The IOC made recommendations for recording and reporting epidemiological data on injuries and illness in sports in 2020, but with little, if any, focus on female athletes. Therefore, the aims of this supplement to the IOC consensus statement are to (i) propose a taxonomy for categorisation of female athlete health problems across the lifespan; (ii) make recommendations for data capture to inform consistent recording and reporting of symptoms, injuries, illnesses and other health outcomes in sports injury epidemiology and (iii) make recommendations for specifications when applying the Strengthening the Reporting of Observational Studies in Epidemiology-Sport Injury and Illness Surveillance (STROBE-SIIS) to female athlete health data.
In May 2021, five researchers and clinicians with expertise in sports medicine, epidemiology and female athlete health convened to form a consensus working group, which identified key themes. Twenty additional experts were invited and an iterative process involving all authors was then used to extend the IOC consensus statement, to include issues which affect female athletes.
Ten domains of female health for categorising health problems according to biological, life stage or environmental factors that affect females in sport were identified: menstrual and gynaecological health; preconception and assisted reproduction; pregnancy; postpartum; menopause; breast health; pelvic floor health; breast feeding, parenting and caregiving; mental health and sport environments.
This paper extends the IOC consensus statement to include 10 domains of female health, which may affect female athletes across the lifespan, from adolescence through young adulthood, to mid-age and older age. Our recommendations for data capture relating to female athlete population characteristics, and injuries, illnesses and other health consequences, will improve the quality of epidemiological studies, to inform better injury and illness prevention strategies.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Injury and illness surveillance is a fundamental element in our efforts to protect the health of athletes. Hence, in 2020 the IOC published a consensus statement that describes standards to monitor and report health problems in sports.1 This consensus aims to ensure consistency in the definitions and methods used, and to guide the collection of comparable epidemiological data across studies. Since then, several sports-specific2–4 and population-specific5 extensions have been produced, further supporting the appropriate and consistent application of the IOC recommendations across different settings.
Consensus statements were traditionally developed and applied to record injuries without focused consideration of the female athlete.6 Indeed, the 2020 IOC consensus statement1 does not mention the female athlete in its recommendations. Historically, injury and illness data that inform the development of injury surveillance systems and consensus statements are typically from male athletes, with such systems then being more frequently used in men’s sport.6 However, female athletes have additional specific biological, sociocultural and environmental considerations that could impact sports exposure or health outcomes. For example, circulating concentrations of both endogenous and exogenous oestrogen and progesterone influence several health conditions,7 which vary with events (eg, puberty, pregnancy, menopause) and across life stages (eg, adolescence, young adulthood, mid-age). Postpuberty population characteristics are rarely reported but may influence injury and illness onset and recovery.8–10 Breast health issues likely go unreported as, like other body regions, the breast does not have a specific diagnosis category in commonly used coding systems11 and, until 2020, did not appear in these coding systems at all. Therefore, female-specific health risks across the lifespan remain largely undocumented, with limited quality data on female athlete health.
To help achieve high quality and consistent female athlete injury and illness epidemiological data, an agreed taxonomy of health domains for categorising health problems is warranted. Using an agreed taxonomy could help identify female-specific injury and illness patterns, and their health and performance interactions. Furthermore, improved reporting methods would aid in determining female-specific risk factors and mechanisms for physical and mental health problems, and lead to female-specific evidence-informed prevention initiatives.
The aims of this supplement to the IOC consensus statement are to: (i) propose a taxonomy for categorisation of female athlete health problems across the lifespan; (ii) make recommendations for data capture to inform consistent recording and reporting of symptoms, injuries, illnesses and other health outcomes in sports injury epidemiology and (iii) make recommendations for specifications when applying the Strengthening the Reporting of Observational Studies in Epidemiology-Sport Injury and Illness Surveillance (STROBE-SIIS) to female athlete health data. This document is intended to be read and applied in conjunction with the IOC consensus statement, rather than as a stand-alone document.
Scope of this supplement
This supplement focuses on cisgender female athletes (assigned female at birth) due to the sex-specific health issues outlined. We also acknowledge mental health and the gendered sports environment, which may influence injury risk12 and health outcomes.13 14 These are not unique to cisgender female athletes, but are more prevalent than in cisgender male athletes.14 Additionally, women and girls with variations of sex development, transgender women and girls and gender diverse athletes may experience some of the same health concerns and gendered experiences as cisgender female athletes. Therefore, much of the data capture and reporting recommendations outlined in this paper will also apply beyond cisgender female athletes. However, a specific and comprehensive supplement should be written in the future to support athletes with variations of sex development and/or who are transgender or gender diverse. The detail required to sufficiently cover the diversity of unique health considerations is beyond the scope of the current paper.
The process of writing this supplement started in May 2021, and followed the six-step process detailed in figure 1. Throughout the process, various expert authors were suggested by the core team and invited by the lead authors (ISM and EV). Two invited experts did not agree to take part. This resulted in a panel of 25 international experts who were selected based on their involvement in various academic and clinical disciplines across different sports. Our panel includes epidemiologists, physiotherapists, sport and exercise physicians, psychologists, sports scientists, an endocrinologist, former elite athletes and coaches with expertise in female athlete or female para athlete health. Six panel members (MM, BC, WD, CAE, AJ, EV) also contributed to the original IOC consensus group.
The working groups were instructed to meet and discuss their tasked theme via video conference as the work was undertaken during the COVID-19 pandemic. Subsequent communication within the working groups was conducted through email or video conference as necessary, leading to the creation of initial drafts for their tasked theme. Each working group was instructed to reach agreement on their initial drafts and ISM and EV subsequently compiled the initial drafts into a first complete draft of this supplement. This was circulated to the entire panel via email and discussed in a group video conference, during which we discussed necessary revisions. All panel members were invited to review the manuscript and suggest edits, and to participate in further video conferences, until consensus was reached. In total, there were seven iterations of the manuscript.
Current and former female athletes were involved in a ‘patient and public involvement’ interview. Athletes from various sports and with lived experiences across the female health domains were identified via the authors and invited by email to participate. For some health domains, we could not find athletes willing to share their experiences. We looked for experiences previously shared in the public domain for those health domains. Before their interview, athletes were provided with a summary of the project, an explanation of the health domains and their role within the project and a draft of this paper. The interview was meant only to allow for a practical illustration of the domains; each athlete only shared about the domain they had experience with. During the interviews, athletes were asked how they experienced the domain during their career and its impact on their health and performance. The interviews were open conversations and hence, no interview guide was necessary. Quotes were selected to illustrate their experiences and were included in the manuscript following review by the interviewed athlete. All athletes provided consent to include their lived experiences and specifically stated they wanted their names to be included in the acknowledgements of this manuscript. They were also invited to review the final version of this text.
Supplemental recommendations to the IOC consensus on injury and illness surveillance
The IOC consensus defines an athletic health problem as ‘any condition that reduces an athlete’s normal state of full health, irrespective of its consequences on the athlete’s sports participation or performance or whether the athlete sought medical attention’.1 The injury and illness definitions and examples provided by the IOC recommendations are universal and apply to female and male athletes. However, for the female athlete, we should also account for specific health-related domains that may influence sports participation while the athlete is—in fact—‘healthy’ (eg, the menstrual cycle, pregnancy or menopause). Therefore, using an adaptation of the recommendations by Dodd et al,15 we propose a taxonomy for categorisation of health problems in female athletes, which, when combined with sports participation as the primary risk exposure, may influence overall health or injury and illness risk in the short and long term (tables 1 and 2). In this taxonomy, health problems are grouped into 10 domains that have the potential to impact female athletes’ sports participation, performance or health outcomes, irrespective of their state of full health or whether they sought medical attention. We provide domain definitions in table 1 to enable consistent application of the taxonomy. Each of the 10 domains identified in table 1 offers a unique lens for observing, classifying and documenting a range of health problems that are relevant to female athlete health and performance. While this approach complements the recommendations described in the IOC consensus,1 it extends the reporting requirements to include female athlete health domains.
Domains of female health, associated health problems and their relationship with sports participation
In table 2, we provide an overview of the 10 domains of female health we identified, with frequently associated consequences or health problems, and examples of potential direct and indirect effects on performance and participation. All health domains are illustrated with athletes’ lived experiences (table 3). We have provided abbreviations for each domain to enable consistent reporting across epidemiological studies (table 1).
There are likely to be several ways these health domains and sports participation interact to affect health problems (ie, injury or illness). Some health domains may not necessarily influence sports participation but may be an intermediate variable on the causal pathway between sports participation and health problems (mediator). Other health domains may confound (distort) the relationship between sports participation and health problems, while also affecting participation. Health domains may also be effect modifiers in the causal pathway between participation and health problems. That is to say that the relationship between sports participation and the health problem may differ by the level of a particular health domain (eg, stage of pregnancy or menopause). We have presented female-specific examples for each of these potential relationships between health domains and health problems (ie, causal, mediation, confounding and effect modification) in figure 2.
Classifying female health domains and their related health problems and sports participation
The 10 female health domains proposed will aid improvements in (i) accurately classifying and categorising diagnoses for female athlete monitoring, reporting and/or research, (ii) facilitating recognition of temporal trends and between-group comparisons of female athletes (eg, different teams, leagues, sports) or risk factor studies and (iii) creating databases from which cases can be extracted for research on specific types of injuries and illnesses in female athletes.1 In some health domains, there is a need for further clarification to enable more granular, relevant data on specific health problems to be documented (eg, breast injury, health effects of sexual assault, relative energy deficiency in sport).
We also consider including the 10 domains of female athlete health and their associated health problems in the context of subsequent health problems. These domains and health problems can be recorded within a surveillance system, alongside injuries and illness. Examples of hypothetical prospective injury and illness data from such a system for athletes in a team are shown in figure 3. The classification and definitions of subsequent health problems in sport should be adopted, per the IOC consensus recommendations,1 with inclusion of female-specific data.
Population baseline characteristics
To improve understanding of the factors predisposing female athletes to female-specific health risks, baseline data should be more detailed than the minimum demographic data recommended by the IOC consensus statement1 (eg, date of birth, sex, gender, level of competition). For example, concerning the sports environment, female and male athletes playing at the same level of competition do not necessarily have similar years of sports exposure16 and this may influence injury risk factors (eg, movement patterns17). In general, where it concerns the professionalism of sport, differences apply too (eg, differences in financial rewards and the level and quality of medical support). Other factors that are not recorded for female or male athletes, but may impact sports exposure and health outcomes, are caregiving responsibilities outside sport and time dedicated to sport and/or dual career within the day. We acknowledge that it would be good practice for our recommendations to be recorded for all athletes and this would help further equity across many aspects (eg, gender, culture and ethnicity). But in the context of our paper, we propose that the following should be recorded for female athletes: level of competition, years of exposure to the sport, caregiving responsibilities and contribution of sport to earning a living.
The population baseline characteristics required for research should also consider the specific life stage of the athlete. Standardised definitions of puberty, menopausal status and other characteristics related to domains of female health (eg, pregnancy, postpartum) have been reported elsewhere.7 We recommend using these to define individuals and cohorts under investigation. We identify three overarching stages and example baseline characteristics that could be collected in table 4. The overarching stages are: peripuberty, adult and perimenopause. Peripuberty spans the period when the onset of menarche may occur (prepuberty and puberty), adult spans postpuberty and premenopause and perimenopause spans the period when menstruation surceases (menopause to postmenopause).7 As information relevant to each domain is likely to change over time, it is necessary to revisit certain baseline characteristics elements at the time of injury and illness to ascertain any change in status (eg, pubertal status, menorrhoea, pregnancy and menopause status). The elements of the relevant female health domain will depend on the life stage covered by the study (table 4).
Data collection methods
Given the variety of data relating to each domain of female health, only the data required to address the specific research questions should be collected. Considering the potentially sensitive nature of some female athlete health information, athletes should be provided with an explanation for why certain data are being collected and what efforts have been made to limit the amount and type of required data. If menstrual cycle-related data are collected, the education and language used in any verbal discussion or written materials should normalise the menstrual cycle rather than medicalise it. Consideration should also be given to who (eg, health professional, researcher or coach) should collect female athlete health data, given the potentially sensitive nature of many variables.18 This individual may vary across different sports contexts and religio-cultural backgrounds.
Where they exist, validated questionnaires should be used to capture data relating to female health problems in each domain (eg, the Australian Pelvic Floor Questionnaire19 for pelvic floor health), alongside appropriate characteristics (eg, number of pregnancies, mode of deliveries for childbirth). There are currently no validated questionnaires to record menstrual cycle information. Without standardised data collection methods, researchers must consider the most pertinent menstrual cycle information for their study (table 4).7
Data management and sharing
Integrating health, illness and injury reporting may be helpful to avoid the repetition of data entry and burden on athletes. However, this must be balanced with data confidentiality and sensitivities surrounding sharing such data with researchers and support staff involved in each system. Therefore, clarification of who is involved in each system is needed, in addition to obtaining informed consent from athletes for any data sharing across systems. For example, where strength and conditioning coaches or nutritionists record menstrual cycle data, this information may be shared (with consent) with medical personnel and integrated with the recording of health outcomes (eg, injuries and illnesses). Considering the sensitivity of data collected, it is important to communicate that data cannot be used for any reason (eg, team selection) other than the research described and agreed to as part of the consent-giving process. Electronic systems shared between support staff may facilitate immediate data integration, but paper versions can also be used. Additionally, online platforms owned by third parties must be scrutinised to establish who owns inputted data and how they can be used. This information must be communicated to athletes to ensure they can provide informed consent. Data security of such platforms must conform to data protection laws in the specific jurisdiction(s) within which it is being used.
The STROBE-SIIS checklist should guide study design and report observations following IOC recommendations.1 However, it is worth noting that domains of female health do not feature in the STROBE-SIIS checklist. We recommend the following specifications when applying the STROBE-SIIS checklist:
Include specific domain of female health as part of the study design (SIIS-4.1).
Incorporate justification for the inclusion of the specific domains and health problems when defining outcomes of interest (SIIS-7.1).
Outline criteria used to categorise domains of female health and their associated health problems (SIIS-7.2).
Document the number of female athletes reporting a specific female health problem, but this should be balanced with ethical considerations such as maintaining athlete anonymity (SIIS-13.1).
Report information related to specific domains of female health separately from the illness data and not combined unless appropriate (SIIS-16.1/16.2/17.1). For example, pregnancy is not an illness, but a pulmonary embolus associated with pregnancy is. The 10 proposed domains of female health and their abbreviations are shown in table 1.
Implementation considerations and recommendations
As an overarching principle, stakeholders (eg, researchers, team physicians, coaches) should plan, implement and evaluate both injury and illness surveillance and communication of results. Diligence should be taken to appreciate the ethnic and cultural backgrounds of individuals and cohorts of female athletes and stakeholders, and awareness of sensitivity concerning female-specific issues is imperative.20 21 As an example, requiring consent from athletes for recording each issue, rather than blanket consent which is typically acquired during injury and illness surveillance to record all issues. Engaging stakeholders to play an active role rather than just supplying an opinion is recommended. This is known as co-design, which may enhance the value and success of the project.22 For example, stakeholders could be involved in devising the injury definition23 and the feedback mechanisms of the surveillance system, and could help to pilot its use. Stakeholders also need to consider the intersection of sex, ethnicity, culture and economics.24 One example is understanding athlete access to devices for use on online platforms and having other options available to ensure equity in provision.
In the current sports landscape, modern technologies allow for a breadth of athlete data to be collected and used routinely. Not all data are always pertinent, nor collected, stored and used appropriately. As such, we recommend that ruling standards for collecting and storing data from athletes are followed25 and data protection laws are adhered to. These recommendations can be regional and are likely to update regularly. For female athletes, given the potentially sensitive nature of health data, clear definitions of all team members’ roles and responsibilities and sufficient gender and sociocultural awareness training should occur before implementation. For example, the data collector should be sensitive to the female athlete’s life-stage and gender identity, particularly when surveillance involves gynaecological questions.26 Adapting principles from women’s health, it is important to provide a safe environment for data collection that respects privacy and dignity when discussing sexual and reproductive health issues.27 In addition, the data collector should have trauma-informed training and be knowledgeable of safeguarding policies and procedures, including reporting mechanisms for allegations of harassment and abuse.28 Finally, the implementation team should have the clinical competency to communicate with female athletes concerning their sport-related experiences during puberty, fertility,29 pregnancy,29 30 postpartum31 32 and menopause. The IOC consensus statement lists more general implementation recommendations.1
This paper extends the IOC consensus statement, to include 10 domains of female health that may affect female athletes across the lifespan, from adolescence through young adulthood, to mid-age and older age. Our recommendations for data capture relating to female population characteristics, and to injuries, illnesses and other health consequences, will improve the quality of sports medicine and epidemiological studies on the health risks specific to female athletes. Specifically, implementing our recommendations should result in consistent and more accurate reporting and help identify potential risk factors to inform better injury and illness prevention strategies that will ultimately support female athletes’ physical and mental health.
Patient consent for publication
We thank Jillian Williams, Alice Merryweather, Alyce Wood, Hilary Stellingwerff, Ilse Hayse and Geneviève Lalonde for sharing their experiences as female athletes with the Female Health Domains.
Twitter @IzzyMoorePhD, @kaymcrossley, @margo.mountjoy, @drkateackerman, @ProfWendyBrown, @cs_bolling, @wderman, @DrPaulDijkstra, @CarolynAEmery, @DrNoeMkumbuzi, @docsoph, @janesthornton, @rtomasmd, @phatho_z, @Evertverhagen
Contributors ISM and EV conceived the idea for the consensus statement. ISM, KB, KMC, MM and EV formed the consensus groups and led each working group. CSB led the athlete interviews. All authors contributed to reviewing and giving feedback on each iteration of the consensus draft, with NSM, PCZ, JdSA and AD providing the critical review of consensus draft. All authors reviewed the final manuscript. EV is the guarantor for this manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests ISM and HPD are an Associate Editor of BJSM. JST and NSM are Editors of BJSM. KEA, PCZ and MM are Deputy Editors of BJSM. EV is the Editor in Chief of BMJ Open Sports and Exercise Medicine.
Provenance and peer review Not commissioned; externally peer reviewed.