Hormone Intelligence

Applying machine learning techniques to modelling female hormones enables women to access hormone intelligence at her fingertips

Female hormone networks form the most complex aspect of the endocrine system. The menstrual cycle depends upon a delicate web of feedback mechanisms that trigger significant changes in hormone levels. This intricate physiological process generally operates reliably, but its timing and the hormone levels are affected by internal and external factors going on in a woman’s life. This is why women differ in their experiences of menstrual cycles and why an individual woman may notice differences between cycles.

Apart from being fascinating from a physiological point of view, why is this so important from a practical point of view for women? The reason is that female hormones are not just about fertility. The ovarian hormones oestradiol (most active form of oestrogen) and progesterone have significant effects through the body. Every biological system is dependent on these hormones: bones, muscle, nervous system, including brain function, skin, the cardiovascular and digestive systems [1]. This is why female hormones impact all aspects of health: physical, mental and social [2].

The cyclical fluctuations in female hormones occurring every menstrual cycle will also change over a woman’s lifespan. Completion of puberty is marked by the start of menstrual cycles: menarche. During her adult life a woman can expect regular menstrual cycles. However, subtle hormone disruption can be missed. Although blood testing is the most accurate way of measuring all four of the key female hormones, the standard protocol of taking a blood test at one time point in the cycle, when hormones are at their most quiescent, can miss subclinical menstrual cycle hormone dysfunction.

For example, in subclinical anovulatory cycles, although a woman may experience regular menstrual periods, subtle mistiming of female hormones will not be detected with a routine single blood test. Yet this type of hormone disruption can have potential adverse consequences on health. This is particularly relevant for exercisers, athletes and dancers who are either on the brink of or recovering from low energy availability. Early identification and prevention of relative energy availability in sport (RED-S) is important for both health and exercise performance [3].

A similar situation arises for women in the perimenopause when the responsiveness of her ovaries starts to decline. This is further complicated by the fact that the decrease in ovarian hormone production is not a smooth linear process. A blood test at a single time point may not identify these changes in key female hormone networks. Although perimenopause is a natural physiological process, it can be a challenging time for women, magnified by uncertainty. All change for female hormones

Women need a new, more supportive approach, to take away uncertainty and to empower them with insights into their hormone networks.

How can a woman understand the details of her female hormone network? In theory she could take daily blood tests for the four key hormones: pituitary control hormones follicle stimulating hormone (FSH), luteinising hormone (LH) and ovarian response hormones oestradiol and progesterone. Clearly this is not practical, but it may be possible to use fewer blood tests over a cycle. Machine learning, specifically Bayesian inference, can help by optimally combining test results with background information. This background knowledge includes medical understanding of hormone networks and the characteristics of the individual woman. Machine learning can revolutionise healthcare, as outlined in the report from the Chief Medical Officer of England [4]. It is an approach widely used in modelling biological systems [5]. Artificial intelligence is an important clinical tool to support the optimisation of personalised health.

It has recently become possible to create a personalised digital fingerprint of a woman’s menstrual cycle hormone network from just two finger prick capillary blood samples taken during a cycle. Machine learning combines deep medical and mathematical understanding of female hormone networks with the individual details of a woman’s menstrual cycle length, age and activity levels. An expert report, providing an explanation of results with actionable, evidence-based advice, can be supplemented with a personal clinical medical discussion. This gives women the long-needed opportunity to connect with their personal female hormone networks. It empowers each woman to adopt a personalised, effective and proactive approach to optimise her hormone health.

To learn more about machine learning applied to female hormone networks, have at look at previous discussions and forthcoming events where I am presenting on this topic and application of this approach for female health.Presentations

To take advantage of this exciting opportunity of fingerprinting hormone networks, you can order Female Hormone Mapping™ which combines medical, mathematical and technology expertise via the Forth website (available on a discounted introductory offer for a limited time). Your personal report is delivered at your fingertips on a mobile app. You can book an appointment for a personal discussion of your report with me Female Hormones

Every woman’s hormone network fluctuations are personal to her. Every woman is an individual.


Article St John’s College, Cambridge University

[1] Keay, N. What’s so good about Menstrual Cycles? British Journal of Sport and Exercise Medicine 2019

[2] Keay, N. Of Mice and Men (and Women) British Journal of Sport and Exercise Medicine 2019

[3] Keay, N. Relative energy deficiency in sport (RED-S) British Journal of Sport and Exercise Medicine 2018 and British Association of Sport and Exercise educational website Health4Performance

[4] “Machine learning for individualised medicine” Mihaela van der Schaar, Chapter 10 of the 2018 Annual Report of the Chief Medical Officer. Health 2040 – Better Health Within Reach. Accessed 2021

[5] Van de Schoot, R., Depaoli, S., King, R. et al. Bayesian statistics and modelling. Nat Rev Methods Primers 1, 1 (2021). https://doi.org/10.1038/s43586-020-00001-2

Energy Availability: Concept, Control and Consequences in relative energy deficiency in sport (RED-S)

Relative energy deficiency in sport (RED-S) is an issue of increasing concern in sports and exercise medicine. RED-S impact exercisers of all levels and ages, particularly where low body weight confers a performance or aesthetic advantage. Key to mitigating adverse health and performance consequences of RED-S is supporting athletes and dancers to change behaviours. These infographics aim to assist clinicians in communicating the concepts to exercisers and in implementing effective management of athletes in their care[1].


Figure 1 illustrates the concept of energy availability (EA) in RED-S. Preferentially energy derived from dietary intake covers the demands of training and the remaining energy, EA, is, quantified in Kcal/Kg of fat free mass[2]. In Figure 1, the central bar illustrates adequate EA in an athlete where energy intake is sufficient to cover the demands of training and fundamental life processes to maintain health. Conversely, low energy availability (LEA) is a situation of insufficient EA to cover basic physiological demands. LEA leads to the adverse consequences of RED-S[3]. LEA can arise unintentionally or intentionally, due to a mismatch between energy intake and energy requirement. In Figure 1 the bar on the left shows LEA resulting from reduced energy intake with maintained training load. On the right, LEA is a consequence of increased training load with maintained energy intake.

Figure 2 illustrates that EA is under the control of an athlete[4]. The three behaviours relating to training, nutrition and recovery determine EA. Integrated periodisation of these behaviours results in optimal health and performance. Conversely, an imbalance in these behaviours results in suboptimal functionally. LEA in the case of high training loads relative to nutritional intake. Thus, this figure reinforces the important point in the IOC statements on RED-S that psychological factors which determine these behaviours are key in both the development, continuation and management of RED-S[2,3].


Figure 2 also shows the temporal, synergistic effect of these behaviours to ensure a fully functioning endocrine system. Hormones are key for health and to drive positive adaptations to exercise, to improve athletic performance. Thus hormones can be informative in tracking the response of an individual to these three input variables. Furthermore, endocrine markers relate to the RED-S clinical outcome of stress fracture in athletes, being more reliable as objective, quantifiable indicators of EA than numerical calculation of EA from direct assessment[5].


Nicola Keay1, Gavin Francis2

1 Department of Sport and Exercise Sciences, Durham University

2 Science4Perforamnce, London

Br J Sports Med 2019;0:1–2. doi:10.1136/bjsports-2019-100611


1 http://health4performance.co.uk (accessed 21/01/2019) Health4Performance Educational BASEM website raising awareness of RED-S Working group on RED-S British Association of Sport and Exercise Medicine 2018

2 Mountjoy M, Sundgot-Borgen J, Burke L et al. IOC consensus statement on relative energy deficiency in sport (RED-S): 2018 update Br J Sports Med2018;52(11):687-697

3 Mountjoy M, Sundgot-Borgen J, Burke L et al. The IOC consensus statement: beyond the Female Athlete Triad–Relative Energy Deficiency in Sport (RED-S). Br J Sports Med2014;48(7):491-7

4 Burke L, Lundy B, Fahrenholtz L et al, & Melin. Pitfalls of conducting and interpreting estimates of energy availability in free-living athletes. International Journal of Sport Nutrition and Exercise Metabolism2018; 28(4):350–363. https://doi.org/10.1123/ijsnem.2018-0142

5 2Heikura I, Uusitalo A, Stellingwerff T et al. Low energy availability is difficult to assess but outcomes have large impact on bone injury rates in elite distance athletes. International Journal of Sport Nutrition and Exercise Metabolism2018; 28(4):403–411. https://doi.org/10.1123/ijsnem.2017-0313

Low Energy Availability in Climbers

Listen into a great discussion I had with Dr Nigel Callender an ex competitive climber and climbing coach about the “elephant in the room” in competitive climbing.

Discussion of Low Energy Availability and RED-S

As a gravitational sport, being a light-weight climber confers a performance advantage. However, being alert to low energy availability and the clinical consequences of RED-S on health and performance is important for climbers. With climbing being included the next Olympics, then hopefully this will raise awareness of being alert to athletes at risk of low energy availability and RED-S.


Insights from Dr Nigel Callender, sports scientist turned medical doctor (anaesthetics/critical care trainee) an active researcher, largely into the exercise physiology aspects of climbing and ex-competitor, having represented Ireland at international level and been British bouldering champion before shoulder injuries ended that. Sport climbing is included in the 2020 Tokyo summer games in its three competitive disciplines; bouldering, lead climbing and speed climbing. Each sub-discipline has a slightly different athlete profile and physiological demands, but all are obviously under the heading of gravity dependent sports. Current participation figures put yearly indoor climbing participation at around the one million mark in the UK and it is said to be one of the fastest growing sports worldwide. The sport is being recognised as a great way to improve overall health and fitness, with recent papers citing it as a useful rehab activity for many physical and mental health conditions and also as a health promotion tool.

Although climbing has been a formal competitive sport in some sense since the late 80’s, it still lacks much in the way of formal training and medical guidelines. Being a gravity dependent sport, strength to weight ratio is important, however Dr Callender and his colleagues are seeing a high incidence of restrictive eating patterns at all levels of the sport and a lack of awareness around the performance impairments and health risks associated with a significant or prolonged negative energy balance in some athletes.

The Outdoor Athlete Podcast is a bit of a winter project that came about to establish a gold-standard resource, driven by credible experts in their relevant fields, as an attempt to provide high-quality and evidence-based information amongst the confusing advice that is now the internet. It’s free and always will be and it was inspired by the BJSM Podcasts though broadly aiming at ‘Outdoor Athletes’ e.g. Climbers, Fell/Trail runners, Mountain bikers and anyone happy to listen.

For more information on climbing in the UK, including competition climbing see http://www.thebmc.co.uk

Raising Awareness of RED-S in Male and Female Athletes and Dancers

Health4Performance is a recently developed BASEM open access educational resource

This is a world premier: a resource developed for and by athletes/dancers, coaches/teachers, parents/friends and healthcare professionals to raise awareness of Relative Energy Deficiency in Sport (RED-S)


Optimal health is required to attain full athletic potential. Low energy availability (LEA) can compromise health and therefore impair athletic performance as described in the RED-S clinical model.

Dietary energy intake needs to be sufficient to cover the energy demands of both exercise training and fundamental physiological function required to maintain health. Once the energy demands for training have been covered, the energy left for baseline “housekeeping” physiological function is referred to as energy availability (EA). EA is expressed relative to fat free mass (FFM) in KCal/Kg FFM.  The exact value of EA to maintain health will vary between genders and individuals, roughly equivalent to resting metabolic rate of the individual athlete/dancer. LEA for an athlete or dancer will result in the body going into “energy saving mode” which has knock on effects for many interrelated body systems, including readjustment to lower the resting metabolic rate in the longer term. So although loss in body weight may be an initial sign, body weight can be steady in chronic LEA due to physiological energy conservation adaptations. Homeostasis through internal biological feedback loops in action.

The most obvious clinical sign of this state of LEA in women is cessation of menstruation (amenorrhea). LEA as a cause of amenorrhoea is an example of functional hypothalamic amenorrhoea (FHA). In other words, amenorrhoea arising as a result of an imbalance in training load and nutrition, rather than an underlying medical condition per se, which should be excluded before arriving at a diagnosis of FHA. All women of reproductive age, however much exercise is being undertaken, should have regular menstrual cycles, which is indicative of healthy hormones. This explains why LEA was first described as the underlying aetiology of the female athlete triad, as women in LEA display an obvious clinical sign of menstrual disruption. The female athlete triad is a clinical spectrum describing varying degrees of menstrual dysfunction, disordered nutrition and bone mineral density. However it became apparent that the clinical outcomes of LEA are not limited to females, nor female reproductive function and bone health in female exercisers. Hence the evolution of the clinical model of RED-S to describe the consequences of LEA on a broader range of body systems and including male athletes.

A situation of LEA in athletes and dancers can arise unintentionally or intentionally. In the diagram below the central column shows that an athlete where energy intake is sufficient to cover the demands from training and to cover basic physiological function. However in the column on the left, although training load has remained constant, nutritional intake has been reduced. This reduction of energy intake could be an intentional strategy to reduce body weight or change body composition in weight sensitive sports and dance.  On the other hand in the column on the right, training load and hence energy demand to cover this has increased, but has not been matched by an increase in dietary intake. In both these situations, whether unintentional or intentional, the net results is LEA, insufficient to maintain health. This situation of LEA will also ultimately impact on athletic performance as optimal health is necessary to realise full athletic potential.


Although LEA is the underlying aetiology of RED-S, there are many methodological and financial issues measuring LEA accurately in “free living athletes“. In any case, the physiological response varies between individuals and depends on the magnitude, duration and timing of LEA. Therefore it is more informative to measure the functional responses of an individual to LEA, rather than the value calculated for EA. As such, Endocrine markers provide objective and quantifiable measures of physiological responses to EA. These markers also reflect the temporal dimension of LEA; whether acute or chronic. In short, as hormones exert network effects, Endocrine markers reflect the response of multiple systems in an individual to LEA. So by measuring these key markers, alongside taking a sport specific medical history, provides the information to build a detailed picture of EA for the individual, with dimensions of time and magnitude of LEA. This information empowers the athlete/dancer to modify the 3 key factors under their control of training load, nutrition and recovery to optimise their health and athletic performance.



Who is at risk of developing RED-S? Any athlete involved in sports or dance where being light weight confers a performance or aesthetic advantage. This is not restricted to elite athletes and dancers. Indeed the aspiring amateur or exerciser could be more at risk, without the benefit of a support team present at professional level. Young athletes are at particular risk during an already high energy demand state of growth and development. Therefore early identification of athletes and dancers at risk of LEA is key to prevention of development of the health and performance consequences outlined in the RED-S clinical model. Although there is a questionnaire available for screening for female athletes at risk of LEA, more research is emerging for effective and practical methods which are sport specific and include male athletes.


Early medical input is important as RED-S is diagnosis of exclusion. In other words medical conditions per se need to be ruled out before arriving at a diagnosis of RED-S.  Prompt medical review is often dependent on other healthcare professionals, fellow athletes/dancers, coaches/teachers and parents/friends all being aware and therefore alert to RED-S. With this in mind, the Health4Performance website has areas for all of those potentially involved,  with tailored comments on What to look out for? What to do? Ultimately a team approach and collaboration between all these groups is important. Not only in identification of those at risk of LEA, but in an integrated support network for the athlete/dancer to return to optimal health and performance.


Heath4Performance BASEM Educational Resource

Video introduction to Health4Performance website

2018 UPDATE: Relative Energy Deficiency in Sport (RED-S) BJSM 2018

What is Dance Medicine? BJSM 2018

Identification and management of RED-S Podcast 2018

Low energy availability assessed by a sport-specific questionnaire and clinical interview indicative of bone health, endocrine profile and cycling performance in competitive male cyclists Keay, Francis, Hind. BJM Open Sport and Exercise Medicine 2018

How to Identify Male Cyclists at Risk of RED-S? 2018

Pitfalls of Conducting and Interpreting Estimates of Energy Availability in Free-Living Athletes IJSNM 2018

Low Energy Availability Is Difficult to Assess but Outcomes Have Large Impact on Bone Injury Rates in Elite Distance Athletes IJSNM 2017

The LEAF questionnaire: a screening tool for the identification of female athletes at risk for the female athlete triad BJSM 2013

IOC consensus statement on relative energy deficiency in sport (RED-S): 2018 update BJSM 2018


How to Identify Male Cyclists at Risk of RED-S?

Relative energy deficiency in sport (RED-S) is a clinical model that describes the potential adverse health and performance consequences of low energy availability (LEA) in male and female athletes. Identification of athletes at risk of LEA can potentially prevent these adverse clinical outcomes.

Athletes at risk of RED-S are those involved in sports where low body weight confers a performance or aesthetic advantage. In the case of competitive road cycling, being light  weight results in favourable power to weight ratio to overcome gravity when cycling uphill. How can male cyclists at risk of LEA be effectively identified in a practical manner?

Energy availability (EA) is defined as the residual energy available from dietary intake, once energy expenditure from exercise training has been subtracted. This available energy is expressed as KCal/Kg fat free mass (FFM). A value of 45 KCal/Kg FFM is roughly equivalent to basal metabolic rate, in other words the energy required to sustain health. In order to quantify EA, accurate measurements of energy intake and expenditure, and FFM assessed from dual X ray absorptiometry (DXA), need to be undertaken. However this is not practical or feasible to undertake all these measurements outside the research setting. Furthermore, methodology for assessing energy intake and expenditure is laborious and fraught with inaccuracies and subjectivity in the case of diet diaries for “free living athletes“. Even if a value is calculated for EA, this is only valid for the time of measurement and does not give any insights into the temporal aspect of EA. Furthermore, an absolute EA threshold has not been established, below which clinical symptoms or performance effects of RED-S occur.

Self reported questionnaires have been shown to be surrogates of low EA in female athletes. However there are no such sport specific questionnaires, or any questionnaires for male athletes. Endocrine and metabolic markers have been proposed as quantitative surrogate measures of EA and shown to be linked to the RED-S clinical outcome of stress fractures in runners. In female athletes the clinical sign of regular menstruation demonstrates a functioning H-P ovarian axis, not suppressed by LEA. What about male athletes? Although hypothalamic suppression of the reproductive axis due to LEA can result in low testosterone, high training loads, in presence of adequate EA, can lead to the same negative effect on testosterone concentration.


Male cyclists present a further level of complexity in assessing EA status. In contrast to runners, stress fracture will not be an early clinical warning sign of impaired bone health resulting from low EA. Furthermore cyclists are already at risk of poor bone health due to the non weight bearing nature of the sport. Nevertheless, traumatic fracture from bike falls is the main type of injury in cycling, with vertebral fracture requiring the longest time off the bike. Chris Boardman, a serial Olympic medal winner in cycling, retired in his early 30s with osteoporosis. In other words, in road cycling, the combined effect of the lack of osteogenic stimulus and LEA can produce clinically significant adverse effects on bone health.

What practical clinical tools are most effective at identifying competitive male cyclists at risk of the health and performance consequences of LEA outlined in the RED-S model? This was the question our recent study addressed. The lumbar spine is a skeletal site known to be most impacted by nutrition and endocrine factors and DXA is recognised as the “gold standard” of quantifying age matched Z score for bone mineral density (BMD) in the risk stratification of RED-S. What is the clinical measure indicative of this established and clinically significant sign of RED-S on lumbar spine BMD? Would it be testosterone concentration, as suggested in the study of runners? Another blood marker? Cycle training load? Off bike exercise, as suggested in some previous studies? Clinical assessment by interview?

Using a decision tree approach, the factor most indicative of impaired age matched (Z score) lumbar spine BMD was sport specific clinical assessment of EA. This assessment took the form of a newly developed sports specific energy availability questionnaire and interview (SEAQ-I). Reinforcing the concept that the most important skill in clinical medical practice is taking a detailed history. Questionnaire alone can lead to athletes giving “correct” answers on nutrition and training load. Clinical interview gave details on the temporal aspects of EA in the context of cycle training schedule: whether riders where experiencing acute intermittent LEA, as with multiple weekly fasted rides, or chronic sustained LEA with prolonged periods of suppressed body weight. Additionally the SEAQ-I provided insights on attitudes to training and nutrition practices.

Cyclists identified as having LEA from SEAQ-I, had significantly lower lumbar spine BMD than those riders assessed as having adequate EA. Furthermore, the lowest lumbar spine BMD was found amongst LEA cyclists who had not practised any load bearing sport prior to focusing on cycling. This finding is of particular concern, as if cycling from adolescence is not integrated with weight bearing exercise and adequate nutrition when peak bone mass (PBM) is being accumulated, then this risks impaired bone health moving into adulthood.

Further extension of the decision tree analysis demonstrated that in those cyclists with adequate EA assessed from SEAQ-I, vitamin D concentration was the factor indicative of lumbar spine BMD. Vitamin D is emerging as an important consideration for athletes, for bone health, muscle strength and immune function. Furthermore synergistic interactions with other steroid hormones, such as testosterone could be significant.

What about the effects of EA on cycling performance? For athletes, athletic performance is the top priority. In competitive road cycling the “gold standard” performance measure is functional threshold power (FTP) Watts/Kg, produced over 60 minutes. In the current study, 60 minute FTP Watts/Kg had a significant relationship to training load. However cyclists in chronic LEA were under performing, in other words not able to produce the power anticipated for a given training load. These chronic LEA cyclists also had significantly lower testosterone concentration. Periodised carbohydrate intake for low intensity sessions is a strategy for increasing training stimulus. However if this acute intermittent LEA is superimposed on a background of chronic LEA, then this can be counter productive in producing beneficial training adaptations. Increasing training load improves performance, but this training is only effective if fuelling is tailored accordingly.

Male athletes can be at risk of developing the health and performance consequences of LEA as described in the RED-S clinical model. The recent study of competitive male road cyclists shows that a sport specific questionnaire, combined with clinical interview (SEAQ-I) is an effective and practical method of identifying athletes at risk of LEA. The temporal dimension of LEA was correlated to quantifiable health and performance consequences of RED-S.


Low energy availability assessed by a sport-specific questionnaire and clinical interview indicative of bone health, endocrine profile and cycling performance in competitive male cyclists  Keay, Francis, Hind, BMJ Open in Sport and Exercise Medicine 2018

2018 UPDATE: Relative Energy Deficiency in Sport (RED-S) Keay, BJSM 2018

Fuelling for Cycling Performance Science4Performance

Pitfalls of Conducting and Interpreting Estimates of Energy Availability in Free-Living Athletes International Journal of Sport Nutrition and Exercise Metabolism 2018

IOC consensus statement on relative energy deficiency in sport (RED-S): 2018 update BJSM 2018

The LEAF questionnaire: a screening tool for the identification of female athletes at risk for the female athlete triad BJSM 2013

Low Energy Availability Is Difficult to Assess but Outcomes Have Large Impact on Bone Injury Rates in Elite Distance Athletes International Journal of Sport Nutrition and Exercise Metabolism 2018

Treating exercise-associated low testosterone and its related symptoms The Physician and Sports Medicine 2018

Male Cyclists: bones, body composition, nutrition, performance Keay, BJSM 2018

Cyclists: Make No Bones About It Keay, BJSM 2018

Male Athletes: the Bare Bones of Cyclists

Cyclists: How to Support Bone Health?

Synergistic interactions of steroid hormones Keay BJSM 2018

Fuel for the Work Required: A Theoretical Framework for Carbohydrate Periodization and the Glycogen Threshold Hypothesis Sports Medicine 2018


Health, Hormones and Human Performance Part 1

How hormones determine health and athletic performance

Endocrine and Metabolic aspects of Sports and Exercise Medicine are crucial determinants of health and human performance, from reluctant exerciser through to elite athlete and professional dancer. This is what I set out to demonstrate as the chair of the recent British Association of Sport and Medicine conference, with insightful presentations from my colleagues whom I had invited to share their research and practical applications of their work. The audience comprised of doctors with interest in sport and exercise medicine, representatives from the dance world, research scientists, nutritionists, physiotherapists, coaches and trainers. In short, all were members of multi-disciplinary teams supporting aspiring athletes. The importance of the conference was reflected in CDP awards from FSEM, BASES, Royal College of Physicians (RCP), REP-S and endorsement for international education from BJSM and National Institute of Dance Medicine and Science (NIDMS).

Exercise is a crucial lifestyle factor in determining health and disease. Yet we see an increasing polarisation in the amount of exercise taken across the general population. At one end of the spectrum, the increasing training loads of elite athletes and professional dancers push the levels of human performance to greater heights. On the other side of the spectrum, rising levels of inactivity, in large swathes of the population, increase the risk of poor health and developing disease states. Which fundamental biological processes and systems link these groups with apparently dichotomous levels of exercise? What determines the outcome of the underlying Endocrine and metabolic network interactions? How can an understanding of these factors help prevent sports injuries and lead to more effective rehabilitation? How can we employ Endocrine markers to predict and provide guidance towards beneficial outcomes for health and human performance?

If you weren’t able to come and participate in the discussion, these are some topics presented. My opening presentation (see video below) set the scene, outlining why having an optimally functioning Endocrine system is fundamental to health and performance. Conversely, functional disruption of Endocrine networks occurs with non integrated periodisation of the three key lifestyle factors of exercise/training, nutrition and recovery/sleep, which can lead to adverse effects on health and athletic performance.

In the case of an imbalance in training load and nutrition, this can manifest as the female athlete triad, which has now evolved into relative energy deficiency in sports (RED-S) in recognition of the fact that Endocrine feedback loops are disrupted across many hormonal axes, not just the reproductive axis. And, significantly, acknowledging the fact that males athletes can also be impacted by insufficient energy availability to meet both training and “housekeeping” energy requirements. Why and how RED-S can affect male athletes, in particular male competitive road cyclists, was discussed, highlighting the need for further research to investigate practical and effective strategies to optimise health and therefore ultimately performance in competition.

A degree of overlap and interplay exists between RED-S (imbalance in nutrition and training load), non functional over-reaching and over-training syndrome (imbalances in training load and recovery). Indeed research evidence was presented suggesting that RED-S increases the risk of developing over-training syndrome. In these situations of functional disruption of the Endocrine networks, underlying Endocrine conditions per se should be excluded. Case studies demonstrated this principle in the diagnosis of RED-S. This is particularly important in the investigation of amenorrhoea. All women of reproductive age, whether athletes or not, should have regular menstruation (apart from when pregnant!), as a barometer of healthy hormones. Indeed, since hormones are essential to drive positive adaptations to exercise, healthy hormones are key in attaining full athletic potential in any athlete/dancer, whether male or female. Evidence was presented from research studies for the role of validated Endocrine markers and clinical menstrual status in females as objective and quantifiable measures of energy availability and hence injury risk in both male and female athletes.

Triumvirate of external factors impacting Endocrine system and hence performance

Alongside training metrics, if female athletes recorded menstrual pattern (as Gwen Jorgensen recently showed on her Training Peaks) and all athletes kept a biological passport of selected Endocrine markers; this could potentially identify at an early stage any imbalances in the triumvirate of training load, nutrition and recovery. Pre-empting development of RED-S or over-training syndrome, supports the maintenance of healthy hormones and hence optimal human performance.

Look out for presentations from speakers which will be uploaded on BASEM website shortly.


Video of presentation on the Endocrine and Metabolic Aspects of Sports and Exercise Medicine BASEM conference “Health, Hormones and Human Performance”

Study of hormones, body composition, bone mineral density and performance in competitive male road cyclists Investigation of effective and practical nutrition and off bike exercise interventions

Sports Endocrinology – what does it have to do with performance? Keay BJSM 2017




Temporal considerations in Endocrine/Metabolic interactions Part 1


It is not a simple question of what, but when we eat, sleep and exercise.

The Endocrine system displays temporal variation in release of hormones. Integrating external lifestyle factors with this internal, intrinsic temporal dimension is crucial for supporting metabolic and Endocrine health.

Amplitude and frequency of hormonal secretion display a variety of temporal patterns:

  • Diurnal variation, synchronised with external light/dark. Orchestrated by a specific area of the hypothalamus, the neuroendocrine gatekeeper.
  • Circadian rhythm, roughly 24-25 hours which can vary with season according to duration of release of melatonin from the pineal gland.
  • Infradian rhythms longer than a day, for example lunar month seen in patterns of hypothalamic-pituitary-ovarian axis hormone release during the menstrual cycle.
  • Further changes in these temporal release and feedback patterns occur over a longer timescale during the lifespan.

Hormones influence gene expression and hence protein synthesis over varying timescales outlined above. The control system for hormone release is based on interactive feedback loops. The hypothalamus is the neuroendocrine gatekeeper, which integrates external inputs and internal feedback.  The net result is to maintain intrinsic biological clocks, whilst orchestrating adaptations to internal perturbations stimulated by external factors such as sleep pattern, nutrition and exercise.

Circadian alignment refers to consistent temporal patterns of sleep, nutrition and physical activity. Circadian misalignment affects sleep-architecture and subsequently disturbs the interaction of metabolic and Endocrine health. This includes gut-peptides, glucose-insulin interaction, substrate oxidation, leptin & ghrelin concentrations and hypothalamic-pituitary-adrenal/gonadal-axes. The main stimuli for growth hormone release are sleep and exercise. Growth hormone is essential for supporting favourable body composition. These integrated patterns of environmental factors may have a more pronounced effect on those with a genetic predisposition or during crucial stages of lifespan. For example curtailed sleep during puberty can impact epigenetic factors such as telomere length and thus may predispose to metabolic disruption in later life. Regarding activity levels, there are strong relationships between time spent looking at screens and markers, such as insulin resistance, for risk of developing type 2 diabetes mellitus in children aged 9 to 10 years.

In addition to adverse metabolic effects set in motion by circadian misalignment, bone turnover has also shown to be impacted. Circadian disruption in young men resulted in uncoupling of bone turnover, with decreased formation and unchanged bone resorption as shown by monitoring bone markers. In other words a net negative effect on bone health, which was most pronounced in younger adult males compared with their older counterparts. These examples underline the importance of taking into account changes in endogenous temporal patterns during the lifespan and hence differing responses to external lifestyle changes.

For male and female athletes, integrated periodised training, nutrition and recovery has to be carefully planned over training seasons to support optimal adaptations in Endocrine and metabolic networks to improve performance. Training plans that do not balance these all these elements can result in underperformance, potentially relative energy deficiency in sport and consequences for health in both short and long term.

Part 2 will consider the longer term consequences and interactions of these temporal patterns of lifestyle factors, including seasonal training patterns in male and female athletes, on the intrinsic biochronometry controlling the Endocrine and metabolic networks during lifespan.

For further discussion on Health, Hormones and Human Performance, come to the BASEM annual conference


Sports Endocrinology – what does it have to do with performance? Dr N.Keay, British Journal of Sports Medicine 2017

One road to Rome: Metabolic Syndrome, Athletes, Exercise Dr N. Keay

Metabolic and Endocrine System Networks Dr N. Keay

Endocrine system: balance and interplay in response to exercise training Dr N.Keay

Sleep for health and sports performance Dr N.Keay, British Journal of Sports Medicine 2017

Factors Impacting Bone Development Dr N. Keay

Sleep, circadian rhythm and body weight: parallel developments Proc Nutr Soc

Sleep Duration and Telomere Length in Children Journal of Paediatrics 2017

Screen time is associated with adiposity and insulin resistance in children Archives of Disease in Childhood

Circadian disruption may lead to bone loss in healthy men Endocrine today 2017

Successful Ageing Dr N. Keay, British Association of Sport and Exercise Medicine 2017

Clusters of Athletes – A follow on from RED-S blog series to put forward impact of RED-S on athlete underperformance Dr N. Keay, British Association of Sport and Exercise Medicine 2017

Optimal Health: For All Athletes! Part 4 – Mechanisms Dr N. Keay, British Association of Sport and Exercise Medicine 2017





Sport Performance and Relative Energy Deficiency in Sport

performance-potentialThe Holy Grail of any training program is to improve performance and achieve goals.

Periodisation of training is essential in order to maximise beneficial adaptations for improved performance. Physiological adaptations occur after exercise during the rest period, with repeated exercise/rest cycles leading to “super adaptation”. Adaptations occur at the system level, for example cardiovascular system, and at the cellular level in mitochondria. An increase in mitochondria biogenesis in skeletal muscle occurs in response to exercise training, as described by Dr Andrew Philip at a recent conference at the Royal Society of Medicine (RSM). This cellular level adaptation translates to improved performance with a right shift of the lactate tolerance curve.

The degree of this response is probably genetically determined, though further research would be required to establish causal links, bearing in mind the ethical considerations laid out in the recent position statement from the Australian Institute of Sport (AIS) on genetic testing in sport. Dr David Hughes, Chief Medical Officer of the AIS, explored this ethical stance at a fascinating seminar in London. Genetic testing in sport may be a potentially useful tool for supporting athletes, for example to predict risk of tendon injury or response to exercise and therefore guide training. However, genetic testing should not be used to exclude or include athletes in talent programmes. Although there are polymorphisms associated with currently successful endurance and power athletes, these do not have predictive power. There are many other aspects associated with becoming a successful athlete such as psychology. There is no place for gene doping to improve performance as this is both unethical and unsafe.

To facilitate adaptation, exercise should be combined with periodised rest and nutrition appropriate for the type of sport, as described by Dr Kevin Currell at the conference on “Innovations in sport and exercise nutrition”. Marginal gains have a cumulative effect. However, as discussed by Professor Asker Jeukendrup, performance is more than physiology. Any recommendations to improve performance should be given in context of the situation and the individual. In my opinion women are often underrepresented in studies on athletes and therefore further research is needed in order to be in a position to recommend personalised plans that take into account both gender and individual variability. As suggested by Dr Courtney Kipps at the Sport and Exercise Conference (SEM) in London, generic recommendations to amateur athletes, whether male or female, taking part in marathons could contribute to women being at risk of developing exercise associated hyponatraemia.

For innovation in sport to occur, complex problems approached with an open mind are more likely to facilitate improvement as described by Dr Scott Drawer at the RSM. Nevertheless, there tends to be a diffusion from the innovators and early adapters through to the laggards.

Along the path to attaining the Holy Grail of improved performance there are potential stumbling blocks. For example, overreaching in the short term and overtraining in the longer term can result in underperformance. The underlying issue is a mismatch between periodisation of training and recovery resulting in maladapataion. This situation is magnified in the case of athletes with relative energy deficiency in sport (RED-S). Due to a mismatch of energy intake and expenditure, any attempt at increase in training load will not produce the expected adaptations and improvement in performance. Nutritional supplements will not fix the underlying problem. Nor will treatments for recurrent injuries. As described by Dr Roger Wolman at the London SEM conference, short term bisphosphonante treatment can improve healing in selected athletes with stress fractures or bone marrow lesions.  However if the underlying cause of drop in performance or recurrent injury is RED-S, then tackling the fundamental cause is the only long term solution for both health and sport performance.

Network effects of interactions lead to sport underperformance. Amongst underperforming athletes there will be clusters of athletes displaying certain behaviours and symptoms, which will be discussed in more detail in my next blog. In the case of RED-S as the underlying cause for underperformance, the most effective way to address this multi-system issue is to raise awareness to the potential risk factors in order to support athletes in attaining their full potential.

For further discussion on Endocrine and Metabolic aspects of SEM come to the BASEM annual conference 22/3/18: Health, Hormones and Human Performance


Teaching module RED-S British Association Sport and Exercise Medicine

From population based norms to personalised medicine: Health, Fitness, Sports Performance Dr N. Keay, British Journal of Sport Medicine 22/2/17

Balance of recovery and adaptation for sports performance Dr N. Keay, British Association Sport and Exercise Medicine 21/1/17

Sleep for health and sports performance Dr N. Keay, British Journal of Sport Medicine 7/7/17

Fatigue, Sport Performance and Hormones… Dr N. Keay, British Journal of Sport Medicine

Annual Sport and Exercise Medicine Conference, London 8/3/17

Bisphosphonates in the athlete. Dr Roger Wolman, Consultant in Rheumatology and Sport and Exercise Medicine, Royal National Orthopaedic Hospital

Collapse during endurance training. Dr Courtney Kipps, Consultant in Sport and Exercise Medicine. Consultant to Institute of Sport, medical director of London and Blenheim Triathlons

Innovations in Sport and Exercise Nutrition. Royal Society of Medicine 7/3/17

Identifying the challenges: managing research and innovations programme. Dr Scott Drawer, Head of Performance, Sky Hub

Exercise and nutritional approaches to maximise mitochondrial adaptation to endurance exercise. Dr Andrew Philip, Senior Lecturer, University of Birmingham

Making technical nutrition data consumer friendly. Professor Asker Jeukendrup, Professor of Exercise Metabolism, Loughborough University

Innovation and elite athletes: what’s important to the applied sport nutritionists? Dr Kevin Currell, Director of Science and Technical Development, The English Institute of Sport

Genetic Testing and Research in Sport. Dr David Hughes, Chief Medical Officer Australian Institute of Sport. Seminar 10/3/17

Effects of adaptive responses to heat exposure on exercise performance

Over Training Syndrome, Ian Craig, Webinar Human Kinetics 8/3/17

The Fatigued Athlete BASEM Spring Conference 2014

Relative Energy Deficiency in sport (REDs) Lecture by Professor Jorum Sundgot-Borgen, IOC working group on female athlete triad and IOC working group on body composition, health and performance. BAEM Spring Conference 2015.

Mountjoy M, Sundgot-Borgen J, Burke L, Carter S, Constantini N, Lebrun C, Meyer N, Sherman R, Steffen K, Budgett R, Ljungqvist A. The IOC consensus statement: beyond the Female Athlete Triad-Relative Energy Deficiency in Sport (RED-S).Br J Sports Med. 2014 Apr;48(7):491-7.