Effect of a Nutritional Education Intervention in Female Athletes

1. Materials and Methods 2.1. Study Design This study was conducted as part of the Assessment of Training in Healthy Habits and Nutritional Education for Adolescent Athletes (ATHENEA) project, which received ethical approval from the Bioethics Committee of the University of Barcelona (initial protocol: IRB00003099, June 30, 2023; modified protocol: IRB00003099, October 14, 2025). A parallel-group cluster randomized controlled trial (RCT) in sports club settings was implemented to evaluate a three-week nutritional education intervention targeting adolescent female handball players in Catalonia, Spain.

Prior to the development of the main study, a pilot project was conducted between September 2023 and October 2024, which allowed for the refinement of the intervention protocol and the optimization of both the intervention procedures and the data collection strategies.

The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. All participants, as well as their parents and coaches, received a comprehensive description of the study, both orally and in writing. Participation was voluntary, and written informed consent was obtained from all participants or, in the case of minors under 16 years of age, from their parents or legal guardians.

2.2. Randomization Randomization was performed at the club level to prevent potential contamination between groups and ensure the internal validity of the study. Using the official registry of women's handball clubs in Catalonia, each club was assigned a unique sequential numerical code, and selection was carried out randomly. Initial contact followed this numerical order and was established via email and telephone, culminating in an in-person meeting with club representatives and/or technical directors to explain the study objectives and formally invite participation. If a club declined the invitation, the next club on the sequential list was contacted until a total of four clubs agreed to participate. The four clubs were randomized to the study groups in a 1:1 ratio, allocating two clubs to the intervention group (IG) and two clubs to the control group (CG).

While individual allocation was not concealed, outcome assessors and data analysts were blinded to group assignments. Participants, however, could not be blinded due to the inherent nature of the educational intervention.

2.3. Subjects Participant recruitment was conducted between February and March 2025. Inclusion criteria were: (a) female athletes aged 12 to 19 years, (b) a minimum of one year of training experience at their club, and (c) training at least three days per week for approximately 1.5 hours per day. Exclusion criteria comprised: (a) athletes who had not yet reached menarche, (b) a diagnosis of an eating disorder, (c) the presence of a chronic disease requiring a specific dietary plan, (d) pregnancy, and (e) lack of proficiency in Spanish.

2.4. Procedure The study was conducted from February 2024 to December 2025. The study design included four phases: pre-intervention, intervention, post-intervention, and a follow-up conducted three months after the completion of the intervention. The timing of the study phases was aligned with the competitive calendar of the participating sports clubs. The pre-intervention phase was conducted during the final period of the competitive season. The intervention phase took place across both the competitive and post-competitive periods, while the post-intervention assessments were performed during the post-competitive phase. The follow-up evaluation, conducted three months after the intervention, coincided with the beginning of the subsequent competitive season.

2.5. Intervention During the initial contact with the athletes in the IG, a focus group session lasting 30 to 60 minutes was conducted to thoroughly explore their expectations, interests, perceived difficulties, facilitating factors, levels of self-efficacy, and opinions regarding nutritional improvement. This qualitative information was useful for identifying potential barriers and opportunities perceived by the participants in adopting healthy eating habits.

The topics addressed in the intervention were determined from: (a) findings from the focus group, (b) results of the nutrition knowledge questionnaire administered during the pre-intervention phase, and (c) insights gained from the previously conducted pilot project. Additionally, the overall design of the intervention was reviewed using the Guide for Effective Nutrition Interventions and Education (GENIE), to ensure alignment with evidence-based standards regarding program goals, instructional methods, content, and evaluation.

The nutritional education intervention consisted of three in-person group sessions delivered by a licensed dietitian-nutritionist, each lasting 30 minutes and delivered once per week. Its duration was established trying to maximize its viability in a real-world setting, outside the framework of a research study. Attendance was systematically recorded for each session.

Following the conclusion of the sessions, the visual materials used in each meeting along with a guide entitled "Nutrition for Young Athletes" were sent to the coaches via email, who distributed the guide to the players and their families. This document, prepared by a licensed dietitian in collaboration with final-year students of the Human Nutrition and Dietetics degree program at the University of Barcelona, included information and practical examples designed to optimize nutrition for sports performance.

The CG did not participate in any educational activities during the intervention phase. These participants were instructed to maintain their usual dietary habits and to avoid seeking additional nutritional information or counseling for the duration of the study. This group received the same nutritional education intervention upon completion of the study.

2.6. Data Collection and Measurements 2.6.1. Questionnaires All questionnaires were self-administered and completed online using forms designed with the REDCap (Research Electronic Data Capture) software platform installed in a server of the University of Barcelona. All questionnaires were completed under the supervision of the principal investigator to prevent discussion or exchange of information among respondents. Completion of the socio-demographic questionnaire, which collected baseline personal and demographic information, was required for inclusion in the study. The Eating Attitudes Test (EAT-26), and the Body Shape Questionnaire (BSQ). were applied only at baseline to assess risk of eating disorders and body image, respectively. Sports nutrition knowledge was evaluated using the Nutrition Knowledge Questionnaire for Young and Adult Athletes (NUKYA) at baseline, post-intervention, and follow-up. Adherence to the Mediterranean diet was assessed with the Kidmed index, at baseline and follow-up.

2.6.2. Assessment of Dietary Intake Dietary intake was assessed at baseline and follow-up using the three-day photographic food record following the previously published pilot study protocol. Average daily intakes of energy, macronutrients, and key micronutrients for female athletes were calculated.

2.6.3. Anthropometric Measurements and Body Composition Body weight, height, and body composition (fat mass, muscle mass, and fat-free mass) were assessed at baseline and follow-up following the previously published pilot study protocol. All measurements were performed by a licensed dietitian at the respective sports facilities.

2.7. Statistical Analysis The primary outcome was sport nutrition knowledge (SNK). Secondary outcomes included adherence to the Mediterranean diet (KIDMED index), dietary intake variables, and anthropometric and body composition parameters. Outcome prioritization was informed by a single-arm pre-post pilot study. In the pilot, changes in KIDMED or anthropometric and body composition parameters were not statistically significant, whereas calcium intake was the only dietary variable that showed a trend towards improvement.

Sample size estimation for the primary outcome sports nutrition knowledge assessed at baseline, post-intervention, and follow-up was conducted using the GLIMMPSE online tool (General Linear Mixed Model Power and Sample Size; University of Colorado Anschutz Medical Campus, Aurora, CO, USA), for a cluster-randomized, repeated-measures design. Power calculations targeted the group-by-time interaction for SNK across three assessments. For calcium intake measured at baseline and follow-up, additional calculations targeted the group-by-time interaction. Assumptions on the expected means, standard deviations, within-participant correlation across repeated measures, and intraclass correlation at the club level were based on pilot data and/or conservative estimates. The type I error rate was set at 0.05 and power at 80%. The minimum required total sample size was 14 participants for SNK and 74 participants for the secondary outcome calcium.

Statistical analyses were conducted according to a predefined plan adapted from our previously published pilot study. Data normality was assessed using the Shapiro-Wilk test, and homogeneity of variances was evaluated using Levene's or Bartlett's test, as appropriate. Between-group comparisons were performed using Student's t-test for normally distributed variables or the Mann-Whitney U test for non-parametric variables.

Changes over time in SNK (NUKYA scores), adherence to the Mediterranean diet (KIDMED index), dietary intake variables, and anthropometric and body composition parameters were analyzed using linear mixed-effects models (LMMs). These models included time and group as fixed effects, and clubs and participants as random effects, allowing for inclusion of incomplete observations. This approach was chosen over repeated-measures ANOVA due to the presence of missing data. Prior to model implementation, missing values were assessed and confirmed to be missing at random (MAR). Post hoc pairwise comparisons were adjusted using the Bonferroni correction. Effect sizes were calculated using Hedges' g and interpreted as very small (<0.20), small (0.20-0.49), moderate (0.50-0.79), or large (≥0.80).

All statistical tests were two-tailed, with a confidence level set at 95%. Statistical significance was defined as p < 0.05. Analyses were performed using STATA statistical software (version 16.1; StataCorp, College Station, TX, USA).