Comorbid Depression and Anxiety in Fibromyalgia Syndrome: Relationship to
Somatic and Psychosocial Variables
KATI THIEME,PHD, DENNIS C. TURK,PHD, AND HERTA FLOR,PHD
Objective: The prevalence as well as predictors of psychiatric disorders (Diagnostic and Statistical Manual of Mental Disorders,
4th edition [DSM-IV] axis I and II) in patients with fibromyalgia syndrome (FMS) was evaluated. Method: One-hundred fifteen
patients with FMS participated in the Structured Clinical Interview for DSM-IV to assess current mental disorders. In addition,
patients completed standardized questionnaires regarding pain, pain impact, anxiety, depression, posttraumatic stress disorder-like
symptoms, and sexual and physical abuse. Results: Patients were grouped into one of three psychosocial subgroups based on
responses to the Multidimensional Pain Inventory (MPI)–Dysfunctional (DYS), Interpersonally Distressed (ID), and Adaptive
Copers (AC). Axis I diagnoses were present in 74.8% of the participants overall with the DYS subgroup mainly reporting anxiety
and the ID group mood disorders. The AC group showed little comorbidity. Axis II diagnoses were present in only 8.7% of the FMS
sample. Conclusion: These results suggest that FMS is not a homogeneous diagnosis, but shows varying proportions of comorbid
anxiety and depression dependent on psychosocial characteristics of the patients. The results demonstrate the importance of not
treating patients with FMS as a homogeneous group. Assessment should not only examine the presence of widespread pain and the
number of tender points, but also the presence of affective distress. Treatment should focus both on physical and emotional
dysfunction. Key words: fibromyalgia, comorbidity, DSM-IV, SCID, MPI subgroups.
FMS = fibromyalgia syndrome; MPI = Multidimensional Pain
Inventory; DYS = dysfunctional; ID = interpersonally distressed;
AC = adaptive copers; SCID = Structured Clinical Interview for
DSM-IV; CES-D = Center for Epidemiologic Studies Depression
Scale; STAI-T = State-Trait Anxiety Inventory–Trait scale;
PRSS = Pain-Related Self-Statements Scale.
INTRODUCTION
F
ibromyalgia syndrome (FMS) is characterized by wide-
spread pain, hypersensitivity to palpation at specific body
locations (tender points) (1), and a range of comorbid physical
symptoms and functional limitations, including persistent fa-
tigue, sleep disturbance, feelings of stiffness, headaches, and
irritable bowel disorders (1). Patients also report cognitive
impairment and general malaise sometimes referred to as
“fibro fog” (2).
As is often the case in medicine, in the absence of specific
physical findings that provide an adequate explanation for
symptoms, a number of authors have suggested that FMS is
primarily a psychogenic disorder (3–5). Depression is reported
to be particularly prevalent in FMS, leading some to suggest
that it is a depressive spectrum disorder (6 – 8). Examination
of the prevalence of depression across studies reveals a wide
diversity with current depressive disorders ranging from
28.6% to 70% across studies (9,10). The large differences in
prevalence may be explained by the variation in methods for
assessing depression, different definitions used for depression,
and sampling bias (eg, patients referred to a psychiatrist for
evaluation); reliance on treatment-seekers to establish preva-
lence (treatment seekers are likely to be the most distressed);
or a combination of these factors. Less attention has been
given to the prevalence of anxiety in FMS. However, Kurtze
(11) found support for the hypothesis that depression and
anxiety are independently associated with the severity of pain
symptoms in FMS. Thus, not all patients diagnosed with FMS
experience the same or necessarily any emotional disorders.
Turk and Flor (12) suggested that FMS is likely comprised
of a heterogenous group of people who may differ on impor-
tant variables such as mood, adaptation to symptoms, as well
as presenting symptoms. Turk et al. (13,14) demonstrated that
patients with FMS could be classified into psychosocial sub-
groups based on the scores of the West Haven-Yale Multidi-
mensional Pain Inventory (MPI) (15). In a cluster analysis of
responses to the MPI, these investigators identified three psy-
chosocial subgroups that were characterized by different lev-
els of pain intensity, interference, and affective distress as well
as the reactions of significant others. One group, labeled
dysfunctional (DYS), exhibited the highest level of pain,
emotional distress, and disability. A second group, termed
interpersonally distressed (ID), reported significantly lower
levels of pain, disability, and marital satisfaction than the
other two subgroups. The significant others of ID patients
showed a higher level of negative responses to the patients’
expressions of pain. The third group, adaptive copers (AC),
showed low pain intensity, emotional distress, and interfer-
ence of pain with daily lives and activities. MPI-based sub-
groups have been replicated in several studies performed in
several Western countries (eg, The Netherlands (17), Ger-
many (18), Sweden (19)). These subgroups have been identi-
fied in diverse chronic pain syndromes (eg, headache, back
pain, temporomandibular disorders) (16). The percentages of
patients classified within each group do, however, vary across
diagnoses. These patterns appear to exist independent of med-
ical diagnosis. The subgroups identified are primarily descrip-
tive; there is no information available to clarify what factors,
if any, predispose people to respond to chronic pain in one of
the three patterns identified. Moreover, to date, there have
been no studies that we are aware of that have attempted to
clarify the extent and nature of psychiatric disorders in the
subgroups identified.
The MPI is a general screening instrument and, although it
contains a scale labeled “affective distress,” it does not di-
From the Department of Anesthesiology, University of Washington, Seat-
tle, Washington (K.T., D.C.T.); and the Department of Neuropsychology
(K.T., H.F.), University of Heidelberg, Central Institute of Mental Health
Mannheim, Mannheim, Germany.
Address correspondence and reprint requests to Kati Thieme, PhD, Department
of Anesthesiology at the University of Washington, 1959 NE Pacific Street, Box
356540, Seattle, WA 98195-6540. E-mail: thiemek@u.washington.edu
Received for publication March 10, 2004; revision received June 18, 2004.
DOI: 10.1097/01.psy.0000146329.63158.40
837 Psychosomatic Medicine 66:837– 844 (2004)
0033-3174/04/6606-0837
Copyright © 2004 by the American Psychosomatic Society