Journal of Anxiety Disorders 30 (2015) 17

Contents lists available at ScienceDirect

Journal of Anxiety Disorders

Effectiveness of cognitive-behavioral group therapy for patients with

hypochondriasis (health anxiety)
Florian Weck a, , Maria Gropalis b , Wolfgang Hiller c , Gaby Bleichhardt d
a
Department of Clinical Psychology, Psychotherapy, and Experimental Psychopathology, Johannes Gutenberg University of Mainz, Wallstrae 3,
D-55122 Mainz, Germany
b
Department of Psychology, Johannes Gutenberg University of Mainz, Wallstrae 3, D-55122 Mainz, Germany
c
Department of Clinical Psychology and Psychotherapy, Johannes Gutenberg University of Mainz, Wallstrae 3, D-55122 Mainz, Germany
d
Department of Clinical Psychology and Psychotherapy, Philipps University Marburg, Gutenbergstrae 18, D-35037 Marburg, Germany

a r t i c l e

i n f o

Article history:
Received 29 October 2014
Received in revised form
23 December 2014
Accepted 25 December 2014
Available online 3 January 2015
Keywords:
Cognitive-behavioral therapy
Group therapy
Hypochondriasis
Health anxiety
Predictors

a b s t r a c t
Cognitive behavioral therapy (CBT) has been shown to be highly effective in the treatment of health anxiety. However, little is known about the effectiveness of group CBT in the treatment of health anxiety.
The current study is the largest study that has investigated the effectiveness of combined individual and
group CBT for patients with the diagnosis of hypochondriasis (N = 80). Therapy outcomes were evaluated
by several questionnaires. Patients showed a large improvement on these primary outcome measures
both post-treatment (Cohens d = 0.821.08) and at a 12-month follow-up (Cohens d = 1.091.41). Measures of general psychopathology and somatic symptoms showed signicant improvements, with small
to medium effect sizes. Patients with more elevated hypochondriacal characteristics at therapy intake
showed a larger therapy improvement, accounting for 78% of the variance in therapy outcome. CBT
group therapy has therefore been shown to be an appropriate and cost-effective treatment for health
anxiety.
2015 Elsevier Ltd. All rights reserved.

1. Introduction
Hypochondriasis (health anxiety) is characterized by the preoccupation with fears of having or the idea that one has a
serious disease (American Psychiatric Association, 2000). In general medical samples, the weighted prevalence of the diagnosis of
hypochondriasis is 2.95% (Weck, Richtberg, & Neng, 2014). Health
anxiety is also frequently reported in the general population (e.g.,
Bleichhardt & Hiller, 2007). Health anxiety is a disturbing and persistent condition, which is associated with high costs for the health
care system (Fink, rnbl, & Christensen, 2010).
Cognitive-behavioral therapy (CBT) has been shown to be highly
effective in the treatment of health anxiety. In a recent metaanalysis that considered the treatment of 1081 patients, high
effect sizes (Hedges g = 0.95) were found for CBT at post-treatment
(Olatunji et al., 2014). However, effect sizes at follow-up were
small (Hedges g = 0.35). CBT produces improvements in anxiety,

Corresponding author. Tel.: +49 6131 39 39215.

E-mail addresses: weck@uni-mainz.de (F. Weck), gropalis@uni-mainz.de
(M. Gropalis), hiller@uni-mainz.de (W. Hiller), bleichha@staff.uni-marburg.de
(G. Bleichhardt).
http://dx.doi.org/10.1016/j.janxdis.2014.12.012
0887-6185/ 2015 Elsevier Ltd. All rights reserved.

depression, and somatic symptoms, which often co-occur with

health anxiety (Olatunji et al., 2014; Thomson & Page, 2007).
CBT has demonstrated its superiority in comparison to psychodynamic psychotherapy (Srensen, Birket-Smith, Wattar, Buemann, &
Salkovskis, 2011) and can be delivered effectively via the Internet
(Hedman et al., 2011, 2014).
In contrast to the growing empirical evidence about individual
psychotherapy for health anxiety, there are only eight clinical studies, which treated altogether 192 patients with hypochondriasis in
group settings (Avia et al., 1996; Bleichhardt, Timmer, & Rief, 2005;
Bouman, 2002; Buwalda, Bouman, & van Duijn, 2006; Eilenberg,
Kronstrand, Fink, & Frostholm, 2013; Hedman et al., 2010; Stern
& Fernandez, 1991; Wattar et al., 2005). Sample sizes ranges from
6 (Stern & Fernandez, 1991) to 34 participants (Eilenberg et al.,
2013). Most studies used a CBT approach (Stern & Fernandez, 1991;
Bleichhardt et al., 2005; Wattar et al., 2005; Hedman et al., 2010),
three studies used psychoeducation or problem solving (Avia et al.,
1996; Bouman, 2002; Buwalda, Bouman, & van Duijn, 2007), and
one used acceptance and commitment therapy (ACT; Eilenberg
et al., 2013). Treatments included 6 (e.g., Bouman, 2002) to 23
(Wattar et al., 2005) sessions. Only two studies included a control
condition. One smaller study (N = 17) used a waiting-list control
group and found CBT signicantly superior to the waiting-list

F. Weck et al. / Journal of Anxiety Disorders 30 (2015) 17

(Avia et al., 1996). Buwalda et al. compared a psychoeducational

treatment to problem solving and found no signicant differences
between the two treatments. The prepost effect sizes (Hedgess
g) for group CBT were large (g = 1.032.38). Moreover, psychoeducation (g = 1.011.21), problem solving (g = 0.54), and ACT (g = 1.01)
demonstrated their effectiveness in the group setting, by medium
to high prepost effect sizes. Five studies reported a 6-month
follow-up (Bouman, 2002; Buwalda et al., 2007; Eilenberg et al.,
2013; Hedman et al., 2010; Stern & Fernandez, 1991) and three
studies a 12-month follow-up (Avia et al., 1996; Bleichhardt et al.,
2005; Wattar et al., 2005). Pre-follow-up effect sizes ranged from
g = 0.28 (Wattar et al., 2005) to g = 1.72 (Hedman et al., 2010).
Accordingly, six studies reported large (g > .80) pre-follow-up effect
sizes (Avia et al., 1996; Bouman, 2002; Bleichhardt et al., 2005;
Buwalda et al., 2007; Eilenberg et al., 2013; Hedman et al., 2010).
Group therapy seems to be an appropriate and cost-effective treatment approach for patients with health anxiety. However, the
empirical evidence for group therapy is limited by the number of
studies and their small sample sizes. Moreover, most of the studies have considered only a 6-month follow-up period. This may be
too short to be useful for the evaluation of long-term effects of the
treatments.
The empirical evidence is also limited regarding reliable predictors of treatment outcomes in psychotherapy for health anxiety. For
example, several studies found that hypochondriacal symptoms,
anxiety symptoms, and somatic symptoms at intake are related
to therapy outcome (Buwalda et al., 2007; Olde Hartman et al.,
2009; Nakao, Shinozaki, Ahern, & Barsky, 2011; Olatunji et al.,
2014). However, these ndings are inconsistent; some studies have
found that higher scores at intake lead to a better therapy outcome
(Nakao et al., 2011; Olatunji et al., 2014), while other studies have
found that higher scores at intake lead to a worse therapy outcome
(Buwalda et al., 2007; Olde Hartman et al., 2009). Therefore, further
research on the possible predictors of treatment outcome in health
anxiety is necessary.
The aim of the current study was to investigate the effectiveness
of a combined group and individual CBT in a large sample of patients
with hypochondriasis. In order to study the long-term effects of
the treatment, we decided upon a 12-month follow-up period. We
hypothesized that group CBT leads to a large improvement on primary hypochondriacal measures (Hypothesis 1). We also expected
to nd maintenance of those effects at the 12-month follow-up
(Hypothesis 2). Moreover, we hypothesized a signicant reduction
of general psychopathology and somatic symptoms during treatment time (Hypothesis 3). We hypothesized that the severity of
patients hypochondriacal symptoms at intake is a signicant predictor for therapy outcome (Hypothesis 4).
2. Method
2.1. Study design
The study was conducted at the outpatient unit of the Department of Clinical Psychology and Psychotherapy at the University
of Mainz (Germany). The outpatient unit offers ambulatory psychotherapy for patients with mental disorders. The clinicians use
CBT and are paid by German health insurance. All participants were
diagnosed by the International Diagnostic Checklists (IDCL; Hiller,
Zaudig, & Mombour, 1996; Hiller, Zaudig, Mombour, & Bronisch,
1993). The inclusion criteria of the current study were (a) age of
at least 18 years, (b) a primary diagnosis of hypochondriasis, and
(c) uency and literacy in German. Exclusion criteria were (a) a
major medical illness (e.g., cancer), (b) serious suicidal ideation,
and (c) the clinical diagnosis of substance addiction, schizophrenia, schizoaffective disorder, or bipolar disorders according to the
IDCL. Patients ow chart is shown in Fig. 1.

Completed a diagnostic
interview (N = 126)
Excluded (n = 46)
- declined participation (n = 33)
- not meeting diagnostic criteria (n = 5)
- scheduling problems, were not able to
attend to the group sessions (n = 8)
Received CBT
(n = 80)

Completed CBT
(n = 67)

Drop-out (n = 13)
- lack of motivation (n = 10)
- refused to participate on post
evaluation (n = 2)
- health insurance did not bear the
costs (n = 1)
Lost at follow-up (n = 10)
- declined further participation (n = 6)
- untraceable (n = 4)

Completed 12 month
follow-up
(n = 57)
Fig. 1. Flowchart of participants. CBT, cognitive-behavioral therapy.

2.2. Participants
Eighty patients with the DSM-IV diagnosis of hypochondriasis
participated in the current study. Patients were a mean of 37.06
(SD = 10.89) years old, and 47 (58.8%) were female. Sixty-seven
(83.8%) were married or cohabiting, and 47 (58.8%) had at least 13
years of education. Patients had in mean 2.01 (SD = 0.84) additional
co-morbid disorders, most often affective disorders (45.0%), or anxiety disorders (35.0%). Participants declared that they suffered from
health anxiety for M = 10.98 years (SD = 9.04 years). Sixteen (20.0%)
of the patients received antidepressant medication.

2.3. Measures
2.3.1. Illness attitudes scales (IAS; Kellner, 1986; German: Hiller
and Rief, 2004)
The IAS is a questionnaire containing 27 items, which are evaluated on a 5-point Likert scale (ranging from 0 = no to 4 = most of
the time). The IAS is an internationally well-established instrument for the assessment of hypochondriacal attributes (Sirri,
Grandi, & Fava, 2008). An example item of the IAS is as follows:
Do you worry about your health? Both the original and the German version of the IAS have proved to be highly reliable and valid
(Hing & Weck, 2013; Weck, Bleichhardt, & Hiller, 2009, 2010). In
the current study, the Cronbachs of the IAS was .87.

2.3.2. Cognitions about body and health questionnaire (CABAH;

Rief, Hiller, & Margraf, 1998)
The CABAH assesses cognitions, which are relevant for the maintenance of health anxiety (e.g., Im healthy when I dont have
any bodily sensations.). The response format of the CABAH is
a 4-point Likert scale (ranging from 0 = completely wrong to
3 = completely right). In the current study, we used a 28-item version of the CABAH (items of health habits subscale were excluded
because they failed to demonstrate their validity), which showed
good reliability and validity (Rief et al., 1998). The Cronbachs of
the CABAH in the current study was = .84.

F. Weck et al. / Journal of Anxiety Disorders 30 (2015) 17

2.3.3. Scale for the assessment of illness behavior (SAIB; Rief, Ihle,
& Pilger, 2003)
The SAIB assesses different aspects of illness behavior on 26
items and includes ve subscales (diagnosis verication, expression
of symptoms, medication, consequences of illness, and scanning).
An example item of the SAIB is as follows: When having complaints I very attentively watch the aficted body part. The
response format of the SAIB is a 4-point Likert scale (ranging from
0 = completely wrong to 3 = completely right). In the current
study, the Cronbachs of the SAIB was .82.
2.3.4. Screening for somatoform symptoms (SOMS; Rief & Hiller,
1999; Rief, Hiller, & Heuser, 1997)
The SOMS is a 53-item questionnaire that includes all somatic
symptoms, which are relevant for somatoform disorders (e.g., chest
pain). The response format of the SOMS is a 5-point Likert scale
(ranging from 0 = not at all to 4 = extremely). Five of the SOMSitems are only for women and one only for men. In the current
study, the 7-days version of the SOMS (SOMS-7T) was used, which
asks for somatic symptoms in the last 7 days. In the current study,
the Cronbachs of the SOMS-7T (48 items which are not genderspecic) was .91.
2.3.5. Brief symptom inventory (BSI; Derogatis & Melisaratos,
1983; Franke, 2000)
The BSI is a widely used self-report measure for the assessment of general psychopathology. The questionnaire consists of 53
items, which are evaluated on a 5-point Likert scale (ranging from
0 = not at all to 4 = extremely). The BSI includes nine subscales
(somatization, obsessive-compulsiveness, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation,
and psychoticism) and three global distress measures, the most
important being the General Severity Index (GSI), which consists of
the mean item scores. The German version of the BSI demonstrates
good psychometric properties (Franke, 2000). In the current study,
the Cronbachs of all BSI items was .95.
2.3.6. Satisfaction with treatment and specic interventions
We developed a brief questionnaire to evaluate patients
satisfaction with the treatment and specic interventions. The
questionnaire consists of 11 items, which address different aspects
of the treatment and the overall treatment satisfaction (see Table 3).
The response format is 5-point Likert scale (ranging from 1 = not
at all to 4 = extremely). The questionnaire was developed as a
patients version (How helpful was . . . for you?) and a therapists
version (How helpful was . . . for your patient?). The Cronbachs
for all 11 items was .73 for the patients version and .82 for the
therapists version.
2.4. Cognitive-behavioral group therapy
CBT included 14 treatment sessions (6 individual and 8 group
sessions) and 1 booster session. The individual sessions lasted
50 min, and the group sessions lasted 100 min. Before treatment,
ve individual diagnostic sessions were carried out with no specic
therapeutic interventions (pre-treatment phase). These sessions
were used to gather information for the reviewer (e.g., autobiographical information, information about previous treatments)
who decided whether the health insurance companies should pay
for the treatment. An overview of the interventions is given in
Table 1.
In the group setting, information about the development (e.g.,
risk factors) and maintenance (e.g., attentional bias towards health
threat information, safety behaviors such as repeated healthrelated internet searches) of health anxiety was presented. For
example, in the second group session, the importance of selective

Table 1
Description of the main contents of the cognitive-behavioral therapy.
Session

Setting

Contents

15

Individual

Group

7
8

Group
Group

9
10

Individual
Group

11
12
13

Individual
Group
Individual

14

Group

15
16

Individual
Group

17
18
19

Individual
Group
Individual

Pre-treatment phase: collecting information

for the health insurance company
Development of individual aims, presentation
of the cognitive behavioral model
Attention-based information and exercises
Information about cognitions and health
anxiety as well as cognitive restructuring
Cognitive restructuring
Cognitive restructuring and progressive
relaxation exercises
Cognitive restructuring
Change of safety behavior (body checking)
Change of safety behavior (body checking) and
cognitive restructuring (questioning the wish
for 100% security)
Change of safety behavior (medical
consultations)
Exposure in sensu
Change of avoidance behavior and further
information about exposure
Exposure in vivo
Relapse prevention
Relapse prevention

attention for the perception of bodily symptoms and the maintenance of health anxiety was demonstrated by means of behavioral
experiments. In one behavioral experiment, the patients had to
focus on different areas of the body (e.g., the feet) and were
asked to pay attention to the emerging bodily sensations. Through
this behavioral experiment, the patients could learn that selective attention to the body could lead to the perception of bodily
symptoms (e.g., tingling in the feet).
In the individual sessions, specic dysfunctional beliefs and
fears of the different patients were addressed. For example, in the
rst session, the therapist discussed the patients individual beliefs
about his/her health (e.g., my headache is a sign of a brain tumor)
by using Socratic dialog. Alternative beliefs (e.g., there are many
other possible and more realistic reasons for my headache) were
discussed and should lead to a more appropriate evaluation of
bodily symptoms. The individual sessions were also used to conduct
in sensu exposure (e.g., exposure to fearful health-related images
about cancer) and in vivo exposure (e.g., visit a hospital).
All interventions are described in detail in a treatment manual (Bleichhardt & Weck, 2010). The combination of individual and
group sessions was favored in the previous study of Wattar et al.
(2005) as well. The booster session was conducted in the group
setting and 3 months after therapy ended. The booster session
included a repeated presentation of the cognitive-behavioral model
and reminders of helpful strategies to overcome health anxiety.
2.5. Procedure
The study was conducted from 2005 to 2011. Participants were
recruited from patients with health anxiety who were seeking
treatment at the outpatient unit. Moreover, information about the
treatment study was also published in the local newspapers and
on the Internet. Of the 126 patients who completed the diagnostic
interview, 80 (63.5%) received CBT (see Fig. 1). The group setting
was well accepted by most of the patients. The main concern
regarding the group setting mentioned by the patients was to
receive disturbing information about serious illnesses from other
patients in the group. Patients with those concerns were informed
that the group sessions included mainly structured information
about the development and maintenance of health anxiety and no
unstructured discussion of possible indicators of serious illnesses.

F. Weck et al. / Journal of Anxiety Disorders 30 (2015) 17

Table 2
Mean scores and standard deviations pre-treatment, post-treatment, and at follow-up.
Analysis/measures

CBT M (SD)
Pre

Intention to treat (N = 80)

76.18 (14.00)
IAS
33.94 (10.61)
CABAH
SAIB
38.74 (10.63)
SOMS-7T
33.10 (22.80)
BSI
48.82 (29.74)
Completer (N = 67 at post; N = 57 at follow-up)
76.18 (14.00)
IAS
33.94 (10.61)
CABAH
SAIB
38.74 (10.63)
SOMS-7T
33.10 (22.80)
BSI
48.82 (29.74)

ANOVA

Effect sizes (Cohens d)

Post

Follow-up

Prepost

Pre-follow-up

59.21 (17.16)
24.87 (11.05)
29.82 (11.09)
24.38 (22.30)
32.18 (30.25)

52.29 (19.37)
22.46 (10.38)
26.49 (11.60)
22.67 (22.18)
28.06 (27.66)

103.95
86.47
93.66
13.82
38.23

<.001
<.001
<.001
<.001
<.001

1.08
0.84
0.82
0.39
0.55

1.41
1.09
1.10
0.46
0.72

57.93 (17.61)
23.33 (10.53)
28.34 (11.00)
22.28 (22.37)
29.59 (30.28)

51.47 (19.65)
21.59 (10.89)
25.20 (11.52)
23.09 (24.71)
28.06 (28.12)

83.52
82.47
85.67
8.82
27.51

<.001
<.001
<.001
<.001
<.001

1.15
1.01
0.96
0.47
0.66

1.45
1.15
1.22
0.42
0.72

Note: BSI, brief symptom inventory; IAS, illness attitude scales; CABAH, cognitions about body and health questionnaire; SAIB, scale for the assessment of illness behavior;
SOMS-7T, screening for somatoform symptoms (7-days version).

Dropouts due to scheduling problems of the group sessions

emerged only for eight patients (6.3%).
After the diagnostic interview, participants completed all questionnaires and were assigned a therapist, who conducted the
diagnostic and individual therapy sessions (see Fig. 1). These
individual sessions were conducted by 16 therapists who were
trained in the treatment manual in a 16-hour workshop. Most of
the therapists were female (87.5%) and in psychotherapy training (81.3%). They were M = 28.25 (SD = 2.70) years old and had
M = 2.75 (SD = 2.02) years of clinical experience at the beginning
of the study. Therapists received monthly supervision by experienced and licensed psychotherapists. After the pre-treatment
phase, patients were then treated in the group setting (14 groups
were conducted, with up to eight patients in 1 group). The group
therapy was conducted by two therapists. After treatment, patients
completed all questionnaires. Patients who needed further psychotherapy (73.8%) continued individual therapy by their allocated
therapists. At the 12-month follow-up, patients completed all questionnaires once more.

Post-hoc analyses revealed a signicant time effect between the

pre-treatment and post-treatment measures, as well as between
the post-treatment and follow-up assessment on all measures. The
effect sizes were large for all primary outcome measures posttreatment and at the 12-month follow-up (see Table 2).

3.2. Secondary outcome measures (Hypothesis 2)

The BSI and the SOMS-7T evaluate general psychopathology and
somatic symptoms, and were used as secondary outcome measures in the current study. Table 2 shows the patients scores on
the BSI and SOMS-7T pre-treatment, post-treatment, and followup. Repeated ANOVAs revealed signicant time effects for the BSI
and SOMS-7T. Post-hoc analyses revealed a signicant reduction on
both measures between pre- and post-treatment but not between
post-treatment and the follow-up assessment. For the BSI, a reduction by trend was found between the post-treatment and follow-up
assessments (p = .080). Effect sizes were small to medium for the
SOMS-7T and medium for the BSI (see Table 2).

2.6. Data analysis

Data analyses were conducted with both the completer (N = 67)
and the intent-to-treat (ITT) samples (N = 80). In cases where data
were missing, the last observations were carried forward (LOCF).
The results of ITT analyses are reported in the text, while the results
of completer analyses are presented only when they differ from the
ITT analyses. The results of ITT and completer analyses are reported
in Table 2. For the analyses of therapy outcomes, repeated analyses of variance (ANOVAs) with three measurement times (pre, post,
and follow-up) were conducted. Cohens d was calculated for the
estimation of effect sizes. According to Cohen (1992), effect sizes
between .50 and .80 were considered medium, and effect sizes .80
were considered large. Relationships between different continuous
variables were analyzed by Pearsons correlation coefcients. Relationships between dichotomous and continuous variables were
analyzed by point biserial correlations.
3. Results
3.1. Primary outcome measures (Hypothesis 1)
The IAS, the CABAH, and the SAIB evaluate central hypochondriacal characteristics, and were therefore used as primary outcome
measures in the current study. Table 2 shows the patients scores
in those measures pre-treatment, post-treatment, and follow-up.
Repeated ANOVAs revealed signicant time effects for all measures.

3.3. Consideration of patients who received additional treatment

in the follow-up period
Fifty-nine (73.8%) of the patients continued individual CBT
after the post-assessment measures. Patients who continued treatment received in mean 17.97 (SD = 15.12) additional therapy
sessions [range: 262]. In order to consider the large number of
patients who received additional treatment, we compared patients
who received no additional treatment (n = 21) with patients who
received additional treatment (n = 59). In our analyses, no differences were found between both groups at post-treatment on the
IAS (F(1,79) = 0.40; p = .527), CABAH (F(1,79) = 0.40; p = .530), SAIB
(F(1,79) = 2.69; p = .105), SOMS-7 T (F(1,79) = 0.53; p = .569), and BSI
(F(1,79) = 0.98; p = .326). Furthermore, in repeated ANOVAs with
the two groups (additional treatment vs. no additional treatment)
and two times (post-treatment and follow-up), we found no significant interaction effects on the IAS (F(1,78) = 0.12; p = .731), CABAH
(F(1,78) = 0.27; p = .271), SAIB (F(1,78) = 0.16; p = .694), SOMS-7T
(F(1,78)= 1.02; p = .317), and BSI (F(1,78) = 0.05; p = .819).

3.4. Satisfaction with the treatment and specic interventions

The patients satisfaction with the treatment and therapists
supposed satisfaction with the treatment are reported in Table 3.
The patients satisfaction with the treatment was signicantly

F. Weck et al. / Journal of Anxiety Disorders 30 (2015) 17

Table 3
Patients and therapists evaluation of the satisfaction with the treatment and specic interventions.

How helpful was. . .

Progressive muscle relaxation
Attention-based exercises
Discussing alternative reasons for bodily
symptoms
Discussion reasons for owns illness (as part
of cognitive
restructuring)
Discussion reasons for owns health (as part
of cognitive
restructuring)
Restructuring problematic health-related
cognitions
Exposure in sensu
Information about safety behaviors
Individual therapy sessions
Group therapy session
The whole therapy
Mean

Patient
M (SD)

Therapist
M (SD)

2.96 (1.07)
3.33 (0.96)
3.80 (1.00)

2.22 (1.13)
2.68 (0.96)
2.96 (0.91)

2.82 (1.15)

2.07 (0.87)

3.67 (0.70)

2.85 (0.87)

3.48 (0.96)

2.77 (1.03)

2.98 (1.21)
3.32 (1.10)
4.00 (0.70)
3.80 (0.75)
4.00 (0.67)
3.47 (0.50)

2.65 (1.14)
2.46 (0.98)
2.83 (0.89)
2.83 (0.92)
2.79 (.88)
2.64 (0.59)

Table 4
Correlations between several patients characteristics at pre-treatment and therapy
outcome (pre- and post-treatment and pre-treatment and follow-up changes in the
Illness Attitude Scales)

Age
Gender (0 = male; 1 = female)
Educational level (0 = less than
13 years; 1 = more than 13
years)
Number of diagnoses
Antidepressant medication
(0 = no; 1 = yes)
Duration of health anxiety
Pre-treatment scores
IAS
BSI anxiety subscale
BSI depression subscale
SOMS-7T

Difference
scores
(Prepost)

Difference
scores
(Pre-follow-up)

.100
.119
.003

.090
.084
.021

.164
.199

.063
.037

.152

.019

.280
.018
.102
.015

.260*
.050
.109
.008

Note: *p < .05; BSI, brief symptom inventory; IAS, illness attitude scales; SOMS-7T,
screening for somatoform symptoms (7-days version).

higher than the therapists supposed treatment satisfaction

(p < .001; d = 1.52).
3.5. Predictors of treatment outcome
Several characteristics of the patients were considered as
potential predictors for treatment outcome. Table 4 depicts the
correlations between those characteristics and patients pre- and
post-treatment and pre-treatment and follow-up changes. The IAS
pre-scores were the only signicant predictor of treatment outcomes, accounting for 8% of the variance in the pre-post differences
and 7% of the variance in the pre-follow-up differences.
4. Discussion
The current study is now the largest study that has investigated the effectiveness of group CBT for patients with health
anxiety. We found a large reduction in the primary outcome measures post-treatment (Hypothesis 1) and at the 12-month follow-up
(Hypothesis 2). Moreover, we found a signicant improvement in
the secondary outcome measures, namely a reduction of general
psychopathology and somatic symptoms (Hypothesis 3). Patients

satisfaction with the treatment was signicantly higher than the

therapists supposed satisfaction with the treatment. Highly elevated hypochondriacal symptoms at intake were found to be
a signicant predictor for a better therapy outcome at posttreatment and at follow-up (Hypothesis 4).
Our combined approach of individual and group CBT has been
shown to be highly effective at reducing hypochondriacal symptoms. This included the reduction of dysfunctional health-related
cognitions and behaviors. The effect sizes were large both posttreatment and at the 12-month follow-up. The maintenance or
further improvement of treatment effects in the follow-up assessment is a desirable result. In contrast, in the meta-analysis of
Olatunji et al. (2014) only small mean effect sizes (Hedges g = 0.35)
were found at follow-up. However, in the current study, most of
the patients (73.8%) received additional treatment between the
post-treatment and follow-up assessments, which might explain
the large effect sizes at follow-up. In our analyses, we found no differences between patients who received additional treatment and
patients who did not receive additional treatment in their posttreatment and follow-up scores. Therefore, additional treatment
can rather not be the reason for the large effect sizes at follow-up.
However, it is still possible that patients who received additional
treatment would experience signicantly worse symptoms without additional treatment sessions.
In our study, the number of treatment sessions was not strictly
controlled after the post-treatment assessment. A consequence of
this lack of control was the fact that most of the patients desired
additional treatment sessions. The very high number of patients
who continued treatment can be explained by the specics of
the German health care system. German health insurance organizations typically pay for either 30 or 50 sessions (meaning that
the rst 5 diagnostic sessions serve to gather information for
the health insurance companies which pay for the treatment).
Many of the patients wanted to attend all of their approved treatment sessions. The benet of these additional treatment sessions
regarding the improvement on hypochondriacal symptoms, general psychopathology, and somatic symptoms can be questioned
by our ndings. However, it is possible that patients benet on
other dimensions that were not evaluated in the current study. The
hypothesis that a longer treatment time leads to a higher reduction of symptoms that was reported by Olatunji et al. (2014) and
Thomson and Page (2007) cannot be determined in the current
study. However, the studies in the meta-analyses varied between
4 and 16 treatment sessions. It is possible that the relationship
between treatment time and treatment success only emerged
for short-term treatments when compared to longer treatments,
but not for longer treatment when compared to even longer
treatments. However, in our study, additional treatment was not
assigned by a randomized procedure; therefore, the conclusions
regarding the effects of additional treatment are limited.
We found that hypochondriacal symptoms at pre-treatment
were the only relevant variable regarding the prediction of therapy
outcome. Therefore, higher scores lead to a better treatment outcome, which is in line with Nakao et al. (2011) and Olatunji et al.
(2014), but contradicts Buwalda et al. (2007) and Olde Hartman
et al. (2009). However, only 78% of the variance in therapy outcome was explained, which is relatively little. In addition, an
interesting nding is the lack of a relationship between the duration of health anxiety and the treatment outcome. That means that
patients who have long-lasting and severe health anxiety benet
from CBT as much as, or even more than, patients who have less
severe and less persisting health anxiety.
Overall, patients were satised with the given treatment. Therapists evaluated the patients satisfaction with the treatment lower
than the patients did. Both patients and therapists reported that the
discussion of alternative reasons for bodily symptoms was the most

F. Weck et al. / Journal of Anxiety Disorders 30 (2015) 17

helpful CBT strategy. This result is in line with empirical ndings,

which showed that the misinterpretation of bodily symptoms is a
specic characteristic of hypochondriasis (e.g., Bailer et al., 2013;
Neng & Weck, 2013; Rief et al., 1998; Weck, Neng, Richtberg, &
Stangier, 2012a, 2012b).
Several limitations of the current study should be considered.
First, as in most studies which investigated group therapy for health
anxiety, we did not use a control group. This means that patients
might improve as much after receiving CBT as they would without
treatment. However, health anxiety was found to be a very persistent condition (e.g., Fink et al., 2009) and only a small effect size
(d = .16) was found for the IAS in a waiting group of a randomized controlled trial, which implemented a 3-month waiting period
(Weck, Neng, Richtberg, Jakob, & Stangier, 2014). Therefore, it does
not seem plausible that patients with no treatment would improve
as much as patients receiving active treatment.
Second, treatment integrity was not evaluated. Therefore, it
is not clear whether CBT was conducted as intended, and therefore, the internal validity of the study can be questioned (e.g.,
Perepletchikova & Kazdin, 2005; Weck, Bohn, Ginzburg, & Stangier,
2011; Weck, Grikscheit, Hing, & Stangier, 2014). Therapists
were trained and supervised in the conducted treatment, and a
detailed treatment manual was given to them. Therefore, treatment integrity can be supposed to be high, but it cannot be veried
empirically in the current study.
Third, the treatment outcomes were only assessed by questionnaires. The additional implementation of a clinical interview, such
as the YaleBrown Obsessive Compulsive Scale for Hypochondriasis (H-YBOCS; Weck, Gropalis, Neng, & Witthft, 2013), conducted
by independent assessors would lead to a more comprehensive
assessment of treatment outcomes. This should also be considered
in further studies.
Fourth, the treatment may not be appropriate for all patients
because 16.3% of the patients dropped out during their treatment.
Even higher dropout rates of 2528% were reported in previous
studies, which investigated individual psychotherapy for health
anxiety (Greeven et al., 2007; Visser & Bouman, 2001). Some studies reported lower dropout rates (e.g., Clark et al., 1998; McManus,
Surawy, Muse, Vazquez-Montes, & Williams, 2012). The overall level of treatment satisfaction was high in the current study.
Therefore, the combination of individual and group CBT can be
considered an appropriate treatment approach for treating health
anxiety.
5. Conclusion
The current study has demonstrated that the combination of
individual and group CBT is an effective approach for the treatment
of health anxiety and that patients were satised with the treatment. Moreover, treatment effects persisted or improved in the
12-month follow-up. In clinical practice, the use of group sessions
can therefore be considered an effective and economically efcient
approach for the treatment of patients with elevated health anxiety. In future studies, one important limitation of the current study
should be addressed by establishing a control group.
References
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders DSM-IV-TR (text revision). Washington, DC: American Psychiatric
Association.
Avia, M. D., Ruiz, M. A., Olivares, M. E., Crespo, M., Guisado, A. B., Snchez, A.,
et al. (1996). The meaning of psychological symptoms: effectiveness of a group
intervention with hypochondriacal patients. Behaviour Research and Therapy, 34,
2331. http://dx.doi.org/10.1016/0005-7967(95)52-Y
Bailer, J., Mller, T., Witthft, M., Diener, C., Mier, D., Ofer, J., et al.
(2013). Symptomattributionsstile bei Hypochondrie [Symptom attribution

style in hypochondriasis]. Psychotherapeut, 58, 552559. http://dx.doi.org/

10.1007/s00278-013-1014-4
Bleichhardt, G., & Hiller, W. (2007). Hypochondriasis and health anxiety in
the German population. British Journal of Health Psychology, 12, 511523.
http://dx.doi.org/10.1348/135910706146034
Bleichhardt, G., Timmer, B., & Rief, W. (2005). Hypochondriasis among patients with
multiple somatoform symptomspsychopathology and outcome of a cognitivebehavioral therapy. Journal of Contemporary Psychotherapy, 35, 239249.
http://dx.doi.org/10.1007/s10879-005-4318-z
Bleichhardt, G., & Weck, F. (2010). Kognitive Verhaltenstherapie bei Hypochondrie und
Krankheitsangst (2. Auage) [Cognitive behavioral therapy for hypochondriasis and
health anxiety (2. Edition)]. Heidelberg: Springer.
Bouman, T. K. (2002). A community-based psychoeducational group approach
to hypochondriasis. Psychotherapy and Psychosomatics, 71, 326332.
http://dx.doi.org/10.1159/000065995
Buwalda, F. M., Bouman, T. K., & van Duijn, M. A. J. (2007). Psychoeducation
for hypochondriasis: a comparison of a cogntive-behavioural approach and
a problem-solving approach. Behaviour Research and Therapy, 45, 887899.
http://dx.doi.org/10.1016/j.brat.2006.08.004
Clark, D. M., Salkovskis, P. M., Hackmann, A., Wells, A., Fennell, M., Ludgate, J., et al.
(1998). Two psychological treatments for hypochondriasis. British Journal of Psychiatry, 173, 218225. http://dx.doi.org/10.1192/bjp.173.3.218
Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155159.
Derogatis, L. R., & Melisaratos, N. (1983). The brief symptom inventory: an introductory report. Psychological Medicine, 13(3), 595605.
http://dx.doi.org/10.1017/S0033291700048017
Eilenberg, T., Kronstrand, L., Fink, P., & Frostholm, L. (2013). Acceptance and commitment group therapy for health anxietyresults from a pilot study. Journal of
Anxiety Disorders, 27, 461468. http://dx.doi.org/10.1016/j.janxdis.2013.06.001
Fink, P., rnbl, E., & Christensen, K. S. (2010). The outcome of health anxiety in
primary care. A two-year follow-up study on health care costs and self-rated
health. PLoS ONE, 5, e9873. http://dx.doi.org/10.1371/journal.pone.0009873
Franke, G. H. (2000). Brief symptom inventory by L.R. Derogatis (Kurzform der SCL-90R)-Deutsche Version. Manual. [Short version of the SCL-90-R German version].
Gttingen: Beltz Test GmbH.
Greeven, A., van Balkom, A. J. L. M., Visser, S., Merkelbach, J. W., van Rood, Y. R.,
van Dyck, R., et al. (2007). Cognitive behavior therapy and paroxetine in the
treatment of hypochondriasis: a randomized controlled trial. American Journal
of Psychiatry, 164, 9199. http://dx.doi.org/10.1176/appi.ajp.164.1.91
Hedman, E., Andersson, G., Andersson, E., Ljtsson, B., Rck, C., Asmundson, G. J.
G., et al. (2011). Internet-based cognitive-behavioural therapy for severe health
anxiety: randomised controlled trial. British Journal of Psychiatry, 198, 230236.
http://dx.doi.org/10.1192/bjp.bp.110.086843
Hedman, E., Axelsson, E., Grling, A., Ritzman, C., Ronnheden, M., Alaoui, S.
E., et al. (2014). Internet-delivered exposure-based cognitive-behavioural
therapy and behavioural stress management for severe health anxiety:
randomised controlled trial. British Journal of Psychiatry, 198, 230236.
http://dx.doi.org/10.1192/bjp.bp.110.086843
Hedman, E., Ljtsson, B., Andersson, E., Ruck, C., Andersson, G., & Lindefors, N. (2010).
Effectiveness and cost offset analysis of group CBT for hypochondriasis delivered
in psychiatric setting: an open trial. Cognitive Behaviour Therapy, 39, 239250.
http://dx.doi.org/10.1080/16506073.2010.496460
Hiller, W., Zaudig, M., & Mombour, W. (1996). IDCL-international diagnostic checklists
for ICD-10 and DSM-IV. Seattle, WA: Hogrefe & Huber.
Hiller, W., Zaudig, M., Mombour, W., & Bronisch, T. (1993). Routine psychiatric
examinations guided by ICD-10 diagnostic checklists (International Diagnostic Checklists). European Archives of Psychiatry and Clinical Neuroscience, 242,
218223. http://dx.doi.org/10.1007/BF02189966
Hiller, W., & Reif, W. (2004). Internationale Skalen fr Hypochondrie. [International
scales for hypochondriasis]. Bern: Huber.
Hing, V., & Weck, F. (2013). Assessing bodily preoccupations is sufcinet: clinically
effective screening for hypochondriasis. Journal of Psychosomatic Research, 75(6),
526531. http://dx.doi.org/10.1016/j.jpsychores.2013.10.011
Kellner, R. (1986). Somatization and hypochondriasis. New York: Praeger Publishers.
McManus, F., Surawy, C., Muse, K., Vazquez-Montes, M., & Williams, J. M. G. (2012). A
randomized clinical trial of mindfulness-based cognitive therapy versus unrestricted services for health anxiety (hypochondriasis). Journal of Consulting and
Clinical Psychology, 80, 817828. http://dx.doi.org/10.1037/a0028782
Nakao, M., Shinozaki, Y., Ahern, D. K., & Barsky, A. J. (2011). Anxiety as predictor of improvements in somatic symptoms and health anxiety associated with
cognitive-behavioral interventions in hypochondriasis. Psychotherapy and Psychosomatics, 80, 151158. http://dx.doi.org/10.1159/000320122
Neng, J. M. B., & Weck, F. (2013). Attribution of somatic symptoms in hypochondriasis. Clinical Psychology & Psychotherapy, http://dx.doi.org/10.1002/cpp.1871
Olatunji, B. O., Kauffman, B. Y., Meltzer, S., Davis, M. L., Smits, J. A. J., & Powers, M.
B. (2014). Cognitive-behavioral therapy for hypochondriasis/health anxiety: a
meta-analysis of treatment outcome and moderators. Behaviour Research and
Therapy, 58, 6574. http://dx.doi.org/10.1016/j.brat.2014.05.002
Olde Hartman, T. C., Borghuis, M. S., Lucassen, P. L. B. J., van de Laar,
F. A., Speckens, A. E., & van Weel, C. (2009). Medically unexplained
symptoms, somatisation disorder and hypochondriasis: course and prognosis: a systematic review. Journal of Psychosomatic Research, 66, 363377.
http://dx.doi.org/10.1016/j.psychores.2008.09.018
Perepletchikova, F., & Kazdin, A. E. (2005). Treatment integrity and therapeutic
change: issues and research recommendations. Clinical Psychology: Science and
Practice, 12, 365383. http://dx.doi.org/10.1093/clipsy/bpi045

F. Weck et al. / Journal of Anxiety Disorders 30 (2015) 17

Rief, W., & Hiller, W. (1999). Toward empirically based criteria for the classication of somatoform disorders. Journal of Psychosomatic Research, 46, 507518.
http://dx.doi.org/10.1016/S0022-3999(99)23-9
Rief, W., Hiller, W., & Heuser, J. (1997). SOMSDas Screening fr Somatoforme Strungen: Manual zum Fragebogen [SOMSscreening for somatofrom symptoms: manual
for the self-rating scale]. Bern: Huber Verlag.
Rief, W., Hiller, W., & Margraf, J. (1998). Cognitive aspects of hypochondriasis and
the somatization syndrome. Journal of Abnormal Psychology, 107(4), 587595.
http://dx.doi.org/10.1037/0021843X.107.4.587
Rief, W., Ihle, D., & Pilger, F. (2003). A new apprach to assess illness behaviour. Journal of Psychosoamtic Research, 54, 405414.
http://dx.doi.org/10.1016/S00-23999(02)401-4
Sirri, L., Grandi, S., & Fava, G. A. (2008). The illness attitude scales: a clinimetric index
for assessing hypochondriacal fears and beliefs. Psychotherapy and Psychosomatics, 77, 337350. http://dx.doi.org/10.1159/000151387
Srensen, P., Birket-Smith, M., Wattar, U., Buemann, I., & Salkovskis, P.
(2011). A randomized clinical trial of cognitive behavioural therapy
versus short-term psychodynamic psychotherapy versus no intervention
for patients with hypochondriasis. Psychological Medicine, 41, 431441.
http://dx.doi.org/10.1017/S0033291710000292
Stern, R., & Fernandez, M. (1991). Group cognitive and behavioral treatment for
hypochondriasis. British Medical Journal, 303, 12291231.
Thomson, A. B., & Page, L. A. (2007). Psychotherapies for hypochondriasis. Cochrane
Database of Systematic Reviews, 4, CD006520. http://dx.doi.org/10.1002/
14651858.CD006520.pub2
Visser, S., & Bouman, T. K. (2001). The treatment of hypochondriasis: exposure plus
response prevention vs cognitive therapy. Behaviour Research and Therapy, 39,
423442. http://dx.doi.org/10.1016/S0005-7967(00)22-X
Wattar, U., Sorensen, P., Buemann, I., Birket-Smith, M., Salkovskis, P. M., Albertsen,
M., et al. (2005). Outcome of cognitive-behavioural treatment for health anxiety (hypochondriasis) in a routine clinical setting. Behavioural and Cognitive
Psychotherapy, 33, 165175. http://dx.doi.org/10.1017/S1352465804002000

Weck, F., Bleichhardt, G., & Hiller, W. (2009). The factor structure of the illness attitude scales in a German population. International Journal of Behavioral Medicine,
16(2), 164171. http://dx.doi.org/10.1007/s12529-009-9043-7
Weck, F., Bleichhardt, G., & Hiller, W. (2010). Screening for hypochondriasis with
the Illness Attitude Scales. Journal of Personality Assessment, 92(3), 260268.
http://dx.doi.org/10.1080/00223891003670216
Weck, F., Bohn, C., Ginzburg, D. M., & Stangier, U. (2011). Treatment integrity:
implementation, assessment, evaluation, and correlation with outcome. Verhaltenstherapie, 21, 99107. http://dx.doi.org/10.1159/000328840
Weck, F., Grikscheit, F., Hing, V., & Stangier, U. (2014). Assessing treatment integrity in cognitive-behavioral therapy: comparing session
segments with entire sessions. Behavior Therapy, 45, 541552.
http://dx.doi.org/10.1016/j.beth.2014.03.003
Weck, F., Gropalis, M., Neng, J. M. B., & Witthft, M. (2013). The German version of
the H-YBOCS for the assessment of hypochondriacal cognitions and behaviors:
development, reliability and validity. International Journal of Behavioral Medicine,
20, 618628. http://dx.doi.org/10.1007/s12529-012-9276-8
Weck, F., Neng, J. M. B., Richtberg, S., Jakob, M., & Stangier, U. (2014). Cognitive therapy versus exposure therapy for hypochondriasis (health
anxiety): a randomized controlled trial. Journal of Consulting and Clinical Psychology, http://dx.doi.org/10.1037/ccp0000013. Advance online
publication
Weck, F., Neng, J. M. B., Richtberg, S., & Stangier, U. (2012a). Dysfunctional beliefs about symptoms and illness in patients with hypochondriasis.
Psychosomatics: Journal of Consultation Liaison Psychiatry, 53(2), 148154.
http://dx.doi.org/10.1016/j.psym.2011.11.007
Weck, F., Neng, J. M. B., Richtberg, S., & Stangier, U. (2012b). The restrictive concept of
good health in patients with hypochondriasis. Journal of Anxiety Disorders, 26(8),
792798. http://dx.doi.org/10.1016/j.janxdis.2012.07.001
Weck, F., Richtberg, S., & Neng, J. (2014). Epidemiology of hypochondriasis and health
anxiety: comparison of different diagnostic criteria. Current Psychiatry Reviews,
10, 1423. http://dx.doi.org/10.2174/1573400509666131119004444