Skip to content

Genetics and risk factors

TL;DR — Fibromyalgia (FM) aggregates strongly in families (OR ~8.5 for FM in first-degree relatives of FM vs RA probands; sibling recurrence risk ratio λs ~13.6), and twin studies attribute roughly half the variance in chronic widespread pain (CWP) to genetic plus shared-environmental factors. The candidate-gene era (COMT, serotonergic genes) largely failed replication; its best-studied variant (COMT Val158Met) is null or inconsistent in meta-analyses and was not confirmed at GWAS scale. Modern GWAS changed the picture: a 2026 multi-ancestry meta-analysis of 2.5 million individuals (54,629 FM cases) found 26 risk loci, heritability enrichment exclusively in brain tissue, and genetic correlations >0.7 with low back pain, PTSD, and IBS — genetic evidence that FM is a CNS disorder on a continuum with other pain and affective phenotypes. Environmental triggers have graded evidence: prospective cohorts consistently show poor sleep, somatic symptom burden, and illness behavior predict new-onset CWP; childhood adversity and trauma show moderate, recall-bias-prone associations (OR ~1.5–2.7); physical trauma (whiplash) and specific infections (EBV, Lyme, COVID-19) have contested or biased evidence bases. Association and causation are separated explicitly below.

Familial aggregation

Study Design and criteria Key numbers
Arnold 2004 (PMID 15022338) 533 relatives of 78 FM probands vs 272 relatives of 40 RA probands; ACR criteria, dolorimetry + structured interview OR 8.5 (95% CI 2.8–26) for FM in relatives of FM vs RA probands; relatives also had more tender points and lower myalgic scores; FM coaggregated with major mood disorder (OR 1.8, 95% CI 1.1–2.9)
Buskila & Neumann 1997 (PMID 9150086) 117 close relatives of 30 female FM patients; ACR 1990 FM in 26% of blood relatives (41% of female relatives, 14% of male); 19% of husbands; relatives significantly more tender than controls
Arnold 2013 linkage (PMID 23280346) 116 multicase families, 341 microsatellites Sibling recurrence risk ratio λs = 13.6 (95% CI 10.0–18.5, assuming 2% population prevalence); suggestive linkage at chr 17p11.2–q11.2

Interpretation: familial clustering is real and large, but family studies cannot separate genes from shared environment, and referral-based proband ascertainment (community rheumatology practices in Arnold 2004) may inflate estimates. The coaggregation with mood disorder was the first formal evidence that FM and depression share familial (possibly genetic) determinants (Arnold 2004, PMID 15022338).

Twin studies and heritability

  • Swedish Twin Registry (15,950 complete pairs ≥42 y; CWP screened by telephone interview using ACR 1990 fibromyalgia criteria without examination): CWP prevalence 4.1%, female:male 3.3:1; genetic plus shared-environmental influences explained ~half of total variance, with modest additive genetic effects and no sex differences in the type or magnitude of genetic influence (Kato 2006, PMID 16646040). The commonly quoted "heritability ~50%" for CWP traces to this study; UK Biobank-era papers cite 48–54% (Rahman 2021, PMID 33926923).
  • Multivariate twin modeling of four functional somatic syndromes (CWP, chronic fatigue, IBS, recurrent headache) plus depression and GAD in 31,318 Swedish twins: the best-fitting model (in the women's data) was a common-pathway model with two latent traits — one loading on psychiatric disorders, one loading on all functional somatic syndromes (most heavily CWP) but not the psychiatric disorders; each syndrome also had illness-specific genetic influences (Kato 2009, PMID 18578896). This argues against FM being only somatized distress: the somatic-syndrome liability is partly separable from the affective one.
  • Finnish Twin Cohort: in 8,343 twins followed from 1975/1981 to 1990, membership in the "many FM symptoms" latent class was predicted by frequent headache (OR 8.6, 95% CI 3.8–19.2), persistent back pain (OR 4.7, 3.3–6.7), persistent neck pain (OR 3.3, 1.8–6.0), sleeping problems, and overweight (Markkula 2016, PMID 26772544).

Candidate-gene era and its replication failure

The 1990s–2000s produced many small case-control studies of monoaminergic candidates (COMT, 5-HTT/SLC6A4, 5-HT2A, dopamine genes). Report honestly, the record is poor:

  • COMT Val158Met (rs4680), the most-studied variant: a meta-analysis of 8 case-control studies (589 FM cases/527 controls) found no association in any genetic model with severe between-study heterogeneity (e.g., allele model OR 1.46, 95% CI 0.80–2.66) (Zhang 2014, PMID 25674213). A systematic review across chronic pain conditions concluded FM/CWP was the only chronic pain condition with a possible rs4680 association (Met as risk allele) — while finding none for migraine or musculoskeletal pain (Tammimäki 2012, PMID 22722321). An updated 2025 meta-analysis (16 studies; 2,115 patients/1,867 controls) found a pooled Met-allele association (OR 1.375, 95% CI 1.076–1.757) that disappeared in every regional subgroup (European, Latin American, Asian) — a classic small-study/heterogeneity signature — though Met/Met genotype was associated with higher FIQ scores (WMD 11.24 vs Val/Val) (Lee 2025, PMID 40153322).
  • The decisive test came at GWAS scale: in the UK Biobank CWP GWAS, the COMT locus reached genome-wide significance in discovery but failed replication in six independent European cohorts (n=43,080) (Rahman 2021, PMID 33926923).

Lesson: candidate-gene FM findings were largely underpowered, heterogeneous in phenotype definition, and prone to publication bias; none should be treated as established without biobank-scale replication.

Genome-wide association studies

Study Phenotype and sample Findings
Docampo 2014 (PMID 24582949) First FM GWAS; 313 low-comorbidity FM cases vs 220 controls (+952/644 replication), women, Spain No SNP reached genome-wide significance; nominal signals at MYT1L (rs11127292) and an intronic NRXN3 deletion — both CNS genes; underpowered by modern standards
Rahman 2021 (PMID 33926923) CWP ("pain all over the body" >3 months, self-report), UK Biobank, 6,914 cases/242,929 controls; replication in TwinsUK, ELSA, Rotterdam 1–3, HUNT (n=43,080) 3 genome-wide significant loci: RNF123 (E3 ubiquitin ligase, highly expressed in skeletal muscle), ATP2C1 (Golgi calcium pump hSPCA1; Hailey-Hailey gene), and COMT (3′-UTR variant rs165599). RNF123 replicated (meta-analysis p=0.0002), ATP2C1 suggestive (p=0.0227), COMT not replicated. Partial genetic correlations with depressive symptoms, BMI, age at first birth, years of schooling; tissue-specificity and colocalization highlighted skeletal muscle and dorsal root ganglion
Kerrebijn 2026 (PMID 42521817) Multi-ancestry GWAS meta-analysis, 54,629 FM cases (90% European; 87.7% female)/2,509,126 controls, 11 biobank cohorts (UK Biobank, FinnGen, Estonian Biobank, All of Us, Genes & Health, deCODE Iceland, Danish CHB/DBDS, and 4 US health-system biobanks); uniform records-based case definition (ICD-10 M79.7 in inpatient/primary-care records, ascertained within each cohort) 26 risk loci. Strongest association: a coding variant in HTT (the Huntington disease gene); prioritized genes include GPR52 (HTT regulator), DCC, DRD2/NCAM1, MDGA2, CELF4 — all neural. Heritability enriched exclusively in brain tissue and neural cell types. Genetic correlations >0.7 with low back pain, PTSD, and IBS; broad positive rg with chronic pain, psychiatric, and somatic disorders. Despite the female-predominant prevalence, genetic architecture was nearly identical between sexes (inter-sex rg ≈ 1.0)
Lin 2026 (PMID 41576141) Genomic structural equation modeling across GWAS of FM, insomnia, depression, anxiety Pairwise rg 0.55–0.84; a single shared latent factor fits well; factor GWAS: 49 lead SNPs across 43 loci (32 novel), enriched in synaptic pathways; MR implicated corpus-callosum white-matter integrity and neuroinflammatory markers (CD40)

What GWAS implies: 1. FM is polygenic with many small-effect common variants — no "fibromyalgia gene." 2. Risk biology localizes to brain, not muscle or immune tissue, in the largest study (Kerrebijn 2026, PMID 42521817) — genetic support for the central/nociplastic framing (note the earlier, smaller CWP GWAS had pointed to skeletal muscle; the discrepancy is unresolved). 3. High genetic correlations with mood disorders, PTSD, IBS, and other pain conditions indicate shared inherited liability across the chronic overlapping pain conditions rather than distinct etiologies — see comorbidities-and-overlap. 4. The sex-similar genetic architecture pushes explanation of the female predominance toward hormonal, ascertainment, or environmental mechanisms rather than sex-specific genetics (Kerrebijn 2026, PMID 42521817).

Environmental triggers, graded by evidence quality

Association ≠ causation: most trigger literature is retrospective/cross-sectional and vulnerable to recall bias (patients search their past for causes) and reverse causation. Prospective designs are flagged.

Physical trauma (whiplash / MVA) — contested

  • For: prospective controlled assessment of 102 adults with neck injury vs 59 with leg fractures (ACR 1990 exams ~3 months post-injury, Israel): FM diagnosed in 21.6% vs 1.7% (~13×); insurance claims were not associated with FM, arguing against pure compensation-seeking (Buskila 1997, PMID 9082932).
  • Against: a later prospective Israeli cohort (153 ER-diagnosed whiplash vs 53 road-accident fracture controls, ACR 1990, mean 14.5 months follow-up) found one FM case in the whiplash group and none in controls (Tishler 2006, PMID 16652434).
  • Status: unresolved. The two best studies from the same country, similar design, give ~22% vs ~0.7%. Differences in setting (hospitalized trauma vs ER discharge), timing of assessment, and case ascertainment may explain the gap; no adequately powered multicenter prospective study exists. Evidence grade: low/conflicting.

Infections — mostly association; one strong prospective post-infective model

  • Hepatitis C: FM in 16% of 90 HCV-infected patients vs 0% of 128 healthy controls (ACR 1990) (Buskila 1997, PMID 9385302); a later clinical cohort reported 57% (106/185) (Mohammad 2012, PMID 22499073) — clinic-based selection likely inflates this. Cross-sectional; direction unproven.
  • EBV: 55% of 50 FM patients reported sudden viral-like onset, but EBV antibody titers did not differ from healthy or unhealthy controls (Buchwald 1987, PMID 2823835); a 13-virus serology panel in 548 chronically fatigued patients (including FM subset) found no consistent differences (Buchwald 1996, PMID 8890037). Serologic evidence for EBV as a specific trigger: essentially negative.
  • Lyme disease: 8% of 287 Lyme patients developed FM, usually after early or arthritic Lyme; antibiotics resolved objective Lyme signs but 14/15 kept FM symptoms (Dinerman & Steere 1992, PMID 1637022). Conversely, in 788 referrals to a Lyme clinic, 57% had no evidence of Lyme — the majority of these carrying CFS or FM, and 45% with false-positive outside serologies (Steere 1993, PMID 8459513). Lyme can plausibly trigger FM; "chronic Lyme" is more often FM/CFS misdiagnosis than persistent infection.
  • Post-infective syndrome model (prospective, best-quality): in the Dubbo cohort, 253 patients followed from acute EBV, Q fever, or Ross River virus infection: a stereotyped syndrome of fatigue, musculoskeletal pain, neurocognitive difficulty and mood disturbance persisted in 12% at 6 months (11% meeting CFS criteria), at similar rates after each pathogen, predicted by acute illness severity rather than psychological or microbiological factors (Hickie 2006, PMID 16950834). This supports a pathogen-nonspecific host-response pathway relevant to FM onset.
  • COVID-19: in a web-based survey of 616 people ~6 months after COVID-19, 30.7% met ACR survey criteria for FM; male sex (OR 9.95) and obesity (OR 41.2) predicted post-COVID FM (Ursini 2021, PMID 34426540) — but self-selected web samples cannot give incidence. A systematic review of post-acute COVID musculoskeletal disease reported FM-criteria fulfillment of 31–40% across studies (Ciaffi 2023, PMID 37296705). A 2026 scoping review of 228 records concluded prevalence studies remain ongoing, definitionally inconsistent, and bias-susceptible (Plaut 2026, PMID 42174645). In a TriNetX propensity-matched cohort (90,508 vaccinated vs 90,508 unvaccinated COVID survivors), vaccination was associated with lower new-onset FM risk (HR 0.84, 95% CI 0.71–0.99) (Liu 2025, PMID 40312371). Evidence grade: moderate for an FM-like post-COVID phenotype existing; low for precise incidence.

Psychosocial stress and adverse childhood experiences — consistent but confounded

  • Abuse meta-analysis (18 case-control studies, 13,095 subjects): FM associated with childhood physical abuse (OR 2.49, 95% CI 1.81–3.42), childhood sexual abuse (OR 1.94, 1.36–2.75), adult physical abuse (OR 3.07) and adult sexual abuse (OR 2.24); study quality mostly poor, and lower quality predicted larger effects for childhood sexual abuse (Häuser 2011, PMID 20722042).
  • ACEs → adult chronic pain meta-analysis (85 studies, 826,452 adults): any direct ACE aOR 1.45 (95% CI 1.38–1.53) for chronic pain; dose–response from 1 ACE (aOR 1.29) to ≥4 ACEs (aOR 1.95) (Bussières 2023, PMID 38111090).
  • Psychological trauma/PTSD meta-analysis (71 controlled studies): trauma exposure OR 2.7 (95% CI 2.27–3.10) for functional somatic syndromes; PTSD associations larger than abuse; self-report instruments inflate estimates (Afari 2014, PMID 24336429).
  • Interpretation: effects are modest (OR ~1.5–2.7), dose-dependent, and consistent — but almost entirely retrospective self-report; genetic confounding (shared liability to adversity exposure and pain) is untested at scale.

Prospective predictor studies: who develops chronic widespread pain

The strongest causal-adjacent evidence in this field, because exposure precedes outcome:

Cohort Follow-up Predictors of new-onset CWP/FM
UK population, 1,658 adults free of CWP (McBeth 2001, PMID 11315933) 12 months; new CWP in 2% of pain-free, 8% of regional-pain subjects Somatic Symptom Checklist OR 3.3; Illness Behavior subscale OR 9.0 — independent of baseline pain
UK population, 3,171 adults 25–65 free of CWP (Gupta 2007, PMID 17085772) 15 months; 10% developed CWP Somatic symptoms OR 1.8; illness behavior OR 3.3; sleep problems OR 2.7; all three together → 12× odds
HUNT, 12,350 Norwegian women free of FM/musculoskeletal pain (Mork & Nilsen 2012, PMID 22081440) ~10 years; 327 incident self-reported FM Dose-dependent sleep-problem association; RR 3.43 (95% CI 2.26–5.19) for frequent sleep problems; RR 5.41 in women ≥45 y
UK adults ≥50, 4,326 followed (McBeth 2014, PMID 24574238) 3 years; 18.5% new widespread pain (ACR 1990 definition) Nonrestorative sleep the strongest independent predictor (OR 1.9, 95% CI 1.2–2.8); also anxiety, cognitive complaint, poorer physical HRQoL
Finnish Twin Cohort, 8,343 (Markkula 2016, PMID 26772544) 15 years Frequent headache OR 8.6; persistent back/neck pain OR 4.7/3.3; sleep problems; overweight

Convergent picture: pre-existing regional pain, somatic symptom amplification, illness behavior, and above all disturbed sleep predict transition to CWP/FM. These are risk markers, not proven causes — but sleep is the one modifiable predictor replicated in four independent cohorts (see sleep-fatigue-cognition).

Association vs causation — explicit accounting

  • Established as associated and heritable: familial clustering, twin-based heritability ~50%, 26 GWAS loci, genetic correlations with mood/pain traits.
  • Plausibly causal upstream factors (prospective evidence, dose-response): sleep disturbance, somatic symptom burden, acute-infection severity (Dubbo), childhood adversity (dose-response but retrospective).
  • Contested/unproven as causes: whiplash trauma (conflicting prospective studies), specific pathogens (EBV serology negative; HCV cross-sectional; COVID incidence unsettled), COMT and all single candidate genes.
  • Mendelian randomization attempts exist for biomarker exposures (e.g., CRP shows no causal effect on CWP; Suri 2024 preprint, PMID 39006441) but MR of behavioral exposures (sleep, adversity) on FM remains thin.

Open questions

  • Do the 26 Kerrebijn loci replicate in clinically ascertained (ACR-criteria) FM rather than the ICD-code (M79.7) biobank definition, and what fraction of the familial λs (~13.6; Arnold 2013, PMID 23280346) do they explain? (Kerrebijn 2026, PMID 42521817)
  • Why does the largest FM GWAS localize heritability exclusively to brain (PMID 42521817) while the UK Biobank CWP GWAS implicated skeletal muscle (PMID 33926923) — phenotype-definition artifact or genuine biological difference between CWP and FM?
  • Is the HTT coding-variant association (PMID 42521817) acting through huntingtin biology (GPR52 axis), and does it nominate a druggable pathway?
  • Can within-family or MR designs test whether ACEs cause FM beyond shared familial liability, given that abuse-FM associations weaken with study quality (Häuser 2011, PMID 20722042; Bussières 2023, PMID 38111090)?
  • Does treating sleep disturbance in high-risk (regional-pain, high somatic symptom) individuals prevent transition to CWP — the interventional test the four prospective cohorts (PMIDs: 11315933, 17085772, 22081440, 24574238) collectively motivate?
  • What is the true incidence of post-COVID FM against matched uninfected controls, using standardized criteria? (Plaut 2026, PMID 42174645)

References

  1. Arnold LM, et al. Family study of fibromyalgia. Arthritis Rheum. 2004;50(3):944-52. PMID 15022338
  2. Buskila D, Neumann L. Fibromyalgia syndrome (FM) and nonarticular tenderness in relatives of patients with FM. J Rheumatol. 1997;24(5):941-4. PMID 9150086
  3. Arnold LM, et al. The fibromyalgia family study: a genome-wide linkage scan study. Arthritis Rheum. 2013;65(4):1122-8. PMID 23280346
  4. Kato K, et al. Importance of genetic influences on chronic widespread pain. Arthritis Rheum. 2006;54(5):1682-6. PMID 16646040
  5. Kato K, et al. A population-based twin study of functional somatic syndromes. Psychol Med. 2009;39(3):497-505. PMID 18578896
  6. Markkula RA, et al. Predictors of fibromyalgia: a population-based twin cohort study. BMC Musculoskelet Disord. 2016;17:29. PMID 26772544
  7. Zhang L, et al. Meta-analysis reveals a lack of association between a common catechol-O-methyltransferase (COMT) polymorphism val158met and fibromyalgia. Int J Clin Exp Pathol. 2014;7(12):8489-97. PMID 25674213
  8. Tammimäki A, Männistö PT. Catechol-O-methyltransferase gene polymorphism and chronic human pain: a systematic review and meta-analysis. Pharmacogenet Genomics. 2012;22(9):673-91. PMID 22722321
  9. Lee YH, Song GG. Association of catechol-O-methyltransferase Val158Met polymorphism with fibromyalgia risk and FIQ scores: an updated meta-analysis. Clin Exp Rheumatol. 2025;43(6):1002-1009. PMID 40153322
  10. Docampo E, et al. Genome-wide analysis of single nucleotide polymorphisms and copy number variants in fibromyalgia suggest a role for the central nervous system. Pain. 2014;155(6):1102-1109. PMID 24582949
  11. Rahman MS, et al. Genome-wide association study identifies RNF123 locus as associated with chronic widespread musculoskeletal pain. Ann Rheum Dis. 2021;80(9):1227-1235. PMID 33926923
  12. Kerrebijn I, et al. The genetic architecture of fibromyalgia across 2.5 million individuals. Nat Med. 2026;32(8):3060-3070. PMID 42521817
  13. Lin L, et al. Shared latent genetic liability across fibromyalgia and psychiatric traits: novel insights from genomic structural equation modeling. PLoS Genet. 2026;22(1):e1012034. PMID 41576141
  14. Buskila D, et al. Increased rates of fibromyalgia following cervical spine injury. A controlled study of 161 cases of traumatic injury. Arthritis Rheum. 1997;40(3):446-52. PMID 9082932
  15. Tishler M, et al. Neck injury and fibromyalgia — are they really associated? J Rheumatol. 2006;33(6):1183-5. PMID 16652434
  16. Buskila D, et al. Fibromyalgia in hepatitis C virus infection. Another infectious disease relationship. Arch Intern Med. 1997;157(21):2497-500. PMID 9385302
  17. Mohammad A, et al. Prevalence of fibromyalgia among patients with chronic hepatitis C infection: relationship to viral characteristics and quality of life. J Clin Gastroenterol. 2012;46(5):407-12. PMID 22499073
  18. Buchwald D, et al. The "chronic, active Epstein-Barr virus infection" syndrome and primary fibromyalgia. Arthritis Rheum. 1987;30(10):1132-6. PMID 2823835
  19. Buchwald D, et al. Viral serologies in patients with chronic fatigue and chronic fatigue syndrome. J Med Virol. 1996;50(1):25-30. PMID 8890037
  20. Dinerman H, Steere AC. Lyme disease associated with fibromyalgia. Ann Intern Med. 1992;117(4):281-5. PMID 1637022
  21. Steere AC, et al. The overdiagnosis of Lyme disease. JAMA. 1993;269(14):1812-6. PMID 8459513
  22. Hickie I, et al. Post-infective and chronic fatigue syndromes precipitated by viral and non-viral pathogens: prospective cohort study. BMJ. 2006;333(7568):575. PMID 16950834
  23. Ursini F, et al. Fibromyalgia: a new facet of the post-COVID-19 syndrome spectrum? Results from a web-based survey. RMD Open. 2021;7(3):e001735. PMID 34426540
  24. Ciaffi J, et al. Post-acute COVID-19 joint pain and new onset of rheumatic musculoskeletal diseases: a systematic review. Diagnostics (Basel). 2023;13(11):1850. PMID 37296705
  25. Plaut S. A systematic scoping review and conceptual analysis of new-onset fibromyalgia manifestations after non-hospitalized COVID-19. J Transl Med. 2026;24(1). PMID 42174645
  26. Liu A, et al. COVID-19 vaccination reduces new-onset fibromyalgia risk in survivors. BMC Med. 2025;23(1):255. PMID 40312371
  27. Häuser W, et al. Emotional, physical, and sexual abuse in fibromyalgia syndrome: a systematic review with meta-analysis. Arthritis Care Res (Hoboken). 2011;63(6):808-20. PMID 20722042
  28. Bussières A, et al. Adverse childhood experience is associated with an increased risk of reporting chronic pain in adulthood: a systematic review and meta-analysis. Eur J Psychotraumatol. 2023;14(2):2284025. PMID 38111090
  29. Afari N, et al. Psychological trauma and functional somatic syndromes: a systematic review and meta-analysis. Psychosom Med. 2014;76(1):2-11. PMID 24336429
  30. McBeth J, et al. Features of somatization predict the onset of chronic widespread pain: results of a large population-based study. Arthritis Rheum. 2001;44(4):940-6. PMID 11315933
  31. Gupta A, et al. The role of psychosocial factors in predicting the onset of chronic widespread pain: results from a prospective population-based study. Rheumatology (Oxford). 2007;46(4):666-71. PMID 17085772
  32. Mork PJ, Nilsen TI. Sleep problems and risk of fibromyalgia: longitudinal data on an adult female population in Norway. Arthritis Rheum. 2012;64(1):281-4. PMID 22081440
  33. McBeth J, et al. Predictors of new-onset widespread pain in older adults: results from a population-based prospective cohort study in the UK. Arthritis Rheumatol. 2014;66(3):757-67. PMID 24574238
  34. Suri P, et al. No evidence for causal effects of C-reactive protein (CRP) on chronic pain conditions: a Mendelian randomization study. medRxiv [preprint]. 2024. PMID 39006441