Skip to content

Non-pharmacologic therapy

TL;DR — Exercise is the anchor of fibromyalgia management: it is the only therapy graded "strong for" by EULAR 2017, on moderate-quality evidence that aerobic training improves quality of life, pain (~11/100 points vs control), and physical function, with withdrawal rates no worse than control (Macfarlane 2017, PMID 27377815; Bidonde 2017, PMID 28636204). CBT and ACT produce small but durable effects (pain SMD ~ −0.3 at end of treatment, maintained at 6 months; ACT: FIQ SMD −1.05) (Bernardy 2013, PMID 24018611; Eastwood 2024, PMID 38751564). Multicomponent programs (exercise + psychology/education) improve pain, fatigue, mood, and HRQoL short-term but fade after treatment ends (Häuser 2009, PMID 19177530). Acupuncture mostly fails sham control; rTMS/tDCS show small effects of uncertain durability; hydro-/balneotherapy shows small consistent effects. A phase 3 smartphone-delivered ACT program (PROSPER-FM) achieved 71% vs 22% PGIC response, pointing at the field's central problem — the evidence-based treatments are the least accessible, and adherence, not efficacy, is the rate-limiting step (Gendreau 2024, PMID 38991582).

Exercise: the anchor recommendation

All exercise Cochrane data are on 0–100 scales (lower better); the reviews use a 15% relative difference as the clinical-relevance threshold.

Modality (vs control) Pain HRQoL / multidimensional function Physical function Withdrawals Quality Source
Aerobic (13 RCTs, 839 pts) MD −11.06 (−18.34 to −3.77) MD −7.89 (−13.23 to −2.55) MD −10.16 (−15.39 to −4.94) RR 1.25 (0.89–1.77), ~20% vs 17% Moderate (HRQoL, withdrawals); low (pain, function) Bidonde 2017, PMID 28636204
Resistance (5 RCTs, 219 women) −3.3 cm/10 (−6.35 to −0.26) FIQ total −16.75 (−23.31 to −10.19) −6.29 (−10.45 to −2.13) ns Low Busch 2013, PMID 24362925
Aquatic (16 RCTs, 881 pts) −6.59 (−10.71 to −2.48); NNT 5 −5.97 (−9.06 to −2.88); NNT 5 stiffness −18.34; strength SMD 0.63 ns vs control Low–moderate Bidonde 2014, PMID 25350761
Mixed aerobic+resistance+flexibility (29 RCTs, 2,088 pts) absolute 5% better absolute 7% better absolute 11% better RR 1.02 Moderate Bidonde 2019, PMID 31124142
Flexibility alone (12 RCTs, 743 pts) no clinically important benefit vs aerobic — — ns Very low Kim 2019, PMID 31476271
Whole-body vibration (4 RCTs, 150 pts) uncertain uncertain — — Very low Bidonde 2017, PMID 28950401

Key facts beyond the table: aerobic benefits for pain and function persist 24–208 weeks post-intervention (low quality); benefits for HRQoL and fatigue do not (Bidonde 2017, PMID 28636204). Long-term effects of mixed exercise similarly persist for HRQoL/fatigue/function but not pain or stiffness (Bidonde 2019, PMID 31124142). Adverse-event reporting is uniformly poor across the exercise literature, but withdrawal rates match controls — exercise is tolerable, if patients can start it. Land- and water-based training show no clinically relevant differences (a statistically significant strength advantage for land-based training stayed below the review's 15% clinical-relevance threshold) (Bidonde 2014, PMID 25350761), so modality choice can follow preference and access — a practically important licence given the adherence problem (below).

Meditative movement: tai chi matched or beat aerobic exercise in a 226-patient, 52-week comparative-effectiveness RCT — combined tai chi groups improved FIQR 5.5 points more than aerobic exercise at 24 weeks; same-dose comparison (24 wk, 2×/wk) favored tai chi by 16.2 points; attendance was higher for tai chi; longer duration (24 vs 12 wk) gave greater improvement while frequency did not (Wang 2018, PMID 29563100). The German S3 guideline "strongly recommends" meditative movement therapies (qi gong, tai chi, yoga) (Langhorst 2017, PMID 28493227).

Education and self-management

No adequately powered trial isolates education alone; it is embedded in every guideline as the platform on which everything else is built. The German S3 guideline makes explicit, consensus-based recommendations: communicate the diagnosis explicitly, use shared decision-making based on patient preference and comorbidity, and apply a stepwise approach scaled to severity (Petzke 2017, PMID 28493229). EULAR's graduated model likewise starts all patients with education and non-pharmacologic therapy before any drug (Macfarlane 2017, PMID 27377815). What patients actually report using is different: rest, heat, distraction, and analgesics dominate self-management repertoires (Bennett 2007, PMID 17349056) — see patient-experience-and-advocacy.

Psychological therapies: CBT and ACT

  • CBT (23 studies, 2,031 participants): pain SMD −0.29 (−0.49 to −0.17) at end of treatment (≈0.5/10 points) and −0.40 at ~6-month follow-up (≈0.6 points); negative mood SMD −0.33 end / −0.43 follow-up; disability −0.30 end / −0.52 follow-up. Dropout no different from controls (RR 0.94). Effects are small but — unlike most drugs — grow or hold after treatment ends (Bernardy 2013, PMID 24018611). GRADE quality low; only two included studies were free of bias risk.
  • ACT (6 RCTs, 384 participants; group, online, or individual): FIQ SMD −1.05 (−2.02 to −0.09) post-treatment and −1.43 (−2.17 to −0.69) at follow-up; pain acceptance SMD +1.05; moderate–large improvements in anxiety, depression, pain, disability, fatigue; no attributable adverse events (Eastwood 2024, PMID 38751564). Small trial base; heterogeneous delivery.
  • Mindfulness (MBSR) (6 trials, 674 patients): short-term QoL SMD −0.35 and pain −0.23 vs usual care; effects "not robust against bias," no evidence of long-term effect (Lauche 2013, PMID 24290038). A broader meta of third-wave therapies in bodily distress (16 RCTs, 7 in FM) found somatic-symptom SMD −0.51 (−0.69 to −0.32) with no difference between mindfulness- and acceptance-based variants and no dose–response by session count (Maas genannt Bermpohl 2023, PMID 37151971).

Durability is the psychological therapies' selling point; effect size is their limitation. Note the asymmetry with drugs: CBT's −0.3 SMD is in the same numeric range as SNRIs' −0.36 (see pharmacologic-therapy), without the NNH 14 adverse-event dropout.

Multicomponent and multimodal programs

Multicomponent therapy (≥1 educational/psychological component + ≥1 exercise component; 9 RCTs, 1,119 subjects, median 24 h contact): pain SMD −0.37, fatigue WMD −0.85, depression SMD −0.67, HRQoL SMD −0.59 at end of treatment — the broadest short-term symptom coverage of any intervention — but no evidence of maintained effect on any symptom long-term except physical fitness (SMD 0.30 at median 7-month follow-up) (Häuser 2009, PMID 19177530). The German S3 guideline nonetheless strongly recommends ≥24 h multimodal therapy for severe FM (Schiltenwolf 2017, PMID 28493228), and in the network meta-analysis multicomponent therapy, aerobic exercise, and CBT were the most promising non-drug options (Nüesch 2013, PMID 22739992). The unsolved engineering problem is maintenance: booster sessions, digital follow-on, and community exercise transition are all untested at scale.

Hydrotherapy and balneotherapy

Meta-analysis of hydro-/balneotherapy RCTs: for hydrotherapy, pain SMD −0.42 (−0.61 to −0.24; 8 studies, 462 participants) and HRQoL SMD −0.40 (−0.62 to −0.18; 7 studies, 398 participants) at end of treatment, I²=0–15%, moderate-to-strong evidence for small effects; no end-of-treatment effect on depression or tender points (Naumann 2014, PMID 25000940). In EULAR 2017, balneotherapy sits below exercise — the guideline's sole "strong for" — in the weak-for tier of therapies (Macfarlane 2017, PMID 27377815). Access and cost, not evidence, drive its geographic usage pattern (central Europe, Turkey, Israel) [unverified].

Acupuncture: the sham-control reality

Cochrane (9 trials, 395 participants): versus no treatment, electroacupuncture improved pain by 22/100; versus sham, acupuncture (EA or manual) was no better for pain, fatigue, sleep, or global well-being (moderate-quality evidence), except less stiffness at one month; an EA subgroup (2 studies, 104 participants) showed pain SMD −0.63 vs sham; effects were gone by 6 months; manual acupuncture "probably does not improve pain or physical function" (Deare 2013, PMID 23728665). Guideline handling diverges: German S3 says acupuncture "can be considered" (Langhorst 2017, PMID 28493227); in EULAR 2017 it sits at best in the weak-for tier, since only exercise was graded strong-for (Macfarlane 2017, PMID 27377815); NICE NG193 includes a course of acupuncture among the options that may be offered for chronic primary pain ("Offer exercise, therapy, acupuncture, or antidepressants for chronic primary pain, says NICE" — Kmietowicz 2021, PMID 33824137) while foregrounding exercise and psychological therapy (Carville 2021, PMID 33883123; Korwisi 2021, PMID 34712885). The mechanistic reading — large contextual/ritual effect, small point-specific effect — parallels the drug placebo story (outcomes-and-measurement).

Neuromodulation: rTMS and tDCS

  • rTMS (11 RCTs, 303 patients): pain SMD −0.35 (−0.62 to −0.08) and QoL SMD −0.51 (−0.78 to −0.23) at ~4 weeks post-intervention vs sham; primary motor cortex (M1) stimulation drives the pain effect (SMD −0.57); no effect on depression or anxiety at either M1 or DLPFC (Toh 2022, PMID 35100184).
  • tDCS (6 RCTs, 192 patients): anodal M1 stimulation improves pain and FM-related function vs sham (p<0.05); anodal DLPFC does not; pressure pain thresholds unchanged with anodal or cathodal M1; no significant adverse effects (Zhu 2017, PMID 27983739).
  • Contrast: the German S3 guideline recommends against TMS (Winkelmann 2017, PMID 28493223) — written before most sham-controlled trials — illustrating how quickly neuromodulation evidence is outrunning guideline cycles. Sample sizes remain small, sham adequacy debated, and durability beyond ~4 weeks unmeasured; no neuromodulation modality has regulatory approval for FM [unverified].

Digital therapeutics

PROSPER-FM (phase 3, 25 US sites, n=275): 12-week self-guided smartphone ACT vs active control (symptom tracking + education): PGIC response 71% vs 22%, difference 48.4% (95% CI 37.9–58.9, p<0.0001); no device-related safety events; sponsor Swing Therapeutics (Gendreau 2024, PMID 38991582). This is an unusually large PGIC responder difference for an FM phase 3 trial, though against an active-control design rather than sham therapy, with research staff unblinded — and PGIC is a global-impression endpoint particularly sensitive to expectancy. The marketed app (Stanza, Swing Therapeutics) is documented as one of 13 FDA-cleared prescription digital therapeutics in the postmarket-surveillance literature (Lakhan 2025, PMID 40469887). If digital ACT holds up, it directly attacks the access problem: guideline-recommended CBT/ACT "is inaccessible to most patients" (trial's own framing; Gendreau 2024, PMID 38991582).

The dose/adherence problem — and what patients actually access

  • Trial exercise programs ran supervised sessions for 6–24 weeks (Bidonde 2017, PMID 28636204); unsupervised continuation after a program ends is where several benefits decay (long-term data above).
  • Attendance favors lower-intensity, mind-body formats: tai chi sessions were attended more often than aerobic exercise sessions in direct comparison (Wang 2018, PMID 29563100).
  • Multicomponent programs require ≥24 contact hours (Schiltenwolf 2017, PMID 28493228) — a resource reality that limits them to severe cases in systems that fund them at all.
  • What patients actually use, per a 2,596-respondent survey: rest, heat, pain medications, antidepressants, hypnotics rated most effective; the most used modalities were self-directed and passive, not the supervised active therapies the evidence favors (Bennett 2007, PMID 17349056).
  • In a UK community sample (n≈1,000) mapped against the new NG193 recommendations as a baseline, physical-therapy, exercise, and psychological-therapy use coexisted with 47% still reporting opioid use; the authors predicted slow adoption of the de-prescribing advice (Zambelli 2022, PMID 36032346).

Open questions

  • What maintains exercise benefit after supervision ends? Long-term pain/function effects persist for aerobic but HRQoL/fatigue effects do not (Bidonde 2017, PMID 28636204); no trial has optimized the maintenance phase itself.
  • Is tai chi's superiority over aerobic exercise (FIQR +16.2 points at matched dose; Wang 2018, PMID 29563100) replicable outside a single expert center, and what is the active ingredient (movement, attention, group, expectancy)?
  • Can digital ACT's 71% PGIC response survive a sham-app control and blinded outcome assessment (Gendreau 2024, PMID 38991582)?
  • Does combining exercise with neuromodulation or drugs beat either alone? Combination trials exist for drug–drug (pharmacologic-therapy) but drug–exercise factorial trials are essentially absent (Nüesch 2013, PMID 22739992 notes the evidence gap for combinations).
  • Why does acupuncture's clinical effect exceed its sham-controlled effect so consistently (Deare 2013, PMID 23728665), and should guidelines treat contextual analgesia as a legitimate therapeutic tool?
  • What is the minimal effective exercise dose for patients with severe fatigue/post-exertional worsening — and does the ME/CFS overlap subgroup respond differently (comorbidities-and-overlap)?

References

  1. Macfarlane GJ, Kronisch C, Dean LE, et al. EULAR revised recommendations for the management of fibromyalgia. Ann Rheum Dis. 2017;76:318-328. PMID 27377815
  2. Bidonde J, Busch AJ, Schachter CL, et al. Aerobic exercise training for adults with fibromyalgia. Cochrane Database Syst Rev. 2017;6:CD012700. PMID 28636204
  3. Busch AJ, Webber SC, Richards RS, et al. Resistance exercise training for fibromyalgia. Cochrane Database Syst Rev. 2013;12:CD010884. PMID 24362925
  4. Bidonde J, Busch AJ, Webber SC, et al. Aquatic exercise training for fibromyalgia. Cochrane Database Syst Rev. 2014;10:CD011336. PMID 25350761
  5. Bidonde J, Busch AJ, Schachter CL, et al. Mixed exercise training for adults with fibromyalgia. Cochrane Database Syst Rev. 2019;5:CD013340. PMID 31124142
  6. Kim SY, Busch AJ, Overend TJ, et al. Flexibility exercise training for adults with fibromyalgia. Cochrane Database Syst Rev. 2019;9:CD013419. PMID 31476271
  7. Bidonde J, Busch AJ, van der Spuy I, et al. Whole body vibration exercise training for fibromyalgia. Cochrane Database Syst Rev. 2017;9:CD011755. PMID 28950401
  8. Wang C, Schmid CH, Fielding RA, et al. Effect of tai chi versus aerobic exercise for fibromyalgia: comparative effectiveness randomized controlled trial. BMJ. 2018;360:k851. PMID 29563100
  9. Bernardy K, Klose P, Busch AJ, Choy EH, Häuser W. Cognitive behavioural therapies for fibromyalgia. Cochrane Database Syst Rev. 2013;9:CD009796. PMID 24018611
  10. Eastwood F, Godfrey E. The efficacy, acceptability and safety of acceptance and commitment therapy for fibromyalgia — a systematic review and meta-analysis. Br J Pain. 2024;18:243-256. PMID 38751564
  11. Lauche R, Cramer H, Dobos G, Langhorst J, Schmidt S. A systematic review and meta-analysis of mindfulness-based stress reduction for the fibromyalgia syndrome. J Psychosom Res. 2013;75:500-10. PMID 24290038
  12. Maas genannt Bermpohl F, Hülsmann L, Martin A. Efficacy of mindfulness- and acceptance-based cognitive-behavioral therapies for bodily distress in adults: a meta-analysis. Front Psychiatry. 2023;14:1160908. PMID 37151971
  13. Häuser W, Bernardy K, Arnold B, Offenbächer M, Schiltenwolf M. Efficacy of multicomponent treatment in fibromyalgia syndrome: a meta-analysis of randomized controlled clinical trials. Arthritis Rheum. 2009;61:216-24. PMID 19177530
  14. Schiltenwolf M, Eidmann U, Köllner V, et al. [Multimodal therapy of fibromyalgia syndrome: updated guidelines 2017]. Schmerz. 2017;31:285-288. PMID 28493228
  15. Nüesch E, Häuser W, Bernardy K, Barth J, Jüni P. Comparative efficacy of pharmacological and non-pharmacological interventions in fibromyalgia syndrome: network meta-analysis. Ann Rheum Dis. 2013;72:955-62. PMID 22739992
  16. Naumann J, Sadaghiani C. Therapeutic benefit of balneotherapy and hydrotherapy in the management of fibromyalgia syndrome: a qualitative systematic review and meta-analysis of randomized controlled trials. Arthritis Res Ther. 2014;16:R141. PMID 25000940
  17. Deare JC, Zheng Z, Xue CC, et al. Acupuncture for treating fibromyalgia. Cochrane Database Syst Rev. 2013;5:CD007070. PMID 23728665
  18. Toh EYJ, Ng JSP, McIntyre RS, et al. Repetitive transcranial magnetic stimulation for fibromyalgia: an updated systematic review and meta-analysis. Psychosom Med. 2022;84:400-409. PMID 35100184
  19. Zhu CE, Yu B, Zhang W, et al. Effectiveness and safety of transcranial direct current stimulation in fibromyalgia: a systematic review and meta-analysis. J Rehabil Med. 2017;49:2-9. PMID 27983739
  20. Winkelmann A, Bork H, Brückle W, et al. [Physiotherapy, occupational therapy and physical therapy in fibromyalgia syndrome: updated guidelines 2017]. Schmerz. 2017;31:255-265. PMID 28493223
  21. Langhorst J, Heldmann P, Henningsen P, et al. [Complementary and alternative procedures for fibromyalgia syndrome: updated guidelines 2017]. Schmerz. 2017;31:289-295. PMID 28493227
  22. Petzke F, Brückle W, Eidmann U, et al. [General treatment principles, coordination of care and patient education in fibromyalgia syndrome: updated guidelines 2017]. Schmerz. 2017;31:246-254. PMID 28493229
  23. Gendreau RM, McCracken LM, Williams DA, et al. Self-guided digital behavioural therapy versus active control for fibromyalgia (PROSPER-FM): a phase 3, multicentre, randomised controlled trial. Lancet. 2024;404:364-374. PMID 38991582
  24. Bennett RM, Jones J, Turk DC, Russell IJ, Matallana L. An internet survey of 2,596 people with fibromyalgia. BMC Musculoskelet Disord. 2007;8:27. PMID 17349056
  25. Zambelli Z, Halstead EJ, Iles R, Fidalgo AR, Dimitriou D. The 2021 NICE guidelines for assessment and management of chronic pain: a cross-sectional study mapping against a sample of 1,000 in the community. Br J Pain. 2022;16:439-449. PMID 36032346
  26. Carville S, Constanti M, Kosky N, Stannard C, Wilkinson C. Chronic pain (primary and secondary) in over 16s: summary of NICE guidance. BMJ. 2021;373:n895. PMID 33883123
  27. Korwisi B, Barke A, Kharko A, et al. Not really nice: a commentary on the recent version of NICE guidelines [NG193] by the Pain Net. Pain Rep. 2021;6:e961. PMID 34712885
  28. Kmietowicz Z. Offer exercise, therapy, acupuncture, or antidepressants for chronic primary pain, says NICE. BMJ. 2021;373:n907. PMID 33824137
  29. Lakhan SE. Postmarket safety surveillance of FDA-cleared prescription digital therapeutics using the Manufacturer and User Facility Device Experience (MAUDE) database: a pharmacovigilance study. Cureus. 2025;17(6):e85343. PMID 40469887