Study population
At the beginning of the study (T0), 781 participants were recruited from primary schools. Of these, 525 participants (67%) completed the study at T1, comprising 269 boys (51%) and 256 girls (49%). The drop-out rate was 33% (254 children) primarily due to parents failing to complete the questionnaires at both T0 and T1.
The regional distribution of participants who completed the study at T1 is presented in Fig. 2.
Fig. 2
Regional distribution of participants. Summary of participant distribution across the three Italian regions (Lazio, Piedmont, and Sicily) and by age group (Group A = 6–7 years; Group B = 9–10 years). The map highlights the regional proportions using a color-coded layout for improved visual clarity.
Food frequency questionnaireSection I: general questions
Children who were on a diet:
Among the 525 participants, 95% (n = 497), were not following a specific diet. These participants were evenly distributed across the regions:
Lazio: 90% (Group A: 47%, Group B: 53%),
Piedmont: 96% (Group A: 52%. Group B: 48%).
Sicily: 97% (Group A: 48%, Group B: 52%).
Only 5% (n = 28) of participants across the three regions reported following a specific diet. This included:
These children were following specific diets for medical reasons—such as allergies (to certain fruits and mollusks), lactose intolerance, overweight, insulin resistance, type 1 diabetes, or hypercholesterolemia. However, they were not excluded from the analysis, as these dietary choices did not entail the complete exclusion of entire food groups.
Family meals
In both groups (A and B) and across all three regions, dinner was the meal “always” consumed at home, with the highest percentages observed in Sicily (96% and 95% for Groups A and B, respectively), followed by Lazio (91% and 88%) and Piedmont (90% and 89%), with only minimal differences between the latter two regions. Breakfast was most frequently consumed together with the family in Piedmont (58% and 60% for Groups A and B, respectively) compared to Lazio (57% and 38%) and Sicily (55% and 53%). Lunch was the meal least frequently eaten at home in Lazio and Piedmont, where many children had lunch at the school canteen. In contrast, in Sicily, the majority of children (80%) reported always having lunch at home.
Predominant figure in meal Preparation
In all three regions, the mother was the predominant figure in meal preparation, especially for dinner. This was particularly evident in Sicily (92% and 91% for Groups A and B, respectively) and in Lazio (90% in Group B).
In Piedmont, 40% of Group A participants reported that a third person (“other”)—neither the mother nor the father—prepared meals, which was higher than in the other regions. Fathers were more involved in preparing lunch and dinner in Group B of Piedmont, accounting for 10% and 18%, respectively, which was more than in the other regions.
Having lunch at the school canteen:
Group A: In Lazio, all children attended the school canteen with varying regularity: 16% regularly, 4% frequently, and 80% occasionally. In Piedmont, most children (73%) used the school canteen, with 59% attending regularly, 9% frequently, and 5% occasionally.
Group B: In Lazio, 99% of children had lunch at school, with 29% attending regularly, 7% frequently, and 63% occasionally. In Piedmont, 71% of children used the school canteen, with 54% attending regularly, 15% frequently, and 2% occasionally.
In contrast, in Sicily, only 3% of children in both groups reported having lunch at the school canteen.
Physical activity
In general, most children in the three regions practiced physical activity.
Group A: Piedmont had the highest percentage of children participating in sports (81%), followed by Sicily (77%), while Lazio had the lowest (61%).
Group B: Lazio had the highest percentage of children involved in sports (86%), followed by Piedmont (84%), while Sicily had the lowest percentage (75%).
Section II: portion sizes and frequencies of consumptionPortion sizes
The consumption of age-appropriate portion sizes in Group A (small portion size) is summarized in Tables 3 and 4. Comprehensive results for all regions (Lazio, Sicily, and Piedmont), including both statistically significant and non-significant differences between T0 and T1 for Groups A and B, are available in the Supplementary Materials (Tables S1–S6). To streamline data presentation, only the significant within-region variations observed in Sicily and Piedmont for Group A are reported in the main text.
Table 3 Group A. Percentage of children consuming the correct age-specific portion size. The table reports only the food items with a statistically significant difference between baseline (T0) and post-intervention (T1) in Piedmont. Statistical test: Chi-square. Effect size was calculated using cramer’s V.Table 4 Percentage of children consuming the correct age-specific portion size. The table reports only the food items with a statistically significant difference between baseline (T0) and post-intervention (T1) in Sicily. Statistical test: Chi-square. Effect size was calculated using cramer’s V.
In Group B, no significant variations were detected across the three regions between T0 and T1. Overall, inter-regional comparisons showed no significant differences in portion sizes for any food category at either time point, indicating consistent portion patterns across regions.
Frequencies of consumption
Group A:
Between T0 and T1, children in Piedmont significantly reduced their consumption of red pizza (from 1.58 ± 1.44 to 1.33 ± 1.39, p = 0.0078), cold cuts (from 1.81 ± 1.61 to 1.61 ± 1.56, p = 0.0042), and meat (from 2.77 ± 1.74 to 1.36 ± 1.32, p = 0.0022), while significantly increasing their intake of vegetables (from 4.08 ± 2.58 to 4.50 ± 2.44, p = 0.0022) and the number of fruits consumed per week (from 7.81 ± 4.44 to 9.99 ± 6.01, p = 0.0022).
In Lazio, children significantly increased the frequency of consumption of crackers (from 2.88 ± 2.11 to 2.91 ± 2.21, p = 0.0039) and fish (from 3.81 ± 1.35 to 4.19 ± 1.09, p = 0.0073) and significantly reduced the consumption of French fries (from 3.36 ± 1.80 to 2.93 ± 1.94, p = 0.0291).
In Sicily, younger children significantly decreased the consumption of French fries (from 1.47 ± 1.43 to 1.14 ± 1.29, p = 0.0076), crisps (from 1.48 ± 1.72 to 1.05 ± 1.23, p = 0.0026), and cold cuts (from 1.95 ± 1.79 to 1.73 ± 1.29, p = 0.0336), while significantly increasing the intake of vegetables (from 1.74 ± 1.58 to 3.30 ± 2.61, p = 0.0136) and the number of fruits per week (from 6.76 ± 4.39 to 8.04 ± 4.94, p < 0.0001) (Fig. 3).
Mean values ± SD for all food categories, along with corresponding p values and effect sizes, are reported in Table S7 of the Supplementary Materials.
Fig. 3
Weekly frequencies of consumption between T0 and T1 in each region in Group A. Statistical test: Wilcoxon and paired t-test (only for number of fruits/week). Effect size was calculated using Rosenthal and Cohen’s (only for number of fruits/week.
Group B:
Between T0 and T1, children in Piedmont significantly increased their consumption of vegetables (from 3.95 ± 2.82 to 4.27 ± 2.60, p = 0.0176) and the number of fruits consumed per week (from 6.88 ± 4.32 to 9.63 ± 6.05, p < 0.0001).
In Lazio, children significantly increased only the frequency of consumption of nuts (from 1.88 ± 2.18 to 2.33 ± 2.39, p = 0.0052). In Sicily, children significantly reduced the consumption of crisps (from 1.30 ± 1.41 to 1.01 ± 1.19, p = 0.0130) (Fig. 4).
Mean values ± SD for all food categories, together with p-values and effect sizes, are reported in Table S8 of the Supplementary Materials.
Fig. 4
Weekly frequencies of consumption between T0 and T1 in each region in Group B. Statistical test: Wilcoxon and paired t-test (only for number of fruits/week). Effect size was calculated using Rosenthal and Cohen’s (only for number of fruits/week.
Group A:
Children in Lazio reported higher consumption of pizzas (plain and red), soups, potatoes (boiled, roasted, mashed), French fries and crisps compared with Piedmont and Sicily across both T0 and T1 (all p < 0.0001). Crackers consumption was also greater in Lazio than in Sicily, both at T0 (p < 0.05) and at T1 (p < 0.01). Rusks were also more frequently consumed in Lazio compared with Piedmont, already higher at T0 (p < 0.001) and further increased at T1 (p < 0.0001). Other cereals (non-wheat grains) were more commonly consumed in Lazio compared with Sicily at T0 (p < 0.05), while bread intake was higher in Sicily than in Piedmont (p < 0.01). No significant differences were observed for pasta or biscuits.
Lazio consistently showed a higher intake of fish, cold cuts, eggs, meat, yogurt, fresh cheeses compared with both Piedmont and Sicily at T0 and T1 (all p < 0.0001). Legume intake was consistently higher in Lazio compared with Piedmont and Sicily, with highly significant differences at both time points for Lazio vs. Piedmont (p < 0.0001), and for Lazio vs. Sicily at T0 (p < 0.0001) and T1 (p < 0.001). Nut consumption was consistently higher in Lazio compared with Piedmont at both time points (p < 0.01), while aged cheese intake was higher in Lazio compared with Piedmont at T0 (p < 0.01) and T1 (p < 0.001), and also higher than in Sicily at T0 (p < 0.01). Milk consumption was greater in Sicily than in Lazio at T0 (p < 0.01). Vegetable drinks were consumed more frequently in Lazio than in Piedmont at T0 (p < 0.05).
Vegetable intake was higher in Piedmont than in Sicily at T0 (p < 0.01) and remained higher at T1 (p < 0.05). Lazio exceeded Sicily at T0 only (p < 0.01). At T1, fruit consumption was significantly higher in Piedmont than in Lazio (p < 0.001).
Details in Table 5 and in Supplementary Materials Figure S2.
Table 5 Group A. Frequencies of food consumption, stratified by region (Lazio, Piedmont, Sicily) at baseline (T0) and follow-up (T1). The table reports pairwise comparisons between regions. Arrows indicate the direction of difference (↑ higher, ↓ lower) in the first region listed compared with the second. The Kruskal-Wallis test, with dunn’s post hoc tests for pairwise comparisons, was applied. Effect sizes (ε2) are reported to indicate the magnitude of observed differences.
Group B:
Red pizza was consumed more frequently in Lazio than in both Piedmont and Sicily at T0 and T1 (all p < 0.0001).
Plain pizza showed a distinct pattern: consumption was higher in Lazio compared with Sicily at both T0 and T1 (all p < 0.0001), and higher in Lazio than in Piedmont at T0 (p < 0.001) and T1 (p < 0.01). Piedmont, in turn, reported a greater intake of plain pizza than Sicily at T0 (p < 0.01) and T1 (p < 0.001). Soups were more frequently consumed in Lazio: at T0 intake was higher than in both Piedmont and Sicily (p < 0.0001), while at T1 it remained higher compared with Piedmont (p < 0.001) and Sicily (p < 0.01). Similarly, potato consumption (boiled, roasted, mashed) was higher in Lazio than in Piedmont at T0 (p < 0.001) and T1 (p < 0.01), and higher than in Sicily at both T0 and T1 (p < 0.0001). French fries were consumed more frequently in Lazio than in Sicily at T1 (p < 0.05).
Other cereals (non-wheat grains) were more commonly consumed in both Lazio and Piedmont compared with Sicily at T0 (p < 0.05 for Piedmont vs. Sicily; p < 0.001 for Lazio vs. Sicily).
Crackers were consumed more frequently in Piedmont than in Sicily at T1 (p < 0.01). No significant differences emerged for bread, rusks, biscuits, or crisps.
In Lazio, the intake of meat and egg was significantly higher than in Sicily and Piedmont at both T0 and T1 (all p < 0.0001). Fish intake followed a similar pattern, being higher in Lazio than in Piedmont at both T0 and T1 (p < 0.0001), and also higher than in Sicily at T0 (p < 0.01) and T1 (p < 0.0001).
Cold cuts were more frequently consumed in Lazio than in Piedmont (p < 0.001 at T0; p < 0.01 at T1) and then in Sicily (p < 0.01 at T0; p < 0.001 at T1). Fresh cheeses intake was also higher in Lazio compared with Sicily at both time points (p < 0.01). Legumes were more often consumed in Lazio than in Piedmont at both T0 (p < 0.01) and T1 (p < 0.001). No significant differences were observed for milk, vegetable drinks, yogurt, aged cheeses, or nuts.
As for fruit and vegetables, Piedmont consistently showed higher vegetable intake than Sicily at both T0 and T1 (p < 0.01). Lazio also reported greater vegetable consumption than Sicily at T0 (p < 0.001) and at T1 (p < 0.05). At T1 fruit intake was higher in Piedmont than in Lazio (p < 0.0001).
Details in Table 6 and in Supplementary Materials Figure S3.
Table 6 Group B. Frequencies of food consumption, stratified by region (Lazio, Piedmont, Sicily) at baseline (T0) and follow-up (T1). The table reports pairwise comparisons between regions. Arrows indicate the direction of difference (↑ higher, ↓ lower) in the first region listed compared with the second. The Kruskal-Wallis test, with dunn’s post hoc tests for pairwise comparisons, was applied. Effect sizes (ε2) are reported to indicate the magnitude of observed differences.Section III: frequency of consumption of seasonings, drinks and snackSeasonings
This section assessed the consumption of seasonings, including extra virgin olive oil, seed oil, butter, margarine, cream, and lard. A significant reduction in the frequency of consumption was observed in Piedmont for both Group A and Group B across all items, except margarine (p < 0.05). Conversely, in Group B in Lazio, a significant increase in seed oil consumption was observed (p = 0.0391). No significant differences were observed in Sicily in both group A and group B from baseline to T1.
Beverages
This section assessed the consumption of beverages, including fruit juice, milkshakes, orange juice, sugary drinks, tea, and light beverages.
Between T0 and T1, a significant reduction in fruit juice consumption was observed only in Group A in Piedmont (p = 0.0450).
Regarding daily water intake, no significant changes were observed in Piedmont in either Group A or Group B between T0 and T1. In contrast, in Lazio, a significant increase in the number of glasses of water consumed per day was observed in both Group A (from 6.62 ± 2.93 to 7.48 ± 3.68, p = 0.0276) and Group B (from 6.66 ± 3.42 to 7.35 ± 3.73, p = 0.0264). In Sicily, a highly significant increase in daily water consumption was observed in both Group A (from 6.27 ± 2.39 to 8.05 ± 2.57, p < 0.0001) and Group B (from 6.93 ± 2.76 to 8.87 ± 3.06, p < 0.0001).
Children in Group B from Sicily consistently reported higher water consumption compared to those from the other two regions. At T1, their intake was significantly higher than in both Lazio (p < 0.0001) and Piedmont (p = 0.0070).
No significant differences in water consumption were observed among the three regions in Group A.
Snacks
This section evaluated the consumption of snacks, including dark chocolate, candies, and chewing gum. Group A in Piedmont and Sicily showed changes in candy consumption, but with opposite trends: in Piedmont, consumption significantly increased (p = 0.0312), whereas in Sicily, it significantly decreased (p = 0.0079). Additionally, in Group B in Sicily, chewing gum consumption also showed a significant decrease (p = 0.0098). No other significant changes were observed in both groups in the three regions.
Details regarding beverages, seasonings and snacks are reported in Tables S9 and S10 in Supplementary Materials.
Comparing beverages, seasonings and snacks consumption across the three regions ant both time (T0 and T1) no significant differences were observed (p > 0.9999 for all comparisons, partial η2 = 0.00).
Section IV: effectiveness of “Nutripiatto”
The final section of the questionnaire, present only at T1, evaluated the effectiveness of the Nutripiatto. Despite the absence of statistically significant differences, the tool appeared to be equally effective in the three regions as parents reported improvements in their children’s lifestyles (an increasing trend in consumption of whole grains, vegetables, and water, and a reducing trend in protein portions and sedentary behaviour).
In fact, in Group A, the effectiveness of Nutripiatto as a visual guide was observed in 96% of children in Sicily, 92% in Piedmont and 88% in Lazio, while in Group B the rates were 97% in Sicily, 94% in Piedmont and 69% in Lazio.
Moreover, in Group A, the consumption of low-refined grains increased in 61% of children in Piedmont, 25% in Lazio, and 21% in Sicily while in Group B, the percentages were 56% in Piedmont, 55% in Lazio, and 33% in Sicily. Additionally, a reduction in meat and fish portions after using Nutripiatto was observed. In Group A, the percentages were 52% in Piedmont, 30% in Sicily and 25% in Lazio while in Group B, the percentages were 63% in Lazio, 49% in Piedmont and 46% in Sicily.
Regarding the consumption of vegetables, in Group A, 83% of children in Piedmont, 61% in Lazio, and 60% in Sicily increased their consumption. In Group B, the percentages were 81% in Piedmont, 79% in Lazio, and 59% in Sicily.
Furthermore, most children increased their water consumption, with percentages of 86% in Piedmont, 75% in Sicily and 57% in Lazio in Group A, and 77% in Sicily, 76% in Piedmont, and 71% in Lazio in Group B.
Lastly, the majority of children demonstrated increased physical activity following the use of Nutripiatto. In Group A, the percentages of children who reduced sedentarity were 80% in Piedmont, 66% in Sicily and 61% in Lazio, while in Group B, 73% in Lazio and 66% both in Piedmont and in Sicily.
Dining and Cooking