ME/CFS & post-viral illness — clinical management

What landed in the literature, and what it changes at the bedside

A screened digest of new ME/CFS and post-exertional-malaise research, weighted toward diagnosis, management, and prognosis rather than bench mechanism.

Edition
15 Nov 2026
Window
7 days to 15 Nov
Screened
9
Flagged
4

Nine new records, four carried, and the week's two trials both missed. A CDC-funded cluster-randomised trial in a San Diego safety-net clinic gave half its primary-care clinicians weekly case conferences with ME/CFS and long COVID experts; on its own primary outcome, physical function, it found no difference, and the abstract does not say so. Its one positive result, on pain interference, is smaller than the threshold the authors set for a meaningful change. A Basque pilot trial of home cognitive and motor training also found no between-group difference. The item she is most likely to be asked about is a Stanford chart review of low-dose aripiprazole in 50 long COVID patients: their function score improved, but none of their symptoms did, PEM included, there was no control group, and most were also taking low-dose naltrexone. In the pipeline, Harvard and the Open Medicine Foundation have published the protocol of a 160-person factorial trial of low-dose naltrexone and pyridostigmine. Tier 1 is empty for the fifth edition running.

Changes what you'd do

Findings with a direct line to a clinical decision — a workup step, a referral, a conversation before surgery.

Nothing this week, and two trials that found nothing are not a reason to change anything either

Tier 1 empty·9 records screened, 0 promoted·fifth consecutive empty edition

The two randomised trials in the window are both null on their main comparisons, and both are about how care is delivered rather than about a drug: one trained primary-care clinicians, the other trained patients' attention and memory at home. A null result can belong in this tier when it removes something from practice, but neither trial tested anything she is likely to be doing, and the larger one tested it in a population that was 91% long COVID with only 49 of 587 patients carrying an ME/CFS diagnosis. The aripiprazole paper is uncontrolled. All three are in Tier 2.

The 160-person low-dose naltrexone trial (NCT05430152), which this digest committed on 2 November to promoting here on peer review, has still not entered PubMed. Its bottom line stands as stated then: no benefit over placebo at up to 4.5 mg daily for sixteen weeks, and well tolerated.

Bottom lineNothing new to act on. The practical value this week is defensive: two items patients may bring in — a low-dose antipsychotic “that helped at Stanford”, and brain-training at home — now have evidence she can describe accurately, and in both cases it is weaker than the headline.

Worth knowing

Shapes how you counsel, set expectations, or answer the question a patient brings in from the internet.

Effectiveness of an Extension for Community Health Outcomes case-based virtual community of practice model on patient outcomes for the management of long COVID and other fatiguing illnesses: a randomized control trial

Front Health Serv·Cluster RCT, 20 clinicians, 587 patients·PMID 43016714·NCT05167227·open access·full text and registry record read; supplementary tables not opened

Fourteen authors from Family Health Centers of San Diego, San Diego State, the University of Washington, the University of Colorado, Ata Health Strategies, the ECHO Institute and the CDC — the last including Elizabeth Unger and Jeanne Bertolli — randomised 20 primary-care clinicians at a large federally qualified health centre to one of two levels of education in long COVID and ME/CFS. Both arms were offered monthly webinars and quarterly online courses. The intervention arm was also offered weekly Project ECHO case conferences, each with at least two national experts and a person with lived experience. Each clinician was assigned about 30 patients, 587 of whom completed a baseline survey (298 control, 289 intervention), and patients were followed by telephone survey for a year. The population was low-income (94.4% under $50,000 a year), 40% preferred Spanish, and 91.3% had long COVID alone; 36 had long COVID with ME/CFS and 13 ME/CFS alone. Two patients with severe ME/CFS could not complete the telephone survey at all.

What they report. The primary outcome, stated in the methods and in the sample-size calculation, was change in PROMIS-29 physical function. The treatment effect was 1.49 T-score points (95% CI −0.04 to 3.02; p = 0.06): no significant difference. Of the nine PROMIS scales modelled, only pain interference separated, by −1.95 points (−3.79 to −0.10; p = 0.04), and at 12 months the arms scored 57 against 60 from a shared baseline of 61. Among patients who completed both baseline and 12-month surveys, 43% of intervention patients (63 of 145) against 29% of controls (50 of 174) improved by at least 5 points on pain interference. Both arms improved on fatigue, physical function, social roles, dyspnoea and cognition over the year. Intervention patients received fewer referrals (median 9 against 10) and completed fewer encounters (16 against 21). No harms were reported to the review board.

Bottom lineThe only randomised evidence so far that training primary-care clinicians in this field helps patients, and it mostly shows that it did not change function. Report the primary outcome first: case conferences on top of webinars did not improve physical function. The pain signal is plausible and modest: about 1 in 7 more patients reached a meaningful improvement in how much pain interfered with life, which fits the idea that clinicians who learn from colleagues treat pain more actively. It comes from one of nine secondary scales with no correction for multiple comparisons, and the average effect is under half the authors' own 5-point threshold. Both arms improved substantially, which reflects natural recovery from long COVID in the first years after infection as much as anything either arm did, and is a reason not to read before-and-after improvement in an uncontrolled long COVID clinic report as a treatment effect. For her, the practical points are two. Peer case discussion is cheap and may help with pain. And this trial tells her almost nothing about ME/CFS specifically: 49 patients, with no analysis by diagnosis.

Weigh it accordingly. The abstract omits the primary outcome. It reports that both arms improved, then the pain-interference result, and concludes that the model “showed evidence” of benefit; the null physical-function result appears only in Table 3. The registry does not settle which outcome was primary: NCT05167227 lists the whole “PROMIS-29”, with no subscale, as its primary outcome, so the paper's choice of physical function cannot be checked against the registration, though it is consistent with the published protocol's sample-size reasoning as described. The registry also lists, as a secondary outcome, whether symptoms worsen for at least 24 hours after activity — a PEM question — which is not reported in the main text (it may be in the symptom-screener tables in the supplement, which this digest did not open). The discussion overstates its own results. It says the intervention arm “reported meaningful improvement in pain interference” when the results section says the average effect “might not be clinically meaningful”, and it describes the 43% and 29% responder proportions as though they were the two arms' scores. Fewer referrals and encounters are presented as a health-system benefit; they could equally mean patients received less care, and the paper has no way to tell the two apart. Retention was poor and unequal: 12-month surveys were completed by 50.5% of the intervention arm and 60.4% of controls (146 of 289; 180 of 298), and the authors say they may have lost both the sickest and the recovered. Patients were assigned to clinicians by clinic and language rather than at random; clinicians attended 42% of the case conferences offered; the control arm received active education, which the authors note may have shrunk the difference; and long COVID was defined as symptoms at four weeks, the definition current in 2021. The comparison with the University of Washington clinic is not randomised and its finding that San Diego patients did better says nothing about the intervention. Two small inconsistencies: the abstract says 590 patients “completed surveys” where 587 did; and Table 2 gives 180 and 146 completers at 12 months where the responder analysis uses 174 and 145. The registry's start date, 30 November 2021, precedes its first submission on 17 December; the paper says recruitment began in December. Interests: funded by the CDC, with CDC-affiliated authors, evaluating a programme the CDC paid for; one author was employed by a consultancy in the consortium. ChatGPT was used for spelling and grammar, declared. Arithmetic that checks: 298 + 289 = 587; 536, 36, 13 and 2 sum to 587; 63/145 = 43.4%; 50/174 = 28.7%; both treatment-effect intervals are symmetric about their estimates (1.49 ± 1.53; −1.95 −1.84/+1.85). Published 28 October, entered PubMed 12 November.

Low-dose aripiprazole for long COVID symptom management: A retrospective observational study

Brain Behav Immun Health·Retrospective chart review, single centre, no control group, n = 50·PMID 43032818·open access·full text read; supplementary tables not opened

Seventeen authors, most at the Stanford long COVID clinic, with Grace McComsey of Case Western, reviewed the records of the 76 of 1,134 clinic patients prescribed aripiprazole at 0.1–2 mg daily between May 2021 and September 2023 for long COVID symptoms. 50 had usable before-and-after questionnaires and form the analysis; anyone diagnosed with ME/CFS before their COVID infection was excluded. Follow-up was whenever the patient was next seen. Outcomes were a 29-symptom checklist the clinic designed itself, rated 0–5, and a five-level functional status scale adapted from the Post-COVID-19 Functional Status scale.

What they report. Functional level improved from a mean of 3.78 to 3.28 (p = 0.001): before treatment 8 patients were at level V (bedbound) and 23 at IV; afterwards 3 and 15, with 7 reaching level II. No symptom improved significantly — not fatigue, brain fog, unrefreshing sleep, headache, lightheadedness, nor post-exertional malaise (3.68 to 3.44, p = 0.34) — and neither did the symptom count or the composite score; 27 patients' composite improved, 22 worsened and 1 was unchanged. The median final dose was 1 mg; 18 reached 2 mg. Twelve of the 76 (15.8%) reported side effects. Weight changed in most: 25 gained up to 42 pounds and 18 lost up to 71. Most patients were also taking low-dose naltrexone. In a subgroup of 33 with two visits before and after, the authors report that the improvement in function persisted after allowing for the pre-treatment trend.

Bottom lineIf a patient asks about low-dose aripiprazole: the only data in long COVID are this chart review, in which patients said they were functioning a little better while reporting the same symptoms, with no comparison group, during the period when long COVID tends to improve on its own, and with most also on naltrexone. The authors themselves call for a randomised trial before it is used. Its PEM item did not move. Weight change and the stigma of an antipsychotic label are real costs, and one patient in six reported a side effect. The rationale was an earlier ME/CFS case series from some of the same clinicians, and that was uncontrolled too. It is reasonable to tell a patient that it is being tried, that it is not known to work, and that a trial is the right place to try it.

Weigh it accordingly. Single-centre, retrospective and uncontrolled, as the authors state, adding selection bias, concomitant treatments, uncontrolled follow-up duration, multiple exploratory comparisons and placebo effects to their own list of limitations. A third of those treated are missing (26 of 76) for incomplete records. Both outcome instruments are unvalidated: the methods say the functional scale was modified and “has not been independently validate[d]”, while the discussion calls it “a validated 5-point ordinal instrument”. Internal inconsistencies: the abstract describes symptom ratings from 1 to 5 where the body uses 0 to 5; the composite score is described as an average of seven symptoms divided by 7, but the table values (20.36 and 18.62) are exactly the sums of the seven symptom means; duration on treatment is given in consecutive sentences as an average of 315 days (range 31–1165) and a mean of 204.4 days (range 31–1025); the weight percentages (48.1%, 34.6%, 3.8%) and the 53.8% concomitant naltrexone figure correspond to a denominator of 52, not 50; and 43 of 50 on naltrexone is 86.0%, not the 86.6% printed. PubMed indexes the paper as a review, which it is not. Arithmetic that checks: the functional-level distributions sum to 50 before and after and give the stated means exactly (189/50 and 164/50); 27 + 22 + 1 = 50; 12/76 = 15.8%. No funding statement was found in the text read; the authors declare no conflicts. Ethics: Stanford IRB 62996, consent waived. Received 5 April, accepted 27 September, online 6 October, entered PubMed 14 November.

Personalized cognitive and physical Telerehabilitation in post-COVID-19 condition and ME/CFS: A pilot randomized controlled trial

Acta Psychol (Amst)·Pilot RCT, 79 analysed (46 vs 33)·PMID 43025223·abstract and declarations only; full text not read

Read the meta line first. The paper is on ScienceDirect, which has served this digest bot checks on earlier runs, and has no PubMed Central copy; the full text was not attempted, so everything here comes from the MEDLINE abstract. Eighteen authors at Biobizkaia, Cruces, Donostia and Araba university hospitals in the Basque Country randomised people with post-COVID condition or ME/CFS to individualised home cognitive and motor training four times a week, adjusted weekly, or to standard care, for 16 weeks. 79 had both assessments: 46 intervention, 33 control. Outcomes were neuropsychological tests, self-reported symptoms and the six-minute walk.

What they report. No between-group difference in any comparison and no domain-level time-by-group interaction after false-discovery-rate correction. Within the intervention arm alone, sustained attention and semantic verbal fluency improved (both corrected p = 0.028). An exploratory component explaining 27.2% of cognitive change was higher in the intervention arm (β = 0.715, p = 0.043, uncorrected) and tracked motor adherence. Median recorded adherence was 99.75% for motor and 100% for cognitive tasks; median usability score 77.5. The authors' conclusion is restrained: feasibility is supported, differential benefit was not established.

Bottom lineAnother null for cognitive training in post-viral illness, worth having when a patient asks whether a brain-training programme will help her brain fog: in this randomised comparison, which included ME/CFS, it did no better than standard care. Within-arm improvement in an untreated-control design is what practice effects on repeated testing look like. The abstract says nothing about whether training four times a week provoked post-exertional symptoms, although the background names them, and that is the question she would most want answered before suggesting it.

Weigh it accordingly. Abstract only. The number randomised, the case definitions, how many had ME/CFS rather than post-COVID condition, severity, dropouts and whether any adverse events were recorded are all unknown. 46 against 33 analysed suggests either unequal allocation or more loss from the control arm; the abstract does not say which. Adherence of 99.75–100% is platform-recorded and implausibly near-perfect for a fatiguing illness; the authors note that the clinical support behind it was not recorded. The one positive between-group signal is an uncorrected exploratory component. No funder appears in the record; the authors report no conflicts. Published and entered PubMed 13 November.

PEM and pain

A standing section, carried every edition regardless of evidence grade, with what each item actually rests on stated up front. This week PEM was rated in one study, as a single unvalidated item that did not change, and is planned as a secondary outcome in one protocol. Pain carried the only positive between-group result of the week, in a trial whose primary outcome was null.

PEM: named in four abstracts, measured properly in none

Standing check·4 of 9 PubMed abstracts mention PEM; 1 rated it; 0 used a validated PEM instrument as an outcome

The counts. PEM appears in four of nine abstracts. The aripiprazole review rated it as one 0–5 item of the clinic's own unvalidated checklist: 3.68 before, 3.44 after, p = 0.34. The LIFT protocol plans the DePaul PEM questionnaire as a secondary outcome. The telerehabilitation pilot names post-exertional symptom exacerbation in its background and, in its abstract, does not measure it. The excluded Bulgarian T-cell study lists a “dePol” symptom score among its measures, plausibly a DePaul questionnaire, used as a correlate. The ECHO trial's registration includes a 24-hour post-activity worsening question as a secondary outcome that its main text does not report. On medRxiv the post-exertional search returned only a revised posting of the Manchester PASS scale, covered last week.

A correction to last week, from the PASS revision. The 8 November edition said the PASS dropped its item on delayed PEM onset. The full text of the revised posting says the item was removed from the score but kept in the questionnaire, unscored, because delayed onset “is an important clinical issue, particularly when it comes to activity management”. The scale also avoids the words “post-exertional malaise” altogether, asking instead about “a worsening of symptoms when you overdo it”, because patients and clinicians found “exertion” and “malaise” hard to explain and the latter “redolent of malingering”.

Bottom lineThe practice point is unchanged: record PEM with a named instrument and a dated note of the exertion and what followed. The PASS wording is worth borrowing in clinic: “what happens to your symptoms when you overdo it, and how long until it shows?” gets a better history than asking about “post-exertional malaise”, and keeps the timing question the scale itself could not score.

Pain: the one result that separated, in a trial that otherwise did not

Standing check·2 of 9 records measured pain with a validated instrument (1 completed, 1 planned)

The ECHO trial's pain-interference result is in its entry above: a mean difference of 1.95 PROMIS points, with 43% against 29% reaching a 5-point improvement. Its CDC symptom screener points the same way — muscle-pain severity at 12 months of 5.8 against 7.8 (p = 0.01), joint pain 6.1 against 7.6 (p = 0.06) — which the authors present as independent support, though it comes from the same patients and the same surveys. LIFT plans PROMIS-29 pain interference and intensity as secondary outcomes. The microbiome study and the JAK review, both excluded, mention pain without measuring it.

Bottom lineA modest, plausible signal that clinicians who discuss cases with experts manage pain better, and the first controlled pain result this digest has carried in some weeks; it is still a secondary outcome without multiplicity correction, in mostly long COVID patients. The standing pain items from previous editions — the naltrexone trial's null on pain, the Leeds and Brussels findings, and the endometriosis prompt — are unchanged and not repeated.

In the pipeline

Registered trials and protocols in view this window — nothing to act on, but these are the results that will arrive next. For the first time in five weeks a PubMed record is itself a trial protocol.

Pyridostigmine and low-dose naltrexone for ME/CFS: study protocol for the Life Improvement Trial (LIFT), a randomized, double-blind, placebo-controlled clinical trial

Trials·Protocol, 2×2 factorial RCT, target n = 160·PMID 43000413·NCT06366724·funded by the Open Medicine Foundation·full text and registry record read

David Systrom's group at Brigham and Women's, with Massachusetts General and Uppsala, is randomising adults aged 18–65 who meet both the National Academy of Medicine and Canadian Consensus criteria, show orthostatic intolerance, and fell ill before December 2019 — so no post-COVID cases — to low-dose naltrexone (4.5 mg daily), pyridostigmine (60 mg three times daily), both, or neither, for 13 weeks after a four-week titration. There are four primary outcomes: the FUNCAP-55 functional-capacity questionnaire and three measures from a submaximal step test (the SHAPE test): extrapolated peak oxygen uptake, oxygen-uptake efficiency slope and one-minute heart-rate recovery. The DePaul PEM questionnaire, PROMIS-29, a reaction-time app and Garmin step count and heart-rate variability are secondary. Each active arm is compared with double placebo, Bonferroni-corrected across 12 primary tests, which the authors calculate needs an effect size of about d = 0.85 for 80% power.

Bottom lineThis is the trial to watch for pyridostigmine, which is used off-label for exercise intolerance on the strength of invasive-CPET physiology from this same group, and the second placebo-controlled test of 4.5 mg naltrexone after last month's null. Two features make the result easier to use than most: FUNCAP is the functional measure last week's Dutch review recommended, and the criteria demand a defined, pre-COVID ME/CFS population. Two make it harder: the step test requires travel to Boston, so housebound and bedbound patients are excluded, as the authors say; and at d = 0.85 only a large effect will register, so a null will not rule out a modest one.

Weigh it accordingly. Timing: the protocol (version 4, June 2024) says recruitment began in August 2024 and was planned to finish in May 2026, but the registry, updated October 2026, still shows the trial recruiting with primary completion in September 2027, so it is running well behind. The protocol was published in August 2026, after its own planned end of recruitment. Terminology: the abstract calls one primary outcome “peak oxygen utilization”; the registry describes it as extrapolated maximal uptake from a submaximal test, which is not the same measurement. FUNCAP is timed at 15 weeks in the registry and the physiological outcomes at 13. Harm: the adverse-event section read does not name a post-exertional crash as an event to record, and the step test is itself an exertion. Treatment-experienced patients may be enrolled if treatment-na√Øve ones run short, which the authors acknowledge could unblind them. Interests, which are substantial and declared: funded by an Open Medicine Foundation grant; OMF's founder and chief executive is an author, a second author is an OMF employee, and the senior author, the Uppsala collaborator and the computational lead each direct an OMF-funded centre, one also serving as its chief medical officer. Recruitment is partly through OMF's own registry. Ethics: Mass General Brigham 2024P000087, registered 16 April 2024, before the first patient.

  • A randomised trial of low-dose aripiprazole in long COVID, Stanford. Called for by this week's chart review; not registered anywhere in view. What will make it easy to apply: the clinic already has a titration protocol and a large cohort. What will make it hard: the only outcome that moved in the chart review was an adapted, unvalidated functional scale, so the trial will need a validated function measure and a placebo arm long enough to outlast natural improvement.
  • An adequately powered trial of personalised telerehabilitation, Basque Country. Stated as the next step by the pilot's authors; no registration in the abstract. What will make it hard: the pilot found no between-group effect to power it on, and will need to count post-exertional worsening as a harm.
  • Still standing from previous editions. NCT05430152, the low-dose naltrexone trial: preprinted 26 October, still not in PubMed, with its registered responder analysis unreported (2 Nov). Psychometric validation of the Vienna PEM questionnaire (8 Nov). The Manchester PASS, SoMES and MEOSS scales, now revised once on medRxiv (8 Nov). Chart validation of hEDS coding in the US national COVID cohort, requested and not done (8 Nov). A sham-controlled trial of hyperbaric oxygen in ME/CFS, still not registered (2 Nov). The exertion-paired EEG programme, unregistered (2 Nov). Enterovirus persistence in tissue rather than blood, Cornell (2 Nov). VITA Monitoring, the FIOCRUZ post-COVID cohort (26 Oct). Reservoir-directed and biologically stratified long COVID trials as the ERJ guideline revision frames them (26 Oct). The Alberta vaccination-as-immune-perturbation proposal, and rs6971 genotyping for TSPO imaging (26 Oct). The rTMS RCT in PASC (ZonMw 11080022420018), still without a registration identifier (19 Oct). The COMET core outcome set for CFS/ME (study 2103), unreported (19 Oct). “ME/CFS Kids Bavaria”, and Germany's National Decade against Post-Infectious Diseases (19 Oct). NCT05498493, Mount Sinai's cognitive rehabilitation trial in long COVID (12 Oct). APAV-ME/CFS parts 6 and 7. The COVID-CNS connectivity secondary analysis (1 Oct). The McMaster qualitative protocol on mind–body approaches (25 Sep). ChiCTR2400082268, the Shanghai Tai Chi trial's unreported outcomes (28 Sep). The Munich adolescent stand-test replication (21 Sep). NCT04740736, Mount Sinai's two-day-CPET saline-versus-sham infusion study, completed July 2025 and unreported (14 Sep). RECOVER-VITAL's viral-persistence analyses (7 Sep). The two Tuina protocols for CFS (31 Aug; 7 Sep). REenergizeME, ADDRESS-LC, the ME/CFS rehabilitation-subgroup study and the auricular vagal neuromodulation trial in POTS (16 Aug). The decentralised tirzepatide phase 2 in long COVID fatigue (24 Aug).

Preprint watch

medRxiv, 8 – 15 Nov. Four searches returned four distinct postings, one of them a revision of a preprint covered last week; one new posting is carried, the revision is noted, and one was set aside. None is peer reviewed. Industry funding: none.

Effect of starting metformin on severe and long COVID-19 among people living with type 2 diabetes in England: a new user cohort study in OpenSAFELY-TPP

medRxiv preprint·Not peer reviewed·Target-trial-style new-user cohort, n = 52,341·posted 10 Nov 2026·full text and declarations read from the source XML

Bristol and Basel epidemiologists, with the OpenSAFELY collaborative, took every English adult newly diagnosed with type 2 diabetes between August 2018 and August 2019 and compared those who started metformin within six months (19,599, 37.4%) with those who started no diabetes drug — with follow-up beginning before the pandemic, so that nobody's treatment could depend on their COVID course. Starting metformin was associated with slightly more COVID-19 hospitalisation or death (HR 1.17, 95% CI 1.02–1.35; per-protocol 1.11, 0.96–1.29). Long COVID: 33 cases (0.17%) against 33 (0.10%), HR 1.39 (0.83–2.33), imprecise. The negative control (fracture, HR 1.00) was null but the positive control (diabetes-related death, HR 1.41) was raised, which the authors read as residual confounding: people started on metformin were sicker in ways the records do not capture.

Bottom lineUseful when a patient asks whether her diabetes drug protects against long COVID: metformin taken long-term before infection showed no sign of doing so here. It does not contradict the randomised evidence patients may have read about, which tested metformin started within days of a COVID diagnosis and found less long COVID afterwards; the authors make that distinction themselves. The question this cannot answer is whether metformin helps someone who already has long COVID.

Weigh it accordingly. Preprint, not peer reviewed; medRxiv's notice applies. Long COVID here means a diagnosis code, and 66 codes among 52,338 people over two years is a severe undercount, as the authors acknowledge, so that estimate can neither show benefit nor harm. Metformin exposure is prescribing, not dispensing or taking. The design is strong for its question and the authors are candid that unmeasured confounding remains. Arithmetic: 33/19,599 = 0.17%, 33/32,739 = 0.10% and 19,599/52,341 = 37.4% check; the two groups sum to 52,338, three short of the 52,341 eligible, unexplained in the text read. Interests: none declared; OpenSAFELY is funded by NHS England, the Wellcome Trust, the MRC and NIHR. Ethics: Health Research Authority REC 22/PR/0095 and University of Bristol 117269; protocol and amendments public on GitHub; reported to RECORD-PE.

  • Four searches, four distinct postings, one carried. The myalgic encephalomyelitis and post-exertional searches each returned the Manchester PASS preprint, as version 1 (posted 8 November, covered last week) and a revised version 2 posted 10 November. The abstract is unchanged. Its full text, read this week from the source XML, adds that participants were UK adults with a self-reported professional diagnosis recruited through support groups, the 25% ME Group and the ME Association's channels, that diagnoses were not checked against records, and that the recall period is the last month; the delay correction is in the PEM section above. It does not resolve last week's question about the minimal detectable difference: the sample size sits in Table 1 and the formula in an equation, both images not present in the text, and the text says only that the MDD was derived from the standard error of measurement on a Rasch-transformed 0–100 scale. Last week's caution about quoting the 1.93 figure stands. The post-exertional search has now returned only this one posting in seven windows. The long COVID phrase returned two: the metformin cohort above, and an Australian study of herpesvirus-specific immunity (Kirby Institute, UNSW) profiling people at 4, 8 and 24 months after infection and reporting lower early EBV antibody avidity and smaller virus-specific CD4 responses in long COVID, with no viral DNA in plasma or stool, and lower EBV avidity associated with poorer quality of life. It is set aside as mechanism, and because medRxiv served a challenge page to this digest's request for its full text and declarations, which was not pursued; it is noted because “EBV reactivation” is a common patient question and its finding, no detectable virus, is worth knowing. The post-COVID search returned nothing.

What was screened out, and why

Five of the nine new PubMed records didn't make the cut — a 56% exclusion rate, near last week's 57%. The grounds, so the filter stays auditable:

  • Self-reported diagnosis, no clinical outcome — Complementary Health Approach Use Among Adults Reporting a Clinician Diagnosis of Chronic Fatigue Syndrome in a US Online Quota Sample (J Clin Med 2026). Six authors at Campbell University's osteopathic medical school reanalysed a 2024 online quota survey of 3,022 US adults, 167 of whom (5.5%) reported a clinician diagnosis of CFS; they reported more use of physical, nutritional and “systems” approaches (adjusted odds ratios about 2), with any complementary use 98.2% against 94.6% (aOR 2.46, 0.97–8.91). Out because no diagnosis was verified and no ME/CFS criteria applied, and because the authors themselves show that adjusting for the number of other diagnoses removed the category associations. Their abstract is unusually honest that the findings “do not establish CFS-specific treatment use, substitution, or effectiveness”. One author practises acupuncture, one of the modalities assessed, declared. The practical point survives: nearly everyone in the sample used something, so asking about it is routine history-taking, not a special question.
  • Biomarker discovery, no clinical use — Treg and Oxidative Stress Biomarker Profiles in Long COVID and Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (Int J Mol Sci 2026). Sofia and Plovdiv authors, with Michael Maes, profiled T-cell subsets and serum oxidative-stress markers in 89 long COVID, 125 ME/CFS and 80 controls and report a five-marker signature separating ME/CFS from controls (AUC 0.907–0.924) but long COVID barely (0.577–0.727). Out on design: a single-centre discovery study, case definitions not in the abstract, markers not associated with symptom severity, and no validation cohort; AUCs from the derivation sample overstate performance. Funded by an EU NextGenerationEU programme; no conflicts declared.
  • Pilot, null after adjustment — Gut microbiota dysbiosis in Post-COVID condition symptoms: relation with chronic fatigue and functional limitation (BMC Microbiol 2026). 30 post-COVID patients and 20 controls in Aragon; no difference in diversity, and the taxon-level differences (Escherichia/Shigella in the ME/CFS-like subgroup) all lost significance after adjusting for digestive symptoms and vaccine doses, as the authors report. One author is affiliated with a bioinformatics company, declared.
  • Wrong population — Neuro-immune and metabolic disorders in association with depression, anxiety, and chronic fatigue-fibromyalgia symptoms due to nonalcoholic fatty liver disease (CNS Spectr 2026). Iranian and Iraqi authors, with Maes, studied 108 patients with fatty liver disease and 60 controls; “fibro-fatigue” is a rating-scale score, not ME/CFS, and the record matched on the MeSH heading alone.
  • Narrative review, no data — The role of the JAK/STAT pathway in the pathogenesis and treatment of post-COVID autoimmune syndromes (Front Pharmacol 2026). Perugia and Rome rheumatologists, with Yehuda Shoenfeld, on JAK inhibitors in post-COVID syndromes, including features overlapping fibromyalgia and CFS. No search, no appraisal; two authors sit on Frontiers editorial boards, declared. Recorded because JAK inhibitors are among the drugs patients ask about, and nothing in this record changes the answer that there is no trial evidence for them in ME/CFS.

Three borderline calls worth naming. The first is the window boundary. The entry-date search for 8–15 November returned twelve PMIDs; three of them (the Vienna PEM questionnaire, the CDC infection-associated-illness review and the TSPO genotype preprint) entered PubMed on 8 November and were screened in last week's edition, so they are not counted or carried again. The second is carrying the telerehabilitation trial on its abstract alone. It is in because it is a randomised null on a question patients ask and its abstract reports its own corrected results plainly; every statement about it is drawn from that abstract. The third is the complementary-medicine survey, excluded on design though its subject is a real part of clinic practice; a reader who thought its 98% figure deserved an entry would not be unreasonable.

How this was assembled

PubMed was queried directly through the NCBI E-utilities. An entry-date search for 8 to 15 November returned a count of twelve and twelve PMIDs. Three had entered on 8 November and were already in last week's edition; re-running last week's window (2–8 November) returned exactly last week's seven PMIDs, and the earliest new record (the LIFT protocol) entered late on 8 November, after last week's run, so nothing fell between the windows and nothing is double-counted. The full MEDLINE records fetched for the nine new PMIDs contained nine records, so nothing was truncated. Entry dates ran from 8 to 14 November. Abstracts, affiliations, publication types, grant numbers and competing-interest statements were read for all nine.

Read end to end from PubMed Central: the ECHO trial, including its methods, all three tables, limitations, declarations and AI statement; the aripiprazole chart review, including methods, results, discussion, limitations and declarations; and the LIFT protocol, including eligibility, outcomes, sample size, statistical plan, adverse-event monitoring, trial status and declarations. The supplements of all three were not opened. Registries read through the ClinicalTrials.gov API: NCT05167227 (ECHO) for its registered outcomes and dates, and NCT06366724 (LIFT) for its outcomes, status and dates. Read at abstract level only: the telerehabilitation pilot, not attempted at its publisher for the reason given in its entry, and the five excluded records.

Where a paper's own numbers or statements disagree, the disagreement is named. In the ECHO trial: an abstract that omits a null primary outcome; a registry that names no primary subscale and a PEM secondary outcome the main text does not report; a discussion calling a sub-threshold effect “meaningful”; 590 against 587 patients; and 180/146 against 174/145 completers. In the aripiprazole review: a functional scale called both unvalidated and validated; a 1–5 against 0–5 rating range; a composite described as an average whose values are sums; two different treatment-duration figures; percentages that imply 52 patients rather than 50; 86.6% for 43 of 50; and a PubMed publication type of review. In the LIFT protocol: a recruitment end date already passed while the registry shows recruitment continuing, and “peak” oxygen uptake that the registry describes as extrapolated from a submaximal test. In the metformin preprint: groups that sum to three fewer than the eligible total. Arithmetic that checked out is also recorded: the ECHO arm and diagnosis totals, its responder proportions and the symmetry of both treatment-effect intervals; the aripiprazole functional-level distributions and means, its 27/22/1 split and its 15.8% side-effect rate; and the metformin long COVID proportions and starter percentage.

Each finding carries its study design, sample size and the authors' own stated limitations. Post-exertional malaise and pain carry standing extra weight: this week PEM was named in four of nine abstracts, rated once as an unvalidated single item, and used as a validated outcome in no completed study; pain gave the week's only positive between-group result. Nothing is filtered for a positive result: the lead item is a trial whose primary outcome was null, reported first, and the second trial carried is null too. Commercial, charitable and institutional interests are stated in each entry: a federal agency evaluating a programme it funded, with authors affiliated to it; a patient foundation funding a trial in which its chief executive and several centre directors are authors; an acupuncturist among the authors of a survey of acupuncture use; and a bioinformatics company affiliation. One declared use of ChatGPT, for spelling and grammar in the ECHO trial, is recorded without comment beyond the observation that this digest is written by a language model. Items from previous editions are referred to, not recycled: the naltrexone trial, the Dutch outcome-measures review and the PASS preprint are referenced, and only the PASS revision's new content is reported.

The standing query, with an entry-date range appended:

("Fatigue Syndrome, Chronic"[MeSH] OR "myalgic encephalomyelitis"[tiab]
 OR "chronic fatigue syndrome"[tiab] OR "ME/CFS"[tiab]
 OR "systemic exertion intolerance"[tiab] OR "post-exertional malaise"[tiab])

medRxiv was searched through its public API. Every posting dated 8 to 15 November — 396 in all, including revisions — was retrieved and filtered on title and abstract for myalgic encephalomyelitis, post-exertional, long covid and post-covid; a broader filter (chronic fatigue, ME/CFS, post-acute, PASC, post-viral, infection-associated, exertion intolerance, post-COVID-19 condition) found nothing further. For the metformin preprint and the PASS revision, the source XML was read in full for text, funder, competing interests, ethics, registration and reporting declarations. For the herpesvirus preprint, medRxiv returned a challenge page and the source XML was not available; only the API abstract was read, and the entry says so.

Every title links to its PubMed record or, for preprints, to the medRxiv abstract page. Previous editions are archived alongside this file. Nothing here is treatment advice; it is a literature digest for a clinician.