The Impact of the COVID-19 Pandemic Outbreak on Eating and Lifestyle Habits of Adolescents in Bosnia and Herzegovina: a Cross-sectional Study

Background: Pandemic caused by the COVID-19 virus brought tremendous changes in the lifestyle of adolescents, about which numerous studies have been published. Due to extended restrictions, long term impact should be investigated. Methods: This cross-sectional study enrolled 953 voluntary participants aged 14 to 21, from different regions of B&H. Participants were asked to complete an online 37-item survey, regarding socio-demographics, geographic, social characteristics, dietary and lifestyle habits, physical activity, including their consumption of dietary, vitamin and mineral supplements. They also needed to provide information about their weight, height and weight change during the entire period of the COVID-19 pandemic from March 2020 until the end of November when the study was completed. For data entry and analysis, SPSS (version 25) and Microsoft Excel were used. Results: At the time of completing the survey, a slightly lower BMI of participants was noticed compared to the time before the pandemic. A statistically signicant difference was determined between males and females BMI, boys BMI was slightly higher. Females gained and men lost BW. Increased intake of unhealthy types of food was associated with weight change. Increased mental stress during the pandemic was associated with dietary changes, respectively with decreased as also increased food amount. High percent of participants (40.4%) who increased their physical activity did not alter their eating in the form of the food amount. This study reported use of some dietary supplements which have not been used before the pandemic by 63.5% participants. Conclusions: This paper presents a unique insight into the changing lifestyle and eating habits of adolescents in B&H during lockdown and post-lockdown period of research. Considering that pandemic is still ongoing, data from study like this may be useful to create further steps in battling the pandemic.

the most socially vulnerable children living in small houses, without possibility of going out, or Internet access and equipment that would give them the opportunity for education [6].
In B&H (Bosnia and Herzegovina) the lockdown was decreed on 16th March 2020, which took children and adolescents out of schools and playgrounds for almost 12 weeks. Closing the schools was a good idea in the sense of stopping the spread of the virus, but it brought a lot of challenges, not only for students and teachers [7]. The interruption of work and school routine caused by the government's measures resulted in boredom, which in turn is associated with a greater energy intake [8], sedentary behavior, modi cation in smoking and alcohol drinking. Sedentary behavior and high mental demands have been associated with increased food intake, suggesting that this may lead to a positive energy balance and weight gain [9,10].
This survey was carried out to explore and analyze lifestyle, eating habits and physical activity of B&H adolescents at the time of lockdown and after getting back to the school, related to mental and emotional mood caused by measures during the COVID-19. The measures have had a huge impact on eating habits, mental and emotional mood and physical activity, not only during these weeks, but also after the abolition of certain measures and return to school. There are already numerous studies on the impact of lockdown on eating habits and behavior, but in the period from March to May 2020.

Methods
The aim of the study As the COVID-19 is a burning topic, the aim of this work is to nd out how it in uences the eating habits and physical activity among adolescents, and compare them to their usual diet. In addition, the use of dietary supplements was investigated. The question about mental and emotional health burden arising from the COVID-19 and related measures, was also investigated. Measures affected the adolescents in their previous habits in diet and physical activity. Although the negative effects of school closures on the lives of children and young people are daily, globally documented, a thorough assessment of the longterm consequences of the pandemic is yet to come.

Participants
The target population for this research was students of different high schools (grammar school, and different vocational high schools) and regions of B&H, which contributed to the representativeness of the sample. A total of 953 students between 14-21 years old (from 1st to 4th grade of high school) were evaluated through an online survey. Respondents with missing or implausible data were excluded from analyses, which led to a study sample of n = 945.

Survey development and content
The data were obtained through an online survey composed of 37 questions, covering sociodemographic, geographic, and social characteristics as well as eating and lifestyle habits. The survey is Page 4/26 divided into ve sections that are presented in Table 1 (Appendix). • Yes

• No
The rst section comprised nine questions on demographic, geographic and social characteristics of the students. The second section consisted of 4 questions focused on weight change during COVID-19. BMI (body mass index) was calculated as the body mass divided by the square of the body height and categorized based on the distribution of the data (< 18 kg/m2; 18-25 kg/m2; >25 kg/m2; >30kg/m2). The third section included 6 questions related to eating habits before and in the time of the pandemic while the fourth section was focused on dietary and herbal supplements. The end of the survey comprised questions about increased mental stress, physical activity, tobacco and alcohol use before and in the time of COVID-19.
The survey was prepared using Google Forms survey administration software, and distributed via accessible mailing list through school administration and staff. Teachers and parents were informed of the procedures and objectives of the study. The research was anonymous and was done exclusively according to the wishes and consent of the students themselves. The study was active for 4 weeks in November 2020. Due to physical restrictions, the online survey was one of very few possibilities to reach as many respondents as possible.

Statistical analysis
Statistical analysis and graphic design were done by SPSS version 25 for Windows (SPSS inc., Chicago, USA) and Microsoft Excel (Microsoft Corporation). Absolute and relative frequencies were calculated for data description. Normality of distribution was tested using Shapiro-Wilk tests. According to these tests, distribution of BMI before and during the COVID-19 pandemic variables as well as BW change variable (in kg) did not show a normal distribution (p < 0.05). But after assessment of the magnitude of skewness and kurtosis of data distribution with graphical methods it is observed that any kurtosis coe cient is greater than 10, and any skewness coe cient is greater than 3 [11]. According to this and due to the fact that the sample size is large and all extreme values as incorrect were excluded, parametric tests were used. Categorical variables (gender, age, BMI) were compared using Chi-squared test. Non categorical variables were compared using t Test and ANOVA. To compare the means of two independent groups (BMI, gender, employment of mother or father, place of residence, physical activity), in order to determine whether there is statistical evidence that the associated sample means are signi cantly different, independent samples t Test was used. ANOVA was performed to compare continuous variables (BMI, weight change) among groups. One-way ANOVA was performed to compare the difference between groups ("decreased", "unchanged", "increased", "never consumed") on weight change (in kg). Since the homogeneity of variances was violated, assessed by Levene's Test of the homogeneity of variance (p < 0.05), Welch ANOVA was used and Games-Howell post hoc test. Data are mean ± standard deviation, unless otherwise stated. Statistical signi cance was determined at p < 0.05. Since multiple response questions were present in the survey, frequencies and percentages for these questions were taken directly from Google Forms survey. week and usually more than 30 minutes. Outside of school, 84.7% (n = 260) of participants exercise more than 3 times per week and more than 30 minutes per activity. This means for most of the participants walk house -school -house, cycling, hiking, running and training in a sport club (soccer, basketball, handball) or some other activities or their combinations. It is interesting that larger percent of participants 33.0% (n = 312) had increased their physical activity during the pandemic in comparison with 27.5% (n = 260) who decreased. There was no statistical signi cance between physical activity modi cation and BMI category, nor between weight change (in kg) and physical activity modi cation or its duration.

Results
Most of the participants (46.3% n = 438) felt more stressed than before the pandemic. Statistically signi cant difference (p = 0.000) was obtained between genders, where 53.0% (n = 366) girls felt more stressed than before pandemic in comparison to 28.3% (n = 72) of males, as well as adolescents from urban areas (49.7%, n = 281) compared to adolescents from rural (41.3%, n = 157) (p = 0.039). 4.6% of students (n = 43) started to drink alcohol during the pandemic and those who started smoking during this time was 3.8% (n = 36). But, almost a similar number of students changed their lifestyle habits for the better in the form of quitting smoking and alcohol consumption (3.1%, n = 29; 3.9%, n = 37 respectively).
Regarding the use of their phone or watch TV while eating, 86.4% of participants do it regularly or occasionally, and 15% of all participants stated that this increased the amount of consumed food.
The presence of one of the chronic non-communicable diseases (NCD) in the immediate family of 24.1% (n = 228) participants was established in this study. This was important since proper lifestyle and dietary habits can prevent many NCD [13].

Eating habits
Most of the participants (41.8%, n = 395) stated they did not alter their food consumption during the COVID-19 pandemic, while 25.9% (n = 245) of the participants increased their food intake, and 14.2% (n = 134) decreased, compared to the time before. Other participants were not able to conclude if their food intake had changed during speci ed time. Statistical analyses showed signi cant correlation between gender, increased mental stress, physical activity and change in food amount ( Table 2). We found that male participants did not alter their food amount as much as girls (decreased: 16.5%, n = 114; increased: 20.4%, n = 141). Increased mental stress during the pandemic was associated with decreased (18.9%, n = 83) as well as increased (31.3%, n = 137) food amount. Signi cantly high percent of participants (40.4% n = 126) who increased their physical activity did not alter their eating in the form of the food amount.

BMI -body mass index
The present study also examined the frequency at which particular food groups were consumed at the time of completing the survey (Table 3). Interesting results were obtained in change of food categories.
The vegetable and fruit consumption increased by 15.9% (n = 150) and 27.9% (n = 264) respectively, which showed increased fruit and vegetable intake of adolescents in a signi cant way during con nement. Respectively, our results also showed an increase in consumption of nuts (walnuts, hazelnuts, peanuts, almonds) by 16.2% (n = 153). In fact more than half of the participants reported an unchanged amount in these categories (Table 3). Soft and soda drinks consumption as also confections (chocolates, biscuits, wa es, ice creams, candies, etc.) were decreased by 26.9% (n = 254) and 19% (n = 180) participants, respectively. Snack food consumption (chips, crackers, ips) was also decreased by 24.1% (n = 228), and fast food like pizza, hotdog, burgers, traditional pie i.e. "pita" by 25.3% (n = 239) participants. BMI groups had no relevant effect on decrease or increase in food categories. Using ANOVA, a statistically signi cant effect in the change of BW with regard to the change to the consumption of confections, soft and soda drinks, snack food and fast food was determined (Table 4).

Use of dietary supplements
It is an interesting fact that 63.5% (n = 600) of respondents did not use supplements before the COVID-19 pandemic. Although most of the participants 88.5% (n = 836) did not have any de cit diagnosed by specialists, 29.4% (n = 278) they used supplements based on their own or their parents' initiative.
According to specialists recommendation, 10.5 %, n = 99 of participants used some supplement (Fig. 3). 7.1% (n = 67) of them have been diagnosed with iron de ciency anemia and 3.1% (n = 29) de cit of vitamin D. Other de ciencies include vitamin B de ciency and some other conditions that require vitamin and mineral supplementation. The most common nutrients supplemented during the period of pandemic were vitamin C (36.6%), magnesium (28.9%), vitamin D (22.9%), calcium (10.8%), iron supplements (11.6%), omega-3 fatty acids (5.9%). These nutrients were followed by probiotics, vitamin B supplements, zinc and selenium. Additionally, adolescents (30.3%, n = 286) are also prone to use different natural bioactive compounds from herbs, tinctures, and oils, and as the main reason they state improving immunity (17.1%). Other reasons they state are weight loss, increasing stool motility and other -non speci ed. Herbal products generally included herbal teas (Matricaria chamomilla, Thymus serpyllum, Mentha piperita, cranberry and ginger tea and tea of Artemisia annua), black seed oil (Nigella sativa) and bee products (propolis, royal jelly, pollen).

Discussion
This cross-sectional study investigated the impact of changes caused by the COVID-19 pandemic on lifestyle and eating habits from the beginning of government measures in March 2020 to the end of the research in November 2020. We hypothesized that measures taken by the government to slow the transmission of COVID-19, affected the change in food consumption, food groups, changes in BW, physical activity and stress. As some researchers con rmed that different natural disasters can affect, not only lifestyle, but physical and mental health in the children and adolescents [14] it was assumed that measures caused by the pandemic COVID-19 may have similar effects. Research made by Medrano et al. [15] show that measures that were taken in Spain, negatively affected physical activity and sedentary behaviors among children.
Government measures in B&H included changes in the education system, the closure of sport clubs, tness centers, caffes, bars, restaurants. All primary and secondary schools have switched to online teaching from March to June 2020, and for the current school year, from September 2020 to the combined type of teaching (online and contact). All of this affected the student's previous habits in diet and physical activity. They spent more time at home and after they enrolled back in the education process in September, classes lasted about three hours, a half time than usual classes. They did not have physical activity at school due to avoiding physical contact. Also it is important to mention that from September 2020, B&H had less restrictive measures in comparison to Western European countries (for example Italy, Germany, France), which means open bars, restaurants, fast foods, cafes as also sports facilities.
Similar to the research of Huber et al. [16] and Di Renzo et al. [17], in this study the female respondents represent 73.1% of the population. Almost one third (28.4%) of our study participants reported a weight gain, and females gained statistically signi cantly more on BW compared to males. Research by Di Renzo et al [17] has also detected positive association between perception of weight gain to the female gender. But ultimately, in the present study, BMI before the COVID-19 was slightly higher than BMI of participants at the time of completing. The reason for this could be found in the large number (40.4%) of participants who increased their physical activity but unchanged intake of food. Most of participants (41.8%) stated they did not alter their food consumption during the COVID-19 pandemic and that correlates with the current literature [16,17]. Also, percent of participants who increased their food consumption is similar to these studies. The vegetable and fruit consumption increased in accordance with results of Ruiz-Roso et al. [18], as well with the results of Di Renzo et al [17]. Results from our study also showed increased consumption of nuts, which is a positive effect as their consumption is associated with less long-term weight gain and a lower risk of obesity in adults [19]. A review by Jayawardena et al. [20] underlines that a balanced nutrition, which can help in maintaining immunity, is essential for prevention and management of viral infections. Overall, increased consumption of vegetables and fruits and decreased consumption of sweets, unhealthy snacks, fast food, sugary and soda drinks was an unexpected positive outcome. Survey by Di Renzo et al. [17] also showed decreased consumption of packaging sweets and baked products as well as delivery food, during con nement. A decrease in snack food consumption (chips, crackers, ips) and fast food like pizza, hot-dog, burgers, pies is in accordance with results of Ruiz-Roso et al. [18] where adolescents from 5 countries decreased their weekly consumption of fast food. In the present study a statistically signi cant effect in the change of BW with regard to the change to the consumption of confections, soft and soda drinks, snack food and fast food was determined ( Table 4).
Study of Pišot et al. [21] also reported healthier meals and less unhealthy food. The reason for these changes can be justi ed by the obstacles that prevent students from choosing a healthy meal at school.
That includes time management for eating quickly, unhealthy food environment around the school and unpleasant food service at school [22] like absence of canteen in Bosnia and Herzegovinian schools. In absence of school canteen, young people are forced to take breakfast or lunch in fast food objects and bakeries around the school, which are usually nutrient-poor, energy-dense food. Moreover, it should be stressed that the nutritional quality of the eaten breakfast is likely to be of higher importance than only eating breakfast on a daily basis [23] and in the present study the rst three places are engaged by dairy products, bakery products, meat products and eggs, respectively. In comparison to research by Sidor et al. [24], in our study it was not reported that speci c BMI groups statistically decrease or increase food amounts which correlate with Italian study [17]. It seems that in B&H, restrictions in schools led to increased physical activity outdoors which may be due to additional free time. That is different from work of Moore et al. [25] who found that children and youth were less active, played outside less, were more sedentary, and engaged in recreational screen-based activities during the initial COVID-19 virus outbreak compared with before the restrictions.
The lack of treatment options for the COVID-19 has led to many attempts to nd alternative options to decrease the probability of getting infected, prevent the transmission of the disease or to alleviate the progression of the infection by the use of herbal products or nutritional supplements [26]. It is an interesting fact that 63.5% (n = 600) of respondents did not use supplements before COVID-19 pandemic, even it was known that use of dietary supplements is often justi ed by the desire to improve immunity, overall health, to prevent illness, or to treat viral infections [20,27]. Only 10.5% of participants used some supplement according to specialist's recommendations which is quite worrying. The most popular and commonly used supplements during the March-November period were vitamin C, magnesium, vitamin D, calcium, iron supplements and omega-3 fatty acids, followed by probiotics, zinc, selenium and vitamin B supplements, which is in accordance with results of PLifeCOVID-19 Online Studies [28]. Other supplements that overlap with the above study include herbal products like ginger tea, black seed oil (Nigella sativa) and bee products. Effective education of consumers in its rational use and healthprotecting behaviors against COVID-19 should be developed and introduced at local and/or national levels.
Situation during 2020 has created a sense of uncertainty, stress and anxiety that might lead to unfavorable outcomes regarding the psychological health of students and indeed, they have been identi ed as a vulnerable group [29,30]. This is con rmed by numerous studies [30,31,32]. Although study by Cao et al. [31] reported living in urban areas as a protective factor against anxiety, our study has shown that more adolescents from the city area felt more stressed compared to the period before the Covid-19, as well as female gender. It has been indicated that compared to adults, this pandemic may have long-term adverse consequences on children and adolescents [29].
Strengths, limitations, and future directions To our knowledge, this is the rst study conducted in B&H on eating habits, physical activity and the use of supplements in adolescents, in the entire period from the outbreak of the pandemic to the conclusion of the study in November. Most of the studies do not take into account BMI and weight change in currently published research, and this is one of the strengths of this research. The limitation of the study was that respondents were mostly females. Also, the presence of the COVID-19 infection in students was not reported. More clear and convincing studies need to clarify the role of dietary supplementation in the COVID-19 prevention and the existing risk of elevated intake of some nutrients due to increased use of dietary supplements. As the measures taken by the government affected physical and eating habits in adolescents, in the future more attention should be paid to measures in the same or similar situations.
Also as the COVID-19 pandemic is ongoing, our data need to be more investigated in future with extensive population studies. These ndings should be taken into consideration for future regulations incense disease is now again spreading in B&H with possible subsequent lockdowns and social distancing measures being enforced by the government.

Conclusion
Quarantine is an essential measure to protect public health and control the transmission of the virus, but this study shows the other side of the lockdown. The results of the study indicate that increased consumption of unhealthy types of food (sweets, sugary and soda drinks, snacks and fast food) affected weight gain in adolescents. Although BMI before the COVID-19 pandemic was slightly higher, this indicates that more attention should be paid to promote healthy eating practices, as this may prevent further weight gain. Positive outcome was recorded in consumption of vegetables, fruits, and nuts (by 15.9%, 27.9% and 16.2% respectively) and among 33.0% of participants who increased their physical activity. Since most participants feel more stressed than before the pandemic, governments should focus more on the psychological health of young adults during the lockdown and whole period of pandemic.  Supplement use between males and females