Abstract
Background
This study aimed to evaluate the effects of a dietitian-led, school-based nutrition education programme on primary school students’ nutrition knowledge, attitudes, behaviours, and anthropometric measurements.
Methods
A randomized controlled, prospective design was conducted in a primary school in Mardin with 67 fourth-grade students (32 intervention, 35 control). The intervention group received 8 weeks of classroom-based nutrition education, and their parents received 4 weeks of education. Data were collected via sociodemographic forms, anthropometric measurements, Nutrition Knowledge Test (NKT), Nutrition Attitude Scale (NAS), Nutrition Behaviour Scale (NBS), and Child Physical Activity Questionnaire (CPAQ).
Results
The intervention group showed significant improvements in knowledge, attitudes, and behaviours (p < .001). A significant reduction in waist circumference was observed (p = .023). Physical activity was negatively correlated with waist circumference and positively correlated with nutrition behaviour scores. ANCOVA results confirmed significant group effects favouring the intervention group in NKT and NAS post-test scores after adjusting for baseline differences (p = 0.007 and p < 0.001, respectively).
Conclusions
School-based nutrition education programs conducted by dietitians can improve children’s nutrition knowledge, attitudes, and behaviors and support healthier anthropometric outcomes such as waist circumference. These findings support the inclusion of nutrition education provided by dietitians to promote the development of healthy eating habits in children in national health and education strategies.
Trial registration
The trial is registered at ClinicalTrials.gov (CT.gov identifier: NCT07168928, Registered 11 September 2025).
Supplementary Information
The online version contains supplementary material available at 10.1186/s12889-026-26194-x.
Keywords: School-based nutrition education, School-aged children, Nutrition knowledge, School dietitian, Anthropometric measurements
Introduction
Childhood overweight and obesity have increased significantly worldwide over the past twenty years [1, 2]. Global trends confirm that unhealthy eating habits among children and adolescents are leading to an increasing prevalence of overweight and obesity [3]. In many low- and middle-income countries, this situation further exacerbates the dual burden of malnutrition, which is often accompanied by micronutrient deficiencies such as anemia [2, 4]. Studies conducted in Peru report that overweight rates have reached 38.4% and anemia prevalence has reached 20% [4].
National monitoring data in Turkey indicate a similar trend. The overweight rate among children aged seven to eight is 12.5%, while the obesity rate is 9.9%; this means that one in five children is outside the normal BMI range [5]. In the Southeast Anatolia Region, including Mardin, these rates are similar to the national average [5]. Childhood obesity and malnutrition have significant consequences, such as increased risk of chronic disease, reduced quality of life, decreased school performance, and increased healthcare costs [6–8].
Nutrition knowledge gaps, negative attitudes, and unhealthy eating behaviors contribute to the exacerbation of these issues [4]. Studies conducted in Nepal and other developing countries indicate that adolescents have insufficient knowledge about nutrition or are influenced by misinformation [9]. Therefore, schools are considered a critical setting for promoting healthy eating due to their capacity to directly reach the young population and their potential to support behavioral change [2, 9].
The international literature indicates that school-based nutrition education interventions can be effective in influencing students’ knowledge, attitudes, and nutrition behaviors. Programs conducted in Peru [4], Nepal [9], and Ethiopia [10] reported significant improvements in knowledge and attitude levels and, in some cases, healthy nutrition behaviors. However, results regarding anthropometric measurements are quite heterogeneous; while some studies show positive changes, systematic reviews indicate that these effects are inconsistent [2, 4, 10]. The largely multi-component nature of the interventions, which include both educational and school environment-based measures, makes it difficult to analytically disentangle the independent effect of the educational component [2]. This highlights the need for randomized controlled trials evaluating the effectiveness of short-term, standardized, school-based nutrition education programs with well-defined content [1].
Current evidence indicates that, despite the positive effects of school-based nutrition education programs, studies that evaluate both behavioral and anthropometric measurements together are limited. Therefore, this study aimed to evaluate the effects of a school-based nutrition education program on elementary school students’ nutrition knowledge, attitudes, behaviors, and anthropometric outcomes using a randomized controlled trial design.
Methods
Study design and sample
This study was conducted as a pre-test, post-test, and follow-up randomized controlled trial. Data were collected from 4th-grade students attending a public primary school located in an urban district of Mardin, a culturally diverse city in Türkiye’s Southeastern Anatolia region. The study population consisted of 8 classes with approximately 35 students each, for a total of 280 students. Sample size was calculated using G*Power 3.1 for a two-tailed independent-samples t-test, based on an expected medium-to-large effect size (f = 0.566) derived from Raut et al. (2024), with α = 0.05 and power = 0.80 [9]. The minimum required sample was 22 students per group (n = 44). To account for potential attrition, additional participants were recruited, and the final analysed sample (n = 67) exceeded this requirement, ensuring adequate statistical power. The final analysed sample exceeded the minimum requirement, ensuring adequate statistical power for the comparisons. 4th grade students were included in the study due to the recommendations in the literature that intervention studies related to nutrition should be carried out in the 7–14 age range, that they are generally carried out with 10-11-year-old students, and that the education to be given in the study will be understandable and applicable by the students[46,47]. Data were collected between November 2024 and April 2025. The researcher collecting the data was blinded to group allocation throughout the study. The study was completed with a total of 67 participants, 32 intervention and 35 control group students, who met the inclusion criteria and participated voluntarily. The inclusion criteria were being a 4th-grade student aged 10–11 years, having no chronic disease or dietary restriction that could affect eating habits, attending school regularly during the study period, and voluntarily participating with parental informed consent. The exclusion criteria were having a chronic or metabolic condition that could influence nutritional status, being absent during the pre-test or post-test periods, having previously participated in any formal nutrition education programme, transferring to another school during the study, or choosing to withdraw from participation. As shown in the flowchart (Fig. 1), 70 students were initially randomized (35 intervention, 35 control). Three students from the intervention group were excluded during follow-up—two due to relocation and one due to voluntary withdrawal—resulting in a final analytical sample of 67 students (32 intervention, 35 control). The total intervention period of nutrition education for students was 8 weeks, and for parents, it was applied for a total of 4 weeks. The study was conducted as a pre-test (1st test before the training), nutrition education to the students and parents included in the intervention group, post-test to be applied after the training (2nd test) and follow-up test to be applied 2 months after the training (3rd test). The pre-test and post-test were applied to all participants and the follow-up test was applied only to the intervention group. The primary results of this study are the evaluation of the Nutrition Knowledge Test, Nutrition Attitude Scale, Nutrition Behavior Scale, and Children’s Physical Activity Questionnaire scores. The secondary results are the evaluation of differences in demographic characteristics, dietary habits, and anthropometric measurements.
Fig. 1.
Flowchart of participants
Participants and randomisation
Randomisation was carried out using a random numbers table (https://www.randomizer.org) among individuals who met the inclusion criteria. Participants were assigned to the control (n = 35) and intervention (n = 35) groups by the researchers according to the numbers generated in the programme. A total of 3 participants in the intervention group left the study, 2 because of a change of city and 1 because he/she did not want to continue the study (Fig. 1).
Data collection tools
The data were collected in the classroom environment by obtaining the necessary permissions during school hours. The questionnaire form consisted of socio-demographic characteristics (age, gender, family structure, parental education and employment status), Nutrition Knowledge Test, Nutrition Attitude Scale, Nutrition Behaviour Scale, Food Consumption Frequency Form, Child Physical Activity Questionnaire. Anthropometric measurements of the students were also taken. The questionnaire form was collected 3 times: at the beginning of the study (pre-test), at the end of the nutrition education (post-test) and 2 months after the education (follow-up test). After the training, a 2-month waiting period was foreseen in order for the training to be assimilated and transformed into behaviour. To minimise recall and social desirability bias, several procedural safeguards were implemented during data collection. All questionnaires were administered in quiet classroom environments under the supervision of both the researchers and the class teachers. Students were seated separately to prevent peer influence and were instructed to complete the forms independently. Before administering the questionnaires, the purpose of each tool was explained in age-appropriate language, and sample items were demonstrated to ensure comprehension. Students were explicitly informed that the questionnaires were anonymous, that their answers would not be shared with teachers or parents, and that there were no right or wrong responses. The presence of familiar teachers was intended to support comfort and trust, while researchers ensured that no guidance or cues were given that could influence responses. These procedures were used to reduce both recall-related inaccuracies and social desirability tendencies among young participants. Data collection forms;
Nutrition knowledge test (NKT)
In order to evaluate the nutritional behaviours, knowledge level and attitudes of the students, questions prepared by making use of the literature and in parallel with the nutrition education given were used [11, 12]. In the evaluation, + 1 point was given for each correct answer, 0 points for each incorrect answer, and the total score was reported. Higher scores obtained from the test indicate higher nutritional knowledge.
Nutrition attitude scale (NAS)
Children’s Cardiovascular Health Attitude Scale developed by Arvidson (1990) [13] was adapted to the Turkish language by Haney and Bahar (2014) [14]. The scale consists of exercise, nutrition, smoking and stress control sub-dimensions. In this study, only the ‘nutrition’ sub-dimension was used. The overall internal consistency coefficient of the scale was reported as 0.79 and the internal consistency coefficient of the nutrition subscale was reported as 0.68. The nutrition sub-dimension assesses the child’s attitude towards behaviours to reduce fat intake, increase healthy food consumption and diets that support heart health. Scale items are scored between 1 (strongly disagree) and 4 (strongly agree) and the total score ranges from 4 to 16. The high scores obtained reflect favourable nutritional attitudes. In the analyses conducted within the scope of this study, the internal consistency coefficient of the nutrition sub-dimension was reported as 0.73.
Nutrition behaviour scale (NBS)
It was developed by Edmundson et al. (1996) [15] to determine children’s food preferences. It consists of 14 items with healthy and unhealthy food visuals. The Turkish validity and reliability study of the Nutrition Behaviour Scale was conducted by (Öztürk & Erdoğan, 2010) [16] and the internal consistency reliability coefficient was 0.68. In the scale, -1 point is given for each unhealthy food choice and + 1 point for each healthy food choice and the total score varies between − 14 and + 14. High scores indicate healthy eating habits.
Child physical activity questionnaire (CPAQ)
The questionnaire was used to determine the physical activity levels of the students. The reliability study of this questionnaire was conducted by Crocker et al. (1997) [17] and the validity study was conducted by Kowalski et al. (1997) [17] and the Cronbach alpha reliability coefficient was found to be 0.80. Erdim et al. (2012) [18] reported the psychometric analysis data of the form and reported Cronbach’s alpha coefficient as 0.86. CPAQ was developed to assess the physical activity levels of children aged 8–14 years. The form is filled in by the child. The form consists of 10 questions and the physical activity status of the child in the last seven days is questioned with the form. Except for the question questioning the disease status of the questionnaire, each item is evaluated as 5 points and an activity score between 1 and 5 is found. ‘1’ indicates low physical activity and ‘5’ indicates high physical activity.
Anthropometric measurements
Body weight, height, waist circumference and upper middle arm circumference of the students were measured and recorded by the researcher in accordance with the technique [19, 20].
Body weight
Weight was recorded in kilograms and measured using a digital scale with the students standing with bare feet, arms at their sides, wearing light clothing [20].
Height
It was measured using a portable stadiometer (Seca 213) (accurate to 0.1 cm), without shoes, in an orthostatic position, arms at the side, feet together, knees straight, head aligned along the Frankfort horizontal plane [20].
Body mass index (BMI)
It is calculated by dividing body weight in kilograms by the square of height in metres [20]. The evaluation was conducted using the BMI percentile table according to age recommended by the World Health Organization (WHO-2007) for children and adolescents aged 5–19 years, through the WHO AnthroPlus software. Accordingly, individuals with BMI values between the 5th and 15th percentiles were classified as “underweight,” those between the 15th and 85th percentiles as having “normal weight,” those between the 85th and 95th percentiles as “overweight,” and those at or above the 95th percentile as “obese” [21].
Upper mid arm circumference measurement
The measurement is performed with the participant standing and in the anatomical position. The upper extremity is flexed 90° at the elbow joint. The midpoint of the distance between the acromial process at the shoulder and the olecranon process at the elbow is determined as the anatomical reference and circumference measurement is taken from this point with an inflexible tape measure [20].
Waist circumference (cm)
Measurements were taken with a non-stretch tape measure at the level of the navel while the children were in an upright position with their stomachs relaxed [22].
Intervention
After the pre-test was completed, the students and parents included in the intervention group were given “nutrition education” by the dieticians in the study team. Nutrition education was given to the students for eight weeks, one lesson hour (35 min) per week in the school during school hours, and to the parents face-to-face in the school conference room for one hour per week for four weeks. The intervention lasted eight weeks for students and four weeks for parents. Student sessions were delivered once a week during a regular class hour. Considering that children have shorter attention spans and that learning retention requires reinforcement over time, delivering the training over an eight-week period was deemed appropriate to strengthen knowledge and behavioural change. Parent sessions were condensed into four weekly meetings focusing on reinforcing the same themes at home, as their participation opportunities within the school calendar were more limited. The nutrition education tools given to children consisted of an 8-week education programme, brochure and activities such as educational games and puzzles developed to reinforce the information [11, 23, 24]. Visual presentations and brochures were used in the nutrition education given to parents. Since parents play an important role in shaping the child’s eating habits, it was deemed appropriate to provide nutrition education to students and parents [23]. During the study period, the control group did not receive any nutrition education, educational materials, or intervention-related content. Students in the control group continued their normal school curriculum without any additional contact from the research team. To minimise potential contamination between groups, the intervention and control classes were located in separate classrooms, and none of the intervention materials were made accessible in shared school areas. The research team entered control classrooms only for data collection sessions, and classroom teachers were instructed not to share intervention-related content with control group students. To ensure ethical fairness, the full nutrition education programme was provided to the control group after the completion of the study.
The nutrition education for children in this study was delivered through computer-aided visual presentations and supported by reinforcement activities. The content was designed to be age-appropriate, engaging, and participatory, aiming not only to increase nutrition knowledge but also to promote positive attitudes and behaviours. During and after the educational sessions, students were encouraged to ask questions related to the topic to ensure interactive participation and strengthen learning. The educational programme provided to children and their parents is presented in Fig. 2. All intervention sessions were conducted by one instructor, a Ph.D. faculty member (Assistant Professor) in Nutrition and Dietetics, with prior experience in school-based nutrition education. The programme was structured in accordance with the Türkiye Nutrition Guide (TUBER-2022) [23]. Standardised weekly lesson plans and presentation materials were used throughout the sessions to ensure content consistency. These standardised materials were designed to provide all participants with the same content and conditions, aiming to ensure consistency and comparability during the assessment process. The intervention was implemented during regular class hours, which enabled full attendance of students. As shown in Fig. 1, 32 of the 35 students in the intervention group and all 35 students in the control group completed the study. Parent sessions also achieved full participation. Intervention fidelity was ensured through a comprehensive set of adherence and quality control procedures. All educational sessions for students and parents were delivered by the same PhD-level dietitian using standardized weekly lesson plans, scripted explanations, and identical visual materials. The instructor strictly followed the weekly curriculum, documenting the content and duration of each session. Lessons were conducted interactively under the supervision of both the instructor and classroom teachers, and student engagement was monitored through teacher checklists to assess the extent to which the education was received. To prevent contamination, intervention and control classes were located in separate classrooms, and no educational materials were displayed in shared school areas. Students were encouraged to participate actively through questions, discussions, visual aids, educational games, and short reflective activities, all of which supported reinforcement and contributed to consistently high levels of participation. No deviations from the planned curriculum were reported.
Fig. 2.
Nutrition education programmes for children and parents
The effectiveness of nutrition education was evaluated in three stages: pretest, posttest and follow-up test by using nutrition knowledge test, nutrition attitude scale, nutrition behaviour scale, food consumption frequency form, child physical activity questionnaire and anthropometric measurements. Follow-up assessments were conducted only in the intervention group. The main objective of this assessment was to examine whether the effects of the training were maintained over time. The follow-up measurement aimed to evaluate the long-term retention of nutrition knowledge, attitudes, and behavioural changes among students who received the educational intervention. Since the control group did not participate in any training activities during the study period, conducting a follow-up in that group would not have yielded meaningful information regarding programme sustainability.
Statistical analysis
Data were analyzed using SPSS 26.0 (IBM Corp., Armonk, NY, USA). Descriptive statistics including mean, standard deviation, median (min–max), and percentages were used to summarise the data. For normally distributed variables, paired sample t-tests were used to compare pre- and post-intervention values within groups, and independent t-tests were used for between-group comparisons. For variables not normally distributed, the Wilcoxon signed-rank test was used for within-group comparisons and the Mann–Whitney U test for between-group comparisons. Chi-square or Fisher’s exact test was applied for categorical data. In addition, to control for potential baseline differences between the intervention and control groups, Analysis of Covariance (ANCOVA) was conducted using post-test scores as dependent variables, group assignment as the fixed factor, and baseline (pre-test) scores as covariates. Partial eta squared (ηp²) values were calculated to evaluate effect sizes. Statistical significance was set at p < .05.
Ethical approval
Ethical approval of the study was obtained from Mardin Artuklu University Scientific Research and Publication Ethics (Date: 11.07.2024, Decision No: 7). In addition, the study was conducted within the framework of the Declaration of Helsinki Principles and informed consent and approval were obtained from the participants. Necessary institutional permission was obtained from Mardin Artuklu District Directorate of National Education for the conduct of the study. Within the scope of the study, parents and students were informed about the study, and participants who volunteered to participate in the study and signed the Informed Parental Consent Form were included in the study. To ensure institutional anonymity, the name of the participating school is withheld in accordance with the confidentiality requirements of the ethics approval. The protocol of the study was registered at www.clinicaltrials.gov (NCT07168928, Registered 11 September 2025).
Results
Table 1 presents the sociodemographic characteristics. The study included 32 intervention and 35 control group students. Mean ages were similar (10.25 ± 0.44 vs. 10.14 ± 0.35). Gender distribution and family structure were comparable between groups (p > .05). However, significant differences were found in parental education: more mothers and fathers in the control group had higher education levels (p = .017 and p = .044, respectively). Employment status showed no significant differences.
Table 1.
Comparison of the Socio-demographic characteristics of the students in the intervention and control groups
| Variable | Category | Control group (n = 35) | Intervention group (n = 32) | p | ||
|---|---|---|---|---|---|---|
| Age (year) X̅±SD (Min-Max) | 10.14 ± 0.35 (10–11) | 10.25 ± 0.44 (10–11) | ||||
| n | % | n | % | |||
| Gender | Boy | 17 | 48.6 | 17 | 53.1 | 0.710 |
| Girl | 18 | 51.4 | 15 | 46.9 | ||
| Family | Nuclear Family | 33 | 94.3 | 30 | 93.8 | 0.609 |
| Large Family | 2 | 5.7 | 2 | 6.3 | ||
| Mother’s Education Status | Illiterate | 3 | 8.6 | 4 | 12.5 | 0.017 |
| Primary | 7 | 20 | 16 | 50 | ||
| High School | 10 | 28.6 | 8 | 25 | ||
| Undergraduate and higher | 15 | 42.9 | 4 | 12.5 | ||
| Mother Employment Status | Not-employed | 24 | 68.6 | 27 | 84.4 | 0.130 |
| Employed | 11 | 31.4 | 5 | 15.6 | ||
| Father’s Education Status | Illiterate | 1 | 2.9 | 3 | 9.4 | 0.044 |
| Primary | 4 | 11.4 | 11 | 34.4 | ||
| High School | 9 | 25.7 | 8 | 25 | ||
| Undergraduate and higher | 21 | 60 | 10 | 31.3 | ||
| Father Employement Status | Not-employed | 4 | 11.4 | 2 | 6.3 | 0.458 |
| Employed | 31 | 88.6 | 30 | 93.8 | ||
X̅ mean, SD Standart deviation, Min minimum, Max maximum, χ2test was used in statistical analyses, p < .05
Findings related to eating habits are given in Table 2. In the control group, significant decreases were observed in breakfast, lunch and dinner consumption. Especially in breakfast consumption, the rate of 65.7% in the pre-test decreased to 62.9% in the post-test and this change was found to be statistically significant (χ²=6.818, p = .009). Similarly, significant decreases were found in lunch (χ²=3.988, p = .046) and dinner (χ²=5.833, p = .016) consumption rates. No statistically significant change was observed in the consumption rates of breakfast (χ²=0.006, p = .938), lunch (χ²=0.286, p = .593) and dinner (χ²=0.069, p = .793) in the intervention group. In the follow-up test after the intervention, breakfast (78.1%), lunch (71.9%) and dinner (93.8%) rates were maintained at a high level.
Table 2.
Comparison of regular meal consumption between groups
| Control group n (%) |
Intervention group n (%) |
|
|---|---|---|
| Regularly consumed meals | ||
| Breakfast | ||
| Pre-test | 23 (65.7) | 26 (81.3) |
| Post-test | 22 (62.9) | 27 (84.4) |
| χ² / p (pre-test – post test) | χ²=6.818,p = .009 | χ²=0.006, p = .938 |
| Follow-up test | - | 25 (78.1) |
| Lunch | ||
| Pre-test | 20 (57.1) | 25 (78.1) |
| Post-test | 11 (31.4) | 21 (65.6) |
| χ² / p (pre-test – post test) | χ²=3.988,p = .046 | χ²=0.286, p = .593 |
| Follow-up test | - | 23 (71.9) |
| Dinner | ||
| Pre-test | 28 (80.0) | 31 (96.9) |
| Post-test | 27 (77.1) | 30 (93.8) |
| χ² / p (pre-test – post test) | χ²=5.833,p = .016 | χ²=0.069, p = .793 |
| Follow-up test | - | 30 (93.8) |
n (%) Number (Percentage), χ2test was used in statistical analyses, p < .05
Table 3 presents the findings related to anthropometric measurements of the individuals. A statistically significant increase in height and body weight was observed in the intervention group (p < .001). In intergroup comparisons, no significant difference was found in terms of both height and body weight (p > .05). When upper arm circumference measurements were analysed, both intra- and inter-group differences were not significant (p > .05). A significant decrease in waist circumference measurements was observed in the intervention group (p = .023). In the evaluation made in terms of BMI percentile groups, a significant change was observed in the intervention group. In the intervention group, there was an increase in the proportion of children in the ‘normal’ category in the pre-test and post-test comparison, whereas there was a decrease in the “obese” and ‘overweight’ groups. While significant differences were found for both groups in intra-group analyses (p < .001), differences in BMI percentile groups were not statistically significant in inter-group comparisons (p > .05).
Table 3.
Comparison of anthropometric measurements and BMI percentile classifications of students in the intervention and control groups
| Control group | Intervention group | Between Groups t/p | ||
|---|---|---|---|---|
| Variable | Time | X̅±SD | X̅±SD | |
| Height length (cm) | Pre test | 137.1 ± 6.1 | 134.9 ± 6.6 |
t = 1.461, p = .149 t = 1.052, p = .297 |
| Post test | 138.0 ± 6.0 | 136.4 ± 6.6 | ||
| Within group | t= -4.900,p < .001 | t= -17.278,p < .001 | ||
| Body weight (kg) | Pre test | 34.9 ± 9.2 | 32.4 ± 8.1 |
t = 1.195, p = .236 t = 0.878, p = .383 |
| Post test | 34.9 ± 9.2 | 33.1 ± 7.7 | ||
| Within group | t= -4.089,p < .001 | t= -5.013,p < .001 | ||
| Upper arm circumference (cm) | Pre test | 21.7 ± 3.8 | 20.1 ± 3.3 |
t = 1.864, p = .067 t = 0.186, p = .853 |
| Post test | 21.2 ± 3.5 | 20.2 ± 3.6 | ||
| Within group |
t = 1.616 p = .115 |
t = 0.131 p = .897 |
||
| Waist circumference (cm) | Pre test | 65.9 ± 10.6 | 62.0 ± 9.9 |
t = 1.574, p = .125 t = 2.315, p = .023 |
| Post test | 66.3 ± 10.5 | 60.6 ± 9.3 | ||
| Within group |
t = 1.555 p = .068 |
t = 2.330 p = .023 |
||
| n (%) | n (%) | |||
| BMI Persentile Groups | Pre test | p = .744 | ||
| Obese | 10 (28.6) | 6 (18.8) | ||
| Overweight | 3 (8.6) | 4 (12.5) | ||
| Normal | 20 (57.1) | 19 (59.4) | ||
| Underweight | 2 (5.7) | 3 (9.4) | ||
| Post test | p = .389 | |||
| Obese | 12 (34.3) | 6 (18.8) | ||
| Overweight | 3 (8.6) | 3 (9.4) | ||
| Normal | 19 (54.3) | 20 (62.5) | ||
| Underweight | 1 (2.9) | 3 (9.4) | ||
| Within group | p < .001 | p < .001 | ||
Intragroup comparisons were analysed by paired sample t test
X̅ Mean, SD Standard Deviation, t Independent Groups t test, p < .05
Table 4 presents the descriptive scores of the NKT, NAS, NBS and the CPAQ for both groups. While unadjusted pre- and post-test means are reported in the table for descriptive purposes, all inferential interpretations are based on ANCOVA-adjusted results.
Table 4.
Evaluation of NKT, NAS, NBS and CPAQ scores of students in the intervention and control groups
| Control group | Intervention group | Between Groups t/p | ANCOVA (baseline-adjusted) |
||
|---|---|---|---|---|---|
| Variables | Time | X̅±SD | X̅±SD | F(1,64) = 7.83, p = .007, ηp²=0.109; adj. M ± SE: Int 12.12 ± 0.22, Ctrl 11.23 ± 0.20 | |
| Nutrition Knowledge Test (NKT) | Pre test | 12.03 ± 1.04 | 10.44 ± 1.52 | t = 5.027,p < .001 | |
| Post test | 11.57 ± 1.09 | 11.75 ± 1.39 | t=-0.587, p = .559 | ||
| Follow-up test | - | 11.34 ± 1.38 | - | ||
| Within group | t = 21.668,p < .001 | F (2, 62) = 10.87, p < .001 | |||
| Nutrition Attitude Scale (NAS) | Pre test | 13.89 ± 1.66 | 13.56 ± 1.72 | t = 0.782, p = .437 | F(1,64) = 13.17,p < 0.001, ηp²=0.171; adj. M ± SE: Int 14.82 ± 0.18, Ctrl 13.90 ± 0.18” |
| Post test | 13.97 ± 1.71 | 14.72 ± 1.11 | t = − 2.101,p = .040 | ||
| Follow-up test | - | 14.25 ± 1.50 | - | ||
| Within group | t = − 0.442, p = .661 | F (2, 62) = 8.68, p < .001 | |||
| Nutrition Behaviour Scale (NBS) | Pre test | 6.29 ± 4.66 | 2.47 ± 4.10 | t = 3.546, p < .001 | F(1,64) = 3.47, p = .067, ηp²=0.051; adj. M ± SE: Int 7.65 ± 0.88, Ctrl 5.29 ± 0.84 |
| Post test | 6.51 ± 5.91 | 7.28 ± 4.37 | t = − 0.599, p = .551 | ||
| Follow-up test | - | 4.94 ± 5.59 | - | ||
| Within group | t = − 0.289, p = .774 | F (2, 62) = 14.27, p < .001 | |||
| Child Physical Activity Questionnaire (CPAQ) | Pre test | 3.27 ± 0.55 | 3.09 ± 0.51 | t = 1.391, p = .169 | F(1,64) = 3.39, p = .070, ηp²=0.050; adj. M ± SE: Int 3.51 ± 0.09, Ctrl 3.28 ± 0.08 |
| Post test | 3.36 ± 0.71 | 3.52 ± 0.75 | t=–0.871, p = .387 | ||
| Follow-up test | - | 3.49 ± 0.74 | - | ||
| Within group | t= − 1.090, p = .283 | F (2, 62) = 5.35,p = .007 | |||
SD Standard Deviation, t: Independent Groups t-test, p: Significance level, ANCOVA Analysis of Covariance (baseline-adjusted), ηp² Partial Eta Squared, adj. M Adjusted Mean, SE Standard Error, Int Intervention group, Ctrl Control group, p < .05
Between-group comparisons at post-test, adjusted for baseline values, indicated significantly higher NKT scores in the intervention group compared with the control group (F(1,64) = 7.83, p = .007, ηp² = 0.109). A similar pattern was observed for the NAS, where adjusted post-test scores were significantly higher in the intervention group (F(1,64) = 13.17, p < .001, ηp² = 0.171). For the NBS, the analysis suggested a moderate but non-significant trend favouring the intervention group (F(1,64) = 3.47, p = .067, ηp² = 0.051). The CPAQ also demonstrated a borderline effect (F(1,64) = 3.39, p = .070, ηp² = 0.050).
Within-group patterns over time, shown descriptively in Table 4, indicated increases in knowledge, attitudes, behaviours and physical activity among intervention students, with some decline at follow-up but still above baseline levels. However, these patterns should be interpreted cautiously, as between-group contrasts rely on ANCOVA-adjusted post-test comparisons.
According to the results of the correlation analysis between the post-test data of the intervention group and the variables, it was found that there was a positive and highly significant relationship between height and body weight (r = .779, p < .001). There was also a positive and moderately significant relationship between height and waist circumference (r = .494, p = .004). The correlation between body weight and waist circumference was very strong and significant (r = .877, p < .001). A moderate positive and significant correlation was found between CPAQ scores and NKT scores (r = .421, p = .016). Likewise, there was a significant positive correlation between the CPAQ scores and the NAS scores (r = .395, p = .025). The relationship between the NAS and the NBS was also significant (r = .390, p = .027). However, no significant correlation was found between height and the NKT, NBS, CPAQ and NAS (p > .05). Similarly, no significant correlations were observed between body weight and waist circumference and these three scales (Table 5).
Table 5.
Correlation analysis between anthropometric measurements (height, body weight, waist circumference) and scale scores (NKT, NBS, NAS, and CPAQ) among students in the intervention group at post-test
| Variables | 1 Height length | 2 Body weight | 3 Waist circumference | 4 NKT Total | 5 NBS Total | 6 CPAQ | 7 NAS Total |
|---|---|---|---|---|---|---|---|
| 1 Height length | — | ||||||
| 2 Body weight | r = .779 | — | |||||
| p < .001 | |||||||
| 3 Waist circumference | r = .494 | r = .877 | — | ||||
| p = .004 | p < .001 | ||||||
| 4 NKT Total | r = –.009 | r = –.006 | r = –.105 | — | |||
| p = .961 | p = .973 | p = .568 | |||||
| 5 NBS Total | r = .085 | r = .101 | r = .172 | r = .314 | — | ||
| p = .645 | p = .584 | p = .346 | p = .080 | ||||
| 6 CPAQ | r = –.273 | r = –.355 | r = –.370 | r = .421 | r = .314 | — | |
| p = .131 | p = .046 | p = .037 | p = .016 | p = .080 | |||
| 7 NAS Total | r = –.029 | r = –.074 | r = –.063 | r = –.053 | r = .390 | r = .395 | — |
| p = .874 | p = .686 | p = .731 | p = .773 | p = .027 | p = .025 |
r Pearson correlation coefficient, NKT Nutrition Knowledge Test, NBS Nutrition Behaviour Scale, CPAQ Child Physical Activity Questionnaire, NAS Nutrition Attitude Scale, p < .05
According to multiple linear regression analysis, approximately 28.4% of the variance (R² = 0.284, Adj. R² = 0.234; F (2, 29) = 5.739, p = .008) was explained by the variables of waist circumference and total score of feeding behaviour (NBS - post test). The results of the analyses showed that waist circumference made a negative and significant contribution to the regression model (β = -0.437, p = .010). This indicates that children’s physical activity scores decreased significantly as waist circumference increased. The total score of the NBS is a positive and significant predictor (β = 0.389, p = .021) (Table 6).
Table 6.
Multiple linear regression analysis examining the predictors of CPAQ scores among students in the intervention group
| Model | Variable | B (Unstd.) | SE | β (Std.) | t | R² | p | 95% CI (Min-Max) |
|---|---|---|---|---|---|---|---|---|
| M₀ | (Intercept) | 3.508 | 0.132 | — | 26.577 | < .001 | [3.239, 3.777] | |
| M₁ | (Intercept) | 5.154 | 0.778 | — | 6.628 | < .001 | [3.563, 6.744] | |
| M₁ | Waist circumference | -0.035 | 0.013 | -0.437 | -2.737 | 0.137 | .010 | [-0.061, -0.009] |
| M₁ | NBS | 0.066 | 0.027 | 0.389 | 2.437 | 0.098 | .021 | [0.011, 0.122] |
CPAQ Child Physical Activity Questionnaire, B Unstandardized Coefficient, SE Standard Error, β Standardized Coefficient, t Independent Groups, R2Coefficient of determination, t test, 95% Confidence Interval, p < .05
Discussion
This study, a randomized controlled trial evaluating the effects of structured school-based nutrition education on children’s nutrition knowledge, attitudes, behaviors, and anthropometric measurements, has revealed important findings. Pre-test results showed that a significant proportion of students consumed irregular meals, had unhealthy eating habits, and that 34.3% were overweight/obese and 7.5% were underweight. The significant improvements achieved after the intervention suggest that structured school-based nutrition education can be effective on multidimensional nutrition outcomes and are consistent with previous studies reporting that school-based interventions involving family participation yield stronger results [12, 25].
The significant increases observed in the intervention group are consistent with the literature indicating that repeated and structured training enhances cognitive learning [4, 9]. Visual materials, group discussions, and interactive activities may have enabled students to better encode nutrition concepts. Additionally, parent education sessions may have reinforced nutrition messages in the home environment and strengthened attitude change. This finding parallels the literature showing that family involvement increases behavioral change [12].
The slight decline in knowledge and attitude scores in follow-up tests can be explained by the end of external reinforcement for education. This is consistent with the “knowledge loss” effect commonly seen in educational interventions [26, 27]. Therefore, it is clear that regular reinforcement sessions are necessary for long-term behavioral change.
It was found that 73.1% of students ate breakfast daily, which falls within the range reported in the literature (42–92%) [12, 26]. Since breakfast is critical for metabolic regulation and cognitive performance, it is important that interventions focus on sustaining this behavior.
However, it is noteworthy that lunch is the most frequently skipped meal and that intervention has limited effect on this behavior. This situation can be explained by structural factors such as short recess periods at school, insufficient food options, or a restricted meal environment. The literature emphasizes that such behavioral changes cannot be achieved through education alone, but also require policy-level measures such as regulating school meal times or improving food options [28, 29]. Furthermore, it is reported that education must be long-term and integrated into the curriculum for sustainable change [30–32].
The significant decrease in BMI percentile and waist circumference observed in the intervention group is consistent with previous studies showing that school-based nutrition education may have positive effects on body composition [33]. The faster change in waist circumference compared to BMI may be due to abdominal fat being a sensitive indicator in the short term [4].
Additionally, the positive correlation between physical activity and dietary behaviors is consistent with studies indicating that physical activity and dietary behaviors reinforce each other [34]. The fact that multi-component interventions show greater anthropometric effects supports this mechanism.
The decline in follow-up test scores can be explained by a decrease in motivation over time and the loss of external reinforcement. This situation is frequently reported in the literature, and “behavioral regression” is an expected process when reinforcement is not renewed [9, 26].
The fact that some differences between the groups did not completely disappear after the intervention may be related to initial sociodemographic differences. Higher parental education levels in the control group may positively influence children’s behavior by providing a more regular eating environment at home. Similarly, lower baseline NKT and NBS scores may have created more room for improvement in the intervention group. Although ANCOVA statistically controlled for these differences, it cannot be expected to eliminate them entirely.
The results indicate that school-based nutrition education, especially when combined with parental support, can improve children’s nutrition knowledge, attitudes, and behaviors. These findings are consistent with international evidence indicating that school-based interventions can play a critical role in reducing childhood obesity [1, 2, 10].
However, it is clear that structural barriers must also be addressed. Extending lunch periods, improving school meal services, and implementing long-term nutrition education policies integrated into the curriculum are essential for sustainable impact.
Strengths and limitations
This study has several important strengths. The randomized controlled design allowed for robust inferences regarding causal relationships. The use of multidimensional assessment tools (NKT, NAS, NBS, CPAQ) and the collection of anthropometric measurements using standard procedures increased methodological reliability. Implementing the education program during school hours strengthened the ecological validity of the intervention. Furthermore, the inclusion of parent sessions potentially created a broader impact on behavior change by supporting social reinforcement in the family context.
However, the study has some limitations. First, the NKT was developed by the researchers and has not undergone validity–reliability analyses; this may limit the psychometric sensitivity of the measurements. Second, the sample size is relatively small, which may reduce the generalizability of the findings. Third, as the intervention was limited to eight weeks, it was not possible to assess whether longer-term behavioral effects would emerge. Furthermore, interaction and knowledge sharing among students could not be fully controlled; this may have created a diluting effect, particularly in studies conducted in the same school setting.
Although multiple procedures were used to minimize recall and social desirability bias (e.g., anonymous administration, standardized instructions, supervision without guidance), a small degree of bias cannot be fully excluded due to the self-reported nature of the data.
Another important limitation of the study is that follow-up measurements were only taken from the intervention group. Therefore, it was not possible to directly compare changes over time with the control group, and conclusions regarding sustainability should be interpreted with caution. While the continued improvements observed in knowledge, attitudes, behaviors, and anthropometric indicators in the intervention group are promising, the lack of follow-up data in the control group prevents the clear identification of long-term group differences.
Despite all these limitations, the study provides valuable preliminary evidence regarding the feasibility and potential effects of school-based nutrition education. Future research should include larger samples with different sociodemographic groups, long-term follow-up measurements, and multi-component intervention models that encompass both educational and environmental components.
Conclusions
This randomized controlled study suggests that a structured, dietitian-led, school-based nutrition education programme may positively influence primary school students’ nutrition knowledge, attitudes, and behaviours, and support modest improvements in anthropometric outcomes such as waist circumference. Based on the findings, several practical recommendations can be offered to enhance the delivery of school-based nutrition programmes. Schools may integrate structured, age-appropriate nutrition education modules developed and delivered by trained dietitians into their existing curriculum, ensuring weekly reinforcement activities such as short discussions, interactive games, or visual reminders. Collaboration between schools and families can be strengthened through parent-focused education sessions, the distribution of culturally adapted educational materials, and the use of simple home-school communication tools (e.g., take-home activity sheets or weekly goal-setting forms). Child nutrition practitioners can localise intervention content by aligning examples, food models, and activities with regional eating habits, food availability, and socio-cultural characteristics. Implementing routine monitoring systems—such as brief monthly follow-ups, teacher checklists, or classroom observations—may also help maintain long-term behaviour change. Together, these practices can guide schools and practitioners in designing more effective, context-sensitive, and sustainable nutrition education programmes. It should be noted that follow-up data were available only for the intervention group; therefore, statements regarding sustained improvements over time should be interpreted cautiously, as the absence of follow-up assessments in the control group limits firm conclusions about long-term between-group effects. To more clearly determine the durability and broader applicability of the programme’s impact, future studies—preferably large-scale, longitudinal, and incorporating follow-up measurements in both groups—are needed to evaluate the sustainability and scalability of dietitian-based nutrition education interventions.
Supplementary Information
Abbreviations
- ANCOVA
Analysis of covariance
- BMI
Body mass index
- CI
Confidence interval
- CPAQ
Child physical activity questionnaire
- NAS
Nutrition attitude scale
- NBS
Nutrition behaviour scale
- NKT
Nutrition knowledge test
- TUBER
Türkiye nutrition guide
- UN
United nations
- WHO
World health organization
Authors’ contributions
Conceptualization, Ç.A., H.K., and N.A.; methodology, Ç.A., H.K., and N.A.; software, Ç.A.; formal analysis, Ç.A.; investigation, Ç.A., H.K. and N.A.; data check, Ç.A. and H.K.; writing—original draft preparation, Ç.A., H.K., and N.A.; writing—review and editing, Ç.A., H.K., and N.A. All authors have read and agreed to the published version of the manuscript.
Funding
This study was supported by Mardin Artuklu University Scientific Research Projects (Project No: MAÜ.BAP.24.SBF.034).
Data availability
The data supporting the findings of this study are available from the corresponding author upon reasonable request. Due to ethical and confidentiality considerations, the data cannot be made publicly available but can be shared in anonymised form for research purposes upon request (Author 1).
Declarations
Ethics approval and consent to participate
Ethical approval of the study was obtained from Mardin Artuklu University Scientific Research and Publication Ethics (Date: 11.07.2024, Decision No:7). In addition, the study was conducted within the framework of the Declaration of Helsinki Principles and informed consent and approval were obtained from the participants. Necessary institutional permission was obtained from Mardin Artuklu District Directorate of National Education for the conduct of the study. Within the scope of the study, parents and students were informed about the study, and participants who volunteered to participate in the study and signed the Informed Parental Consent Form were included in the study.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Bogataj Š, Trajković N, Cadenas-Sanchez C, Sember V. Effects of school-based exercise and nutrition intervention on body composition and physical fitness in overweight adolescent girls. Nutrients. 2021;13(1):238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Carducci B, Oh C, Keats EC, Roth DE, Bhutta ZA. Effect of food environment interventions on anthropometric outcomes in school-aged children and adolescents in low- and middle-income countries: a systematic review and meta-analysis. Curr Developments Nutr. 2020;4(7):nzaa098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.World Health Organization. Report of the Commission on Ending Childhood Obesity. 2017. Available from: https://iris.who.int/bitstream/handle/10665/259349/WHO-NMH-PND-ECHO-17.1-eng.pdf. Accessed 19 Jan 2025.
- 4.Vasquez-Mamani LN, Cueva-Calizaya LA, Gálvez-Díaz NDC, Saintila J, Calizaya-Milla YE. Effect of a nutrition education program on knowledge, attitudes, dietary practices, anthropometry and hemoglobin in Peruvian adolescents. J Multidisciplinary Healthc. 2025;18:1679–93. [DOI] [PMC free article] [PubMed]
- 5.Republic of Türkiye Ministry of Health. Türkiye Çocukluk Çağı Obezite Araştırması (COSI-TUR) 2022 Raporu. 2023. Available from: https://hsgm.saglik.gov.tr/depo/. Accessed 10 Jan 2025.
- 6.Clark H, Coll-Seck AM, Banerjee A, et al. A future for the world’s children? A WHO–UNICEF–Lancet commission. Lancet. 2020;395(10224):605–58. [DOI] [PubMed] [Google Scholar]
- 7.Abate KH, Belachew T. Chronic malnutrition among under-five children of Ethiopia May not be economic: a systematic review and meta-analysis. Ethiop J Health Sci. 2019;29(2):123–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Pal A, Pari AK, Sinha A, Dhara PC. Prevalence of undernutrition and associated factors: a cross-sectional study among rural adolescents in West Bengal, India. Int J Pediatr Adolesc Med. 2017;4(1):9–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Raut S, Kc D, Singh DR, Dhungana RR, Pradhan PMS, Sunuwar DR. Effect of nutrition education intervention on nutrition knowledge, attitude, and diet quality among school-going adolescents: a quasi-experimental study. BMC Nutr. 2024;10(1):35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Dargie F, Henry CJ, Hailemariam H, Regassa N. A peer-led pulse-based nutrition education intervention improved school-aged children’s knowledge, attitude, practice and nutritional status in Southern Ethiopia. J Food Res. 2018;7(3):38–51. [Google Scholar]
- 11.Oosthuizen D, Oldewage-Theron WH, Napier C. The impact of a nutrition programme on the dietary intake patterns of primary school children. South Afr J Clin Nutr. 2011;24(2):75–81. [Google Scholar]
- 12.Aktaç Ş, Kızıltan G, Avcı S. The effect of family participation in nutrition education intervention on the nutritional status of preschool-age children. Eğitim Ve Bilim. 2019;44(199):415–31.
- 13.Arvidson CR. Children’s cardiovascular health promotion attitude scale: an instrument development. Texas Woman’s University; 1990.
- 14.Haney MÖ, Bahar Z. Çocuk Kalp sağlığını geliştirme tutum ölçeği’nin geçerlik ve güvenirliği. Dokuz Eylül Üniversitesi Hemşirelik Fakültesi Elektronik Dergisi. 2014;7(2):92–7. [Google Scholar]
- 15.Edmundson E, Parcel GS, Perry CL, et al. The effects of the child and adolescent trial for cardiovascular health intervention on psychosocial determinants of cardiovascular disease risk behavior among third-grade students. Am J Health Promotion. 1996;10(3):217–25. [DOI] [PubMed] [Google Scholar]
- 16.Öztürk M. ES. Investigation of children’s eating habits according to the health behavior interaction model. İstanbul Üniversitesi Sağlık Bilimleri Enstitüsü Doktora Tezi; 2010.
- 17.Crocker PR, Bailey DA, Faulkner RA, Kowalski KC, McGrath R. Measuring general levels of physical activity: preliminary evidence for the physical activity questionnaire for older children. Med Sci Sports Exerc. 1997;29(10):1344–9. [DOI] [PubMed] [Google Scholar]
- 18.Kowalski KC, Crocker PRE, Faulkner RA. Validation of the physical activity questionnaire for older children. Pediatr Exerc Sci. 1997;9(2):174–86. [Google Scholar]
- 19.McCarthy HD, Jarrett KV, Crawley HF. The development of waist circumference percentiles in British children aged 5.0–16.9 years. Eur J Clin Nutr. 2001;55(10):902–7. [DOI] [PubMed] [Google Scholar]
- 20.Pekcan G. Determination of nutritional status. In: Diet Handbook. 2008:67–141.
- 21.World Health Organization. Growth reference data for 5–19 years. 2007.
- 22.Hatipoğlu N, Öztürk A, Mazicioğlu M, Kurtoğlu S, Seyhan S, Lokoğlu F. Waist circumference percentiles for 7–17-year-old Turkish children and adolescents. Eur J Pediatrics. 2007;166(10):973–8. [DOI] [PubMed] [Google Scholar]
- 23.TUBER. Türkiye Nutrition Guide 2022 (TUBER). Ministry of Health. 2022.
- 24.Türk M, GŞT EI. Nutritional habits of first-year high school students in an urban area. J Gen Med. 2007;17(2):81–7. [Google Scholar]
- 25.Antwi J, Ohemeng A, Boateng L, Quaidoo E, Bannerman B. Primary school-based nutrition education intervention on nutrition knowledge, attitude and practices among school-age children in Ghana. Global Health Promotion. 2020;27(4):114–22. [DOI] [PubMed] [Google Scholar]
- 26.Moss A, Smith S, Null D, Long Roth S, Tragoudas U. Farm to school and nutrition education: positively affecting elementary school-aged children’s nutrition knowledge and consumption behavior. Child Obes. 2013;9(1):51–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lytle LA, Jacobs DR Jr, Perry CL, Klepp KI. Achieving physiological change in school-based intervention trials: what makes a preventive intervention successful? Br J Nutr. 2002;88(3):219–21. [DOI] [PubMed] [Google Scholar]
- 28.Pongutta S, Ajetunmobi O, Davey C, Ferguson E, Lin L. Impacts of school nutrition interventions on the nutritional status of school-aged children in asia: a systematic review and meta-analysis. Nutrients. 2022;14(3):589. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Chamberland K, Sanchez M, Panahi S, Provencher V, Gagnon J, Drapeau V. The impact of an innovative web-based school nutrition intervention to increase fruits and vegetables and milk and alternatives in adolescents: a clustered randomized trial. Int J Behav Nutr Phys Activity. 2017;14:1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Franciscato SJ, Janson G, Machado R, Lauris JRP, Andrade SMJD, Fisberg M. Impact of the nutrition education program Nutriamigos on levels of awareness on healthy eating habits in school-aged children. J Hum Growth Dev. 2019;29(3):390–402. [Google Scholar]
- 31.Kendel Jovanović G, Janković S, Pavičić Žeželj S. The effect of nutritional and lifestyle education intervention program on nutrition knowledge, diet quality, lifestyle, and nutritional status of Croatian school children. Front Sustainable Food Syst. 2023;7:1019849. [Google Scholar]
- 32.Selmin K, Yıldırım G. The effect of a nutrition education program on nutrition behavior and body mass index of secondary school students. Int J Caring Sci. 2020;13(1).
- 33.Silveira JA, Taddei JA, Guerra PH, Nobre MR. Effectiveness of school-based nutrition education interventions to prevent and reduce excessive weight gain in children and adolescents: a systematic review. Jornal De Pediatria. 2011;87:382–92. [DOI] [PubMed] [Google Scholar]
- 34.Dewi RS, Fikawati S, Syafiq A. The effect of school-based nutrition education combined with physical activity on overweight and obesity in children. Malaysian J Nutr. 2019;25(1):113–23. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data supporting the findings of this study are available from the corresponding author upon reasonable request. Due to ethical and confidentiality considerations, the data cannot be made publicly available but can be shared in anonymised form for research purposes upon request (Author 1).