Knowledge of low energy availability in biathlon: Why coaches need more targeted education. Results from a national,cross-sectional study

Abstract

This study explores the knowledge of coaches working with adolescent biathletes regarding eating disorders (ED), low energy availability (LEA), and Relative Energy Deficiency in Sport (REDs). METHODS: Recruited coaches completed a questionnaire assessing self-reported knowledge (scale 0–10) of ED, LEA and REDs, and secondly by articulating their knowledge (scored by three experts, scale 0–2). RESULTS: Thirty-three coaches participated (mean age 40.6 ± 12.2 years), most without formal coach education (n = 25, 75.8%), and 13 (39.4%) were female. Self-reported knowledge of LEA/REDs was lower than ED knowledge (4.75 ± 2.9 vs. 5.74 ± 2.5; p = 0.02, g = 0.4), with a similar pattern in expert-rated scores (0.72 ± 0.6 vs. 1.12 ± 0.6; p < 0.01, g = 0.8). A regression model explained 45% of variance in self-reported LEA/REDs knowledge, with education level and course participation positively associated, while sex had no significant effect. For expert-rated scores, the model explained 33% of variance, with only course participation significant. For knowledge of guidelines on addressing LEA/REDs concerns self-reported scores were 3.6 ± 2.9, and 4.5 ± 2.9 for ED concerns (p = 0.04, g = 0.4); and expert-assessed scores were 0.7 ± 0.7 and 0.9 ± 0.8 (p = 0.11), respectively. Regression models did not significantly explain variability in these scores CONCLUSION: Biathlon coaches are less aware of LEA/REDs symptoms and less confident in addressing them than symptoms of ED. Importantly, education and course attendance raised knowledge scores, and preferred learning methods were virtual or physical courses with experts over scientific papers.

Introduction

Coaches and trainers play a crucial role in supporting athletes’ health. ¹ They observe athletes on a weekly basis, serve as trusted allies, and hold significant influence over their career opportunities. As such, their competency in observing and identifying early signs and symptoms of challenges related to physical and mental function and wellbeing are crucial. ² Low energy availability (LEA) poses a risk to athletes’ health, and occurs when energy intake is insufficient to meet the body's total energy needs after accounting for energy expended through exercise. ³ This imbalance can lead to Relative Energy Deficiency in Sport (REDs), a syndrome with a wide range of negative physical, mental, and performance-related effects. Awareness and understanding of the mental and physical consequences of REDs, both with and without an accompanying eating disorder (ED), have grown within scientific communities. ³ However, the translation of evidence-based knowledge into real-world settings remains limited, as research consistently highlights a lack of awareness among coaches and trainers.^4–9 Furthermore, while LEA and REDs are considered important topics within many national sport governing bodies, few have sufficient education or policies in these areas. ¹⁰ There is also an indication that female coaches may be more aware of these health challenges and associated symptoms than male coaches, still evidence is very limited. ⁸ The International Olympic Committee (IOC) medical working group on REDs updated their consensus statement for the third time in 2023, but it remains unclear whether the knowledge translation to sport has improved. Coach influence has been identified as an external mediating factor of intentional LEA exposure for athletes, ² highlighting the need for coaches to have this knowledge.

A distinctive feature of Norwegian sports medical support system is the strong emphasis on “health before performance,” with some federations routinely assessing athletes’ health and well-being before international competitions. ¹¹ If an athlete exhibits symptoms of disordered eating (DE), LEA or REDs, they may be deemed ineligible for international representation, regardless of their recent performance. This does not seem to impede Norwegian sport performance achievements, as Norway is one of the best nations in the Olympics, ranked by medals in total or also gold medals won per capita. ¹² Part of this work, emphasizing the importance of health for athletes, are the specific efforts made by the national “the Healthy Sport Initiative” to increase knowledge on nutrition and awareness of ED, LEA and REDs. ¹³

Adolescent sports represent a critical stage where competition and rankings take on greater significance. This phase is also characterized by the increased physical energy demands of puberty, growth, and development, which can heighten the risk of LEA and REDs. ¹⁴ Additionally, adolescence is a time when ED-symptoms often first appear and can worsen, and the prevalence has increased.^15–17 As such, it is essential for coaches to possess the knowledge and awareness necessary to prevent symptoms, identify early warning signs, and take appropriate action. One of the federations complying with the aforementioned measures, is the Norwegian biathlon federation, who benefit from the numerous courses also organized by the international biathlon federation.^18,19 A key question, therefore, is whether these measures make a meaningful difference to coaches’ knowledge – i.e., whether Norwegian biathlon coaches possess sufficient knowledge of EDs and REDs. Another key question is whether coaches of adolescent athletes engage with courses on EDs and REDs or is it primarily coaches working with elite athletes. This study examines the knowledge of Norwegian coaches working with adolescent biathletes on ED, LEA, and REDs, whether there are sex differences in knowledge, and whether the knowledge is associated with formal coach education or participation in relevant courses. The study further examines the types of courses coaches typically attend and the formats they prefer for acquiring knowledge.

Material and methods

This is a cross-sectional study of ED, LEA and REDs awareness and knowledge among national biathlon coaches in Norway, conducted between September and December 2024. This study is part of a research project aimed at investigating the prevalence of symptoms related to ED, LEA, and REDs among competitive adolescent biathletes (Clinical Trial id-No NCT06487260). All recruited participants were asked to respond to a digital questionnaire (Nettskjema.no, an online form service at the University of Oslo, Norway).

Participants

Participants were Norwegian coaches for adolescent athletes within biathlon from different clubs. They were invited by three means: direct invitation by email correspondence to all listed teams (n = 127) at the national biathlon federations (NBF) webpage, by individual coach invitation (n = 89) based on lists provided by the NBF, and by recruitment via social media. Participants reported their personal sport participation based on the sport categories defined by McKay et al. (2022), while the performance level of their supervised biathletes was classified according to the Tier 0–5 framework outlined in the same publication. ²⁰

Survey design

The digital questionnaire collected demographic information about coaches and assessed their awareness and knowledge regarding DE, ED, LEA, and REDs (Supplementary File 1; translated from Norwegian to English). All questions included both Likert scale (rated from 0 to 10) to capture nuances in participants’ perceived competence and open-ended text responses, allowing participants to articulate their understanding in their own words. The articulated knowledge was assessed by three experts separately and rated on a scale ranging from 0 (low knowledge) to 2 (high knowledge). This simplified scale was chosen to ensure consistency and interrater reliability, which is particularly suitable for studies with smaller sample sizes. The three investigators (TFM, CSB, AMR) were all professionals in female health, athlete health, EDs and REDs, and did their evaluations following the scoring guideline. The intraclass correlation coefficient (ICC) for the three investigators scores was estimated using a two-way random-effects model with absolute agreement for average measures. The ICC was 0.86 (95% CI: 0.814–0.893, p < 0.001), demonstrating good reliability. Any disagreement was discussed in a calibration meeting in which the three investigators discussed their evaluations to set a final score for all respondents on each relevant questionnaire item.

In addition to the individual subscale scores on items related to DE/ED and on LEA/REDs, the average score for knowledge on LEA/REDs (5 items: definitions of LEA and of REDs, risks, health consequences, performance consequences) and on DE/ED (5 items: definitions of DE and of ED, risks, consequences, symptoms) were calculated. These overall mean scores on LEA/REDs knowledge and DE/ED knowledge were calculated both for self-reported knowledge and expert-assessed knowledge. Finally, one question on awareness of guidelines on how to approach an athlete with suspected LEA (knowing how to address the concerns and how to proceed further), and one similar question related to suspected EDs, were both examined separately by self-reported knowledge and expert-assessed knowledge.

Designing the questionnaire

In the development and refinement of the questionnaire, recommended procedures ²¹ and the overarching methodology described previously^5,822–24 were generally adhered to. Specifically, this process comprised the following five steps: (1) drafting items, (2) reviewing the scale, (3) designing a scoring manual, (4) reviewing the scoring manual, and (5) consulting with a Professor in biomedicine statistics to design a quantitative scoring manual. The questionnaire items and scoring manual were drafted by two master's students in sports medicine under the supervision of Professor Jorunn Sundgot-Borgen (JSB). The draft questionnaire was reviewed by a panel of multidisciplinary experts with extensive clinical and research expertise in female athlete health, EDs, and REDs. After refinement following the review rounds, the expert group evaluated the instrument to have content validity (covering the entire domain of knowledge intended to be assessed). The final digital questionnaire was tested with regards to functionality by three coach-students.

The questionnaire was piloted in a preliminary study as part of a graduate research study (MSc), which was reviewed by two experts in the field. ²⁵ That study assessed coaching students’ knowledge of LEA/REDs and ED and evaluated the impact of an intervention designed to enhance their competence.

Evaluation of internal consistency by Cronbach's α for the five knowledge items on LEA and REDs demonstrated good reliability with α = 0.97, with mean inter-item correlation 0.85. Similarly, α for the five knowledge items on DE and EDs was 0.96 with mean inter-item correlation 0.83.

Ethics

This study was prospectively registered and approved according to GDPR regulations (SIKT ID: 219687, 22.07.2024). All participants signed informed consent before entering the study.

Analyses

Results were analysed by IBM SPSS statistics version 28.0.1.0 (142). Data was visually inspected for normality and baseline and descriptive differences between sexes were analysed by Students T-test and Chi-square test, or Mann-Whitney U-test as appropriate.

Two subgroups were created within the sample: academic education (those with a bachelor or master degree, or without any such academic coaching education) and course attendance on LEA/REDs (any of responses 2, 3, 4 and/or 6 in question 13), or DE/ED (any of responses 5–8 in question 13) to assess such impact on knowledge and attitudes. To examine whether sex, education, or course participation were associated with knowledge scores (LEA/REDs knowledge and DE/ED knowledge, separately), multiple linear regression analyses were conducted. Additionally, a MANOVA was conducted to examine the effects of sex, education, or course participation on 17 statements on body mass - and appearance attitude and comments. Knowledge scores or statements were the dependent variables in each model, while sex (0 = female, 1 = male), education level (0 = low, 1 = high), and course participation in the respective topics (0 = no, 1 = yes) were included as independent variables. For the analyses on knowledge scores, course-attendance on LEA/REDs topics was the grouping variable to evaluate LEA/REDs knowledge, and vice versa for ED knowledge. For the analyses on body mass- and appearance comments, one common grouping variable for course attendance was created as it is relevant that ‘body comments’ are addressed in both type of courses (hence, course attendance represented all who had been to courses on either LEA/REDs or/and DE/ED). Regression assumptions were checked and found to be acceptable: linearity and homoscedasticity were confirmed via residual plots, residuals were approximately normally distributed, multicollinearity was low (VIF < 10), and observations were independent.

Finally, associations between self-reported Likert scale scores and expert-assessed knowledge are analysed by Pearson correlations or Spearman's Rho as appropriate. Given the exploratory nature of this study, a significance threshold of p ≤ 0.05 was applied for sensitivity. Hedges g was given as estimates of effect sizes, in which 0.2 is considered small, 0.5 is considered medium, and 0.8 is considered large.

Two respondents reported using webpages to retrieve information for their answers. Consequently, their responses were excluded from the analysis of knowledge and attitude scores.

Results

A total of 33 coaches responded to the questionnaire (73.0% of consenting coaches), with a mean (SD) age of 40.6 years (12.2) and with no significant difference in age between male coaches (n = 20, 60.6%) and female coaches (n = 13, 39.4%) (p = 0.8). The sample is considered an experienced sample, with all having a formal coaching education (Figure 1), and 84.7% with at least 4–6 years of coaching experience in sport and 74.7% with experience specifically in biathlon. All except two coaches had prior experience competing in sport themselves, of which 10 (30.3%) had experience from two or more sports. Most coaches had personal experience as an endurance athlete (75.0% of male coaches and 76.9% of female coaches), whereas team sports were the next most common category of sport, with 45.0% of male coaches and 38.5% of female coaches. A total of 15.0% male coaches and 7.7% female coaches had practiced technical sports, and 7.7% female coaches had participated in another type of sport (equestrian sport).

Figure 1.

Educational background of coaches by sex. Participants could choose multiple categories. NOTE: BSc, bachelor's degree; EHP, Exercise, Health & Performance; MSc, master's degree; NSF, Norwegian Sport Federation (The Norwegian Olympic and Paralympic Committee and Confederation of Sports); “sport federation courses” are courses held by the sport specific federations; “other sport education” is elite coach courses held by the national Olympic Committee.

Coaching responsibilities

All female and all but one of the male coaches had athletes of both sexes in their coaching groups, and the most typical age group and performance level was teenagers (13–15 years) and Tier 1–3 athletes (recreational, developmental or national level), respectively (Figure 2).

Figure 2.

Distribution of age groups (A) and performance levels (B) of athletes that coaches were coaching, presented by sex of coaches. NOTE: yrs, years; Tier 1–5, performance levels standardized by McKay et al. 2022.

The mean (SD) number of athletes coached per team was 17.7 (11.7), with female coaches overseeing an average of 17.0 (10.0) athletes and male coaches overseeing 18.1 (12.9) athletes. The median (IQR) number of co-coaches who contributed with the shared responsibility of athlete teams were 2.0 (4.0), a number that was similar for female and male coaches.

Coaches’ course attendance

The percentage of male and female coaches who reported to have participated in courses on themes related to mental health and REDs, and who reported a request for such courses, are presented in Figure 3 by sex. The most frequent, previously attended courses were on mental health, menstrual cycle, puberty, and nutrition. There was no significant difference between sex on previously attended courses with topics related to DE/ED or LEA/REDs (p = 0.48). The most frequently requested courses were on REDs, communication about LEA, REDs and DE/ED, body image, and menstrual cycle (no significant differences between sex). There appeared to be a saturation of interest in topics already covered by previously attended courses, as the most commonly requested courses were those that had been least addressed in prior courses.

Figure 3.

The percentage of female coaches (A) and male coaches (B) who reported to have previously attended relevant educational courses (black columns), and correspondingly: who were interested in attending such courses (striped columns). NOTE: LEA, low energy availability; REDs, relative energy deficiency in sport; FAT, female athlete triad; DE, disordered eating; ED, eating disorders; Communic. LEA+, communication about LEA, REDs and DE/ED.

Coaches were asked about the preferred course delivery method, giving them the opportunity to choose multiple options. The majority (55.0% males and 69.2% females) reported to prefer digital synchronous lectures, about half (30.0% males and 69.2% females) also reported a preference for asynchronous online self-study tools, 40.0% males and 46.2% females rated physical workshops as acceptable alternatives, whereas 55.0% males and 30.8% females rated physical lectures as good options. Concurrently, about ¼ would like an interactive app and less than 20% rated books and reports as relevant alternatives.

Coaches’ awareness of REDs and body image issues

Most coaches were not aware of the updated REDs consensus statement 2023 (72.7%), whereas four (12.1%) had heard about it, one (3.0%) had heard about the previous version from 2018, and four (12.1%) were unsure whether they had seen these statements. Most coaches had not read the full REDs 2023 consensus statement, whereas two (6.1%) had read the original statement, none had seen the nationally translated version, and three (9.1%) were unsure if they had read any of these. Similarly, three (9.1%) had read the “Best Practice Guidelines on Body Composition Assessment” supplementing the REDs 2023 consensus statement. There were no differences between sex in any of these questions (p > 0.3).

A total of 13 (39.4%) reported that issues with body image, body appearance pressure or DE/ED were present among their own athletes, whereas 20 (60%) did not report any such issues among their own athletes (Figure 4). There were no differences between sex (p = 0.5) nor between those who had or had not participated in courses related to DE/ED, LEA/REDs or body image (p > 0.3). Among those who reported seeing issues related to body appearance pressure or DE/ED among their athletes, no one specifically rated the period before or after puberty as more troublesome. Among those who reported not seeing such issues among their own athletes, 13 (39.4%) reported pre-puberty as the most typical vulnerable period, whereas nine (27.3%) reported post-puberty as the more typical vulnerable period.

Figure 4.

Percentage of coaches reporting body image, body appearance pressure and/or disordered eating as recognizable issues among biathletes (upper row, figure 4A), and total scores on subscales of the self-reported knowledge related to low energy availability (LEA) and relative energy deficiency in sport (REDs) in left figure (lower row, figure 4B), and self-reported knowledge related to disordered eating (DE) and eating disorders (ED) in right figure (lower row, figure 4C). NOTE: “not an issue” is reported for the athletes the coach is directly working with/is responsible for; “others” is reported for other coaches’ athletes; “pre-puberty” is reported issues before puberty/in younger adolescents; “post-puberty” is reported issues after puberty/in older adolescents; LEA health, symptoms of health issues due to LEA; LEA perf, symptoms of performance impairments due to LEA; Risk, risk factors for DE/ED; Conseq, consequences of DE/ED; Symptoms, symptoms of DE/ED.

Coaches’ knowledge of DE, ED, LEA and REDs

The mean (SD) scores in self-reported and expert-assessed overall LEA/REDs and DE/ED knowledge are presented in Table 1. There was a significantly higher mean (SD) self-reported score in DE/ED knowledge versus LEA/REDs knowledge by 0.99 (2.5), p = 0.02, g = 0.4. Similarly, difference was found between the expert-assessed scores in DE/ED knowledge versus LEA/REDs knowledge by 0.40 (0.5), p < 0.01, g = 0.8.

Table 1.

Self-reported and expert-assessed knowledge scores on LEA/REDs and on DE/ED, and on awareness and knowledge on how to approach athletes with symptoms of LEA or EDs, respectively (guidelines). Results are for total group and values are mean (SD).

	Self-reported, Score range 0–10		Expert-assessed, Score range 0–2
	Overall LEA/REDs knowledge	Overall DE/ED knowledge	Overall LEA/REDs knowledge	Overall DE/ED knowledge
Total group (n = 33)	4.75 (2.9)	5.74 (2.5)	0.72 (0.6)	1.12 (0.6)
	LEA guidelines awareness	ED guidelines awareness	LEA guidelines awareness	ED guidelines awareness
	3.6 (2.9)	4.5 (2.9)	0.7 (0.7)	0.9 (0.8)

NOTE: DE, disordered eating; ED, eating disorder; LEA, low energy availability; REDs, relative energy deficiency in sport.

The knowledge on how to approach an athlete with suspected LEA or symptoms of ED (knowing how to address the concerns and how to proceed further) is presented as mean scores in Table 1. Comparing the self-reported knowledge of these guidelines revealed a significant mean (SD) difference by 0.88 (2.3), p = 0.04, g = 0.4, still there was no difference in the expert-assessed scores (p = 0.11).

The effects of sex, academic education or course-attendance on overall LEA/REDs knowledge

The regression model for overall LEA/REDs self-reported knowledge was significant, F(3,27) = 7.24, p = .001, explaining 45% of the variance in knowledge scores (R² = 0.45). Examination of the individual explanatory variables revealed that education level (B = 2.45, SE = 0.96, β = 0.39, p = .02) and course participation (B = 2.23, SE = 0.85, β = 0.40, p = .02) were significant explanatory variables to knowledge scores. Sex was not a significant explanatory variable (B = -0.9, SE = 0.11, β = -0.17, p = .26).

The regression model for overall LEA/REDs expert-assessed knowledge was also significant, F(3,27) = 4.33, p = .013, explaining 33% of the variance in knowledge scores (R² = 0.325). Examination of the individual explanatory variables revealed that course participation (B = 0.44, SE = 0.20, β = 0.38, p = .03) contributed as a significant explanation to knowledge scores. Neither education level (B = 0.38, SE = 0.22, β = 0.29, p = .09) nor sex (B = -0.15, SE = 0.19, β = -0.13, p = .42) were significant explanatory variables.

The regression model for self-reported knowledge on guidelines for addressing concerns on LEA was not significant, F(3,27) = 2.24, p = .11, explaining 20% of the variance in knowledge scores (R² = 0.199). Similarly, no significant model explained the variance in expert-assessed knowledge on these guidelines, F(3,27) = 1.40, p = .26, explaining 14% of the variance in knowledge scores (R² = 0.135). This indicates that the combination of sex, education, and course participation did not significantly explain variation in guideline awareness in these analyses.

The effects of sex, academic education or course-attendance on overall DE/EDs knowledge

The overall regression model for self-reported DE/ED knowledge was not significant, F(3,27) = 2.23, p = .11, explaining 20% of the variance in knowledge scores (R² = 0.199). Neither was the overall regression model for expert-assessed DE/ED knowledge significant, F(3,27) = 1.12, p = .32, only explaining 12% of the variance in knowledge scores (R² = 0.119).

The regression model for self-reported knowledge on guidelines for addressing concerns on symptoms of EDs was significant, F(3,27) = 9.32, p < .001, explaining 51% of the variance in knowledge scores (R² = 0.509). Examination of the individual explanatory variables revealed that course participation (B = 0.43, SE = 1.07, β = 0.54, p < .001) contributed as a significant explanation to guideline awareness. Education level (B = 1.73, SE = 0.89, β = 0.26, p = .06) was not a significant explanatory variable, whereas gender (B = -1.66, SE = 0.81, β = -0.28, p = .05) was close to being significant in explaining the variability in ED-guideline awareness.

On the contrary, the regression model for expert-assessed knowledge on guidelines for addressing symptoms of EDs was not significant, F(3,27) = 1.79, p = .17, explaining 17% of the variance in knowledge scores (R² = 0.166).

Association between self-reported knowledge and expert-assessed knowledge

While some variation existed between self-reported knowledge and expert-assessed knowledge on subscales of DE/ED and LEA/REDs (Supplementary File 2, Figure S-1), there were significant correlations between most of the topics. More importantly, the correlation between the overall self-reported and expert-assessed scores on DE/ED and on LEA/REDs were significant (r = 0.61, p < 0.001, and r = 0.67, p < 0.001, respectively) (Figure 5).

Figure 5.

Correlation between A) average overall self-reported score in LEA/REDs knowledge and expert-assessed score, and B) average overall self-reported score in DE/EDs and expert-assessed score. Note: DE, disordered eating; EDs, eating disorders; LEA, low energy availability; REDs, relative energy deficiency in sport.

Communication about body mass and -composition

Coaches were asked to rate their response from 0 (strongly disagree) to 10 (strongly agree) on statements related to importance of body mass and body composition for performance in biathlon. The median (IQR) score for supporting the claim that body mass is an important performance aspect in biathlon was 5.0 (4.0). In contrast, the score for considering body mass as an important performance aspect to address directly in dialogues with athletes, was 2.0 (5.0). While the median (IQR) scores for the importance of controlling athletes’ body mass either within or beyond the competitive season were 0.0 (3) and 0.0 (5.0), respectively, a quarter of respondents rated these aspects between 4 and 8.

Similarly, most rated the need to do regular assessments of body composition as unnecessary by a median (IQR) scoring of 0.0 (3.0). The median (IQR) score on whether assessment of body composition may trigger LEA or DE/ED, was 7.0 (3.0). Providing detailed nutritional advice to help regulate body mass or composition was rated with a median (IQR) score of 2.0 (5.0). Overall, there were no statistical differences between the sexes on these ratings. Few were inclined to comment on athletes’ bodies in relation to the demands of the sport, with a median (IQR) score of 0.0 (1.0), but the median (IQR) score on the rated importance of addressing concerns about athletes being underweight were high; 8.0 (5.0). In contrast, few felt comfortable commenting to an athlete on the need to reduce body mass, with a median (IQR) score of 0.0 (3.0). Concurrently, the median (IQR) score on willingness to comment on frequent body mass fluctuations was 7.0 (4.0).

Most coaches unanimously agreed not to comment on body appearance, either through negative remarks, median (IQR) score 10.0 (0), or positive remarks, median (IQR) score 10.0 (1). Instead, most preferred to focus on body functionality rather than appearance, median (IQR) score of 10.0 (0).

The coaches were asked whether they facilitated individual talks with each athlete (≥2 times per season), fostered a culture of openness for discussing sensitive topics, and created a safe sporting environment. The median (IQR) scores for these aspects were 9.0 (3.0), 9.0 (2.0), and 10.0 (0.0) for total group.

A MANOVA was conducted to examine the effects of sex, education, or course participation on these 17 communication aspects. The multivariate tests indicated that none of the factors had a significant effect (Pillai's Trace: Sex = 0.67, F(9, 17) = 1.07, p = .48; Education = 0.77, F(9, 17) = 1.76, p = .20; Course = 0.67, F(9, 17) = 1.07, p = .48).

However, follow-up univariate analyses revealed that Q11 (“If I find that an athlete should reduce their body mass, I discuss this with the athlete”) was significantly influenced by sex (F(1, 52) = 7.46, p = .011) with women scoring higher (mean = 2.67, SD = 4.1) than men (mean = 1.42, SD = 1.8). But because Q11 was significantly influenced by the interaction of sex×course attendance, F(1, 53) = 7.6, p = .01), the effect from sex seems dependent on course-attendance. This may result by an extreme value from a single female (mean = 10.0) compared to two males (mean = 0.0) with no course attendance, as the means for course-attending participants were relatively similar for females (mean [SD] = 2.0 [3.5]) and males (mean [SD] = 1.59 [1.8]). Furthermore, Q2 (“It is important to discuss body mass as a crucial performance factor with the athletes”) was significantly influenced by education (F(1, 44) = 5.24, p = .03), with higher education resulting in higher score (mean = 4.75, SD = 3.9) than lower educational levels (mean = 1.83, SD = 2.2). Nevertheless, these sub-results should be interpreted cautiously, as the overall multivariate test was not significant and functions as a control to ensure that spurious effects on individual items are not reported.

When asked to provide examples of the practices they prioritized to foster openness and safety, the responses could be categorized into four main topics, presented by representativeness: (1) building trust and strengthening the coach-athlete relationship, (2) promoting openness, inclusion, and acceptance, (3) facilitating team building and incorporating knowledge translation through theme days and group dialogues on relevant topics, and (4) demonstrating their own human vulnerability in the coaching role to lower the barrier for honesty and encourage seeking help.

Discussion

This study found that coaches from a leading country in biathlon demonstrated moderate knowledge regarding DE/ED, LEA, and REDs, and possessed better knowledge on DE/ED compared to LEA/REDs in both self-reported and expert-assessed scores. There were nuances to this, as education seemed to be associated with better awareness of LEA/REDs both in terms of academic education and relevant course attendance. The coaches seemed to be less confident about what guidelines existed on how to address concerns about LEA compared to addressing DE/ED, still the expert-evaluated results revealed no difference. Most coaches expressed a strong interest in further education in these areas, and specifically regarding communication about LEA and ED. Preferred learning formats included digital synchronized lectures, asynchronous online self-study tools (like videos with experts), and physical workshops, with books or other literature (i.e., reading) seen as less engaging options. Responses to questions on attention to body mass and -composition and communication of attitudes around these factors, demonstrated that coaches were aware of the link with LEA, REDs and DE/ED. They also reported strivings to mitigate the risk through their approaches. No significant differences were found by sex, educational level, or course attendance.

The coaches demonstrated significantly better knowledge about DE/ED compared to LEA/REDs. Contrary to knowledge on LEA/REDs, academic education or course attendance did not seem linked to higher self-reported or expert-assessed knowledge scores on DE/ED. Some explanations may be the frequent articles about DE/ED cases in the news, social media and information by interested organizations, and the more longstanding research on EDs and public discourse on DE/ED, potentially consumed by coaches without the specific education. In contrast, little has been presented about LEA/REDs to the general public by mass media or have been made accessible in Norwegian language, a factor that decreases the likelihood of this information reaching the coaches who need it. Further, The Norwegian initiative Healthy Sport Initiative ¹³ provides sport-specific information on nutrition, including LEA/REDs. However, the titles and content of their information primarily emphasize general nutrition and energy needs, rather than explicitly highlighting LEA/REDs. As a result, many coaches—despite recognizing the need for and expressing a desire to learn more about LEA/REDs—may overlook these educational resources. Supposing they have already covered sports nutrition in previous courses, they may not realize that these programs now also include critical information on LEA/REDs, leading them to engage with the materials less frequently than needed. Low knowledge about REDs may also lead coaches to misinterpret or overlook REDs-specific content in courses, mistaking it for general nutrition-, injury-, or eating disorder education. This lack of understanding can result in underreporting of REDs-specific training, thereby biasing the current results. It may also be an argument that many may believe that LEA/REDs is similar to DE/EDs, or is only caused by DE/EDs, and as such think they understand the concept until they are asked to articulate their specific knowledge. Qualitative follow-up studies are needed to further explore this apparent interpretive confusion among coaches. Those with academic degrees in coaching reported better self-evaluated and expert-evaluated knowledge of LEA and REDs, and similar was found for coaches that had attended courses on these topics. These results indicate that attendance of relevant courses may result in higher knowledge levels of the relevant topics. However, without scores prior to education courses it is not possible to comment on causation. Evaluation of coaches’ knowledge of REDs pre and post workshop has been shown to remain unchanged in some contexts ²⁶ and improved in others. ²⁷ Nevertheless, improved knowledge does not guarantee competence in effective dissemination of knowledge,^26,27 and many coaches seek training and support in how to address their concerns with athletes. According to Kolb's experiential learning theory, learning is most effective when it involves a cycle of concrete experience, reflective observation, conceptual understanding, and active experimentation. ²⁸ Participation in a single course often provides knowledge but lacks opportunities for reflection and practical application, which are essential for translating learning into behavioural change. ²⁸ This may be reflected in what the coaches expressed as their preferred learning method, where self-study of written material was less favored, whereas various ‘meet-the-expert’ opportunities and workshops were rated more highly. A reasonable interpretation is that such formats may provide greater opportunities for discussion and interaction. The Social Cognitive Theory emphasizes that behaviour change occurs through a dynamic interaction between the individual, their behaviour, and the surrounding environment, and that individuals are more likely to apply new knowledge when they feel a sense of social responsibility or commitment to others. ²⁹ Therefore, it is crucial that course participants perceive a consensus and shared agreement on the need for change, and that they do not feel alone in wanting to implement change but rather experience that others are likely to do the same. In some contexts, there still exists an ‘old school’ coaching mentality which prioritises performance over aspects of athlete health and wellbeing ¹⁰ which can make it challenging to create this shared consensus and agree need for change. Without repeated practice, feedback, and a supportive environment that reinforces this social accountability, coaches may struggle to find motivation and competence to implement new concepts effectively in their daily work with athletes.

A few respondents admitted to looking up the terms online (LEA/REDs) to complete the knowledge questionnaire (these responses were excluded from the analysis). This aligns with previous findings in the literature, where interviewed coaches reported unfamiliarity with the terms. ⁴ These findings underscore the importance of making evidence-based information about LEA/REDs accessible to coaches through education, whether through academic studies or sport federation courses to safeguard the quality of information. The high number of coaches in this study without formal academic education but who attended courses highlights the importance of national and international federations continuing to prioritize these educational opportunities.

There was generally a strong correlation between self-reported knowledge and expert-evaluated knowledge, suggesting a good correspondence between perceived and professionally validated knowledge. However, for knowledge specifically related to LEA/REDs, there was considerable variation between self-reported and expert-assessed knowledge. This indicates that some individuals may perceive themselves as knowledgeable about a critical health issue, while having limited insight into the actual extent of their understanding. Such a discrepancy could have serious consequences for the athletes being guided, as symptoms may go unnoticed and health problems may develop.

Importantly, whereas the coaches acknowledged that body mass is a key performance variable in biathlon, they appeared to be highly aware that comments about body mass and appearance can be triggering for LEA and DE/ED. Furthermore, most believe that regular assessment is not required and are careful to focus on body functionality rather than appearance. This approach is in accordance with the recent IOC consensus and supplementary paper.^3,30 Few sex differences in knowledge were identified, which contrasts to one previous study, pointing to better awareness of mental health challenges among female versus male coaches. ⁸ Keeping in mind that our findings may be influenced by a limited sample size, the observed sex differences in knowledge were minimal, underscoring a shared need for enhanced competence. We also found no significant differences in behaviour and attitudes between those who had attended courses or had higher education. While courses and education would be expected to improve awareness, the particular demands and expectations of sport may still take precedence over common sense and health considerations. While we do not know the details of the course content or teaching methods - the intention of education to reduce risk does not seem to have been met. Future work should evaluate the effectiveness of education and incorporate methods that are known to impact practice so as to further reduce risk for athletes. It is encouraging from the results that coaches in biathlon appear inclined to raise concerns of risk factors relating to body mass (underweight or fluctuations) and do not engage in practices that may increase risk (e.g., not controlling body mass, assessing regularly, commenting on BM in relation to the sport demands).

Implications

Our findings indicate that, regardless of educational background, coaches often feel unprepared to address issues related to DE, ED, LEA, and REDs. Furthermore, the course topics that coaches have attended on previously (mental health, puberty, nutrition and menstrual cycle) differ to those they wish to attend (REDs, communication of LEA, REDs and DE/ED). Therefore, and in line with previous findings,^10,26,27 the next step in advancing coach education should focus on equipping them with the skills to initiate conversations about body image, DE/ED, and LEA/REDs. This includes guidance on and practice in how to approach athletes, how to take the first step in communication, a clearer understanding of their role, and when and how to refer athletes to appropriate health services.

To enhance both knowledge and skills among coaches, other teaching methods than purely didactic approaches may be needed.^26,28 More effective approaches are interactive and practice-based teaching, combined with mentoring, reflection, and application in real-world situations, ²⁸ and should be considered into education to make coaches feel confident in their role and with their responsibilities.^27,31 This contrasts with how information, especially about REDs and guidelines on how to approach athletes, is generally disseminated (reports and scientific articles). The latter does not reach out to coaches and represents the least relevant means for dissemination.

Strength and limitations

The cooperation with the national federation within biathlon must be considered a strength to this study, increasing the likelihood of reaching out to all relevant coaches. The multidisciplinary group cooperating on the design of the questionnaire, the internal consistency by items demonstrating good reliability, and the piloting of the questionnaire by a master's in science project is a further strength to the reliability and content validity of the non-validated questionnaire. Furthermore, a notable strength was the collaboration of three researchers in scoring the open-ended knowledge questions, with ICC demonstrating good reliability. Nevertheless, a limited sample size, and the use of a formally unvalidated questionnaire may have significant limitations to the validity of these results.

Conclusion

This study highlights the need for improved knowledge of LEA and REDs among biathlon coaches working with adolescent athletes. Coaches, irrespective of sex, demonstrated limited familiarity with symptoms and knowledge on how to address related concerns. Education was associated with better knowledge scores. When it came to learning preferences, they favoured asynchronous or synchronous digital and in-person lectures or workshops over books, reports, or self-study resources such as apps. These findings offer valuable guidance for coach education programs and sport federations, particularly within biathlons, to shape future training initiatives.

Supplemental Material

sj-docx-1-spo-10.1177_17479541251407845 - Supplemental material for Knowledge of low energy availability in biathlon: Why coaches need more targeted education. Results from a national, cross-sectional study

Supplemental material, sj-docx-1-spo-10.1177_17479541251407845 for Knowledge of low energy availability in biathlon: Why coaches need more targeted education. Results from a national, cross-sectional study by Therese Fostervold Mathisen, Jorunn Sundgot-Borgen, Lindsay Macnaughton, Anne Mette Rustaden, Inger Johanne Løkkevik, Helge Einar Lundberg and Christine Sundgot-Borgen in International Journal of Sports Science & Coaching