Fibromyalgia revisited: do latent class analyses of symptom profiles in the general population confirm 2016 fibromyalgia diagnostic criteria?

Page created by Darrell Mendez
 
CONTINUE READING
Fibromyalgia revisited: do latent class analyses
                   of symptom profiles in the general population
                  confirm 2016 fibromyalgia diagnostic criteria?
               B. Schmalbach1, W. Häuser2,3, E. Brähler4, P. Henningsen2, F. Wolfe5

1
  Department of Medical Psychology          ABSTRACT                                    Introduction
and Medical Sociology, University           Objective. The definition of the 2016       The definition and content of fibromy-
Medical Centre of the Johannes              diagnostic criteria of fibromyalgia         algia (FM) syndrome have changed re-
Gutenberg University Mainz;
                                            (FM) syndrome and of FM severities          peatedly in the 110 years of its existence
2
  Department of Psychosomatic Medicine
and Psychotherapy, Technische               was based on studies with clinical sam-     (1). The most important change arose
Universität München;                        ples. We tested if somatic symptom pro-     in the 1990s by the American College
3
  Department of Internal Medicine 1,        files consistent with the symptom pat-      of Rheumatology (ACR) classification
Klinikum Saarbrücken;                       tern of the FM 2016 diagnostic criteria     criteria which defined FM by symp-
4
  Department of Psychosomatic Medicine      and of severities of FM can be found in     toms (chronic widespread pain [CWP])
and Psychotherapy, University Medical       the general population.                     and findings (tenderness at palpation of
Centre of the Johannes Gutenberg
                                            Methods. Somatic symptom burden was         muscles and tendons) (2). By 2010, a
University Mainz, Germany;
5
  National Data Bank for Rheumatic          measured by the Somatic Symptom Scale       second shift occurred that excluded ten-
Diseases, Wichita, USA.                     - 8 in 2,531 persons aged ≥14 years rep-    der points. These new criteria overcame
Bjarne Schmalbach, PhD*                     resentative for the general German pop-     the requirement for specialist medical
Winfried Häuser, MD*                        ulation. We used latent class analysis of   examinations. Some patient-reported
Elmar Brähler, PhD                          SSS-8 items to identify somatic symptom     non-musculoskeletal pain symptoms
Peter Henningsen, MD                        profiles. The profiles were described       (headache, abdominal pain) and psy-
Frederick Wolfe, MD                         by their association with age, gender,      chological symptoms (fatigue, cogni-
*These authors contributed equally.         self-reported disabling somatic disease,    tive problems, depression) were added
Please address correspondence to:           psychological symptom burden, illness       as minor diagnostic criteria (3). FM be-
Winfried Häuser,                            worries and self-perceived health.          came a symptom-based diagnosis that
Department of Psychosomatic                 Results. We identified five somatic         included multiple somatic and psycho-
Medicine and Psychotherapy,                 symptom profiles. The majority of the       logical symptoms. In the 2016 diagnos-
Technische Universität München,
Langerstrasse 3,
                                            population (40.9%) had a profile char-      tic criteria (4), the criteria of CWP was
81675 München, Germany.                     acterised by the absence of bothering       tightened compared to the 1990 classifi-
E-mail:                                     symptoms. 5.9% had a profile defined by     cation criteria (2) requiring pain sites in
whaeuser@klinikum-saarbruecken.de           “considerable bothering” back and ex-       at least four of five body regions.
Received on January 30, 2021; accepted      tremities pains, fatigue and sleep prob-    All studies defining the ACR 1990 clas-
in revised form on April 22, 2021.          lems. This symptom profile was associat-    sification (2), the 2010 ACR prelimi-
Clin Exp Rheumatol 2021; 39 (Suppl. 130):   ed with older age, self-reported somatic    nary diagnostic (3), the 2011 (5) and
S128-S136.                                  diseases, psychological symptom burden      the 2016 (4) criteria were conducted
© Copyright Clinical and                    and fair to poor general health. 63.2%      with people with various rheumatic dis-
Experimental Rheumatology 2021.             of persons meeting FM 2016 criteria         eases included in the US National Data
                                            belonged to this profile. 17.8% of the      Bank of Rheumatic Diseases (5). These
Key words: fibromyalgia, somatic            sample were characterised by little per-    selections might have led to consider-
symptoms, psychosomatic medicine,           turbation by multiple somatic symptoms      able bias in the identification of symp-
latent class analysis, representative       and good to fair general health. 36.8%      tom classes including higher symptom
survey                                      of persons meeting FM 2016 criteria be-     prevalence in the study population (6).
                                            longed to this profile.                     Previous studies have shown a lower
                                            Conclusion. Two somatic symptom             symptom burden of FM-cases in the
                                            profiles consistent with the 2016 FM        general population compared to the
                                            diagnostic criteria were identified in      ones of clinical settings (7). A symp-
                                            the general German population. These        tom profile in the general population
Competing interests: W. Häuser
has received honoraria for a CD with        symptom profiles differed in somatic        consistent with the one defined by 2016
medical hypnosis for fibromyalgia by        and psychological symptom burden            FM diagnostic criteria (4) would sup-
Hypnos publisher. The other authors         and general health supporting the dis-      port its use in making a clinical diag-
have declared no competing interests.       tinction of FM severities.                  nosis of FM.

S-128                                                                                    Clinical and Experimental Rheumatology 2021
Fibromyalgia revisited / B. Schmalbach et al.

Studies with FM patients in clinical          258 living areas was randomly select-      musculoskeletal and general symptoms
care have demonstrated that FM is a           ed from a non-overlapping stratum of       during the last four weeks. Thus, the
heterogenous condition with regards           all area units: 210 areas were sampled     BDS-25 asks for negative appraisal
to the amount of somatic and psycho-          from Western Germany and 48 areas          of somatic symptoms, but not explic-
logical symptom burden, disability            from Eastern Germany. The random se-       itly for psycho-behavioural symptoms.
and comorbid diseases (7). Therefore,         lection of households was implemented      Each symptom can be scored on Likert
a distinction of severities of FM has         in the second step. Finally, one person    Scales from 0 (bothering not at all) to 4
been suggested, e.g. based on clinical        matching the inclusion criteria was ran-   (bothering a lot) (17). We used the vali-
criteria such as the extent of disability     domly selected from each household.        dated German version of the BDS-25
and /or symptom scores, e.g. of the Fi-       Sociodemographic data were collected       (13).
bromyalgia Impact Questionnaire (9)           by trained interviewers face-to-face. In
or the Patient Health Questionnaire           addition, participants completed a bat-    The Patient Health Questionnaire-4
(PHQ) 15 (7, 10). The studies on se-          tery of self-report questionnaires. The    (PHQ-4) was used to assess psycho-
verities of FM were conducted with            interviewers waited until the partici-     logical symptom burden. On Likert
clinical populations (7, 9) and require       pants answered all questionnaires, and     scales from 0 (not at all) to 3 (nearly
testing in the general population, too.       offered help in case of ambiguities.       every day), respondents rate how often
The latent class approach (LCA) has                                                      they have been bothered by little inter-
proven to be a powerful analytical ap-        Instruments                                est or pleasure in doing things and feel-
proach for diagnosing symptom pat-            Demographics: Age, gender, family          ing down, depressed or hopeless, feel-
terns in the general population (6, 11).      status, educational level, and net fam-    ing nervous, anxious or on edge, or not
Previous studies have found “healthy”,        ily income per month were assessed by      being able to stop or control worrying
specific symptom and multi-symptom            a standardised questionnaire used pre-     over the last two weeks. The total score
profiles in the general population (6, 12).   viously in German health surveys (15).     ranges from 0 to 12. Scores are rated
We studied profiles of somatic symp-                                                     as normal (0-2), mild (3-5), moderate
toms by LCA in the general population         The Somatic Symptom Scale-8 (SSS-8)        (6-8), and severe (9-12) psychologi-
in order to assess:                           is the short form of the Patient Health    cal symptom burden (18). We used the
• If profiles which are consistent with       Questionnaire PHQ-15 (10) and asks         validated German version (19).
   the symptom pattern of the 2016 FM         for eight somatic symptoms during the
   criteria can be found;                     past 7 days (stomach or bowel prob-        The Michigan Body Map (MBM) is a
• If these somatic symptom profiles           lems; back pain; pain in arms, legs, or    graphic mannequin for the assessment
   differ in the amount of somatic and        joints; headaches; chest pain or short-    of chronic pain. It offers 35 checkbox
   psychological symptom burden and           ness of breath; dizziness; feeling tired   body areas covering all 19 areas from
   in general health supporting the con-      or having low energy; trouble sleep-       the Widespread Pain Index (WPI) (4)
   cept of severities of FM.                  ing). Symptoms are scored on Likert        plus 16 other pain sites (20). Subjects
                                              Scales from 0 (not bothered at all) to     are asked to mark all areas where they
Materials and methods                         4 (bothered very much) (16). We re-        have felt persistent or recurrent pain
Design and subjects                           coded as follows: 0=0; 1 and 2=1; 3        present for the last three months or
The study is part of a larger cross-sec-      and 4=2 because a) 4-scores were very      longer. We used the German version of
tional survey on physical and mental          rare (
Fibromyalgia revisited / B. Schmalbach et al.

and pains?”; “Do you find that you          Table I. Fit criteria for latent class models with 1-10 components.
are bothered by many different symp-
                                            # of latent clusters     LL     df    BIC                                        CAIC
toms?”) because they capture symptom
quantity rather than illness conviction     1                    -14016.07 2363 28156.53                                   28172.53
(WI-5). We used the validated German        2                    -11944.48 2346 24145.52                                   24178.52
version (22).                               3                    -11567.86 2329 23524.44                                   23574.44
                                            4                    -11408.47 2312 23337.82                                   23404.82
                                            5                    -11302.05 2295 23257.16                                   23341.16
The self-administered comorbidity           6                    -11239.02 2278 23263.26                                   23364.26
questionnaire (SCQ) is a validated self-    7                    -11189.62 2261 23296.63                                   23414.63
rating instrument in clinical and health    8                    -11160.14 2244 23369.82                                   23504.82
services research to assess common dis-     9                    -11131.24 2227 23444.19                                   23596.19
                                            10                   -11103.44 2210 23520.76                                   23689.76
eases. It asks about the presence, treat-
ment, and functional limitations of thir-   LL: log likelihood; df: degrees of freedom; BIC: Bayesian information criterion; CAIC: consistent
teen common diseases (heart disease;        Akaike information criterion.
high blood pressure; lung disease; dia-     Model with best fit is printed in bold.
betes; ulcer or other stomach disease;
kidney disease; liver disease; anaemia      of FM (5) were recorded as follows:        should only be related to a negligible
or other blood disease; cancer; rheuma-     Fatigue and sleeping problems when         degree (similar to the error terms in
toid arthritis. We substituted osteoar-     reported by SSS-8 items 7 and 8 as at      confirmatory factor analysis). The pri-
thritis by pancreas disease and low back    least “somewhat bothering”; cognitive      mary measures of interest in the LCA
pain by inflammatory bowel disease to       problems when reported BDS item 24         are then the Bayesian Information Cri-
a) avoid overlap with pain sites assessed   was rated as at least “somewhat both-      terion (BIC) and the Consistent Akaike
by the WPI b) to increase the number of     ering; headache when BDS 25 item 23        Information Criterion (CAIC). In line
somatic diseases which might contrib-       was rated as at least “somewhat bother-    with the typical recommendation, we
ute to somatic symptoms captured by         ing”; pain or cramps in the lower ab-      also ran bootstrapped (i=100) LCAs to
the BDS 25 checklist. Three subscales       domen when BDS 25 item 8 was rated         ascertain the initial results (29).
(present disease, present disease with      as at least “somewhat bothering”; de-      Subsequently, we compared the result-
drug treatment, present disease with as-    pression when PHQ 4 item 2 was rated       ing classes with regard to their sociode-
sociated disability) are available (23).    as at least “at several days”. Scores of   mographic characteristics as well as
We used the subscale “present disease       ACRSSS range from 0-12. Polysymp-          their descriptive statistics for both, the
with associated disability” (range 0-12)    tomatic Distress Scale score (PDS) is      SSS-8 items and scales, and other relat-
of the validated German version (24).       the sum of the WPI and SSS. Scores         ed measures and variables. For metric
                                            range from 0-31. We used these PPS         variables, we used univariate ANOVAs
The Short Health Survey 12 (SF-12)          severity categories: none (0-3), mild (4-  and report the η² effect size, which ac-
General Health concept was used to as-      7), moderate (8-11), severe (12-19), and   cording to Cohen (30, 31) should be
sess self-perceived health on a 5-point     very severe (20-31). FM was defined as     interpreted as signifying small, moder-
Likert scale. Subjects were straight        1) WPI ≥ 7 and SSS ≥5 OR WPI 4-6           ate, and large effects for values exceed-
asked “In general, would you say            and SSS ≥9, AND 2) pain in 4 of 5 body     ing 0.01, 0.09, 0.25. In addition, we
your health is….”. The answers are as       regions (4).                               conducted post-hoc group comparisons
follows: 1=excellent; 2=very good;          Self-reported FM: Participants were        using Holm-corrected t-tests.
3=good; 4= moderate; 5=poor self-per-       asked if they have been diagnosed with     For count variables, we utilised the χ²
ceived health’ (25). We used the vali-      FM by a physician in the past.             test to investigate whether there are
dated German version (26).                                                             significant between-group differences.
                                            Statistical analyses                       We report Cramer’s V as an effect size
Case definitions of FM                      All analyses were conducted in R. Only for these analyses, which is defined as
We used two case definitions of FM          complete cases were included: n = the root of χ² divided by the product
because: a) there is no gold standard       2,379. We then utilised poLCA (28) to of the sample size and dfA, where dfA
for FM diagnosis; b) considerable disa-     perform a latent class analysis with the is the length of the smaller of the two
greement between clinical diagnosis         items of the SSS-8. This technique al- dimensions minus 1 (30). In addition,
and criteria-based diagnosis of fibro-      lows for the clustering of observations we applied the correction suggested by
myalgia was found in an US rheumatol-       with regard to a given number of char- Bergsma (32) which delivers a more
ogy clinic (27).                            acteristics. Prior to computing the LCA accurate estimate of the population
2016 criteria: MBM pain sites and pain      model, we tested the assumption of con- effect size. For dfA = 1, Cramer’s V is
regions were counted (excluding jaw,        ditional independence, which was met interpreted analogously to the Pearson
chest, and abdominal pain). In addi-        with only minor deviations for the indi- r coefficient with small, moderate, and
tion, the Somatic Severity Scale (SSS)      cators. This means that, after accounting large effects being identified by values
criteria of the 2011 diagnostic criteria    for classification the indicator variables of 0.10, 0.30, and 0.50, respectively.

S-130                                                                                           Clinical and Experimental Rheumatology 2021
Fibromyalgia revisited / B. Schmalbach et al.

Table II. Relative response frequencies for the SSS-8 items.

		                                                 Total sample    1      2      3        4        5     Group comparison

Item                                 n                2379        973   435     424      407     140
                                     %                 100         41    18      18       17       6
Stomach or bowel problems            Not at all         72         95    63      76       34      28     χ²(8) = 140.47, p
Fibromyalgia revisited / B. Schmalbach et al.

Table III. Sociodemographic characteristics for the overall sample and the latent classes.

  Total 		                                                    Profile Profile   Profile   Profile    Profile   Group-comparison
			                                                             1       2         3         4          5

n                                    2379             -         973     435      424         407      140
%                                       -           100          41      18       18          17        6
Sex								                                                                                                    χ²(4) = 7.68, p=0.104, V = 0.086
 Male   1116 47 54 36 53 39 35
 Female 1263 53 46 64 47 61 65
Age								                                                                                                    F(4, 2374) = 74.70, p
Fibromyalgia revisited / B. Schmalbach et al.

Table IV. Mean values and standard deviations SSS-8 scale score and external criteria.

  Total                                           Profile       Profile       Profile        Profile       Profile Group comparison             Pairwise
		                                                  1             2             3              4             5		                              comparisons*

SSS-8 (0-16) (Mean, SD) 3.05 (3.06) 0.39 (0.55) 3.57 (1.24) 2.84 (0.94) 6.57 (1.46) 10.33 (2.07) F(4, 2374) = 4135.52, 1 < 3 < 2 < 4 < 5
							                                                                                          p
Fibromyalgia revisited / B. Schmalbach et al.

Table V. Relative response frequencies for the SSS-8 scale score and external criteria.

  Total                                                Profile    Profile   Profile    Profile    Profile         Group comparison
		                                                       1          2         3          4          5

%                                         100            41         18        18          17         6
SSS-8 (0-15)
  0-3                                       64    100 54 75  0  0 χ²(16) = 588.38, p15                                        0      0  0  0  0  1
WPI (0-19)
 0                                          72    96 80 63 31 20 χ²(12) = 166.61, p5                                          6     0  0  2 18 39
PSD (0-31)
 0-3                                        72    100 69 83 23  1 χ²(16) = 409.71, p
Fibromyalgia revisited / B. Schmalbach et al.

six symptom classes by including 15                   4. WOLFE F, CLAUW D, FITZCHARLES MA et                   J Psychosom Res 2015; 78: 536-45.
                                                         al.: Revisions to the 2010/2011 fibromyalgia      18. KROENKE K, SPITZER RL, WILLIAMS JB,
symptoms in LCA (12).
                                                         diagnostic criteria. Semin Arthritis Rheum            LOWE B: An ultra-brief screening scale for
The identified profiles may describe                     2016; 46: 319-29.                                     anxiety and depression: the PHQ-4. Psycho-
clinical recognisable symptom patterns                5. WOLFE F, CLAUW D, FITZCHARLES MA et                   somatics 2009; 50: 613-21.
from well-defined somatic diseases,                      al.: Fibromyalgia criteria and severity scales    19. LÖWE B, WAHL I, ROSE M et al.: A 4-item
                                                         for clinical and epidemiological studies: A           measure of depression and anxiety: valida-
symptom-based diagnoses and short-                       modification of the acr preliminary diag-             tion and standardization of the Patient Health
term unspecific symptoms. The symp-                      nostic criteria for fibromyalgia. J Rheumatol         Questionnaire-4 (PHQ-4) in the general pop-
tom profiles should not be interpreted                   2011; 38: 1113-22.                                    ulation. J Affect Disord 2010; 122: 86-95.
as definitions of diseases or new clas-               6. ELIASEN M, JØRGENSEN T, SCHRÖDER A et             20. BRUMMETT CM, BAKSHI RR, GOESLING J et
                                                         al.: Somatic symptom profiles in the general          al.: Preliminary validation of the Michigan
sifications of disorders but as complex-                 population: a latent class analysis in a Danish       Body Map. Pain 2016; 157: 1205-12.
es of considerably bothering somatic                     population-based health survey. Clin Epide-       21. FINK P, EWALD H, JENSEN J et al.: Screen-
symptoms of different origin (6).                        miol 2017; 9: 421-33.                                 ing for somatization and hypochondriasis in
                                                      7. HÄUSER W, BRÄHLER E, WOLFE F, HEN-                    primary care and neurological in-patients: a
                                                         NINGSEN P: Patient Health Questionnaire               seven-item scale for hypochondriasis and so-
Conclusions                                              15 as a generic measure of severity in fibro-         matization. J Psychosom Res 1999; 46: 261-
Latent class analyses separated the                      myalgia syndrome: surveys with patients of            73.
continuum of somatic symptom burden                      three different settings. J Psychosom Res         22. HINZ A, RIEF W, BRÄHLER E: Hypochondrie
in the general German population into                    2014; 76: 307-11.                                     in der Allgemeinbevölkerung: Teststatistische
                                                      8. FINK P, EWALD H, JENSEN J, SØRENSEN L,                Prüfung und Normierung des Whiteley-Index
clinically meaningful profiles.                          ENGBERG M, HOLM M: Screening for soma-                [Hypochondria in the general population:
FM appears to be an invisible experi-                    tization and hypochondriasis in primary care          Test statistical assessment and standardiza-
ence without any visible biomarker                       and neurological in-patients: a seven-item            tion of the Whiteley-Index.] Diagnostica
                                                         scale for hypochondriasis and somatization.           2003; 49: 34-42.
to exhibit to healthcare professionals
                                                         J Psychosom Res 2006; 60: 137-43.                 23. SANGHA O, STUCKI G, LIANG MH, FOSSEL
(38). We identified a somatic symp-                   9. SCHAEFER C, CHANDRAN A, HUFSTADER M                   AH, KATZ JN: The Self-Administered Comor-
tom profile consistent with the major                    et al.: The comparative burden of mild, mod-          bidity Questionnaire: a new method to assess
symptoms of FM according to the 2016                     erate and severe fibromyalgia: results from a         comorbidity for clinical and health services
                                                         cross-sectional survey in the United States.          research. Arthritis Care Res 2003; 49: 156-
diagnostic criteria (widespread pain,                    Health Qual Life Outcomes 2011; 9: 71.                63.
fatigue and sleep problems) in the gen-              10. KROENKE K, SPITZER RL, WILLIAMS JB:               24. STREIBELT M, SCHMIDT C, BRÜNGER M,
eral population supporting a specific                    The PHQ-15: validity of a new measure for             SPYRA K: Comorbidity from the patient per-
diagnostic code for FM in the upcom-                     evaluating the severity of somatic symptoms.          spective - does it work? Validity of a ques-
                                                         Psychosom Med 2002; 64: 258-66.                       tionnaire on self-estimation of comorbidity
ing International Classification of Dis-             11. KATO K, SULLIVAN PF, PEDERSEN NL:                     (SCQ-D). Orthopade 2012; 41: 303-10.
eases (ICD-11) of the World Health                       Latent class analysis of functional somatic       25. WARE J, KOSINSKI M, KELLER SD: A 12-Item
Organisation (39).                                       symptoms in a population-based sample of              Short-Form Health Survey: construction of
As in clinical populations, FM 2016                      twins. J Psychosom Res 2010; 68: 447-53.              scales and preliminary tests of reliability and
                                                     12. WIRTZ MA, MORFELD M, BRÄHLER E, HINZ                  validity. Medical Care 1996; 34: 220-33.
cases in the general population differed                 A, GLAESMER H: Association of physical            26. WIRTZ MA, MORFELD M, GLAESMER H,
in the amount of additional and psy-                     morbidity and health-related quality of life          BRÄHLER E: Normierung des SF-12 Ver-
chological symptom burden, subjec-                       in a representative sample of older German            sion 2.0 zur Messung der gesundheitsbezo-
tive health and comorbid somatic dis-                    people. Eur J Health Psychol 2019; 25: 140-           genen Lebensqualität in einer deutschen bev-
                                                         51.                                                   ölkerungsrepräsentativen Stichprobe. [Stand-
eases. The heterogeneity of FM should                13. SCHMALBACH B, ROENNEBERG C,                           ardization of the SF-12 for assessment of
be addressed by targeted and graduated                   HAUSTEINER-WIEHLE C: Validation of the                health-related quality of life in a representa-
management approaches (35, 36).                          German version of the Bodily Distress Syn-            tive German population sample.] Diagnostica
                                                         drome 25 checklist in a representative Ger-           2018; 64: 215-26.
                                                         man population sample. J Psychosom Res            27. WOLFE F, SCHMUKLER J, JAMAL S et al.:
Acknowledgements                                         2020; 132; 109991.                                    Diagnosis of fibromyalgia: disagreement
Elmar Brähler was previously affili-                 14. HÄUSER W, HENNINGSEN P, BRÄHLER E,                    between fibromyalgia criteria and clinician-
ated with the University of Leipzig at                   SCHMALBACH B, WOLFE F: Prevalence and                 based fibromyalgia diagnosis in a university
                                                         overlap of somatic symptom disorder, bod-             clinic. Arthritis Care Res 2019; 71: 343-51.
the time of ethics approval.                             ily distress syndrome and fibromyalgia syn-       28. LINZER DA, LEWIS JB: poLCA: An R Pack-
                                                         drome in the German general population: A             age for Polytomous Variable Latent Class
References                                               cross sectional study. J Psychosom Res 2020;          Analysis. J Statistical Software 2011; 42:
 1. WOLFE F, WALITT B: Culture, science and              133: 110111.                                          1-29.
    the changing nature of fibromyalgia. Nat Rev     15. WOLFE F, BRÄHLER E, HINZ A, HÄUSER W:             29. AITKIN M, ANDERSON D, HINDE J: Statistical
    Rheumatol 2013; 9: 751-5.                            Fibromyalgia prevalence, somatic symptom              modelling of data on teaching styles. J R Stat
 2. WOLFE F, SMYTHE HA, YUNUS MB et al.:                 reporting, and the dimensionality of poly-            Soc Series A General 1981; 144: 419-48.
    The American College of Rheumatology                 symptomatic distress: results from a survey       30. COHEN J: Quantitative methods in psycholo-
    1990 Criteria for the Classification of Fibro-       of the general population. Arthritis Care Res         gy: A power primer. Psychol Bull 1992; 112:
    myalgia. Report of the Multicenter Criteria          2013; 65: 777-85.                                     1155-9.
    Committee. Arthritis Rheum 1990; 33: 160-        16. GIERK B, KOHLMANN S, KROENKE K et al.:            31. COHEN J: Statistical power analysis for the
    72.                                                  The somatic symptom scale-8 (SSS-8): a                behavioral sciences. (2nd ed.). Mahwah, NJ,
 3. WOLFE F, CLAUW D, FITZCHARLES MA et                  brief measure of somatic symptom burden.              Lawrence Erlbaum, 1988.
    al.: The American College of Rheumatology            JAMA Intern Med 2014; 174: 399-407.               32. BERGSMA W: A bias-correction for Cramér’s
    Preliminary Diagnostic Criteria for Fibromy-     17. BUDTZ-LILLY A, FINK P, ØRNBØL E et al.:               V and Tschuprow’s T. J Korean Stat Soc
    algia and measurement of symptom severity.           A new questionnaire to identify bodily dis-           2013; 42: 323-8.
    Arthritis Care Res 2010; 62: 600-10.                 tress in primary care: The ‘BDS checklist’.       33. HÄUSER W, SCHMUTZER G, BRÄHLER E,

Clinical and Experimental Rheumatology 2021                                                                                                           S-135
Fibromyalgia revisited / B. Schmalbach et al.

   GLAESMER E: A cluster within the continuum      35. HÄUSER W, PERROT S, CLAUW DJ,                      tion in counseling and clinical psychology:
    of biopsychosocial distress can be labeled        FITZCHARLES MA: Unravelling fibromy-                A meta-analysis. Educ Psychol Measurement
    “fibromyalgia syndrome”- evidence from a           algia-steps toward individualized manage-          2009; 69: 389-403.
    representative German population survey.           ment. J Pain 2018; 19: 125-34.                 38. PERROT S: Fibromyalgia: A misconnection
    J Rheumatol 2009; 36: 2806-12.                 36. MacFARLANE GJ, KRONISCH C, DEAN LE et              in a multiconnected world? Eur J Pain 2019;
34. KATZ RS, WOLFE F, MICHAUD K: Fibro-                al.: EULAR revised recommendations for             23: 866-73.
    myalgia diagnosis: a comparison of clinical,       the management of fibromyalgia. Ann Rheum      39. NICHOLAS M, VLAYEN JWS, RIEF W et al.:
    survey, and American College of Rheumatol-         Dis 2017; 76: 318-28.                              The IASP classification of chronic pain for
    ogy criteria. Arthritis Rheum 2006; 54: 169-   37. HORN PSV, GREEN KE, MARTINUSSEN M:                 ICD-11: chronic primary pain. Pain 2019;
    76.                                                Survey response rates and survey administra-       160: 28-37.

S-136                                                                                                  Clinical and Experimental Rheumatology 2021
You can also read