Elite female athlete research: stop searching for the ‘magic P’

In recent years, the call for a priority need for research in female athletes has grown louder (Bruinvels et al., 2017). Practitioners, coaches and athletes are crying out for more informed support. Certainly, there are signs of positive change, with a visible increase in female athlete specific publications. Yet, whilst the gap may be closing, it remains significant in elite populations where studies in elite female athletes remain limited. Studies in truly elite female athletes are often statistically underpowered and therefore not considered publication worthy, as peer-review feedback we and others have experienced has suggested. But isn't that the point? The elite population is a limited number – these are the top 5% performers of their sport – but if academia continues to work with a mindset that elite sport requires an evidence-based practice approach, then more often than not this means meeting the magic 'P < 0.05′ or alternative statistical approval to be considered of publication value. Yet, by their very nature, elite athletes are elite precisely because they are statistical outliers. The nature of elite sport does not easily lend itself to traditional experimental design; the population is very small, and the characteristics of elite sport environments mean researchers have little room for manoeuvre when it comes to delivering interventions and requesting athlete time commitments. Yet, academia appears steadfast in the belief that change in practice still largely relies on published outcomes, which can only come in the form of a cohort study. Do the outputs from traditional cohort studies reflect the individual nature of the menstrual cycle? The short answer is no. Current approaches often fail to recognise the between-subject variation of oestradiol and progesterone and the influence this might have on outcome measures, even when participants are grouped in the same phase. Therefore, any association made between the cycle phase and health/performance outcome is based on the mean of a highly variable group and tells us very little about the influence of the hormones. Do the hormones matter? Yes, of course they do. Yet, the narrative often focuses on cycle phase rather than hormonal concentration. Rather than concluding cycle phase influences performance, should we not actually be trying to find out if hormone concentration influences performance? Whether it is an absolute hormone concentration or the magnitude of change from a relative baseline, it is the change in the hormone that will influence physiological systems. Do individuals matter? Again, yes of course they do. In elite sport, individuals not group means, matter the most. There is so much about the menstrual cycle that we do not understand in relation to sport and exercise, but one of the few things we do know is that there is nothing more individual than a menstrual cycle in both the underpinning physiology and the individual's own experiences. We must do a better job of capturing this and using it to inform future practice. We have first-hand experience of the limitations we are outlining here, so we understand the challenges of conducting female athlete research. There is collective and excellent intent amongst the research community to increase the scope and quality of female athlete specific research (Elliott-Sale et al., 2021), but perhaps it is time to reconsider our approach. Rather than assigning increasingly scarce financial and time resources to research investigations that merely show how the average concentration of a hormone effects the average of a particular physiological response, a greater effort should be made in developing the technology and know-how to adequately capture data that provide information that is specific and meaningful to individuals. Perhaps journals and journal editors can be more open minded about the value of case studies on elite athletes, and specifically for elite female athlete research. Experimental Physiology was one of the first physiology journals to recognise this and offer a case study option, but unfortunately it remains one of only a few sport and exercise science-related journals to do so. If case studies are an accepted way of sharing and informing practice in clinical environments, why can they not play the same role in the sport and exercise sciences? Is it time, therefore, to consider a culture shift, where changes to practice and improvements in performance are not reliant on the publication of cohort studies, which often require unrealistic numbers of elite athletes to provide an improbable homogeneous group simply to meet statistical significance and therefore be worthy of publication? Instead of chasing under-powered cohort studies that make large assumptions and giant leaps to application, perhaps case studies provide a valuable vehicle for the accurate transfer of knowledge to 'real world' practice, much like you would see in clinical practice where publications, such as BMJ Case Reports, are valued resources providing information on clinically important conditions. Case studies provide meaningful data, related specifically to the individual in question, which is vital in elite sport and should be a pre-requisite when studying the influence of the menstrual cycle. Broad brush generalisations no longer cut it in elite sport. Coaches, support staff and the athletes themselves cannot use group data averages to move their performances forward. We need more precise, individualised athlete support and it would be helpful if more research reflected this. None declared.