The use of texture-adapted liquids is a widespread intervention in the clinical management of dysphagia (Tolstrup Andersen et al., 2013). Thickened beverages are included as one of the options for the required minimum total fluid intake, along with other alternatives such as enteral or parenteral fluid substitution. By adapting viscosity to an appropriate flow rate, thickened beverages can reduce dysphagic symptoms such as penetration or aspiration and can promote a safe oral fluid intake (Steele et al., 2015). Although prethickened drinks are available, German rehabilitation facilities mostly use thickening agents to reach necessary stages of thickening. In clinical practice, a nectar-like viscosity seems to be the most common stage of thickening. The study of Leder et al. (2013) showed that a nectar-like texture was adequate for patients who were able to swallow puree sufficiently but aspirated thin liquids. Caregivers and therapists often experience that thickened beverages are perceived as unappetizing and evoking unpleasant associations even before first-time consumption (Horwarth et al., 2005) associated with worse quality of life (Swan et al., 2015). Poor fluid intake and decreasing patient compliance based on disgust are issues in dysphagia management. Inadequate fluid intake of thickened beverages was found in three studies (Crary et al., 2015 ; Vivanti et al., 2009 ; Whelan, 2001). To prevent dehydration, enteral or parenteral fluid substitution is required if minimum total fluid intake is not reached orally. Nevertheless, naturally thick nectars and smoothies are vastly popular. Although featuring a comparable nectar-like thickness, consumers seem to tolerate this degree of viscosity better (Costell et al., 2000).
Previous research found that the kind of beverage and thickening agent, composed of starch and/or powdered gum, influences the acceptability of thickened beverages (Horwarth et al., 2005 ; Matta et al., 2006). Besides the desired change in viscosity, commercial thickeners often add textural properties such as slickness, sliminess, or graininess, which appear to be disliked characteristics of texture (Matta et al., 2006 ; Szczesniak, 2002). Generally, food acceptance and preferences are affected by multiple cultural, situational, and psychological factors. In particular, familiarity and expectations about the food are central for the willingness to ingest (Martins & Pliner, 2005). Cardello (1994) established the construct of two levels of expectations: (a) a sensory expectation about sensory characteristics of a certain food and (b) a hedonic expectation based on the anticipated liking/disliking of the food and the joy induced by the intake. Confirmation or disconfirmation of sensory and/or hedonic expectations is formative for assessing a product to be good or poor.
Although an alternative use of naturally thick beverages has been suggested by Horwarth et al. (2005) and Castellanos et al. (2004), no research has yet been carried out investigating the acceptability differences between thickened and naturally thick drinks, which is an area this study aims to investigate. For this purpose, healthy adult volunteers rated a selection of texture-modified and prototypic beverages with nectar-like viscosity. In addition, the effects of age and gender were investigated.
Materials and Methods
In this exploratory, cross-sectional study, healthy volunteers rated their preferences for six beverages with nectar-like viscosity: three usually fluid beverages thickened to nectar-like viscosity and three beverages of naturally nectar-like viscosity. The procedure comprised two measurements: expected acceptance preconsumption and actual acceptance postconsumption.
A total of 128 healthy adult volunteers (61 male, 67 female) participated in the study. The average age was 55.4 years, ranging from 30 to 93 years. The population included 44 participants of a pilot study (Gerschke, 2015) carried out under the same conditions by the first author at Fresenius University of Applied Sciences, Hamburg, Germany.
Exclusion criteria were present or former dysphagia, neurological disorder, or intolerance to fruit juices or dairy products. All subjects were recruited from the public and senior residences in Hamburg. As this study investigated healthy consumers’ preferences for everyday beverages in combination with a commercial thickener, which is rated in Germany as foodstuff, no ethics board approval was needed. In accordance with the World Medical Association Declaration of Helsinki, all participants received verbal and written information about the procedure and signed an informed consent statement.
Samples and Measurement
The six samples constituted three usually fluid beverages thickened to nectar-like viscosity (mineral water, homogenized milk, apple juice) and three beverages of naturally nectar-like viscosity (peach nectar, pear nectar, banana nectar). In order to create a selection of clinically relevant and generally well-accepted samples, we focused on the most consumed nonalcoholic, noncarbonated cold beverages according to the Federal Statistical Office of Germany (2011). Regarding the high consumption of water, milk, and fruit juices in the German population, it can be assumed that these are generally well-accepted in their original condition. The naturally thick beverages are commonly available in German food markets. In relation to real-life situations in German rehabilitation facilities, all samples were served in transparent plastic cups and consumed at room temperature. For preparation of the thickened samples, we used Nutricia Nutilis Powder dosed as specified by the manufacturer. Participants’ ratings were measured using the Hybrid Hedonic Scale (HHS; Figure 1; Villanueva et al., 2005) translated into German by the first author. Hedonic rating scales are commonly used to assess consumers’ acceptance regarding overall liking/disliking of a product (Hein et al., 2008). The continuous, numeric scale provides verbal affective labels in the middle and at the endpoints of the scale. A scale value of five points is set as a neutral assessment (“Neither liked nor disliked”). Descending values from four to zero (“Disliked extremely”) issue negative evaluations, whereas increasing values from six to ten (“Liked extremely”) issue positive ratings.
Each participant was examined separately in a quiet, undisturbed room by one of the authors. To avoid bias, the purpose of the study was only revealed after completing the task. The samples were presented subsequent in mixed order together with verbal information about the beverage (e.g., “This is peach nectar” or “This is thickened milk”). First, the subjects were instructed to rate how much they expected to like/dislike each beverage overall. After the following consumption of the sample, the subjects rated the actual perceived overall liking/disliking. Ratings were given by naming a scale value on the HHS.
Mean ratings and standard deviations (SD) are presented for the different samples pre- and postconsumption. We used the paired sample t test to assess mean differences between expected and perceived acceptance (significance level: p < .05). Mean differences between each thickened sample compared to the naturally thick samples were also assessed with the paired sample t test. To account for the fact that there were multiple tests conducted (three series), Bonferroni correction was used (adjusted significance level: p < .017). The independent sample t test was used for mean differences between age groups and gender (significance level: p < .05). Considering that even small mean differences may be statistically significant but not necessarily relevant for clinical practice, an additional clinical significance level was set: A mean difference of at least two points on the HHS was determined as clinically significant. Statistical analysis was carried out with SPSS Version 21 (SPSS, Inc., Chicago, IL).
Expected Acceptance Preconsumption
Mean ratings for expected acceptance are given in Table 1. For the thickened samples, water and milk received poor ratings, whereas apple juice was stated as neutral. All naturally thick samples were rated good.
Statistical and clinical significance was found for mean differences between all samples of natural viscosity versus thickened water and thickened milk (p < .001, mean differences >2 HHS). Comparing mean ratings of thickened apple juice, as the best rated thickened sample, with the poorest rated naturally thick sample (banana juice), no statistical or clinical significance was found (p = .031, mean difference <2 HHS).
Perceived Acceptance Postconsumption
The results postconsumption (Table 1) show negative mean ratings for thickened water and thickened milk and a neutral rating for thickened apple juice. The group of naturally thick samples received good ratings.
Comparing mean differences of the thickened samples and the natural samples showed statistical significance (p < .001) and clinical significance displaying a difference greater than two points on the HHS.
Comparison of the Results Pre- and Postconsumption
We found statistically significant decreased mean ratings post consumption for the thickened samples water and apple juice and an increase for all naturally viscous samples; there were no differences in thickened milk (Table 1). In view of clinical significance, the differences turned out as not being significant.
Age and Gender Effects
To assess the effect of age, we performed an analysis by age groups. Therefore, the population was divided into a middle age group, aged 30–59 years (n = 69, 34 male, 35 female, average age was 41 years), and an elderly group, aged 60–93 years (n = 59, 27 men, 32 female, average age was 72 years). We found no statistical or clinical significant difference by age groups (Table 2).
The comparison of mean ratings between male and female participants of the total population showed statistical significance for only one sample: banana nectar pre-(p < .001) and postconsumption (p = .002). However, the differences did not reach clinical significance (Table 3).
The study regarded the acceptability of thickened and naturally thick beverages pre- and postconsumption. Participants already showed distinct preference differences before consumption. Thickened water and thickened milk were associated with negative expectations. The deviation from the prototypic viscosity standard seemed to lower expectations. Thickened apple juice was anticipated neutrally. All naturally thick samples elicited positive expectations. For these, viscosity appeared to be tolerable.
Ratings postconsumption show considerable differences between thickened and naturally thick samples. All thickened beverages received significant lower ratings in comparison to those with naturally thickness, reaching the set levels for statistical and clinical significance. Thickened beverages, excluding milk, disappointed expectations, whereas naturally thick samples exceeded expectations.
Reflections about the modification of viscosity as a disconfirmation of consumers’ expectations of sensory attributes can be made. According to the previously described construct of expectations (Cardello, 1994), pro-totypic thickness confirmed sensoric expectations in the naturally thick group as divergent thickness violated sensoric expectations in the thickened group. Although displaying a comparable viscosity, participants preferred naturally thick to thickened beverages. The results support the findings of Martins and Pliner (2005), stating that familiarity and expectations about sensory properties influence food-related preferences. However, viscosity modification did not lead automatically to poor acceptance: participants excused viscosity in apple juice rather than in water. Apple juice potentially profited from consumers’ experiences with viscosity variations in fruit juices, whereas water usually features no variations. It cannot be ruled out that the thickening agent added a taste on its own that possibly was ameliorated by those samples with flavor. Furthermore, the thickener caused a degree of turbidity to the clear fluids as an additional visual aberration, which we assume to occur equally in clinical everyday life when beverages are served in glasses. Supporting the findings of Horwarth et al. (2005) and Matta et al. (2006), acceptability of the thickened samples seems to be affected by the kind of beverage.
No age effect was observed between the subgroups. The only gender difference for banana nectar fell below the clinical significance level and can therefore be stated as being negligible for clinical practice.
Some limitations of the study must be mentioned. The fact that viscosity measurements were not carried out is a main limitation of this study. Despite following the dosage recommendations given by the manufacturer, slight variations of viscosity were observed for the thickened samples as well as the naturally thick samples. Changes to the thickening effect of commercial thickeners to different kinds of beverages have been described by Garin et al. (2014) and are consistent with the experience of various actors in dysphagia management. The choice of the thickened beverages is based on the most consumed nonalcoholic, noncarbonated cold beverages in Germany, which may differ from individual food preferences, especially in other cultures. Thus, the study regards only one segment of everyday life beverages, excluding many beloved and highly ritualized beverages like hot coffee or tea. Another major limitation is that it was conducted on healthy individuals. Therefore, the results cannot be inferred to different pathological states of dysphagia patients. Intraoral sensory perception can change due to neurological disorders, structural changes in the context of head and neck cancer, and irradiation therapy (Elfring et al., 2014 ; Sciubba & Goldenberg, 2006 ; De Wijk & Prinz, 2005). Furthermore, a nectar-like viscosity may not be suitable for each patient, and naturally thick beverages may differ in terms of their viscosity. Adjustments into individually required grades of thickness must be considered to ensure a safe oral fluid intake.
Yet, little is known about how long-term exposure affects the acceptability of thickened fluids. The observation that drinks with unnatural viscosity were valued lower than those with prototypic viscosity indicates an alternative offer of naturally thick beverages. These allow the intake of “normal” beverages with expectable texture according to dietetic requirements. Higher costs for such beverages need to be compared with the costs for the thickening agent itself and possible personnel and material costs for enteral or parenteral fluid substitution in case of insufficient oral fluid intake.
Ensuring a safe and adequate hydration of people with dysphagia is the main goal of the use of thickened beverages. Alongside the value of thickening agents in dysphagia management, patients’ willingness to consume an appropriate quantity of fluid is required. The results of this study among 128 healthy subjects indicate that textural deviations from the usual appearance of everyday beverages have a significant impact on their acceptability. Beverages of naturally nectar-like thickness showed significant better acceptance postconsumption than thickened beverages. No age effect and no clinical relevant gender effect were found in this study. The findings support the suggestion for a higher supply of naturally thick liquids, such as fruit nectars, smoothies, yogurt drinks, or buttermilk for people with dysphagia. Considering these beverages as an alternative that combines the experience of normality with dietetic requirements could contribute to prevent complications based on noncompliance or poor fluid intake. Thickened water was rated to be aversive and cannot be recommended for clinical practice.
Further research on the effect of naturally thick beverages to patients’ with dysphagia oral fluid intake, long-term acceptance, and additional acceptance-related factors are needed.
Key Practice Points
- Thickened liquids are often perceived as unnatural and unappetizing.
- Appealing alternatives are needed to increase the acceptability of dietetic requirements, especially in patients with dysphagia.
- Naturally thick beverages are more attractive than thickened ones.
- Thickened water is unappealing to consumers.
Conflict of interest
The authors claim no conflict of interest.
Cardello A. V. (1994). Consumer expectations and their role in food acceptance
. In H. J. H. MacFie and D. M. H. Thomson (Eds.), Measurement of food preferences (pp. 253–297). London, UK: Blackie Academic.
Castellanos V. H., Butler E., Gluch L., & Burke B. (2004). Use of thickened liquids in skilled nursing facilities. Journal of the American Dietetic Association
, 104(8), 1222–1226.
Costell E., Pastor M. V., Izquierdo L., & Durán L. (2000). Relationships between acceptability and sensory attributes of peach nectars using internal preference mapping. European Food Research and Technology
, 211(3), 199–204.
Crary M. A., Carnaby G. D., Shabbir Y., Miller L., & Silliman S. (2015). Clinical variables associated with hydration status in acute ischemic stroke patients with dysphagia
, 31(1), 1–6.
De Wijk R. A., & Prinz J. F. (2005). The role of friction in perceived oral texture. Food Quality and Preference
, 16(2), 121–129.
Elfring T., Boliek C. A., Winget M., Paulsen C., Seikaly H., & Rieger J. M. (2014). The relationship between lingual and hypoglossal nerve function and quality of life in head and neck cancer. Journal of Oral Rehabilitation
, 41(2), 133–140.
Garin N., De Pourcq J. T., Martín-Venegas R., Cardona D., Gich I., & Mangues M. A. (2014). Viscosity differences between thickened beverages suitable for elderly patients with dysphagia
, 29(4), 483–488.
Gerschke M. (2015). Konsistenzadaption bei Dysphagie. Eine Pilotstudie zur Akzeptanz nektarartig angedickter und naturdicker Getraenke bei aelteren Menschen. Forum Logopädie
, 29(5), 24–28.
Hein K. A., Jaeger S. R., Carr B. T., & Delahunty C. M. (2008). Comparison of five common acceptance
and preference methods. Food Quality and Preference
, 19(7), 651–661.
Horwarth M., Ball A., & Smith R. (2005). Taste preference and rating of commercial and natural thickeners. Rehabilitation Nursing
, 30(6), 239–246.
Leder S. B., Judson B. L., Sliwinski E., & Madson L. (2013). Promoting safe swallowing when puree is swallowed without aspiration but thin liquid is aspirated: Nectar is enough. Dysphagia
, 28(1), 58–62.
Martins Y., & Pliner P. (2005). Human food choices: An examination of the factors underlying acceptance
/rejection of novel and familiar animal and nonanimal foods. Appetite
, 45(3), 214–224.
Matta Z., Chambers E., Garcia J. M., & Helverson J. M. (2006). Sensory characteristics of beverages prepared with commercial thickeners used for dysphagia
diets. Journal of the American Dietetic Association
, 106(7), 1049–1054.
Sciubba J. J., & Goldenberg D. (2006). Oral complications of radiotherapy. The Lancet Oncology
, 7(2), 175–183.
Steele C. M., Alsanei W. A., Ayanikalath S., Barbon C. E. A., Chen J., Cichero J. A. Y., … Wang H. (2015). The influence of food texture and liquid consistency modification on swallowing physiology and function: A systematic review. Dysphagia
, 30(1), 2–26.
Swan K., Speyer R., Heijnen B. J., Wagg B., & Cordier R. (2015). Living with oropharyngeal dysphagia
: Effects of bolus modification on health-related quality of life—A systematic review. Quality of Life Research: An International Journal of Quality of Life Aspects of Treatment, Care and Rehabilitation
, 24(10), 2447–2456.
Szczesniak A. S. (2002). Texture is a sensory property. Food Quality and Preference
, 13(4), 215–225.
Tolstrup Andersen U., Beck A. M., Kjaersgaard A., Hansen T., & Poulsen I. (2013). Systematic review and evidence based recommendations on texture modified foods and thickened fluids
for adults (≥18 years) with oropharyngeal dysphagia
. e-SPEN Journal
, 8, e127–e134.
Villanueva N. D. M., Petenate A. J., & Da Silva M. A. A. P. (2005). Performance of the hybrid hedonic scale as compared to the traditional hedonic, self-adjusting and ranking scales. Food Quality and Preference
, 16(8), 691–703.
Vivanti A. P., Campbell K. L., Suter M. S., Hannan-Jones M. T., & Hulcombe J. A. (2009). Contribution of thickened drinks, food and enteral and parenteral fluids to fluid intake in hospitalised patients with dysphagia
. Journal of Human Nutrition and Dietetics
, 22(2), 148–155.
Whelan K. (2001). Inadequate fluid intakes in dysphagic acute stroke. Clinical Nutrition
, 20, 423–428.