THE RED-S PROTOCOL

1

The RED-S Protocol: A Standardized Approach for Female College Athletics
Eric K. Walker
Edson College of Nursing and Health Innovation, Arizona State University

Author Note
Eric Walker is a graduate student in the Edson College of Nursing and Health Innovation
at Arizona State University.
I have no known conflict of interest to disclose.
Correspondence concerning this article should be addressed to Eric Walker, Edson
College of Nursing and Health Innovation, Arizona State University, 550 N. 3rd Street, Phoenix,
AZ 85004, email: ekwalke2@asu.edu

THE RED-S PROTOCOL

2
Abstract

Relative Energy Deficiency in Sport (RED-S) is an insidious but sinister condition that will
affect a significant amount of female college athletes during their careers. Despite its prevalence,
there is no standardized approach to screening for and managing this disorder in college
athletics. In a novel protocol implemented at a college in the Mountain West, 86% (n=44) of the
total female student-athlete population was screened for RED-S and its precursors using
validated questionnaires (LEAF-Q and EAT-26). Inclusion criteria were all female athletes over
the age of 18. Following the screening questionnaires, subjects were encouraged to attend a
follow-up interview to discuss their results. The questionnaires identified 70% (n=31) of
participants deemed at risk for this syndrome and should receive follow-up guidance. Of the 70%
considered at risk, only 41% (n=13) attended the follow-up interview. Only 3% (n=1) sought
help from the Wellness Center or Athletic Trainer after the interview. The results of this project
confirmed that female college athletes are at risk for this threatening condition and are not
receiving appropriate screening and care. This project also showed that a standardized approach
to RED-S is a critical yet efficient and feasible option to improve the health and performance of
female college athletes.
Keywords: Relative Energy Deficiency in Sport, RED-S, Low Energy Availability,
screening questionnaires, female college athletes

THE RED-S PROTOCOL

3

The RED-S Protocol: A Standardized Approach for Female College Athletics
The Female Athlete Triad is a well-known and well-researched syndrome that affects
female athletes. Unfortunately, Relative Energy Deficiency in Sport (RED-S) is not as widely
known, even though it impacts significantly more athletes. Most athletes do not realize they are
suffering from this disorder, and most providers do not know about it or how to screen for it.
Background and Significance
RED-S and its causative issue, Low Energy Availability (LEA), are precursors to the
Female Athlete Triad (FAT). FAT is characterized by low energy due to disordered eating (DE)
or eating disorders (ED), osteoporosis or osteopenia, and menstrual cycle dysfunction (Cabre et
al., 2022; Logue et al., 2020). A substantial literature review and synthesis showed that although
RED-S has been public knowledge for nearly a decade, it was not until the last four years that
RED-S became a focal point of discussion and study (see Appendices A and B).
RED-S is a deficiency in caloric intake compared to energy expenditure that prevents an
athlete’s body from functioning optimally (Cabre et al., 2022; Logue et al., 2020; Mehta et al.,
2018). Five days of LEA can lead to symptom onset and complications (Cabre et al., 2022;
Mehta et al., 2018). Unrecognized LEA/RED-S can lead to long-term complications, including
osteoporosis, osteopenia, amenorrhea, cardiovascular dysfunction, immune system suppression,
muscular atrophy, orthostatic hypotension, infertility, anemia, anxiety, depression, poor athletic
performance, and other comorbidities (Cabre et al., 2022; Mehta et al., 2018; Melin et al., 2018).
As a result of inconsistent definitions of RED-S, non-standardized questionnaires and
their use, reporting bias in the responses to screening questionnaires, and a general lack of
awareness of RED-S/LEA, it is difficult to obtain an accurate estimate of RED-S and LEA in
female college athletes (Cabre et al., 2022; De Souza et al., 2014, Logue et al., 2020; Mehta et

THE RED-S PROTOCOL

4

al., 2018). However, Logue et al. (2020) state that 22-58% of female athletes will experience
LEA or RED-S during their athletic careers. In 2022, the National Collegiate Athletic
Association (NCAA) published a student demographic report for the last 50 years demonstrating
a continuous increase in the number of female athletes participating in college sports. As more
females participate in collegiate athletics, it should be assumed that more female athletes will
suffer from LEA or RED-S. This is partly because female athletes, compared to male athletes
and the general population, have a higher risk of developing RED-S secondary to an ED or DE
because of the pressure to perform well, the public scrutiny of their bodies in form-fitting and
revealing uniforms, misinformation about healthy eating as an athlete, and mental illnesses
(Mehta et al., 2018; Melin et al., 2019).
Screening questionnaires are the most common interventions to assist in identifying
ED/DE, RED-S, and LEA (Mehta et al., 2018; Melin et al., 2019; Rogers et al., 2021; Sim &
Burns, 2021). Nevertheless, they are under-utilized in the college female athlete population, most
likely due to the unfamiliarity with RED-S and its prevalence and a high level of discomfort in
treating the syndrome among healthcare professionals (Logue et al., 2020; Mehta et al., 2018;
Tenforde et al., 2020). Mehta et al. (2020) reported that 63% of physicians were unaware of
RED-S, and Tenforde et al. (2020) reported that 73% of sports medicine healthcare providers
were unaware of it. There is a problem with this lack of familiarity, as providers can neither
screen for nor treat a syndrome if they are unaware of it.
Setting and Stakeholders
A small college in the western United States envisions its students entering with a passion
and leaving with a purpose. They pride themselves in offering affordable and quality education
to traditional, non-traditional, non-degree-seeking, and trade school students. They serve an

THE RED-S PROTOCOL

5

estimated 3,000 students annually. Of those 3,000 students, there are 75 student-athletes, and
roughly 50 are female athletes who play basketball, soccer, or volleyball. There are no protocols
or regulations for screening female athletes for RED-S/LEA or ED/DE in the athletics
department.
Numerous stakeholders are involved in this project: coaches and coaching staff, athletic
director, athletic trainer, institution administration, and wellness center counselors. The coaches
and coaching staff are stakeholders invested in the success of this project since it has the
potential to positively impact their athletes both on and off the playing field. Coaches, trainers,
and staff who support the project can advocate for its continued use, and it will serve as the first
line of defense against RED-S. With the trainer’s support, athletes, coaching staff, and all other
parties involved will be more convinced of the project’s success. Due to the athletic director’s
commitment to improving the health and well-being of student-athletes at the institution, she is a
stakeholder in this project. In addition to assisting with institutional administration, she and the
site champion are on hand to navigate administrative hurdles. The wellness center counselors are
another critical stakeholder in this project. Counselors are more likely to support the project if
they are aware of the prevalence and importance of these disorders. Athletes determined to be at
risk for RED-S or ED/DE will be referred to counselors for an initial therapeutic intake exam.
Institution administration is the gatekeeper for implementing the project. If they believe in the
project and understand the positive impact that it will have on the athletics program, they have
the power and authority to approve it and ensure the athletics department has the appropriate
resources to maintain it.
Project Purpose

THE RED-S PROTOCOL

6

The purpose of this quality improvement project was to increase the awareness and
management of RED-S and LEA among female college athletes and athletic staff by establishing
a protocol at a small institution in the Southwest United States. This protocol had three
components: an educational seminar, appropriate screening questionnaires, and a standardized
treatment plan that utilized campus resources. By following the protocol, RED-S was identified
and awareness increased, thus decreasing long-term complications.
PICO Question
A review of the literature leads to the clinically relevant PICOT question: For a college
athletics program, how does implementing a RED-S education and screening protocol compare
with no education and screening during the competitive sports season provide an accurate
assessment of athletes at risk for Relative Energy Deficiency in Sport?
Search Strategy and Literature Review
To best answer the PICO question, an intensive literature review exhausted three
databases: SAGE Journals, CINAHL, and PubMed. All three databases are peer-reviewed and
only publish articles that meet their standards. By systematically searching these three databases
for articles relevant to the PICO question, there is little doubt that all applicable studies were
discovered and reviewed.
Initial and Final Search Yields
All searches were limited to the English language, full-text, and published within the last
five years. The initial search contained the phrases: female athletes or women athletes and
relative energy deficiency in sport or RED-S. This search yielded 175 results in PubMed, 689 in
SAGE Journals, and 70 in CINAHL. The terms college athletes and LEA or low energy
availability and screening, test, management, diagnosis, or assessment were added to refine the

THE RED-S PROTOCOL

7

search. These additions produced 300 PubMed results, 52 in SAGE Journals, and 101 in
CINAHL. To encompass a broader range of studies, the term eating disorder was added to the
PubMed and CINAHL databases, which yielded 520 and 431 articles, respectively. One last
search was performed in all three databases where the population term female athletes or women
athletes was changed to junior college athletes or community college athletes, which produced
zero results overall. Gray literature was reviewed but was excluded due to poor design or
methodology, published more than five years ago, or significant bias. After reviewing the titles
and abstracts of the results, 80 articles were selected for further review. After completing a rapid
critical appraisal checklist, the top 10 most relevant studies were selected for inclusion.
Limitations, Inclusion, and Exclusion Criteria
Explicit inclusion and exclusion criteria were used to identify all potentially relevant
studies. The inclusion criteria included studies that were (1) published in the English language;
(2) were limited to human subjects; (3) available in full text; (4) and performed from 2001 to
2023. The exclusion criteria comprised of studies that were (1) not available in full text; (2) had
poor design or methodology; (3) and had significant bias. The ten studies included: two
systematic reviews, one randomized controlled trial, and seven cross-sectional studies. There is a
paucity of high-quality level one and two evidence due to the only recent significant interest in
RED-S and LEA, which is why level three evidence is sufficient for this literature review.
Critical Appraisal and Evidence Synthesis
The articles selected for inclusion were graded on strength and quality by a rapid critical
appraisal tool (Melnyk & Fineout-Overholt, 2019). All studies that underwent appraising were
quantitative, with two systematic reviews for level one evidence, one randomized controlled trial
for level two evidence, and seven cross-sectional studies for level four evidence (see Appendix A

THE RED-S PROTOCOL

8

Table A1). There is a paucity of high-quality level one and two evidence due to the only recent
significant interest in RED-S and LEA, which is why having a majority of level four evidence is
sufficient for this literature review.
The focus of each study has a wide range of heterogeneity; three studies examined the
prevalence of RED-S/FAT in female athletes, three studies explored RED-S/FAT knowledge
among athletes, coaches, and trainers, and both systematic reviews explored the efficacy of
screening questionnaires to identify RED-S/FAT/ED/DE (see Appendix A Table A2). One study
examined the relationship between the LEAF-Q and biophysical manifestations of RED-S/FAT
(Rogers et al., 2019). The final study examined the relationship between ED symptom severity
and a discussion-based intervention (Steward et al., 2019). While most studies used validated
instruments for measurement, a small portion created tools because a validated one does not exist
yet. Note that the internally created tools were based on valid instruments but were modified to
fit the study’s goal and to gather the appropriate data.
Prevalence and intervention-based studies had a homogenous population of females aged
17-44 at various levels of athletic competition, from recreational to professional. Two studies
were international; one occurred in Ireland, one in the United Kingdom, and the rest in the
United States. The knowledge-based studies introduced more heterogeneity to the demographic
pool by including healthcare professionals and coaches, which introduced males and individuals
>44 years old.
After carefully reviewing the top 10 most relevant studies, three themes became apparent
and should be focused on. One, there is a high risk of RED-S/FAT in female athletes, with
increasing prevalence in higher levels of competition, thus putting college athletes at significant
risk for LEA and RED-S. Two, the ability to identify, prevent, and treat LEA/RED-S/FAT is low

THE RED-S PROTOCOL

9

among athletes, coaches, and healthcare professionals, leading to poor outcomes if not addressed.
Three, screening questionnaires are an appropriate and validated tool to identify athletes at risk
for LEA/RED-S/FAT/ED/DE, providing a feasible intervention to prevent or improve poor
outcomes. Based on the main themes, it should be assumed that validated tools used to screen
college athletes for LEA/RED-S/FAT/ED/DE will identify high-risk individuals who should be
referred to a healthcare provider with experience managing those conditions, thus improving
outcomes for female athletes.
Theory and Theoretical Framework Application
Orem’s Self-Care Theory assumes a balance between Self-Care, Self-Care Agency, and
Self-Care Demands after adjusting for age, ability, socioeconomic status, and other variables
(Gonzalo, 2023; Petiprin, 2023) (see Appendix B Figure B1). Self-Care is the actions taken to
maintain health, and Self-Care Agency is the ability of an individual to perform those actions.
Self-Care Demands are the volume of actions that must occur in a certain amount of time to
maintain health and wellness (Gonzalo, 2023; Petiprin, 2023). When there is an imbalance in one
of those areas, it is termed Self-Care Deficit, and that is when Nursing Agency comes into effect.
Essentially, this means the nurse determines how to get an individual to be self-reliant again by
correcting the imbalance in self-care (Gonzalo, 2023; Petiprin, 2023). Some fundamental tenets
of universal Self-Care principles are to have an appropriate food intake, a proper ratio of activity
and rest, and to prevent self-harm (Gonzalo, 2023; Petiprin, 2023). These tenets describe
LEA/RED-S/FAT almost exactly, as they result from a deficient calorie intake coupled with an
inappropriate ratio of activity and rest which can lead to self-harm. When an athlete is in a SelfCare Deficit, i.e., when experiencing LEA/RED-S/FAT, this theory recommends that Nursing
Agency be employed to fix the deficit and help the athlete return to a homeostatic state.

THE RED-S PROTOCOL

10

Nursing Agency actions can be Wholly Compensatory, Partial Compensatory, or part of a
Supportive-Education System (Gonzalo, 2023; Petiprin, 2023). Wholly Compensatory is when
the nurse does everything for an individual. Partial Compensatory actions assist an individual in
completing their tasks, and Supportive-Education System actions are when a nurse educates the
individual on how to perform the actions independently (Gonzalo, 2023; Petiprin, 2023). By
understanding the scope of actions that a nurse can take to correct a Self-Care Deficit, an
appropriate intervention can be created to combat LEA/RED-S/FAT.
This project falls into the Supportive-Education System because it educates athletes with
a Self-Care Deficit about available resources so they can manage their disorders with or without
professional help. As this project becomes engrained in the foundation of the athletic network
over the next few years, it could become a Partial Compensatory action. This means that after an
athlete is identified as at-risk for RED-S, the nurse would take a more hands-on approach to
ensure the athlete is utilizing available resources and helping to manage the condition.
Implementation Framework
Rosswurm and Larrabee (1999) created a Quality Improvement Framework model to
implement organizational change based on best practices and evidence (see Appendix B Figure
B2). In summary, the model involves identifying a need within an organization, linking the
problem with desired outcomes, synthesizing the best evidence, developing a proposed
intervention, implementing and evaluating the proposed intervention and outcomes, and finally
integrating and maintaining the intervention if it is beneficial to the organization (Rosswurm &
Larrabee, 1999). This framework is ideal for creating an organizational change targeting
LEA/RED-S/FAT.

THE RED-S PROTOCOL

11

According to our assessment of the institution’s needs, there is a deficiency in screening
for LEA, RED-S, and ED among female athletes. Linking this deficit with the desired end-state,
healthy athletes that perform well academically and athletically, the evidence search and
literature review showed that screening tools effectively and efficiently identify athletes at risk
for these conditions. The next step is to create and implement a protocol that uses validated
screening tools to identify high-risk athletes and standardize the treatment and management
process. After that intervention is implemented on a trial basis, it is reviewed for effectiveness
and value. If deemed valuable to the organization, it is integrated into the standard operating
procedure and will be continually evaluated for benefit and effectiveness, with changes and
modifications made as needed.
Methods
This project was approved by the WWCC and Arizona State University’s Institutional
Review Board (IRB), with final approval granted on November 1st, 2023. It was noted to pose
minimal to no risk to participants who receive the education and survey. Three ethical principles
guided this project: respect for person(s), beneficence, and justice.
Participants and Recruitment
The coaches, support staff, athletic director, athletic trainer, and student-athletes received
an email requesting their help with this project at the following intervals prior to initiation of the
project: one month, one week, one day, and the day of the project.
Participants were asked to complete an informed consent form, and no monetary
compensation was given. All participants were encouraged to act in their independent interest,
free from coercion or other prerogatives. The principles of veracity and fidelity were
demonstrated by communication, transparency, informed participation, and advocacy. All

THE RED-S PROTOCOL

12

stakeholders were encouraged to maintain transparency and open communications during the
project. Additionally, there was an open-door policy for questions, concerns, and clarifications
for all participants.
The student-athletes participated in a short presentation that defined RED-S, the Female
Athlete Triad, Eating Disorders, and Low Energy Availability. It discussed the epidemiology of
these disorders, common signs and symptoms, potential long-term complications, and when to
see a medical provider. As a result of the presentation, participants understood the possible
adverse effects of these disorders, why screening is important, and should be more comfortable
talking about these disorders with coaches and healthcare professionals. The presentation took
place in a large conference room at the project site and lasted approximately 45 minutes.
Data Collection and Outcomes Management
Student-athletes who enrolled in the study completed the Low Energy Availability in
Females Questionnaire (LEAF-Q) (Appendix C Figure 1) and the Eating Attitudes Test- 26
(EAT-26) (Appendix C Figure 2). The LEAF-Q is a validated and sensitive instrument that
screens for self-reported symptoms potentially caused by low energy availability, including
frequent injury, gastrointestinal dysfunction, and menstrual dysfunction. A score is generated
based on the participant’s response to questions in those domains. A score ≥8 indicates the
individual is at risk for LEA (Melin et al., 2014).
The EAT-26 is a validated and sensitive instrument that measures eating behaviors to
identify individuals who might be at risk for an eating disorder. The EAT-26 measures
participants’ self-reported behaviors and beliefs about food and eating to generate a score with a
score of ≥20 indicating a potential eating disorder or disordered eating (Garner et al., 1982). It
also screens for red flag eating behaviors such as binge eating, vomiting after eating, laxative or

THE RED-S PROTOCOL

13

diuretic use to control weight, and excessive exercise. A positive response to any of those
questions was deemed a high-risk behavior (Garner et al., 1982).
As part of these questionnaires, demographic information such as age, ethnicity, sport,
and grade was also collected. It was possible for participants to skip questions they did not wish
to answer. The questionnaires were answered directly after the presentation and took
approximately 20 minutes.
In order to prevent student-athletes from feeling pressured to participate, coaches and
athletic trainers were not allowed to attend the presentation or follow-up interview. Coaches,
staff, athletic trainers, and the Wellness Center did not receive individual results of the
screenings unless a participant voluntarily disclosed them. Student-athletes were also reminded
that their participation in the project, or non-participation, would not impact their playing time,
scholarships, or other benefits associated with competing at this institution. At the end of the
competitive season, a report, with de-identified information, was provided to the coaches, staff,
and athletic trainer detailing which percentage of participants are considered at-risk for these
syndromes. However, it did not include personally identifiable information.
The project investigator reviewed, scored, keyed, and de-identified the questionnaires
after all participants completed them. To clarify any questions that participants may have
regarding the results of the questionnaires, follow-up interviews were scheduled, lasting
approximately 15 minutes. Participants considered at risk for these syndromes were encouraged
to use the appropriate resources at the project site. Participants at risk for Low Energy
Availability or RED-S were encouraged to meet with the Athletic Trainer to create food logs and
increase caloric intake. Participants at risk for eating disorders were encouraged to meet with the
Wellness Center to discuss management and prevention. All participants identified as at risk for

THE RED-S PROTOCOL

14

any disorder were encouraged to follow up with a primary care provider for further evaluation
and comprehensive management. Those not at risk for these syndromes were provided with
information regarding campus resources. It should be noted that these questionnaires and
interviews were not sufficient for the diagnosis of these disorders; rather, they were only
designed to identify student-athletes at risk for developing these disorders.
After all the data was collected and de-identified, it was entered into a statistics software
program for data analysis. Data analysis was done to identify which age groups, college grade
levels, and sports teams have the highest percentage of at-risk athletes.
The prospective outcomes of this project included identifying the student-athletes
experiencing these syndromes, educating student-athletes on available resources, and ways to
prevent the progression of the issues, as well as aggregate data regarding student-athletes who
contacted the appropriate resource within one month of the initial screening. At the conclusion of
the competitive season, the athletic staff received the results after analyzing the collected data.
Following their review, a short meeting was held with the key stakeholders to discuss the
outcomes and recommendations for how to improve the protocol. Appendix D provides a brief
overview of the project.
Budget and Funding
The budget for this project was maintained by the project investigator. It included the
cost of gas for travel, investigator meals and lodging, and printing costs for the surveys (See
Appendix E).
Results
Introduction

THE RED-S PROTOCOL

15

Intellectus statistics was used to manage all data and analysis. Descriptive statistics
analysis was utilized for all analyses.
Demographics
A total of 86% (n=44) of the total female student-athlete population participated in the
project protocol, which included an educational seminar, two screening questionnaires, a followup interview, and referral to other resources as needed. An analysis of descriptive statistics was
conducted on participants and is presented in the following Table 1 (See Appendix F Table 1).
Each of the 44 athletes completed the LEAF-Q and EAT-26 screening questionnaires.
The LEAF-Q utilizes injury history, and gastrointestinal and menstrual functions to generate a
score. A score ≥8, out of a possible 49, indicates the individual is at risk for LEA (Melin et al.,
2014). The LEAF-Q scoring does not indicate the severity of symptoms. The score should only
be used to identify not-at-risk and at-risk athletes (Melin et al., 2014). The average LEAF-Q
participant score was 8.50. There was a high score of 19, and one participant scored 0. (See
Appendix E Table 2).
The EAT-26 measures participants’ self-reported behaviors and beliefs about food and
eating to generate a score with a score of ≥20 indicating a potential eating disorder or disordered
eating (Garner et al., 1982). In contrast to the LEAF-Q, the EAT-26 score does correlate to the
severity of behaviors and attitudes. A higher score indicates a more disordered relationship with
food and eating. The average EAT-26 score was 8.07. The highest score was 37, and 10
participants scored 0. Seven participants scored ≥20. Additionally, it screens for red flags related
to eating behavior, such as binge eating, vomiting after eating, laxative or diuretic use to control
weight, and excessive exercise. A positive response to any of those questions was deemed a
high-risk behavior (Garner et al., 1982). Among the participants, 22.7% (n=10) of participants

THE RED-S PROTOCOL

16

responded positively to one or more questions about high-risk behavior. EAT-26. (See Appendix
E Table 2).
ANOVAs
An analysis of variance (ANOVA) was conducted to determine whether there were
significant differences in EAT-26 Score by Ethnicity, Grade, and Sport. A second ANOVA was
conducted to determine whether there were significant differences in LEAF-Q Score by
Ethnicity, Grade, and Sport.
Before interpreting the results of the ANOVAs, they were evaluated for normality,
homoscedasticity, and outliers. Significant deviations from the predicted or theoretical data could
indicate an unreliable data set (Field, 2017; Osborne & Waters, 2002; Pituch & Stevens, 2015).
After reviewing the Q-Q scatterplots, homoscedasticity scatterplots, and Studentized residuals
plots, the data was deemed appropriate and reliable for analysis and interpretation. (See
Appendix E Figures 1-6).
Results
The first ANOVA evaluating the relationship between EAT-26 Score and Ethnicity,
Grade, and Sport was examined based on an alpha value of .05. The results of this ANOVA were
not significant, F(8, 35) = 0.85, p = .565, indicating the differences in EAT-26 Score among the
levels of Ethnicity, Grade, and Sport were all similar. (See Appendix E Table 3). The main
effect, Ethnicity was not significant, F(4, 35) = 0.62, p = .652, indicating there were no
significant differences of EAT-26 Score by Ethnicity levels. The main effect, Grade was not
significant, F(1, 35) = 0.58, p = .451, indicating there were no significant differences of EAT-26
Score by Grade level. The main effect, Sport was not significant, F(3, 35) = 0.76, p = .524,
indicating there were no significant differences of EAT-26 Score by Sport levels.
The second ANOVA evaluating the relationship between LEAF-Q Score and Ethnicity,
Grade, and Sport was examined based on an alpha value of .05. The results of this ANOVA were
not significant, F(8, 35) = 1.67, p = .142, indicating the differences in LEAF-Q Score among the
levels of Ethnicity, Grade, and Sport were all similar. (See Appendix E Table 4). The main

THE RED-S PROTOCOL

17

effect, Ethnicity was significant, F(4, 35) = 2.98, p = .032, ηp2 = 0.25, indicating there were
statistically significant differences in LEAF-Q Score by Ethnicity levels and will be evaluated
post-hoc. (See Appendix E Figure 7). The main effect, Grade was not significant, F(1, 35) =
0.12, p = .734, indicating there were no significant differences of LEAF-Q Score by Grade
levels. The main effect, Sport was not significant, F(3, 35) = 0.53, p = .667, indicating there were
no significant differences of LEAF-Q Score by Sport levels.
Post-hoc
There were no significant effects in the model for the first ANOVA. As a result, posthoc
comparisons were not conducted.
For the second ANOVA, a t-test was calculated between each group combination to
further examine the differences between LEAF-Q score and Ethnicity, Grade, and Sport based on
an alpha of .05. The Tukey HSD p-value adjustment was used to correct for the effect of multiple
comparisons on the family-wise error rate. For the main effect of Ethnicity, the mean of LEAF-Q
Score for White (M = 7.74, SD = 4.22) was significantly smaller than for Other (M = 15.33, SD =
3.51), p = .036. No other significant effects were found.
Fisher’s Exact Tests
Introduction
A Fisher’s exact test was conducted to examine whether Sport and Scores Indicated
Referral were independent. The variable, Scores Indicated Referral, means that one of the two
questionnaires the participants filled out classified the participant as high risk (LEAF-Q ≥ 8,
EAT-26 ≥ 20, or Red Flag Eating Behaviors). There were four levels in Sport: Volleyball,
Basketball, Soccer, and Multi-Sport. There were two levels in Scores Indicated Referral: No and
Yes.

THE RED-S PROTOCOL

18

A second Fisher’s exact test was conducted to examine whether Grade and Scores
Indicated Referral were independent. There were two levels in Grade: Freshman and Sophomore.
There were two levels in Scores Indicated Referral: No and Yes.
A third Fisher’s exact test was conducted to examine whether Ethnicity and Scores
Indicated Referral were independent. There were five levels in Ethnicity: White, Black, Other,
American Indian or Alaskan Native, and Multi-Ethnic. There were two levels in Scores Indicated
Referral: No and Yes.
Results
The results of the first Fisher exact test were not significant based on an alpha value of
.05, p = .937, suggesting that Sport and Scores Indicated Referral could be independent of one
another. This implies that the observed frequencies were not significantly different than the
expected frequencies. (See Appendix E Table 5).
The results of the second Fisher exact test were not significant based on an alpha value of
.05, p = .170, suggesting that Grade and Scores Indicated Referral could be independent of one
another. This implies that the observed frequencies were not significantly different than the
expected frequencies. (See Appendix E Table 6).
The results of the Fisher exact test were not significant based on an alpha value of .05, p
= .211, suggesting that Ethnicity and Scores Indicated Referral could be independent of one
another. This implies that the observed frequencies were not significantly different than the
expected frequencies. (See Appendix E Table 7).
Follow Up Interview
Out of the 44 participants, 70% (n=31) had scores that indicated referral or further
evaluation. Only 39% (n=17) of all participants attended the follow-up interview to review their

THE RED-S PROTOCOL

19

results with the primary investigator. Of the 31 participants with a score indicating referral, 42%
(n=13) participated in the follow-up interview. Only one person utilized campus resources
(athletic trainer or Wellness Center) for education or help after the follow-up interview.
The follow-up interview was important because it allowed the principal investigator to
ask clarifying questions and identify areas of interest that were not previously known. All three
participants who marked their Ethnicity as “Other” were international students and participated
in the follow-up interview. “Other” students have higher LEAF-Q scores when compared to
“White” students: 15.33 compared to 7.74. When reviewing their LEAF-Q scores, it became
apparent that their gastrointestinal dysfunction was less likely related to RED-S/LEA and more
likely associated with the drastic change in diet after moving to the United States.
Project Impact- Clinical Significance
The implementation of the RED-S Protocol had a significant impact on the project site. In
the educational seminar, most female athletes at the project site were taught the warning signs of
RED-S and LEA and the importance of increasing calorie intake. In the follow-up interviews,
participants made multiple comments that indicated retention and learning for the project
investigator despite the absence of a pre- and post-test to assess understanding and retention.
After analyzing the project data and notifying the athletic department staff that 70% of
female athletes, irrespective of grade, sport, and ethnicity, were at risk for RED-S and LEA, they
immediately began to discuss how to assist them. A discussion surrounding the process to
implement the protocol at the beginning of the next school year when sports resume has begun.
Additionally, they began working to make it part of the athletic department’s standard practices.
This process included tentatively planning dates to educate and screen the athletes when they

THE RED-S PROTOCOL

20

return to campus in August 2024. It also included solidifying the roles and responsibilities of the
athletic department staff members in the protocol.
Discussion
There are four key findings from this project. First, the average female college athlete at
the project site is experiencing RED-S/LEA. Seventy percent of participants were deemed at
risk. According to the literature, the majority of female athletes experience these syndromes
throughout their careers (Logue et al., 2022). Second, student-athletes are not seeking help for
these syndromes. Only 39% of student-athletes met with the principal investigator to receive
their results and get appropriate guidance for their next steps. Only one participant utilized the
Wellness Center after meeting with the project investigator. No participant met with the Athletic
Trainer for direction, help, and support. Third, female athletes are at high risk for developing an
eating disorder or disordered eating. Twenty three percent of participant female athletes engaged
in red flag eating behaviors that could be concerning for an eating disorder. Fourth, screening
questionnaires are an efficient and cost-effective way to identify at-risk athletes and refer them to
the appropriate resources.
An additional finding found was those students who classified their Ethnicity as “Other”
were international students who scored higher on the LEAF-Q. If those students had not
participated in the follow-up interview, it could have been assumed that “Other” students were at
higher risk of developing RED-S/LEA. However, it is more likely that their LEAF-Q score was
falsely elevated due to the change in diet after moving to the United States, subjectively causing
increased bloating and cramping.
A significant limitation of this project was the low participation rate in the follow-up
interview. By not participating in the follow-up interview, student-athletes did not receive their

THE RED-S PROTOCOL

21

scores and guidance if they were deemed at risk. This project flaw could have been mitigated by
collecting contact information to briefly discuss the participant’s scores and the appropriate next
steps without requiring a follow-up interview.
Identifying the reasons why participants may hesitate to participate in follow-up
interviews is one of the objectives of future studies. Future studies should also explore the
barriers athletes must overcome to use free campus resources. In this project, only one
participant used campus resources.
RED-S is a common yet underdiagnosed syndrome that impacts a majority of female
athletes. This efficient and cost-effective protocol educates athletes about the signs and
symptoms of RED-S, LEA, and ED/DE and how to prevent or recover from them. It uses
validated screening tools to identify at-risk athletes and standardizes the management of females
suffering from these insidious syndromes. Immediate implementation of this protocol in all
institutions with female athletes is needed to safeguard them from short- and long-term
complications.

THE RED-S PROTOCOL

22
References

Cabre, H., Moore, S., Smith-Ryan, A., & Hackney, A. (2022). Relative energy deficiency in sport
(RED-S): Scientific, clinical, and practical implications for the female athlete. Deutsche
Zeitschrift Fur Sportmedizin, 73(7), 225–234. https://doi.org/10.5960/dzsm.2022.546
De Souza, M. J., Nattiv, A., Joy, E., Misra, M., Williams, N. I., Mallinson, R. J., Gibbs, J. C.,
Olmsted, M., Goolsby, M., Matheson, G., & Panel, E. (2014). 2014 Female athlete triad coalition
consensus statement on treatment and return to play of the female athlete triad: 1st International
Conference held in San Francisco, California, May 2012 and 2nd International Conference held
in Indianapolis, Indiana, May 2013. British Journal of Sports Medicine, 48(4), 289–289.
https://doi.org/10.1136/bjsports-2013-093218
Field, A. (2017). Discovering statistics using IBM SPSS statistics: North American edition. Sage
Publications
Garner, D., Olmsted, M. P., Bohr, Y., & Garfinkel, P. E. (1982). The eating attitudes test:
psychometric features and clinical correlates. Psychological Medicine, 12(4), 871-878.
doi:10.1017/S0033291700049163
Gonzalo, A. (2023). Dorothea Orem: Self-care deficit theory. Nurseslabs.
https://nurseslabs.com/dorothea-orems-self-care-theory/
Intellectus Statistics [Online computer software]. (2023). Intellectus Statistics.
https://statistics.intellectus360.com
Logue, D. M., Madigan, S. M., Melin, A., Delahunt, E., Heinen, M., Donnell, S.-J. M., & Corish, C.
A. (2020). Low energy availability in athletes 2020: An updated narrative review of prevalence,
risk, within-day energy balance, knowledge, and impact on sports performance. Nutrients, 12(3),
835. https://doi.org/10.3390/nu12030835

THE RED-S PROTOCOL

23

Mehta, J., Thompson, B., & Kling, J. M. (2018). The female athlete triad: It takes a team. Cleveland
Clinic Journal of Medicine, 85(4), 313–320. https://doi.org/10.3949/ccjm.85a.16137
Melin, A. K., Heikura, I. A., Tenforde, A., & Mountjoy, M. (2019). Energy availability in athletics:
Health, performance, and physique. International Journal of Sport Nutrition and Exercise
Metabolism, 29(2), 152–164. https://doi.org/10.1123/ijsnem.2018-0201
Melin, A. K., Tornberg, Å. B., Skouby, S., Faber, J., Ritz, C., Sjödin, A., & Sundgot-Borgen, J.
(2014). The LEAF questionnaire: A screening tool for the identification of female athletes at risk
for the female athlete triad. British Journal of Sports Medicine, 48(7), 540–545.
https://doi.org/10.1136/bjsports-2013-093240
NCAA. (2022). The state of women in college sports.
https://s3.amazonaws.com/ncaaorg/inclusion/titleix/2022_State_of_Women_in_College_Sports_
Report.pdf
Osborne, J., & Waters, E. (2002). Four assumptions of multiple regression that researchers should
always test. Practical Assessment, Research & Evaluation, 8(2), 1-9.
Petiprin, A. (2023). Orem’s self-care deficit nursing theory. Nursing Theory. Retrieved April 24,
2023, from https://nursing-theory.org/theories-and-models/orem-self-care-deficit-theory.php
Pituch, K. A., & Stevens, J. P. (2015). Applied multivariate statistics for the social sciences (6th ed.).
Routledge Academic. https://doi.org/10.4324/9781315814919
Rogers, M. A., Drew, M. K., Appaneal, R., Lovell, G., Lundy, B., Hughes, D., Vlahovich, N.,
Waddington, G., & Burke, L. M. (2021). The utility of the low energy availability in females
questionnaire to detect markers consistent with low energy availability-related conditions in a
mixed-sport cohort. International Journal of Sport Nutrition & Exercise Metabolism, 31(5), 427–
437. https://doi.org/10.1123/ijsnem.2020-0233

THE RED-S PROTOCOL

24

Rosswurm, M. A., & Larrabee, J. H. (1999). A model for change to evidence-based practice. Image:
Journal of Nursing Scholarship, 31(4), 317-322. https://doiorg.ezproxy1.lib.asu.edu/10.1111/j.1547-5069.1999.tb00510.x
Sim, A., & Burns, S. F. (2021). Review: Questionnaires as measures for low energy availability
(LEA) and relative energy deficiency in sport (RED-S) in athletes. Journal of Eating Disorders,
9(1), 41. https://doi.org/10.1186/s40337-021-00396-7
Tenforde, A. S., Beauchesne, A. R., Borg-Stein, J., Hollander, K., McInnis, K., Kotler, D., &
Ackerman, K. E. (2020). Awareness and comfort treating the female athlete triad and relative
energy deficiency in sport among healthcare providers. German Journal of Sports Medicine
71(3), 76–80. https://doi.org/10.5960/dzsm.2020.422

THE RED-S PROTOCOL

25
Appendix A Table A1
Evaluation Table for Quantitative Studies

Citation
Feltner et al.,
(2022),
Screening for
eating
disorders in
adolescents
and adults:
Evidence
report and
systematic
review for the
US
Preventative
Services Task
Force

Design/
Method/
Purpose
Design:
Systematic
Review
Purpose:
Review the
literature to
understand the
benefits and
harm of
screening adults
and adolescents
for EDs

Sample/Setting

Variables

Measurement/
Instrumentation

Data Analysis

Results/ Findings

N= 57

IV1: ED screening
tests
DV1: Accuracy of
screening tests for
ED
DV2: Effectiveness
of interventions for
improving health
outcomes
DV3: Harms of
interventions

Tools: SCOFF
questionnaire

Statistical Tests
Used:
None were
explicitly stated
in the article, but
the reviewed
literature used
statistical tests
and analysis.

DV1: SCOFF test
was researched most
often and has a high
specificity/sensitivity
for ED. Screening
tools are valid
clinical instruments.
False positive rate
20%
DV2:
Lisdexamfetamine
and topiramate are
effective in reducing
ED symptom
severity. Guided selfhelp was more
effective than
unguided self-help.
DV3: Pharmacologic
interventions were
associated with
adverse effects such
as tremors, nausea,
vomiting, &
confusion.

Design: Crosssectional study

N= 285
Demographics:
Head athletic

Statistical Tests
Used:

DV1: Women
trainers had greater
knowledge/awareness

Demographics:
English,
adolescent or
adult population,
primary care
type settings,
countries that are
very high on the
United Nations
Human
Development
Index

Yale-Brown
Obsessive
Compulsive Scale
Validity/
Reliability:
Both tests are highly
reliable and have
been validated in
multiple populations

Country: US
Funding:
Agency for
Healthcare
Research and
Quality, US
Department of
Health and
Human
Services
Bias: not
explicitly
stated or
implied
Kroshus et al.,
(2018),
Collegiate

IV1: Gender of
athletic trainer

Tools: Internally
created survey

Level of Evidence;
Application to practice;
Generalization
LOE: 1
Strengths: Broad
literature search.
Thorough.
Weakness: 2 key
questions were not
answered due to a lack of
research
Application: Screening
questionnaires are an easy
next step that can be
applied in any setting.

LOE: 1

THE RED-S PROTOCOL
athletic
trainers’
knowledge of
the female
athlete triad
and relative
energy
deficiency in
sport
Country: US

Funding: Not
stated
Bias: Possible
b/c authors are
likely
employed at
schools that
participated in
the study.
Funding was
not stated, so
bias could also
be attributed to
this.

Lodge et al.,
(2022),
Knowledge of
the female
athlete triad
and relative
energy
deficiency in

Purpose: To
determine Head
Athletic
Trainers’
knowledge of
FAT and RED-S
and to see the
variability of
screening and
referral actions
among the
different
national
collegiate
athletic
association
institutions

trainers at
National
Collegiate
Athletic
Association
institutions
Setting:
Collegiate
athletic training
departments
Exclusion: It
was not stated
but implied that
non-head ATs
were prohibited
from
participating.
Attrition: 0%

Design: Crosssectional
Purpose:
Understanding
the depth and
breadth of
knowledge about
FAT/RED-S

N= 260
Demographics:
175 athletes
55 coaches
30 trainers

IV2: Institution
Level
DV1: FAT/RED-S
Knowledge
DV2: Referral
DV3: Screening for
ED/DE/LEA/REDS
DV4: Prescribe
hormonal
contraception
Definitions:
Institution Level is
either Division 1, 2,
or 3 in the National
Collegiate Athletic
Association
Referral: Pt referred
to a sports medicine
physician,
counselor,
psychologist,
nutritionist, or
dietitian

Validity/
Reliability:
Not stated

IV1: Survey
DV1: FAT/RED-S
knowledge of
athletes
DV2: FAT/RED-S
knowledge of
coaches

Tools:
Internally created a
survey using
questions from
multiple validated
surveys.
Validity/

26

Fisher Exact
test
Mann-Whitney
Test
Kruskal-Wallis
Test

Statistical Tests
Used: Shapirowilk
Komogorovsmirnov test
1-way analysis
of variance

of FAT, including
diagnostic criteria.
Four men had not
heard of it. Over 60%
of respondents had
not heard of RED-S.
DV2: 47% of schools
always refer to a
sports medicine
provider if there is
menstrual
dysfunction. Division
1 schools were more
likely to refer athletes
to nutritionists.
DV3: Division 1
schools screen for
ED/DE more often
than 2 and 3 schools
(70% vs. 48% vs.
49%). Division 3
schools had a higher
frequency of annual
screenings (80%)
than 1 and 2 schools
(71% and 57%).
Division 2 and 3
schools were more
likely never to assess
menstrual function
(14.3% and 16.9%
vs. 4.7%)
DV4: Division 1
schools were more
likely to prescribe
hormonal
contraception.
DV1: lowest
knowledge and
confidence scores of
all three groups. The
only group that
scored higher on
RED-S knowledge
than FAT knowledge

Strengths: Large sample
size. Addressed many
variables.
Weakness: Limited
response rate possibly
overestimates RED-S
awareness. The internal
survey is not validated
and could cause response
bias.
Feasibility: At larger
institutions, it is more
feasible to implement
screenings because of the
availability of staff to
review screenings and
meet with athletes. The
US Preventative Services
Task Force cautions
against screenings if there
is no plan to refer and care
for an individual at risk
for LEA/RED-S.
Application: With a clear
and proper protocol, it
should be easily
applicable at institutions.
Especially if partnering
with campus health
services or community
partners.

LOE: 4
Strengths: Looked at
three different groups of
people, a large sample
size from diverse
locations
Weakness: Limited to
cross country, used a non-

THE RED-S PROTOCOL
sport among
female crosscountry
athletes and
support staff
Country: US
Funding: Not
stated
Bias: None
noted

Logue et al.,
(2019),
Screening for
risk of low
energy
availability in
athletic and
recreationally
active females
in Ireland
Country:
Ireland
Funding: Irish
Research
Council and
Sport Ireland
Bias: None
noted

amongst college
cross-country
athletes,
coaches, and
trainers

Setting:
National
Collegiate
Athletic
Association
schools

DV3: FAT/RED-S
knowledge of
trainers

Reliability:
Not stated

27

Tukey post hoc
analysis
Paired-samples ttest
Independent ttests

Exclusion: had
to be currently
involved in a
cross-country
program

Design: Crosssectional
Purpose:
Screening for
risk of LEA in
different
competition
levels of athletes
in Ireland

Attrition: 81
surveys were
improperly filled
out
N= 833
Demographics:
18-44 y/o;
international,
provincial/intercounty,
competitively,
and recreational
athletes; many
sports
Setting: selfadministered
online
questionnaire
Exclusion:
failing to finish
the questionnaire
Provincial/intercounty athletes
are similar to
semi-pro or D1
college athletes.
Competitive
athletes are
comparative to
lower levels of
college athletes.

IV1: LEAF-Q
DV1: LEA risk in
international
athletes
DV2: LEA risk in
provincial/intercounty athletes
DV3: LEA risk in
competitive athletes
DV4: LEA risk in
recreational athletes

Tools:
LEAF-Q sensitivity
78% specificity 90%
Validity/
Reliability:
Validated in athletes

Statistical Tests
Used:
Chi-square
analysis
Logistic
regression
analysis

DV2: The majority
of coaches had little
knowledge of FAT
and RED-S
DV3: Trainers had
more knowledge of
FAT than RED-S

DV1: 45% at risk
DV2: 47% at risk
DV3: 38% at risk
DV4: 33% at risk
40% of total
participants were at
risk for LEA
Every extra hour of
practice a week
increases the risk of
LEA by 1.06x

validated tool for
information gathering.
Feasibility: Interventions
need to be aimed at all
three levels of impact
Application: Create highimpact players to create
the most change

LOE:4
Strengths: large sample
size, many different
sports, results are
consistent with current
literature
Weakness: Utilized nonvalidated questions to
gather more information,
self-selection, and
response bias
Feasibility: LEAF-Q is an
easy tool to screen for
LEA in athletes
Application: Easy to
apply to college programs
as a preparticipation
physical or during periods
of intense activity

THE RED-S PROTOCOL
Rogers et al.,
(2021), The
utility of the
low energy
availability in
females
questionnaire
to detect
markers
consistent with
low energy
availabilityrelated
conditions in a
mixed-sport
cohort
Country:
Australia
Funding: AIS
High
Performance
Research
Fund,
University of
Canberra
Research
Institute for
Sport and
Exercise
Bias: Many
authors work
for or with the
AIS, which
could be a
source of bias
due to the
pressure to
produce
favorable
results.

Design: Crosssectional study
Purpose:
Determining if
the LEAF-Q can
identify
biological
manifestations
of LEA or FAT
in elite and preelite female
athletes

Sharps et al.,
2021.

Design: crosssectional study

N= 75
Demographics:
18-32 (mean 23)
years old,
Female, elite or
pre-elite athletes,
members of a
national sporting
organization
(athletics,
basketball,
boxing, netball,
rowing, water
polo,
weightlifting)
Setting:
Australian
Institute of Sport
Exclusion: Not
explicitly stated
Attrition: 0%

N=112
Demographics:

28

IV1: LEA/FAT
DV1: LEAF-Q
Score
DV2: DXA Scan
DV3: Energy
Availability
DV4: Psychiatric
Conditions
DV5: DE
DV6: Menstrual
function
DV7: Bloodwork
DV8: Time loss
Definitions:
DXA Scan – dualenergy x-ray
absorptiometry to
calculate
fat/lean/muscle
mass in individuals
Energy
Availability –
calorie availability
based on resting
metabolic rate and
fat-free mass.
Time Loss- missed
training/competition
time due to
gastrointestinal
issues, illnesses, or
physical injuries

Tools:
- LEAF-Q
(sensitivity 78%,
specificity 90%)
- SCOFF
Questionnaire
(validated in female
athletes)
- Internally created
questionnaire (Not
validated)
- DXA Scan
(validated in a wide
range of populations)
- Semistructured
diagnostic
psychiatric interview
using an abridged
version of the Mini
International
Neuropsychiatric
Interview
- Blood test

Statistical Tests
Used:
Spearman’s
correlation
coefficient

IV1: Age
IV2: Sport Level

1. Female athlete
screening tool

Shapiro Wilks

Wilcoxon rank
sum
Fisher’s Exact Test

DV1: LEAF-Q
scores cannot
diagnose an
LEA/RED-S but
have a high negative
predictive value.
DV2: Lumbar and
Femoral Z scores are
highly predictive of
LEA/RED-S
DV3: True energy
availability was
difficult to calculate
due to poor calorie
tracking and
expenditure
compliance in certain
sports. <30 kcal/kg
fat-free mass/day was
indicative of LEA
DV4: 70% of
participants did not
have a mental health
diagnosis.
DV5: Increased DE
risk correlated w/ a
LEAF-Q score >8
DV6: Menstrual
dysfunction has a
negative predictive
value of 94.7%. A
positive value of
30.8%. 80%
sensitivity and 59.3%
specificity.
DV7: Time loss
100% negative
predictive value
DV8: Blood works
100% negative
predictive value for
low free
triiodothyronine.
16% of athletes were
at risk for ED. 44%

LOE: 4
Strengths: Large sample
size and results are
consistent with the current
literature. It was one of
the first studies to
examine multiple sports at
one time and increase the
validity of the LEAF-Q in
non-endurance sports
athletes.
Weakness: Difficult to
determine actual EA.
Feasibility: Using the
LEAF-Q to determine
which athletes are at low
risk for RED-S is very
feasible, and it is a
relatively short survey
with extremely high
negative predictive value.
Application: Initial
screening is easily
applicable, and the
follow-up for not low-risk
athletes would be more
time and resourceintensive.

LOE: 4

THE RED-S PROTOCOL
Prevalence of
Disordered
Eating, Eating
Disorders, and
Risk of Low
Energy
Availability in
Professional,
Competitive,
and
Recreational
Female
Athletes based
in the United
Kingdom

Purpose: To
determine the
prevalence of
ED/DE and LEA
in female
athletes in the
UK and if age
and level of
competitiveness
(recreation,
competitive,
professional) are
indicators for
ED/DE or LEA

Country UK
Funding: No
funding was
received
Bias is
unlikely due to
the subject
population and
relationship to
researchers,
and the
researchers do
not stand to
gain anything
from a specific
result.
Sim et al.,
2021, Review:
questionnaires
as measures
for low energy
availability
(LEA) and
relative energy
deficiency in
sport (RED-S)
in athletes

Systematic
Review
Purpose: To
identify and
critique
validated
LEA/RED-S risk
questionnaires

Female athletes
in the UK. Age
18-40. Rec,
competitive, and
professional
athletes in many
different sports.
Setting:
Online,
anonymous
questionnaire
Exclusion:
incomplete
questionnaire

DV1: LEA
DV2: ED/DE Risk

(validated in female
athletes)
2. Low energy
availability in
females –
questionnaire
(sensitivity 78%,
specificity 90%)

29

One way
ANOVA
Fischers Exact
Spearmans Rank
Correlation

Attrition: 0%
129 responses,
17 incomplete =
total 112
participants

N= 13
Demographics:
Unique
questionnaires
that were
focused on
athletes and were
meant to identify
LEA/REDS/FAT/ED/DE

were at risk for DE.
53% were at risk for
LEA.
Pro athletes have
higher rates of ED
and LEA.
Comp athletes have
higher rates of DE.
25-30-year-olds had
the highest rate of ED
and LEA; 32%, 61%.
18-24-year-olds had
the highest rate of
DE, 56%.
All competitive
levels of female
athletes are at risk for
DE and LEA.

IV1: Individuals w/
LEA/REDS/FAT/DE/ED
DV1: Screening
Questionnaires
Definitions:
See study one for
LEA/ED/DE
definitions.
RED-S is a
syndrome
characterized by

Tools:
1. Brief Eating
Disorder in Athletes
Questionnaire ––
82.1%, & 84.6%
2. Eating Disorder
Examination
Questionnaire- 83%
& 96%
3. Eating Disorder
Inventory- Drive for

Statistical Tests
Used: This paper
did not use
statistical tests
because the
questionnaires
had already been
evaluated.

1. Validated in elite
female adolescent
athletes.
2. Good tool to look
for LEA and possibly
RED-S in males and
females.
3. LEA can be
inferred based on the
scores.

Strengths: Utilized
validated screening tools.
Data is consistent with
other studies. A wide
variety of sports are
represented.
Weakness: Anonymous,
convenience sample,
which could lead to false
responses. The sample
sizes of pro athletes and
general athletes were both
small.
Conclusions: 16% of
athletes were at risk of
ED. 44% likely had DE.
53% had LEA.
Feasibility:
Implementing these
screenings is highly
feasible. Low cost and
minimal time
requirements can help
prevent long-term, lasting
damage to athletes.

LOE: 1
Strengths: Thorough and
the first systematic review
of these screenings.
Addresses their
shortcomings.
Weakness: Only searched
one database.
Feasibility: Depends on
the screening that is used.
Some are lengthy, while
some are short. Overall, it

THE RED-S PROTOCOL
Country: Not
stated
Funding:
None
Bias: Likely
none. Not
attached to an
organization,
no competing
interests, or
obvious
opportunities
to benefit from
a specific
result.

Exclusion: Had
to be within the
last 10 years,
English, and
full-text
Attrition: NA

low-calorie intake,
high-calorie output,
and physiological
manifestations of
calorie deficiency
FAT is the Female
Athlete Triad, a
syndrome that
presents with
amenorrhea,
osteoporosis, and an
ED.

Thinness Score 86%
& 80%
4. Eating Disorder
Screening for
Primary Care- 100%
& 71%
5. Female Athlete
Triad Risk Scale –
NV
6. Female Athlete
Triad Screening
Questionnaire- NV
7. Low Energy
Availability in
Females
Questionnaire- 78%
& 90%
8. Meal Attitudes and
Body Weight
Questions- NV
9. RED-S Risk
Measurement for
Cyclists- NV
10. RED-S Specific
Screening Tool
(female version)Female version
validated
11. Sport-specific
Energy Availability
Questionnaire and
Interview- Validated
internally
12. Three-Factor
Eating Questionnaire
– Dietary Cognitive
Restraint- 72% &
70.1%
13. Triad Consensus
Panel Screening
Questions by the
Female Athlete
Coalition- NV

30

4. Best used with
another screening
tool.
5. Not validated, but
great potential for
mass screening
6. Measures triad risk
before the
competitive season.
7. Used in large
groups. It is best if
used with ED/DE
screening tools as
well.
8. Not validated but
has the potential for
easy mass screening.
9. Short and concise.
It is not validated and
focuses only on
symptoms.
10. Produces risk
score. Designed for
both sexes and a
wide age range,
which fits a gap in
the literature but
needs more
validation.
11. Specific to
cycling. Attempts to
fill a gap in the
literature for male
screening. It is not
validated.
12. May
underestimate the
risk of LEA in
females.
13. Pre-participation
screen but cannot be
used independently
without follow-up
assessments. Not
validated.

is an easy intervention to
implement in any sports
training program.
Application: Highly
applicable. Finding the
right screening tool could
be challenging, but it is
easy to apply when it is
found.

THE RED-S PROTOCOL
Stewart et al.,
(2019), The
female athlete
body project
study: 18month
outcomes in
eating disorder
symptoms and
risk factors
Country: US
Funding:
National
Institute of
Mental Health
Bias: Funding
is from a
mental health
organization,
and so there
could be a
benefit to the
authors to
conclude that
this
intervention
benefits mental
health
Tenforde et al.
(2020),
Awareness and
comfort
treating the
female athlete
triad and
relative energy
deficiency in
sport among
healthcare
providers.
Country: US
Funding: No
funding was
provided
except for Dr.

Design: RCT
Purpose: Does
the Female
Athlete Body
project reduce
the symptoms,
severity, and
occurrence of
EDs

N= 481
Demographics:
Female college
athletes, 17-27
Setting: 3
colleges across
the US
Exclusion:
failure to attend
follow-up
meetings, being
removed from an
athletic team or
the school

IV1: Female
Athlete Body
Project intervention
DV1: EDE-Q
scores
DV2:
objective/subjective
binge eating
episodes
DV3: New onset
ED

Tools:
EDE-Q

IV1: Screening tool
DV1: Knowledge
of FAT
DV2: Knowledge
of RED-S
DV3: Comfort
treating FAT
DV4: Comfort
treating RED-S

Tools:
Internally created
survey

Validity/
Reliability: EDE-Q
values are not
explicitly stated, but
it is a valid and
reliable tool

31

Statistical Tests
Used:
Priori power
analysis
Power
calculations
Linear mixedeffects models

DV1: All scores were
lower, but the only
statistically
significant subscore
was dietary restraint
DV2: There were
fewer episodes of
objective/subjective
binge-eating episodes
DV3: Prevention
levels were lower but
not statistically
significant

Attrition: 24%
at 18 months

Design: crosssectional
Purpose:
Ascertain the
awareness and
comfort in
treating REDS/FAT in
healthcare
professionals

N= 163
Demographics:
20-60+. 54%
female. 33%
were physicians,
60% were allied
health
professionals
Setting: Sports
Medicine
Conference
Exclusion: Had
to be a certified
health
professional

Validity/
Reliability:
It is not stated, but it
is based on valid and
reliable
questionnaires

Statistical Tests
Used:
Chi-squared
testing
Wilcoxon rank
sum

DV1: 76% had heard
of it.
DV2: 29% had heard
of it. Physicians were
most likely to have
heard of it, especially
with a fellowship.
DV3: International
practitioners were
more comfortable
than US practitioners
DV4: Physicians and
non-physicians had
similarly low rates.

LOE: 2
Strengths: A large sample
size helps fill a gap in the
literature.
Weakness: Most of the
data is related to binge
eating or bulimia,
anorexia, and other
disorders are not
mentioned
Feasibility: It would be a
great tool to implement,
but it is not commercially
available due to continued
research. Application:
Peer-led group
discussions can positively
impact ED severity and
symptoms.

LOE:4
Strengths: Fits a gap in
the literature. Decent
sample size.
Weakness: Response bias
and overrepresentation of
specific populations in the
study. Did not use a
validated tool.
Application:
Interventions should be
aimed at educating
providers.

THE RED-S PROTOCOL
Hollander’s
participation.
Bias: None
noted
TorresMcGehee et
al., 2021.
Energy
Availability
With or
Without Eating
Disorder Risk
in Collegiate
Female
Athletes and
Performing
Artists
Country: US
Funding: not
stated
Bias: All
authors are
athletic
trainers, and
all the subjects
came from one
school. There
could be bias
with an
athletic trainer
skewing
results at their
school for
recognition,
funding, or
other
imponderables.

Design:
Cross-sectional
study
Purpose:
Determine LEA
with and without
ED risks in
college athletes.

32

N= 121
Demographics:
D1 female
athletes and
ballet dancers
17-22 y/o
Setting:
Free living in
sport-specific
settings at a D1
university in the
US
Exclusion: not a
current
athlete/artist,
history of eating
disorder, injury
preventing them
from exercising,
or did not
complete the
study
Attrition: 0%

IV1: Calories
consumed
DV1: LEA
DV2: ED Risk
Definitions:
LEA: when energy
intake is less than
energy output in an
athlete, putting
them in a low
energy availability
state
ED Risk is when a
subject presents
with attitudes and
behaviors that may
put them at risk for
having or
developing an
eating disorder.
ED is a diagnosed
eating disorder such
as anorexia,
bulimia, etc.
DE is disordered
eating, which is a
subclinical eating
disorder.

Tools:
1. Microlite
MedGem
2. ESHA food
processor
3. Sensewear
armband
4. Eating Disorder
Inventory-3
5. Eating disorder
inventory-3 symptom
checklist
Validity/
Reliability:
1. Reliability 0.910.97
2. Not reported
3. 0.81
4. 0.90-0.97
5. Not reported

Statistical Tests
Used:
ANOVAs
Post hoc Tukey
test
Chi-square
analysis

DV1: 81% displayed
LEA
DV2: 76% had ED
risk

LOE: 4
Strengths: Controls in
place to eliminate
response bias from
subjects. Very thorough.
Utilized highly reliable
instruments. Obtained a
substantial sample size—
0% attrition.
Weakness: Only at one
institution. Possible
response bias on food
logs. It occurred during
the pre-season for one
team and the season for
the other teams—length of
monitoring.
Conclusions: Most
female athletes at D1
schools suffer from LEA
and ED risk.
Feasibility: This would
be hard to implement at
other institutions with
smaller budgets. The
energy expenditure tools
cost money. The Eating
Disorder Inventory-3 and
symptom checklist are
lengthy and could lead to
inaccurate responses.

THE RED-S PROTOCOL

33
Appendix A Table A2
Synthesis Table

Study
(Author, year)

Feltner et al.,
2022

Kroshus et
al., 2018

Lodge et
al., 2022

Logue et al.,
2019

Rogers et al.,
2021

Sharps et al.,
2021

Sim et al., 2021

Stewart et al., 2019

Tenforde et al.,
2020

Torres-McGehee et
al., 2021

Design

Systematic
Review
Understand
the benefit and
harm of
screening for
ED

Crosssectional
FAT/REDS
knowledge,
screening
rates, and
referral
actions

Crosssectional
FAT/REDS
knowledge

Crosssectional
Understand
the LEA risk
for different
levels of
competition
for FAs

Cross-sectional

Crosssectional
Prevalence
of
LEA/ED/DE

Systematic
Review
Identify and
critique
screening
questionnaires
for LEA/REDS/ED/DE

Randomized
controlled trial
Female Athlete
Body project
influence on ED
symptoms and
severity

Cross-sectional

Cross-sectional

Ascertain the
awareness and
comfort of
treating REDS/FAT

Prevalence of LEA
w/ and w/o ED risks

57
N/A
ED

285
N/A
Head AT
(D1-D3)

260
18-47
College
Crosscountry
FAs,
Coaches,
and ATs

833
18-44
Ireland
International
ProvincialInter- County
Competitive
Recreational

75
18-32
Elite and PreElite Athletes

112
18-40
FA in the
UK (3
levels)

13
Within ten years
6 -ED
3 -FAT
2- LEA
2- REDS

481
17-27
3 D1 Schools with a
Broad Array of
Sports Teams

163
20-65+
Healthcare
Professionals

121
17-22
D1 FA and Dancers

X
X
X

X

X
X
X

X

X
X
X

X

X

X
X

Purpose

Sample
n subjects
M-Age
Other variables

Can the LEAFQ identify
biological
manifestations
of LEA/FAT

Target Demographic
Athletes
Coaches
Health
Professionals
Screening Tools
LEAF-Q
Other Valid
Survey
Internal Survey
Measured Outcomes
LEA
Prevalence
RED-S/FAT
Knowledge

X
X
X

X

X
X
X

X

X

X

X

X
X

X
X

X
X

X

X

X

X
X

THE RED-S PROTOCOL

34
Appendix B
Models and Frameworks

Figure B1
Orem’s Self-Care Nursing Theory

(Gonzalo, 2023)

THE RED-S PROTOCOL

35
Appendix B

Figure B2
Rosswurm and Larrabee’s Model for Evidence-Based Practice

(Rosswurm & Larrabee, 1999)

THE RED-S PROTOCOL

36
Appendix C
Questionnaires

Figure C1
Athlete Questionnaires

THE RED-S PROTOCOL

37

THE RED-S PROTOCOL

38

THE RED-S PROTOCOL
Figure C2
Eating Attitudes Test

39

THE RED-S PROTOCOL

40

THE RED-S PROTOCOL

Appendix D
RED-S Protocol Logic Model

41

THE RED-S PROTOCOL

Appendix E
RED-S Protocol Budget Breakdown

Item

Cost Breakdown

Total Cost

Gas

Approx 400 miles/24 mpg =
16.7 gallons
16.7 gallons*$4.21/gallon =
$15/meal * 3 meals

$70.2

3 Coach/Staff Questionnaire
Pages * 15 people = 45 prints
*$0.58 = $26.10

$110.10

Investigator Meals
Survey Printing

Participant Snacks

Investigator Lodging (if
needed)
Coach/Staff Time
Wellness Center Counseling
Athletic Trainer

2 Athlete Questionnaire
Pages (double sided) *75
people = 150 double-sided
prints * $0.56 = $84
3 Cases of water from
Costco*$3.50
3 Variety boxes of chips from
Costco *$23.99
IHG Hotel Military Rate 1
Night
Unknown Rate * 1-2 Hrs
Salary Employees, Unknown
Quantity of Time
Salary Employee, Unknown
Quantity of Time

$45

$82.47

$119

Total ~$426.77

42

THE RED-S PROTOCOL

43

Appendix F
Statistical Tables and Figures

Table F1
Descriptive Statistics of Participants
Descriptive Statistics of Participants
Variable
Ethnicity
White
Black
Other
American Indian or Alaskan Native
Multi-Ethnic
Grade
Freshman
Sophomore
Sport
Volleyball
Basketball
Soccer
Multi-Sport
Note. Due to rounding errors, percentages may not equal 100%.

n

%

34
1
3
1
5

77.27
2.27
6.82
2.27
11.36

30
14

68.18
31.82

9
13
21
1

20.45
29.55
47.73
2.27

Table F2
Questionnaires Scores and Distribution
Questionnaires Scores and Distribution
Variable

M

SD

n

Min

Max

LEAF-Q Score

8.50

4.44

44

0.00

19.00

EAT-26 Score

8.07

8.95

44

0.00

37.00

Table F3
ANOVA Table for EAT-26
ANOVA Table for EAT-26 Score by Ethnicity, Grade, and Sport
Term
Ethnicity
Grade
Sport

SS
204.16
47.79
187.84

ηp2
4 0.62 .652 0.07
1 0.58 .451 0.02
3 0.76 .524 0.06

df

F

p

THE RED-S PROTOCOL
Residuals 2,883.60 35

44

Table F4
ANOVA Table for LEAF-Q
ANOVA Table for LEAF-Q Score by Ethnicity, Grade, and Sport
Term

SS

Ethnicity 208.56
Grade
2.05
Sport
27.67
Residuals 613.27

ηp2
4 2.98 .032 0.25
1 0.12 .734 0.00
3 0.53 .667 0.04
35
df

F

p

Table F5
Observed and Expected Frequencies for Fisher’s Exact Test For Scores Indicated Referral and Sport
Observed and Expected Frequencies of the First Fisher’s Exact Test
Scores Indicated Referral
Sport
No
Yes
p
Volleyball
2[2.66]
7[6.34] .937
Basketball
4[3.84]
9[9.16]
Soccer
7[6.20]
14[14.80]
Multi-Sport
0[0.30]
1[0.70]
Note. Values formatted as Observed[Expected].
Table F6
Observed and Expected Frequencies for Fisher’s Exact Test For Scores Indicated Referral and Grade
Observed and Expected Frequencies of the Second Fisher’s Exact Test
Scores Indicated Referral
Grade
No
Yes
p
Freshman
11[8.86]
19[21.14] .170
Sophomore
2[4.14]
12[9.86]
Note. Values formatted as Observed[Expected].
Table F7
Observed and Expected Frequencies for Fisher’s Exact Test For Scores Indicated Referral and Ethnicity
Observed and Expected Frequencies of the Third Fisher’s Exact Test
Ethnicity

Scores Indicated Referral
No
Yes

p

THE RED-S PROTOCOL
White
12[10.05]
Black
0[0.30]
Other
0[0.89]
American Indian or Alaskan Native
1[0.30]
Multi-Ethnic
0[1.48]
Note. Values formatted as Observed[Expected].

22[23.95] .211
1[0.70]
3[2.11]
0[0.70]
5[3.52]

Figure F1
Q-Q scatterplot for normality of the residuals for the regression model in ANOVA 1
Q-Q scatterplot for normality of the residuals for the regression model in ANOVA 1

Figure F2
Q-Q scatterplot for normality of the residuals for the regression model in ANOVA 2
Q-Q scatterplot for normality of the residuals for the regression model in ANOVA 2

45

THE RED-S PROTOCOL

Figure F3
Residuals scatterplot testing homoscedasticity in ANOVA 1
Residuals scatterplot testing homoscedasticity in ANOVA 1

Figure F4
Residuals scatterplot testing homoscedasticity in ANOVA 2
Residuals scatterplot testing homoscedasticity in ANOVA 2

46

THE RED-S PROTOCOL

Figure F5
Studentized residuals plot for outlier detection in ANOVA 1
Studentized residuals plot for outlier detection in ANOVA 1

Figure F6

47

THE RED-S PROTOCOL
Studentized residuals plot for outlier detection in ANOVA 2
Studentized residuals plot for outlier detection

Figure F7
Means of LEAF-Q Score by Ethnicity with 95.00% CI Error Bars
Means of LEAF-Q Score by Ethnicity with 95.00% CI Error Bars

AI/AN

Note. AI/AN are abbreviations for American Indian/Alaskan Native

48