ulander K. Minimal eating observation
intervening towards this group, but since food preferences are form: reliability and validity. J Nutr Health Aging 2009;13(1):6e12.
highly influenced by a person's culture and lifelong habits, the re- [13] Beck AM, Beermann T, Kjaer S, Rasmussen HH. Ability of different screening
tools to predict positive effect on nutritional intervention among the elderly
sults can only be extrapolated within a Danish context [35]. How-
in primary health care. Nutrition 2013;29(7e8):993e9.
ever, the perspective of application is wide, since these in-between- [14] Marcason W. What is the international dysphagia diet standardisation
meals may possibly also be suited for people with dysphagia in the initiative? J Acad Nutr Diet 2017;117(4):652.
community in general, as well as in hospitals. [15] Sørensen LB, Møller P, Flint A, Martens M, Raben A. Effect of sensory
perception of foods on appetite and food intake: a review of studies on
humans. Int J Obes 2003;27:1152e66.
5. Conclusion [16] Chen AW, Resurreccion AVA, Paguio LP. Age appropriate hedonic scales to
measure food preferences of young children. J Sens Stud 1996;11(2):141e63.
[17] Okkels SL, Bredie WLP, Klausen TW, Beck AM. An investigation into between-
In conclusion, it is promising that fatty and energy-rich in-be- meal food desires among hospitalised haematological cancer patients. Clin
tween-meals of sweet and sour taste and cold or frozen tempera- Nutr 2016;35(2):440e5.
[18] Sullivan GM, Artino Jr AR. Analyzing and interpreting data from likert-type
ture were mostly preferred. A way to design meals for scales. J Grad Med Educ 2013;5(4):541e2.
undernourished people with dysphagia could be to offer them [19] de Winter JCF, Dodou D. Five-point likert items: t test versus Mann-Whitney
these nutritious in-between-meals in order to offer what is both Wilcoxon 2010;15(11):1e12.
[20] Olin AO,€ Osterberg
€ P, Hådell K, Armyr I, Jerstro €m S, Ljungqvist O. Energy-
liked and needed.
enriched hospital food to improve energy intake in elderly patients
1996;20(2):93e7.
[21] Sorensen J, Holm L, Frøst MB, Kondrup J. Food for patients at nutritional risk: a
Author contributions
model of food sensory quality to promote intake. Clin Nutr 2012;31(5):
637e46.
All authors declare that there are no conflicts of interest. SLO [22] Price RJG, Mcmurdo MET, Anderson AS. A personalized snack-based inter-
and TWK mainly provided the statistical analysis. SLO and MS vention for hip fracture patients: development, feasibility and acceptability
2006;19(2):139e45.
collected the data, and SLO, AMB, SB and AO was responsible for the [23] Mingioni M, Mehinagic E, Siucin  ska K, Konopacka D, Artigas G, Symoneaux R,
study design and for drafting the manuscript. All authors have read et al. Sweet and sour discrimination abilities of elderly people compared to
those of young adults in apple pure e. Food Qual Prefer 2017;59:59e67.
and approved the final manuscript submitted for publication.
[24] De Jong N, de Graaf C, Van Staveren WA. Effect of sucrose in breakfast items
on pleasantness and food intake in the elderly 1996;60(6):1453e62.
Acknowledgements [25] Song X, Giacalone D, Bølling JSM, Frøst MB, Bredie WLP. Changes in oro-
sensory perception related to aging and strategies for counteracting its in-
fluence on food preferences among older adults. Trends Food Sci Technol
The study is part of the ELDORADO project “Preventing mal- 2016;53:49e59.
nourishment and promoting well-being in the elderly at home [26] Beelen J, de Roos NM, de Groot LCPGM. Protein enrichment of familiar foods
as an innovative strategy to increase protein intake in institutionalized
through personalised cost-effective food and meal supply”. elderly. J Nutr Health Aging 2017;21(2):173e9.
ELDORADO is supported by grant (4105-00009B) from the Inno- [27] Kremer S, Bult JH, Mojet J, Kroeze JH. Compensation for age-associated che-
vation Fund Denmark. mosensory losses and its effect on the pleasantness of a custard dessert and a
tomato drink. Appetite 2007;48(1):96e103.
[28] Mojet J, Heidema J, Christ-Hazelhof E. Effect of concentration on taste-taste
Conflicts of interest interactions in foods for elderly and young subjects. Chem Senses
2004;29(8):671e81.
[29] Ettinger L, Keller HH, Duizer LM. A comparison of liking of pureed food be-
None declared. tween two groups of older adults. J Nutr Gerontol Geriatr 2014;33(3):
198e209.
[30] Hall G, Wendin K. Sensory design of foods for the elderly. Ann Nutr Metab
References 2008;52(Suppl. 1):25e8.
[31] Jaros D, Rohm H, Strobl M. Appearance properties e a significant contribution
[1] Andersen UT, Beck AM, Kjaersgaard A, Hansen T, Poulsen I. Systematic review to sensory food quality? Food Sci Technol 2000;33(4):320e6.
and evidence based recommendations on texture modified foods and thick- [32] Keller H, Chambers L, Niezgoda H, Duizer L. Issues associated with the use of
ened fluids for adults (18 years) with oropharyngeal dysphagia. e-SPEN J modified texture foods 2012;16(3):195e200.
2013;8(4):e127e34. [33] Mojet J, Heidema J, Christ-Hazelhof E. Taste perception with age: generic or
[2] Madhavan A. Prevalence and risk factors for dysphagia in the community specific losses in supra-threshold intensities of five taste qualities? Chem
dwelling elderly: a systematic review. J Nutr Health Aging 2016;20(8): Senses 2003;28:397e413.
806e15. [34] Methven L, Jime nez-Pranteda ML, Lawlor JB. Sensory and consumer science
 M, Campins L, García-Peris P, et al. Diag-
[3] Rofes L, Arreola V, Almirall J, Cabre methods used with older adults: a review of current methods and recom-
nosis and management of oropharyngeal dysphagia and its nutritional and mendations for the future. Food Qual Prefer 2016;48:333e44.
respiratory complications in the elderly. Gastroenterol Res Pract 2011;2011: [35] Birch L. Development of food preferences. Annu Rev Nutr 1999;19:41e62.
1e13.

Please cite this article in press as: Okkels SL, et al., Acceptance of texture-modified in-between-meals among old adults with dysphagia, Clinical
Nutrition ESPEN (2018), https://doi.org/10.1016/j.clnesp.2018.03.119