Introduction
In the face of the COVID-19 pandemic, the Brazilian Ministry of Health, in February 2020, following guidelines from the World Health Organization (WHO), prepared the National Contingency Plan for Human Infection by the new Coronavirus COVID-19. As a palliative measure not to paralyze teaching in schools that adopted the face-to-face model of formal education, the Ministry of Education (MEC), in turn, published regulations to enable the flexibility of educational actions by educational institutions in Brazil 1-3.
Thus, there was a global change in the way of teaching and teachers were challenged to adapt to a new working arrangement and most of them faced challenges related to pedagogical and technological skills and the home context, environment, often not suitable for professional activities (4,5,6). Teachers were pressured in a short period of time to carry out a quick and unstructured change of work at home, to know the platforms used, to adapt their face-to-face content to the remote system, to restructure their teaching to be in contact with students per screen tasks that in the context of the pandemic have become a necessity 7,8,9. The lack of quality internet access, availability of computers and cell phones in the family nucleus are examples of obstacles that are intertwined with this new work configuration and the domestic routine and, the teaching work, which is characterized by being exhausting, can be even more stressful in view of the exceptional health situation 7,10.
Given this complex context, the following question emerged: how did the abrupt changes in teaching work and in the teaching-learning process, as well as work overload, influence your eating habits, as this relationship has already been determined in different contexts 11.
Studies have shown that the period of social isolation due to COVID-19 influenced eating behaviors and food choices of several populations with different outcomes in relation to changes in eating patterns 12,13,14. For example, some of these studies show healthier eating behaviors, including organic farming and fruit and vegetable consumption 15. Nevertheless, others show unhealthy eating habits including less frequent consumption of vegetables,
fruits, and legumes, greater adherence to fast food, increased consumption of frozen foods and reduced consumption of natural foods 12,13,16. There are also those who have shown that the pandemic has not significantly changed their diet 15 and who have shown a greater proportion of increased consumption of different food groups, healthy or unhealthy, due to social isolation 17. Although these studies show a variety of scenarios in relation to food behavior, little is known about the relationship between the teaching work and food consumption in this context of pandemic.
Thus, the present study had the purpose of describing associations between home-working conditions and food consumption of teachers from the primary education of Minas Gerais during suspension of faceto-face classes.
Materials and methods
Population
The study population was composed of teachers working in primary education at state schools in the state of Minas Gerais, Brazil and that took part in Projeto ProfSMoc- Etapa Minas Covid (Profs Moc Project - Minas Covid Phase), 2020.
Sample planning and inclusion criteria
For sample calculation, a formula based on the prevalence of disease or event, considering an infinite population 18. A prevalence of 50% was considered in order to obtain a larger sample size and consequently a greater inference power to infer different variables. Tolerable error adopted was 3%. Furthermore, the sample was duplicated (Deff=2), since it came from conglomerates. The sample size was increased in 20% to compensate for possible losses (rate of non-respondents) which might compromise the validity of the study. Thus, the need to collect data from 2,564 teachers to ensure the representativeness for the state of Minas Gerais. However, the questionnaire was answered by 15,641 teachers. Of this number, 269 individuals were excluded from the analysis, as they did not meet the study inclusion criteria, totaling a final valid sample of 15,372. Inclusion criteria were regent teachers in the year of 2020, working in the primary education and accepting to take part in the research. The study was approved by Comitê de ética em Pesquisa Envolvendo Seres Humanos (Research Ethics Committee Involving Human Beings) under protocol n. 4.200.389/2020 and was conducted according to the guidelines of the Helsinki Declaration.
Data collection
Authorizations and partnerships were obtained from Secretaria de Estado de Educação de Minas Gerais (Secretary of State of Education of Minas Gerais). Data collection occurred from August 20th to September 11th, 2020. Cross-sectional study using online self-applied questionnaire, available for participants through the Google Forms platform and widely circulated by the Secretary of State of Education of Minas Gerais in social media. The questionnaire addressed sociodemographic questions as well as those related to remote working conditions, and food consumption. Teachers signed the written consent form to take part in the research. Anonymity was guaranteed by the platform, not being possible to associate participants’ e-mails to their responses. In order to prevent auto fill the research form, a reCAPTCHA test was used thus avoiding that a form was filled by a robot.
Food consumption in this study was measured using a validated food frequency questionnaire (FFQ) used by the telephone survey system for surveillance of risk and protective factors for chronic diseases (Vigitel) of the Ministry of Health, Brazil. Participants were asked about the frequency of their consumption of healthy food: raw or cooked vegetables and legumes, fruits, beans, whole foods (bread, rice, cookies, pasta, whole grains, oats, granola, and flaxseed). And they were also asked about the frequency of their consumption of unhealthy food: processed meats (ham, salami, mortadella, frankfurter, sausage, or hamburger), frozen dishes (pizza, lasagna, etc.), pack snacks, sweets, pies and chocolate, in addition to soft drinks and artificial juices. Dependent variables analyzed were healthy and unhealthy food consumption.
For analysis, responses were recategorized as follows: each group of healthy food receive a score (0 to 4), with 0 as the group with the highest frequency of consumption (every day), 1 (5 to 6 days a week), 2 (3 to 4 days a week), 3 (1 to 2 days a week) and 4 (never or almost never). For each group of food regarded as unhealthy, this score was inverted, so that, the higher the score of each participant, the worse is their diet. For further analysis, a cut-off point was established to represent these two standards of interpretation: above P75 increased consumption of unhealthy foods and below the 75th percentile with no changes in diet or increased healthy food consumption during the pandemic.
Independent variables included sociodemographic factors, remote working conditions, understanding of online technologies, access to a computer and quality of internet connection. Sociodemographic variables studied were as follows: age (in years: 21 to 40; 41 to 60; 61 or over); gender (male; female) and income (1 to 2 minimum wages; 3 to 5 minimum wages; 5 minimum wages or more). Regarding remote working conditions, the following aspects were assessed: weekly working hours (up to 20 h; 21 to 40 h; 40 h or over); feeling about the teachers’ work during the pandemic (satisfied; indifferent; dissatisfied); workload during the pandemic (worked the same amount; worked more; worked less; overworked) and difficulties to teach online (none; a few/moderate; many). The understanding of online technologies was assessed in three categories (excellent/good; regular; poor/terrible). The access to a computer and the quality of internet connection were assessed considering whether the teacher had a computer at home (yes, for personal use; yes, shared with other people; does not have a computer) and the quality of internet connection at home (excellent/good; regular/ poor; not available).
Data analysis
Data were analyzed using descriptive statistics and relative frequencies. The following were considered as independent variables: sociodemographic variables, working conditions, understanding of online technologies, access to a computer and quality of internet connection. Outcome variable was food consumption, dichotomized as healthy or no changes (score up to P75) and unhealthy (score over P75). Chi-square test was performed to check the association between the variable of eating behavior and independent variables. Variables with p-value Prevalence Ratio (PR), at the 0.05 significance level. Poisson regression analysis was performed to assess which variables were predictive of either healthy or unhealthy food consumption during the pandemic. The test of deviance was used in order to evaluate the quality of the adjusted model. All analyses were carried out using the computerized package Statistical Package for the Social Sciences (SPSS), version 20.0.
Results
The sample was composed of 15,372 participants, with a predominance of females (81.9%). Most aged 40 or older (58.8%) and income from 3 to 5 minimum wages (59.5%). During the pandemic 62.2% of those interviewed worked from 20 to 40 hours per week; 33% were dissatisfied with the remote work and 79.7% reported working more or much more than usual during the pandemic. Regarding computer availability at home, virtually half of the participants (46.7%) shared the computer with other people and 45.6% informed that the quality of their internet connection at home was regular or poor. Regarding food consumption, 20% of teachers had an unhealthy food consumption during the pandemic, 80% did not change their eating habits or had a healthy food consumption (Table 1).
In the bivariate analysis, virtually all variables were seen to be associated with food consumption, except for the following variables: income, understanding of online technologies and computer availability at home (Table 2).
In the adjusted multivariate model analysis, there was a significant association between eating habits and the following variables: gender, age, working hours (per week), feeling about the teachers’ work during the pandemic, workload during the pandemic and quality of internet connection. Regarding gender, males had a higher prevalence ratio (PR=1.26; 1.17-1.36) of unhealthy food consumption,
Variables | n | % | ||
---|---|---|---|---|
Sociodemographic characteristics | ||||
Gender | ||||
Female* | 12817 | 81.9 | ||
Male | 2824 | 18.1 | ||
Age | ||||
21 to 40 years* | 6447 | 41.2 | ||
41 to 60 years | 8793 | 56.2 | ||
60 or older | 401 | 2.6 | ||
Income | ||||
1 to 2 min. w.*, a | 3969 | 25.4 | ||
3 to 5 min. w. | 9301 | 59.5 | ||
5 min. w. or more | 2371 | 15.2 | ||
Working conditions | ||||
Working hours (per week) | ||||
Up to 20 h* | 4433 | 28.3 | ||
21 to 40 h | 9735 | 62.2 | ||
40 h or over | 1471 | 9.4 | ||
Feeling about the teachers’ work during the pandemic | ||||
Satisfied* | 3375 | 21.6 | ||
Indifferent | 6995 | 44.7 | ||
Dissatisfied | 5271 | 33.7 | ||
Workload during the pandemic | ||||
Worked the same amount* | 2143 | 13.7 | ||
Worked more | 6755 | 43.2 | ||
Worked less | 1037 | 6.6 | ||
Overworked | 5706 | 36.5 | ||
Difficulties to teach online during the pandemic | ||||
None* | 1472 | 9.4 | ||
A few/moderate | 10961 | 70.1 | ||
Many | 3208 | 20.5 | ||
Understanding of online Technologies | ||||
Excellent/good* | 9840 | 62.9 | ||
Regular | 5019 | 32.1 | ||
Poor/terrible | 782 | 5.0 | ||
Have a computer at home | ||||
Yes, for personal use* | 7757 | 49.6 | ||
Yes, shared with other people | 7310 | 46.7 | ||
No | 574 | 3.7 | ||
Quality of internet connection at home | ||||
Excellent/good* | 8423 | 53.9 | ||
Regular/poor | 7126 | 45.6 | ||
Not available | 92 | 0.6 | ||
Food consumption | ||||
Healthy or no changes (up to P75) * | 12509 | 80.0 | ||
Unhealthy (over P75) | 3132 | 20.0 | ||
Total | 15641 | 100.0 | ||
Lower than P25 | 3325 | 21.0 | ||
Between P25 and P50 | 4821 | 31.0 | ||
Between P50 and P75 | 4363 | 28.0 | ||
Over P75 | 3132 | 20.0 |
*reference category. a1 min. w.* = 198.15 US dollars
Independent Variables | Food consumption | |||
---|---|---|---|---|
Healthy n (%) | Unhealthy n (%) | Total n | P-value | |
Sociodemographic characteristics | ||||
Gender | ||||
Female* | 10399(81.1) | 2418(18.9) | 12817 | - |
Male | 2110(74.7) | 714(25.3) | 2824 | < 0.001 |
Age (years) | ||||
21 to 40* | 4779(74.1) | 1668(25.9) | 6447 | - |
41 to 60 | 7371(83.8) | 1422(16.2) | 8793 | < 0.001 |
60 or older | 3591(89.5) | 42(10.5) | 401 | < 0.001 |
Income (Minimum wage) a | ||||
1 to 2* | 3162(79.7) | 807(20.3) | 3969 | - |
3 to 5 | 7459(80.2) | 1842(19.8) | 9301 | 0.49 |
5 or over | 1888(79.6) | 483(2.4) | 2371 | 0.97 |
Working conditions | ||||
Working hours (per week) | ||||
Up to 20 h* | 3554(80.2) | 879(19.8) | 4433(100.0) | - |
21 to 40 h | 7832(80.5) | 1903(19.5) | 9735(100.0) | 0.70 |
40 h or over | 1123(76.3) | 348(23.7) | 1471(100.0) | 0.002 |
Feelings about the teachers’ work during the | ||||
pandemic Satisfied* | ||||
2806(83.1) | 569(16.9) | 3375(100.0) | - | |
Indifferent | 5672(81.1) | 1323(18.9) | 6995(100.0) | 0.01 |
Dissatisfied | 4031(76.5) | 1240(23.5) | 5271(100.0) | < .001 |
Workload during the pandemic | ||||
Worked the same amount* | 1798(83.9) | 345(16.1) | 2143(100.0) | - |
Worked more | 5464(80.9) | 1291(19.1) | 6755(100.0) | 0.002 |
Worked less | 857(82.6) | 180(17.4) | 1037(100.0) | 0.37 |
Overworked | 4390(76.9) | 1316(23.1) | 5706(100.0) | < .001 |
Difficulties to teach online during the pandemic | ||||
None* | 1218(82.7) | 254(17.3) | 1472(100.0) | - |
A few/moderate | 8796(80.2) | 2165(19.8) | 10961(100.0) | 0.03 |
Many | 2495(77.8) | 713(22.2) | 3208(100.0) | < .001 |
Understanding of online technologies | ||||
Excellent/good* | 7840(79.7) | 2000(20.3) | 9840(100.0) | - |
Regular | 4049(80.7) | 970(19.3) | 5019(100.0) | 0.15 |
Poor/Terrible | 620(79.3) | 162(20.7) | 782(100.0) | 0.79 |
Having a computer at home | ||||
Yes, for personal use * | 6168(79.5) | 1589(20.5) | 7757(100.0) | - |
Yes, shared with other people | 5880(80.4) | 1430(19.6) | 7310(100.0) | 0.16 |
No | 461(80.3) | 113(19.7) | 574(100.0) | 0.65 |
Internet quality at home | ||||
Excellent/good* | 6819(81.0) | 1604(19.0) | 8423(100.0) | - |
Regular/poor | 5617(78.8) | 1509(21.2) | 7126(100.0) | 0.001 |
Not available | 73(79.3) | 19(20.7) | 92(100.0) | 0.69 |
*reference category, a1 min. w.* = 198.15 US dollars
when compared to females. The age of 40 or older was associated with healthy food consumption (PR=0.63; 0.59-0.67) and (PR=0.79; 0.74-0.84), respectively. The following variables were associated with an unhealthy food consumption: working 40 hours a week or over (PR=1.18; 1.06-1.31), dissatisfaction about work (PR=1.26; 11.15-1.39), working more or much more during the pandemic (PR=1.34; 1.21-1.50) and regular or poor internet connection (PR=1.09; 1.02-1.16) (Table 3).
Independent Variables | Unhealthy food consumption | |||
---|---|---|---|---|
Adjusted PR | IC 95% | P-value | ||
Gender | ||||
Female* | 1 | - | ||
Male | 1.26 | 1.17-1.36 | < 0.001 | |
Age (years) | ||||
21 to 40* | 1 | - | ||
41 to 60 | 0.63 | 0.59-0.67 | < 0.001 | |
60 or older | 0.41 | 0.31-0.55 | < 0.001 | |
Working hours (per week) | ||||
Up to 20 h* | 1 | - | ||
21 h to 40 h | 1.03 | 0.96-1.11 | 0.36 | |
40 h or over | 1.18 | 1.06-1.31 | 0.003 | |
Feelings about the teachers’ work during the pandemic | ||||
Satisfied* | 1 | - | ||
Indifferent | 1.08 | 0.99-1.19 | 0.08 | |
Dissatisfied | 1.26 | 1.15-1.39 | < 0.001 | |
Workload during the pandemic | ||||
Worked the same amount * | 1 | - | 0.76 | |
Worked less | 1.03 | 0.87-1.21 | 0.76 | |
Worked more | 1.18 | 1.06-1.31 | 0.003 | |
Overworked | 1.34 | 1.21-1.50 | < 0.001 | |
Quality of internet connection at home | ||||
Excellent/good* | 1 | - | ||
Regular/poor | 1.09 | 1.02-1.16 | 0.008 | |
Not available | 1.06 | 0.71-1.59 | 0.77 |
*reference category - Deviance= 9715.960 P=0.622
Discussion
Our data has shown that teachers’ remote working conditions are associated with a worse food consumption, and that weekly working hours, the feeling of dissatisfaction and the workload during the pandemic, as well as the quality of their internet at home were the variables that correlate with unhealthy food consumption. Several studies have already described changes in eating behavior during the COVID-19 pandemic, which is quite reasonable 15,16,19-22. The stress associated with the coronavirus pandemic, subsequent closing of schools, moving and adapting to homeschooling work, combined with the convenience of easily accessible, palatable, and emotionally comforting foods were reported in another study of school teachers Basic 8. Nevertheless, our study analyzed aspects relating teachers’ remote working conditions to their diet for the first time. Studies conducted prior to the pandemic had already identified the absence of data on teachers’ and school staff’s food consumption and food choices 23. In another context and with different objectives from our study, Dias et al. (2020) showed that temporary contract teachers have the worst eating habits 11.
Changes in teachers' eating habits during the COVID-19 pandemic were evaluated in some studies that demonstrated opposite situations, one of them recorded inadequate eating habits 24, the teachers interviewed rarely consumed fruits and vegetables and the other observed a positive change in most eating habits 25. However, literature is still scarce on teachers’ working conditions and food consumption, therefore, results from the present study are of great relevance and may guide studies on nutritional intervention with the target population, in order to analyze and better understand whether the working conditions per se may change the food consumption, or trigger feelings and unfavorable eating behaviors. The relationship between stress at work and food choices has been explored and explained through physiological mechanisms, which favor, for example, the consumption of comfort food (high-calorie food, also those that are rich in carbohydrates and fats) 26,27. Thus, work overload and dissatisfaction with remote work, as potential sources of stress, may shed some light on associations with unhealthy food consumption by participants in this study.
Since professional activities have been online due to the pandemic, interviewed teachers were shown to have difficulties with the quality of their internet at home and their working conditions. These results have been corroborated by other authors who verified similar challenges and issues faced by school teachers during the pandemic, such as difficulties to teach online due to internet connection instability, among other issues 28-30. Mishra et al. (28) have also noticed damage to overall health of teachers who have been working online for a long time. In a recent study, Ekpanyasku and Padungtod 31 investigated a magnitude of occupational health problems and lifestyle changes among workers from different companies who started working from home during the pandemic. Irregular food consumption has been identified as one of the lifestyle changes related to the intensity of remote work 22,31).
In the present study, less healthy eating habits were observed among younger teachers. This result is corroborated by others research that considered adults and elderly people more careful and aware of the influence of diet on health 32,33. Remote work during the COVID-19 pandemic did not inaugurate the precarious situation of teaching work, however it brought even more work overload, with a transformation of pedagogical approaches and the need to learn new ways of teaching, situations capable of causing greater dissatisfaction with the work among teachers and impact on behaviors and habits during the pandemic 34
Since we have seen in the present study associations between remote teaching work conditions and a higher consumption of unhealthy food, it is important to look more closely at the health of such workers. Studies on relations among remote work, food consumption and health of teachers are beginning and there is a need for continuous monitoring of these teachers in the current context and when returning to face-to-face classes 35.Previous researchers have also highlighted the need to adopt public policies aimed at this class of workers, focusing on improvements in working conditions and restructuring strategies to promote health, encouraging preventive measures, encouraging mainly the regular practice of physical activities and change in eating habits 36-38.
The present study is limited. Among these limitations are the fact that it is a self-applied questionnaire, the fact that it is a crosssectional study, which limits the causal inference, as well as the assessment of food consumption, which is assessed according to food groups (healthy or unhealthy) and not according to consumption assessment questionnaires such as the 24-hour dietary recall, for example. The strengths of the study are the large sample size and the fact that it is the first to analyze teaching working conditions associated with changes in diet during the COVID-19 pandemic with a representative sample from the state of Minas Gerais, Brazil. It is concluded that remote working conditions during the pandemic influenced basic education teachers’ food choices, with working hours and workload, as well as dissatisfaction with remote work and the quality of the internet connection as variables that were associated with unhealthy food consumption. Further studies are required to deepen issues related to both remote and face-to-face teaching working conditions and their implications in their food choices.