NB: When considering information concerning groups of people with ME/CFS and those with long COVID, it is important to remember that ME/CFS is a symptom-based clinical diagnosis not a mechanistic one. It is clear there is a high degree of shared pathophysiology between ME/CFS and long COVID, and the two diagnostic labels are not mutually exclusive. Importantly, some individuals with long COVID meet ME/CFS diagnostic criteria or could have a dual diagnosis.

Research suggests that hormones—chemical messengers that help regulate processes throughout the body—may play a role in both ME/CFS and in long COVID.

However, the results of hormone studies have been inconsistent, so firm conclusions cannot yet be drawn. This may be partly because many studies collect data at only one point in time, do not analyse males and females separately, fail to account for hormonal changes across the life course such as puberty, pregnancy, and menopause, or include small yet highly varied groups of people with ME/CFS or long COVID.

Recognising this, a team of researchers led by Dr Natalie Thomas, and including Dr Beth Pollack – who has written a review paper looking at female reproductive health in long COVID ad related illnesses including ME/CFS – have published an article which summarises the evidence relating to ME/CFS, long COVID, and hormones, and recommends a framework for research studies going forward.

The central argument in the paper is that hormones should not be studied as isolated, fixed measurements. Hormone levels rise and fall over hours, days and—in people who menstruate—across the menstrual cycle. If researchers take a single blood sample without carefully recording when it was collected, they may miss a genuine abnormality or mistake normal biological variation for illness-related change.

Why hormones matter in ME/CFS and long COVID research?

Hormones help coordinate many functions affected in ME/CFS and long COVID, including energy use, sleep and wakefulness, stress responses, immune activity, blood pressure, temperature regulation, and cognitive function.

Much of this coordination is organised through the hypothalamus and pituitary gland in the brain. Together with hormone-producing organs such as the adrenal glands, thyroid and ovaries or testes, they form feedback systems – known as “axes” –  in which the brain signals to an organ, the organ releases hormones, and those hormones feed information back to the brain. One example if this is the hypothalamic-pituitary-adrenal (HPA) axis which helps regulate the body’s response to stress – notably, dysregulation of HPA axis has been implicated in ME/CFS.

In their paper, Thomas and colleagues argue that disruption anywhere in these networks could have effects across several body systems at once, and that this ‘systems view’ may be useful for illnesses – like ME/CFS and long COVID – that are marked by a changing combination of fatigue, post-exertional malaise (PEM), sleep disturbance, cognitive problems, pain, immune symptoms, and autonomic difficulties.

It is essential to note that the paper does not claim that one hormone explains either ME/CFS or long COVID. Rather, it asks whether altered timing, coordination, signalling or feedback across several hormone systems could contribute to symptom patterns and suggests that incorporating methodologies that allow researchers to capture this information may improve research going forward – more on this later in the article.

What the existing evidence suggests

In ME/CFS and in long COVID research, studies have reported differences in steroid hormones like cortisol or progesterone, peptide hormones such as vasopressin, and neuropeptides like orexin. However, findings are not always consistent due to a number of factors, including:

• Use of only small groups of participants, which may lack the power to detect significant associations.
• Diverse groups of participants analysed together – this can mask associations through ‘effect modification’ (where the association of interest, for example the link between a specific hormone and ME/CFS severity level, differs by a third factor such as sex).
• Analysis that does not consider differences in factors such as illness stage, severity of disease, medication use, and reproductive status.
• Use of samples which collect data at one point on time only (cross-sectional studies) rather than looking at repeated measures.
• Variation in laboratory methods used between studies.
• Failure to account for time of day or menstrual phase.

Additionally, the team note that although a hormone level may look “normal”, there could still be an underlying problem. For example, a hormone level may be within the ‘normal range’ but peak at the wrong time, change less than expected during the day, or fail to work properly because the body’s cells do not respond to it as they should.

Therefore, Thomas and the team suggest that researchers need to look at more than the amount of a hormone in the blood. They should also consider whether the body produces and responds to a hormone properly, whether cells can receive a hormones message, and if different hormones still follow their usual rhythms and work together.

The importance of biological timing

As noted in the paper, there are three key types of biological rhythm that hormonal changes in the body typically follow:

• Ultradian: rhythms that repeat more than once within 24 hours.
• Circadian: rhythms that follow an approximately 24-hour cycle, such as the sleep–wake cycle.
• Infradian: rhythms lasting longer than 24 hours, such as the menstrual cycle.

Symptoms of ME/CFS, and of long COVID, have been found to change over a 24-hour period across the menstrual cycle, and before, during, and after the menopause. The paper therefore recommends measuring hormones at different points during these periods where appropriate, or to take into account these phases in the analysis in other ways, for example grouping male participants into pre-puberty, middle age and ‘elderly’, and grouping female participants into pre-pubertal, menstruating (grouped into menses, follicular phase (early, mid, and late subphase), luteal phase (early, mid, and late subphase), and ovulation), peri-menopausal, post-menopausal and old age.

The researchers also recommend taking several samples at set times and recording factors such as sleep, meals, activity, symptoms, medicines and menstrual phase. This could show hormone patterns that a single sample might miss.

Sex hormones and female predominance

Sex-hormone signalling is highlighted as an especially important area because both ME/CFS and long COVID are more commonly reported in women, and because reproductive hormones interact with systems commonly implicated in ME/CFS and long COVID disease mechanisms, including:

• The immune system.
• The brain.
• Metabolism.
• Blood vessels.

Hormonal transitions and fluctuations may help explain why symptoms differ between individuals or change at particular life stages or menstrual phases.

Importantly, the authors are not saying that ME/CFS and long COVID are caused by female hormones. Rather, they argue that failing to measure reproductive timing and hormone signalling properly may conceal biologically meaningful subgroups and patterns.

A proposed research framework

• Measure hormone levels repeatedly and at planned times: collect samples across the day and, where relevant, across the menstrual cycle rather than relying on one snapshot.
• Record context: document sleep, light exposure, food intake, activity, exertion, medicines, hormonal contraception or therapy, reproductive stage, illness duration, and symptom severity.
• Test how well hormones are working as well as concentration: use stimulation or suppression tests where appropriate, and study receptors and downstream cellular responses.
• Connect hormones to symptoms: gather symptom and physiological data at the same times as biological samples so that changing hormone patterns can be compared with changing health.
• Study interacting systems: combine hormone measures with immune, autonomic, neurological, and metabolic information to enable links to be made across different systems of the body.
• Use multi-omics: integrate measurements such as genes, gene activity, proteins and metabolites to identify pathways and patient subgroups rather than searching for a single universal marker.
• Design inclusive, adequately powered studies: account for sex, age, reproductive stage, ethnicity, disease severity, and other clinically relevant subgroups, and confirm findings in independent cohorts.

Takeaway message

The review argues that the most informative question may not be simply whether a hormone is “high” or “low”, but whether multiple hormone systems are producing the right signals, at the right times, and responding appropriately. By measuring endocrine rhythms repeatedly and linking them with symptoms and other biological systems, researchers may gain a more accurate picture of ME/CFS, and of long COVID.

Importantly, considering hormones in this way may help researchers to account for biological sources of variation (heterogeneity) in those with ME/CFS, and in those with long COVID such as circadian rhythms, menstrual cycle phase reproductive stage, sex, age, and hormone exposure. In tun, this may help to create a stronger foundation for personalised treatments in the future, but more research is needed.

Importantly, in their study the researchers recognise that there is a lack of research in males with ME/CFS, and in those with long COVID stating:

“Males as a cohort remain an under characterised group and deserves focused attention”