Cooking habits and food choices of individuals with Type 1 Diabetes Mellitus during the pandemic of COVID-19

To verify the association between cooking habits, socioeconomic data, and food choices of individuals with Type 1 Diabetes Mellitus during the pandemic of COVID-19.


I N T R O D U C T I O N
In February 2020, the COVID-19 pandemic started in Brazil [1]. A report of 72.314 cases of COVID-19 published by the Chinese Center for Disease Control and Prevention showed an increase in the mortality of diabetic people during the pandemic [2]. If contaminated with COVID-19, individuals with Type 1 Diabetes Mellitus (T1DM) and Type 2 Diabetes Mellitus (T2DM) without glycemic control [glycated hemoglobin (HbA1C)≥10%] are more likely to have a worse prognosis, and cases may evolve to death more easily compared to those with good glycemic control (HbA1c ≤ 7%) [3].
Current studies with people diagnosed with T1DM, which focus on the pandemic period, emphasize the importance of maintaining an appropriate glycemia control in this population, assessing a few behaviors that may influence the glycemia management [4][5][6].
One of the most important aspects of controlling glycemia is adhering to healthy eating habits that ensure proper food consumption, preventing and correcting nutritional deficiencies. This may boost immunity and grant better prognoses in cases of infection by COVID-19 [7].
Quarantine and social distancing were carried out in Brazil as prevention measures aimed to diminish the disease's impact [8]. However, it is possible that these social distancing measures may impair the glycemic control in patients with T1DM due to restrictions of outdoor physical activities, psychological issues, deregulated sleep, and ingestion of unhealthy food [9]. Isolation can impact the individuals' eating routine, leading to a considerable increase of high-energy dense foods' ingestion [10].
According to Fundação Oswaldo Cruz (Fiocruz, Oswaldo Cruz Foundation), there was a decrease in the consumption of fresh foods during the pandemic, while the consumption of industrialized foods (ultraprocessed and processed) increased, such as packaged snacks and frozen foods, chocolates and sweets, mainly by young adults [11]. Still, some studies show that, in this period, individuals with diabetes lowered their adherence to diets, with rising consumption of carbohydrates and snacks, although an increase in the consumption of fruits has been reported [5,12]. Thus, the prolonged permanence at home may affect individual choices regarding cooking habits and buying ready-made foods in a higher frequency [13].
In this context, one of the behaviors studied as a form of adhering to eating better are cooking habits, which have been associated with improvements in eating attitudes and self-efficiency, promoting the adoption of healthier eating habits in adults and children [14,15]. In Brazil, the reformulation of the Guia Alimentar para a População Brasileira (Dietary Guidelines for the Brazilian Population) was a step toward stimulating this habit in the attempt to improve the quality of the population's food [16].
Jomori et al. [17] mention that the term "cooking skills" is still not very clearly defined in the literature. However, it is generally understood as the individual's ability to perform tasks that involve all the stages of food preparation, having as fundamental requirements cooking confidence and individual knowledge.
According to Marco de Referência de Educação Alimentar e Nutricional para as Políticas Públicas (Reference of Food and Nutritional Education for Public Policies) [18], the habit of preparing their own meals contributes to autonomy, which, according to national and international guidelines for diabetes, is related to self-management [19;20]. In this context, the American Association of Diabetes Educators mentions that part of diabetes self-care is linked to healthy behaviors, including the adoption of healthy eating [21].
Grabia et al. [22] aimed to assess the impact of the COVID-19 pandemic, nutrition, and health behaviors in individuals with T1DM and T2DM in Poland, and observed that more than half of the participants started cooking during this period, though the associated factors were not assessed. Regarding the period before the pandemic, it was observed that people with T2DM obtained significantly improved levels of glycated hemoglobin and, consequently, improved glycemia control, from interventions focusing on culinary and economic skills [23]. Nonetheless, despite these benefits, in a study with adult Spanish people during isolation due to COVID-19, less than half of the participants (44.4%) cooked [24].
Hence, aware of the benefits of preparing their own meals in promoting a healthy diet, and given the scarcity of studies on this theme that assess the associated factors in individuals with T1DM, it is expected that people with T1DM who cooked more during the period of social distancing, present better eating habits. Therefore, the present study intended to verify the association between cooking habits, socioeconomic and demographic data, and food choices of individuals with T1DM during the pandemic of COVID-19 in Brazil.

M E T H O D S
This is a transversal, descriptive, and analytical study. It was carried out in July 2020, when measures of social distancing were in place in Brazil. The sample was selected by convenience sampling with nonprobabilistic methods. Participation was voluntary, anonymous, and subject to acceptance of the Informed Consent Form (ICF). The study included individuals over 18 years old, of both sexes, and diagnosed with T1DM. To reduce the chances of another audience taking the online survey, participants chose their condition on the form (T1DM, not diagnosed with diabetes, having other types of diabetes, being a child/ adolescent, or caregiver). When the answer was not equivalent to the expected audience, the survey closed automatically.
Data from people who marked an alternative besides diagnosed T1DM were excluded at the beginning of the online form. When people did not complete the survey or did not agree with the terms of the ICF, choosing the option "I do not accept to participate" available at the beginning of the form, the survey closed automatically.
This research was approved by the Research Ethics Committee of the Center for Tropical Medicine of the Federal University of Pará, under process number 4.147.663, following the Helsinki Declaration.
Data collection was performed using an online form built on a digital platform in the opinion survey format. A pilot study was carried out with two participants to assess the time taken to complete the form and proper understanding of the questions. Participants were attracted by virtual invitations published on the social media networks Facebook ® (Menlo Park, United States of America), Instagram ® (Menlo Park, United States of America), Whatsapp ® (Menlo Park, United States of America), and in social media groups targeted at people with T1DM.
The following information was collected: (a) Socioeconomic, demographic, and social distancing data: sex; age; schooling (no higher education and higher education, complete or ongoing); income during the pandemic (remained the same or increased; decreased); the Brazilian macro region where they lived; type of distancing measures adopted (total, those who were did not leave their homes for any activity; partial, leaving only to buy food and medicines; no distancing because they needed to work outside; no distancing because they did not agree with the measures; distancing for family reasons, despite not agreeing); (b) Eating habits during the quarantine, compared to the pattern before the pandemic: cooking habits (cooking more than before; cooking as much as before; cooking less than before; they did not know how to cook; they did not like to cook, but someone else was cooking); increase in food consumption (increased a lot; increased a little; remains the same; decreased); number of daily meals (less than five meals; five or more meals a day); daily fruit consumption (appropriate, consumption of two to three servings; and inappropriate, consumption of less than two servings or more than three); daily vegetable consumption (appropriate, corresponding to two or more servings; and inappropriate, less than two servings); sweets consumption (much more than before; a little more than before; remains the same; decreased); consumption of ultraprocessed foods, considering frozen ready-made foods such as nuggets, cheese bread, pizza, or pasta (more than before; the same or less than before); delivery meals (does not order or orders less; orders the same or more than before); and water intake (equal; less; more than before).
For statistical analysis, we used the Statistical Package for Social Science software, version 21 (Chicago, United States of America). The Kolmogorov-Smirnov normality test was applied, and the descriptive results were expressed in absolute frequency and proportion. For the analytical assessment, the cooking variable had its categories grouped in order to classify it as yes, if they were cooking more, or no, regardless of the reason. The Chi-Square test of independence was applied with adjusted residual analysis, considering a level of statistical significance of p<0.05.

R E S U LT S
The research evaluated 472 individuals with T1DM. Most were female (86.0%), aged between 25 and 44 years old (57.0%), had reached higher education (78.4%), kept or increased their incomes during the pandemic (53.2%), and lived in the Southeastern region of Brazil (47.1%).
Regarding the cooking habits, 50.9% of the participants reported that they were cooking more during the quarantine, followed by 23.5% who were cooking as much as before. About 11.2% reported that they did not like to cook, but someone else cooked at home. Only 8.3% affirmed they were cooking less than before, and 6.1% said they did not know how to cook ( Figure 1).
An association was found between cooking more during the pandemic and having higher education (p<0.000). Not adhering to social distancing because of work was directly associated with not cooking (p=0.006) ( Table 1).
Most of the participants reported a general increase in the amount of consumed food (61.2%): 50.4% reported having five or more daily meals; 42.6% had appropriate fruit consumption; 53.4% had appropriate vegetable consumption; and 47.9% reported having increased their sweets consumption to some degree; 31.4% affirmed to be consuming more ultra-processed foods; 28.4% reported an increase in water intake; and 46.0% ordered delivery meals in a lower frequency or did not order such meals at all. 8 Were cooking more than before Cooking more during quarantine was associated with consuming less than five meals a day (p=0.04), having appropriate fruit (p=0.02) and vegetable consumption (p=0.04), and increased water intake (p=0.01) ( Table 2).

D I S C U S S I O N
The present study found associations between the cooking habits, socioeconomic and demographic data, and food choices of individuals with T1DM during the pandemic of COVID-19 in Brazil. Regarding cooking habits, half of the sample cooked more during social distancing. In the study by Grabia et al. [22], 65% of participants with T1DM and T2DM reported that they started preparing their meals during social distancing due to COVID-19 in Poland. No other studies were found that verified cooking habits in people with diabetes during the pandemic. However, regarding the general population, Carroll et al. [25], a study with 256 families with young children in Canada during the pandemic found that 70% of mothers and 68% of fathers reported spending more time cooking compared to the pre-pandemic period. For the French population, Deschasaux-Tanguy et al. [26] described the eating behavior of 37.252 adults during the period of social distancing. They observed that 47.6% of the participants were cooking more. In Spain, Note: Residual analysis: (+) Significant association (-)Negative significant association; *p<0,05; **Chi-square test.
Pérez-Rodrigo et al. [27] found that 14.1% of individuals who did not cook started cooking. Therefore, an observable increase in the cooking habits during the quarantine corroborated the study of Ruiz-Roso et al. [28], in which people spent more time preparing food and improving their eating habits due to measures of social distancing. In the present study, the reasons for cooking more during the pandemic were not investigated. However, we must highlight the great inequalities concerning income and access to food in Brazil [29]. During the pandemic, according to Silva Filho and Gomes Júnior [30], social distancing measures directly affected the income of a large part of Brazilians, contributing to an increase in food insecurity.
Mazzonetto, Dean, and Fiates [31] conducted an integrative review of the perception of individuals about cooking in the domestic environment in a period before the pandemic. They observed that for Brazilians, cooking also contributes to the control of family expenses and livelihood, even food resources and scarce financial resources. Therefore, Uggioni et al. [29] affirm that because of social inequalities in Brazil, food preparation can represent an increase in domestic work for some people, while for others, it can help reduce emotional stress, being a form of leisure and relaxation. It is also important to note that, according to Uggioni et al. [29], there was an increase in the number of influencers and chefs transmitting cooking classes online due to the need for social distancing. That may contribute to the development of culinary skills, promotion of healthy eating, and exchange of recipes, which reinforce social connections and affective memories involving food [29]. In the study by Byrne et al. [23] the authors noted that educational plans that include the development of culinary skills (recipe preparation and meal planning) aimed at people with diabetes can positively impact the treatment of these patients, increasing self-efficacy and improving glycemic control.
Regarding socioeconomic and demographic data, the present study showed that most participants had completed higher education courses, and there was a significant association between cooking more during the pandemic and this category of schooling. Grabia et al. [22] observed that 49% of the participants with T1DM also had higher education. In the study by Pal et al. [32] assessing the knowledge, attitudes, and practices of young adults with T1DM concerning COVID-19 during the lockdown in India, it was observed that 41% of participants had completed undergraduate-level schooling, and 32% had completed graduatelevel schooling.
We did not find any studies that analyzed the association between schooling and cooking habits in subjects with T1DM. However, according to the Pesquisa Nacional por Amostra de Domicílios Contínua (PNAD, Continuous National Household Sample Survey) [33], in Brazil, performing household chores increases in direct proportion with the level of schooling. So, in 2018, 90.1% of Brazilians with complete higher education levels performed domestic activities, compared with 82.2% of activities performed by people with no education or with incomplete basic education [33].
In the study by Deschasaux-Tanguy et al. [26] carried out in France, the authors found that those who reported spending more time cooking had higher levels of education. One hypothesis for these results is that individuals with higher levels of schooling may be more concerned with their health; therefore, they seek ways to improve their eating behaviors, such as preparing their meals. Another hypothesis is that people with higher educational levels may have greater chances of working at home, thus causing an increase in domestic responsibilities, among them, the preparation of food.
As for the type of social distancing measures adopted, not doing it because of work was associated with not cooking. In the study by Bracale and Vaccaro [34], the authors mention that food preparation, including items like bread, pizzas, and cakes, can be analyzed as a way to rediscover family ties, to spend time together during the period of social distancing, besides being a positive reaction associated with boredom due to the reduction of activities and routine under the policy of mandatory homestay. It is important to highlight that, in the present study, we did not investigate which foods were prepared by the evaluated subjects. However, we suggest that this association may be related to the lack of changes in the participants' routines. Thus, it may not have been necessary to seek alternatives, such as food preparation, to spend time and ease the boredom of staying at home.
In this study, a positive and significant association was found between the cooking habits and the consumption of fewer than five meals a day during social distancing. Besides observing that participants with T1DM and T2DM started cooking during the pandemic, Grabia et al. [22] found that 3% of individuals were consuming one to two meals a day, 50% consumed three to four meals a day, and 47% consumed more than five meals a day. Nevertheless, the authors did not assess the association between the variables. Hence, from the results of the present study, we hypothesize that when individuals prepare their meals, they decrease in daily fractioning due to the process of cooking requiring more time and daily organization. Besides, the fact that the individual remains at home for long periods may generate a decrease in daily physical activities, with a consequent reduction in energy expenditure, causing less hunger and, therefore, less fractioning of meals. On the other hand, staying at home for long periods can increase demands, such as home office, care for children, and monitoring of online classes, which are also factors that can influence less fractionation of meals due to decreased time availability. A significant association was observed between the habit of cooking more and the appropriate consumption of fruits and vegetables. According to the Brazilian Diabetes Society [19], appropriate consumption of fruits and vegetables contributes to a better supply of antioxidants, helping to modulate the oxidative stress observed in individuals with diabetes. No studies were found to relate these variables in people with T1DM in the context of the pandemic. However, Grabia et al. [22] also observed an increase in the consumption of fresh fruits and vegetables. Similarly, Carroll et al. [25] observed that in addition to mothers and fathers spending more time cooking during the pandemic, there was an increase in fruit consumption for parents and children. Even so, both studies did not assess associations between cooking and the consumption of fruits and vegetables. In a study of subjects with T2DM during the lockdown in India, the authors found that 20% of the sample reported increased fruit consumption, and 7% started eating fruit during the lockdown [12]. Ruiz-Roso et al. [28] examined the impact of the lockdown on eating habits in patients with T2DM and found a significant increase in vegetable consumption. For the authors, this can be explained as families have more free time at home to prepare more elaborate meals. However, none of the studies investigated cooking habits.
A systematic review and meta-analysis investigated the effects of culinary interventions on dietary intake, anthropometric measures, behavioral change, cardiometabolic outcomes, and quality of life in children and adults in the general population before the pandemic discovered that studies that used culinary workshops as interventions found that these are associated with improvements in eating attitudes, selfefficiency, and a healthier food intake, which includes the increase in the ingestion of vegetables, fruits, and food rich in fiber, besides reducing the injection of fatty food after cooking workshops [15]. These data corroborate the findings of the present study, indicating that the habit of cooking is related to an improvement in the quality of the food consumed. Hence, the importance of Food and Nutrition Education to encourage the habit of preparing meals during and after the period of social distancing is demonstrated as a strategy to motivate the adoption of healthier eating practices.
An association between the increase in water intake and the habit of cooking more was also observed. Not cooking was also associated with the maintenance of the water intake pattern prior to quarantine. In the study by Grabia et al. [22], an increase in water intake in individuals diagnosed with T1DM and T2DM was also observed. Deschasaux-Tanguy et al. [26] found as well that during confinement, in addition to the increase in cooking habits, 13% of the participants reported an increase in water intake, but no association between the two variables was assessed. Based on the results of the present research and the authors cited, the hypothesis suggested is that during the preparation of meals, individuals remember to drink more water because they are in the storage and supply place, unlike those who were not cooking any more than before and maintained the same water intake.
One of the present study's limitations is being an online survey, which limits the participation only to individuals with internet access. In addition, the instrument used has not been validated. However, the questions were based on the pillars of the treatment of people with T1DM recommended by the Brazilian Diabetes Society [19], in addition to sociodemographic information that we consider to interfere with the habit of cooking. Another limitation is not having evaluated the glycemic control of individuals, making it impossible to analyze whether there was an association between increased frequency of cooking and improvement in glycemic control. In addition, we did not investigate which foods were prepared by the participants and the reasons that led to the increase in the cooking habits.
However, this study was the first to make associations between the habit of cooking, socioeconomic and demographic factors, and eating habits in individuals with T1DM. No other studies analyzing these associations were found. Therefore, the need for further studies on this topic involving people diagnosed with T1DM is emphasized, since the relation between cooking and improving the quality of food choices in the general population and glycemic control in patients with T2DM is well established in the literature [14,15,23]. Other studies are suggested to identify the determinants involved in increasing the practice of preparing meals during the pandemic, since it can occur due to social inequalities, increased domestic activities, decision to improve eating habits, to have better glycemic control, and others.

C O N C L U S I O N
The present study observed the associations between the increased cooking habits during the pandemic of COVID-19, and high schooling, consuming less than five meals a day, appropriate consumption of fruits and vegetables, and increased water intake. Not cooking was associated with not practicing social distancing because of work and with a water intake equal to the previous period of the pandemic.
Despite its limitations, the study importantly showed the scenario in which people with T1DM are inserted and emphasized the lack of similar studies in the literature. It is also important to highlight that during the pandemic in Brazil, the habit of cooking may represent an increase in domestic activities, reinforced by social inequalities. Therefore, understanding these individuals' cooking habits and food choices can broaden health professionals' views, in addition to pointing out ways to modify public policies, even in the post-pandemic period.
It is extremely important to increase investments in public policies in the country, which encourage the development of cooking skills, as well as the habit of cooking as part of the actions of Food and Nutrition Education and Education in Diabetes, once the preparation of the food may not only encourage healthier eating, but also contribute to autonomy and self-care for people with diabetes.