Effect of a health claim on consumer acceptance of milk-based dessert containing omega-3

This study was to evaluate the influence of information on the addition of omega-3 and its health benefits in the acceptance of dairy dessert. The hedonic thresholds methodology was applied in two trials: blind and open-label. In the blind trial, no information was provided to consumers regarding the formulation of the flans. In the open-label trial, consumers received the information on the addition of omega-3 and its health benefits. The value of compromised acceptance threshold (CAT) was not influenced by the information on the addition of omega-3 and its health benefits. However, there was a high increase in the value of hedonic rejection threshold (HRT) when the consumers were informed of the benefits of omega-3, and it was possible to increase the addition of the compound in the flans by up to 21.03% without causing product rejection, confirming the positive effect of information on health benefits in accptance.


Introduction
Traditional sensory analysis, which studies the sensory characteristics arising from the balance of food composition, is essential to ensure its quality standard. However, it is insufficient to meet the quality requirements of the dynamic food market.
The optimized product formulation is indeed necessary for successful innovation and acceptance. Consumers, however, are also influenced by other factors during the process of choosing, purchasing, or accepting food (Minim, 2018).
Among the non-sensory characteristics, the concern with health and the consequent interest of consumers in more nutritious foods stand out. Studies show that information about the addition of the nutrient and its health benefits has great relevance in product acceptance (Monahan et al., 2017;Vidigal et al., 2011).
Lima Filho et al (2015Filho et al ( , 2017Filho et al ( e 2018 aimed to meet the demand for hedonic thresholds proposed and validated the hedonic thresholds methodology (HTM). This methodology allowed us to determine two new affective thresholds: the compromised acceptance threshold (CAT) and the hedonic rejection threshold (HRT). CAT is defined as the intensity of the stimulus in which changes in the sensory acceptance of the product begin to occur. The HRT is the stimulus intensity at which sensory rejection of the product begins, that is, the transition point between acceptance and sensory rejection.
HTM proved to be effective to assess the reduction of sucrose, butter, and sodium in foods (Gamba et al., 2021;Gamba et al., 2020, Santana et al., 2020, and for the addition of nutrients such as ferrous sulfate (Simiqueli et al., 2019). However, there are no reports in the literature on the application of HTM to assess the influence of non-sensory characteristics on hedonic thresholds.
However, because it gives an unpleasant taste to foods, its insertion becomes a challenge for the food industry. Thus, it is of fundamental importance to study the influence of the health benefits of omega-3 on hedonic thresholds, aiming at the development of a new product with functional properties without compromising sensory acceptability.

Methodology
This study was conducted at the Laboratório de Propriedades Tecnológicas e Sensoriais dos Alimentos at Universidade Federal de Viçosa (UFV), under the approval of the Human Research Ethics Committee of UFV (process number 077164/2019).
Employing two trials, blind and open-label, the influence of the information on the addition of omega-3 and its health benefits on the acceptance and on the values of hedonic thresholds in a strawberry-flavored dairy dessert was evaluated. In preliminary tests with several products, strawberry-flavored dairy dessert was selected, because in this product the characteristic flavor of ômega-3 was better masked.
In the blind trial, the flan samples were encoded with random three-digit numbers and no information was given to consumers regarding the formulation of the flans. In the open-label trial, consumers received the information on the addition of omega-3 and its health benefits when evaluating the samples. The results were analyzed by comparing the hedonic thresholds (CAT and HRT) obtained in both tests, and the analysis of variance. The flan-type dairy dessert was chosen due to the high consumption and acceptance of individuals of different age groups and simple preparation.

Preparation of strawberry-flavored flans
The flans formulations control (without addition of omega-3) and stimulus (with addition of omega-3) were composed of condensed milk, cream, filtered water, gelatin with strawberry flavor and strawberry essence ( Table 1). The addition of omega-3 was performed only in the stimulus samples (Table 1). In preparation, gelatin was previously dissolved in water at 100 °C and then added to the other ingredients and homogenized for two minutes in a Philips Walita Duravita 800 W blender, at speed 5.
The flans were placed in 50 ml plastic cups at a temperature of 8º C until the sensory test was conducted.
Omega-3 oil had its mass discounted in the mass of sour cream in each formulation. Omega-3 concentrations were defined considering the recommended daily intake (1 g of omega-3).

Consumers
The sensory panels in both trials (blind and open-label) were composed of 120 regular consumers of flan-type dairy dessert, recruited at the UFV, Minas Gerais.
The consumer team in the blind trial was composed of 63 women and 57 men aged between 18 and 62 years. In the openlabel trial, the consumer team consisted of 69 women and 51 men aged between 18 and 67 years.

Hedonic Threshold Methodology -Blind trial
To determine the thresholds of compromised acceptance (CAT) and hedonic rejection (HRT) of flans, we applied the hedonic thresholds methodology (HTM) developed and validated by Lima Filho et al. (2015,2017,2018). First, 120 consumers conducted five acceptance test sessions. In each session, consumers received two flan samples encoded with three-digit numbers: a control sample, with no omega-3, and a stimulus sample, added with omega-3. Within each pair, the samples were randomized, and among the pairs, the samples were served in ascending order of omega-3 concentration in the stimulus sample.
Participants were instructed to taste the samples from left to right and to evaluate how much they liked or disliked each product in relation to the overall impression, using the 9-point hedonic scale form, where 1 corresponds to "I disliked it extremely" and 9 to "I liked it extremely" .

Hedonic Threshold Methodology -open-label trial on the addition of omega-3 and its health benefits
We applied the open-label trial with the information on the addition of omega-3 and its health benefits similarly to the blind trial. However, the 120 consumers were first instructed to read the information on the benefits of omega-3 ( Figure 1), which remained in the consumer cabin throughout the assessment.

Figure 1.
Information about the benefits of omega-3.

Source: Authors (2021).
In addition to the informative, we also code the samples differently to the blind trial. The control samples received the identification of "Flan" and the other samples, with the addition of omega-3 (0.050 g; 0.225 g; 0.400 g; 0.575 g and 0.750 g), were identified with "Flan with 0.050 g of omega-3", "Flan with 0.225 g of omega-3", "Flan with 0.400 g of omega-3", "Flan with 0.575 g of omega-3", and "Flan with 0.750 g of omega-3", respectively.

Comparison of hedonic averages in each session
To assess the significance between hedonic notes of the stimulus samples obtained in the blind and open-label trails, the analysis of variance was performed in a randomized block design (α = 5%). In this way, the hedonic scores of the stimulus samples of each session were compared in both trials, blind and with open-label on the addition of omega-3 and its health benefits.
Statistical analysis was performed using the free software R, version 3.5.3.

Determination of CAT and HRT in blind and open-label trials
Through the hedonic notes obtained with the analysis of the samples, one CAT and one HRT were determined for each assay.
In each test, for statistical analysis of CAT determination, for each session, the t-test for paired samples was performed, comparing the hedonic note of the control sample with the hedonic note of the stimulus sample. With the t values obtained in each session, a graph of the t value (y-axis) was drawn up as a function of the omega-3 concentrations of the stimulus samples (x-axis). A regression model was fitted to the data, considering the significance of the regression parameters and the determination coefficient R 2 (SSregression/SStotal).
The CAT value was calculated using the regression model equation, corresponding to the concentration of omega-3 in which the calculated t value is equal to the tabulated t value (p = 0.05; 119 consumers). This means that from that point on, the change in sensory acceptance of the product begins. Thus, the tabulated t value (1.9801) was replaced by the y in the equation to obtain the value of x, which corresponds to the CAT, that is, the added omega-3 concentration in which sensory acceptance impairment begins to occur of the flan.

Benefits of Omega-3:
ü Reduces total cholesterol and LDL (bad cholesterol) ü Anti-inflammatory properties ü Improves athletic performance ü Acts in the prevention of heart disease, diabetes, Alzheimer, and neurodegenerative diseases ü Helps in cancer control In each trial, for statistical analysis and determination of the HRT, a graph was drawn up with the average of the hedonic notes of the stimulus samples obtained in each session (y-axis) as a function of the omega-3 concentrations (x-axis). A regression model was adjusted, considering the same criteria used to select the CAT model.
The HRT was calculated using the regression model equation, with the cut-off point corresponding to the hedonic note 5 (hedonic term "indifferent"), indicating the beginning of the sensory rejection of the product. The variable y of the adjusted model was replaced by 5 and the value of x, corresponding to the HRT, was calculated, that is, it is the concentration of omega-3 at which sensory rejection of the strawberry flan begins to occur. Table 2 shows the averages of the hedonic scores of the stimulus samples obtained in each session, in the blind trials and open-label trails on the addition of omega-3 and its health benefits. We observe that the information had no influence on the hedonic notes of the samples with 0.05% and 0.75% omega-3, the samples having the lowest and highest concentration of omega-3, respectively. That is, for such samples there was no influence of the information on the addition of omega-3 and its health benefits on the acceptance of the product at the level of 5% probability by the F test.   open-label trail on the addition of omega-3 and its health benefits. For both tests, the model that best fitted the data was the first order (Table 3), considering the significance of the regression parameters (p ≤ 0.05) and determination coefficients greater than or equal to 92%. The dotted line represents the tabulated t-value after which there is a significant difference in relation to acceptance (GL = 119; p = 0.05). To determine the compromised acceptance threshold (CAT), the y of the model was replaced by the tabulated t value For both tests, the model that best fitted the data was the first order (Table 4), considering the significance of the regression parameters (p ≤ 0.05), the determination coefficients greater than 96%. The dotted line represents the hedonic note 5, corresponding to the term "indifferent", the beginning of sensory rejection. The value of the HRT was calculated by replacing the value of y in the model with 5 (corresponding to the hedonic term "indifferent") and by calculating the value of variable y (value of the HRT) (Table 4) It was observed that when consumers were informed about the addition of omega-3 and its health benefits, the HRT value was higher than when consumers were unaware of the addition of the compound. Therefore, by informing the consumer about the addition of omega-3 and its health benefits, it is possible to add even more omega-3 without resulting in sensory rejection of the strawberry flan.

Hedonic thresholds
The HRT models (Table 4)

Discussion
In the evaluation of the average of hedonic notes of the stimulus samples (Table 2), we observe that the acceptance of flans containing 0.05 g of omega-3/100 g of flan does not differ between blind and open-label trails, therefore, information influenced the acceptance of flans containing 0.05 g/100 g. This may have happened due to the low concentration of omega-3, which is not enough to alter the sensory characteristics and, therefore, resulted in high acceptance by the samples, regardless of the presence of the information.
Numerically, the hedonic scores of all stimulus samples were higher in the open-label trials in relation to the blind trials.
In addition, in sessions 2, 3 and 4, this difference is significant (Table 2). In sessions 2 and 3 the product was accepted, and this This shows that despite the unpleasant taste conferred by omega-3 in this concentration, consumers were influenced by information about the addition of omega-3 and its health benefits.
In session 5, when 0.75g/100 g of omega-3 was added to the product, no influence of the information on the hedonic note was observed, due to the unpleasant taste conferred by the high concentration of omega-3. Thus, the fact that consumers are aware of the health benefits of omega-3 was not enough to alter their rejection, also demonstrating the influence of sensory characteristics. This result corroborates with the results found by Vidigal et al. (2011) when assessing the influence of information on the health benefits of exotic juices. The authors concluded that camu-camu juice was rejected in both blind and open-label trials due to the unpleasant taste of the fruit.
When investigating the influence of information on the compromised acceptance threshold (CAT), it was observed that when consumers were unaware of the addition of the compound, the CAT value was 0.079 g of omega-3/100 g of flan. When consumers were informed about the addition of omega-3 and its health benefits, the CAT value was 0.082 g of omega-3/100 g of flan ( Figure 2). Thus, we observe that value of the CAT was not influenced by the information about the addition of omega- FeSO4/100 g flan) and HRT (0.1274 g FeSO4/100 g) were lower than those found in this study (0.079 g omega-3/100 g and 0.477 g of omega-3/100 g), which indicates the unpleasant taste of ferrous sulfate, even more intense than that of omega-3.
With the increase in the concentration of omega-3 added to the flan, the thresholds of hedonic rejection (HRTs) are obtained, which consists of the intensity of the stimulus from which the sensory rejection of the product begins. As shown in The addition of omega-3 in the flans from the concentration referring to the CAT and lower than the one referring to the HRT indicates that the product had its acceptance altered in relation to the control sample, although it is still accepted. Thus, the flan added with 0.225 and 0.400 g of omega-3/100 g has acceptance, but this differs from the acceptance of the control flan.
Regardless of the assay, samples containing concentrations of 0.575 and 0.750 g of omega-3 per 100 g of flan, showed concentrations of omega-3 higher than the HRT (0.447 and 0.541 g of omega-3 per 100 g of flan) (Figures 2 and 3), that is, these samples were sensorially rejected and the presence of information about the addition of omega-3 and its health benefits did not allow the reversion of the rejection for sensory acceptance by these samples.
Considering the recommended daily intake (RDI) of omega-3 (EPA and DHA) of 1 g (Couet et al., 2001;Agostoni et al., 2012), the CATs values obtained in the blind and open-label trials, for 100 g of product, correspond to 7.9 and 8.2% of the RDI of omega-3, respectively, which allows it to be considered by the legislation as an "added" product of omega-3. For the product to be considered "enriched" or "fortified" the addition of the compound must occur at least 30% of the RDI in the case of solid foods. Using the equations in Table 3, it appears that a product containing 0.30 g of omega-3 per 100 g of flan (x = 0.3) would have a hedonic score equal to 5.9 in the blind test, next to the hedonic term "I liked it slightly", and 6.6 in the essay with information, placing between the terms "I liked it slightly" and "I liked it moderately". Thus, it is observed that it is possible to add omega-3 in sufficient concentrations to be considered "enriched" or "fortified" maintaining the sensory acceptance of the flan. In addition, it appears that the acceptance of this product will be even greater if the consumer is informed about the benefits of adding omega-3. Although the addition of strawberry aroma was the only strategy in an attempt to mask the unpleasant sensory characteristics inherent in omega-3, it was possible to add the compound in sufficient concentrations so that the flan could be characterized as enriched or fortified without causing sensory rejection to the product.
Comparing the values of the HRTs with the RDI, it is observed that it is possible to add concentrations lower than 44.7 and 54.1% of the omega-3 RDI to the flan without and with information, respectively. This allows to characterize the flan as an "enriched" or "fortified" product with omega-3 (Brasil, 1998).
The effect of health benefits on the acceptance of flans containing omega-3 can be associated with the fact that, currently, the population has been increasingly concerned with health, and has tried to live healthier, being, therefore, willing consuming healthier foods (Euromonitor Intrnational, 2017). Several studies have also found the influence of non-sensory characteristics, especially information on the addition of compounds to health benefits in the acceptance of products, showing that when consumers are informed about the presence of such compounds, acceptance tends to increase (Monahan et al., 2017;Vidigal et al., 2011).
It is worth mentioning that the consumption of omega-3 can bring health benefits, acting in different functions in the body and in the control of neurodegenerative and allergic diseases, as well as diabetes, anemia, Parkinson's, Alzheimer's, cancer, among others. Thus, the increased consumption of omega-3 tends to bring benefits to consumers. However, the industries still face difficulties regarding the fortification of omega-3 in foods, since omega-3 (EPA and DHA) is majorly obtained from oily fish the nutrient brings with it the characteristic flavor from its source, considered unpleasant to the taste of consumers (Kolanowski et al., 1999).
Therefore, it is of great interest to industries to determine to what concentration they can add omega-3 without affecting the acceptance and without resulting in the sensorial rejection of products by consumers. This information was provided for the first time through this study, with the determination of the hedonic thresholds for concentrating omega-3 in strawberry-flavored flan.

Conclusion
Information on the health benefits of omega-3 positively influenced the acceptance of the strawberry-flavored flan when omega-3 was added in intermediate concentrations (0.225; 0.400 and 0.575 g/100 g). However, at extreme concentrations (0.05 and 0.75 g/100 g) the acceptance was not influenced. This occurred because at the lowest concentration, the product was accepted, and at the highest concentration the unpleasant taste characteristic of omega-3.
Through HTM, it was identified that it is possible to increase the addition of omega-3 by 21% without sensory rejection of the flan. Therefore, by providing information on health benefits, it is possible to make greater additions of omega-3 without causing sensory rejection of the flan.
Thus, it is concluded that the information on health benefits can positively influence sensory acceptance, as long as there is sensory pleasure.