Myalgic Encephalomyelitis/Chronic Fatigue Syndrome

Myalgic encephalomyelitis (ME) and chronic fatigue syndrome (CFS) are overlapping names for a chronic, disabling, multisystem illness whose defining clinical pattern is not ordinary tiredness but impaired function, post-exertional malaise (PEM), disordered sleep, and cognitive or autonomic symptoms.^[1] ME is the older term and foregrounds neurological illness; CFS foregrounds persistent fatigue. Neither label is fully satisfactory: “encephalomyelitis” implies a specific inflammatory lesion that has not been established as a universal pathological finding, whereas “fatigue” suggests a symptom that improves with rest and obscures the characteristic worsening after exertion. The combined term ME/CFS therefore preserves both clinical traditions while signalling that fatigue alone is an inadequate shorthand.^[1][10]

PEM is the feature that most sharply defines the illness. It is a disproportionate, delayed worsening of existing symptoms or appearance of new symptoms after physical, cognitive, emotional, orthostatic, or sensory activity that was previously tolerated; recovery may take days or longer.^[3] The trigger may be modest, and the relevant exertion is not limited to exercise. A patient who reports severe tiredness but no reproducible post-exertional deterioration has not yet described the cardinal ME/CFS phenotype, even if the fatigue is chronic and disabling.^[3][12] Unrefreshing sleep means that sleep fails to restore function, not merely that the patient sleeps for too few hours. Cognitive impairment commonly presents as slowed processing, impaired concentration, or “brain fog,” while orthostatic intolerance refers to symptoms provoked or aggravated by upright posture, such as light-headedness, palpitations, presyncope, or cognitive worsening.^[1][3]

The term “fatigue” also compresses several clinically distinct experiences. ME/CFS fatigue is persistent or recurrent, substantially reduces activity, is not reliably relieved by rest, and is coupled to limited physiological recovery after exertion.^[1][10] Chronic fatigue is only a duration-based symptom and can accompany many diseases; idiopathic chronic fatigue denotes persistent fatigue for which no cause has been identified but does not, by itself, establish PEM, unrefreshing sleep, cognitive impairment, or orthostatic intolerance. CFS therefore cannot be inferred from a fatigue score alone, and a high fatigue score is neither necessary nor sufficient to identify the characteristic syndrome.^[10][11]

ME/CFS remains a clinical diagnosis. No blood test, imaging study, physiological measurement, or other biomarker has sufficient sensitivity, specificity, reproducibility, and clinical practicality to confirm or exclude it.^[10] Proposed metabolomic, immunological, electrophysiological, and neuroimaging markers are research findings rather than validated diagnostic tests.^[1][10] The diagnosis consequently rests on the history and examination, application of a stated case definition, and assessment that another condition does not better explain the presentation; the absence of a confirmatory biomarker does not make the illness a diagnosis of exclusion in the sense of being medically unreal.^[3][10]

Case definitions answer different clinical questions and identify overlapping, not identical, populations. The 1994 Fukuda definition is broad: it requires unexplained, persistent or recurrent fatigue of at least 6 months that substantially reduces activity, plus at least 4 of 8 accompanying symptoms. PEM is one of those eight symptoms rather than a mandatory feature, and neither unrefreshing sleep nor cognitive impairment is independently required. This breadth supported epidemiological work but permits cohorts with chronic fatigue and little or no PEM, thereby reducing phenotypic specificity.^[1][12]

The 2003 Canadian Consensus Criteria (CCC) require PEM, pathological fatigue with reduced activity, sleep dysfunction, pain, and neurological or cognitive manifestations, together with symptoms from at least two of three domains: autonomic, neuroendocrine, and immune manifestations. The usual duration threshold is at least 6 months in adults and 3 months in children and adolescents.^[3] The CCC consequently gives PEM a central position and captures the multisystem nature of the illness, but its greater specificity can miss patients early in the course or patients whose symptoms are incompletely documented.^[3]

The 2011 International Consensus Criteria (ICC) also require PEM and organize the illness around neurological, immune, gastrointestinal or genitourinary, and energy-production or transport abnormalities. Cognitive impairment, sleep disturbance, pain, sensory symptoms, and autonomic or orthostatic manifestations are treated as clinically meaningful components rather than incidental complaints.^[1][13] The ICC is deliberately stringent and is often useful for research phenotyping, but its complexity and narrower case ascertainment limit uniform use in general practice.^[11]

The 2015 Institute of Medicine , now the National Academy of Medicine , criteria require all three of the following: substantial reduction or impairment in the ability to engage in usual activities, accompanied by fatigue that is profound, new or of definite onset, not substantially relieved by rest, and present for at least 6 months; PEM; and unrefreshing sleep. The patient must also have at least one of cognitive impairment or orthostatic intolerance.^[1][20] This framework made PEM mandatory while allowing either of two common manifestations to complete the phenotype. Its accessibility makes it useful for case finding, although its broader structure can include a more heterogeneous population than the CCC or ICC.^[3]

The 2021 National Institute for Health and Care Excellence (NICE) guideline recommends suspecting ME/CFS when debilitating fatigue, PEM, unrefreshing sleep, and cognitive difficulty occur together and persist for at least 3 months, with symptoms not better explained by another condition.^[2] NICE therefore makes all four features central and permits diagnosis earlier than the traditional 6-month threshold. Orthostatic intolerance, pain, flu-like symptoms, and sensory sensitivity support the diagnosis but are not required core symptoms.^[2] This approach is clinically practical and recognizes that waiting 6 months can delay support, but its shorter duration threshold and symptom-based structure may increase heterogeneity if applied without careful clinical judgment.^[2][12]

Post-COVID condition is not synonymous with ME/CFS. It is defined by symptoms arising after probable or confirmed COVID-19 , usually within 3 months, lasting at least 2 months, and not explained by another diagnosis; ME/CFS is defined by its symptom phenotype and does not require SARS-CoV-2 infection.^[8] The syndromes overlap substantially: PEM, cognitive dysfunction, sleep disturbance, pain, and autonomic symptoms may occur in both, and a subset of people with post-COVID condition fulfills ME/CFS criteria.^[7][17] The correct formulation is therefore that post-COVID condition may cause an ME/CFS phenotype, coexist with ME/CFS, or remain a distinct post-infectious illness; the temporal association with COVID-19 does not substitute for applying ME/CFS criteria.^[7][8][20]

Pearl: Ask what happens after exertion, not only how tired the patient feels. PEM, together with unrefreshing sleep and cognitive or orthostatic symptoms, is what turns “fatigue” from a nonspecific complaint into a recognizable ME/CFS phenotype.^[3][10]

Prevalence estimates for ME/CFS are not interchangeable. Population surveys, administrative records, specialist cohorts, and self-selected patient surveys identify different parts of the illness spectrum; estimates also change when investigators use a symptom-based fatigue definition, a more stringent ME/CFS definition, or a clinician-recorded diagnosis. Reported prevalence ranges from 0.4-0.6% in one synthesis to 0.3-0.9% in another, whereas the 2021-2023 US National Health Interview Survey found that 1.5% of adults had ever received an ME/CFS diagnosis. ^[21][22][23] The last estimate measures recognition or past labeling rather than population disease prevalence, and the post- SARS-CoV-2 expansion of post-COVID illness makes temporal comparisons particularly unstable. ^[21][23]

Incidence is even more dependent on ascertainment. Norwegian registry studies found the greatest frequency of recorded diagnoses in adolescence and early-to-middle adulthood, but registry age at diagnosis is not age at onset and is distorted by diagnostic delay. ^[21] A large European patient survey and a DecodeME subcohort showed a reproducible bimodal distribution, modeled at approximately 16 and 37 years; these findings support clinically meaningful early- and later-onset groups but do not establish population incidence because participants were recruited through patient organizations, social media, or an existing research cohort. ^[21] Early-onset disease may carry a greater severity burden, a finding that should prompt clinicians to recognize illness in adolescents rather than treating school absence or reduced participation as merely behavioral. ^[21]

Women are diagnosed approximately four times as often as men. ^[21] This sex ratio may reflect biological susceptibility, differences in exposure and health-care use, diagnostic expectations, or some combination; it should not be used to exclude ME/CFS in men. Evidence for ethnic inequity is clearest where diagnosis and benefits are used as proxies for disease. In a New Zealand cohort restricted to working-age benefit recipients, the recorded ME/CFS population was disproportionately European and underrepresented Māori, Pacific, and Asian people; because the cohort excluded people not receiving a health or disability benefit, it cannot estimate ethnic prevalence. ^[26] Socioeconomic disadvantage can also reduce ascertainment: Norwegian and UK data found fewer recorded diagnoses in more deprived areas, a pattern more consistent with unequal access to assessment than with lower disease occurrence. ^[21]

Familial aggregation should be assessed without implying a simple inherited disorder. Genetic studies support polygenic susceptibility, but the available population evidence does not provide a clinically usable recurrence risk for relatives. ^[25] Ask about affected first-degree relatives because the answer may inform family counseling and research referral, while emphasizing that shared infections, household exposures, and health-care access can mimic genetic clustering.

An infection-associated onset is common but not universal. In the European ME Association survey, 49% of respondents reported an infection before illness onset; 62.4% of DecodeME participants reported infection as the trigger, and 27% of those infection-triggered reports identified infectious mononucleosis or glandular fever. ^[21] Prospective follow-up after infectious mononucleosis has reported ME/CFS in approximately 10% of participants at 12 months, but this estimate applies to a defined, infected cohort and cannot be multiplied by the population prevalence of Epstein-Barr virus infection. ^[21] Acute infections other than infectious mononucleosis, including influenza-like illnesses, are also reported antecedents. A rise in reported onset around the 2009 influenza pandemic is a temporal association rather than evidence that influenza caused the subsequent cases. ^[21]

SARS-CoV-2 infection has added a large, separately ascertained postinfectious population. In a prospective cohort of hospital employees with persistent post-COVID fatigue, 7 of 221 participants fulfilled the Canadian Consensus Criteria for ME/CFS, with an extrapolated estimate of 3.2-9.0%; the broad range reflects staged recruitment through persistent fatigue and assumptions about participants who were not fully assessed. ^[22] Post-COVID condition and ME/CFS therefore overlap but should not be counted as synonymous epidemiologic categories: a post-COVID cohort requires a preceding SARS-CoV-2 infection, whereas ME/CFS does not. ^[22][25]

Patients also report onset after surgery, injury, pregnancy or childbirth, vaccination, traumatic life events, and periods of major physiological or psychological stress. In the European survey, accident/injury/surgery and pregnancy/birth were reported more often among later-onset respondents, while infection was more common in early-onset respondents; every trigger category occurred in both age groups. ^[21] These observations are useful when reconstructing the illness timeline, but they are retrospective associations subject to recall and selection bias and do not establish causation. Avoid describing a trigger as the cause unless a prospective study supports that inference.

Underrecognition is not a minor adjustment to prevalence; it changes who appears to have the disease. In one UK survey, only 22.1% of patients reported diagnosis within 1-2 years, while 12.9% reported a delay exceeding 10 years; the EMEA report described average delays of 5 years in the UK, 6 years in Norway, and up to 12 years in other European countries. ^[21] Limited medical education contributes to missed diagnosis, and a point-of-care algorithm introduced at Mayo Clinic was associated with increased referrals and better agreement between referral coding and specialist diagnosis. ^[33] Patients may reach specialist care after years of fragmented evaluations; a Minnesota specialty-clinic study found that disease-directed pharmacotherapy had usually not been tried before referral, suggesting a substantial primary-care and specialist-access gap rather than absence of patient need. ^[31]

Stigma compounds delayed recognition. ME/CFS has been described as misunderstood, denied, and insufficiently represented in clinical attention and research investment, while patients may encounter care environments that do not validate the illness. ^[25][32] A fluctuating disability, socioeconomic disadvantage, and restrictive benefit systems can further separate recorded cases from the true affected population. In New Zealand benefit data, only 18.3% of the ME/CFS cohort was currently employed, but that figure describes a selected group already receiving health or disability support and must not be generalized to all people with ME/CFS. ^[26]

ME/CFS is best approached as a disorder of impaired physiological adaptation rather than as a disease with one demonstrable lesion. The leading models converge on a loss of resilience: immune, metabolic, vascular, autonomic, sleep-related, and central nervous system abnormalities may remain subtle at rest but become evident when the patient is challenged by exertion, upright posture, cognitive load, or sensory stimulation. This is a mechanistic hypothesis, not an established unified diagnosis; current reviews describe coordinated abnormalities across several systems but also emphasize cohort dependence and limited reproducibility. ^[34]

Postinfectious immune perturbation: trigger, persistence, and causation

An infection may initiate ME/CFS in a susceptible person without remaining the sole driver of illness. A plausible sequence is incomplete resolution of the acute immune response, persistent danger signalling or restricted viral activity, immune-cell remodeling, and downstream effects on metabolism, vascular regulation, and neural function. Epstein-Barr virus, enteroviruses, and SARS-CoV-2 have been associated with prolonged fatigue syndromes or ME/CFS-like phenotypes, but infection is more appropriately treated as a potential trigger than as a universally established cause. ^[35]

The distinction between association and causation matters clinically and scientifically. A cytokine, metabolite, autoantibody, microbial pattern, or imaging signal may differ between patients and controls because it contributes to disease, results from disease, reflects medication or inactivity, or identifies only a subgroup. A reproducible biomarker is not automatically causal: reproducibility establishes that a signal is consistently measurable under defined conditions, whereas causation requires temporal evidence, a credible mechanism, dose or perturbation relationships, and ideally reversal of the phenotype when the pathway is modified. Current ME/CFS studies rarely satisfy all of these requirements. Even transcriptomic and proteomic reanalyses have found little convergence at the level of individual molecules, although higher-order mitochondrial dysregulation may recur across datasets. ^[34]

Long-COVID research supports the plausibility of postinfectious immune-metabolic illness but does not prove that ME/CFS has an identical mechanism. ME/CFS and long COVID can show overlapping autonomic, small-fiber, cognitive, immune, vascular, and metabolic findings, while comparative immunophenotyping also identifies distinct immune profiles between the two conditions. ^[41][42] The appropriate inference is mechanistic overlap in some patients, not disease equivalence or proof of a persistent pathogen.

Abnormal responses to exertion and impaired bioenergetic adaptation

The exertion model proposes that ME/CFS is characterized less by a fixed inability to generate energy than by an abnormal transition between metabolic states. During activity, skeletal muscle, immune cells, endothelium, and the brain must increase substrate use, oxygen delivery, ATP production, redox buffering, and autonomic support. If that coordinated response is delayed, blunted, or poorly recovered, a previously tolerated activity can expose a temporary system-wide failure. Two-day cardiopulmonary exercise testing (CPET) has therefore been used as a physiological challenge rather than as a routine diagnostic test. Some studies have reported reproducible reductions in peak oxygen consumption, workload, and anaerobic threshold on the second day, whereas a 2026 replication found no significant day-to-day change in peak oxygen consumption, ventilatory threshold, or ventilatory efficiency and concluded that the protocol should not define PEM or disability. ^[34][39] The conflicting results make impaired recovery a plausible disease feature in some cohorts, not a validated standalone test.

Cellular bioenergetic models provide a possible explanation for this stress dependence. Multi-omics studies have reported disturbances involving the citric-acid cycle, fatty-acid oxidation, amino-acid metabolism, redox balance, and mitochondrial function, while chronic inflammatory states can reprogram immune-cell metabolism and alter the balance between glycolysis and oxidative phosphorylation. ^[35] The “cell danger response” model interprets these changes as an evolutionarily conserved shift toward defense and repair after infection or injury; the proposed problem is persistence of that state after the initiating threat has resolved, with reduced metabolic flexibility rather than a single inherited respiratory-chain defect. ^[34] These observations support a model of constrained adaptation, but serum-transfer experiments, metabolomic signatures, and cellular respiration findings remain mechanistic associations unless they are reproduced across well-phenotyped cohorts and shown to precede or reverse illness.

Autonomic, circulatory, endothelial, and microvascular models

Orthostatic intolerance offers a clinically observable route by which circulatory dysregulation could amplify exertional and cognitive symptoms. Upright posture shifts blood into dependent venous compartments; normal compensation requires venous return, skeletal-muscle pumping, increased heart rate, vasoconstriction, and appropriate regulation of circulating volume. Proposed abnormalities in ME/CFS include low cardiac output associated with a small left ventricle, reduced renin-aldosterone and antidiuretic-hormone responses, excessive or ineffective sympathetic activation, reduced cerebral blood flow, and vestibular or postural disequilibrium. ^[40] These mechanisms are not universal, and orthostatic findings vary with hydration, medications, conditioning, disease severity, and the protocol used.

Endothelial and microvascular hypotheses extend this model from blood-flow regulation to tissue oxygen delivery. Retinal vascular studies in post-COVID cohorts, including participants meeting ME/CFS criteria, have found reduced venular dilation, narrower retinal arterioles, altered arteriolar-to-venular ratios, and associations with inflammatory markers; the most pronounced abnormalities occurred in the ME/CFS-phenotype subgroup. ^[46] Such findings are compatible with impaired vascular reserve or microvascular regulation, but they come mainly from observational post-COVID research and do not establish that microvascular dysfunction initiates ME/CFS or explains every patient’s illness. A vascular bottleneck could also be downstream of immune, autonomic, metabolic, or coagulation abnormalities.

Immune and inflammatory signaling

Immune models propose persistent remodeling rather than one uniform inflammatory state. Candidate processes include altered monocyte and dendritic-cell phenotypes, impaired immune-cell trafficking, abnormal costimulation, T-cell exhaustion or suppression, immune-complex effects, and low-grade cytokine signaling that alters mitochondrial and endothelial function. In a cross-sectional comparison, long COVID showed M2-like monocyte polarization, increased CD80 expression, expanded dendritic-cell populations, and features of immune exhaustion, whereas ME/CFS showed reduced costimulatory expression, impaired CCR7-mediated trafficking, and less coordinated activation. ^[42] These differences argue against a single “ME/CFS cytokine profile” and favor biological subgroups.

Small studies have reported higher serum interleukin-11 and matrix metalloproteinase-9 in ME/CFS, and in vitro Epstein-Barr-virus protein stimulation induced matrix metalloproteinase-9 release from cultured mast cells. ^[53] The finding is hypothesis-generating: it does not establish active Epstein-Barr-virus infection, mast-cell disease in vivo, or a treatment target. Similarly, altered extracellular vesicles and mitochondrial membrane potential in selected immune-cell-derived vesicles may link immune signaling to bioenergetics, but the reported profiles remain candidate biomarkers rather than diagnostic tests. ^[52]

Hypothalamic-pituitary-adrenal-axis and neuroendocrine models

The hypothalamic-pituitary-adrenal (HPA) axis coordinates corticotropin-releasing hormone, adrenocorticotropic hormone, and cortisol responses to physiological stress. A proposed ME/CFS model is a maladaptive or blunted stress response in which chronic immune signaling, altered sleep, autonomic dysfunction, and metabolic stress change HPA-axis set points and reduce the ability to respond to challenge. HPA-axis dysregulation has historically been invoked in fatigue biology, but current evidence does not establish a uniform abnormality in ME/CFS, and an isolated cortisol measurement cannot diagnose the illness. ^[35] Neuroendocrine findings should therefore be interpreted as potential state markers or modifiers, not as proof of adrenal disease or a reason to prescribe glucocorticoids.

Gastrointestinal and microbiome models

The gut model links intestinal symptoms, barrier function, microbial metabolites, mucosal immunity, and systemic signaling. Cross-sectional studies report altered but overlapping microbial communities in ME/CFS and long COVID, including shared enrichment of Sutterella and depletion of Terrisporobacter and Lachnospiraceae compared with healthy controls; alpha diversity may remain unchanged despite disease-associated differences in community structure. ^[44] These findings do not demonstrate dysbiosis as a cause, because diet, transit time, medications, activity, geography, and illness-related behavior can all alter the microbiome.

A clinically relevant association is that gastrointestinal symptom burden correlates with fatigue, cognitive difficulty, pain, sleep disturbance, sensory symptoms, and inflammatory measures; one study found a prior irritable bowel syndrome diagnosis in 53% of ME/CFS participants compared with 8% of controls. ^[45] The observations support gut-immune-brain communication as a research model, not routine microbiome-directed treatment. Short-chain fatty acids and other microbial metabolites can influence immune signaling in experimental systems, but their role in human ME/CFS remains incompletely defined. ^[35]

Sleep and circadian disruption

Sleep disturbance may be both a symptom amplifier and a biological contributor. Sleep regulates autonomic tone, endocrine rhythms, immune signaling, synaptic recovery, and energy metabolism; unstable or nonrestorative sleep could therefore worsen the response to exertion without being the primary lesion. In free-living monitoring, people with ME/CFS spent more time in bed, had lower sleep efficiency, and showed greater night-to-night variability in sleep efficiency than matched controls, despite similar total sleep time. ^[50] Observational light-exposure data also associate irregular or nocturnal light patterns with worse clinical measures, but causality cannot be inferred. ^[51] These findings justify careful assessment for coexisting sleep and circadian disorders while leaving the direction of causation unresolved.

Central nervous system processing, sensory gain, and cognition

Central models propose altered integration of interoceptive, autonomic, sensory, motor, and cognitive signals. A modest peripheral disturbance may become disabling if the brain inaccurately estimates available energy, amplifies bodily threat signals, or fails to increase motor output during sustained activity. In a small multimodal grip-force study, people with ME/CFS developed fatigability earlier despite similar maximal force, and unlike controls they showed little increase in brain and muscle activity to maintain the task; the authors interpreted this as evidence for a central contribution. ^[47] Cerebrospinal-fluid measurements in a cross-sectional postinfectious cohort also found lower noradrenergic-pathway activity, but this was an association and does not establish a neurotransmitter lesion or treatment indication. ^[43]

Sensory and cognitive processing abnormalities may arise at higher-order stages rather than at the first stage of stimulus registration. In an auditory event-related-potential study, early P50 suppression did not differ from controls, whereas the later P300 potential was reduced in ME/CFS participants with greater sensory problems. ^[49] Diffusion imaging studies have also reported microstructural differences in selected brain regions, including the cingulum, supplementary motor areas, and corpus callosum, although small samples and cross-sectional designs limit causal interpretation. ^[48] These results support altered central processing as one component of a multisystem model, not a single neuroanatomical explanation.

Why no routine test establishes the diagnosis

Routine clinical tests generally sample resting blood counts, chemistry, endocrine measures, inflammatory markers, and organ function. They are valuable for identifying alternative diagnoses and comorbid disease, but they do not reproduce the dynamic conditions under which ME/CFS abnormalities may emerge. The absence of a consistently abnormal resting result therefore neither proves normal physiology nor confirms a particular mechanism. Reviews continue to describe the absence of a validated biomarker, and candidate findings such as cytokines, extracellular vesicles, metabolites, vascular measures, and neuroimaging signals have not yet achieved adequate validation across independent, clinically defined cohorts. ^[34][52][53]

Research heterogeneity compounds the problem. Broad definitions that do not require PEM can combine patients with different biology, while more stringent definitions enrich for exertional and multisystem phenotypes but reduce comparability with older studies. The resulting mixture changes effect sizes, obscures subgroup-specific signals, and can make a true abnormality appear irreproducible. ^[34][36] Until studies use transparent case definitions, objective longitudinal phenotyping, challenge paradigms that are safe and ethically justified, and prespecified replication, mechanistic findings should be reported as associations or candidate models rather than diagnostic criteria.

Pearl: ME/CFS biology is most coherently modeled as state-dependent failure of coordinated adaptation across immune, metabolic, vascular, autonomic, sleep, and central nervous systems; no proposed pathway currently has sufficient specificity or causal proof to replace clinical diagnosis or serve as a standalone treatment target. ^[34][36][53]

The clinical signature of myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is not ordinary tiredness but loss of the ability to sustain and recover from previously tolerated activity. Post-exertional malaise (PEM), also called post-exertional symptom exacerbation, is a delayed and disproportionate increase in symptoms after physical, cognitive, emotional, orthostatic, or sensory demand. Even a modest task, showering, reading, a conversation, standing in a queue, or attending an appointment, may be followed several hours later by a multisystem deterioration that is more severe than the initiating activity would predict. The exacerbation may persist for days or weeks, and rest does not promptly restore pre-exertional function. ^[56]

PEM has a characteristic temporal structure. The patient may feel unchanged, or even temporarily better, immediately after an activity, then deteriorate later that day or the next day. The delayed phase can include intensified fatigue, cognitive slowing, pain, unrefreshing sleep, orthostatic symptoms, gastrointestinal disturbance, and flu-like sensations. A 2025 exercise study found that symptom changes were greatest during the first 72 hours after a 25-minute moderate-intensity exercise task, although individual responses varied and some participants had little or no measurable increase. ^[54] This variability does not invalidate the history: PEM is defined by the patient’s reproducible loss of post-exertional capacity, not by a uniform response to a laboratory challenge. A single exercise test therefore cannot establish or exclude PEM, particularly when the provoking workload, baseline state, medication exposure, sleep, orthostatic stress, and recovery environment differ from ordinary life. ^[54]

The practical distinction from ordinary fatigue is recovery. Ordinary tiredness generally improves with sleep or rest and does not produce a delayed, disproportionate collapse after routine activity. In ME/CFS, the patient often describes a reduced threshold for both physical and cognitive exertion: an activity tolerated before illness, or tolerated on one day, may provoke symptoms after illness onset or during a flare. The patient may therefore alternate between activity and enforced inactivity, not because of lack of motivation but because exceeding the available physiological capacity produces a later worsening. Functional capacity fluctuates over hours, days, and longer intervals; some patients have relapsing-remitting courses, whereas others experience persistent deterioration or gradual progression. A small 16-year follow-up of previously diagnosed women reported progression in 15%, but its size and follow-up participation limit generalization. ^[27]

Postinfectious onset is common but not required. ME/CFS may begin after an acute infectious illness, including infectious mononucleosis, or after another physiological stressor; the resulting phenotype is defined by persistent impaired function and PEM rather than by the identity of the initiating event. Contemporary reviews describe ME/CFS as a disorder of impaired adaptive capacity within postinfectious disease biology, while also emphasizing that the underlying biology remains heterogeneous. ^[34] A post-COVID syndrome may contain an ME/CFS-like phenotype, but the two labels are not interchangeable: post-COVID illness requires preceding SARS-CoV-2 infection, whereas ME/CFS does not. ^[34]

The symptom cluster

Unrefreshing sleep is not simply short sleep. Patients may sleep for a conventional duration yet awaken unrefreshed, with the same exhaustion or cognitive dysfunction present before sleep. Sleep may be fragmented, delayed, excessively prolonged, or associated with an abnormal sleep-wake pattern. In survey data, sleep disturbance was among the most frequently reported ME/CFS symptoms, alongside fatigue, PEM, cognitive dysfunction, and pain. ^[16] Sleep quality may worsen during PEM, creating a reinforcing cycle in which the patient begins the next day with less reserve.

Cognitive impairment , often called “brain fog,” is usually described in functional rather than purely neurological terms: slowed processing, impaired attention, word-finding difficulty, poor working memory, reduced ability to multitask, and difficulty planning or following a conversation. Cognitive demand can provoke PEM just as physical exertion can. Patients may read a page repeatedly, lose the thread of a discussion, or be unable to complete familiar administrative tasks. A case-control study found poorer self-reported cognitive and neurological well-being in ME/CFS than in healthy controls, although its objective cognitive tests did not show a group effect; subjective cognitive disability therefore remains clinically important but is not captured by every brief test. ^[58]

Orthostatic intolerance is an inability to remain upright without symptoms such as light-headedness, palpitations, weakness, blurred vision, nausea, dyspnoea, cognitive worsening, or presyncope. Symptoms may occur during standing, prolonged sitting, warm environments, bathing, or after meals, and may improve when the patient lies down. Postural orthostatic tachycardia syndrome and orthostatic hypotension are recognizable subtypes, but orthostatic symptoms can occur without either formal haemodynamic diagnosis. In an outpatient cohort of 193 adults seeking ME/CFS care, 32.5% demonstrated postural orthostatic tachycardia syndrome or orthostatic hypotension on tilt-table testing; those with either finding reported more sleep problems, PEM, and functional limitation. ^[59]

Pain is often widespread and may include myalgia, arthralgia, neuropathic or burning sensations, and tenderness. Headache may be new, migrainous, pressure-like, or worsened by upright posture, sensory stimulation, sleep loss, or PEM. Patients may also report heightened sensitivity to light, sound, touch, odours, temperature, or movement. These symptoms can amplify the cost of an otherwise modest activity: a noisy clinic, bright screen, or prolonged conversation may be as exhausting as walking. Survey and cohort data consistently place pain among the dominant symptom domains, but the distribution and intensity vary substantially between patients. ^[16][67]

Autonomic symptoms extend beyond orthostatic intolerance and may include palpitations, abnormal sweating, altered bowel motility, urinary frequency or urgency, temperature dysregulation, and an unstable response to exertion. Thermoregulatory disturbance may present as chills, heat intolerance, cold extremities, episodic sweating, or difficulty maintaining a comfortable temperature. These symptoms often fluctuate with posture, meals, sleep, ambient temperature, and PEM rather than remaining constant. The breadth of autonomic complaints helps explain why patients may describe simultaneous cardiovascular, gastrointestinal, urinary, and cognitive changes during a flare. ^[18][59]

Flu-like or immune-type symptoms include sore throat, tender lymph nodes, malaise, chills, feverishness without documented fever, and a sense of systemic illness. They commonly intensify after exertion and may occur together with diffuse pain, headache, sleep disruption, and cognitive slowing. Gastrointestinal symptoms range from nausea, abdominal discomfort, bloating, altered bowel habit, and food-related symptom worsening to a coexisting diagnosis of irritable bowel syndrome . In a 2026 case-control study, 53% of patients with ME/CFS versus 8% of controls reported a prior diagnosis of irritable bowel syndrome; gastrointestinal burden correlated with fatigue, cognitive difficulty, flu-like symptoms, pain, sleep disturbance, neurological complaints, and sensory sensitivity. ^[45] Genitourinary complaints, including urinary frequency, urgency, nocturia, pelvic discomfort, and altered sexual function, may accompany the autonomic and gastrointestinal phenotype, although they require assessment for infection, inflammatory disease, medication effects, and other urological or gynaecological causes. ^[18]

Mood symptoms require careful interpretation. Frustration, grief, fear of another crash, demoralization, and social withdrawal may be understandable reactions to fluctuating disability, diagnostic delay, and loss of role function. Anxiety or depressive disorder may also coexist and deserves independent assessment, but neither should be presumed to explain PEM, orthostatic intolerance, unrefreshing sleep, or the multisystem pattern. In a comparative study, people with ME/CFS reported worse anxiety and depression-related well-being than controls, alongside worse pain, sleep, mobility, breathlessness, and neurological symptoms; the cross-sectional design cannot determine whether mood symptoms were causes, consequences, comorbidities, or interacting factors. ^[58]

The phenotype is therefore best recognized as a pattern: impaired baseline function, unrefreshing sleep, cognitive and/or autonomic symptoms, and a reproducible delayed loss of capacity after activity, accompanied by variable pain, sensory, gastrointestinal, genitourinary, thermoregulatory, flu-like, and mood symptoms. The defining clinical question is not “How tired is the patient?” but “What happens after the patient spends the available energy?” ^[56][67]

Begin with function, not the word “fatigue.” Ask the patient to describe what they could do before illness, what changed, and what they can do now on a good day and a bad day. Establish onset, abrupt or gradual, with a specific date or illness if possible, and confirm that the reduction has persisted for the duration required by the chosen case definition. Quantify ordinary activities: work or school attendance, walking, bathing, shopping, meal preparation, social contact, and time spent recumbent. Ask which activities have been abandoned, which require assistance, and what happens afterward. A patient who reports severe tiredness but unchanged activity has not yet demonstrated the functional impairment required for an ME/CFS diagnosis. ME/CFS diagnostic frameworks differ in their duration requirements and symptom composition, so the clinician should record the framework used rather than treating “chronic fatigue” as a diagnosis. ^[1]

History that establishes the phenotype

Characterize post-exertional malaise (PEM) as a reproducible deterioration after activity that was previously tolerated. Ask: “What happens if you do more than usual?” and then clarify the activity, the delay before worsening, the symptoms that appear or intensify, the degree of functional loss, and the time required to return to baseline. Include physical exertion, prolonged standing, cognitive work, emotional stress, travel, sensory stimulation, and medical appointments; patients may not recognize these as exertion. Ask specifically whether symptoms are worse later the same day or the next day, because PEM may be delayed rather than immediate. Record a concrete example, for instance, the consequence of a shower, a short walk, a school day, or a family event, rather than accepting “I crash.” Patient-reported PEM is central to current diagnostic frameworks, but symptom questionnaires remain subjective and can be affected by recall bias. ^[60]

Distinguish baseline symptoms from the post-activity change. Ask the patient to describe function and symptoms before the trigger, during the first several hours afterward, at 24-72 hours, and at recovery. Document whether the patient must reduce activity, remain in bed, avoid upright posture, or obtain help with basic activities. A brief activity-and-symptom diary can reveal delayed deterioration and fluctuation that a single office history misses; do not prescribe a provocation test to demonstrate PEM. Exercise-based physiological testing is burdensome and is not required for routine diagnosis. ^[60]

Assess the associated domains in the same time-linked manner. Ask whether sleep is unrefreshing despite adequate opportunity, whether sleep onset or maintenance is difficult, and whether the patient has snoring, witnessed apnoeas, restless legs, circadian reversal, or abnormal nocturnal behaviours. Ask about slowed thinking, impaired attention, word-finding difficulty, reduced working memory, and the consequences for reading, conversation, driving, medication management, or work. Screen for orthostatic symptoms, light-headedness, palpitations, tremulousness, blurred vision, nausea, weakness, disequilibrium, or presyncope when standing, and ask whether symptoms improve on sitting or lying down. Ask about syncope and falls. Orthostatic intolerance may occur without a formal diagnosis of postural orthostatic tachycardia syndrome or orthostatic hypotension; the symptom history therefore matters even when a bedside measurement is normal. ^[1]

Ask about pain in the muscles, joints, head, chest, abdomen, and pelvis; sensory hypersensitivity; sore throat or tender lymph nodes; temperature dysregulation; gastrointestinal and genitourinary symptoms; and recurrent flu-like sensations. Record fluctuation across the day, week, menstrual cycle, seasons, infections, and changes in sleep or medication. Pain and cognitive symptoms may be primary manifestations, comorbid disorders, or consequences of repeated PEM; the temporal relationship helps separate these possibilities. ^[60]

Take a chronology of preceding infection and other precipitants without making an infection a prerequisite. Ask about febrile illness, infectious mononucleosis, respiratory or gastrointestinal infection, probable or confirmed SARS-CoV-2 infection, vaccination, surgery, injury, pregnancy and childbirth, and major physiological or psychological stress. Record the date of the event and the interval to loss of function. A prior SARS-CoV-2 infection supports consideration of post-COVID-19 condition , but ME/CFS does not require COVID-19; the two diagnoses may coexist. ^[1]

Review prescribed, over-the-counter, and recreational drugs, including sedatives, antihistamines, stimulants, opioids, cannabis, alcohol, and supplements. Ask about dose changes, withdrawal, adherence, and drugs that worsen sleep, cognition, orthostatic symptoms, or exercise tolerance. Clarify sleep schedule and sleep opportunity; mood, trauma, anxiety, and depression; eating pattern, weight change, food restriction, vomiting, diarrhoea, and hydration; pregnancy possibility; and symptoms of endocrine, inflammatory, infectious, cardiopulmonary, neurological, and systemic disease. Ask what the patient believes is happening and what they fear has been missed. This both improves diagnostic accuracy and prevents psychological symptoms from being used to explain PEM or orthostatic intolerance without examining the rest of the phenotype. ^[60]

Examination

Measure temperature, pulse, blood pressure, respiratory rate, oxygen saturation, weight, and body habitus. Repeat pulse and blood pressure after standing when orthostatic symptoms, presyncope, syncope, palpitations, or marked activity intolerance are present and the patient can do so safely; record posture, timing, symptoms, heart-rate response, and blood-pressure response rather than reporting a single number. Formal autonomic testing is reasonable for unexplained syncope, recurrent presyncope, disabling orthostatic symptoms, or an equivocal bedside assessment. In ME/CFS cohorts, orthostatic testing identifies heterogeneous haemodynamic patterns, so an abnormal result can characterize a comorbidity but does not establish ME/CFS by itself. ^[73]

Perform a focused neurological examination: mental status and speech, cranial nerves, eye movements, strength, tone, reflexes, sensation, coordination, gait, tandem gait, and Romberg testing when safe. Examine for focal deficits, movement disorder, objective weakness, sensory level, cerebellar signs, or progressive gait impairment. Examine the heart and lungs for rhythm abnormality, murmurs, signs of heart failure, wheeze, and impaired ventilation; inspect the thyroid and palpate cervical, axillary, and inguinal lymph nodes. Assess joints, muscles, range of motion, tenderness, skin, oral mucosa, hydration, and signs of nutritional deficiency. Examine for pallor, fever, oedema, cachexia, or inflammatory features that demand another diagnostic pathway.

Assess function without provoking PEM. Observe transfers, gait, speech endurance, ability to sit upright, and the cognitive load tolerated during the consultation. Ask whether the examination itself is likely to worsen symptoms and offer breaks, a quiet room, hydration, recumbency, or remote review. Do not use a normal brief office examination to discount severe disability: resting examination may be normal between relapses, whereas the clinically decisive abnormality is delayed deterioration after activity. Candidate physiological tests and biomarkers have not replaced clinical assessment; current diagnosis still relies predominantly on patient-reported symptoms. ^[1]

Applying criteria in practice

Choose one case definition before concluding, document it in the assessment, and apply every required element explicitly. For routine adult practice, the 2015 National Academy of Medicine / Institute of Medicine framework is useful because it requires substantial functional impairment with new, persistent fatigue; PEM; unrefreshing sleep; and either cognitive impairment or orthostatic intolerance. The NICE framework is an alternative when its three-month duration and required symptom set are being used. Fukuda criteria may identify broader historical cohorts but do not require PEM, so a patient who meets Fukuda alone should not automatically be labelled as having the narrower PEM-defined phenotype. The evidence base remains difficult to compare because studies frequently use different or unspecified criteria. ^[12]

For each required feature, write the patient’s evidence in observable terms: “can walk 10 minutes but develops flu-like symptoms and marked cognitive slowing the following day, requiring two days of reduced activity,” rather than “positive PEM.” Record the duration, severity, frequency, functional consequence, and whether the feature is present at baseline or only after exertion. Then document the assessment of alternative explanations and comorbid conditions without presenting an exhaustive differential in the ME/CFS assessment note. A coexisting condition does not exclude ME/CFS if it does not adequately explain the complete pattern, particularly delayed PEM and the combination of sleep, cognitive, autonomic, and multisystem symptoms.

When criteria are incompletely met, communicate uncertainty precisely. Use terms such as “probable ME/CFS,” “ME/CFS phenotype under evaluation,” or “does not currently meet the selected criteria,” and state which element is absent, uncertain, or lacks the required duration. Do not use “mild” or “atypical” to conceal a missing mandatory feature. Arrange review after an interval appropriate to the clinical problem, reassess function and PEM, and revise the diagnosis if a better explanation emerges. A negative or normal routine test does not exclude ME/CFS, and a positive autonomic or sleep test does not confirm it; each finding must be interpreted against the longitudinal history. ^[1]

Clinical assessment checklist

Pearl: Diagnose the pattern, not the symptom label: persistent functional reduction plus characteristic delayed PEM, supported by the selected criteria and a deliberate search for better explanations, is more informative than any isolated fatigue score, normal examination, or abnormal ancillary test. ^[1]

The purpose of investigation is to identify a treatable disorder that better explains the patient’s illness, detect an important comorbidity, and establish a safe baseline, not to produce a laboratory signature for ME/CFS . No single laboratory finding confirms or excludes the diagnosis, and patients may have substantial illness despite normal routine results. ^[83][86] Order the initial panel once, interpret it in the clinical context, and repeat it only when symptoms, examination, treatment, or elapsed time creates a new indication.

For most adults, obtain a complete blood count , electrolytes, renal function, liver tests, calcium, glucose or HbA1c , thyroid-stimulating hormone , C-reactive protein or erythrocyte sedimentation rate , ferritin, and urinalysis . These tests screen for anemia or cytopenias, renal or hepatic disease, calcium and glucose disorders, thyroid disease, systemic inflammation, iron deficiency, and urinary or renal abnormalities. They do not establish ME/CFS; an abnormal result should trigger focused evaluation rather than automatic attribution of every symptom to the abnormality. ^[83][86]

The history and examination determine the second tier. Add vitamin B12 and folate for macrocytosis, neuropathy, cognitive change, restricted diet, malabsorption risk, or relevant medication exposure; vitamin D when deficiency risk, bone or muscle symptoms, limited sunlight exposure, or malnutrition is present; and creatine kinase for objective weakness, myalgia with muscle tenderness, dark urine, or a myopathic or drug-related concern. Use celiac serology , usually tissue-transglutaminase IgA with total IgA, for chronic diarrhoea, weight loss, iron deficiency, malabsorption, autoimmune disease, compatible gastrointestinal symptoms, or unexplained fatigue; serology is the recommended first screening step. ^[88]

Perform a pregnancy test when pregnancy is possible and the result would alter the diagnostic or treatment plan. Test for HIV, hepatitis B, hepatitis C, or other infections when exposure, constitutional features, examination findings, laboratory abnormalities, or geography raises the pre-test probability; do not use broad infectious panels merely because illness began after an infection. Remote viral serologies, including isolated evidence of past Epstein-Barr virus infection, neither prove persistent infection nor explain the ME/CFS phenotype in the absence of a compatible acute syndrome. ^[83][86]

Investigate a specific symptom domain rather than ordering an indiscriminate battery. Arrange polysomnography for loud snoring, witnessed apnoea, nocturnal choking, resistant hypertension, marked obesity, parasomnias, or persistent unrefreshing sleep that suggests a primary sleep disorder; objective sleep findings in ME/CFS are heterogeneous, so a sleep study is not a confirmatory test. ^[89] Obtain an electrocardiogram for palpitations, syncope or presyncope, chest discomfort, unexplained dyspnoea, abnormal pulse, a relevant medication, or cardiovascular risk. Add echocardiography for abnormal cardiac examination, abnormal ECG, exertional or positional dyspnoea, oedema, syncope with cardiac features, or suspected structural heart disease, and use ambulatory rhythm monitoring when episodic palpitations, unexplained syncope, or intermittent tachyarrhythmia is suspected.

Measure supine and standing pulse and blood pressure when orthostatic symptoms are prominent, while protecting the patient from falls and avoiding prolonged provocation. Refer for tilt-table testing or formal autonomic testing when syncope, recurrent presyncope, disabling orthostatic intolerance, or equivocal bedside measurements require haemodynamic characterization. Tilt testing may document postural orthostatic tachycardia syndrome , orthostatic hypotension, or another autonomic disorder, but it does not confirm or exclude ME/CFS; autonomic symptoms can occur without a formal haemodynamic diagnosis. ^[13][86]

Use neurological studies selectively. Obtain magnetic resonance imaging of the brain or spine for focal neurological deficits, seizures, a new objective cognitive or motor syndrome, papilloedema, a progressive course suggesting structural disease, or another specific neurological indication, not for uncomplicated cognitive dysfunction, headache, or ME/CFS itself. Arrange nerve-conduction studies, electromyography, or specialist neurological assessment for objective weakness, sensory loss, progressive neuropathy, abnormal reflexes, fasciculations, or a suspected neuromuscular disorder. Do not provoke post-exertional malaise with exercise testing; a cardiopulmonary exercise test may be justified only for a separate cardiopulmonary or occupational question in a setting able to manage the patient’s risk and exertional consequences.

Do not order commercial cytokine panels, unvalidated metabolomic signatures, remote-infection viral serologies, routine brain imaging, or exercise-challenge tests to confirm ME/CFS. Current biomarker studies have not yielded a universal or specific diagnostic marker; even urinary biomarker findings show substantial heterogeneity and limited evidence for a consistent diagnostic signal. ^[92] A normal or abnormal result from one of these assays should not replace longitudinal clinical assessment, and an abnormality discovered incidentally should be interpreted on its own clinical merits rather than treated as proof of ME/CFS. ^[83][86]

Reassess when the clinical pattern changes: new fever, progressive focal findings, objective weight loss, persistent night sweats, bleeding, sustained deterioration outside the patient’s usual fluctuation, or a new organ-specific symptom warrants fresh evaluation. Coexisting disease remains possible; the practical endpoint is not an exhaustive negative workup, but a documented rationale for the tests performed, the abnormalities found, and why the remaining pattern still requires clinical assessment for ME/CFS. ^[86]

Fatigue becomes diagnostically useful only when its pattern is defined. Ask whether function has fallen, whether sleep restores capacity, and whether physical, cognitive, emotional, sensory, or upright activity produces delayed, disproportionate worsening. Post-exertional malaise (PEM) is the most discriminating clinical feature: it is a reproducible change in symptoms and function after exertion, often delayed, and may last well beyond ordinary tiredness.^[95][96] A competing diagnosis should explain the whole syndrome, including PEM, unrefreshing sleep, cognitive dysfunction, orthostatic symptoms, pain, and multisystem fluctuation, or be recorded as a coexisting disorder rather than substituted for ME/CFS.

Use the presentation to direct evaluation. Investigate bleeding, dietary restriction, malabsorption, endocrine symptoms, fever, weight loss, focal neurological signs, cardiopulmonary symptoms, sleep-disordered breathing, medication exposure, and substance use when the history or examination supports them. Do not repeat broad testing without a new clinical indication; a normal routine assessment neither confirms nor excludes ME/CFS.^[95] Targeted testing should answer a specific diagnostic question, and a normal result should not be used to discount PEM or substantial functional impairment.

Coexistence and diagnostic attribution

Do not force a single diagnosis when two conditions are present. Chronic overlapping pain conditions are frequent in specialty cohorts: one study found at least one of six such conditions in 76.1% of participants with ME/CFS, with chronic migraine or headache, fibromyalgia, chronic low back pain, and irritable bowel syndrome among the most frequent; comorbidity was associated with greater symptom burden and worse functioning.^[96] Record which disorder explains each symptom, which findings remain unexplained, and whether treatment or correction of the comorbidity changes PEM.

Fibromyalgia , irritable bowel syndrome , migraine, sleep disorders, POTS, orthostatic hypotension, endocrine disease, anemia, and psychiatric illness can all coexist with ME/CFS. Their presence should prompt domain-specific assessment, not diagnostic dismissal. Conversely, ME/CFS should not be used to absorb new objective findings such as progressive weakness, persistent fever, bleeding, focal neurological deficits, sustained weight loss, or organ-specific inflammation.

Pearl: The diagnostic question is not “Which illness causes fatigue?” but “Which illness explains the delayed, disproportionate response to exertion and the complete multisystem pattern?”

Severity in ME/CFS is a functional judgment, not a symptom count. Record what the patient can actually sustain, how long recovery takes after ordinary demands, and how much assistance is required. A patient may have substantial symptom burden while retaining part-time work, whereas another may have fewer reported symptoms but be unable to remain upright, prepare food, communicate, or tolerate light and sound. Capacity also fluctuates; a single clinic observation can therefore overestimate usual function. ^[110]

These categories are descriptors rather than immutable stages. Record the date, the patient’s current baseline, the degree of fluctuation, and the principal limiting domain, mobility, upright tolerance, cognition, communication, sensory tolerance, or self-care. Record dependence explicitly: who prepares food, administers medicines, assists with bathing or toileting, supervises transfers, provides transport, and monitors the patient after activity. For children and adolescents, document school attendance, reduced timetable, home tuition, missed lessons, cognitive recovery after school, and participation in age-appropriate activities. ^[26]

Assess against an individualized activity baseline. Ask what the patient could do before illness, what is done now on a relatively stable day, what is abandoned, and what happens during the subsequent 24-72 hours after physical, cognitive, emotional, upright, or sensory activity. Include time out of bed, walking distance, transfers, speech, reading, screen use, meals, hygiene, and social contact. Do not use a clinic task, step target, standing challenge, or exercise test as a prerequisite for recognizing disability. When assessment itself may exceed the patient’s tolerance, shorten the encounter, use remote or home assessment, allow recovery periods, and document the unperformed component and the reason. ^[110]

Objective functional documentation should combine patient report with observable or recorded behavior. Activity diaries, wearable activity data when tolerated, school or employment records, work accommodations, sick leave, care plans, prescription or medication-administration records, and caregiver accounts can show reduced and fluctuating capacity. Interpret wearable activity cautiously: low activity confirms inactivity, not its cause, and a brief period of increased activity does not establish sustainable capacity. A two-day cardiopulmonary exercise test should not be used to define PEM or disability; a 2026 replication found no significant day-to-day change in peak oxygen consumption or ventilatory threshold and concluded that the protocol did not support that use. ^[39]

Use outcome measures to answer a defined clinical question, not to manufacture a diagnosis. None of the following instruments, alone or in combination, establishes or excludes ME/CFS; they quantify selected domains and should be interpreted alongside the history of delayed PEM and functional loss. ^[110]

• Activity and symptom diaries: Record activity type, duration, posture, cognitive or sensory load, symptoms before activity, delayed worsening at 24-72 hours, recovery time, sleep, and assistance required. A diary is particularly useful when capacity fluctuates, but completion itself may be burdensome; use a caregiver or a shortened format when necessary. ^[110]
• Orthostatic symptom scales: Record dizziness, presyncope, palpitations, weakness, nausea, cognitive worsening, and inability to remain upright, together with posture and duration tolerated. Pair symptom ratings with safely obtained supine and standing pulse and blood pressure when indicated; symptoms may be disabling even when formal POTS or orthostatic hypotension is not demonstrated. ^[40]
• DePaul Symptom Questionnaire (DSQ): Use its symptom-frequency and symptom-severity ratings to standardize the patient’s account of PEM and associated symptoms, particularly in research or longitudinal review. It is a patient-reported phenotype measure, not a diagnostic test. ^[110]
• PROMIS measures: Select relevant Patient-Reported Outcomes Measurement Information System domains, such as physical function, fatigue, pain interference, cognitive concerns, sleep, and ability to participate in social roles. Use the same domains and administration mode at follow-up; do not infer improvement in overall illness from a change in one domain. ^[111]
• Fatigue Severity Scale and Chalder Fatigue Questionnaire: These can quantify perceived fatigue and its physical or mental dimensions, but neither captures the defining temporal relationship between exertion and delayed PEM. Interpret scores beside activity loss, recovery time, and orthostatic or cognitive limitations. The Chalder Fatigue Scale has been used as an outcome in ME/CFS research, but an apparent score change does not by itself demonstrate improved functional capacity. ^[114]
• Bell disability scale: Use this ordinal disability scale to communicate broad functional status, especially when identifying severe or very severe patients; its value is greatest when accompanied by a narrative description of mobility, self-care, communication, sensory tolerance, and dependence. A score should not replace the underlying functional account. ^[110]
• Work and Social Adjustment Scale (WSAS): Use it to quantify perceived impairment in work, home management, social leisure, private leisure, and close relationships. In unemployed patients, clarify whether the score reflects inability to work, absence of a job, or both; in students, supplement it with attendance and participation records. ^[26]

At each review, repeat the same core measures only if the burden is acceptable and report the context: baseline state, recent infection or comorbidity, sleep disruption, orthostatic symptoms, medication changes, and any activity-related worsening after the assessment. A lower fatigue score with unchanged dependence, school absence, or time confined to bed is not a global functional recovery; conversely, improved attendance achieved by exertion followed by prolonged PEM may represent greater effort rather than greater capacity. ^[111]

Pearl: Classify severity by what the patient can sustain and recover from, document dependence and fluctuation in concrete functional terms, and never make the assessment itself the cause of the disability being measured. ^[110]

Management of ME/CFS is rehabilitative in the broad sense, preserving function, preventing avoidable deterioration, and supporting participation, not restorative exercise therapy. No curative treatment has been established, so the immediate clinical target is to reduce post-exertional malaise (PEM), maintain nutrition and hydration, treat relevant comorbidities, and adapt care to the patient’s fluctuating capacity. ^[121]

Energy management and pacing

Energy management means matching physical, cognitive, emotional, social, orthostatic, and sensory demands to the patient’s current capacity. The energy envelope is the range of activity that can be performed without provoking delayed, disproportionate symptom and functional worsening. It is not a fixed daily allowance: the envelope may contract during infection, poor sleep, pain flares, hormonal changes, emotional stress, or relapse, and may expand only gradually and unpredictably. Ask the patient to identify activities that reliably precede PEM, the usual delay before deterioration, the symptoms that worsen, and the time required to return to baseline. A brief activity-and-symptom diary can make this pattern visible, but monitoring must not become an additional cognitive burden. Pacing is intended to stabilize illness and reduce exertion-related symptoms; the supporting literature remains limited and includes consensus work and small exploratory studies rather than definitive efficacy trials. ^[120][123]

Use pacing as a collaborative process rather than a prescription for a target number of steps, minutes, or calories. Break essential tasks into smaller components, alternate demanding and low-demand activities, perform tasks seated or recumbent when possible, delegate or postpone nonessential work, and schedule rest before, not only after, predictable exertion. Include cognitive and sensory tasks in the same plan as walking or household activity; reading, screen use, conversation, upright posture, travel, and clinical appointments can all consume the available envelope. Avoid the boom-and-bust cycle , in which a relatively good day leads to compensatory overactivity followed by delayed PEM and prolonged inactivity. During a flare, reduce activity to the level that is currently tolerated, prioritize eating, drinking, toileting, hygiene, medicines, and essential communication, and resume other activities only after symptoms have stabilized. NICE advises remaining within the individual energy limit, balancing activity with regular rest, and adjusting the plan during relapse. ^[121]

Pacing differs fundamentally from graded exercise therapy (GET). Pacing is symptom-contingent and permits reduction, substitution, or cessation of activity when symptoms increase; GET traditionally uses planned, fixed or quota-based increments, often with the aim of reversing presumed deconditioning. NICE’s 2021 guideline states that no programme using fixed or quota-based incremental increases in physical activity or exercise, including GET, should be offered as treatment for ME/CFS. A fixed incremental programme or advice to push through PEM can worsen symptoms and function, and should not be prescribed to reverse presumed deconditioning. ^[121] The NICE position has been challenged in a 2023 methodological critique, which argued that the guideline interpreted GET too narrowly and recommended energy management despite limited supportive research; clinicians should recognize this controversy while avoiding forced progression in a patient who develops PEM. ^[125]

A physiotherapist or occupational therapist with ME/CFS expertise may help patients test safer ways to perform necessary activities, but the intervention must remain individualized, reversible, and symptom-contingent. Increased activity is an observation to be earned by sustained stability, not a treatment target. Do not use a normal-looking clinic performance, a single good day, or a wearable activity score as evidence that the patient can sustain a higher workload.

Collaborative, flexible care

Begin with shared decision-making. Agree on a small number of priorities, such as preventing crashes, maintaining reliable meals, improving sleep timing, attending essential education, or reducing orthostatic symptoms, and define success in functional terms chosen by the patient. Explain the rationale and possible burden of each intervention, ask what the patient is willing and able to try, and document the patient’s baseline, current envelope, warning signs, and preferred response to deterioration. A named clinician should coordinate primary care, relevant specialty services, rehabilitation professionals, social care, and, when the patient consents, family or carers. NICE recommends individualized multidisciplinary care, regular review, and flexibility in delivery, including remote consultations and home visits for people who cannot safely travel. ^[121]

Schedule follow-up according to severity and instability rather than relying on routine clinic intervals. Review delayed responses over the preceding 24-72 hours, time out of bed, upright tolerance, self-care, meals, communication, school or work participation, rest periods, sleep timing, and caregiver burden. Reassess after any intervention that increases demand, and reduce or stop it if PEM appears. New persistent fever, bleeding, focal neurological findings, sustained weight loss, progressive organ-specific symptoms, or a marked change from baseline warrants reassessment for another disorder rather than attribution to ME/CFS. Remote, telephone, video, written, or caregiver-assisted review may be safer than requiring travel; minimize waiting-room noise, bright light, prolonged upright posture, and repeated history-taking.

Occupational and school planning should aim for sustainable participation, not rapid return to a preset schedule. Consider reduced or flexible hours, asynchronous or home-based work, rest breaks, reduced sensory and cognitive load, modified deadlines, remote attendance, quiet spaces, permission to leave early, and postponement of nonessential assessments. Liaise with employers, schools, occupational health, disability services, and social-security agencies with informed consent. NICE advises support for reasonable adjustments and recognizes that returning to work, education, or training can worsen symptoms when demands exceed the person’s envelope. ^[121]

Provide mobility aids when they reduce orthostatic load, falls risk, pain, or energy expenditure. A wheelchair, walking aid, shower chair, stair aid, or other adaptation is a means of conserving capacity and preserving independence, not proof of deconditioning or treatment failure. Assess transfers, pressure risk, upper-limb tolerance, home layout, and caregiver support; involve occupational therapy or physiotherapy when the patient can tolerate assessment without PEM. NICE includes aids and adaptations in the individualized care plan when they improve independence or quality of life. ^[121]

Symptom-compatible routines

Encourage a regular sleep-wake schedule without demanding a rigid bedtime or prolonged daytime activity. Schedule quiet rest periods according to individual need, distinguish restorative rest from unplanned collapse after overexertion, and review whether daytime sleep is worsening nocturnal sleep. Use low-stimulation wind-down routines, comfortable positioning, and reduction of evening light and screen exposure when tolerated. Sleep advice must not become a reason to withdraw necessary daytime rest in a patient whose upright or cognitive capacity is very limited. NICE recommends gradual development of regular sleep habits while balancing daytime rest against night-time sleep. ^[121]

Plan food and fluids around the patient’s available energy. Favor simple, nutrient-dense meals, smaller and more frequent portions when nausea or early satiety limits intake, and practical adaptations such as prepared food, delivery, softer textures, or eating aids when shopping, cooking, chewing, or swallowing consumes excessive energy. Screen for weight loss, dehydration, restrictive eating, and difficulty obtaining or preparing food; involve a dietitian for nutritional risk. Encourage adequate fluid intake and a balanced diet, but do not present vitamins, supplements, or restrictive diets as treatments for ME/CFS without a specific nutritional indication. ^[121][133]

For orthostatic symptoms, reduce prolonged standing, use seated or recumbent tasks, rise slowly, keep essential items within reach, and plan bathing, meals, and appointments around the time of day with the greatest tolerance. Consider compression garments, fluid or salt strategies, and specialist treatment only after assessing contraindications such as hypertension, heart failure, kidney disease, or relevant medication effects. Do not prescribe upright exercise as conditioning when it provokes PEM. Management of documented postural orthostatic tachycardia syndrome or orthostatic hypotension should follow the relevant autonomic assessment and specialist advice; orthostatic treatment does not replace pacing because it does not explain or prevent PEM. Evidence for conservative dysautonomia protocols remains preliminary. ^{[121][128][129]}

Reduce sensory load when light, sound, touch, smell, or visual motion provokes symptoms. Offer dimmable lighting, noise reduction, sunglasses or eye shades, quiet rooms, unscented products, reduced screen brightness, text-based communication, and shorter consultations with planned breaks. These adaptations are not avoidance of recovery; they reduce competing demands so that essential activity remains within the energy envelope.

Pearl: In ME/CFS, the safest activity prescription is not “more over time”; it is the greatest participation the patient can sustain without delayed PEM, revised whenever the energy envelope changes. ^[120][121]

Treat symptoms as comorbid clinical problems, not as evidence that ME/CFS is primarily psychiatric or as a reason to increase activity. NICE advises individualized management and states that cognitive behavioral therapy (CBT) should be offered only as supportive treatment; neither CBT nor antidepressant treatment cures ME/CFS. ^[121] EUROMENE describes care as symptom-oriented because no causal treatment exists. ^[138]

Sleep

Ask whether sleep is unrefreshing, delayed, fragmented, phase-shifted, or shortened by pain, pruritus, nocturia, medication effects, or orthostatic symptoms. Review caffeine, alcohol, cannabis, sedating antihistamines, stimulants, hypnotics, and opioid use. A sleep diary or actigraphy over several days is more informative than a single “good” or “bad” night because sleep efficiency can fluctuate substantially from night to night in ME/CFS. ^[50]

Use a stable, individualized sleep-wake schedule; reduce evening light, noise, screens, and temperature extremes; reserve the bed for sleep when feasible; and avoid prescribing rest or sleep routines that require prolonged upright activity or provoke PEM. Screen for loud snoring, witnessed apnoeas, gasping, morning headache, resistant hypertension, obesity, or marked nocturnal restlessness. Arrange polysomnography or validated home sleep-apnoea testing when obstructive sleep apnea or another primary sleep disorder is suspected. Ask specifically about an urge to move the legs, discomfort worse at rest and in the evening, and relief with movement; check ferritin and transferrin saturation when restless legs syndrome is suspected and correct iron deficiency before dopaminergic treatment.

Offer cognitive behavioral therapy for insomnia (CBT-I) when insomnia persists after reversible contributors have been addressed. Deliver it in a low-burden, flexible format, with shortened sessions or remote care for patients whose travel, prolonged sitting, or cognitive load provokes PEM. NICE describes CBT as supportive rather than curative, and psychotherapy has no established curative effect in ME/CFS. ^[121][141]

If medication is required, start below usual doses, change one drug at a time, and review delayed sedation, worsened cognition, falls, orthostatic symptoms, and PEM. Avoid assuming that a sedating drug produces restorative sleep; stop it if the patient is more impaired the following day. The D-A-CH consensus states that medication doses and administration times should be adapted to ME/CFS because some patients respond to low doses. ^[3]

Pain and headache

First determine whether pain is nociceptive, neuropathic, nociplastic, migraine-related, pelvic, gastrointestinal, or caused by a coexisting inflammatory or structural disorder. Use heat or cold according to preference, supported positioning, gentle range-of-motion within the patient’s activity envelope, massage or manual therapy only if it does not provoke PEM, desensitization for allodynia, and occupational adaptation of tasks. Treat migraine according to standard headache practice, but account for dehydration, prolonged upright posture, sleep disruption, and medication-overuse headache.

Use analgesics for a defined pain target and a time-limited trial. Avoid routine opioids when possible because sedation, constipation, cognitive impairment, respiratory depression, and orthostatic worsening can compound ME/CFS disability. For generalized pain or fibromyalgia-like pain, low-dose tricyclic antidepressants, duloxetine, or gabapentinoids may be reasonable specialist-directed trials; they treat pain or sleep and do not treat the underlying disease. A 2026 specialty-clinic review found that serotonin-norepinephrine reuptake inhibitors, gabapentin, and tricyclic antidepressants were among the medications most often used for symptom management, but this was retrospective evidence rather than proof of efficacy. ^[31]

Avoid NSAIDs with active ulcer disease, significant kidney disease, decompensated heart failure, anticoagulation-associated high bleeding risk, or NSAID hypersensitivity. Check blood pressure, renal function, gastrointestinal symptoms, sedation, cognition, falls, and functional change after analgesic initiation; taper rather than abruptly stopping long-term tricyclics, duloxetine, or gabapentin.

Orthostatic intolerance

Measure symptoms together with safely obtained supine and standing pulse and blood pressure. POTS requires orthostatic symptoms and a sustained heart-rate increase of at least 30 beats/min in adults, or at least 40 beats/min in children, without orthostatic hypotension. Orthostatic hypotension is a sustained fall in systolic blood pressure of at least 20 mmHg or diastolic blood pressure of at least 10 mmHg within 3 minutes of standing. ^[129] Neurally mediated syncope is a reflex episode of hypotension, bradycardia, or both that causes presyncope or syncope; it is not interchangeable with POTS or chronic orthostatic symptoms. Formal autonomic testing is useful when syncope, recurrent presyncope, disabling symptoms, or equivocal bedside measurements require clarification, but an abnormal result does not diagnose ME/CFS.

Start with recumbent or seated strategies: rise in stages, avoid prolonged still standing and hot showers, sit or lie down at the first warning symptom, elevate the legs, use a shower chair, and consider abdominal or lower-limb compression. If there is no hypertension, heart failure, kidney disease, cirrhosis, or other contraindication, increase fluids and salt with individualized targets; do not impose a fixed fluid or salt prescription on every patient. A small 2026 feasibility study used a supervised protocol that cumulatively introduced up to 3 L/day of fluid and 10 g/day of salt, but adherence was variable, one participant was withdrawn after a severe crash, and the authors cautioned that the changes might not be clinically meaningful. ^[128]

Monitor standing symptoms, falls, supine and standing blood pressure, pulse, edema, weight, potassium, and renal function when using volume-expanding therapy. Monitor for supine hypertension with midodrine, bradycardia or bronchospasm with beta-blockers, visual phenomena and bradycardia with ivabradine, and hypokalemia or edema with fludrocortisone. Do not prescribe autonomic medication to increase exercise capacity or replace pacing; reduce or stop a drug that worsens delayed PEM.

Comorbid symptoms

Treat coexisting disease on its own diagnostic merits while preserving the patient’s limited cognitive, upright, and sensory capacity. For irritable bowel syndrome , use regular tolerated meals, adequate hydration, individualized soluble fiber or dietetic guidance, and targeted treatment of constipation or diarrhea; avoid unnecessarily restrictive diets and investigate weight loss, bleeding, nocturnal symptoms, anemia, or other alarm features. For allergic rhinitis or urticaria, reduce confirmed triggers and use the least sedating effective antihistamine; sedating antihistamines may worsen cognition, sleep architecture, and orthostatic symptoms. Manage pelvic pain, dysmenorrhea, endometriosis, vulvodynia, bladder pain, and menstrual-related symptom flares with gynecologic or urologic assessment, heat, positioning, hormonal or analgesic treatment when appropriate, and a plan for PEM-sensitive appointments.

Treat depression , anxiety, trauma-related symptoms, and suicidality when present, without attributing PEM or orthostatic intolerance to mood disorder. Offer supportive psychotherapy or CBT for coping, grief, insomnia, fear, and illness-related distress; do not use psychotherapy to challenge the reality of the physical illness or to prescribe fixed activity increases. ^[141] Select antidepressants for the comorbid indication, start low, titrate slowly, and monitor activation, nausea, sleep, orthostatic symptoms, suicidality, and delayed functional deterioration. Urgent psychiatric assessment is required for suicidal intent, psychosis, mania, or inability to maintain safety.

Pearl: Treat the symptom that limits function, begin with the least burdensome intervention, and judge benefit by sustainable function and recovery over the subsequent 24-72 hours, not by a transient improvement during the consultation.

Severe ME/CFS leaves only a narrow margin for clinical intervention. A patient may tolerate eating, toileting, or brief communication yet develop delayed deterioration after the assessment itself; approximately 25% of people with ME/CFS are largely housebound, and patients with severe disease are frequently excluded from inpatient protocols because of exertional intolerance and PEM risk ^[110]. Treat the patient’s sustainable baseline, not the performance observed during a brief visit, as the clinical reference point.

Adapt the clinical encounter

Arrange home visits when travel, transfer, waiting-room exposure, or transport would exceed the patient’s energy envelope. Before attending, ask the patient or caregiver about the best time of day, tolerable duration, positioning, lighting, sound, touch, temperature, communication method, and the activities that must be preserved afterward. A home assessment should be staged, with supine examination and observation preferred to standing or walking tests. Obtain only information that will change management; defer non-urgent examination, phlebotomy, and testing rather than combine them into one burdensome encounter. Home-based assessment is a feasibility strategy, not permission to reproduce a hospital work-up in the bedroom ^[110].

Use telemedicine for review of symptoms, medication administration, hydration and nutrition, bowel function, falls, pressure areas, caregiver strain, and safety. Offer telephone, text, email, captioning, speech-to-text, communication boards, eye-gaze or augmentative communication, and a trusted interpreter when speech, reading, screens, or auditory processing are limited. Send questions in advance and allow asynchronous replies. Keep encounters short, with one clinician speaking at a time, pauses, no requirement for video, and an agreed stop signal. A caregiver may report observations, but confirm the patient’s preferences and consent wherever possible; reduced speech or eye contact is not evidence of incapacity.

Reduce sensory load with dimmable indirect light, quiet surroundings, minimal staff, neutral clothing, ear protection only if tolerated, and avoidance of unnecessary touch or conversation. Record the patient’s preferred name, pronouns, communication method, positioning, and emergency contact in the care plan. Transport should be avoided when a home alternative exists; if unavoidable, use the shortest route, recumbent or reclined positioning when safe, prearranged access, no waiting, and recovery time after arrival. Stop an assessment for rising symptoms, cognitive slowing, pallor, presyncope, distress, or the patient’s request, and review for delayed PEM over the next 72 hours ^[110].

Prevent predictable complications

Assign a named clinician or nurse to coordinate primary care, nutrition, pharmacy, occupational therapy, physiotherapy, social care, and specialist review. At each contact document oral intake, fluid access, urine output, weight trend, swallowing, bowel movements, skin, transfers, falls, medication doses actually taken, sleep, orthostatic symptoms, infection symptoms, dental care, and caregiver capacity. Severe or very severe ME/CFS can require long-term care and artificial nutrition, and nutritional difficulty may arise because the effort of eating and drinking is itself intolerable, not because of food refusal ^[147]. Do not attribute weight loss, dysphagia, vomiting, dehydration, fever, pain, or functional decline automatically to ME/CFS; investigate a new or disproportionate change for a coexisting disorder.

Screen for malnutrition early with weight history, body-mass trend, intake, swallowing, gastrointestinal symptoms, dentition, and the patient’s ability to obtain, prepare, chew, and swallow food. Offer small, frequent, energy-dense foods and fluids in the least stimulating form tolerated; consider purée or liquid nutrition when chewing, odour, sound, or upright posture is limiting. Refer promptly to a dietitian and speech and language therapist for dysphagia or recurrent aspiration. If oral intake is inadequate, involve gastroenterology and nutrition support early to consider enteral feeding; if the gastrointestinal route cannot meet requirements, specialist parenteral nutrition may be appropriate after an individualized assessment of goals, risks, venous access, thrombosis, infection, metabolic complications, and monitoring. In severe malnutrition, check and replace phosphate, potassium, and magnesium and introduce calories cautiously because refeeding syndrome can cause arrhythmia, cardiac failure, delirium, seizures, or death, usually during the first five days of renewed feeding ^[147]. Do not label malnutrition as an eating disorder without a proper assessment of beliefs, intent, swallowing, motility, autonomic symptoms, and physical barriers.

Use an individualized positioning schedule, pressure-redistributing mattress or cushions, heel protection, moisture control, and daily skin inspection when mobility is markedly reduced. Reposition only within the patient’s tolerated range and use two-person or mechanical assistance when needed; a turn that produces PEM is not preventive care. Occupational therapy can prescribe transfer aids, bedside commodes, bathing adaptations, pressure-relieving equipment, communication aids, and environmental changes. Physiotherapy should be limited to tolerable passive or active-assisted range of motion, breathing comfort, contracture prevention, positioning, and safe transfers; do not prescribe progression toward exertion or fixed exercise quotas. Assess falls risk without repeated walking tests, address orthostatic symptoms and clutter, and use a wheelchair or hoist when it conserves energy and prevents injury.

Review venous thromboembolism risk individually, considering immobility, prior thrombosis, cancer, pregnancy or postpartum status, estrogen exposure, dehydration, infection, and central venous access. Encourage tolerated ankle movement and position changes only when they do not provoke PEM; do not use exercise as prophylaxis. Decide on pharmacological prophylaxis with the responsible clinician after balancing thrombosis risk against bleeding, falls, renal function, body weight, drug interactions, and patient goals. Assess constipation through stool frequency, consistency, pain, distension, vomiting, medicines, hydration, and intake. Use a bowel plan that may include fluids when safe, tolerable dietary fibre, a bedside commode, and prescribed laxatives; urgent assessment is required for severe pain, persistent vomiting, marked distension, or inability to pass stool or flatus.

Provide nursing support for personal care, continence, skin, nutrition, medicines, observations, and escalation plans. Arrange dental review through a home-capable or low-stimulation service when possible; use a small soft toothbrush, adapted handle, fluoride toothpaste, and caregiver assistance only with consent. Reduce infection risk through hand hygiene, vaccination according to current clinical advice, attention to dental and skin integrity, safe catheter and line care, and an individualized plan for early review of fever or localizing symptoms. Pharmacy should simplify regimens, synchronize dispensing, use blister packs or liquid formulations where suitable, label medicines clearly, and assess swallowing, dexterity, cognition, nausea, orthostasis, sedation, and whether a caregiver is administering doses. Never conceal medicines in food without a lawful best-interest process and appropriate consent.

Multidisciplinary and anticipatory care

Occupational therapy should target energy conservation and safety: place food, fluids, medicines, communication devices, call systems, and continence equipment within reach; arrange bathing, toileting, bed, and transfer adaptations; and reduce sensory and cognitive demand. Physiotherapy should preserve comfort, joint range, respiratory function, positioning, and contracture prevention within the patient’s current tolerance. A successful session is one followed by stable function, not one that achieves more repetitions. If symptoms or function worsen later that day or over the next 24-72 hours, reduce or stop the intervention ^[110].

Social care assessment should address personal care, meal preparation, cleaning, shopping, transport, equipment funding, housing, benefits, education or employment accommodations, and respite. Ask separately about caregiver sleep, physical strain, finances, isolation, and emotional health. Provide a written contingency plan for caregiver illness and a 24-hour contact route. Safeguarding is clinical care: severe disability can increase dependence and vulnerability, while exhausted caregivers may need support rather than blame.

Discuss advance care planning while the patient can participate, revisiting it after major changes. Record preferred place of care, acceptable and unacceptable investigations, emergency transport requirements, communication needs, nutrition and hydration preferences, medication administration, treatment-escalation limits, nominated decision-makers, and any advance decision or legally recognized directive. Assess decision-specific capacity; severe cognitive or communication impairment does not by itself establish incapacity. Include the patient in brief, low-burden decisions and use supported communication before relying on proxy reports.

Urgent assessment

Do not explain new acute symptoms as PEM without considering ordinary medical emergencies and coexisting disease. Arrange emergency assessment for chest pain, severe or new dyspnoea, hypoxia, cyanosis, syncope with injury or persistent abnormal observations, suspected pulmonary embolism, severe dehydration, significant bleeding, rapidly progressive infection, airway compromise, or acute confusion. New focal neurological deficits, unilateral weakness or sensory loss, seizure, severe sudden headache, or altered consciousness require urgent neurological assessment. Immediate psychiatric or emergency support is required for suicidal intent, a plan, imminent risk, inability to remain safe, or concern that a caregiver or patient may act on such intent. A marked change from baseline, persistent fever, sustained weight loss, recurrent vomiting, or progressive organ-specific symptoms warrants prompt medical review rather than an increase in pacing restrictions alone ^[147].

Children and adolescents require assessment against developmental expectations rather than adult work capacity. Ask what the child could do before illness and what is now sustainable across school attendance, travel between lessons, stairs, sport, homework, screen use, self-care, and social contact; then examine delayed worsening over the subsequent 24-72 hours. Early-onset disease is clinically consequential: in a large European survey, the early-onset peak was 16.0 years, and early onset was associated with greater odds of severe or very severe illness (OR 2.15, 95% CI 1.84-2.51). ^[21] A school absence may therefore represent PEM, orthostatic intolerance, cognitive overload, pain, sleep disruption, or an untreated comorbidity, not poor motivation or school refusal.

Obtain the young person’s account privately as well as the family history. Clarify who supervises medicines, food, fluids, toileting, mobility, and online access; ask about bullying, punishment for absence, pressure to exercise, missed medical care, family conflict, and fear of disclosure. Safeguarding assessment must remain proportionate: disability-related dependence is not evidence of neglect, while unexplained injury, coercion, unsafe restraint, food or fluid restriction, or a discrepancy between accounts warrants the usual child-protection process. Invite caregivers to contribute diaries, school records, and observations of delayed PEM, but do not let an adult answer every question for an adolescent. Family education should explain that a good hour does not establish sustainable capacity and that pushing through symptoms can produce a delayed crash. ^[119]

Coordinate a written education plan with the school, family, and, when appropriate, the child. Useful accommodations include reduced or flexible attendance, remote access, a quiet room, permission to recline, elevator access, shortened lessons, rest between cognitive tasks, reduced homework, alternatives to timed examinations and compulsory sport, extra examination time, and a single named school contact. Review accommodations against function rather than attendance alone. Do not prescribe coercive exercise-based rehabilitation or fixed increases in activity. NICE-aligned energy management is symptom-contingent; the adolescent may reduce, substitute, or stop an activity when PEM appears, and any physiotherapy or occupational therapy should be reversible and tied to safety, transfers, positioning, or participation rather than an exercise quota. A prospective pediatric study found that repeated handgrip testing was feasible, but it did not discriminate ME/CFS from other PEM-reporting illnesses; such tests are therefore research or adjunctive measures, not diagnostic requirements. ^[76]

Use developmentally appropriate communication and measures. Explain PEM in concrete terms, “what happens later today or tomorrow after a task?”, rather than asking only whether the child feels tired. Separate activity from participation: a young person may manage a brief conversation yet be unable to attend school, travel, or complete a full day. The MCFC Activity and Participation scales were developed for ages 10-25 years and showed different relationships to function, although they remain adjuncts rather than diagnostic tests. ^[162] Cognitive testing, questionnaires, and examination should be divided into short sessions or completed remotely when necessary. The Wood Mental Fatigue Inventory correlated strongly with the adolescent cognitive-fatigue subscale of the Pediatric Quality of Life Multidimensional Fatigue Scale, but neither instrument establishes ME/CFS. ^[112]

Pregnancy should be managed as a period of changing physiological demand, not as a predictable treatment for ME/CFS. Do not assume that pregnancy will improve or worsen symptoms; establish pre-pregnancy function, orthostatic tolerance, nutrition, sleep, migraine and pain pattern, medication exposure, and available support, then review these domains at each trimester and after delivery. Pregnancy may magnify the consequences of reduced upright tolerance, vomiting, dehydration, anaemia, sleep loss, and pelvic or musculoskeletal pain, while postpartum sleep deprivation, feeding demands, infection, and loss of practical support may reduce the energy envelope. These are reasons for anticipatory planning, not evidence that ME/CFS follows one uniform reproductive trajectory.

Before conception, reconcile all prescribed, over-the-counter, and supplementary treatments with an obstetric clinician or pharmacist. Use the fewest medicines necessary, choose agents with established pregnancy and lactation safety when treatment is required, avoid abrupt withdrawal of drugs on which the patient depends, and document the risk-benefit discussion. Review sedating, blood-pressure-lowering, serotonergic, anti-inflammatory, migraine, antihistamine, and autonomic drugs individually; do not extrapolate safety from adult ME/CFS practice or from a nonpregnant patient. Coordinate hydration, nutrition, venous-thromboembolism risk, mobility aids, antenatal appointments, mode of communication, and a postpartum plan with obstetric, midwifery, primary-care, and ME/CFS teams. Breastfeeding decisions should account for infant exposure, maternal sleep and hydration, feeding position, orthostatic safety, and whether another adult can assist; medication compatibility must be checked for each drug rather than inferred from the diagnosis. Flexible, accessible, and modular support is particularly relevant when symptoms fluctuate or clinic attendance itself provokes PEM. ^[119]

In older adults, do not attribute new functional loss to age or ME/CFS without reassessment. Establish premorbid mobility, cognition, continence, nutrition, vision, hearing, falls, polypharmacy, sleep, mood, cardiopulmonary reserve, and social support. Recheck for anaemia, thyroid disease, infection, malignancy, heart failure, arrhythmia, sleep-disordered breathing, neurological disease, and medication toxicity when the pattern changes; new focal findings, fever, bleeding, weight loss, or sustained deterioration require directed investigation. Start symptom medicines below usual doses when frailty, renal or hepatic impairment, falls, orthostasis, or cognitive vulnerability increases harm, and judge benefit by safe sustainable function rather than transient alertness. An apparent remission also requires functional confirmation: population survey data show that many people reporting past ME/CFS remained impaired, and some had symptom and function levels similar to those reporting current disease. ^[23]

Neurodevelopmental conditions, intellectual disability, acquired brain injury, sensory impairment, and physical disability can alter how PEM, pain, orthostatic symptoms, or cognitive difficulty are expressed. Use the person’s preferred communication system, concrete questions, visual scales, communication aids, caregiver observations, and changes from the individual’s own baseline. Do not require fluent speech, written questionnaires, independent travel, upright examination, or exercise performance to establish disability. Distinguish pre-existing motor or communication limitations from new delayed deterioration, and involve occupational therapy, speech and language therapy, physiotherapy, social care, and disability services when they improve access or safety. If a standard questionnaire has not been validated for the person’s communication or cognitive profile, treat its result as incomplete rather than falsely precise.

Limited health literacy, language discordance, poverty, unstable housing, migration, and cultural explanations of illness can all obstruct assessment and care. Use a trained interpreter rather than a child or family member, offer translated written plans, use teach-back, ask what the patient calls the illness and what treatments are acceptable, and allow extra time or remote review. A diagnosis should not depend on familiarity with the term PEM or on access to specialist vocabulary. Underrepresentation of Māori, Pacific, and Asian groups among New Zealand benefit recipients with recorded ME/CFS illustrates that service data can reflect access and ascertainment inequity rather than lower disease burden. ^[26]

Overlap is common, but overlap is not diagnostic substitution. Long COVID requires a compatible history of probable or confirmed SARS-CoV-2 infection and may produce an ME/CFS phenotype; ME/CFS does not require COVID-19. Persistent post-COVID illness and ME/CFS can show broadly overlapping autonomic, sensory, cognitive, and fatigue profiles, but available longitudinal data do not establish identical trajectories or convergence. ^[41] Record both diagnoses when both are justified, and retain diagnostic discipline by documenting delayed PEM, unrefreshing sleep, functional reduction, and cognitive or orthostatic symptoms rather than using “long COVID” as a catch-all.

Fibromyalgia and migraine may account for widespread pain, sensory sensitivity, or headache, but they do not by themselves explain delayed PEM. POTS or orthostatic hypotension may amplify dizziness, tachycardia, nausea, and cognitive symptoms, but an abnormal orthostatic test does not establish ME/CFS. Hypermobility spectrum disorder warrants assessment of joint instability, injury, pain, proprioception, and autonomic symptoms; avoid assuming that hypermobility explains every systemic complaint. Mast-cell-related symptoms such as flushing, pruritus, urticaria, wheeze, abdominal symptoms, or episodic hypotension should be assessed on their own merits, with treatment directed at documented manifestations rather than unvalidated panels. Autoimmune disease requires attention to objective inflammatory, organ-specific, or serological findings; a positive antibody without a compatible syndrome should not replace clinical reasoning. Chronic overlapping pain conditions are associated with greater pain, fatigue, sleep disturbance, disability, and healthcare use, supporting coordinated rather than one-disease/one-drug care. ^[143]

Across every context, document the patient’s sustainable baseline, the delayed response to care, and the practical support required. A plan that improves attendance for one day but produces a crash over the next 72 hours is not rehabilitation; it is excessive demand. ^[128]

The course of ME/CFS is heterogeneous. Some patients improve substantially or enter sustained remission; others remain persistently impaired, follow a relapsing-remitting pattern, or deteriorate gradually into severe or very severe disease. A 2026 study protocol summarizes the available literature as showing that fewer than 10% of cases return to pre-illness functioning without treatment, but this estimate should not be treated as a population prognosis: cohorts differ in referral severity, follow-up duration, treatment access, and case definition, and the protocol itself emphasizes that prognosis varies widely ^[168]. Broader Fukuda-based cohorts may include patients without obligatory post-exertional malaise , whereas more stringent criteria select a different clinical phenotype; outcome estimates therefore cannot be transferred uncritically between studies ^[168].

Interpret apparent improvement against sustainable function, not a single good day or a lower symptom score. A patient may increase activity briefly and then experience delayed functional loss, or may report fewer symptoms while remaining unable to work, study, prepare meals, travel, or tolerate upright activity. Conversely, a reduction in work or social participation may reflect new comorbidity, loss of accommodation, or caregiver exhaustion rather than biological progression. Record the patient’s baseline, the activity that remains reliably tolerable, the delay and duration of PEM, recovery time, and the help required for activities of daily living.

Studies associate outcome, inconsistently, with illness duration, onset pattern, initial severity, comorbidity, age, and access to appropriate care. These are probabilistic associations, not individual predictions. Early-onset illness may be more disabling in some pediatric cohorts, while long illness duration and severe baseline functional loss may reduce the opportunity for recovery; neither finding justifies therapeutic pessimism. Depression history and multimorbidity are associated with current ME/CFS in cross-sectional survey data, but the temporal direction is unresolved, so these findings cannot establish that depression or comorbidity causes ME/CFS or determines its course ^[169].

Review patients according to the burden and volatility of illness. Mild or stable disease may permit review every 3-6 months; moderate disease, recent deterioration, medication changes, nutritional risk, or recurrent orthostatic events usually warrants review within 4-12 weeks. Severe or very severe disease requires a home-based, telemedicine, or caregiver-assisted plan at an interval dictated by nutrition, hydration, medication safety, pressure-injury risk, falls, swallowing, and caregiver capacity. Use a shorter interval after any intervention that could alter alertness, blood pressure, sleep, intake, or PEM. Do not require travel, prolonged sitting, standing, exercise, or repeated testing to demonstrate deterioration.

Mental-health assessment must neither psychologize the illness nor omit suicide prevention. Ask directly and privately when possible; a negative answer does not end assessment if there is marked hopelessness, isolation, uncontrolled pain, medication access, recent bereavement, or a sudden loss of function. Treat depression, anxiety, trauma-related symptoms, and suicidality on their own merits while preserving the patient’s limited cognitive, upright, and sensory capacity. Persistent disbelief and stigmatization are recognized burdens for people with ME/CFS, so validation is a clinical intervention: state that the illness and disability are real, describe what remains uncertain, and explain why new or changing findings still require medical review ^[170].

Mortality data should be interpreted cautiously. Available cohort and survey designs do not provide a dependable individual mortality estimate, and a fluctuating course does not guarantee safety. Ask about falls, dehydration, malnutrition, medication mismanagement, untreated comorbidity, and suicidal thinking; document advance preferences and the acceptable burden of investigations for patients who are severe or very severe. Caregiver strain is a clinical outcome, not a peripheral social issue: ask whether care is sustainable, whether the caregiver can leave the patient safely, and whether respite or additional nursing and social support is required.

Disability certification should describe observable, reproducible function rather than rely on the label or a brief examination. Document sustainable hours of activity, upright tolerance, recovery after exertion, frequency of PEM, cognitive and sensory limits, self-care, transfers, communication, attendance reliability, and the assistance or accommodations required. A patient’s ability to complete a single appointment, test, or household task does not establish capacity for repeated activity across a working week. The same functional record supports benefits applications, workplace or school accommodations, equipment provision, and safe discharge planning.

Research should move from small cross-sectional case-control comparisons toward harmonized, longitudinal phenotyping. Priority areas are validated biomarkers that distinguish ME/CFS from mimics and identify mechanistic subgroups; agreed case definitions with transparent reporting of PEM, duration, severity, and comorbidity; and repeated measurements that capture delayed post-exertional trajectories rather than only resting physiology. Current biomarker literature remains heterogeneous, with variation in analytical methods and frequent study of only selected molecular pathways; larger cohorts and longitudinal designs are required to validate candidate markers ^[174].

Mechanistic studies should link immune, metabolic, vascular, autonomic, sleep, gastrointestinal, and central nervous system measures within the same participants and test temporal direction rather than infer causality from association. They should include children and adolescents, people with severe and very severe disease, underrepresented communities, and patients who cannot travel to research centers. Patient and caregiver-reported outcomes should be paired with low-burden physiological measures, but no candidate test should be adopted as a clinical biomarker until it replicates across independent cohorts and case definitions.

Trials need pragmatic designs, meaningful follow-up, transparent harms reporting, and outcomes that include PEM, recovery time, sustainable function, participation, sleep, orthostatic symptoms, nutrition, and quality of life. Interventions must be tested in the severity ranges in which they will be used, with remote or home-based options for those excluded by travel and exertion demands. Treatments should not require fixed activity increments or provoke PEM. A placebo-controlled trial of intermittent hypoxia-hyperoxia illustrates the desired direction by incorporating PEM, fatigue, cognition, function, autonomic symptoms, mechanistic measures, and follow-up to 12 months, although its results remain unavailable ^[172].

Pearl: Prognosis is individualized and often uncertain. Validate the patient’s lived disability, measure recovery and sustainable function over time, investigate meaningful change rather than chase biomarkers, and treat every new red flag as a reason to reconsider the diagnosis or identify a coexisting disorder.