A cross-sectional online survey was conducted to collect data from Swedish and Norwegian female athletes (aged 16–24) in football, horseback riding, handball, swimming, gymnastics, floorball, athletics, cycling, and cross-country and alpine skiing. Sport disciplines were selected based on the highest participation rates among girls aged 13–16 years in Sweden (Skåne County) (24) and Norway (25). The 6-tiered Participant Classification Framework was applied to categorize the level of sport participation (26). To identify potential group differences (27), both team and individual sports were included. No comparative analyses between Swedish and Norwegian participants were conducted due to the uneven distribution of participants. Data were collected from February to June 2024.

The survey focused on perceived knowledge and knowledge needs regarding BPS puberty-related topics during adolescence. In this study, perceived knowledge refers to the level of knowledge the respondents believed they had during adolescence, while knowledge needs refer to their perceived need for additional knowledge during adolescence. Both perceived knowledge and knowledge needs are related to subjective well-being in sports, referring to “a self-rated assessment of well-being in which an individual considers any events, circumstances, or experiences he or she is facing currently” (28, p. 2). This definition was further applied in sports, acknowledging that subjective well-being is a dynamic and multidimensional state, which can change depending on various circumstances (28, 29).

The survey was developed and reviewed by members of an interdisciplinary research group with expertise in survey construction. The process was guided by existing literature on female athletes BPS puberty development and sports participation to ensure that the items effectively captured the intended topics (30). The review enhanced content validity and ensured that the survey did not contain common errors, such as ambiguous or double-phased questions (31). The survey was deigned following Stockemer and Bordeleau's (31) guidelines, including ordering, groupings, and sections. Initially developed in Swedish, it also was translated into Norwegian by the project group using backward translation continuously during the process (32).

A pilot-test was conducted on a diverse subset of female athletes (n = 34) within the target age group to assess face validity (30). It also provided insights into readability, potential linguistic errors, and the time it took to complete the survey (31). The pilot test led to minor language adjustments and ensured that the translation was more likely to be correct (32).

The items in the survey were close-ended and included five items in the introduction, including (a) age, (b) age at entry into sports, (c) average training hours per week, (d) level of sport, and (e) type of sport. This was followed by 35 close-ended items divided into five main variables: biological, psychosocial, behavioral, coaches and parents, utilizing response options using a 7-point Likert-Scale.

The main variables biological, psychosocial, and behavioral included 24 items, addressing both perceived knowledge and knowledge needs of 12 BPS puberty-related topics: changes in sex hormones (e.g., breast development, pubertal fat mass, mood swings), individual differences (e.g., height growth and pubertal development), the influence of biological puberty on sports development (e.g., decreased motor control), changes in body composition, menstruation (e.g., bleeding frequency, menstrual pain, premenstrual symptoms such as rapid mood swings, fatigue, and low mood), social cohesion (e.g., sense of community, trust, and conversation climate), social comparisons, acceptance within the group, feedback, unhealthy norms around body shape and weight, sports nutrition, and recovery (e.g., sleep, rest). Perceived knowledge was assessed on a scale ranging from 1 (very poor) to 7 (very good) [e.g., “How would you rate your perceived knowledge during adolescence about how changes in sex hormones (e.g., breast development, pubertal fat mass, mood swings) could influence your well-being in sports?”]. Knowledge needs were assessed on a scale from 1 (no degree) to 7 (high degree) [e.g., “To what extent do you think increased knowledge about how changes in sex hormones (e.g., breast development, pubertal fat mass, mood swings) influence athletes could have enhanced your well-being in sports during adolescence?”].

The main variables coaches and parents consisted of five items each, addressing the following topics in a more general manner: biological puberty development (e.g., sex hormones, individual differences, impact on athletic development/performance, body composition and weight), menstruation (e.g., bleeding frequency, menstrual pain, premenstrual symptoms such as rapid mood swings, fatigue, and low mood), psychosocial factors (e.g., social cohesion, security within the group), sports nutrition, and recovery (e.g., sleep, rest). These items were assessed using a scale from 1 (no degree) to 7 (high degree) (e.g., “To what extent do you think increased knowledge about biological puberty development (e.g., sex hormones, individual differences, impact on athletic development/performance, body composition and weight) among your coaches could have enhanced your well-being in sports during adolescence?”).

Finally, the respondents were asked to assess the need among young female athletes' (13–16 years old) for increased knowledge regarding being a female athlete during puberty, rated on a scale from 1 (no need) to 7 (very high need).

Due to the EU's General Data Protection Regulation [SFS 2018:218; (33)], which prohibits associations from sharing personal information such as athletes' email addresses, the survey was primarily distributed through a two-step process across national districts in Sweden and Norway. Districts acted as gatekeepers, forwarding the survey to relevant sports clubs, which facilitated broader distribution and enhanced the potential sample representativeness.

The study received ethical approval by the Ethics Review Authority in Sweden (Dnr. 2023-05264-01), and by the Norwegian Agency for Shared Services in Education and Research (ref: 930534), and the Research Ethics Committee at the Faculty of Health and Sport Science at the University of Agder (ref: RITM0225563) in Norway. The respondents gave their informed consent before responding the survey and could withdraw at any time.

Data were analyzed through both descriptive and analytical statistical methods in JASP (34). Descriptive statistics provided an overview of the data and response distribution, including central measures such as median and interquartile range for Likert scales.

To investigate potential differences between athletes in team and individual sports, the Mann–Whitney U test was used. A p < .05 was set to indicate statistically significant results. The Rank-Biseral correlation (rb), a non-parametric measure of effect-size based on correlation coefficient, was the calculated effect-size indicator for group differences (35). This effect-size indicator measures strength of associations in the r-family, varying from −1.0 to +1.0, with 0 indicating no effect. The strength of relationship can be interpreted through the following effect size stages: small (r = 0.10), medium (r = 0.24), and large (r = 0.37) (35).

To explore interconnections in knowledge across BPS domains, a Network Analysis was employed. This analysis combines multivariate statistics with network science, allowing for the exploration of associations within the data by identifying the system components—nodes—and the relationships between them—edges or ties (36, 37). This type of analysis provides insight into the structure and strength of these interconnections, highlighting which topics are central or peripheral within the network.

The analysis involved the generation of a network plot and a centrality plot. The network plot illustrates the relationships among nodes. In this study, nodes represented the 12 BPS puberty-related topics across all networks. The edges indicated the undirected relationships between them. Blue edges indicate positive connections, orange edges indicate negative connections, and thicker edges represent stronger relationships. The grouping of the nodes reflects their interconnectedness and clustering (36, 37).

The centrality measures betweenness centrality, closeness centrality, strength centrality, and expected influence were used to measure the centrality of the nodes in the networks (36). The node strength values are depicted on the x-axis of centrality plots, illustrating the node strengths of connections within the networks. Negative values indicate weaker or inhibitory connections, while positive values reflect stronger connections. Values near 0 suggest minimal importance in the network structure.

To illustrate network data, several figures were merged. Thus, the centrality plots display centrality measures for all respondents and differences between athletes in team and individual sports. Additionally, network plots for both athletes in team and individual sports were merged into one for clarity.

The results indicated that the respondents generally perceived their knowledge to be low (scores 2–3) across most of the included BPS puberty-related topics during adolescence (see Table 2). Scores were 2–3 for all topics, except for feedback among all respondents (score 4) and social cohesion among athletes in team sports (score 4; see Table 2). The results also revealed knowledge needs, as the respondents generally believed (scores of 5–6) that increased knowledge of all BPS puberty-related topics could have enhanced their subjective well-being in sports during adolescence (see Table 2).

Additionally, most respondents agreed on the need to increase knowledge among young female athletes (aged 13–16 years) about being a female athlete during puberty (scores of 6–7; see Supplementary Appendix S1).

The network analysis of perceived knowledge among all respondents (see Figures 1, 2) identified 54 out of 66 non-zero edges, indicating that several topics were correlated. The overall network sparsity in Figure 1 was 0.182, corresponding to that 18.2% of the possible links were missing. The network analysis revealed a relatively dense network with distinct clusters, while some connections between topics remained absent.

According to the network plot (see Figure 1) and centrality plot (see Figure 2), individual differences appeared to be the most central node, characterized by strong connections and a high degree of influence on other nodes in the network. Menstruation and social comparisons were also central, with menstruation showing strong links to multiple nodes and high influence, whereas social comparisons had many connections but weaker associations. The least connected node in the network was unhealthy norms related to body shape and weight.

The network plot revealed three clusters of nodes, as illustrated in Figure 1. Due to strong interconnections, psychological and social topics merged into a single psychosocial cluster (9). This resulted in three clusters: a biological cluster (nodes 1–5), a psychosocial cluster (nodes 6–9, except for node 10 in perceived knowledge), and a behavioral cluster (nodes 11–12). These clusters indicated that perceived knowledge within each domain was more strongly related than across domains, suggesting that potential connections are currently not utilized. For instance, the node acceptance within the group appeared to be strongly linked to other psychosocial nodes (6, 7, 9) but was neither highly connected nor central within the overall network in Figure 1.

The network analysis of knowledge needs among all respondents (see Figure 4 and Figure 5) identified 51 out of 66 possible non-zero edges and an overall sparsity of 0.227. Like previous network analyses (see Figures 1–3), this network (see Figure 4) appeared relatively dense although it still had some unconnected potential connections.

According to the network plot in Figure 4 and the centrality plot in Figure 5, changes in body composition emerged as the most central node in this network. Overall, the correlations between the nodes were weaker compared to previous network analyses. The clusters of biological (nodes 1–15), psychosocial (nodes 6–10), and behavioral nodes (11–12) persisted, with the node unhealthy norms around body shape and weight approaching the psychosocial cluster (see Figure 4).

The robust data material of >1,300 female athletes enabled a comprehensive understanding of Swedish and Norwegian female athletes' perceived knowledge and knowledge needs about BPS puberty-related topics in sports during adolescence. The interdisciplinary approach, including the BPS framework, and the network analysis enabled a nuanced perspective on knowledge of BPS puberty-related topics in sports (6–8, 38).

Nevertheless, the use of a non-validated questionnaire is acknowledged as a limitation and future research is encouraged to refine and validate the survey instrument. However, considering the absence of validated BPS questionnaires suitable for the intended study, this methodological choice was necessary to conduct the present study. The inability to contact respondents directly due to GDPR restrictions, necessitating the use of district gatekeepers for survey distribution, was also a limitation. While this approach ensured broad national coverage, it may have introduced non-response bias as certain districts were not willing to participate in this study (31). Additionally, self-selection bias may also have occurred due to some individuals being more likely to respond than others (31). Moreover, the retrospective design may have introduced response bias in the respondents recalls and self-reports (31). Lastly, the differences observed between athletes in team and individual sports may have been influenced by the specific sports included in this study (27). It is uncertain whether the results would have been the same if other sports, such as tennis or golf, had been considered instead of gymnastics and swimming—two leanness-focused sports (27).

The results of the present study highlight the urgent need of, and thus, a considerable opportunity for a more holistic and integrated understanding of BPS puberty in the sports contexts of female athletes. Differences in perceived knowledge and knowledge needs were identified between athletes in team and individual sports, suggesting potential advantages in tailoring future educational programs to their respective contexts. Moreover, the generally low perceived knowledge and the high knowledge needs observed among all respondents indicate that interventions are necessary for all athletes. Athletes recognized these knowledge needs not only within themselves but also in relation to their coaches and parents, suggesting that future educational programs should extend beyond athletes to include key stakeholders such as coaches and parents in addressing these knowledge gaps.

The findings of low perceived knowledge and high knowledge needs across BPS domains, and their clustering within these domains, suggest that it is insufficient to limit the scope of future educational programs to a single domain. Future educational programs should therefore be holistic and interdisciplinary in their scope when addressing these knowledge gaps in the sporting context of female athletes (6, 7, 38).