Dietary Patterns and Health Trends Among Students in Portugal: A Cross-Sectional Analysis

1. Introduction

The dietary and lifestyle habits of students play a fundamental role in their overall health and development [1]. There has been a marked decrease in physical activity levels among children and adolescents, partly due to increased screen time and sedentary behaviours [2,3]. Lifestyle changes among students need to be understood, in order to promote early interventions to prevent obesity and reduce the risk of having chronic diseases later in life [4].

Obesity and overweight are among the most significant public health challenges facing the modern world [5,6]. Defined by the World Health Organization (WHO) as an abnormal or excessive fat accumulation that may impair health, obesity is linked to numerous non-communicable diseases (NCDs), including type 2 diabetes, cardiovascular diseases, hypertension, and certain types of cancer [5,7]. The global burden of NCDs linked to obesity continues to increase, with recent forecasts predicting that, if trends persist, over 1.31 billion people worldwide will have diabetes by 2050 [8,9]. Type 2 diabetes has been increasing among children and adolescents and can lead to severe complications such as atherosclerotic cardiovascular disease, stroke, myocardial infarction, renal insufficiency, neuropathy, vasculopathy, or retinopathy leading to blindness [10,11]. Cardiovascular diseases, including heart disease and stroke, are also more prevalent among obese individuals, contributing to higher mortality rates [12]. The incidence of cardiovascular events is expected to more than double in the next decade in some countries [13,14]. The number of obesity-related cancer cases is projected to exceed 2 million globally by 2070 [15]. Obesity is also linked to various musculoskeletal disorders, including osteoarthritis, which can cause chronic pain and disability [16]. Additionally, obesity can lead to psychological problems, such as depression, anxiety, and low self-esteem, particularly among children and adolescents [17].

The commonness of obesity has reached alarming levels globally. According to the WHO, the prevalence of obesity has more than doubled since 1990. In 2022, more than 2.5 billion adults were overweight, and of these, over 890 million were obese. Similarly, in the same year, 37 million children under the age of five were overweight or obese [5]. In 2019, in Portugal, 17% of adults was classified as obese, a proportion slightly above the EU average of 16% [18]. According to the Childhood Obesity Surveillance Initiative (COSI) conducted in Portugal, although there has been a reversing trend in the prevalence of childhood obesity and overweight (-8.3% between 2008 and 2019), one in every three children is still overweight, and 10.6% are obese [19]. These trends indicate a public health crisis that requires urgent and comprehensive intervention.

Body Mass Index (BMI) is a widely recognized tool used to classify individuals' weight categories and assess potential health risks associated with body weight. It is a simple measure derived from a person’s weight relative to their height and it is typically used to categorize weight status, identifying if someone is underweight, normal weight, overweight, or obese [20]. Overall, BMI is an essential tool in public health for monitoring population weight trends and it serves as a valuable indicator for healthcare professionals to initiate discussions about nutrition, physical activity, and lifestyle choices that contribute to overall well-being [21].

Considering this evidence, our study focuses on the nutritional and lifestyle habits of students in the city of Guarda, Portugal, and estimates the prevalence of overweight and obesity among students. In addition, our research aims to identify contributing factors to obesity and propose effective strategies to combat this growing health concern. The findings will provide insights into the current state of student health in Guarda and allow to contemplate future public health initiatives aimed at reducing obesity rates and improving overall well-being.

2. Materials and Methods

To assess the nutrition and associated behaviours among students from various educational levels (1^st to 12^th grade and higher education), a cross-sectional study (Figure 1) was conducted from January to July of 2024. The study focused on two groups of schools, namely, (1) Agrupamento de Escolas Afonso de Albuquerque and (2) Agrupamento de Escolas da Sé, and one higher education institution, (3) Instituto Politécnico da Guarda, located in Guarda, Portugal. Initially, a comprehensive survey was developed [22], which included questions related to demographics, physical activity, sleep, sedentarism, dietary habits, substance consumption and health status (Supplementary Figure S1). Ethical approval was obtained from the Ethical Commission of the XXX (Document no. 8/2024) on 21 May 2024.

A total of 3,599 surveys were distributed to the entire target population, the students in the two school groups, provided in both paper or online formats and with a prior request for parental permission to participate. Out of the 3,599 distributed surveys, 1,950 responses were received. After a meticulous validation process, 1,762 responses were deemed valid, resulting in a response rate of 54%, since the remaining participants did not have authorization from the parents to participate or did not complete the survey in its entirety. Regarding the higher education, a survey link was shared with 2,483 students. From this cohort, 321 students completed the survey, with a response rate of 13%. All responses to the survey were collected anonymously and kept confidential to ensure privacy of the participants and encourage honest and accurate reporting.

The collected data was separated in the following age groups: children and adolescents (from 6 to 19 years old) and adults (over 19 years old) [5]. The study was stratified in age intervals (5-12 years, 13-19 years, 20-39 years, 40-59 years). We defined overweight and obesity using the Body Mass Index (BMI) for each participant, which was calculated based on the reported height and weight, using the following formula:

$$\mathit{BMI}={\displaystyle \frac{\mathit{weight}\left(\mathit{kg}\right)}{{\mathit{height}\left(m\right)}^2}}$$

[23]

BMI categories for adults were defined as: underweight (<18.5), normal weight (18.5–24.9), overweight (25–29.9), and obese (≥30, with further classes for obesity) (Table 1). For children and adolescents, BMI is age- and sex-specific, following WHO guidelines that use percentiles: healthy weight (15^th–85^th percentile), overweight (85^th–97^th percentile), obese (>97^th percentile), and thinness (<15^th percentile) [24,25]. This approach accounts for varying growth patterns and maturity rates.

As for the frequency of physical activity practice, it was classified as low if the individual exercises less than twice per week, as moderate if three to four times per week and heavy if more than five times per week, since the WHO recommends practising vigorous-intensity physical activity at least 3 days per week [27].

Statistical Analysis

Data collected from the surveys was organized and analysed using IBM SPSS Statistics version 29.0.0.0 for Windows. An exploratory analysis of the data was performed using frequency distributions, graphical representations, box plots and descriptive measures (mean, median, quartiles, standard deviation, minimum, maximum). The statistical analysis included the Fisher’s exact test, chi-square tests and Independent Samples t-test to examine associations between BMI, demographic factors, dietary habits and physical activity. When appropriate, the Relative Risk (RR) was used to quantify the probability of an event occurring in the exposed group compared to its probability in the non-exposed group. Mann-Whitney U test and the Spearman's Coefficient (SC) were applied for non-parametric and correlation analyses, respectively. Lastly, binary logistic regression was used to identify factors associated with BMI categories. The 95% Confidence Intervals (CI) was carried out in situations where it was considered relevant to the study. In the statistical analysis of the results, a significance level of 5% was considered.

3. Results and discussion

3.1. Demographic Data

Regarding children and adolescents, the results included 1545 individuals, of which 43.4% are male and 56.6% are female. This group includes 63.6% of children aged from 6 to 12 years old and 36.4% adolescents from 13 to 19 years old. 41.4% of answers are from the first cycle of education (1^st to 4^th grade), 17.4% belong to the second cycle (5^th to 6^th grade) and 18.1% to the third cycle (7^th to 9^th grade). 16.3% of the students are currently in high school (10^th to 12^th grade) and a percentage of 6.7% is enrolled in higher education.

We obtained data from a group of 217 adult students, from a higher education institution, that included 35% males and 65% females. From the participants, 93.5% belong to the age interval of 20 to 39 and 6.5% are 40 to 59 years old.

3.2. Prevalence of Overweight and Obesity Among Students

After the calculation of the BMI, we found that the prevalence of overweight among the children and adolescents is 17%, while 9.1% are obese (Figure 2a). A high percentage of students has a normal weight (70.1%). In Portugal, according to the COSI study, in 2021/2022, the prevalence of excess weight for children was 31.9%, from which 13.5% were obese [19]. Data from the COSI study in 33 countries from the European Region reveals a 29% prevalence of obesity in children aged 7 to 9 years old [28].

According to the collected data, 17.5% adults are overweight, while 8.3% suffer from class I obesity, 0.9% are class II obese and 0.5% have morbid obesity. It was observed that 68.7% of adult students fell within the normal weight range (Figure 2b). A study with European college students from the Western Balkans describes a prevalence of 12% of overweight and 2.3% obesity [29]. According to the WHO European obesity report from 2022, obesity affects almost 60% of adults [30].

3.3. Weight Status According To Sociodemographic Characteristics And Life Habits Of The Students

A dependency was found between sex and the weight classification (P < 0.001), according to Fisher’s exact test. Following this information, Table 2 represents the prevalence of some characteristics and life habits in females and males aged from 6 to 19 years old, in relation to their weight status, according to the BMI. In male children and adolescents, variations in BMI classification are observed across different age groups, showing a statistically significant relationship between BMI and age. Boys aged 6 to 12 show a higher prevalence of overweight and obesity, at 33.4%, compared to boys aged 13 to 19, who show a prevalence of 25.8% (P = 0.05). A similar trend is seen among female children, where 27.4% of girls aged 6 to 12 are overweight or obese, whereas this prevalence decreases to 15.0% in girls aged 13 to 19 (P < 0.001). Furthermore, an Independent Samples t Test confirmed that overweight/obese children tend to be younger in both male (P < 0.001; t(472.223) = 2.732) and female students (P < 0.001; t(376.01) = 5.171). Accordingly, data from the WHO in Europe 2022 reveals that one in three school-aged children are overweight or obese, while in 10-19 years the prevalence decreases to one in four [30]. These findings suggest that younger children are more susceptible to higher BMI, which is worrying, potentially due to variations in physical activity levels [31]. Younger children may not be as involved in structured physical activity as adolescents, who possibly participate in school sports and extracurricular activities more often. In fact, we found a statistically significant positive and weak relationship between age and frequency of physical activity for both sexes (female: SC=0.095, P = 0.005; male: SC=0.340, P < 0.001), this indicates that adolescents practice more exercise than younger children.

The frequency of weekly physical activity also shows a correlation with BMI classification (male: P = 0.031; female: P = 0.002). The Mann-Whitney Test confirmed that males with low physical activity have a higher prevalence of overweight/obesity (34.3%) compared to those with moderate (32.3%) and heavy physical activity (22.4%) (U = 42 954.5; P = 0.024). Similarly, females with low physical activity had a 26.3% prevalence of overweight/obesity, while those with moderate and heavy physical activity had 19.3% and 11.9% respectively (U = 57 423.5; P = 0.002). This suggests that higher physical activity levels are associated with lower rates of overweight and obesity in both sexes, which is in accordance with previous studies. A systematic review of longitudinal studies found that individuals with higher physical activity levels had a significantly reduced risk of developing obesity compared to those with lower activity levels, with a lower risk of developing coronary heart disease or diabetes [32]. The Rotterdam Study, a large-scale study involving middle-aged and elderly adults, found that higher levels of physical activity significantly reduce the risk of cardiovascular disease associated with overweight and obesity [33]. The results suggest that female children that are overweight/obese spend more time sitting down during the week, (SC= 0.067, P = 0.049), which is consistent with what was mentioned above, since spending more time sitting down logically implies less physical activity.

However, among the individuals who have a low frequency of physical activity, a substantial proportion maintains a healthy weight. We need to consider the interaction with other factors, such as genetic predisposition or metabolic rates.

Regarding disease incidence, there is an association between the weight status and having been diagnosed with diseases in males (P = 0.035). According to our following tests, the incidence of health conditions among overweight or obese boys is 1.27 times higher than the incidence of health conditions among boys with a normal weight (RR = 1.27; 95% CI: 1.0 – 1.6). A relationship between obesity and health complications has been intensively described before. Studies show that obese children are more likely to be diagnosed with diseases such as mental disorders, gastrointestinal disorders, metabolic syndrome, insulin resistance and non-alcoholic fatty liver disease [34,35,36]. Another study showed that children who are morbidly obese have even higher prevalence of diabetes/prediabetes and use more medications for asthma than obese children [37].

Lastly, we found an association between the consumption of food supplements and the weight status in male children (P = 0.036). Further tests determined that the incidence of overweight/obesity among boys who consume supplements is 1.4 times higher than the incidence of overweight/obesity among those who do not consume supplements (RR = 1.4; 95% CI: 1.1 – 1.9). A possible explanation could be the inappropriate usage of these products, coming from the parents’ erroneous assumption that the children need to take supplements to boost their growth, when in fact they don´t need them. Inappropriate usage of supplements may increase a child's calorie consumption without making up for real nutritional needs, which might result in weight gain. However, since the numbers found do not imply a direct cause-and-effect relationship, it is also possible that obese or overweight students are taking supplements to aid in weight loss, which could explain the higher incidence [38].

For other variables, there were no relations with the weight status of the children. Therefore, there was no significant relationship between the sleep and the weight status of children. However, the descriptive data analysis suggests that the people that sleep less tend to be overweight, with 44.4% of the males who sleep less than 6 hours per day being overweight or obese, compared to 30.5% who sleep 6 hours or more. For females, the percentages were, respectively, 25.0% and 22.2%. A study analysed data from the National Health Interview Survey in the US and stated that individuals who sleep less than the recommended 7-8 hours per night have significantly higher odds of being overweight or obese, increasing in the past years. Short sleepers (5-6 hours per night) had a 57% greater risk of obesity, while very short sleepers (less than 5 hours per night) were twice as likely to be obese compared to those who get adequate sleep [39]. Although no significant relationship was found between sleep duration and the weight, it is important to consider that we did not evaluate directly the sleep quality. Research suggests that the quality of sleep plays a significant role in weight management, disrupting the metabolism and increasing appetite which might lead to weight gain and obesity [40].

The characteristics of adult students were categorized similarly (Table 3). Among males aged 20-39, 30.0% were overweight/obese, compared to 24.1% of females in the same age group. The sample size for the 40-59 years old group was not relevant.

A relationship between smoking and weight was found in women (P = 0.002). The incidence of obese woman is much higher in the smoking group (48.1%) than in the no smokers’ group (18.4%). The relative risk of being obese/overweight in the group of women who smoke is 2.6 times the risk in the group of non-smokers (RR = 2.6; 95% CI: 1.5 – 4.5). Studies suggest that the relationship between smoking and obesity may vary with age and smoking intensity. Younger or heavy smokers tend to show higher rates of obesity, while older or light smokers are more likely to exhibit lower BMIs, often attributed to the appetite-suppressing effects of nicotine [41,42]. However, given the small sample size of the smoking group in our study (only 27 smokers out of 141 women), this estimate lacks precision, and no definitive conclusions can be drawn. Larger studies are needed to clarify these results.

Alcohol consumption was significantly associated with the weight status among adult males, with 46.9% of those who consumed alcohol being overweight/obese compared to 22.7% of non-drinkers (P = 0.047). This association suggests that alcohol consumption may be a significant risk factor for overweight and obesity. However, previous research seems to lack consensus on this relationship. A review about alcohol and obesity stated that moderate alcohol consumption does not appear to be a significant risk factor for obesity, while heavy drinking and changes in alcohol consumption patterns can possibly contribute to weight gain and obesity. However, long-term effects and gender-specific responses are not yet well understood [43]. In another study of the Irish population, the individuals who drank heavily were more likely to be obese (high BMI and large waist circumference (WC)), but binge drinkers were more likely to have a larger WC and frequent alcohol consumers were less likely to be obese, suggesting a complex relationship between alcohol use and body weight [44].

The last variable that showed a significant relationship with the weight status was the consumption of supplements for memory enhancement in males. Among male students who use this kind of supplementation, 61.5% are overweight or obese (P = 0.024). However, even though the difference seems to be large, the small number of participants using memory aid supplements (13 men and 26 women) does not allow us to draw conclusions about this relationship

The findings highlight critical relationships between weight status and several factors, such as age, physical activity, and lifestyle habits. Younger children, especially boys, demonstrate higher rates of overweight and obesity. Physical activity is strongly associated with healthier BMI in both male and female children, reinforcing the need for an active lifestyle. Although correlations with supplement intake, smoking, and alcohol consumption were observed, the limited sample sizes in some categories limit the conclusions.

Other factors should be considered when evaluating a population’s habits and health outcomes. The socioeconomic status, for example, is one of the factors that can influence dietary habits (for example, higher consumption of more affordable energy dense foods) and access to physical activity resources, possibly influencing behaviours and, consequently, the prevalence of obesity and overweight [45]. Furthermore, psychological variables like academic stress (which can lead to depression and anxiety) may have an indirect impact on weight, by influencing certain behaviours like the practise of regular exercise or affecting sleep [46].

It is important to sensitise parents, in an educational way, about the importance of balanced diets and regular physical activity for their children. Schools should implement programs that encourage active play and provide nutritious meals, promoting healthier habits from a young age. Furthermore, community-based programs that provide access to nutritious foods and exercise opportunities could enhance children's health outcomes and support the fight against the rising incidence of childhood obesity.

3.4. Food Habits of the Student Community

We then analysed the frequency of food consumption among children and adults, identifying notable trends. Regarding children (Figure 3a), in the cereal, sugar and sweets category, the preferred foods are white bread, breakfast cereals and cookies, with a median consumption of about 2 to 4 days per week. The consumption of bread shows a wide distribution among children. Whole wheat bread is less consumed than white bread, with a lower distribution, showing that a higher percentage of children does not consume this type of bread. In western diets, it is an often practice to consume bread in breakfast, sandwiches, snacks, and other. However, whole wheat bread should be chosen over white bread for its content in whole grains, which have been associated with lower BMI [47].

Whole grains, being a source of fibres, vitamins, minerals, phenolic compounds and other important components, have been related to health benefits and improvement of insulin sensitivity [48]. Cakes and salted snacks are consumed moderately by most children, with a median of 1 to 3 days per month. Perhaps they are associated with special occasions, and they are considered unhealthy foods, leading to a limited consumption. Salted snacks show a few outliers with more frequent consumption, which could be of concern regarding unhealthy eating habits. Honey has the lower frequency (median at less than once per month), suggesting it might not be such a common ingredient in children’s diets.

Regarding fruit and vegetables, fruits and soups are being consumed regularly, up to 2 times in the same day. Some outliers can be noticed in the fruit category, showing that some children still do not include fruits in their diets. Natural fruit juices are consumed less often by children, which could possibly reflect the growing awareness among parents of the higher sugar content of these fruit products or due to the increased time and effort that the preparation of these juices requires. The World Health Organization recommends a daily intake of at least 400 g (5 portions) of fruit and vegetables, in order to prevent noncommunicable diseases (NCDs), such as diabetes, cancer, heart disease and stroke [26,49].

In the meat section, it is noticeable that the consumption of red and white meat is consistent, given the median of about 2 to 4 times per week, but 50% of the participants eat each one less/more than 2 to 4 times per week. This could mean that the meals include a balance between the two types of meat, however, there is a wider distribution for the red meat, with more children consuming less but also more often. Given the health risks associated with the higher levels of saturated fats in these products, consumption of red meat needs to be moderated, having been linked to a higher risk of cardiovascular disease development [50]. The more consistent consumption pattern of white meats, may reflect an inclination of the parents towards healthier and leaner proteins, recommended by nutritional guidelines [51]. Processed meats have a slightly lower consumption with 50% of participants doing so between one to three days per month, and 1 day per week (median), which is positive, given the known health risks associated with processed meats, such as their high sodium and saturated fat content and the presence of preservatives like nitrates, which have been linked to increased risk of breast, colorectal, colon, rectal and lung cancers [52]. Consuming less than 500 g of red meat per week and no to little processed meat is advised by the World Cancer Research Fund (WCRF) [53]. The UK Scientific Advisory Committee on Nutrition (SACN)’s advised for the consumption of red and processed meat fewer than 70 g per day [54].

While fish seems to be present in the diets of children, it is not consumed as frequently as meat. Clearly, there is a general tendency to prioritize meat consumption over fish. Fatty and lean fish are consumed in a median of once per week, with 50% of participants doing so less than once per week. Both types of fish should be included in the diets of children, particularly, fatty fish, including salmon and mackerel. These fishes are rich in omega-3 fatty acids, essential for brain development and health [55,56]. Shellfish has the lowest consumption frequency, with most children consuming it less than once per month, possibly due to the high costs of this type of food, availability questions or even the distinctive and strong taste that might not be too appealing for younger palates.

Eggs are generally consumed around 2 to 4 days of the week by most children, with a narrow distribution, which is in accordance with European recommendations for moderate egg consumption by children [57].

Milk and yogurt stand out as the most frequently consumed dairy products. Milk has a median consumption of once per day, having a significant role in children’s diets. Daily intake is recommended by European guidelines for a healthy diet [58] as a source of calcium and protein. The consumption of cheeses is far less frequent, and soy alternatives show almost no consumption. Plant-based dairy alternatives are not yet widely adopted by children, most likely being consumed more often in households with specific restrictions.

While sugary sodas and juices are present in children’s diets, their consumption is moderate, with a median of 1 to 3 days per month, which suggests an awareness of their health impacts. Outliers indicate that a small part of the community can be consuming these drinks as often as in a daily dose, which is concerning due to their sugar content. Given the extensively studied relationship between free sugars intake and obesity and dental caries. The WHO recommends a reduction of the consumption of these sugars, to less than 10% of the total energy intake, if possible, under 5% of the total energy intake, for children and adults [59]. The consumption of light juices and sodas is even less frequent, with some outliers that could indicate the presence of some concerns and awareness about sugar intake.

The total distribution of tea and infusions consumption, which ranges from once a month to every day, suggests that they occasionally drink them, maybe as a result of their health benefits. Coffee and coffee substitutes show no significant consumption among children, as expected and recommended. Caffeine is not advised for children consumption, since their potential negative effects in younger populations are yet unknown [60].

Fast-food showed a relatively low median, with consumption frequency of one to three times per month, with 50% of participants doing so less times, indicating that it is viewed as occasional treat instead of a regular meal option. However, some outliers are concerning, with some children possibly consuming fast food on a daily basis. Fast-food meals have a higher content of solid fat (23.9% of total energy) than foods from retail stores (17.6%) or schools (20.9%) [61]. Also, the consumption of these foods is also often associated with consumption of sweetened beverages and lower intake of vegetables, fruit and milk, increasing total energy intake and leading to a poor-quality diet [62,63]. Some families may rely on this type of meals due to their convenience, accessibility and socioeconomic factors [64].

The trends for adults are very similar, with a few exceptions (Figure 3b). Regarding the first food group, the adults eat breakfast cereal, cookies and other sweets slightly less frequently than children’s students. Given that these products are frequently more closely linked to children and are widely promoted to younger audiences, this outcome is to be expected. Adults may also avoid these snacks, since they can be more conscious of their dietary choices.

Regarding vegetables and fruits, adult students seem to consume all the categories (fruit, soup, fruit natural juices and cooked vegetables) slightly less often than the children. Adults might be more likely to choose quicker and less nutritious options due to time constraints, while children may have more balanced meals with more structured mealtimes at school or home. Additionally, schools and parents tend to promote the consumption of fruits and vegetables as part of healthy eating habits for children, while adults may not receive the same external encouragement or monitoring. Socioeconomic factors could also play a role, limiting the access to fresh produce for some adult students.

Adults consume eggs at a more varied frequency than children, ranging from daily intake to no consumption, perhaps due to some dietary restrictions. The median frequency of consumption of light sodas and juices is slightly higher in adults, who may choose these beverages as an effort to manage weight or reduce sugar intake, which typically is not often a concern of children.

The consumption of coffee among adults is frequent, with a median of 4 to 6 days per week. The frequency has a high diversity, with 75% students drinking from zero to one time per day. However, coffee is a regular part of the daily routine of adults, with 25% of the students drinking coffee every day, from one to two or more times per day. Coffee, as a major source of caffeine, is well-known for its stimulating effects that help with concentration, fatigue management and alertness, which are particularly important for students in their demanding studying schedules. Coffee also has a role in adults’ social and work environments, often being a central part of routine and social gatherings, which is a particularly common practice in Portugal. Improving of attentiveness and the taste of caffeine products are the main reasons for their consumption in the country [65]. Adding to the energy related benefits of coffee, moderate consumption as health benefits giving its content in antioxidants, which has been linked with a reduced risk of diseases such as Parkinson and Alzheimer [66]. However, high caffeine intake has been associated in students with negative health effects, such as depression, anxiety and sleep issues [67].

Statistically, we did not find any relevant associations between the consumption of certain groups of food and the weight status of the students (Supplementary Figures S2 and S3). Cooked vegetables show a P-value of 0.029 for children and 0.016 for adults. Adding cooked vegetables to meals may significantly increase the daily vegetable intake of the individuals, beyond other vegetable sources such as soup. Vegetables are a good source of essential vitamins, minerals, and fibre, which promote satiety and help regulate caloric intake, which can help maintaining a healthy body weight [68]. However, other foods within these two food groups did not show significant differences between the weight groups, indicating the absence of a consistent trend across them. We also found a significance for coffee (P = 0.003) and for tea and infusions (P = 0.034) in the case of children, however, these beverages are not often consumed by them, particularly coffee, which limits the relevance of this finding. For the adults, we also found a significance for coffee consumption (P < 0.001). An association between a higher BMI and the consumption frequency of coffee (Spearman’s correlation coefficient = 0.260; P < 0.001) has been observed. The literature did not provide any evidence of an association between caffeine consumption and weight changes. We should consider examining whether individuals drink their coffee with sugar or not, as this could lead to an increase in daily caloric intake, which can be a possible explanation for weight gain related to coffee consumption. A study about coffee intake inversely associated an increase in intake of unsweetened caffeinated and decaffeinated coffee with weight gain [69].

Lastly, there is significance for the adults in the soy alternatives (P = 0.030), but the consumption of these foods is not significant for this group. While these few values show significance, they are not enough to suggest meaningful associations between food consumption and the weight status of the students.

3.5. Disease Incidence In The Student Community

Lastly, we analysed the incidence of diseases in the student community, through questioning the students if they have ever been diagnosed with any disease. Firstly, we found that 66% of the children have never been diagnosed with health problems before, and 34% have been diagnosed with at least one (Table 4). These results are a positive indicator of general health in the community. The most prevalent diseases found among the community of children and adolescent students were allergic (18.1%), pulmonary (8.2%), skin (7.9%), gastrointestinal (3.4%) and mental diseases (2.4%). Furthermore, we found no significant association between the diagnosis of any of the diseases and the weight status (Supplementary Figures S4-S8). An association between the age and the diagnosis of diseases was observed, with older children having a higher disease incidence than younger ones (U = 201381.5; Z = -8.019; P <0.001).

Regarding adult students, 55.8% has been diagnosed with one or more diseases before, while 44.2% has not. The most common diseases among this group, similarly to the children, are allergic (32.7%), pulmonary (11.1%), skin (10.1%), mental (9.2%) and gastrointestinal diseases (7.8%). In this case, we found a significant relationship between the weight status and the diagnosis of pulmonary diseases (P = 0.03) (Supplementary Figures S9-S13). When we calculate the relative risk, we can observe that the incidence of pulmonary diseases in the overweight/obese group is 2.3 higher than the underweight/normal group. A higher BMI is associated with reduced lung volume, which can manifest as a restrictive ventilatory pattern on spirometry, including reductions in expiratory reserve volume and functional residual capacity. Obese individuals may also experience decreased vital capacity, total lung capacity, and altered expiratory flow rates due to increased residual volume and airway resistance. Clinically, these patients tend to have a higher respiratory rate, leading to increased oxygen consumption [70].

An allergy epidemic has been blooming, with increased prevalence of atopic diseases such as allergic rhinitis, asthma, food allergies, conjunctivitis and atopic dermatitis, attributed to factors like diet, the hygiene hypothesis (reduced exposure to microbes early in life), air pollution, climate change and urbanization [71]. A cross-sectional study from primary schools in Belgium 2019 describes a prevalence of allergic rhinitis of 29.3%, with asthma being the most significant related comorbidity [72]. According to Asher M. et al. (2020), asthma is the most frequent chronic disease among children, worldwide. A systematic review from 2023 describes that prenatal and early-life exposure to traffic-related air pollution can significantly increase the risk of allergic rhinitis in children [73]. This explains why the allergic, pulmonary and skin diseases stand out in our study, with environmental pollutants probably playing an important role.

3.6. Logistic Regression Model

Lastly, we performed a binary logistic regression model to analyse all the variables in our study in relation to the BMI, separately for children and adults. The pre-requisites for performing the model were verified in both cases, and no multicollinearity was found between the independent variables, meaning that the variables used in the model are not highly correlated. Additionally, no outliers were identified in the model and there is a sufficient number of records per variable. The model performed 5 steps for the children’s data and 6 for adults.

The analysis used the Forward method, revealing low R² values (Table S1), which indicates moderate predictive capacity. In our case, this may not be critical, as our primary objective is to identify associations between dependent and independent variables rather than to make precise predictions.

The Hosmer-Lemeshow test indicated that the model is adequate, in both cases, with a significance value of 0.267 for children and 0.365 for adults, which is higher than the significance level of 0.05. This suggests that the model satisfactorily explains the dependent variable (obesity/overweight) with the independent variables.

In line with our previous analysis, the model presented some significant relationships. Regarding children (Table 5), the likelihood of overweight/obesity decreases with age (B = -0.137) indicating 12.8% of reduction in the odds if obesity per year (Exp(B) = 0.872). The amount of sleep was associated with a lower probability of obesity (B = -0.561). Sleeping more hours reduces the likelihood of obesity by 42.9% (Exp(B) = 0.571). A positive coefficient (B = 0.288) suggests that more time sitting down during the weekend increases the probability of obesity, with a 33.4% increase in the likelihood of obesity for more sitting hours during the weekend (Exp(B) = 1.334).

Lastly, females have a 43.8% reduction in the odds of being obese, compared to males (B = -0.576; Exp(B) = 0.562). In the past, data from Poland revealed a greater prevalence of obesity in boys than girls [74]. Another study with national Canadian data observed a higher prevalence of obesity in boys compared with girls aged 3–19 [75]. According to Shah et al., some biological variations in the body of girls and boys could explain this difference: girls tend to have greater fat mass and lower fat-free mass (e.g., muscle), which affects energy requirements, with girls requiring fewer calories than boys, thus reducing the risk of excess weight gain; girls have higher levels of leptin, a hormone that suppresses appetite and promotes energy utilization, while boys have higher levels of androgens, which suppress leptin production, leading to potentially higher appetite; some studies suggest that brown adipose tissue, which helps burn calories, is more prevalent in females [76].

In the case of adults, the significance found for soy alternatives and coffee substitutes (Table 5) does not seem relevant, as previously we observed a low consumption of these products. A positive coefficient (B = 0.329) suggests that consuming light sodas increases the likelihood of obesity by 39% (Exp(B) = 1.390). Although these beverages contain less sugar, their consumption could potentially be associated with other unhealthy habits, such as the consumption of fast-foods or regularly alternating between sugary and light beverages. For coffee, we found that a higher consumption increases the likelihood of obesity by 29.5% (B = 0.258; Exp(B) = 1.295). In the case of tea and infusions, their consumption is associated with a 35.9% increase in the likelihood of obesity (B = 0.307; Exp(B) = 1.359). As we previously suggested, these associations could be related to the addition of sugar to these beverages, which needs to be further analysed.

4. Limitations and Future Perspectives

Although our study provides valuable insights into the health and nutritional status of a group of students, it is important to acknowledge some limitations. First, the cross-sectional design limits our ability to establish causality between nutritional and lifestyle habits and obesity. Since the data is collected at a single point in time, we cannot analyse changes over time or determine the direction of relationships. Additionally, the use of self-reported data introduces potential biases, since participants might overestimate or underestimate their responses due to social constraint or inaccurate memory, which can affect the validity of our findings. Some participants may feel uncomfortable disclosing information about sensitive topics, such as weight, dietary habits and consumption of substances, especially the younger individuals. Furthermore, using a single measure like BMI does not allow us to fully evaluate the health status of the participants, since it does not account for body fat and muscle and, in the case of adults, for sex and age.

While the initial number of surveys was high, not all of them were answered, which may affect the results, as the people who chose not to participate may differ significantly from those who did. The fact that a part of the surveys was delivered in paper also complicates the process, since it is impossible to make the answers mandatory similarly to the online version, leading to a lack of some information and to incomplete data. Also, our study is geographically focused on students from the city of Guarda, Portugal, which may limit the applicability of our findings to other regions with different socioeconomic, cultural, and environmental contexts.

As the next steps, performing longitudinal studies could be a good approach to try to establish a relationship between nutritional and lifestyle habits and obesity, identifying trends and changes over time. The accuracy of the data can be improved by using equipment like accelerometers to measure physical activity [77], and therefore obtain more objective measurements. Other geographic areas with different socioeconomic and cultural aspects should also be studied and compared. Incorporating new technologies, such as mobile health apps and online platforms can be helpful for monitoring behaviours.

5. Conclusions

The findings of our study indicate a complex relationship between BMI and several factors, such as age, physical activity, sex and dietary habits among the students. According to our survey, the younger students, particularly boys, have a higher rate of overweight and obesity, which highlights the need for an early intervention. Higher physical activity and reduced sedentary behaviours were associated to a healthier BMI, which emphasizes the need for promoting a more active lifestyle in schools. Despite the limitations of the study, it is undeniable that public health strategies to prevent obesity are needed, educating the community for improved dietary habits and providing them with access to physical activity resources. In the future, broader longitudinal studies are essential to provide a better insight into the lifestyle of these students and the factors affecting their health outcomes.