Skip to content

Diagnostic Criteria

TL;DR — Five criteria generations coexist: ACR 1990 (widespread pain + ≥11/18 tender points, examiner-dependent; Wolfe 1990, PMID 2306288), ACR 2010 preliminary (WPI + symptom severity, physician-administered; Wolfe 2010, PMID 20461783), the 2011 self-report modification for research (Wolfe 2011, PMID 21285161), the 2016 revision adding a generalized-pain requirement (Wolfe 2016, PMID 27916278), and AAPT 2019 (multisite pain + sleep problems or fatigue; Arnold 2019, PMID 30453109); ICD-11 separately codes FM under chronic primary pain (Treede 2019, PMID 30586067). The sets agree only moderately: in the same Scottish population, prevalence was 1.7% (1990), 1.2% (2010) and 5.4% (modified 2010), with female:male ratios of 13.7:1, 4.8:1 and 2.3:1 respectively (Jones 2015, PMID 25323744) — the tender point, not the disease, generated the extreme female predominance. In practice most clinicians apply no criteria at all: clinic diagnosis agrees with criteria at κ ≈ 0.3–0.4, missing about half of criteria-positive patients while labeling criteria-negative ones (Wolfe 2019, PMID 30724039; Srinivasan 2019, PMID 31777779), and at population level ~73% of criteria-positive adults are undiagnosed (Walitt 2015, PMID 26379048). Over- and under-diagnosis are simultaneously large.

Criteria genealogy

Criteria set Case definition (essentials) Administration Intended use Key performance figures
ACR 1990 (Wolfe 1990, PMID 2306288) Widespread pain ≥3 months (axial + left/right + above/below waist) and ≥11/18 tender points at ~4 kg pressure Physician exam (dolorimetry/digital palpation) Classification (research) Sensitivity 88.4%, specificity 81.1% vs clinical diagnosis in 293 FM vs 265 rheumatic controls
ACR 2010 preliminary (Wolfe 2010, PMID 20461783) WPI ≥7 & SS ≥5, or WPI 3–6 & SS ≥9; symptoms ≥3 months; no tender points Physician interview/exam (WPI 0–19; SS 0–12) Clinical diagnosis + severity measurement Correctly classifies 88.1% of 1990-classified cases; 25% of previously diagnosed FM patients no longer met 1990 criteria at study time
2011 modification ("modified 2010" / research criteria) (Wolfe 2011, PMID 21285161) Same WPI/SS structure; physician estimate of somatic symptoms replaced by 3 self-reported items (headache, lower-abdominal pain/cramps, depression); Fibromyalgia Symptom scale (FS = WPI+SS, 0–31) Fully self-report Epidemiology and research; not self-diagnosis FS ≥13 classified 93.0% correctly (sens 96.6%, spec 91.8% in study population); criteria met by 60% of prior FM, 21.1% RA, 16.8% OA, 36.7% SLE
2016 revision (Wolfe 2016, PMID 27916278; Galvez-Sánchez 2020, PMID 32340369) WPI ≥7 & SSS ≥5, or WPI 4–6 & SSS ≥9; generalized pain in ≥4 of 5 regions (jaw/chest/abdominal pain excluded from region count); ≥3 months; diagnosis valid irrespective of other diagnoses (exclusion clause removed) Physician or self-report (self-report valid for research only) Unifies 2010/2011 into one set Vs 1990/clinical criteria across 14 validation studies: median sensitivity 86%, specificity 90%; generalized-pain criterion cuts regional-pain misclassification (from 1.8% to 0.4% by one definition)
AAPT 2019 (Arnold 2019, PMID 30453109; Galvez-Sánchez 2020, PMID 32340369) Core criteria: multisite pain (≥6 of 9 body sites) plus moderate-to-severe sleep problems or fatigue, both present ≥3 months; embedded in a 5-dimension taxonomy (core criteria, common features, comorbidities, consequences, mechanisms) Clinical assessment Clinically oriented taxonomy (ACTTION-APS/FDA) Not yet validated at publication; in one head-to-head: sensitivity 73.8%, specificity 91.7%, correct classification 81.8% vs rheumatologist diagnosis — the weakest of the compared sets (Salaffi 2020, PMID 32206792)
ICD-11 (2019) (Treede 2019, PMID 30586067; Nicholas 2019, PMID 30586068) FM sits under chronic primary pain: pain >3 months + significant distress or functional disability, not better accounted for by another condition; FM itself is coded MG30.01 within this block (Berwick 2022, PMID 36427885) Coding classification, not a criteria set Health records, epidemiology, policy Replaces ICD-10 M79.7-era coding (WHO recognition since 1992; Galvez-Sánchez 2020, PMID 32340369)

Supporting instrument: the polysymptomatic distress (PSD) scale (= FS, 0–31) grades severity continuously (none 0–3, mild 4–7, moderate 8–11, severe 12–19, very severe 20–31) and extends measurement beyond the diagnostic threshold (Galvez-Sánchez 2020, PMID 32340369; Wolfe 2011, PMID 21285161).

Instrument anatomy (2010-lineage scales)

All post-1990 criteria are built from two scales (details per the criteria papers and the full-text review Galvez-Sánchez 2020, PMID 32340369):

  • Widespread Pain Index (WPI, 0–19): count of 19 predefined body areas painful in the last week (Wolfe 2010, PMID 20461783).
  • Symptom Severity scale (SS/SSS, 0–12): part A grades fatigue, waking unrefreshed, and cognitive symptoms 0–3 each (0–9 subtotal); part B grades somatic-symptom extent 0–3 — in 2010 from a physician-assessed 41-item checklist, replaced in 2011 by three self-reported yes/no items (headaches, lower-abdominal pain or cramps, depression) (Wolfe 2010, PMID 20461783; Wolfe 2011, PMID 21285161; Galvez-Sánchez 2020, PMID 32340369).
  • Polysymptomatic distress / Fibromyalgia Symptom scale (PSD = FS = WPI + SS, 0–31): continuous severity measure; bands none 0–3, mild 4–7, moderate 8–11, severe 12–19, very severe 20–31; every criteria-positive patient scores ≥12, and FS ≥13 was the optimal single cut-off (93.0% correct classification) (Wolfe 2011, PMID 21285161; Galvez-Sánchez 2020, PMID 32340369).
  • Fibromyalgia Survey Questionnaire (FSQ): the packaged self-report version (symptom severity items + 19-area pain diagram + duration item) used in epidemiologic studies; validated in Spanish, German, Japanese, and Persian; explicitly not for self-diagnosis (Galvez-Sánchez 2020, PMID 32340369).
  • Generalized pain regions (2016): 5 regions (4 quadrants + axial); ≥4 required; jaw, chest, and abdominal pain do not count toward regions (Wolfe 2016, PMID 27916278; Galvez-Sánchez 2020, PMID 32340369).

Reporting recommendation from the criteria authors: studies should always report FM and generalized-pain positivity plus WPI, SSS, and PSD (Wolfe 2018, PMID 30212526).

What each revision was fixing

  • 1990 → 2010: the tender point exam was rarely performed or was performed incorrectly in primary care, and tender points ignored the symptom burden (fatigue, sleep, cognition); ~25% of previously diagnosed patients failed the 1990 criteria at re-examination while remaining symptomatic (Wolfe 2010, PMID 20461783). The 1990 set was also criticized as an all-or-nothing construct at odds with continuum data (Galvez-Sánchez 2020, PMID 32340369; Wolfe 1997, PMID 9166001).
  • 2010 → 2011: the physician-dependent somatic-symptom estimate blocked survey research; three self-reported symptoms replaced it, making the instrument fully self-administered for epidemiology (Wolfe 2011, PMID 21285161).
  • 2011 → 2016: the WPI floor allowed regional pain syndromes to qualify; 6.2% of 2011-positive patients did not satisfy 1990-style widespread pain, and a ≥4-of-5-region "generalized pain" requirement retained 98.8% of cases while excluding regional-pain false positives (Wolfe 2016 [widespread pain analysis], PMID 27370877; Wolfe 2016, PMID 27916278). The 2016 set also deleted the confusing "exclude other explanations" clause: FM is diagnosable alongside other diseases (Wolfe 2016, PMID 27916278; Galvez-Sánchez 2020, PMID 32340369).
  • AAPT (2019): an attempt to align FM with a cross-condition chronic-pain taxonomy for trials and clinical use, adding dimensions (course, risk factors, mechanisms) that ACR-lineage criteria omit (Arnold 2019, PMID 30453109).

Agreement between criteria sets

Comparison Setting / N Result Source (PMID)
2010/2011 vs 1990 & clinical criteria 14 validation studies (review) Median sensitivity 86%, specificity 90% 27916278
2010 vs 1990 829 patients (criteria development) 88.1% of 1990-classified cases captured 20461783
2011 vs clinical FM diagnosis Spanish samples Sensitivity 88.3%, specificity 91.8% (FM vs healthy); Japanese validation: 64% / 96% (FM vs RA/OA) 32340369
1990 vs 2010 vs modified 2010, same population NE Scotland, 1,604 survey → 104 examined Prevalence 1.7% / 1.2% / 5.4%; the three sets identify different people, not nested subsets 25323744
2011 vs 2016 vs AAPT vs rheumatologist diagnosis 732 chronic-pain referrals, Italy Correct classification 85.1% / 83.6% / 81.8%; "considerable agreement," AAPT weakest 32206792
Criteria (2010/2011 self-report) vs physician ICD diagnosis 497 university-clinic patients Agreement 79.2% but κ=0.41 (fair); κ=0.32 in RA subset 30724039
Criteria (2016) vs physician diagnosis 3,276 primary-care patients κ=0.296; only 32.2% of physician-diagnosed met criteria; only 35.4% of criteria-positive had the diagnosis 31777779

Interpretation: within clinic populations selected for FM-like presentations, ACR-lineage sets agree reasonably well with each other (≈80–90%). Across the general population, the sets diverge sharply in prevalence and in who qualifies (Jones 2015, PMID 25323744). Agreement between any criteria set and real-world physician behavior is poor everywhere measured.

The tender-point sex bias

The 1990 tender point requirement built a sex artifact into the case definition:

  • In the general population, women have systematically higher tender point counts (median 6 vs 3 in men), and tender points track distress independent of pain (Croft 1994, PMID 7950521).
  • Consequently the 1990 criteria select women at extreme ratios: F:M 13.7:1, vs 4.8:1 under 2010 and 2.3:1 under modified 2010 criteria in the same population (Jones 2015, PMID 25323744).
  • 1990-criteria-era population studies accordingly report large sex gaps (women 3.4% vs men 0.5%, Wolfe 1995, PMID 7818567; women 4.9% vs men 1.6%, White 1999, PMID 10405947), and a meta-analysis dominated by such studies estimates female prevalence 3.98% vs male 0.01% (Heidari 2017, PMID 28447207) — a near-zero male figure incompatible with symptom-based criteria studies.
  • With tender points removed and unbiased sampling, the female share of criteria-positive FM falls to ≈59–60% (58.7% among RA-cohort FM cases; 59.2% in the German population study), against >90% among clinically referred/diagnosed patients — the ≥90%-female picture is a product of selection and confirmation bias plus the old criteria, not of the underlying condition (Wolfe 2018, PMID 30212526). In a German population sample, criteria prevalence in women (2.4%) and men (1.8%) did not differ significantly (Wolfe 2013, PMID 23424058).
  • Clinician behavior preserves the artifact: women are more likely to receive a physician diagnosis than to meet criteria (diagnosis OR for women 3.2 vs criteria OR 1.9, Srinivasan 2019, PMID 31777779; same direction in Wolfe 2019, PMID 30724039).

Criteria misuse in practice

  • Most health professionals apply no published criteria; digital palpation without pressure control persists as the de facto method long after tender points were dropped (Galvez-Sánchez 2020, PMID 32340369).
  • Referral-based assessment found the initial FM label correct in only 34% of cases, with error mostly over-diagnosis of other conditions as FM (Fitzcharles & Boulos 2003, PMID 12595620).
  • The 2011/2016 self-report versions are explicitly not valid for individual clinical diagnosis, yet their simplicity invites exactly that use (Wolfe 2011, PMID 21285161; Wolfe 2016, PMID 27916278).
  • Physician diagnosis appears "subjective and unrelated to fibromyalgia criteria," with no common community definition (Srinivasan 2019, PMID 31777779) — a problem for every ICD-code-based study in epidemiology.

Over- and under-diagnosis: both are real

Direction Evidence Source (PMID)
Under-diagnosis 73% of criteria-positive US adults (NHIS surrogate criteria) reported no FM diagnosis despite severe symptoms and high disability 26379048
Survey-criteria prevalence 6.4% vs 1.1% clinically diagnosed in Olmsted County; criteria-positive men especially unlikely to be diagnosed 23203795
Rheumatologists failed to diagnose 49.6% of criteria-positive clinic patients 30724039
Only 35.4% of criteria-positive primary-care patients carried the diagnosis 31777779
Over-diagnosis Only 32.2% of physician-diagnosed FM met 2016 criteria; 68.3% of the physician-diagnosed group received FM-specific drugs regardless 31777779
11.4% of criteria-negative university-clinic patients nonetheless carried an FM diagnosis 30724039
Referral diagnosis accurate in 34%; "disturbing inaccuracy, mostly... overdiagnosis" 12595620
Physician-diagnosed community FM has markedly lower symptom burden than criteria-positive FM (mean PSD 12.4 vs 18.4) 31777779

The two errors are not contradictory: clinicians appear to diagnose a different construct — influenced by sex, presentation style, and symptom count rather than pain distribution (Wolfe 2019, PMID 30724039) — so the diagnosed population and the criteria-defined population overlap only partially. Which population the evidence base (trials, cohorts, claims data) actually describes depends on the study's case ascertainment; see outcomes and measurement and epidemiology.

Open questions

  • Which criteria set best predicts treatment response and prognosis? All validation to date is against other criteria or clinical diagnosis, not against outcomes (Wolfe 2016, PMID 27916278; Salaffi 2020, PMID 32206792).
  • AAPT and ACR-2016 disagree at the margins (Salaffi 2020, PMID 32206792) — do the discordant patients differ biologically or prognostically?
  • Can a criteria set be designed whose case definition is stable over time within patients? ~25–30% of diagnosed patients drift below threshold on re-examination (Wolfe 2010, PMID 20461783; Galvez-Sánchez 2020, PMID 32340369).
  • What intervention actually changes clinician criteria use? The κ ≈ 0.3 clinic–criteria gap has persisted across settings and decades (Fitzcharles 2003, PMID 12595620; Wolfe 2019, PMID 30724039; Srinivasan 2019, PMID 31777779).
  • Should ICD-11 chronic-primary-pain coding (Treede 2019, PMID 30586067) be reconciled with ACR-lineage criteria for research use, and what is lost if administrative and research definitions continue to diverge?
  • history-and-nosology — why the definition kept changing; the continuum the thresholds cut.
  • epidemiology — prevalence and sex-ratio consequences of each criteria era.
  • outcomes-and-measurement — severity scales (FIQ/FIQR, PSD) beyond the diagnostic threshold.
  • biomarkers — what an objective diagnostic test would need to displace symptom criteria.
  • guidelines — which criteria the major guidelines endorse.

References

  1. Wolfe F, et al. The American College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum. 1990;33(2):160-72. PMID 2306288
  2. Wolfe F, et al. The American College of Rheumatology preliminary diagnostic criteria for fibromyalgia and measurement of symptom severity. Arthritis Care Res (Hoboken). 2010;62(5):600-10. PMID 20461783
  3. Wolfe F, et al. Fibromyalgia criteria and severity scales for clinical and epidemiological studies: a modification of the ACR Preliminary Diagnostic Criteria for Fibromyalgia. J Rheumatol. 2011;38(6):1113-22. PMID 21285161
  4. Wolfe F, et al. 2016 Revisions to the 2010/2011 fibromyalgia diagnostic criteria. Semin Arthritis Rheum. 2016;46(3):319-329. PMID 27916278
  5. Wolfe F, Egloff N, Häuser W. Widespread Pain and Low Widespread Pain Index Scores among Fibromyalgia-positive Cases Assessed with the 2010/2011 Fibromyalgia Criteria. J Rheumatol. 2016;43(9):1743-8. PMID 27370877
  6. Arnold LM, et al. AAPT Diagnostic Criteria for Fibromyalgia. J Pain. 2019;20(6):611-628. PMID 30453109
  7. Treede RD, et al. Chronic pain as a symptom or a disease: the IASP Classification of Chronic Pain for ICD-11. Pain. 2019;160(1):19-27. PMID 30586067
  8. Nicholas M, et al. The IASP classification of chronic pain for ICD-11: chronic primary pain. Pain. 2019;160(1):28-37. PMID 30586068
  9. Galvez-Sánchez CM, Reyes del Paso GA. Diagnostic Criteria for Fibromyalgia: Critical Review and Future Perspectives. J Clin Med. 2020;9(4):1219. PMID 32340369
  10. Salaffi F, et al. Diagnosis of fibromyalgia: comparison of the 2011/2016 ACR and AAPT criteria and validation of the modified Fibromyalgia Assessment Status. Rheumatology (Oxford). 2020;59(10):3042-3049. PMID 32206792
  11. Jones GT, et al. The prevalence of fibromyalgia in the general population: a comparison of the ACR 1990, 2010, and modified 2010 classification criteria. Arthritis Rheumatol. 2015;67(2):568-75. PMID 25323744
  12. Croft P, et al. Population study of tender point counts and pain as evidence of fibromyalgia. BMJ. 1994;309(6956):696-9. PMID 7950521
  13. Wolfe F, et al. The prevalence and characteristics of fibromyalgia in the general population. Arthritis Rheum. 1995;38(1):19-28. PMID 7818567
  14. White KP, et al. The London Fibromyalgia Epidemiology Study: the prevalence of fibromyalgia syndrome in London, Ontario. J Rheumatol. 1999;26(7):1570-6. PMID 10405947
  15. Heidari F, et al. Prevalence of fibromyalgia in general population and patients, a systematic review and meta-analysis. Rheumatol Int. 2017;37(9):1527-1539. PMID 28447207
  16. Wolfe F, et al. Fibromyalgia diagnosis and biased assessment: Sex, prevalence and bias. PLoS One. 2018;13(9):e0203755. PMID 30212526
  17. Wolfe F, et al. Fibromyalgia prevalence, somatic symptom reporting, and the dimensionality of polysymptomatic distress. Arthritis Care Res (Hoboken). 2013;65(5):777-85. PMID 23424058
  18. Wolfe F, et al. Diagnosis of Fibromyalgia: Disagreement Between Fibromyalgia Criteria and Clinician-Based Fibromyalgia Diagnosis in a University Clinic. Arthritis Care Res (Hoboken). 2019;71(3):343-351. PMID 30724039
  19. Srinivasan S, et al. The Problematic Nature of Fibromyalgia Diagnosis in the Community. ACR Open Rheumatol. 2019;1(1):43-51. PMID 31777779
  20. Fitzcharles MA, Boulos P. Inaccuracy in the diagnosis of fibromyalgia syndrome: analysis of referrals. Rheumatology (Oxford). 2003;42(2):263-7. PMID 12595620
  21. Walitt B, et al. The Prevalence and Characteristics of Fibromyalgia in the 2012 National Health Interview Survey. PLoS One. 2015;10(9):e0138024. PMID 26379048
  22. Vincent A, et al. Prevalence of fibromyalgia: a population-based study in Olmsted County, Minnesota. Arthritis Care Res (Hoboken). 2013;65(5):786-92. PMID 23203795
  23. Wolfe F. The relation between tender points and fibromyalgia symptom variables: evidence that fibromyalgia is not a discrete disorder in the clinic. Ann Rheum Dis. 1997;56(4):268-71. PMID 9166001
  24. Berwick R, Barker C, Goebel A. The diagnosis of fibromyalgia syndrome. Clin Med (Lond). 2022;22(6):570-574. PMID 36427885