- Email: [email protected]
Sensory impact of lowering sugar content in orange nectars to design healthier, low-sugar industrialized beverages
Appetite 96 (2016) 239e244
Contents lists available at ScienceDirect
Appetite journal homepage: www.elsevier.com/locate/appet
Sensory impact of lowering sugar content in orange nectars to design healthier, low-sugar industrialized beverages Lívia de Lacerda de Oliveira Pineli a, *, Lorena Andrade de Aguiar a, Anndressa Fiusa a, ~o Botelho a, Renata Puppin Zandonadi a, Lauro Melo b Raquel Braz de Assunça a b
College of Health Sciences, University of Brasilia, Campus Darcy Ribeiro, Asa Norte, Brasilia, 70910-900, Brazil Department of Biochemical Engineering, School of Chemistry, Federal University of Rio de Janeiro, Rio de Janeiro, RJ, 21941-909, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 May 2015 Received in revised form 23 September 2015 Accepted 24 September 2015 Available online 30 September 2015
The presence of added sugars (AS) in the diet is associated with increased risk of obesity and other chronic diseases. We assessed sensory impact of lowering AS in orange nectar, aiming at new WHO sugar guideline. Ideal sweetness by just-about-right (JAR) tests (60 consumers), difference and rejection thresholds (36 and 35 assessors), and acceptance and sensory profile by Check-all-that-apply (CATA) tests (100 consumers) were performed. JAR test comprised six concentrations of AS from 12% down to 4.5%. Thresholds tests comprised orange nectars at reference sugar concentration (10%) and at lower sugar levels. Acceptance and CATA tests compared reference, ideal sweetness and thresholds concentrations. There were two groups of consumers; one with ideal sweetness lower at 5.5% AS and another with ideal sweetness at standard 10.5% AS. The average ideal sweetness among all consumers was 7.3% AS. The difference threshold from the reference at 10.5% AS was at 8.5% AS and the rejection threshold was 7.2%. Overall acceptance of nectar with 8.5% and 7.2% AS was similar to reference and higher than acceptance of nectar with 5.5%. However, after cluster analysis, nectars with 5.5% AS did not differ from nectars with 8.5% or 7.2% AS, suggesting the possibility of a gradual reduction until 5.5% in the long term. Lowering AS to 7.2% or 5.5% caused significant changes in viscosity, sweet odor, bitterness and sweetness in comparison to the reference concentration. Lowering sugar from 10% to 8.5% did not affect acceptance or sensory attributes, and could be indicated for a first reduction. Results indicate that a gradual reduction to 7.2% and 5.5% would be feasible. Reductions can remove 3150e9450 tons of sugar per year from the Brazilian diet resulting in healthier beverages. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Sugar lowering Orange nectar Sensory analysis Acceptance Public health policies
1. Introduction The presence of added sugars (AS) in the diet is associated with an increasing risk of several chronic diseases, including obesity. About 36% of the American adult population is obese and the increasing consumption of beverages, particularly sugar sweetened ones are thought to play a role in the etiology of obesity (Hu, 2013; Jeong, Gilmore, Bleakley, & Jordan, 2014). About 50% of Americans consume sugar sweetened beverages on any given day. In addition to obesity, this beverage intake has been associated with increased risk for type 2 diabetes, cardiovascular disease, and decreased diet quality (Park, Onufrak, Sherry, & Blanck, 2014).
* Corresponding author. Department of Nutrition, College of Health Sciences, University of Brasilia, Campus Darcy Ribeiro, Asa Norte, Brasilia, 70910-900, Brazil. E-mail address: [email protected] (L.L.O. Pineli). http://dx.doi.org/10.1016/j.appet.2015.09.028 0195-6663/© 2015 Elsevier Ltd. All rights reserved.
The glycemic and metabolic effects of beverages containing rapidly absorbable carbohydrates are likely to increase body weight and fat gain, especially with beverages containing about 10% carbohydrate in the form of AS. Limiting sugar sweetened beverages consumption has been acknowledged as one of the primary preventive strategies to control obesity. Reduction of added sugar in beverages is likely to benefit weight control, to reduce fat accumulation and obesity risk (Zheng et al., 2015). Therefore, the nutritional recommendation of the World Health Organization (WHO, 2015) is to reduce AS intake to up to 5% of the total calorie diet. According to the last Brazilian Household Budget Survey, juices and soft drinks are among the most commonly consumed foods by the Brazilian population, with consumption of 145.0 g/day and 94.7 g/day, respectively (Brazilian Institute of Geography and Statistics, 2010). The juices and nectars sector grew 14.4% in 2010
240
L.L.O. Pineli et al. / Appetite 96 (2016) 239e244
(Kist et al., 2012). According to the Brazilian Association of Nonalcoholic Beverages (ABIR), orange nectar consumption in 2014 in Brazil was estimated to 210 million liters per year (personal communication). In 2007, Brazilian Department of Health and food sector signed the Cooperation Agreement to promote healthy lifestyles that include a nutritionally healthy and adequate diet (Brazilian Department of Health, 2007). According to this agreement, the Health Department is responsible for establishing a gradual strategy to improve the nutritional profile of industrialized foods, focusing on lowering the levels of sugar, sodium, saturated fats and trans fatty acids in processed foods. Such agreements follow the trends of public policies to reduce sugar consumption worldwide, such as those practiced in the UK (Markey, Lovegrove, & Methven, 2015). The discussion on the gradual reduction of sugar levels by 2020 started in 2013, but to the best of our knowledge, no sensory study was carried out to test reductions in priority foods, such as nectars. At the same time, the Brazilian Department of Agriculture (2013) determined an increase in the minimum percentage of orange juice in nectars from 30% to 50%, which is likely to impact the sensory characteristics of the products. Therefore, it is necessary to evaluate sensory quality of orange nectars with reduced added sugars (AS) after increasing the concentration of orange juice. For juices, Brazilian legislation allows the addition of up to 10% of AS in the product (Brazilian Department of Agriculture, 2000). This amount is currently added to nectars as well, similar to other beverages, which contain about 10% carbohydrate in the form of AS (Zheng et al., 2015). Few studies to date have evaluated the impact of lowering sugars in food without the replacement by sweeteners. Hoppert, Zahn, Puschmann, Ullmann, & Rohm (2012) and Chollet, Gille, Schmid, Walther, and Piccinali (2013) lowered sugar without adding sweeteners in yogurt and Biguzzi, Schlich, and Lange (2014) in biscuits. In these studies the use of hedonic and just about right scales were used to assess the sensory impact of sugar reduction. Hoppert et al. (2012) associated sugar lowering with an increase in fiber, probably resulting in products that cannot be compared with standard counterparts. In yogurts with visible fiber, sugar content affected acceptance. Chollet et al. (2013) found that in spite of the significant difference in acceptance of yogurts with 10% AS and 7% AS, both were acceptable (mean score higher than 5 in 9-point hedonic scale). The increase of strawberry flavor concentration impaired the acceptance of yogurts with 7% of sugar and increased the percentage of consumers considering it “not sweet enough” in Just About Right scale. Biguzi et al. (2014) evaluated biscuits with reduced content of sugar (lowering of 11e29%) associated or not with fat reduction. Polydextrose was increased in formulations with lower sugar contents. The authors found that it was more acceptable to reduce fat than sugar in biscuits. Considering the complexity of the studied food matrices (yogurt and biscuits) and the changes in other food components (fibers, fat, polydextrose), these works cannot support a reduction in fruit juices and nectars, where sugar is the main macronutrient. A recent study by Lima Filho, Minim, Silva, Della Lucia, and Minim (2015) determined the compromised acceptance threshold (CAT) and rejection threshold (RT) for the concentration of sucrose in grape nectars. The reduction of sucrose concentration from 9.00% to 6.87% was found to decrease acceptance of the product (CAT) whereas the decrease of sucrose concentration from 9.00% to 3.83% was the threshold for sensory rejection of grape nectar. However, the authors did not determine the difference threshold, an important value in a context where sucrose reductions are designed to be undetected. Thus, considering the great public health concern on the
influence of AS beverages consumption, the goal of this study was to evaluate the impact of AS lowering in orange nectars driven by discrimination, affective and descriptive sensory methods to design healthier industrialized beverages. This is a first study focused on using sensory analysis to reduce the sugar content in nectars, and hence to support scientifically the agreements between government and food sector to improve nutritional composition of processed food aiming new WHO sugar guideline. 2. Material and methods 2.1. Nectar formulation Nectars were formulated by dilution of a reconstituted concentrated orange juice brand Maguary previously analyzed by HPLC-RID for sugar profile, presenting a ratio of 2.1/2.2 g of glucose/ fructose in 100 mL. Dilution was performed with 500 mL of reconstituted concentrated orange juice and granulated sugar (99.8% purity) was added according to each amount of AS in the final product, for each experiment. Water was used to make up the volume to 1 L. The studies were previously approved by Ethic Committee (CAAE: 22986714.8.0000.0030/2014). 2.2. Determination of ideal sweetness sugar concentration Ideal concentration of AS was determined using a Just-aboutright (JAR) category scale for sweetness (1-much less sweet than ideal; 5-just-about right; 9-much sweeter than ideal). Sixty panelists, regular consumers of industrialized fruit nectars (at least once a week), among students, professors, employees and other frequenters of University of Brasilia, (Federal District, Brazil), 55% females, from 18 to 34 years old, were recruited by means of social network announcements and dissemination of posters. They evaluated samples (50 mL, 11 ± 1 C) of orange nectars with AS at six different concentrations (12.0%, 10.5%, 9.0%, 7.5%, 6.0% and 4.5%) at 10 ± 1 C. Monadic sequential presentation of the samples was carried out in booths, in randomized order. Water and cream cracker biscuits were served to cleanse the palate. Linear regression (confidence interval of 95%) was performed on the JAR scores versus AS concentration and the AS concentration for ideal sweetness was then calculated by substituting “five” into the regression equation for the JAR score. Cluster analysis (CA) of JAR data was carried out in order to see different behaviors regarding preference of sweetness in the product. CA was performed by using Euclidian distance as a dissimilarity data proximity type and Ward's agglomeration method. All analyses were run in XLSTAT 2014 (Addinsoft, France). 2.3. Determination of thresholds Studies of rejection and detection thresholds were performed in order to determine the acceptable and noticeable levels of sugar reduction, thus providing significant information for the design of strategies of gradual sugar lowering by industrialized orange nectars. Consumer rejection thresholds (CRT) were performed according to Prescott, Norris, Kunst, and Kim (2005), who determined rejection thresholds by assessing consumer preference using paired preference tests against a constant stimuli and defining the CRT as the point where preference was significantly lower than reference. Samples (50 mL, 11 ± 1 C) were presented in booths, in ascending order of AS reduction and randomized within each pair. Panelists were asked to indicate the sample they preferred in each pair. Two series of three paired tests were presented, with a 3 min interval between them to avoid sensory fatigue. Water and cream cracker biscuits were served to cleanse the palate.
L.L.O. Pineli et al. / Appetite 96 (2016) 239e244
Similarly, difference threshold (DT) was determined by means of a sequence of one-sided paired difference tests for sweetness with a constant stimuli method. In both cases, samples were orange nectar with 10% AS (reference) versus samples with gradual reductions of AS (9.3%, 8.6%, 7.9%, 7.2%, 6.5% and 5.8%). Procedures of sample presentation were the same of CRT test. Panelists were asked to indicate the sample with the sweeter taste in each pair. The collected data were used to calculate proportions of the total 35 (70% females) and 36 (67% females) panelists, for CRT and DT, respectively. All panelists were regular consumers of orange nectars (at least once a week), among students, professors, employees and other frequenters of University of Brasilia, (Federal District, Brazil), from 18 to 35 years old, recruited by means of social network announcements and dissemination of posters. The thresholds were calculated by interpolation using AS levels and their corresponding proportion of assessors who preferred reference sample for CRT or proportion of correct answers in difference tests for DT to calculate the AS level corresponding to critical value for statistical significance (p < 0.05), according to binomial distribution table for one-sided paired-preference test for difference threshold and two-sided preference test for consumer rejection threshold (Prescott et al., 2005; ISO Standard 5495, 1983; Roessler, Pangborn, Sidel, & Stone, 1978). Calculation was done in Excel with support of XLSTAT 2014 software (Addinsoft, France) for the obtainment of the critical values. 2.4. Acceptance and descriptive tests of orange nectars with JAR and threshold concentrations of AS For acceptance and descriptive tests, orange nectars were prepared at the concentrations identified in the tests of thresholds and JAR scales. One-hundred regular consumers of orange nectars (at least once a week) aged between 18 and 40 years, 60% female, were recruited by means of social network announcements and dissemination of posters. Samples were served in balanced presentation, according to Macfie, Bratchell, Greenhoff, and Vallis (1989). Monadic sequential presentation of the samples of orange nectar with reference concentration (10%) and nectars at AS concentrations identified as ideal sweetness for two consumer clusters (50 mL, 11 ± 1 C) were carried out in booths. Acceptance was assessed with 9-point hedonic scale anchored from “1 e disliked extremely” to “9 e liked extremely”. Afterwards, consumers were asked to complete CATA question comprising12 attributes, which had been previously elicited by Kelly's Repertory grid method (Kelly, 1955; Lawless & Heymann, 1998) applied to six groups of 6 consumers. For the CATA lexicon, pairs of orange nectar samples with different characteristics in relation to pulp concentration (30% or 50%), reduction of sugar (10% or 5.5%), were displayed in different combinations in order to elicit attributes regarding appearance, aroma, flavor and texture. In the CATA test, the instruction given to assessors was: “Please, check all that applies to the orange nectar you tasted”. The 12 selected attributes were organized in appearance, odor, flavor and texture categories and attribute order was randomized between samples and across rrega, Izquierdo, and Jaeger (2014). consumers, according Ares, Ta Acceptance data were analyzed by ANOVA followed by multiple pairwise comparisons with Fisher's LSD to compare AS concentrations before and after Cluster analysis (CA) of acceptance data. Student's t test was performed to compare acceptance between clusters for specific AS concentration. Non-parametric Cochran's test analysis was carried out for each CATA descriptor to evaluate if there were differences in consumer perception of the nectars. Multiple pairwise comparisons were carried out using the Marascuilo procedure. Statistics analyses were performed with XLSTAT 2014 (Addinsoft, France).
241
3. Results A combination of discrimination, affective and descriptive sensory tests were carried out aiming to evaluate how gradual lowering in AS could be established for orange nectars without decreasing sensory quality of the product. 3.1. Concentration of AS for ideal sweetness Just-about-right (JAR) test was performed in order to evaluate the concentration of AS to achieve ideal sweetness (Table 1). The concentration of AS to achieve ideal sweetness across all consumers was 7.3%. However, 35% of the consumers (cluster 1) preferred the sweetness of orange nectars with more AS (10.4%), whereas 65% of them (cluster 2) considered ideal to be a lower concentration of AS (5.5%). Our results indicate that there is a possible margin for reducing AS in orange nectars from reference concentration of 10% AS to the ideal concentration of the majority of consumers. These results are promising to support agreements of sugar reduction, if associated with other sensory indicators such as difference and rejection thresholds as well as acceptance and descriptive tests for a better comprehension of the sensory impacts in the product. 3.2. Difference and consumer rejection thresholds Table 2 shows difference and rejection thresholds of AS. According to the results of constant stimuli tests, consumers perceived the difference in sweetness when AS was reduced from 10% to 8.5%, corresponding to a reduction of 15% of AS. However, consumer rejection threshold was reached once AS was reduced to 7.2%, a reduction of 28% in relation to reference AS concentration. The results indicate that gradual lowering in AS by the orange nectar industry could be carried out according to these levels, starting at the perceived difference limit (8.5% AS) and then further reducing to a minimum acceptable difference limit (7.2% AS). 3.3. Acceptance Levels of sugar recorded for the difference threshold (8.5% AS), CRT (7.2% AS) and the lower ideal JAR sweetness level (5.5% AS) were evaluated in comparison to the reference concentration of 10% AS in acceptance and CATA tests. The average ideal JAR sweetness concentration (7.3%) as well as the ideal JAR concentration of cluster group 2 (10.5%) were not evaluated in these analyses because the values were very close to rejection threshold (7.2%) and higher than the reference (10%), respectively, therefore they would not contribute to other insights on the sensory quality of orange nectars with reduced AS. The overall acceptances of nectar with 8.5% and 7.2% of AS did not differ from the reference nectar, but they were liked significantly more than the nectar with 5.5% of AS (Table 3). Cluster analysis resulted in a first group of consumers (n ¼ 46) who gave higher liking results for all samples and showed similar acceptance to the reference nectar and orange nectar with 5.5% of
Table 1 Added sugar levels to achieve justdaboutdright (JAR) sweetness for all consumers and within consumers clusters determined by clustering the JAR data.
Overall (n ¼ 60 consumers) Cluster 1 (n ¼ 21 consumers) Cluster 2 (n ¼ 39 consumers)
Equation
R2
JAR (% added sugar)
y ¼ 0.44x þ 1.78 y ¼ 0.48x þ 0.15 y ¼ 0.43x þ 2.66
0.97 0.98 0.92
7.3 10.4 5.5
JAR AS levels were six concentrations calculated with decreases of 1.5% AS (12.0%, 10.5%, 9.0%, 7.5%, 6.0% and 4.5%).
242
L.L.O. Pineli et al. / Appetite 96 (2016) 239e244
Table 2 Difference (n ¼ 36 assessors) and rejection (n ¼ 35 assessors) thresholds. % Added sugar (AS)
Difference threshold
Rejection threshold
% Sugar reduction
% Correct answers
p
% Sugar reduction
Preference answers for the highest concentration
p
9.3 8.6 7.9 7.2 6.5
7 14 21 28 36
63.9 83.3 69.4 97.2 91.7
0.096 <0.0001 0.02 <0.0001 <0.0001
7 14 21 28 35
54.3 42.9 54.3 68.6 77.1
0.612 0.398 0.612 0.028 0.001
Critical value (a ¼ 0.05) Threshold
Difference
66.7 8.5%
68.6 Rejection
15% of reduction
7.2%
28% of reduction
Table 3 Acceptance of orange nectars with different concentrations of AS (n ¼ 100 consumers). Concentration of AS
10%
8.5%
7.2%
5.5%
Overall acceptance Cluster 1 acceptance (n ¼ 46) Cluster 2 acceptance (n ¼ 54)
5.86A ± 1.54 6.59Ca ± 1.34 5.34Ab ± 1.44
6.12A ± 1.84 7.13ABa ± 1.33 4.76ABb ± 1.68
6.05A ± 1.65 7.26Aa ± 1.06 5.10Ab ± 1.39
5.37B ± 1.70 6.63BCa ± 1.34 4.39Bb ± 1.22
In rows, means followed by the same capital letter do not differ according to Fisher test (p < 0.05). Different lowercase letters in a column indicate statistical difference between clusters according to Student's t-test (p < 0.05).
AS. At the same time their preferred nectar was that with 7.2% of AS which did not differ from nectar with 8.5% of AS acceptance. This cluster indicated that optimum concentration of AS is in the intermediary points tested, showing a Type II functions. As stated by Moskowitz (1982), exhibiting Type II functions show maximum pleasantness value (optimum points) at an intermediate sensory level and include acceptability of sugar and salt. The second group of consumers (n ¼ 54) liked all nectars less than overall and showed higher acceptance for nectars with the highest sugar concentrations tested (10e7.2% of AS). However, acceptance of 8.5% of AS was intermediary and did not differ from the acceptance of nectar with 5.5% of AS. From the hedonic analysis it is possible to observe that the concentration of 7.2% of AS is in the top of positive evaluations in all situations. Sensory characteristics of all nectars were then assessed in order to further obtain associations between acceptance and sensory characteristics of the nectars. 3.4. Descriptive CATA analysis Significant differences were found in the frequencies of 4 out of the 12 terms of the CATA question used to describe samples indicating that the reduction of AS in orange nectars causes changes in
Table 4 Frequency of checked attributes for orange nectars with different concentrations of added sugar (n ¼ 100 consumers).
Viscosity Homogeneity Yellowness Citric odor Sweet odor Orange odor Artificial odor Bitterness Sweetness Sourness Orange flavor Orange juice flavor
10%
8.50%
7.20%
5.50%
p Value
54a 91a 97a 67a 62a 57a 65a 29c 77a 54a 67a 54a
43ab 93a 99a 64a 58a 63a 71a 33bc 68ab 58a 71a 53a
35b 92a 97a 70a 49ab 58a 75a 49ab 52bc 63a 70a 49a
40b 94a 97a 74a 40b 59a 69a 53a 43c 67a 63a 48a
0.002 0.753 0.392 0.361 0.001 0.763 0.261 <0.0001 <0.0001 0.124 0.409 0.650
In rows, frequency counts followed by the same letter do not differ by Cochran's test (p < 0.05). Multiple pairwise comparisons were carried out using the Marascuilo procedure.
viscosity, sweet odor, bitterness and sweetness (Table 4). Lowering AS did not affect the overall association of the following terms; homogeneity, yellowness, citric, orange and artificial odors, sourness, orange flavor and orange juice flavors. Viscosity was identified less in nectars with 7.2% and 5.5% of AS in comparison with the reference nectar. Sweet odor did not differ for nectars with 10%, 8.5% and 7.2% of AS; however, this attribute was lower in nectars with 5.5% of AS in comparison with nectars with 10 and 8.5% of AS. The evaluations of bitterness and sweetness were similar, but inverted. Studies of binary and more complex mixtures have demonstrated that bitterness can be suppressed by sweet tasting stimuli (Green, Lim, Osterhoff, Blacher, & Nachtigal, 2010; Lawless, 1979). According to Lawless (1979), mutual suppression of bitter and sweet tastes is related to neural inhibition, such as antidromic inhibition or occlusion, rather than chemical interactions in solution or competition of molecules for common receptor sites. Green et al. (2010) have shown that sucrose sweetness strongly suppressed the other tastes and at the same time, it was the least suppressed taste in ternary or quaternary mixtures. Therefore, the sugar lowering procedures probably reduces the bitterness-suppressing effect thus increasing bitterness in orange nectars. Reference nectar and nectar with 8.5% of AS were associated with sweetness with no difference between them; however nectar with 8.5% AS did not differ from nectar with 7.2% of AS. Nectar with 7.2% of AS was intermediary whereas nectar with 5.5% was associated least with sweet and most with bitter. As the basic tastes are very important attributes and easily recognized in all foods and beverages, these results should be taken into account in a gradual lowering AS approach with orange nectars.
4. Discussion Although consumers usually do not know the difference between juice and nectar, fruit juice concentration differentiates them. Nectars are diluted in water thus containing less fruit content, presents added sugar, and optionally colors and preservatives (Brazilian Department of Agriculture, 2009) that are generally cheaper than fruit. Therefore nectars are more accessible to intermediate or low income populations, and hence the majority of the Brazilian population (Longo-Silva et al., 2015). Thus, the standards
L.L.O. Pineli et al. / Appetite 96 (2016) 239e244
of the Brazilian Department of Agriculture by raising from 30% to 50% the minimum content of orange juice in fruit nectars and the possibility of sugar reduction as demonstrated in this study may contribute to consumers' health by increasing the amounts of bioactive compounds from orange (Milenkovic, Deval, Dubray, Mazur, & Morand, 2011) in orange nectars and by reducing the ingestion of sugars in the beverage which is inversely associated with body fat (Zheng et al., 2015). The use of approximately 10% of AS in several highly consumed beverages, such as juices and nectars, contributes to the high intake of AS, associated to obesity and chronic diseases (Hu, 2013; Jeong et al., 2014). Indeed, if a person consumes 145 g/day of juice (Brazilian Institute of Geography and Statistics, 2010), as a typical nectar of 10% AS, this equates to 14.5 g AS per day, corresponding to 58% of total sugar ingestion as recommended by WHO (2015). Although this level of AS is frequently practiced in beverages, ideal sweetness for several drinks have been recorded in literature and sometimes JAR concentrations of AS are different from 10%. JAR concentrations were 9.4% AS in passion fruit nectars (Rocha & Bolini, 2015), 9.5% of AS for instant coffee (Moraes & Bolini, 2010), 8.3% AS for hot tea (Cardoso, Battochio, & Cardello, 2004), 10% AS for peach nectar (Cardoso & Bolini, 2007) 8.0 and 7.5% for mango nectars (Umbelino, 2005) and 8.2% or 13.1% for orange juices, by 9-point JAR or 3-point JAR tests (Osornio & Hough, 2010). On average, ideal sweetness found in this work for orange nectar is lower than those found for peach, passion fruit and pineapple nectars, but close to mango nectar. However, when cluster analysis is applied, it is clear that two groups of Brazilian consumers exist, one group (35% of panelists) was probably adapted to the usual concentration of AS, as they preferred nectars with 10.4% of AS, whereas another group (65% of panelists) found 5.5% of AS the optimum concentration. Preference of the second group could be related to health attitudes towards the consumption of sugar in beverages or just sensory preference, but in both hypotheses indicating that acceptance can be achieved for lower concentrations of AS for several reasons, including nutritional education, parental rules or exposure to a new level of the sensory stimulus for a period of time (Bouhlal, Bernard, Issanchou, & Nicklaus, 2011; Liem, Mars, & De Graaf, 2004; Sartor et al., 2011). To our knowledge, only few studies have been carried out to assess difference thresholds for sweetness (Chang & Chiou, 2006; Chiou, Yeh, & Chang, 2009; Hoppert et al., 2012). None of them were carried out with fruit juices and nectars. The difference and rejection thresholds found in this study encourage AS lowering to levels close to JAR concentrations of AS for some nectars, thus being feasible and easily adaptable. According to the results presented, a proposed strategy to gradually lower AS content in orange nectars could enable a reduction to 8.5% (difference threshold). Followed by a further reduction to 7.2% (rejection threshold). The concentration of 5.5% was indicated as ideal by a group of consumers and, although different from the reference in some sensory aspects, it did not significantly differ from the nectar with 7.2% or 8.5% of AS in any of the performed sensory tests. Therefore, one can believe it is feasible the reduction to 5.5% after a period of exposition and adaptation of consumers to nectars with 7.2% of AS. Association to health claims could also be interesting to support such reduction (Chiou et al., 2009). In many countries guidelines and labeling has been developed to help consumers on healthier choices such as the Daily Intake Guide launched by Australian Food and Grocery Council (AFGC, 2006) and the traffic lights used in UK (Malan et al., 2009). Considering lowering strategies and annual orange nectar consumption in Brazil of 210 million of liters, the Brazilian population would reduce consumption of AS per year by 3150 tons, 5880 tons and 9450 tons of sugar if AS were adjusted to
243
8.5%, 7.2% or 5.5% of nectar, respectively. In a 2000 kcal diet, WHO guidelines suggest a maximum of 50 g of sugar a day, and an ideal maximum daily intake of 25 g. Orange nectars' intake of 145 g/day with 5.5% of AS would provide 7.97 g or 31.9% of AS intake, according to WHO guidelines, making it feasible to keep the appropriate consumption limits. A limitation of this study is the number of panelists involved in sensory tests. Although considered suitable for sensory tests (Stone & Sidel, 2004), the use of a larger number of assessors, mainly in the acceptance tests, could provide more information about the behavior of different consumer groups in future studies. CATA method is not the most recommended method to assess differences in sensory characteristics of very similar samples (Ares et al., 2014). In spite of that, it was possible to find differences in four attributes that are usually associated to sugar contents in food. More specifically sensory differences could be determined by the use of trained sensory panel profiling in further studies. Further studies should also be carried out to assess lowering AS in nectar from the standpoint of children. Frequently considered big users of these products, it is known that children's sensory perception can differ from adults' perception. James, Laing, and Oram (1997) evaluated age and gender influences in the detection of taste stimuli and observed that male children showed significantly higher thresholds than those of female adults. However, in another study using different rating scales, James, Laing, Jinks, and Oram NHutchinson (2003) concluded that the sense of taste has reached maturity for the perception of suprathreshold sucrose stimuli by mid-childhood (8e9 years old), once similar sweetness response functions for adults and children were obtained with two of the three procedures. On the other hand, studies with elderly indicate an increase in sweetness threshold in this age group (Mojet, Heidemma, & Christ-Hazelhof, 2003), which demands specific studies. Additionally, a high effect of age and gender on food choices and acceptance (Shepherd & Raats, 2006), is expected. In spite of that, the AS lowering in nectars as a public health policy in Brazil may contribute for sensory adaptation of the population from early ages, making it easier to achieve WHO recommendations in the future on sugar consumption. 5. Conclusion The sensory results of this study suggest that it is possible to reduce the AS in orange nectars by between 15 and 45% with very little impact on acceptance and sensory characteristics of the product. According to our results on JAR scale, thresholds, acceptance and CATA tests, these differences may not be perceived or may be acceptable. The reduction in sugar intake by the Brazilian population could reach almost 10 million tons per year after conclusion of such a lowering process within juices and nectars, if considered at current consumption levels. This work may contribute to public health policies of reduction of sugar in diets, which are mandatory in the face of obesity and chronic diseases profile associated with the consumption of AS. References rrega, A., Izquierdo, L., & Jaeger, S. R. (2014). Investigation of the number Ares, G., Ta of consumers necessary to obtain stable sample and descriptor configurations from check-all-that-apply (CATA) questions. Food Quality and Preference, 31, 135e141. Australian Food & Grocery Council. (2006). Health, nutrition and scientific affairs. http://www.afgc.org.au/our-expertise/health-nutrition-and-scientific-affairs/ Accessed 30.04.15. Biguzzi, C., Schlich, P., & Lange, C. (2014). The impact of sugar and fat reduction on perception and liking of biscuits. Food Quality and Preference, 35, 41e47. Bouhlal, S., Bernard, C., Issanchou, S., & Nicklaus, S. (2011). Sugar content impacts food intake in toddlers, but could be reduced. Appetite, 57S.
244
L.L.O. Pineli et al. / Appetite 96 (2016) 239e244
Brazilian Department of Agriculture. (2000). Decree nº 3.510, June 16, 2000. Change of regulation devices approved by decree no. 2314 of September 4, 1997, which provides for the standardization, classification, registration, inspection, production and supervision of beverages. http://extranet.agricultura.gov.br/sislegis-consulta/ consultarLegislacao.do?operacao¼visualizar&id¼1012 Accessed 05.05.15. Brazilian Department of Agriculture. (2009). Decree Nº 6.871, June 4th , 2009. Regulates law no. 8918 of July 14, 1994, which provides for the standardization, classification, registration, inspection, production and supervision of beverages. http:// www.planalto.gov.br/ccivil_03/_Ato2007-2010/2009/Decreto/D6871.htm Accessed 26.07.15. Brazilian Department of Agriculture. (2013). Normative ruling no. 42, of September 11, 2013. http://crn3.org.br/wp-content/uploads/2015/01/INSTRUCAO-NORMATIVAN42-11.07.13.pdf Accessed 05.04.15. Brazilian Department of Health. (2007). Technical cooperation agreement by Federal Government, between the Ministry of Health and the ABIA, in order to join forces and work together to implement actions to promote healthy lifestyles, which includes a healthy, balanced and nutritionally adequate diet. http://189.28.128.100/ nutricao/docs/geral/acordodecooperacaoabia_ms.pdf. Published November 2007. Accessed 05.04.15. Brazilian Institute of Geography and Statistics. (2010). Pesquisa de Orçamentos lise da disponibilidade domiciliar de alimentos e do Familiares 2008e2009: ana estado nutricional no Brasil. Rio de Janeiro: IBGE. Cardoso, J. M. P., Battochio, J. R., & Cardello, H. M. A. B. (2004). Equi-sweetness and sweetening power of different sweetening agents in differents temperatures of consumption of tea drink in soluble power. Brazilian Food Science and Technology, 24, 448e452. Cardoso, J. M. P., & Bolini, H. M. A. (2007). Different sweeteners in peach nectar: Ideal and equivalent sweetness. Food Research International, 40, 1249e1253. Chang, M. H., & Chiou, W. B. (2006). Differential threshold and psychophysical power function of sweetness sensation: applied psychophysics and prospect theory on formulating baking products. Journal of Sensory Studies, 21, 534e551. Chiou, W. B., Yeh, L. T., & Chang, M. H. (2009). Effects of health-related claims on the differential threshold of consumers' sweetness sensation. Journal of Sensory Studies, 24, 621e633. Chollet, M., Gille, G., Schmid, A., Walther, B., & Piccinali, P. (2013). Acceptance of sugar reduction in flavored yogurt. Journal of Dairy Science, 96, 5501e5511. Green, B. G., Lim, J., Osterhoff, F., Blacher, K., & Nachtigal, D. (2010). Taste mixture interactions: suppression, additivity, and the predominance of sweetness. Physiology Behavior, 101, 731e737. Hoppert, K., Zahn, S., Puschmann, A., Ullmann, I., & Rohm, H. (2012). Quantification of sensory difference thresholds for fat and sweetness in dairy-based emulsions. Food Quality and Preference, 26, 52e57. Hu, F. B. (2013). Resolved: there is sufficient scientific evidence that decreasing sugar-sweetened beverage consumption will reduce the prevalence of obesity and obesity-related diseases. Obesity Review, 14, 606e619. ISO. (1983). Standard 5495: Sensory analysiseeMethodologyeePaired comparison test. James, C. E., Laing, D. G., Jinks, A. L., & Oram N,Hutchinson, I. (2003). Taste response functions of adults and children using differentiating scales. Food Quality and Preference, 15, 77e82. James, C. E., Laing, D. G., & Oram, N. (1997). A comparison of the ability of 8e9-yearold children and adults to detect taste stimuli. Physiology & Behavior, 62, 193e197. Jeong, M., Gilmore, J. S., Bleakley, A., & Jordan, A. (2014). Local news media framing of obesity in the context of a sugar-sweetened beverage reduction media campaign. Journal of Nutritional Education and Behavior, 46, 583e588. Kelly, G. A. (1955). The psychology of personal constructs. New York: Norton. Kist, B. B., Vencato, A. Z., Santos, C., Carvalho, C., Reetz, E. R., Poll, H., et al. (2012). rio brasileiro da fruticultura 2012. Santa Cruz do Sul: EditoraGazeta Santa Anua Cruz (p. 128). Lawless, H. T. (1979). Evidence for neural inhibition in bittersweet taste mixtures. Journal of Comparative and Physiological Psychology, 93, 538e547. Lawless, H. T., & Heymann, H. (1998). Sensory evaluation of food. Principles and practices. New York: Chapman & Hall. Liem, D. G., Mars, M., & De Graaf, C. (2004). Sweet preferences and sugar consumption of 4- and 5-year-old children: role of parents. Appetite, 43, 235e245.
Lima Filho, T., Minim, V. P. R., Silva, R. C. S. N., Della Lucia, S. M., & Minim, L. A. (2015). Methodology for determination of two new sensory thresholds: compromised acceptance threshold and rejection threshold. Food Research International, 76, 561e566. Longo-Silva, G., Toloni, M. H. A., Menezes, R. C. A., Asakuraa, L., Oliveira, M. A. A., & Taddei, J. A. A. C. (2015). Introduction of soft drinks and processed juice in the diet of infants attending public day care centers. RevistaPaulista de Pediatria, 33, 34e41. Macfie, H. J., Bratchell, N., Greenhoff, K., & Vallis, L. (1989). Designs to balance the effect of order of presentation and first-order carry-over effects in hall tests. Journal of Sensory Studies, 4, 129e148. Malam, S., Clegg, S., Kirwan, S., McGinigal, S., Raats, M., Shepherd, R., et al. (2009). .Comprehension and use of UK nutrition signpost labelling schemes. London, UK: Food Standards Agency (UK). Available http://webarchive.nationalarchives.gov. uk/20131104005023/http://www.food.gov.uk/multimedia/pdfs/pmpreport.pdf Accessed 30.04.15. Markey, O., Lovegrove, J. A., & Methven, L. (2015). Sensory profiles and consumer acceptability of a range of sugar-reduced products on the UK market. Food Research International, 72, 133e139. Milenkovic, D., Deval, C., Dubray, C., Mazur, A., & Morand, C. (2011). Hesperidin displays relevant role in the nutrigenomic effect of orange juice on blood leukocytes in human volunteers: a randomized controlled cross-over study. Plos One, 6, e26669. Mojet, J., Heidemma, J., & Christ-Hazelhof, E. (2003). Taste perception with age: generic or specific losses in supra-threshold intensities of five taste qualities? Chemical Senses, 28, 397e413. Moraes, P. C. B. T., & Bolini, H. M. A. (2010). Different sweeteners in beverages prepared with instant and roasted ground coffee: ideal and equivalent sweetness. Journal of Sensory Studies, 25, 215e225. Moskowitz, H. R. (1982). Sensory intensity versus hedonic functions: classical psychophysical approaches. Journal of Food Quality, 5, 109e137. Osornio, M. M. L., & Hough, G. (2010). Comparing 3-point versus 9-point just-aboutright scales for determining the optimum concentration of sweetness in a beverage. Journal of Sensory Studies, 25, 1e17. Park, S., Onufrak, S., Sherry, B., & Blanck, H. M. (2014). The relationship between health-related knowledge and sugar-sweetened beverage intake among US adults. Journal of the Academy of Nutrition and Dietetics, 114, 1059e1066. Prescott, J., Norris, L., Kunst, M., & Kim, S. (2005). Estimating a “consumer rejection threshold” for cork taint in white wine. Food Quality and Preference, 16, 345e349. Rocha, I. F. O., & Bolini, H. M. A. (2015). Different sweeteners in passion fruit juice: ideal and equivalent sweetness. LWT e Food Science and Technology, 62, 861e867. Roessler, E. B., Pangborn, R. M., Sidel, J. L., & Stone, H. (1978). Expanded statistical tables for estimating significance in paired preference, paired difference, duotrio and triangle tests. Journal of Food Science, 43, 940e941. Sartor, F., Donaldson, L. F., Markland, D. A., Loveday, H., Jackson, M. J., & Kubis, H. P. (2011). Taste perception and implicit attitude toward sweet related to body mass index and soft drink supplementation. Appetite, 57, 237e246. Shepherd, R., & Raats, M. (2006). The psychology of food choice (p. 399). Oxfordshire: CABI. Stone, H., & Sidel, J. L. (2004). Sensory evaluation practices (3rd ed., p. 408). New York: Academic Press. Umbelino, D. C. (2005). Sensory characterization by descriptive quantitative analysis juice time-intensity and mango pulp (Mangiferaindica L.) sweteners with different edulcorants. PhD thesis (Food Science and Nutrition) e FEA. Campinas, Brazil: UniversidadeEstadual de Campinas. World Health Organization. (2015). WHO guideline: Sugars intake for adults and children. http://apps.who.int/iris/bitstream/10665/149782/1/9789241549028_ eng.pdf?ua¼1. Published February 2015. Accessed 05.04.15. Zheng, M., Rangan, A., Olsen, N. J., Andersen, L. B., Wedderkopp, N., Kristensen, P. … Heitmann, B. L. (2015). Substituting sugar-sweetened beverages with water or milk is inversely associated with body fatness development from childhood to adolescence. Nutrition, 31, 38e44.
Contents lists available at ScienceDirect
Appetite journal homepage: www.elsevier.com/locate/appet
Sensory impact of lowering sugar content in orange nectars to design healthier, low-sugar industrialized beverages Lívia de Lacerda de Oliveira Pineli a, *, Lorena Andrade de Aguiar a, Anndressa Fiusa a, ~o Botelho a, Renata Puppin Zandonadi a, Lauro Melo b Raquel Braz de Assunça a b
College of Health Sciences, University of Brasilia, Campus Darcy Ribeiro, Asa Norte, Brasilia, 70910-900, Brazil Department of Biochemical Engineering, School of Chemistry, Federal University of Rio de Janeiro, Rio de Janeiro, RJ, 21941-909, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 May 2015 Received in revised form 23 September 2015 Accepted 24 September 2015 Available online 30 September 2015
The presence of added sugars (AS) in the diet is associated with increased risk of obesity and other chronic diseases. We assessed sensory impact of lowering AS in orange nectar, aiming at new WHO sugar guideline. Ideal sweetness by just-about-right (JAR) tests (60 consumers), difference and rejection thresholds (36 and 35 assessors), and acceptance and sensory profile by Check-all-that-apply (CATA) tests (100 consumers) were performed. JAR test comprised six concentrations of AS from 12% down to 4.5%. Thresholds tests comprised orange nectars at reference sugar concentration (10%) and at lower sugar levels. Acceptance and CATA tests compared reference, ideal sweetness and thresholds concentrations. There were two groups of consumers; one with ideal sweetness lower at 5.5% AS and another with ideal sweetness at standard 10.5% AS. The average ideal sweetness among all consumers was 7.3% AS. The difference threshold from the reference at 10.5% AS was at 8.5% AS and the rejection threshold was 7.2%. Overall acceptance of nectar with 8.5% and 7.2% AS was similar to reference and higher than acceptance of nectar with 5.5%. However, after cluster analysis, nectars with 5.5% AS did not differ from nectars with 8.5% or 7.2% AS, suggesting the possibility of a gradual reduction until 5.5% in the long term. Lowering AS to 7.2% or 5.5% caused significant changes in viscosity, sweet odor, bitterness and sweetness in comparison to the reference concentration. Lowering sugar from 10% to 8.5% did not affect acceptance or sensory attributes, and could be indicated for a first reduction. Results indicate that a gradual reduction to 7.2% and 5.5% would be feasible. Reductions can remove 3150e9450 tons of sugar per year from the Brazilian diet resulting in healthier beverages. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Sugar lowering Orange nectar Sensory analysis Acceptance Public health policies
1. Introduction The presence of added sugars (AS) in the diet is associated with an increasing risk of several chronic diseases, including obesity. About 36% of the American adult population is obese and the increasing consumption of beverages, particularly sugar sweetened ones are thought to play a role in the etiology of obesity (Hu, 2013; Jeong, Gilmore, Bleakley, & Jordan, 2014). About 50% of Americans consume sugar sweetened beverages on any given day. In addition to obesity, this beverage intake has been associated with increased risk for type 2 diabetes, cardiovascular disease, and decreased diet quality (Park, Onufrak, Sherry, & Blanck, 2014).
* Corresponding author. Department of Nutrition, College of Health Sciences, University of Brasilia, Campus Darcy Ribeiro, Asa Norte, Brasilia, 70910-900, Brazil. E-mail address: [email protected] (L.L.O. Pineli). http://dx.doi.org/10.1016/j.appet.2015.09.028 0195-6663/© 2015 Elsevier Ltd. All rights reserved.
The glycemic and metabolic effects of beverages containing rapidly absorbable carbohydrates are likely to increase body weight and fat gain, especially with beverages containing about 10% carbohydrate in the form of AS. Limiting sugar sweetened beverages consumption has been acknowledged as one of the primary preventive strategies to control obesity. Reduction of added sugar in beverages is likely to benefit weight control, to reduce fat accumulation and obesity risk (Zheng et al., 2015). Therefore, the nutritional recommendation of the World Health Organization (WHO, 2015) is to reduce AS intake to up to 5% of the total calorie diet. According to the last Brazilian Household Budget Survey, juices and soft drinks are among the most commonly consumed foods by the Brazilian population, with consumption of 145.0 g/day and 94.7 g/day, respectively (Brazilian Institute of Geography and Statistics, 2010). The juices and nectars sector grew 14.4% in 2010
240
L.L.O. Pineli et al. / Appetite 96 (2016) 239e244
(Kist et al., 2012). According to the Brazilian Association of Nonalcoholic Beverages (ABIR), orange nectar consumption in 2014 in Brazil was estimated to 210 million liters per year (personal communication). In 2007, Brazilian Department of Health and food sector signed the Cooperation Agreement to promote healthy lifestyles that include a nutritionally healthy and adequate diet (Brazilian Department of Health, 2007). According to this agreement, the Health Department is responsible for establishing a gradual strategy to improve the nutritional profile of industrialized foods, focusing on lowering the levels of sugar, sodium, saturated fats and trans fatty acids in processed foods. Such agreements follow the trends of public policies to reduce sugar consumption worldwide, such as those practiced in the UK (Markey, Lovegrove, & Methven, 2015). The discussion on the gradual reduction of sugar levels by 2020 started in 2013, but to the best of our knowledge, no sensory study was carried out to test reductions in priority foods, such as nectars. At the same time, the Brazilian Department of Agriculture (2013) determined an increase in the minimum percentage of orange juice in nectars from 30% to 50%, which is likely to impact the sensory characteristics of the products. Therefore, it is necessary to evaluate sensory quality of orange nectars with reduced added sugars (AS) after increasing the concentration of orange juice. For juices, Brazilian legislation allows the addition of up to 10% of AS in the product (Brazilian Department of Agriculture, 2000). This amount is currently added to nectars as well, similar to other beverages, which contain about 10% carbohydrate in the form of AS (Zheng et al., 2015). Few studies to date have evaluated the impact of lowering sugars in food without the replacement by sweeteners. Hoppert, Zahn, Puschmann, Ullmann, & Rohm (2012) and Chollet, Gille, Schmid, Walther, and Piccinali (2013) lowered sugar without adding sweeteners in yogurt and Biguzzi, Schlich, and Lange (2014) in biscuits. In these studies the use of hedonic and just about right scales were used to assess the sensory impact of sugar reduction. Hoppert et al. (2012) associated sugar lowering with an increase in fiber, probably resulting in products that cannot be compared with standard counterparts. In yogurts with visible fiber, sugar content affected acceptance. Chollet et al. (2013) found that in spite of the significant difference in acceptance of yogurts with 10% AS and 7% AS, both were acceptable (mean score higher than 5 in 9-point hedonic scale). The increase of strawberry flavor concentration impaired the acceptance of yogurts with 7% of sugar and increased the percentage of consumers considering it “not sweet enough” in Just About Right scale. Biguzi et al. (2014) evaluated biscuits with reduced content of sugar (lowering of 11e29%) associated or not with fat reduction. Polydextrose was increased in formulations with lower sugar contents. The authors found that it was more acceptable to reduce fat than sugar in biscuits. Considering the complexity of the studied food matrices (yogurt and biscuits) and the changes in other food components (fibers, fat, polydextrose), these works cannot support a reduction in fruit juices and nectars, where sugar is the main macronutrient. A recent study by Lima Filho, Minim, Silva, Della Lucia, and Minim (2015) determined the compromised acceptance threshold (CAT) and rejection threshold (RT) for the concentration of sucrose in grape nectars. The reduction of sucrose concentration from 9.00% to 6.87% was found to decrease acceptance of the product (CAT) whereas the decrease of sucrose concentration from 9.00% to 3.83% was the threshold for sensory rejection of grape nectar. However, the authors did not determine the difference threshold, an important value in a context where sucrose reductions are designed to be undetected. Thus, considering the great public health concern on the
influence of AS beverages consumption, the goal of this study was to evaluate the impact of AS lowering in orange nectars driven by discrimination, affective and descriptive sensory methods to design healthier industrialized beverages. This is a first study focused on using sensory analysis to reduce the sugar content in nectars, and hence to support scientifically the agreements between government and food sector to improve nutritional composition of processed food aiming new WHO sugar guideline. 2. Material and methods 2.1. Nectar formulation Nectars were formulated by dilution of a reconstituted concentrated orange juice brand Maguary previously analyzed by HPLC-RID for sugar profile, presenting a ratio of 2.1/2.2 g of glucose/ fructose in 100 mL. Dilution was performed with 500 mL of reconstituted concentrated orange juice and granulated sugar (99.8% purity) was added according to each amount of AS in the final product, for each experiment. Water was used to make up the volume to 1 L. The studies were previously approved by Ethic Committee (CAAE: 22986714.8.0000.0030/2014). 2.2. Determination of ideal sweetness sugar concentration Ideal concentration of AS was determined using a Just-aboutright (JAR) category scale for sweetness (1-much less sweet than ideal; 5-just-about right; 9-much sweeter than ideal). Sixty panelists, regular consumers of industrialized fruit nectars (at least once a week), among students, professors, employees and other frequenters of University of Brasilia, (Federal District, Brazil), 55% females, from 18 to 34 years old, were recruited by means of social network announcements and dissemination of posters. They evaluated samples (50 mL, 11 ± 1 C) of orange nectars with AS at six different concentrations (12.0%, 10.5%, 9.0%, 7.5%, 6.0% and 4.5%) at 10 ± 1 C. Monadic sequential presentation of the samples was carried out in booths, in randomized order. Water and cream cracker biscuits were served to cleanse the palate. Linear regression (confidence interval of 95%) was performed on the JAR scores versus AS concentration and the AS concentration for ideal sweetness was then calculated by substituting “five” into the regression equation for the JAR score. Cluster analysis (CA) of JAR data was carried out in order to see different behaviors regarding preference of sweetness in the product. CA was performed by using Euclidian distance as a dissimilarity data proximity type and Ward's agglomeration method. All analyses were run in XLSTAT 2014 (Addinsoft, France). 2.3. Determination of thresholds Studies of rejection and detection thresholds were performed in order to determine the acceptable and noticeable levels of sugar reduction, thus providing significant information for the design of strategies of gradual sugar lowering by industrialized orange nectars. Consumer rejection thresholds (CRT) were performed according to Prescott, Norris, Kunst, and Kim (2005), who determined rejection thresholds by assessing consumer preference using paired preference tests against a constant stimuli and defining the CRT as the point where preference was significantly lower than reference. Samples (50 mL, 11 ± 1 C) were presented in booths, in ascending order of AS reduction and randomized within each pair. Panelists were asked to indicate the sample they preferred in each pair. Two series of three paired tests were presented, with a 3 min interval between them to avoid sensory fatigue. Water and cream cracker biscuits were served to cleanse the palate.
L.L.O. Pineli et al. / Appetite 96 (2016) 239e244
Similarly, difference threshold (DT) was determined by means of a sequence of one-sided paired difference tests for sweetness with a constant stimuli method. In both cases, samples were orange nectar with 10% AS (reference) versus samples with gradual reductions of AS (9.3%, 8.6%, 7.9%, 7.2%, 6.5% and 5.8%). Procedures of sample presentation were the same of CRT test. Panelists were asked to indicate the sample with the sweeter taste in each pair. The collected data were used to calculate proportions of the total 35 (70% females) and 36 (67% females) panelists, for CRT and DT, respectively. All panelists were regular consumers of orange nectars (at least once a week), among students, professors, employees and other frequenters of University of Brasilia, (Federal District, Brazil), from 18 to 35 years old, recruited by means of social network announcements and dissemination of posters. The thresholds were calculated by interpolation using AS levels and their corresponding proportion of assessors who preferred reference sample for CRT or proportion of correct answers in difference tests for DT to calculate the AS level corresponding to critical value for statistical significance (p < 0.05), according to binomial distribution table for one-sided paired-preference test for difference threshold and two-sided preference test for consumer rejection threshold (Prescott et al., 2005; ISO Standard 5495, 1983; Roessler, Pangborn, Sidel, & Stone, 1978). Calculation was done in Excel with support of XLSTAT 2014 software (Addinsoft, France) for the obtainment of the critical values. 2.4. Acceptance and descriptive tests of orange nectars with JAR and threshold concentrations of AS For acceptance and descriptive tests, orange nectars were prepared at the concentrations identified in the tests of thresholds and JAR scales. One-hundred regular consumers of orange nectars (at least once a week) aged between 18 and 40 years, 60% female, were recruited by means of social network announcements and dissemination of posters. Samples were served in balanced presentation, according to Macfie, Bratchell, Greenhoff, and Vallis (1989). Monadic sequential presentation of the samples of orange nectar with reference concentration (10%) and nectars at AS concentrations identified as ideal sweetness for two consumer clusters (50 mL, 11 ± 1 C) were carried out in booths. Acceptance was assessed with 9-point hedonic scale anchored from “1 e disliked extremely” to “9 e liked extremely”. Afterwards, consumers were asked to complete CATA question comprising12 attributes, which had been previously elicited by Kelly's Repertory grid method (Kelly, 1955; Lawless & Heymann, 1998) applied to six groups of 6 consumers. For the CATA lexicon, pairs of orange nectar samples with different characteristics in relation to pulp concentration (30% or 50%), reduction of sugar (10% or 5.5%), were displayed in different combinations in order to elicit attributes regarding appearance, aroma, flavor and texture. In the CATA test, the instruction given to assessors was: “Please, check all that applies to the orange nectar you tasted”. The 12 selected attributes were organized in appearance, odor, flavor and texture categories and attribute order was randomized between samples and across rrega, Izquierdo, and Jaeger (2014). consumers, according Ares, Ta Acceptance data were analyzed by ANOVA followed by multiple pairwise comparisons with Fisher's LSD to compare AS concentrations before and after Cluster analysis (CA) of acceptance data. Student's t test was performed to compare acceptance between clusters for specific AS concentration. Non-parametric Cochran's test analysis was carried out for each CATA descriptor to evaluate if there were differences in consumer perception of the nectars. Multiple pairwise comparisons were carried out using the Marascuilo procedure. Statistics analyses were performed with XLSTAT 2014 (Addinsoft, France).
241
3. Results A combination of discrimination, affective and descriptive sensory tests were carried out aiming to evaluate how gradual lowering in AS could be established for orange nectars without decreasing sensory quality of the product. 3.1. Concentration of AS for ideal sweetness Just-about-right (JAR) test was performed in order to evaluate the concentration of AS to achieve ideal sweetness (Table 1). The concentration of AS to achieve ideal sweetness across all consumers was 7.3%. However, 35% of the consumers (cluster 1) preferred the sweetness of orange nectars with more AS (10.4%), whereas 65% of them (cluster 2) considered ideal to be a lower concentration of AS (5.5%). Our results indicate that there is a possible margin for reducing AS in orange nectars from reference concentration of 10% AS to the ideal concentration of the majority of consumers. These results are promising to support agreements of sugar reduction, if associated with other sensory indicators such as difference and rejection thresholds as well as acceptance and descriptive tests for a better comprehension of the sensory impacts in the product. 3.2. Difference and consumer rejection thresholds Table 2 shows difference and rejection thresholds of AS. According to the results of constant stimuli tests, consumers perceived the difference in sweetness when AS was reduced from 10% to 8.5%, corresponding to a reduction of 15% of AS. However, consumer rejection threshold was reached once AS was reduced to 7.2%, a reduction of 28% in relation to reference AS concentration. The results indicate that gradual lowering in AS by the orange nectar industry could be carried out according to these levels, starting at the perceived difference limit (8.5% AS) and then further reducing to a minimum acceptable difference limit (7.2% AS). 3.3. Acceptance Levels of sugar recorded for the difference threshold (8.5% AS), CRT (7.2% AS) and the lower ideal JAR sweetness level (5.5% AS) were evaluated in comparison to the reference concentration of 10% AS in acceptance and CATA tests. The average ideal JAR sweetness concentration (7.3%) as well as the ideal JAR concentration of cluster group 2 (10.5%) were not evaluated in these analyses because the values were very close to rejection threshold (7.2%) and higher than the reference (10%), respectively, therefore they would not contribute to other insights on the sensory quality of orange nectars with reduced AS. The overall acceptances of nectar with 8.5% and 7.2% of AS did not differ from the reference nectar, but they were liked significantly more than the nectar with 5.5% of AS (Table 3). Cluster analysis resulted in a first group of consumers (n ¼ 46) who gave higher liking results for all samples and showed similar acceptance to the reference nectar and orange nectar with 5.5% of
Table 1 Added sugar levels to achieve justdaboutdright (JAR) sweetness for all consumers and within consumers clusters determined by clustering the JAR data.
Overall (n ¼ 60 consumers) Cluster 1 (n ¼ 21 consumers) Cluster 2 (n ¼ 39 consumers)
Equation
R2
JAR (% added sugar)
y ¼ 0.44x þ 1.78 y ¼ 0.48x þ 0.15 y ¼ 0.43x þ 2.66
0.97 0.98 0.92
7.3 10.4 5.5
JAR AS levels were six concentrations calculated with decreases of 1.5% AS (12.0%, 10.5%, 9.0%, 7.5%, 6.0% and 4.5%).
242
L.L.O. Pineli et al. / Appetite 96 (2016) 239e244
Table 2 Difference (n ¼ 36 assessors) and rejection (n ¼ 35 assessors) thresholds. % Added sugar (AS)
Difference threshold
Rejection threshold
% Sugar reduction
% Correct answers
p
% Sugar reduction
Preference answers for the highest concentration
p
9.3 8.6 7.9 7.2 6.5
7 14 21 28 36
63.9 83.3 69.4 97.2 91.7
0.096 <0.0001 0.02 <0.0001 <0.0001
7 14 21 28 35
54.3 42.9 54.3 68.6 77.1
0.612 0.398 0.612 0.028 0.001
Critical value (a ¼ 0.05) Threshold
Difference
66.7 8.5%
68.6 Rejection
15% of reduction
7.2%
28% of reduction
Table 3 Acceptance of orange nectars with different concentrations of AS (n ¼ 100 consumers). Concentration of AS
10%
8.5%
7.2%
5.5%
Overall acceptance Cluster 1 acceptance (n ¼ 46) Cluster 2 acceptance (n ¼ 54)
5.86A ± 1.54 6.59Ca ± 1.34 5.34Ab ± 1.44
6.12A ± 1.84 7.13ABa ± 1.33 4.76ABb ± 1.68
6.05A ± 1.65 7.26Aa ± 1.06 5.10Ab ± 1.39
5.37B ± 1.70 6.63BCa ± 1.34 4.39Bb ± 1.22
In rows, means followed by the same capital letter do not differ according to Fisher test (p < 0.05). Different lowercase letters in a column indicate statistical difference between clusters according to Student's t-test (p < 0.05).
AS. At the same time their preferred nectar was that with 7.2% of AS which did not differ from nectar with 8.5% of AS acceptance. This cluster indicated that optimum concentration of AS is in the intermediary points tested, showing a Type II functions. As stated by Moskowitz (1982), exhibiting Type II functions show maximum pleasantness value (optimum points) at an intermediate sensory level and include acceptability of sugar and salt. The second group of consumers (n ¼ 54) liked all nectars less than overall and showed higher acceptance for nectars with the highest sugar concentrations tested (10e7.2% of AS). However, acceptance of 8.5% of AS was intermediary and did not differ from the acceptance of nectar with 5.5% of AS. From the hedonic analysis it is possible to observe that the concentration of 7.2% of AS is in the top of positive evaluations in all situations. Sensory characteristics of all nectars were then assessed in order to further obtain associations between acceptance and sensory characteristics of the nectars. 3.4. Descriptive CATA analysis Significant differences were found in the frequencies of 4 out of the 12 terms of the CATA question used to describe samples indicating that the reduction of AS in orange nectars causes changes in
Table 4 Frequency of checked attributes for orange nectars with different concentrations of added sugar (n ¼ 100 consumers).
Viscosity Homogeneity Yellowness Citric odor Sweet odor Orange odor Artificial odor Bitterness Sweetness Sourness Orange flavor Orange juice flavor
10%
8.50%
7.20%
5.50%
p Value
54a 91a 97a 67a 62a 57a 65a 29c 77a 54a 67a 54a
43ab 93a 99a 64a 58a 63a 71a 33bc 68ab 58a 71a 53a
35b 92a 97a 70a 49ab 58a 75a 49ab 52bc 63a 70a 49a
40b 94a 97a 74a 40b 59a 69a 53a 43c 67a 63a 48a
0.002 0.753 0.392 0.361 0.001 0.763 0.261 <0.0001 <0.0001 0.124 0.409 0.650
In rows, frequency counts followed by the same letter do not differ by Cochran's test (p < 0.05). Multiple pairwise comparisons were carried out using the Marascuilo procedure.
viscosity, sweet odor, bitterness and sweetness (Table 4). Lowering AS did not affect the overall association of the following terms; homogeneity, yellowness, citric, orange and artificial odors, sourness, orange flavor and orange juice flavors. Viscosity was identified less in nectars with 7.2% and 5.5% of AS in comparison with the reference nectar. Sweet odor did not differ for nectars with 10%, 8.5% and 7.2% of AS; however, this attribute was lower in nectars with 5.5% of AS in comparison with nectars with 10 and 8.5% of AS. The evaluations of bitterness and sweetness were similar, but inverted. Studies of binary and more complex mixtures have demonstrated that bitterness can be suppressed by sweet tasting stimuli (Green, Lim, Osterhoff, Blacher, & Nachtigal, 2010; Lawless, 1979). According to Lawless (1979), mutual suppression of bitter and sweet tastes is related to neural inhibition, such as antidromic inhibition or occlusion, rather than chemical interactions in solution or competition of molecules for common receptor sites. Green et al. (2010) have shown that sucrose sweetness strongly suppressed the other tastes and at the same time, it was the least suppressed taste in ternary or quaternary mixtures. Therefore, the sugar lowering procedures probably reduces the bitterness-suppressing effect thus increasing bitterness in orange nectars. Reference nectar and nectar with 8.5% of AS were associated with sweetness with no difference between them; however nectar with 8.5% AS did not differ from nectar with 7.2% of AS. Nectar with 7.2% of AS was intermediary whereas nectar with 5.5% was associated least with sweet and most with bitter. As the basic tastes are very important attributes and easily recognized in all foods and beverages, these results should be taken into account in a gradual lowering AS approach with orange nectars.
4. Discussion Although consumers usually do not know the difference between juice and nectar, fruit juice concentration differentiates them. Nectars are diluted in water thus containing less fruit content, presents added sugar, and optionally colors and preservatives (Brazilian Department of Agriculture, 2009) that are generally cheaper than fruit. Therefore nectars are more accessible to intermediate or low income populations, and hence the majority of the Brazilian population (Longo-Silva et al., 2015). Thus, the standards
L.L.O. Pineli et al. / Appetite 96 (2016) 239e244
of the Brazilian Department of Agriculture by raising from 30% to 50% the minimum content of orange juice in fruit nectars and the possibility of sugar reduction as demonstrated in this study may contribute to consumers' health by increasing the amounts of bioactive compounds from orange (Milenkovic, Deval, Dubray, Mazur, & Morand, 2011) in orange nectars and by reducing the ingestion of sugars in the beverage which is inversely associated with body fat (Zheng et al., 2015). The use of approximately 10% of AS in several highly consumed beverages, such as juices and nectars, contributes to the high intake of AS, associated to obesity and chronic diseases (Hu, 2013; Jeong et al., 2014). Indeed, if a person consumes 145 g/day of juice (Brazilian Institute of Geography and Statistics, 2010), as a typical nectar of 10% AS, this equates to 14.5 g AS per day, corresponding to 58% of total sugar ingestion as recommended by WHO (2015). Although this level of AS is frequently practiced in beverages, ideal sweetness for several drinks have been recorded in literature and sometimes JAR concentrations of AS are different from 10%. JAR concentrations were 9.4% AS in passion fruit nectars (Rocha & Bolini, 2015), 9.5% of AS for instant coffee (Moraes & Bolini, 2010), 8.3% AS for hot tea (Cardoso, Battochio, & Cardello, 2004), 10% AS for peach nectar (Cardoso & Bolini, 2007) 8.0 and 7.5% for mango nectars (Umbelino, 2005) and 8.2% or 13.1% for orange juices, by 9-point JAR or 3-point JAR tests (Osornio & Hough, 2010). On average, ideal sweetness found in this work for orange nectar is lower than those found for peach, passion fruit and pineapple nectars, but close to mango nectar. However, when cluster analysis is applied, it is clear that two groups of Brazilian consumers exist, one group (35% of panelists) was probably adapted to the usual concentration of AS, as they preferred nectars with 10.4% of AS, whereas another group (65% of panelists) found 5.5% of AS the optimum concentration. Preference of the second group could be related to health attitudes towards the consumption of sugar in beverages or just sensory preference, but in both hypotheses indicating that acceptance can be achieved for lower concentrations of AS for several reasons, including nutritional education, parental rules or exposure to a new level of the sensory stimulus for a period of time (Bouhlal, Bernard, Issanchou, & Nicklaus, 2011; Liem, Mars, & De Graaf, 2004; Sartor et al., 2011). To our knowledge, only few studies have been carried out to assess difference thresholds for sweetness (Chang & Chiou, 2006; Chiou, Yeh, & Chang, 2009; Hoppert et al., 2012). None of them were carried out with fruit juices and nectars. The difference and rejection thresholds found in this study encourage AS lowering to levels close to JAR concentrations of AS for some nectars, thus being feasible and easily adaptable. According to the results presented, a proposed strategy to gradually lower AS content in orange nectars could enable a reduction to 8.5% (difference threshold). Followed by a further reduction to 7.2% (rejection threshold). The concentration of 5.5% was indicated as ideal by a group of consumers and, although different from the reference in some sensory aspects, it did not significantly differ from the nectar with 7.2% or 8.5% of AS in any of the performed sensory tests. Therefore, one can believe it is feasible the reduction to 5.5% after a period of exposition and adaptation of consumers to nectars with 7.2% of AS. Association to health claims could also be interesting to support such reduction (Chiou et al., 2009). In many countries guidelines and labeling has been developed to help consumers on healthier choices such as the Daily Intake Guide launched by Australian Food and Grocery Council (AFGC, 2006) and the traffic lights used in UK (Malan et al., 2009). Considering lowering strategies and annual orange nectar consumption in Brazil of 210 million of liters, the Brazilian population would reduce consumption of AS per year by 3150 tons, 5880 tons and 9450 tons of sugar if AS were adjusted to
243
8.5%, 7.2% or 5.5% of nectar, respectively. In a 2000 kcal diet, WHO guidelines suggest a maximum of 50 g of sugar a day, and an ideal maximum daily intake of 25 g. Orange nectars' intake of 145 g/day with 5.5% of AS would provide 7.97 g or 31.9% of AS intake, according to WHO guidelines, making it feasible to keep the appropriate consumption limits. A limitation of this study is the number of panelists involved in sensory tests. Although considered suitable for sensory tests (Stone & Sidel, 2004), the use of a larger number of assessors, mainly in the acceptance tests, could provide more information about the behavior of different consumer groups in future studies. CATA method is not the most recommended method to assess differences in sensory characteristics of very similar samples (Ares et al., 2014). In spite of that, it was possible to find differences in four attributes that are usually associated to sugar contents in food. More specifically sensory differences could be determined by the use of trained sensory panel profiling in further studies. Further studies should also be carried out to assess lowering AS in nectar from the standpoint of children. Frequently considered big users of these products, it is known that children's sensory perception can differ from adults' perception. James, Laing, and Oram (1997) evaluated age and gender influences in the detection of taste stimuli and observed that male children showed significantly higher thresholds than those of female adults. However, in another study using different rating scales, James, Laing, Jinks, and Oram NHutchinson (2003) concluded that the sense of taste has reached maturity for the perception of suprathreshold sucrose stimuli by mid-childhood (8e9 years old), once similar sweetness response functions for adults and children were obtained with two of the three procedures. On the other hand, studies with elderly indicate an increase in sweetness threshold in this age group (Mojet, Heidemma, & Christ-Hazelhof, 2003), which demands specific studies. Additionally, a high effect of age and gender on food choices and acceptance (Shepherd & Raats, 2006), is expected. In spite of that, the AS lowering in nectars as a public health policy in Brazil may contribute for sensory adaptation of the population from early ages, making it easier to achieve WHO recommendations in the future on sugar consumption. 5. Conclusion The sensory results of this study suggest that it is possible to reduce the AS in orange nectars by between 15 and 45% with very little impact on acceptance and sensory characteristics of the product. According to our results on JAR scale, thresholds, acceptance and CATA tests, these differences may not be perceived or may be acceptable. The reduction in sugar intake by the Brazilian population could reach almost 10 million tons per year after conclusion of such a lowering process within juices and nectars, if considered at current consumption levels. This work may contribute to public health policies of reduction of sugar in diets, which are mandatory in the face of obesity and chronic diseases profile associated with the consumption of AS. References rrega, A., Izquierdo, L., & Jaeger, S. R. (2014). Investigation of the number Ares, G., Ta of consumers necessary to obtain stable sample and descriptor configurations from check-all-that-apply (CATA) questions. Food Quality and Preference, 31, 135e141. Australian Food & Grocery Council. (2006). Health, nutrition and scientific affairs. http://www.afgc.org.au/our-expertise/health-nutrition-and-scientific-affairs/ Accessed 30.04.15. Biguzzi, C., Schlich, P., & Lange, C. (2014). The impact of sugar and fat reduction on perception and liking of biscuits. Food Quality and Preference, 35, 41e47. Bouhlal, S., Bernard, C., Issanchou, S., & Nicklaus, S. (2011). Sugar content impacts food intake in toddlers, but could be reduced. Appetite, 57S.
244
L.L.O. Pineli et al. / Appetite 96 (2016) 239e244
Brazilian Department of Agriculture. (2000). Decree nº 3.510, June 16, 2000. Change of regulation devices approved by decree no. 2314 of September 4, 1997, which provides for the standardization, classification, registration, inspection, production and supervision of beverages. http://extranet.agricultura.gov.br/sislegis-consulta/ consultarLegislacao.do?operacao¼visualizar&id¼1012 Accessed 05.05.15. Brazilian Department of Agriculture. (2009). Decree Nº 6.871, June 4th , 2009. Regulates law no. 8918 of July 14, 1994, which provides for the standardization, classification, registration, inspection, production and supervision of beverages. http:// www.planalto.gov.br/ccivil_03/_Ato2007-2010/2009/Decreto/D6871.htm Accessed 26.07.15. Brazilian Department of Agriculture. (2013). Normative ruling no. 42, of September 11, 2013. http://crn3.org.br/wp-content/uploads/2015/01/INSTRUCAO-NORMATIVAN42-11.07.13.pdf Accessed 05.04.15. Brazilian Department of Health. (2007). Technical cooperation agreement by Federal Government, between the Ministry of Health and the ABIA, in order to join forces and work together to implement actions to promote healthy lifestyles, which includes a healthy, balanced and nutritionally adequate diet. http://189.28.128.100/ nutricao/docs/geral/acordodecooperacaoabia_ms.pdf. Published November 2007. Accessed 05.04.15. Brazilian Institute of Geography and Statistics. (2010). Pesquisa de Orçamentos lise da disponibilidade domiciliar de alimentos e do Familiares 2008e2009: ana estado nutricional no Brasil. Rio de Janeiro: IBGE. Cardoso, J. M. P., Battochio, J. R., & Cardello, H. M. A. B. (2004). Equi-sweetness and sweetening power of different sweetening agents in differents temperatures of consumption of tea drink in soluble power. Brazilian Food Science and Technology, 24, 448e452. Cardoso, J. M. P., & Bolini, H. M. A. (2007). Different sweeteners in peach nectar: Ideal and equivalent sweetness. Food Research International, 40, 1249e1253. Chang, M. H., & Chiou, W. B. (2006). Differential threshold and psychophysical power function of sweetness sensation: applied psychophysics and prospect theory on formulating baking products. Journal of Sensory Studies, 21, 534e551. Chiou, W. B., Yeh, L. T., & Chang, M. H. (2009). Effects of health-related claims on the differential threshold of consumers' sweetness sensation. Journal of Sensory Studies, 24, 621e633. Chollet, M., Gille, G., Schmid, A., Walther, B., & Piccinali, P. (2013). Acceptance of sugar reduction in flavored yogurt. Journal of Dairy Science, 96, 5501e5511. Green, B. G., Lim, J., Osterhoff, F., Blacher, K., & Nachtigal, D. (2010). Taste mixture interactions: suppression, additivity, and the predominance of sweetness. Physiology Behavior, 101, 731e737. Hoppert, K., Zahn, S., Puschmann, A., Ullmann, I., & Rohm, H. (2012). Quantification of sensory difference thresholds for fat and sweetness in dairy-based emulsions. Food Quality and Preference, 26, 52e57. Hu, F. B. (2013). Resolved: there is sufficient scientific evidence that decreasing sugar-sweetened beverage consumption will reduce the prevalence of obesity and obesity-related diseases. Obesity Review, 14, 606e619. ISO. (1983). Standard 5495: Sensory analysiseeMethodologyeePaired comparison test. James, C. E., Laing, D. G., Jinks, A. L., & Oram N,Hutchinson, I. (2003). Taste response functions of adults and children using differentiating scales. Food Quality and Preference, 15, 77e82. James, C. E., Laing, D. G., & Oram, N. (1997). A comparison of the ability of 8e9-yearold children and adults to detect taste stimuli. Physiology & Behavior, 62, 193e197. Jeong, M., Gilmore, J. S., Bleakley, A., & Jordan, A. (2014). Local news media framing of obesity in the context of a sugar-sweetened beverage reduction media campaign. Journal of Nutritional Education and Behavior, 46, 583e588. Kelly, G. A. (1955). The psychology of personal constructs. New York: Norton. Kist, B. B., Vencato, A. Z., Santos, C., Carvalho, C., Reetz, E. R., Poll, H., et al. (2012). rio brasileiro da fruticultura 2012. Santa Cruz do Sul: EditoraGazeta Santa Anua Cruz (p. 128). Lawless, H. T. (1979). Evidence for neural inhibition in bittersweet taste mixtures. Journal of Comparative and Physiological Psychology, 93, 538e547. Lawless, H. T., & Heymann, H. (1998). Sensory evaluation of food. Principles and practices. New York: Chapman & Hall. Liem, D. G., Mars, M., & De Graaf, C. (2004). Sweet preferences and sugar consumption of 4- and 5-year-old children: role of parents. Appetite, 43, 235e245.
Lima Filho, T., Minim, V. P. R., Silva, R. C. S. N., Della Lucia, S. M., & Minim, L. A. (2015). Methodology for determination of two new sensory thresholds: compromised acceptance threshold and rejection threshold. Food Research International, 76, 561e566. Longo-Silva, G., Toloni, M. H. A., Menezes, R. C. A., Asakuraa, L., Oliveira, M. A. A., & Taddei, J. A. A. C. (2015). Introduction of soft drinks and processed juice in the diet of infants attending public day care centers. RevistaPaulista de Pediatria, 33, 34e41. Macfie, H. J., Bratchell, N., Greenhoff, K., & Vallis, L. (1989). Designs to balance the effect of order of presentation and first-order carry-over effects in hall tests. Journal of Sensory Studies, 4, 129e148. Malam, S., Clegg, S., Kirwan, S., McGinigal, S., Raats, M., Shepherd, R., et al. (2009). .Comprehension and use of UK nutrition signpost labelling schemes. London, UK: Food Standards Agency (UK). Available http://webarchive.nationalarchives.gov. uk/20131104005023/http://www.food.gov.uk/multimedia/pdfs/pmpreport.pdf Accessed 30.04.15. Markey, O., Lovegrove, J. A., & Methven, L. (2015). Sensory profiles and consumer acceptability of a range of sugar-reduced products on the UK market. Food Research International, 72, 133e139. Milenkovic, D., Deval, C., Dubray, C., Mazur, A., & Morand, C. (2011). Hesperidin displays relevant role in the nutrigenomic effect of orange juice on blood leukocytes in human volunteers: a randomized controlled cross-over study. Plos One, 6, e26669. Mojet, J., Heidemma, J., & Christ-Hazelhof, E. (2003). Taste perception with age: generic or specific losses in supra-threshold intensities of five taste qualities? Chemical Senses, 28, 397e413. Moraes, P. C. B. T., & Bolini, H. M. A. (2010). Different sweeteners in beverages prepared with instant and roasted ground coffee: ideal and equivalent sweetness. Journal of Sensory Studies, 25, 215e225. Moskowitz, H. R. (1982). Sensory intensity versus hedonic functions: classical psychophysical approaches. Journal of Food Quality, 5, 109e137. Osornio, M. M. L., & Hough, G. (2010). Comparing 3-point versus 9-point just-aboutright scales for determining the optimum concentration of sweetness in a beverage. Journal of Sensory Studies, 25, 1e17. Park, S., Onufrak, S., Sherry, B., & Blanck, H. M. (2014). The relationship between health-related knowledge and sugar-sweetened beverage intake among US adults. Journal of the Academy of Nutrition and Dietetics, 114, 1059e1066. Prescott, J., Norris, L., Kunst, M., & Kim, S. (2005). Estimating a “consumer rejection threshold” for cork taint in white wine. Food Quality and Preference, 16, 345e349. Rocha, I. F. O., & Bolini, H. M. A. (2015). Different sweeteners in passion fruit juice: ideal and equivalent sweetness. LWT e Food Science and Technology, 62, 861e867. Roessler, E. B., Pangborn, R. M., Sidel, J. L., & Stone, H. (1978). Expanded statistical tables for estimating significance in paired preference, paired difference, duotrio and triangle tests. Journal of Food Science, 43, 940e941. Sartor, F., Donaldson, L. F., Markland, D. A., Loveday, H., Jackson, M. J., & Kubis, H. P. (2011). Taste perception and implicit attitude toward sweet related to body mass index and soft drink supplementation. Appetite, 57, 237e246. Shepherd, R., & Raats, M. (2006). The psychology of food choice (p. 399). Oxfordshire: CABI. Stone, H., & Sidel, J. L. (2004). Sensory evaluation practices (3rd ed., p. 408). New York: Academic Press. Umbelino, D. C. (2005). Sensory characterization by descriptive quantitative analysis juice time-intensity and mango pulp (Mangiferaindica L.) sweteners with different edulcorants. PhD thesis (Food Science and Nutrition) e FEA. Campinas, Brazil: UniversidadeEstadual de Campinas. World Health Organization. (2015). WHO guideline: Sugars intake for adults and children. http://apps.who.int/iris/bitstream/10665/149782/1/9789241549028_ eng.pdf?ua¼1. Published February 2015. Accessed 05.04.15. Zheng, M., Rangan, A., Olsen, N. J., Andersen, L. B., Wedderkopp, N., Kristensen, P. … Heitmann, B. L. (2015). Substituting sugar-sweetened beverages with water or milk is inversely associated with body fatness development from childhood to adolescence. Nutrition, 31, 38e44.