Positive and Negative Changes in Food Habits, Physical Activity Patterns, and Weight Status during COVID-19 Confinement: Associated Factors in the Chilean Population

The association between the changes in lifestyle during coronavirus disease 2019 (COVID-19) confinement and body weight have not been studied deeply. Therefore, the aim of the present study was to determine lifestyle changes, such as eating habits and physical activity (PA) patterns, caused by confinement during the COVID-19 pandemic and to analyze its association with changes in body weight. Seven hundred participants (women, n = 528 and men, n = 172) aged between 18–62 years old of the Chilean national territory participated in the study. Food habits, PA, body weight, and sociodemographic variables were measured through a survey in May and June 2020. The body weight increase presented positive association with the consumption of fried foods ≥ 3 times per week (OR; 3.36, p < 0.001), low water consumption (OR; 1.58, p = 0.03), and sedentary time ≥6 h/day (OR; 1.85, p = 0.01). Conversely, fish consumed (OR; 0.67, p = 0.03), active breaks (OR; 0.72, p = 0.04), and PA ≥ 4 times per week (OR; 0.51, p = 0.001) presented an inverse association with body weight increase. Daily alcohol consumption (OR; 4.77, p = 0.003) was associated with PA decrease. Food habits, PA, and active breaks may be protective factors for weight increase during COVID-19 confinement.


Introduction
Coronavirus disease 2019 (COVID-19) has affected many factors in all countries, including labor, economy, production, health, and consequently, lifestyle. This pandemic has led to strict decisions to control the chain of virus transmission, indicating physical distancing and a significant reduction in mobility as the primary prevention measure, calling on nations to implement quarantines and state plans that promote teleworking [1,2]. The first contagion notified in Chile was on 3 March 2020. By the middle of that month, communal quarantines began to be decreed according to indications of the Coronavirus plan implemented by the Ministry of Health. In this epidemiological panorama, Chile has accumulated up to 16 July 2020, 366,595 cases and 8580 deaths, and the country has one of the highest rates per million inhabitants in the world [2]. Although limitations on free transit and physical distance of more than 1.5 m per person are the most effective strategies to reduce COVID-19 completed in accordance with the Declaration of Helsinki (2013) and was approved by the Department of Physical Education, Universidad de La Frontera, Chile (project number DFP20-0032).

Self-Report Measures
The information was collected by an electronic survey designed by a multidisciplinary team uploaded to the Google forms platform. This instrument was piloted with 20 people to evaluate any unforeseen problems in question design and response collection (i.e., in April 2020). Once the survey was tested and validated by the team, it was shared by institutional emails, Facebook, Instagram, WhatsApp, and Twitter in May and June 2020 (i.e., for eight weeks). The survey was divided into four sections, including informed consent, personal history, food habits, and physical activity patterns.

Study Presentation and Informed Consent
The first section presented the details of the research and informed consent. Participants were asked to be as honest as possible and to report reality in all areas faithfully. The responses were anonymous and confidential, without reporting the name or any personal information. Participants were free to leave the survey at any stage before the submission process. Responses were only accepted and considered in data processing by the research team when "submit" was selected.

Personal History
The second section was related to personal background: sex (female, male), age (years), education level (primary, secondary, university, technical education, postgraduate), number of people who shared the same place for quarantine, socioeconomic level (low, middle-low, middle, middle-high, high), current occupation (medical leave, distance education, telework, unpaid domestic work, retired, blended work, unemployed, normal shift, independent work), marital status (single, married, common law married, separated, widowed), body weight (kg), and size (m). The body mass index (kg/m 2 ) and its classification category (normal weight, overweight, or obese) were calculated. In addition, they were asked for information on body weight before and after confinement (no change, increase, or decrease).

Food Habits
For the collection of antecedents related to eating, a daily and weekly consumption frequency survey was used. This tool consisted of gathering as much information about the frequency of weekly and daily beverage consumption (1, 2, 3, 4, 5, 6, and 7 times a week or does not consume) and food daily (1, 2, 3 or more times a day, or does not consume) from each food group (cereals, legumes, dairy products, red meats (beef, lamb, pork), white meat (chicken, turkey), fish, fruits, vegetables, water, and alcohol. Participants were consulted to determine unhealthy foods and the types of preparations or culinary techniques used, according to "The Dietary Guideline of the Chilean Population" (DGCP) [14]. Regarding behavioral changes compared to before COVID-19 confinement, questions were asked about increasing, maintaining, or decreasing diet in general or if cooking was performed more than before, less than before, or was maintained. Additionally, the general perception of diet was consulted, as to whether it was healthier, less healthy, or maintained compared to before COVID-19 confinement.

Physical Activity Patterns
Regarding PA, we asked about the frequency of PA per week (1, 2, 3, 4, 5, more than 6 times a week, or do not do it) and the number of minutes dedicated per session according to the current references of PA for the population over 18 years [11]. They were asked what type of PA they performed (combined or mixed, yoga or Pilates, aerobics or jogging, calisthenics, resistance training) and if they had checked social networks to find exercise routines (yes, no). Sedentary behavior was reported in hours per

Physical Activity and Food Habits
Regarding anthropometric parameters, 35.86% of the sample reported being overweight and 16.43% reported obesity. Women had a higher prevalence of obesity than men (p = 0.03). The men reported a higher PA (times/week and min/session, p < 0.001) than women. Men also performed a higher number of active breaks than women (p = 0.04). The highest percentage of the sample passed ≥6 h sitting or sedentary (54.4%) ( Table 2). According to eating habits, the highest percentage of the study sample drank 3-5 glasses of water per day (45.3%) and consumed 1-2 portions of vegetables per day (69.1%), legumes 1-2 times per week (83.7%), 1-3 fruits per day (53.3%), and the same amount of vegetables as before (48.4%). Regarding meat consumption, 55.5% consumed red meat, 65.0% consumed white meat, and 75.1% consumed fish 1-2 times per week. Women reported cooking at home more frequently than men (p = 0.04). The majority of participants declared to have maintained their eating habits but in regard to consumption, the majority declared to eat more than before (51.3%). In relation to the negative parameters, 30% of the sample reported consuming alcohol daily and eating junk food and fried foods 1-2 times per week (62.9% and 59.9, respectively) ( Table 3).   Figure 1 shows the change in body weight and PA patterns. Of the men, 25.6% and of the women, 38.1% reported an increase in body weight (men vs. women, p = 0.008); 51.2% of the men and 58.7% of the women reported a decrease in PA levels (men vs. women, p = 0.10).

Body Weight Changes
Note: Data shown represents numbers and proportions. p values < 0.05 are statistically significant. arepresents median and 5 and 95 percentiles, b -represents proportions. Figure 1 shows the change in body weight and PA patterns. Of the men, 25.6% and of the women, 38.1% reported an increase in body weight (men vs. women, p = 0.008); 51.2% of the men and 58.7% of the women reported a decrease in PA levels (men vs. women, p = 0.10).

Discussion
The present study aimed to determine lifestyle changes, such as food habits and PA patterns (i.e., type of PA, time, duration, sedentary time), in the Chilean population during COVID-19 confinement and to analyze its association with changes in body weight and physical status. The main results of the present study were as follows: (a) low water consumption was associated with body weight increase; (b) active breaks may be protective factors for body weight increase; (c) daily alcohol consumption was associated with PA decrease; and (d) sedentary time ≥ 6 h/day presented an association with negative changes in body weight.
Negative eating habits, such as low consumption of legumes and water and high consumption of junk food (i.e., food with low food quality, low contribution of micronutrients and with a high contribution of sugar, saturated fat, and sodium) and fried foods, were associated with negative changes in body weight. In the case of legumes, the high content of dietary fiber, low energy density, high protein intake, and low glycemic index make them a food with high nutritional quality that enables the control of body weight and the prevention of metabolic disease [15,16]. The DGCP [14] recommended the consumption of legumes to be at least two times per week. However, recommendations of dietary guideline for the Spanish [8] and North American [17] populations suggest increasing the minimum consumption of legumes to three times per week in different types of presentation (i.e., salad, stews, sauces, soup, cream of legume's, etc.) In this study, those evaluated had medium to low compliance with the general recommendation (83.7% indicated consuming 1-2 times a week) similar to that reported in another study in the North American population [18]. In contrast to our results, a recent study conducted in Spanish adults reported that the subjects increased consumption of foods, such as olive oil, vegetables, fruits, or legumes, during confinement. Moreover, this study reported a higher Mediterranean diet adherence (MDA) that could have a positive impact on the prevention of COVID-19-related complications [19].
In times of longer homestays and quarantine plans, changes occur in the daily routine and boredom is increased. This has been associated with a greater desire to consume pleasant foods to cope with the stress produced by confinement [20]. Junk food has a high proportion of refined sugars and saturated fats (in addition to frying), making consumption a risk factor for obesity and causing an increase in the pro-inflammatory state [21,22]. The DGCP recommended avoiding fried foods and foods with saturated fats, and to consume sugar sporadically and in small amounts. In this study, the consumption of junk food three times per week was associated with an increase in body weight and the study sample were not complying with DGCP recommendations [14]. Additionally, a recent study reported that food consumption and meal patterns were unhealthier during COVID-19 confinement [23]. Along this line, Bhutani et al. reported an increase in unhealthy foods and snacks during the COVID-19 home confinement [24].
Another important element that stands out as a protective factor in the increase of fat mass and favors the regulatory functions of hunger and satiety is the daily consumption of water [25]. In this case, the study sample presented a frequency of consumption of 3 to 4 glasses of water per day, with men reporting higher consumption than women (i.e., 4 glasses and 3 glasses, respectively), similar to that reported in other studies [26,27]. Water consumption of the sample remained under the recommendation of DGCP (recommendation of 6-8 glasses per day), which is highly related to a higher caloric intake [28].
For the control of cardiovascular diseases and body weight, the MDA has been one of the most accepted worldwide because it is characterized by a high concentration of antioxidants from legumes and vegetables and it contributes significantly with essential fatty acids from nuts and fish [29]. This report presented an inverse association between fish consumption and increased body weight, such as indicated by Jain et al. [30] However, this group had lower consumption in relation to the DGCP recommendations and the American Heart Association [31] (2 times a week) and compared to the Spanish population (2-3 times a week).
Active breaks and a different kind of PA are protective factors for body weight increase during COVID-19 confinement. Jakobsson et al. indicated that maintaining regular PA during self-isolation was important for prevention [32]. Moreover, the authors recommended interrupting sitting time with active breaks during the day. Conversely, lower PA levels increased the risk of gaining weight by reducing energy expenditure [33]. High levels of obesity increase the risk of infection and mortality in viral diseases, so it is advisable to avoid weight gain by including regular exercise [34]. A recent study of the Australian population indicated that 43.4% of the population (n = 5469) exercised less during the COVID-19 pandemic and found a relationship between binge eating and exercise [35]. Quarantine affected body weight increase (around 2.2-4.4 kg), indicating a decrease in PA to be one of the main risk factors [36]. Moreover, the perception of body weight increasing during confinement has been observed in 48.6% of the Italian population [37]. Therefore, it is essential to incorporate activities during the pandemic such as walking around the house, stair climbing, sitting and standing on a chair, raising PA levels, and increasing energy expenditure, thus avoiding body weight increase and mental health problems [38,39].
The findings of our study indicate that a sedentary time ≥6 h/day had a positive association with body weight increase; these results are opposite to those reported by Zachari et al. who found no relationship between sedentary time and body weight gain in times of quarantine [36]. Despite this, extended home quarantine may help generate body weight gain in adults [5]. Similarly, another study reported that COVID-19 home confinement had a negative effect on all PA intensity levels; moreover, the daily sitting time increased [23].
We found that a middle socioeconomic background was associated with body weight increase. Similarly, healthy lifestyle practices such as more favorable modifications of nutritional behaviors and PA levels during COVID-19 confinement are associated with higher incomes in the French population [13]. This could be related to greater financial resources and employment flexibility facilitating healthy choices. Conversely, Australian adults in the lowest income category had significantly higher mental health problems during COVID-19 compared to higher income adults [40]. Lifestyle (i.e., foods habits, PA) and mental health may affect body weight increase during COVID-19 confinement [13,19,23,40].
In the present study, 30% of the sample study reported daily alcohol consumption during confinement; in addition, this consumption was associated with negative changes in lifestyle, such as a decrease in PA. A negative change in alcohol intake was more likely to cause more depression, anxiety, and stress during confinement in the Australian population [40]. Evidence has shown that alcohol drinkers are less able to find anything positive about the pandemic situation and were mentally less able to cope [41]. Moreover, Jurak et al. suggested that being physically active is a simple and effective way of addressing the adverse effects of COVID-19 [42]. In contrast to our results, there is strong evidence for the existence of a positive association between alcohol consumption and PA [43][44][45]. Additionally, another study reported that moderate drinkers and heavy drinkers were more likely than abstainers to have physically active lifestyles in data representative of the U.S. population [46]. Piazza et al. reported that alcohol consumers were more physically active than non-drinking peers; moreover, the authors concluded that these findings were contrary to the hypothesis of the investigators [47]. Despite this, it has been demonstrated that excessive alcohol consumption is associated with an increased risk of mortality [48,49].
In this study, separated and married marital status was associated with body weight increase. One longitudinal study reported that changes in marital status, such as entering or leaving a marriage, influenced body weight; besides, women who were unmarried at baseline and married at follow-up had greater increase in body weight than women who were married at both times [50]. Moreover, Umberson et al. concluded that marital transitions are more important than marital status in predicting changes in body weight [51]. Future studies into COVID-19-related changes in behavior related to COVID-19 mitigation could ask about relationship history.

Strengths and Limitations
The main limitation of this study is its cross-sectional design; these factors should also be measured in a longitudinal study in the future to clarify the direction of the associations. Another limitation would be that the body weight and PA level were self-reported, which could mean that these data are underestimated or overestimated. Likewise, the results presented could have been regardless of the COVID-19 confinement. The main strength was that people from all over the Chilean national territory participated in this study and the study provides novel results applicable to confinement times.

Contributions of this Study
The present study provides different and relevant aspects that must be considered in times of COVID-19 confinement. Among them, water consumption and the development of active breaks can be recommended since they may be protective factors for body weight increase and they are simple factors to support and easy to apply for the population.

Conclusions
A healthy lifestyle that includes good food habits, PA, and active breaks are particularly important since they may be protective factors for body weight increase during COVID-19 confinement. Moreover, the evidence suggests that factors such as increasing water consumption and the performance of active breaks can be developed at the home as an easy way to avoid a body weight increase during COVID-19 confinement.