Food intake,physical activity and quality of life among children and youth

2.1 Participants

A sample of 337 children and youth, aged 8 to 17 (12.61±2.96), mostly from Portugal inland (79.8%) and rural areas (56.6%) of which 58.2%are female, participated in this study. Regarding academic qualifications, 21.3%completed the first cycle of basic education, 22.2%the second, 26%the third and 30.5%completed secondary education. Parents of the sample have mostly secondary education (fathers 32.9%and mothers 35.9%). Lunch is mostly away from home (64.4%), while practically everyone dines at home (96.1%). Only 10 were found to have both meals away from home. It was noted that 34%were underweight while 6%were overweight.

2.2 Measures

A socio-demographic and anthropometric questionnaire with questions about weight, height, time (minutes) sitting and walking daily (physical activity measure), a weekly food ingestion record table (lunch and dinner) and the Portuguese version of Kid-Kindl (a quality of life questionnaire), validated by Ferreira et al. [15], were used in data collection. The walking time, in the absence of reference values, because it is a questionnaire constructed for this purpose, was corrected by analogy with International Physical Activity Questionnaire - IPAQ [16]: only values of 10 or more minutes of activity were included, based on the scientific evidence that episodes of at least 10 minutes are required to achieve health benefits; and the time exceeding 3 hours (180 minutes) was re-coded to 180 minutes. Food consumption was scored in weekly consumption frequency at lunch and dinner, including various categories of food and beverages, as well as types of confectionery. Weekends were excluded. The Kid-Kindl is a short and psychometrically valid and reliable 24-item measurement instrument that evaluates six health-related QoL dimensions (physical well-being, emotional well-being, self-esteem, family, friends, and school) in children and youth. All questions in the instrument refer to the temporal universe of the week prior to completion and were answered on a scale of 1 to 5, with 1 never, 2 rarely, 3 sometimes, 4 often and 5 always. There are partial (by dimension) and global scores (0–100 scale) where higher means better QoL.

2.3 Procedure

The data collection was done through direct administration in domiciliary visits, during December of 2016 and the ethical care guarantees regarding any research process were ensured, including a signed informed consent by the legal guardians of the participants. Authorization of the authors of the instrument was also guaranteed. Parents completed the sociodemographic and anthropometric questionnaire and the food intake registration table. The Kid-Kindl has been completed by the children/youth.

2.4 Data analysis

Descriptive (average, standard deviation, frequencies and percentages) and inferential parametric (One Sample t-Test) and non-parametric (Mann-Whitney Test, Spearman’s rho and Wilcoxon Test) analyses, since there is no resistance to normality criteria, were performed using SPSS-IBM 25, with an established significance of p≤0.05.

3.1 Descriptive analysis

Regarding BMI, the participants are distributed in 3 levels (see Table 1). Similarly, Morales-Suarez-Varela, Julve, and Gonzalez [17] found percentages of 57.3%and 7.4%normal weight and obesity, respectively, in Spanish children aged 5 to 15. Also, Grosso et al. [18] found percentages of 46.7 in normal weight, 41.1 overweight and 11.2 obesity in children and adolescents (4–16 years old) from a rural area of Sicily (Italy). An Asiatic study (Northeast China city), with 3205 students (aged 7–18 years), showed 71.8%of normal weight, 12.5%of overweight and 15.7%of obesity [2].

In terms of eating habits (Table 2) meat is the preferred food of the participants, followed by fish and eggs with low consumption (only 2-3 times a week). 35 individuals reported eating meat at lunch and dinner, at least 5 times a week, while only 4 and 1 eat fish and eggs, respectively. Most (91.2%) consume carbohydrates at lunch (vs 78.4%at dinner), but 246 (74.5%) do so at both meals. Only 2 individuals reported never consuming carbohydrates at lunch and dinner simultaneously. Less than 50%consume soup, at least in one of the meals considered, although it is preferred at lunch (46.4 vs 33.9). However, there are 7%(20) who never eat soup at both meals. The preferred type was vegetables soup (73%at lunch and 66%at dinner).

We highlight the reduced consumption of vegetables. Daily, only 31 individuals consume vegetables at lunch and 19 at dinner, while 83 (25.1%) participants never consume vegetables at any of these meals. In Colombia, Campo-Ternera et al. [19] also obtained a high rate (17.5%), although lower, of children and adolescents who never consume vegetables. Also, fruit intake should be improved, as fruit consumption is reported daily by about 35%at lunch or dinner, but only 23%in both meals. On the other hand, there is a positive note of the reduced consumption of sweet desserts during the week and only 5 individuals consume sweet desserts in the 2 meals. 53 participants do not consume sweets during the 5 days of the week, which can mean that they are reserved for weekend, or less. Borrmann, Mensink and Group [20] conclude that only 12.2%of a sample of German girls and 9.4%of boys consume the recommended five portions of fruits and vegetables per day; on average, girls and boys consume 2.7 and 2.4 portions a day.

Yogurt is not the option of the participants, for lunch and dinner, 5 days a week, since 76%never consume yoghurt. However, this should be an option up to the point that it does not interfere with iron intake, because calcium may reduce the iron absorption. However, the inhibitory effect may be of short duration because the studies on long-term calcium supplementation do not show any adverse effects on iron status [21]. Hence, foods rich in calcium, such dairy, must be part of the diet, but not simultaneously with meals where the iron intake is key (lunch and dinner).

At lunch and dinner, 66.3%and 55%drink water, respectively. However, only 48.6%drink water in these two meals while 9 people never drink water at all. It would make sense to see if water intake occurs at other times of the day or falls short of what is desirable (1.5–2L/day).

The participants reported drinking more natural juices than soft drinks (6.5–8.0%and 4.4–5.9%) at lunch and dinner, but the differences are not very marked. The study by Campo-Ternera et al. [19] showed higher percentages of soft drinks than ours (16.9%).

Bread is preferred at lunch, with approximately 40%of respondents opting for different types. Here too, one can intervene to promote this behaviour. Current nutritional recommendations suggest that about a third of food intake should come from the food group comprising bread, rice, potatoes, pasta and other starchy foods [22]. However, common consumer perceptions that starchy foods are potentially fattening and have a high glycaemic index are eventual barriers to the consumption of such foods [23].

Several cooking techniques, excluding frying, both at lunch and dinner, are already used by about 70%of the participants. There is evidence that frying presents a higher content of fat than the other techniques [24]. García-González, Achón, Alonso-Aperte and Varela-Moreiras [25] concluded, with a Spanish adult sample, that the cooking techniques most frequently used were grilling (66%), stewing (49%), and baking (42%) while 20%preferred deep frying.

Considering PA (Table 3), we highlight the minutes walking daily, whose mean (97.3±66.7) is highest than the minimum of 60 minutes recommended by the WHO [26] and adopted by the European Union. However, the dispersion is high, (standard deviation), and there are 31.8%who walk less than an hour and 13.5%less than half an hour. Campo-Ternera et al. [19] found 65.4%walk daily less than an hour in Colombian children and adolescents. In addition, the mean of almost 7 hours sitting per day is very high and contributes, largely, to the dangers of sedentary lifestyle. We would like to draw attention to the point that this information is only about the time the sample walk per day and these young people can do, concomitantly, other physical activities of moderate intensity. Hulteen et al. [7], in a review article, point out that walking was not classified as a top five activity in Europe. European children and adolescents prefer swimming, running and team sports.

Regarding QoL (Table 3), the scores are higher than those of the Portuguese validation [15], independently of gender and age, but there is similarity in the highest and lowest values, both the global scale and the subscales. Positively, the participants have high scores in health-related quality of life dimensions, but almost all the means are lower in older participants. We highlight the lower values of self-esteem and at the school dimension. In the group aged 8–12, the friend’s dimension is higher than the family, but this relationship is reversed in the older age group. It may will be related to the influence of the group associated with sociomoral development, in the youngest participants. Also, in cultural adaptation and validation studies, Ferreira et al. [15] found this progress, although in the group of 13 to 16 years, there was an approximation, but without inversion.

There are statistical differences between lunch and dinner eating habits (Table 4), with the higher consumption at lunch in most foods, regardless of whether they are healthy. This must be related to an association between greater need for energy consumption at lunch, due to the tasks still to be performed and an association between higher energy intake at lunch with a lower risk of weight gain [27]. In addition, since most meals out of house are at lunch, the preparation of the dishes quantities is not carried out by the sample and is intended for standard consumers. On the positive side, we highlight the higher consumption of water, soup, fruit and vegetables; while negatively, the consumption of sweet desserts. At dinner, there is more charcuterie consumption, but the consumption of natural juices also stands out.

Negatively, at home, the participants consume more soft drinks (U = 10362.0, p = 0.001), charcuterie (U = 10763.0, p = 0.005) and fried food (U = 10509.5, p = 0.029) and consumption outside of the home, includes sweet desserts (U = 14705.5, p = 0.007). In addition, the intake of eggs is more at home if the meal is the lunch (U = 10207.5, p = 0.000), but at dinner, is higher when eating out (U = 2599.5, p = 0.049). Generally, according to Shonkoff et al. [4], restaurant food and beverages are linked to overconsumption of calories, sodium, total and saturated fat, and sugar-sweetened beverages as well as under-consumption of milk, fruits, vegetables, and vitamins, in children. Also, Andegiorgish et al. [2], in a cross-sectional observation, conclude that overweight children consumed significantly more sweet foods and often use food services outside of the home (general and fast-food restaurants or school cafeterias) rather than normal weight children.

Relatively to the statistically significant comparisons between QoL (global score and subscales) means obtained in our study and in the validation sample, our results are generally high. The main differences are in the female group; only in one age group or both (self-esteem and school dimensions). The differences in the males are all favourable to our sample and are verified in the global score (t = 2.724; p = 0.010) and in the dimensions of self-esteem (t = 8.278; p = 0.000), school (t = 2.718; p = 0.010) and friends (t = 7.563; p = 0.000). Only the last one refers to the younger age group and is also exclusive to males, while among the females there are differences in all other scales: aged 13–17, in global score (t = 5.795; p = 0.000), physical well-being (t = 3.452; p = 0.001), self-esteem (t = 8.347; p = 0.000), family (t = 2.450; p = 0.017) and school (t = 8.278; p = 0.000). In self-esteem, there are differences also in the female younger group (t = 5.561; p = 0.000). It would be interesting to explore why the scores of younger female children contradict the tendency towards superior results to those of validation sample in the emotional well-being (t = 2.241; p = 0.029), and school (t = 6.155; p = 0.000) dimensions.

We found statistical differences in PA, EH and QoL regarding the BMI levels. Hence, the low weight group performed better than the normal weight group, in minutes walking (X²KW(2) = 9.06, p = 0.011), and also consumes more soup (X²KW(2) = 11.6435, p = 0.003, at lunch, and X²KW(2) = 16.519, p = 0.000, at dinner), fruit (X²KW(2) = 10.838, p = 0.004), fish (X²KW(2) = 10.279, p = 0.006), carbohydrates (X² KW(2) = 25.206, p = 0.000) and water (X²KW(2) = 10.958, p = 0.004). The literature highlights that the food intake and physical activity are the two primary factors which determine body weight and focused the impact of decreased consumption of fresh fruits and vegetables on weight of children [2]. The results may eventually eliminate the myth of carbohydrate consumption associated to the overweight. This food group is essential in a balanced diet, if the recommendations are met. Recommendations for calorie and macronutrient intake in school-age children includes 45–65%of total daily calories by carbohydrates, with no more than 10%from simple sugars - white bread, candy, chips [28]. Also, the low weight group has more self-esteem (X²KW(2) = 13.365, p = 0.001), but in the family dimension of QoL, the scores are highest in overweight (X²KW(2) = 8.194, p = 0.017). Considering that the dimension ‘self-esteem’ includes being proud of oneself, having felt “the greatest” or satisfied with oneself, we recall the importance, in these ages, of the body image, often associated negatively with being overweight. The ‘family’ dimension measures the relationship of the child / adolescent with the family, having had, in the last week, well-being with their parents, having felt good or having had discussions at home, beyond whether parents have let them to do anything (children) or feel limited/controlled by their parents (adolescents). Thus, parental control of food can often be a factor of lower well-being and potentially conflicting.

There are no gender differences in physical activity and quality of life. In eating habits, there is a difference favouring girls in the consumption of fruit in nature or ready-made fruit (U = 11234.0; p≤0.05), perhaps contemptible, because isolated. However, Hoy, Goldman and Moshfegh [29] showed gender significant differences in estimates of fruit and vegetables intake, higher among females. Galán et al. [6] highlight the positive effect on health produced by PA higher in females compared to males.

In the older group (13–17 years), the results in all variables (QoL, EH and PA) are the lowest (Table 5). In relation to the QoL, this was expected, considering the validation data for the Portuguese population [15]. The authors refer to the coincidence at this stage of growth (adolescence) in which individuals become more interested in everything concerning their health. The increased awareness of diseases and changes in their health status, and consequently worst self-perceptions of QoL, may be because they were more preoccupated about health.

The younger walk more minutes per day. Similarly, Chen et al. [14] found more physical activities or less sedentary behaviour in younger South Australian children and adolescents. Also, Benitez-Porres, Alvero-Cruz, Sardinha, Lopez-Fernandez and Carnero [30] found that children achieved better results than adolescents in the total steps/day. Relatively to the eating habits, Borrmann et al. [20] observed that younger participants consume significantly more portions of fruits than older participants. In contrast, Grosso et al. [18] point out that the older age group was associated with a higher consumption of fruit and vegetables, as well as higher weekly physical activity. We are convinced that if the major concerns about weight gain inherent in the adolescent body image may be associated with fears about more food intake, is also true that it is a period in which the relationship with peers (including love relationship) may consume time which was once used in physical activity. So, differences between these age stages could be expected because adolescents experience several changes in social influences which affect their PA pattern.

We found positive associations between PA and QoL (Table 6) which concur in other several studies: Finne, Bucksch, Lampert and Kolip [31] found, in a German representative sample of children and adolescents (aged 11–17 years) higher frequency of PA in people with higher scores of quality of life in relation to health; Galán et al. [6] also reported this association in a national sample of Spanish children and adolescents; and Lacy et al. [32] already highlighted this association in Australian adolescents. Those and other 28 studies with similar results were systematically reviewed by Wu et al. [8] founding associations between physical activity and/or sedentary behaviour and health-related quality of life among children and adolescents (3–18 years), from databases of MEDLINE, EMBASE, PSYCINFO and PubMed, between 1946 and January 2017. The main conclusions drawn were that the observed associations between PA and QoL in children and adolescents are consistent irrespective of sociodemographic, anthropometric and socioeconomic characteristics.

From the positive correlations of PA with food consumption, we highlight those obtained with sweet desserts, found in both meals (lunch and dinner) which may be attributed to the compensation of energy intake. In fact, Andegiorgish et al. [2] concluded that students who ate sweet desserts more frequently have higher BMI compared with those who ate less frequently.