Low Cooking Skills Are Associated with Overweight and Obesity in Undergraduates

Culinary skills are defined as the confidence, attitude, and the application of one’s individual knowledge in performing culinary tasks, and their development may be associated with better diet quality and better health status. This study aimed to analyze the association between cooking skills, overweight, and obesity in undergraduates. This is a descriptive, observational, and cross-sectional study, with data collected between October 2020 and March 2021, with undergraduate students (n = 823) at the Federal University of Rio Grande do Norte. Participants answered the online Brazilian Cooking Skills and Healthy Eating Questionnaire Evaluation, BCSQ, which included socioeconomic information. Logistic regressions were used to assess the associations of cooking skills with overweight and obesity. From the total of the students, 70.8% were female, with a median age of 23 (21–30) years; 43.6% were with overweight or obesity; 48.8% were eutrophic; and 7.7% underweight. Overweight and obesity were significantly associated with low levels of culinary self-efficacy and self-efficacy in the use of fruits, vegetables, and seasonings in the bivariate analysis. The logistic regressions showed that living with other people and eating out were associated with higher chances of overweight and obesity. Sharing the responsibility for preparing meals and a high self-efficacy in the use of fruits, vegetables, and seasonings were associated with lower chances for overweight/obesity. Overall, our study showed that overweight and obesity were associated with lower cooking skills in the studied undergraduates. Therefore, the study demonstrates that culinary skills can be explored in educational programs that aim to reduce overweight/obesity in students.


Introduction
Cooking skills are defined as the confidence, attitude, and the application of one's individual knowledge in performing culinary tasks, such as planning meals, shopping, and preparing different kinds of food (fresh and/or processed ones) [1]. Studies have shown the importance of the development of cooking skills for adopting healthy eating habits because homemade foods favor the consumption of vegetables, improving the nutritional quality of the meals [2][3][4][5].
Undergraduate students have a high consumption of ultra-processed foods. Fondevila-Gascón [6], for example, observed that 83% of their university student sample consumed snacks and nuggets weekly, as well as soft drinks and pastries. According to Monteiro et al. [7] ultra-processed foods are multi-ingredient formulations with food substances that mimic unprocessed foods' sensory qualities. The high consumption of this type of food, which requires little or no preparation technique, is related to the sporadic use of cooking skills [4,8].

Sociodemographic Characterization, Meal Preparation, and Consumption
The questionnaire had 15 questions about sociodemographic variables, meal preparation, and consumption characteristics. The questions sought information about gender, date of birth, undergraduate courses, when the participant was enrolled in university, and whether he/she was a beneficiary of any local university aid program. They also covered parental education, ethnicity, civil status, and whether they had children. Regarding the characteristics of meal preparation, there were questions about the time available to cook per day, the people responsible for preparing food at home, the equipment and kitchen utensils available for cooking, and where the main meal was consumed [29].

Nutritional Status
Weight (Kg) and height (m) data were self-reported by the participants in the questionnaire, and the values were used to calculate the body mass index (BMI). After calculation, the BMI was classified according to the World Health Organization [30].

Cooking Skills
The Brazilian questionnaire for the assessment of cooking skills and healthy eating (BCSQ) was used. The questionnaire is the result of an adaptation of the Cooking with Chef Program and was validated for Brazil (Cronbach's alpha = 0.70) [31][32][33]. The internal consistency of the BCSQ was also calculated in our sample, and the Cronbach's alpha was 0.70.
Scales 2-6 were considered for calculating the level of cooking skills, and these were classified as low (20-43 points), medium (44-73 points), or high (74-100 points). Availability of fruits and vegetables was classified as low (0-2 points), medium (3-6 points), or high (7-8 points). The knowledge of culinary terms and techniques was classified as high when the participant correctly answered ≥6 items or low when the participant answered ≤ 5 items correctly [29,31].

Statistical Analysis
The data obtained were moved from Google Forms to the Microsoft Excel program (2013) and coded for the analysis performed using the Statistical Package for the Social Sciences SPSS ® , version 18.0 (IBM Corporation, Armonk, NY, USA, 2011).
For the students' characterization variables, a descriptive analysis of categorical variables was performed by the distribution of the absolute (n) and relative (%) frequency and the discrete and continuous variables by the median (Q1-Q3), according to the nonnormality of the data, and verified using the Kolmogorov-Smirnov test.
Association tests were performed between the results of the BCSQ, the socioeconomic and meal preparation variables, and the presence of overweight or obesity. The chi-square test and Fisher's exact test were used, with the latter being used for the 2 × 2 tables.
The variables that showed a significant association in the univariate analysis with the presence of overweight or obesity were used for logistic regression models, first exploring the effect of a single variable in the presence of overweight or obesity (1 = yes, 0 = no), and their unadjusted odds ratios (OR) and respective 95% confidence intervals (95% CI) were demonstrated. Then, adjusted logistic regression models were calculated, considering the presence of overweight or obesity (1 = yes, 0 = no) as the dependent variable. The adjustment of the final model shown was guaranteed by observing the Omnibus test, with p values less than 0.05, and the Hosmer and Lemeshow test, considering p values greater than 0.05. Thus, sex, income, living arrangement, responsibility for preparing meals at home, place where the main meal was consumed, and scales of cooking skills (culinary self-efficacy and self-efficacy in the use of fruits, vegetables, and seasonings) were included in the final model as independent variables. Multicollinearity between the independent variables was tested and tolerance and VIF values in the final adjusted model were approximately 1.0. The adjusted odds ratios (AOR) and their respective 95% CI were presented.

Cooking Skills
Most of the students (65.7%) presented high cooking skills; 33.7% had medium cooking skills; and 0.6% had low cooking skills. Most of the students presented high levels of culinary attitude (54.7%), culinary behavior (84.7%), culinary self-efficacy (66.8%), selfefficacy in the consumption of fruits and vegetables (63.7%), and self-efficacy in the use of fruits, vegetables, and seasonings (71.4%) ( Figure 1A). Almost half of the students (46.5%) had a high availability of fruits and vegetables ( Figure 1B), and half of the students (50.5%) showed a high knowledge of culinary terms and techniques ( Figure 1C).

Cooking Skills
Most of the students (65.7%) presented high cooking skills; 33.7% had medium cooking skills; and 0.6% had low cooking skills. Most of the students presented high levels of culinary attitude (54.7%), culinary behavior (84.7%), culinary self-efficacy (66.8%), self-efficacy in the consumption of fruits and vegetables (63.7%), and self-efficacy in the use of fruits, vegetables, and seasonings (71.4%) ( Figure 1A). Almost half of the students (46.5%) had a high availability of fruits and vegetables ( Figure 1B), and half of the students (50.5%) showed a high knowledge of culinary terms and techniques ( Figure 1C).

Association of Meal Preparation and Consumption with Overweight and Obesity
There were more students with overweight and obesity who ate meals out of the home than those without overweight or obesity (14.8% vs. 8.4%, respectively; Table 3). There were also more students with overweight and obesity who were solely responsible for preparing meals than those without overweight or obesity (26.5% vs. 17.2%, respectively; Table 3). For those who received help from someone else in preparing meals, there

Association of Meal Preparation and Consumption with Overweight and Obesity
There were more students with overweight and obesity who ate meals out of the home than those without overweight or obesity (14.8% vs. 8.4%, respectively; Table 3). There were also more students with overweight and obesity who were solely responsible for preparing meals than those without overweight or obesity (26.5% vs. 17.2%, respectively; Table 3). For those who received help from someone else in preparing meals, there were more students without overweight or obesity who received help from their father than those with overweight or obesity (14.8% vs. 9.1%, respectively; Table 3). Considering the living arrangement, more students with overweight or obesity lived with a partner/spouse (13.1% vs. 7.3%, Table 3) or with a partner/spouse and children (14.5% vs. 5.4%, Table 3).

Association of Cooking Skills with Overweight or Obesity
There was a higher percentage of students with overweight or obesity presenting low levels of cooking self-efficacy (Chi-square, p = 0.026, Table 4) and low self-efficacy in the use of fruits, vegetables, and seasonings (Chi-square, p = 0.042, Table 4) compared with the students without overweight or obesity (5.9% vs. 3.4%, and 4.5% vs. 1.7%, respectively).

Logistic Regression for Variables Associated with Overweight and Obesity in the Studied Undergraduates
The sociodemographic variables and cooking skill scales associated with overweight and obesity were tested using logistic regressions ( Table 5). The logistic regression showed that living with a partner/spouse (AOR = 3.17; 95%CI = 1.42-7.04), living with a partner/spouse and children (AOR = 5.10; 95%CI = 2.17-12.00), and consuming the main meal out of the home (AOR = 1.89; 95%CI = 1.14-3.13) increased the chances of overweight and obesity. On the other hand, some variables reduced the chances of overweight or obesity: sharing the responsibility for preparing meals with one person (AOR = 0.44; 95% CI = 0.26-0.74); having another person responsible for preparing the meals (AOR = 0.42; 95% CI = 0.22-0.80); and a high self-efficacy in the use of fruits, vegetables, and seasonings (AOR = 0.32; 95% CI = 0.11-0.95).

Discussion
This study analyzed the association between cooking skills, overweight, and obesity in university students. Our data showed that living with a partner/spouse and children and eating the main meal out of the home were associated with overweight and obesity. However, sharing responsibility for preparing meals was associated with a lower chance of being with overweight or obesity. In addition, high levels of self-efficacy in the use of fruits, vegetables, and seasonings also reduced the chances of overweight and obesity.
We found that living with a partner/spouse and children increased the chances of overweight and obesity. Consistently with this finding, Lancaster [34] found an association between living with a spouse or partner and overweight/obesity in both men and women but with a stronger association for women. Another study [35] showed that the nevermarried population had the lowest prevalence of overweight and obesity when compared to those married, divorced, and widowed. Thus, the association of the marital status with overweight/obesity can be explained mainly by the increase in moments that encourage the regular consumption of meals, in addition to larger portions, resulting in greater energy intake, as well as a decrease in physical activity and a decrease in the desire to maintain bodyweight to attract a partner [36]. From these perspectives and considering the pandemic scenario, students were overwhelmed by distance learning. Still, they had to dedicate themselves to domestic activities, in addition to having to dedicate themselves to children who were in distance education [37]. These facts might explain why undergraduates living with a partner/spouse and children presented more chances for overweight and obesity in the present study.
However, on the other hand, Davis et al. [38] found that women who lived with their spouses had a lower BMI than those who lived with other people. Van Der Horst et al. [3] found that students who lived alone had a higher consumption of ready-to-eat meals compared to students who lived with other people, and this consumption of ready-to-eat meals was significantly associated with overweight and obesity. Based on these conflicting results, further studies are needed to understand better the association between housing conditions, meal consumption, and nutritional status, which may be also associated with socioeconomic variables.
Another factor associated with overweight and obesity in the present study was the responsibility of preparing meals. The logistic regression showed that sharing this responsibility with one more person or having another person cook the meal was associated with a lower chance of overweight and obesity. The Food Guide for the Brazilian Population highlights the importance of sharing part or all the activities that precede and follow the consumption of meals, including planning, acquisition, and preparation [5]. The responsibility of cooking for the family generally falls on the women, reinforcing the gender roles. Therefore, when cooking is divided among family members, the health status of the people involved is improved [5,39].
Additionally, we found in the bivariate analysis that more students with overweight or obesity reported being solely responsible for preparing their meals compared to those without overweight or obesity. Few studies have addressed the relationship between overweight and obesity and its association with the responsibility to cook at home. Ducrot et al. [40] evaluated the involvement in the choice of dishes for the home meal preparation of over 50 thousand French adults by an ongoing web-based prospective observational cohort study launched in 2009 with a scheduled follow-up of 10 years. The authors found that participants with overweight and obesity were more likely to choose dishes for home meals that were part of diets to lose weight. Additionally, their results suggested that participants with overweight and obesity were less likely to consider the importance of time for cooking, the availability of ingredients, and leftovers at home for planning meals [40].
Thus, in the context of undergraduates, having someone to help with meal preparations or another person to prepare the meals might have decreased the burden of house activities, stress, and help to maintain nutritional status. These facts might explain our results, and further studies should investigate living arrangements that help undergraduates in Brazil have a better quality of life.
In our study, more women reported knowing how to cook and having more time available for cooking than men. Women also presented higher levels of cooking skills and less overweight and obesity. These data are consistent with the results observed by Murakami et al. [41], suggesting that the difference in competence between genders is because women still have the greater responsibility for cooking and buying food. Our data are also consistent with the study by Dezanetti et al. [27], also studying undergraduates and using the BCSQ, which found that in the period before and during the pandemic, women were more likely to cook "several times a week". Over the years, there has been a tendency for an increase in the time men dedicate to cooking (from 35% to 46% in 13 years) [42]. However, women still dedicate more time to cook than men. This fact could have several explanations. For example, men may cook more for pleasure and entertainment, while women, in the majority, cook as a family routine, which strengthens social norms around gender and makes women more involved in cooking, feel more confident, and pass on these skills to their children. Wolfson et al. [43] and Taillie [42] found that women cooked more frequently than men [42,43], like our findings. Therefore, the present study shows that meal preparation remains a highly gender-related task and that sharing this task with other family members contributes positively to health.
Conversely, we observed that men had lower levels of cooking skills than women. This result was also observed in other studies, such as in the research by García-González et al. [44], who divided the study sample into five age groups. In all groups, women were the majority declaring knowing how to cook. Van Der Horst et al. [3] observed that, in addition to men having lower cooking skills than women, a barrier in the preparation of homemade food, being overweight, and poor cooking skills were associated with the consumption of ready-to-eat meals. Thus, these variables are interconnected and all decrease diet quality, which can lead to overweight and obesity, especially in men [11,45].
Furthermore, our study found that eating out, such as eating in a restaurant, a cafeteria, an a la carte restaurant, and eating fast food, increased the chances of overweight and obesity. Frequently eating out compromises diet quality, increasing energy, total and saturated fat, sugar, and sodium intake and contributes to decreasing the consumption of fresh foods and micronutrients [46]. Strengthening this proposition, Tani et al. [2] found that, despite the benefits of homemade food, the consumption of these preparations declined in their residences when out-of-home food, such as fast food and convenience food, emerged. Kim et al. [47] observed a positive association between eating out and a high BMI among women. Therefore, eating out is related to a higher BMI [2,47].
Thus, it is essential to reinforce how the development of cooking skills is a significant modifying factor in encouraging people to cook. By strengthening these skills, diet quality can improve [2]. Some studies have shown an association between high cooking skills and the lower consumption of ready-to-eat meals and ultra-processed foods among adults [3,4,48]. Other studies have shown that interventions aimed at improving cooking skills increase culinary confidence and result in better food choices, such as the increased consumption of fruits and vegetables [48,49].
Effective culinary interventions were observed in the study by Bernardo et al. [8]: a randomized trial in which Brazilian university students participated in a six-week culinary intervention with weekly meetings, five practical cooking classes, and a food shopping selection workshop. The intervention successfully increased the access and availability of fruits and vegetables; culinary attitudes; self-efficacy for using culinary techniques; self-efficacy in using fruits, vegetables, and seasonings; and knowledge of cooking terms and techniques. In addition, Flego et al. [49], using the Jamie's Ministry of Food Program as an intervention in Australian adults, improved culinary confidence and eating/cooking behaviors, contributing to a healthier diet for the participants. The intervention consisted of ten weeks with a 1.5 h weekly class. Like Bernardo et al. [8], the program emphasized cooking from scratch, nutritious meals, and good shopping choices. Thus, intervention programs should consider these aspects.
Interestingly, the bivariate analysis showed that there was a higher percentage of students with overweight or obesity, with both low and high culinary self-efficacy compared to students without overweight or obesity. This result shows that there seemed to be two profiles in the studied population: students with overweight or obesity who have high and low culinary self-efficacy. For students with overweight or obesity with low cooking self-efficacy, targeted actions to increase meal preparation and cooking skills could be interesting. For those with high culinary self-efficacy, actions aimed at preparing and consuming healthy foods in healthy preparations could be beneficial. However, in the multivariate analysis, culinary self-efficacy was not associated with overweight or obesity. Taking our results together (both the bivariate and multivariate analyses), the determinants of culinary self-efficacy could be further explored in other studies.
Culinary self-efficacy demonstrates confidence in cooking with respect to basic cooking skills. Our results suggest that for those with overweight and obesity confidence in cooking seems to decrease. Thus, developing culinary self-efficacy is important for healthy eating habits, such as a reduction in the consumption of processed foods and food autonomy, resulting in a better nutritional status [12,26,50].
We also found that students with overweight or obesity mainly had low-level knowledge of culinary techniques and terms. These students incorrectly answered questions involving roasting food, measuring cups, boiling water, and sautéing food. This technical knowledge and confidence in cooking are related to better diet quality and cooking using basic ingredients, which is an essential determinant of culinary behavior, according to Jomori et al. [1].
Thus, interventions in our population should address the creation of strategies to increase the frequency of cooking at home as a way of improving the quality of what is consumed, increasing the intake of fruits and vegetables, and reducing the consumption of ready-to-eat or convenience foods [8,51,52]. Studies point out that these culinary interventions present themselves as promising strategies to reduce the consumption of ready-to-eat and non-prepared foods at home [8,52].
Considering students' food consumption and its possible relationship with changes in BMI, the COVID-19 pandemic was decisive for changes in people's buying behavior and food consumption patterns [53]. This is because eating habits and behaviors can be affected by sociological, physiological, and psychological factors, and the pandemic period affected all of them [54], mainly due to social distancing to prevent the spread of the virus [24,50], which reduced trips to the supermarket and increased the use of delivery services [25] with a consequent decrease in vegetable intake and increased consumption of ultra-processed foods [26]. In addition, since the beginning of the pandemic, universities have closed their facilities and adopted distance learning, which has resulted in students dealing with an unexpected and potentially anxiety-producing situation [17,18]. As a result, there was an increase in food intake as anxiety can act as a trigger for people to use food in response to unfavorable emotions [19,20]. Thus, the increase in food intake associated with the higher consumption of industrialized foods may be related to the emergence of overweight in university students [20], which was observed in studies, such as those by Palmer et al. [55] and Tan et al. [53], who found weight gain in university students during confinement due to changes in eating patterns.
The present study has some limitations. The study involved students from a public university in Northeast Brazil, which was a particular group in terms of age, parental education, and ethnicity. Therefore, it is not possible to generalize the findings to other populations of undergraduate students from different Brazilian regions. Although we had a large sample size, sampling was not probabilistic, and there may have been selection bias, selecting those most interested in the topic under study. In the present study, the trend toward greater female involvement could be explained by their familiarity with the topic related to gender biases [12,50], as discussed. Nevertheless, several studies conducted online on other topics, such as food choice values, mental health, COVID-19, and fertility, showed greater female participation [41,[56][57][58], which may reflect greater adherence of this group to this method of collecting data. Thus, these limitations do not invalidate the research objectives and the associations found.
For strengths, the study presented a large sample size, allowing a multivariate analysis, and cooking skills were characterized using a scale validated with Brazilian undergraduates [32]. Cooking skills were associated with overweight and obesity, conditions directly related to health. In Brazil, some documents make health promotion strategies viable, such as the Food and Nutrition Education Framework for Public Policies [59] and the Food Guide for the Brazilian Population [5]. However, no public policies promote healthy eating specifically for the university community or community-based culinary skill interventions. Thus, our results can contribute to developing public policies on food and nutrition education at the university. Promoting cooking skills in these students can contribute to healthy habits, diet quality, and better awareness of choosing and making better foods.
The present study showed that better self-efficacy in the use of fruits, vegetables, and seasonings, a culinary skill construct, was associated with less chances for overweight/obesity. Additionally, eating out of the home was associated with higher chances for overweight/obesity. Therefore, the development and encouragement of cooking skills should be part of health promotion activities. The university can be considered a privileged setting for the implementation of cooking courses that help raise awareness about food, culinary ingredients, preparing and eating meals that promote health and well-being, and for teaching students how to prepare foods in the most nutritious way. This can help the students involved apply the technical information in everyday practice, serving as an intervention for improving health via cooking skills.

Conclusions
Our study showed that better self-efficacy in using fruits, vegetables, and seasonings, which is a culinary skill construct, and sharing the responsibility for preparing meals were associated with fewer chances for overweight and obesity. Living with other people and eating out of the home were associated with higher chances for overweight and obesity. Considering the bivariate and multivariate analyses, our findings showed that overweight and obesity were associated with lower cooking skills, so the present study confirmed our hypothesis. Thus, the university environment emerges as a place to develop and encourage health promotion measures by facilitating the development of culinary skills.  Informed Consent Statement: Consent was obtained from all subjects involved in the study. Data Availability Statement: Data described in the manuscript, code book, and analytic code will be made available upon request pending application and approval.