Literature review

Relationship between quinoa, nutrition and COVID-19

The COVID-19 crisis accelerated the trend towards healthy eating globally, which was driven by the greater concern of consumers over the quality and nutrition of their food. The desire to consume nutritional foods emerged in this period because many consumers became increasingly aware of the importance of maintaining good health (^{Borsellino et al., 2020}). ^{Graziano da Silva (2020)} reports that the relationship between nutritional status and immunity has been widely demonstrated, while the ^{WHO (2021)} reported that people with chronic diseases, such as cardiovascular and respiratory diseases, obesity, diabetes and cancer, are particularly vulnerable to the COVID-19 virus. Thus, a good diet provides better protection against COVID-19, both by reducing mortality and increasing recovery without chronic health problems. This good nutrition is seen as necessary to combat the virus and undoubtedly involves access to a variety of more nutritious foods (^{Borsellino et al., 2020}).

Access to nutritious food was a major challenge under the lockdown measures that were imposed in many countries because of the pandemic (Graziano da Silva, 2020). The distribution chains of fresh products such as fruit and vegetables were affected by the constraints of social distancing due to COVID-19 (^{Borsellino et al., 2020}). Moreover, a survey carried out in May 2020 by the Peruvian Institute of Statistics and Informatics (INEI) to citizens from Lima and Callao indicated that 14% of households lacked the economic means to purchase the necessary protein foods for a balanced diet (^{Presidencia del Consejo de Ministros de Perú, 2020}). Nonetheless, the nutritional quality of quinoa as a whole grain stands out when compared to other staple foods, such as wheat, corn or rice (^{Hernández et al., 2020}), for being an amino-acid complete product that is easy to digest, thus meeting the growing expectations of consumers regarding healthy diets (FAO-ALADI, 2014; ^{Gamboa et al., 2018}; ^{Muziri et al., 2021}). Furthermore, quinoa is found in a wide range of processed foods and beverages, such as snacks, bars, functional drinks, breakfast cereals, liqueurs, baby foods and whole grain foods, being also suitable for vegan diets (^{Hernández et al., 2020}; ^{MINCETUR, 2019}). Therefore, to analyze this topic, we formulated a second hypothesis:

H2: Peruvian people who are concerned about their health and nutrition needs tended to consume more quinoa during the COVID-19 outbreak.

Theory of planned behavior (TPB) applied to the health and socioeconomic aspects influenced on quinoa consumption

Based on the precautionary principle alone, choosing quinoa appears to be an entirely rational decision (^{Higuchi & Maehara, 2021}). One of the most important assumptions of TPB is that consumers are rational in their decision-making processes and their actions, so that cognitive approaches can be used to anticipate behaviors (^{Ajzen, 1991}; ^{Lee & Yun, 2015}). This theory presumes that the likelihood of a certain behavior is a function of the human being's conscious intention to perform it (^{Scholderer & Grunert, 2001}), which in turn is assumed to be a weighted average of three conceptually independent variables: (i) attitude towards the behavior, which is determined by beliefs about its outcomes and their evaluation (^{Higuchi & Maehara, 2021}) and entail consideration of the outcomes of performing that behavior; (ii) subjective norms (or social norms), which are determined by beliefs about the expectations of others (perceived social pressure) and motivation to acquire a behavior (^{Ajzen, 1991}); and (iii) perceived behavioral control, determined by self-efficacy and beliefs about the controllability of facilitating conditions, past experience and inhibitory factors or anticipated difficulties (^{Ajzen, 1991}).

The diverse non-sensorial factors have a significant influence on a consumer's food choice. Therefore, understanding the role of these factors enhances the comprehension of dietary behavior (^{Carrillo et al., 2011}). Recent studies have classified the attributes of foods into those that accomplish healthy nutritional and functional attributes (^{Bedoya-Perales et al., 2018}; ^{Lee & Yun, 2015}; ^{Zevallos et al., 2014}), convenience (^{Marquart et al., 2008}), health and safety guarantees (FAO-ALADI, 2014; ^{Nasir & Karakaya, 2014}), and special characteristics like expression of cultural traditions (FAO-ALADI, 2014). As a result, when clients make food purchasing decisions, health-related attributes and non-health-related attributes, such as taste, familiarity and convenience (^{Lee & Yun, 2015}), become relevant. Consumers' perceptions towards the health benefits of foods are important for understanding preferences (^{Carrillo et al, 2011}; ^{Mitterer-Daltoé et al, 2013}; ^{Muziri et al, 2021}; ^{Ragaert et al, 2004}; ^{Rojas-Rivas et al, 2019}; ^{Verbeke & Vackier, 2005}). Quinoa can meet at least two of these features: it is a healthy food and it has ethnic characteristics associated with cultural traditions (FAO-ALADI, 2014; ^{Hernández et al, 2020}).

Additionally, reviews of the socioeconomic variables used in the literature to study the influence of numerous factors on the behavior of specific groups of people have been widely developed (^{Füsun-Tathdil et al, 2009}; ^{Misra et al, 1991}). For instance, ^{Hjelmar (2011)} suggests including demographic characteristics as explanatory variables to describe, for example, a typical organic food consumer, where ^{Nosi et al. (2020)} explored the quinoa case. There is some evidence that demographic characteristics such as age, gender, education and income influence consumers' attitudes and purchasing behavior towards superfoods (^{Encalada-Añazco et al, 2021}), specifically craft food (^{Rivaroli et al., 2020}). In the specific case of quinoa consumption, it is related to the demographic consumers belong to, their regional location, and the price of the product (^{Ayala-Macedo, 2003}). It is also said that the demand for healthy food increases at higher income levels (FAO-ALADI, 2014; ^{Nosi et al, 2020}). As an example, ^{Park et al. (2007)} mention that the "health-aimed stage" correlates income with an increased consciousness of the effect of food on human health, which means that the attainment of beauty and healthy living has a direct association with class level in society (^{Muziri et al, 2021}). Moreover, the income factor seems to influence the choice of the location and the quantity of craft food consumed (^{Rivaroli et al., 2020}). Consequently, we formulated a third hypothesis:

H3: The frequency of quinoa consumption is related to TFB factors and demographics.

Results of the binomial logistic models

Factors and demographic characteristics were incorporated in the four logistic regression models. Socio-demo-graphic characteristics were used to measure the Peruvian frequency of quinoa consumption by means of a logit model, where the dependent variable was quinoa consumption frequency (^{Al-Mazrooei et al., 2003}).

Influence of IYQ on quinoa consumption

It is important to determine whether this consumption pattern occurred before the 2013 IYQ or afterward. Many initiatives aimed at increasing quinoa consumption were carried out, mainly by Peruvian government institutions. Indeed, during this period, there was extensive dissemination campaigns to promoting Peruvian quinoa consumption internationally (^{Higuchi & Maehara, 2021}). Among other activities, there was further research on quinoa and the launch of new varieties, enactment of laws on food provenance and safety, and promotion of food fairs and partnerships with renowned chefs (^{Higuchi & Maehara, 2021}). Following these policies, there was an apparent increase in the internal demand by Peruvian population for Andean grains such as quinoa, mainly because of its multiple nutritional factors. Table 4 shows that 61% of the sample consumed quinoa before the IYQ. This means that consumers were already familiar with and consumed quinoa because they had the perception that quinoa is a traditional functional food. For instance, in the Ecuadorian highlands, particularly in Latacunga, quinoa is consumed as a tradition in typical dishes, but it is not associated with healthy lifestyle elements (^{Hernández et al., 2020}). The remaining 39% consumed quinoa only after the IYQ (table 4), which is associated with nutrition and health factors. It seems that nutrition and health have been the main reasons for consuming quinoa as a superfood (^{Encalada-Añazco et al., 2021}), with influences on the probability of consuming quinoa after the IYQ (table 6).

The perceived behavior control (PBC) factor was linked to quinoa consumption before the IYQ (table 6). The like li-hood of eating traditional meals made with typical ingredients relates to the attitudes of family members (^{Romo & Gil, 2012}). Family members are the most influential in terms of encouraging adolescents to eat healthily (^{Chan et al., 2011}; ^{Grønhøj et al., 2013}). Therefore, it is essential to make parents more aware of their role as socialization agents of quinoa food choices (^{Higuchi & Maehara, 2021}). Moreover, outcomes show that consumers who have the attitude-subjective norms or the attitude or subjective norms independently as the main factors also had a higher probability of consuming quinoa before the quinoa boom (IYQ). In contrast, being married or cohabiting is coupled with the probability of consuming quinoa more after the IYQ. This finding coincides with that of ^{Hernández et al. (2020)}, who mention that married people are more likely to consume quinoa, whereas the opposite was found for widowed or divorced people. For the SN&A-H-PBC and the A-H-PBC models, higher education corresponded to the probability of consuming quinoa before the IYQ, and in the SN-H-PBC model, the subjective norms factor had the strongest influence on the probability of consuming quinoa before the IYQ.

Increased quinoa consumption in Peru during the pandemic

Just 29% of Peruvian consumers mentioned that their quinoa consumption increased due to COVID-19 (table 4). Those consumers who previously had high monthly quinoa consumption (grams/month) showed a high probability of increasing quinoa consumption during the pandemic. Earning between 7,501 and 10,000 PEN/month, the second highest level of income in Peru, also impacted positively the probability of increasing quinoa consumption during the pandemic. Quinoa is a food that is in high demand in more affluent areas, as people with a higher income tend to stay well-informed of health trends and foods that make healthy living possible (^{Encalada-Añazco et al., 2021}; ^{Gonzalez, 2015}). As a result, the quantity and frequency of consumption are related to people's level of income (^{Ayala-Garay et al., 2020}; ^{Encalada-Añazco et al., 2021}), which is particularly the case of people in Top Lima and Modern Metropolitan Lima, who are aware of the nutritive qualities of trendy Andean crops.

The results are in line with the expectation that education or income level (proxied by a district dummy involving household income) would be reasonably homogeneous within small residential areas (^{Hanley & Morgan, 2008}), and that these factors could have a positive effect on quinoa consumption, as is the case of healthy foods such as fish (^{Can et al., 2015}). Additionally, while the health construct connects to the probability of increasing quinoa consumption during the pandemic in the SN&A-H-PBC and the A-H-PBC models, in the SN-H-PBC model, the main reason for consuming quinoa was also health and nutrition (table 7). It can be said that Peruvian people who are concerned about their health and nutrition needs consumed more quinoa during the COVID-19 outbreak. ^{Hernández et al. (2020)} explain that, in Ecuador, knowledge of quinoa's nutritional value and its use as a dietary ingredient increases the probability of consumption. The benefits of consuming more nutritious food emerged particularly during COVID-19 because many consumers became increasingly aware of maintaining good health (^{Hernández et al., 2020}) and boosting the immune system (^{Borsellino et al, 2020}).

Consumption frequency and purchasing at supermarkets

The results of the consumption frequency model are presented in table 8. As the family's primary dietary decision maker, women's purchasing attitudes are important (^{Duncan et al, 2015}). Consequently, in the three proposed models, women were primarily the ones who connected with the probability of quinoa consumption on a daily to weekly basis. It can be said that the frequency of consuming quinoa is related to the TPB factors and demographics. In the SN-H-PBC model, the constructs subjective norms, years of study and monthly quinoa consumption increased the probability of daily to weekly frequency of consumption.

The frequency of quinoa consumption was then cross-referenced with the variable "quinoa consumption during COVID-19" (table 9). It is said that quinoa is eaten in the homes of all socioeconomic strata once a day on average (^{Ayala-Macedo, 2003}). Nonetheless, approximately 46% of the sample stated they eat quinoa only once every fortnight or month (table 4). Moreover, Peruvians in the sample did not increase their quinoa consumption level during the COVID-19 pandemic in Peru. This is because quinoa is perceived as a traditional product, even though it has been shown that many traditional foods have characteristics of functional foods that help to improve health or prevent illness (^{Rojas-Rivas et al., 2020}). Additionally, prior to the pandemic, quinoa consumption had already been promoted by the United Nations Food and Agriculture Organization (FAO) (^{MIDAGRI, 2020}).

Finally, to know where consumers purchase quinoa, table 10 shows the outcomes of purchasing at supermarkets or other shops. It has been found that the general preference for shopping in the central regions of Latin American cities has shifted away from more traditional outlets to more "modern" channels, such as super/hypermarkets and drugstores, due to a perception of cleanliness and safety (^{Borsellino et al, 2020}). This is the main reason consumers prefer to purchase quinoa at supermarkets. Factors such as earning between 7,501 and 10,000 PEN/month and more than 10,000 PEN/month impact on the probability of purchasing at supermarkets during the COVID-19 pandemic. This means that people with the highest income have more probability of purchasing at supermarkets (p-value < 0.001) than other people who earn less income.

The reasons for this trend could be the convenience of shopping at supermarkets compared to other options such as small markets, street fairs or biofairs. Also, since packaged foods are easier to store and prepare, they can be more attractive to consumers (^{Graziano da Silva, 2020}). Due to the lockdown and the increasing need for people to cook at home, some preferences for less healthy foods may have declined (^{Borsellino et al, 2020}; ^{Eftimov et al, 2020}). In addition, being older (up to 65 years) was related to the probability of purchasing at supermarkets during the pandemic. This behavior is likely due to higher consumption rates of superfoods among people with an average age over 56 years, because of the nutritional quality of these foods (^{Delicato et al, 2019}; ^{Encalada-Añazco et al, 2021}) and the wide range of them being sold at supermarkets. Moreover, middle-aged individuals with a higher income level are more likely to purchase craft food (^{Rivaroli et al., 2020}). Corroborating this, ^{Hernández et al. (2020)} found that households headed by younger people tend to consume more of these foods.