Content uploaded by Andreea Vîslă
Author content
All content in this area was uploaded by Andreea Vîslă on Nov 27, 2015
Content may be subject to copyright.
E-Mail karger@karger.com
Special Article
Psychother Psychosom 2016;85:8–15
DOI: 10.1159/000441231
Irrational Beliefs and Psychological
Distress: A Meta-Analysis
AndreeaVîslă a,b ChristophFlückiger b,c MartingrosseHoltforth b–d
DanielDavid a,e
a Department of Clinical Psychology and Psychotherapy, Babeş-Bolyai University, Cluj-Napoca , Romania;
b Department of Psychology, University of Zurich, Zurich ,
c Department of Psychology, University of Bern, and
d Psychosomatic Competence Center, University Hospital Inselspital, Bern, Switzerland; e Department of Oncological
Sciences, Icahn School of Medicine at Mount Sinai, New York, N.Y. , USA
country. Results: Overall, irrational beliefs were positively as-
sociated with various types of distress, such as general dis-
tress, anxiety, depression, anger, and guilt (omnibus: r =
0.38). The following variables were significant moderators of
the relationship between the intensity of irrational beliefs
and the level of distress: irrational belief measure and type,
stressful event, age, educational and clinical status, and de-
veloper/validator status of the author. Conclusions: Irratio-
nal beliefs and distress are moderately connected to each
other; this relationship remains significant even after con-
trolling for several potential covariates.
© 2015 S. Karger AG, Basel
Introduction
Given the high prevalence of mental disorders in the
population and the considerable economic burden asso-
ciated, it is imperative to disseminate effective treatments
that have enduring effects for patients over time
[1] .
However, in order to establish and disseminate psycho-
therapeutic treatments, there is not only a claim to inves-
tigate their efficacy/effectiveness, but also a demand for a
clear understanding of their underlying theory and psy-
Key Words
Irrational beliefs · Meta-analysis · Psychological distress
Abstract
Background: Since the cognitive revolution of the early
1950s, cognitions have been discussed as central compo-
nents in the understanding and treatment of mental illness-
es. Even though there is an extensive literature on the asso-
ciation between therapy-related cognitions such as irratio-
nal beliefs and psychological distress over the past 60 years,
there is little meta-analytical knowledge about the nature of
this association. Methods: The relationship between irratio-
nal beliefs and distress was examined based on a systematic
review that included 100 independent samples, gathered in
83 primary studies, using a random-effect model. The overall
effects as well as potential moderators were examined: (a)
distress measure, (b) irrational belief measure, (c) irrational
belief type, (d) method of assessment of distress, (e) nature
of irrational beliefs, (f) time lag between irrational beliefs and
distress assessment, (g) nature of stressful events, (h) sample
characteristics (i.e. age, gender, income, and educational,
marital, occupational and clinical status), (i) developer/vali-
dator status of the author(s), and (k) publication year and
Received: May 24, 2015
Accepted after revision: September 18, 2015
Published online: November 27, 2015
Andreea Vîslă
Department of Clinical Psychology and Psychotherapy, Babeş-Bolyai University
Republicii 37
RO–400015 Cluj-Napoca (Romania)
E-Mail andreea.visla @ gmail.com
© 2015 S. Karger AG, Basel
0033–3190/15/0851–0008$39.50/0
www.karger.com/pps
Downloaded by:
Universitätsbibliothek Bern
198.143.58.33 - 11/27/2015 9:31:57 AM
Irrational Beliefs and Psychological
Distress
Psychother Psychosom 2016;85:8–15
DOI: 10.1159/000441231
9
chological mechanisms [2, 3] . Therefore, investigating
how fundamental psychological mechanisms postulated
to produce change in these treatments relate to psycho-
logical distress becomes a necessary step towards validat-
ing these treatments.
Irrational beliefs are lasting constructs used over the
past decades and co-developed in the Rational Emotive
Behavior Therapy (REBT), an early cognitive-behavior-
al framework for the treatment of mental disorders
[4] .
After initially proposing 11 irrational or evaluative belief
types
[5] , subsequent developments in REBT [6] as-
signed these types of irrational beliefs to four catego-
ries: demandingness (i.e. absolutistic/inflexible require-
ments), awfulizing (or catastrophizing), frustration
intolerance (or low frustration tolerance) and global
evaluation of one’s own person (self-downing), other
persons (other-downing) and/or the life situation (life-
downing). In stressful activating situations, irrational
beliefs are hypothesized to engender dysfunctional emo-
tions
[7] . Another main approach in clinical psychology
focuses on the role of automatic thoughts and cognitive
schemas (or representations) in generating distress
[8] .
It has been shown that while representations contribute
to emotion, only evaluations (i.e. how relevant the event
is for the individual) have a direct impact on emotional
response
[9, 10] .
Over the years, many studies investigated REBT and
helped to refine the theory and practice of REBT based
on a large number of diverse narrative reviews of em-
pirical research
[11, 12] . However, thus far, the quantita-
tive findings associated with these reviews have not been
summarized using meta-analytic methods. Thus, the
major aim of the present investigation was to examine
the quantitative findings regarding the correlational re-
lationship between irrational beliefs and various types of
psychological distress using meta-analytic methods.
Furthermore, we aimed to investigate the role of the fol-
lowing variables as potential moderators of the relation-
ship between the level of irrational beliefs and the inten-
sity of psychological distress: (a) distress measure; (b)
irrational belief measure; (c) irrational belief type; (d)
method of assessment of distress (i.e. self-report vs. ob-
server report); (e) nature of irrational beliefs (i.e. gen-
eral vs. specific, and naturally varying vs. manipulated);
(f) time lag between irrational beliefs and distress assess-
ment; (g) nature of stressful events (i.e. present vs. ab-
sent, real/naturalistic vs. experimentally induced, per-
sonally relevant vs. not personally relevant); (h) sample
characteristics (e.g. age, gender, income, and education-
al, marital, occupational and clinical status); (i) devel-
oper/validator status of the author(s), and (k) publica-
tion year and country.
We expect distinct patterns of association between
irrational beliefs and distress, considering that some dis-
tress measures contain mostly emotional items, while
others contain a mixture of emotional, behavioral and
somatic items
[13] . Second, we expect to obtain a sig-
nificantly stronger irrational belief-distress relationship
when irrational beliefs were measured using scales that
contain emotional items (e.g. the Irrational Beliefs Test,
IBT
[14] ), as opposed to newer scales containing only
cognitive items (e.g. the Survey of Personal Beliefs, SPB
[15] ; the Attitude and Belief Scale, ABS [16] , and the Ir-
rational Belief Scale, IBS
[17] ). In addition, reliability
indices of the different measures might also impact this
relationship
[18] . Third, we expect that the various ir-
rational belief types (e.g. demandingness and awfuliz-
ing) would show significantly different associations with
different types of psychological distress (e.g. anxiety and
depression
[19] ). Fourth, we expect that the strength of
the irrational belief-distress association will differ de-
pending on the observational perspective (i.e. self-re-
port vs. clinician report
[20] ). Fifth, we expect signifi-
cant differences in the irrational belief-distress associa-
tion depending on whether general versus specific
irrational beliefs were measured
[21] . Sixth, we expect
the association between irrational beliefs and distress
would be significantly stronger in studies where irratio-
nal beliefs were not only measured occurring naturally
but also manipulated experimentally (e.g. instructions
to repeat irrational beliefs
[22] , and randomized clinical
trials that included REBT as a treatment group and mea-
sured irrational beliefs as a central mechanism of change
[23] ). Seventh, we expect the irrational belief-distress as-
sociation will significantly differ depending on the time
lag between the measurement of the two constructs
[21] .
Moreover, the association between irrational beliefs and
distress may vary depending on the level of stress poten-
tially being induced by particular events
[7] ; it is as-
sumed that the level of induced stress may also be max-
imal when the stress is real and personally relevant
[7,
24] . The strength of the irrational belief-distress asso-
ciation may also vary depending on characteristics of
specific samples (e.g. participants’ age
[25] ). Research-
ers’ conflicts of interest may systematically impact the
(predictive) validity of irrational beliefs, i.e. developers
or validators of a particular irrational belief measure
may be more motivated to demonstrate a high predic-
tive validity of the measure examined. When (i.e. publi-
cation year) and where (i.e. publication country) a re-
Downloaded by:
Universitätsbibliothek Bern
198.143.58.33 - 11/27/2015 9:31:57 AM
Vîslă/Flückiger/grosse Holtforth/David
Psychother Psychosom 2016;85:8–15
DOI: 10.1159/000441231
10
spective study was conducted may also influence the re-
lationship between irrational beliefs and psychological
distress
[26] .
Methods
Selection of Studies
Potentially relevant studies were queried in the PsycINFO and
Medline databases for all years covered through April 2014. We
used the following search terms: ‘irrational beliefs’, ‘distress’, ‘neg-
ative feelings’, ‘negative emotions’, ‘anxiety’, ‘depression’, ‘anger’
and ‘guilt’. We also examined the reference sections of all articles
included. The search procedure led to the identification of 182 re-
cords (online suppl. fig.1; for all online suppl. material, see www.
karger.com/doi/10.1159/000441231). After removing duplicates,
the remaining studies were analyzed in detail for relevance based
on their abstract. Following the exclusion of irrelevant publica-
tions (i.e. the screened abstracts indicated that either irrational be-
liefs or distress were not measured, irrational beliefs were not mea-
sured according to the REBT theory, or the authors did not com-
pute an association between irrational beliefs and distress), a total
of 89 potentially relevant articles were inspected for relevance
based on their full text. Only studies that fulfilled the following
criteria were included: (a) assessed irrational beliefs according to
the REBT theory; (b) assessed at least one type of psychological
distress (e.g. general distress and anxiety), and (c) reported a nu-
merical relationship between irrational beliefs and distress that
was amenable to meta-analytic methods. Eighty-three articles were
included for analyses (online suppl. fig.1).
Coding of Studies
The authors created a coding system based on the theoretical
conceptualization of irrational beliefs and psychological distress
(online suppl. table1). All articles were coded by the first author; in
addition, a random sample of 10 studies was coded by a research
assistant. The interrater agreement was high (r ≥ 0.95 for categorical
variables and r ≥ 0.99 for continuous variables). The few cases with
diverging assessments were discussed until a consensus was reached.
Meta-Analytic Procedures
The correlation coefficient, r , was the measure of choice to as-
sess the effect size for most analyses. We used only one effect size
from each study for each analysis
[27] , and we randomly selected
one within-study level for each categorical moderator to base our
analyses on a fully independent sample
[28] . All computations
were performed on the basis of Fisher’s z transformation of r be-
fore the sample effect sizes were included in our meta-analysis
[29] . The weighted mean effect sizes were converted back to r for
interpretive purposes. When interpreting the magnitude of r, we
adopted Cohen’s
[30] guidelines: r values of 0.10, 0.30 and 0.50
indicate small, medium and large effect sizes, respectively. In the
effect size analyses, we used a random-effect model. All analyses
were conducted using the R statistical software package for meta-
analysis ‘MAc’
[29] and ‘metafor’ [31] . Heterogeneity was assessed
using the Q and I
2 statistics [32] . To identify publication bias,
asymmetry was tested based on rank correlation
[33] and regres-
sion tests
[34] . Furthermore, a funnel plot was examined using
trim and fill procedures
[35] .
Results
Descriptive Characteristics
Participant and Study Characteristics
The total number of participants across all 83 studies
comprising 100 different samples was 16,110. The weight-
ed mean age by sample size was 29.4 years (SD = 10.45;
range = 12.5–72.5 years). Sixty-six samples were com-
posed of student populations and 34 samples were com-
posed of populations other than high-school or college
students. Twenty-two were clinical samples, and 78 were
subclinical or nonclinical samples. The included studies
were published between 1972 and 2014. They were con-
ducted in 13 different countries; the first 3 places were
occupied by the United States, Romania, and the United
Kingdom. Irrational beliefs were manipulated before be-
ing measured in only 5 samples. In only 17 samples, a
stressful event was present.
Irrational Belief and Psychological Distress Measures
The most frequently used irrational belief measures
were the IBT
[14] (k = 23), SPB [15] (k = 18), IBS [17]
(k = 15) and ABS
[16] (k = 15). The most frequently used
measures for psychological distress were the Beck De-
pression Inventory
[36] (k = 25), the State-Trait Anxiety
Inventory
[37] (k = 21), the Profile of Mood States [38]
(k = 10), and a single item rating (i.e. emotional items
rated on a Likert scale; e.g. anxiety and depression; k = 8).
Irrational Belief and Psychological Distress Types
Irrational belief types were measured as follows: de-
mandingness was measured in 40 samples, awfulizing/
catastrophizing in 22 samples, frustration intolerance/
low frustration tolerance in 24 samples and global evalu-
ation in 22 samples. The psychological distress types were
measured as follows: general distress was measured in 26
samples, depression in 47, anxiety in 44, anger in 17, and
guilt in 6.
Relationship between Irrational Beliefs and Overall
Psychological Distress
The overall effect of the unconditional model analysis
(k = 100) was r = 0.38 (95% CI = 0.34–0.42). There was
significant heterogeneity in the effect sizes (Q = 904.88,
p < 0.001; I
2 = 89%, 95% CI = 0.34–0.42). Significance
tests of asymmetry indicated that publication bias was not
present in the included studies (p > 0.22). The trim and
fill procedure
[34] estimated that the number of missing
studies needed to attain complete symmetry was one; the
imputed study is depicted in the online supplementary
Downloaded by:
Universitätsbibliothek Bern
198.143.58.33 - 11/27/2015 9:31:57 AM
Irrational Beliefs and Psychological
Distress
Psychother Psychosom 2016;85:8–15
DOI: 10.1159/000441231
11
figure 2. The distress measure was not a significant mod-
erator of the relationship between irrational beliefs and
psychological distress on the overall sample (p > 0.05).
Relationship between Irrational Beliefs and
Psychological Distress Types
The relationship between irrational beliefs and each
distress type reached statistical significance, with anxiety
showing the highest association ( table1 ). Moreover, there
was significant heterogeneity for each distress type.
Therefore, we tested potential moderators for each dis-
tress type.
Moderation Analyses within Distress Types
Irrational belief measure moderated the association
between irrational beliefs and depression as well as the
association between irrational beliefs and anxiety (p <
0.05), with some measures (e.g. IBT and ABS) showing
significantly higher associations than other measures
(e.g. SPB; table2 ). For an analysis of direct contrast of
measures in anxiety and depression, see online supple-
mentary table2. The type of irrational beliefs was not a
significant moderator of the irrational belief-distress as-
sociation for depression and anxiety (p > 0.05). However,
due to the observed variation
among the irrational belief-
distress association for different irrational belief types
(with frustration intolerance showing the highest associa-
tion), we computed an exploratory analysis contrasting
frustration intolerance and all other irrational belief types
for each psychological distress type. We obtained a sig-
nificant moderation effect for frustration intolerance ver-
sus all other irrational belief types in depression, anxiety,
anger, and guilt (p < 0.05), but not in general distress
(p > 0.05). As shown in table2 , the relationship between
frustration intolerance and distress was significantly
higher than the relationship between all other irrational
belief types combined and distress.
Regarding stressful events, we obtained significant re-
sults in the case of depression (p < 0.05); the association
between irrational beliefs and depression was higher
when a stressful event was present (r = 0.67, p < 0.001,
k = 2) versus when not (r = 0.30, p < 0.001, k = 42; table2 ).
We also obtained a significant result for general distress
(p < 0.05); a higher association between irrational beliefs
and general distress was reported when the stressful event
was experimentally induced (r = 0.55, p < 0.001, k = 2)
versus when the stressful event was real/naturalistic (r =
0.32, p < 0.001, k = 7). We did not obtain a significant
moderation effect for the personal relevance of the stress-
ful event for either anxiety or depression (p > 0.05). Age
was a significant moderator of the relationship between
irrational beliefs and anger; specifically, for every unit in-
crease in age, there was a 0.04 increase in the association
between irrational beliefs and anger (p < 0.05). Educa-
tional status was also a significant moderator of the asso-
ciation between irrational beliefs and distress in general
distress and anger (p < 0.05); the association with both
general distress and anger was significantly smaller in stu-
dents (r = 0.30, p < 0.001, k = 16 for general distress; r =
0.19, p < 0.001, k = 14 for anger) compared with subjects
in nonuniversity samples (r = 0.46, p < 0.001, k = 10 for
general distress; r = 0.52, p < 0.001, k = 3 for anger). We
did not obtain significant results for depression or anxiety
(p > 0.05). The clinical status of the participants was a
moderator of the association between irrational beliefs
and anger (p < 0.05); the association was higher in clinical
samples (r = 0.54, p < 0.001, k = 3) compared with non-
clinical samples (r = 0.21, p < 0.001, k = 14). However,
clinical status was not a significant moderator in general
distress, depression or anxiety (p > 0.05).
The developer/validator status of the author(s) was a
significant moderator in depression and anger (p < 0.05;
table2 ). Notably, the association between irrational be-
liefs and depression was significantly smaller when any of
the authors were a developer/validator (r = 0.22, p < 0.001,
k = 15) versus when none of the authors were a developer/
validator (r = 0.38, p < 0.001, k = 32). Similarly, the asso-
ciation between irrational beliefs and anger was signifi-
cantly smaller when an author was a developer/validator
(r = 0.15, p < 0.05, k = 7) versus when not (r = 0.33, p <
0.001, k = 10). The developer/validator status of the
author(s) did not moderate the association between irra-
tional beliefs and psychological distress for either general
distress or anxiety (p > 0.05). As can be observed in ta-
ble2 , we did not obtain significant results for the follow-
Table 1. Overall effect sizes for different psychological distress
types
Distress type k n r 95% CI Q I2
General
distress 26 4,290 0.36** 0.27, 0.44 243.2** 89%
Depression 47 8,278 0.33** 0.26, 0.39 463.06** 90%
Anxiety 44 5,911 0.41** 0.31, 0.5 752.99** 94%
Anger 17 3,046 0.25** 0.17, 0.32 72.86** 76%
Guilt 6 1,270 0.29*0.02, 0.52 122.15** 95%
k is the number of effect sizes included in each analysis. * p <
0.05, ** p < 0.001.
Downloaded by:
Universitätsbibliothek Bern
198.143.58.33 - 11/27/2015 9:31:57 AM
Vîslă/Flückiger/grosse Holtforth/David
Psychother Psychosom 2016;85:8–15
DOI: 10.1159/000441231
12
Table 2. Moderation analyses within distress types
Distress types
general distress depression anxiety anger guilt
Moderators Qb (d.f.) k r Qb (d.f.) k r Qb (d.f.) k r Qb (d.f.) k r Q
b (d.f.) k r
Irrational belief measure 1.13 (4) 16.36 (4)*12.38 (4)*––
IBT 2 0.50*10 0.46** 12 0.50**
IBS 5 0.30*70.41** 70.45**
ABS 6 0.40** 60.45** 30.73**
SPB 4 0.37*11 0.17*90.28*
Others 9 0.33** 13 0.25** 13 0.29*
Irrational belief type – 5.81 (3) 0.54 (3) – –
DEM 8 0.19** 90.32*
AFW/CAT 7 0.20** 2 0.30
FI/LFT 4 0.39** 70.40*
GE 6 0.20*2 0.24
FI/LFT vs. other 1.67 (1) 4.48 (1)*3.97 (1)*3.89 (1)*6.37 (1)*
FI/LFT 4 0.43** 60.38** 90.47** 50.30** 30.36**
Other IB types150.28** 19 0.22** 11 0.24*80.16** 3 0.00
Distress reporting – 0.69 (1) – – –
Self-reported distress 44 0.32**
Observer-reported distress 30.43*
Irrational belief specificity 0.65 (1) 0.001 (1) 0.26 (1) – –
General irrational belief 21 0.34** 45 0.33** 42 0.42**
Specific irrational belief 50.43** 20.33 20.29
Irrational belief 2.91 (1) – 0.04 (1) – –
Measured 24 0.34** 42 0.41**
Manipulated 20.61** 20.36
Time lag 0.58 (1) 1.40 (1) 0.03 (1) 0.00 (1) –
Same time 23 0.35** 42 0.34** 41 0.41** 15 0.25**
Different time 30.45 20.11 30.44*20.25
Stressful events
(A) Present vs. absent 0.14 (1) 7.87 (1)*1.67 (1) 3.17 (1) –
Present 90.38** 20.67** 80.54** 30.41**
Absent 17 0.35** 42 0.30** 36 0.38** 14 0.22**
(B) Real vs. experimental 4.77 (1)*– 0.42 (1) – –
Real 70.32** 30.59**
Experimentally induced 20.55** 50.49**
(C) Personally relevant vs. not – 0.22 (1) 1.002 (1) – –
Relevant 60.43*60.56**
Not relevant 20.26 40.39*
Age 1.12 (1) 0.008 (1) 0.10 (1) 4.57 (1)*–
Gender 0.35 (1) 0.11 (1) 0.0008 (1) 2.10 (1) –
Clinical status 0.29 (1) 0.1 (1) 0.52 (1) 10.62 (1)*–
Clinical 50.41** 12 0.35** 10 0.47** 30.54**
Non(sub)clinical 21 0.35** 35 0.32** 34 0.39** 14 0.21**
Developer/validator 0.96 (1) 5.97 (1)*0.91 (1) 6.01 (1)*–
Yes 10 0.31** 15 0.22** 14 0.34** 70.15*
No 16 0.40** 32 0.38** 30 0.44** 10 0.33**
Publication year 1.13 (1) 2.97 (1) 0.22 (1) 0.07 (1) –
Country 0.05 (1) 3.05 (1) 0.13 (1) 2.54 (1) –
US 10 0.35** 26 0.38** 29 0.42** 12 0.21**
Other countries 16 0.37** 18 0.25** 15 0.39** 50.43**
k is the number of effect sizes included in each analysis. Others = Other instruments used to measure irrational beliefs beside the core ones; DEM = de-
mandingness; AWF/CAT = awfulizing/catastrophizing; FI/LFT = frustration intolerance/low frustration tolerance; GE = global evaluation. 1 Demandingness,
awfulizing/catastrophizing, and global evaluation. * p < 0.05, ** p < 0.001.
Downloaded by:
Universitätsbibliothek Bern
198.143.58.33 - 11/27/2015 9:31:57 AM
Irrational Beliefs and Psychological
Distress
Psychother Psychosom 2016;85:8–15
DOI: 10.1159/000441231
13
ing proposed moderators: self-reported versus observer-
reported distress, general/core versus specific irrational
beliefs, naturally varying versus manipulated irrational
beliefs, time lag between irrational beliefs and distress as-
sessment, gender, publication year and publication coun-
try (p > 0.05).
Discussion
The present meta-analysis empirically tested the asso-
ciation between irrational beliefs and psychological dis-
tress. The study was based on 100 independent samples
of 83 primary studies being conducted in 13 different
countries over the last 60 years. Overall, our results cor-
roborate a moderate (overall r = 0.38) but robust relation-
ship between psychological distress and irrational beliefs.
Some authors
[39, 40] have warned against conducting
meta-analyses that include highly heterogeneous popula-
tions. Despite including heterogeneous studies reflecting
different populations with the aim of avoiding data frag-
mentation, the present meta-analysis also reported over-
all effects for the association with different emotional dis-
turbances and conducted moderation analyses within
these specific categories. The following variables were sig-
nificant moderators of the relationship between irrational
beliefs and different types of psychological distress: the
irrational belief measure chosen in a particular study, ir-
rational belief types (specifically frustration intolerance),
stressful events providing the context of assessment, age,
educational status, clinical status, and the developer/vali-
dator status of the author.
The distress measure did not significantly moderate
the association between irrational beliefs and psychologi-
cal distress. Nonetheless, when split into the various types
of distress, the irrational belief measure was a moderator
for depression and anxiety. These results may be explained
by comparing scales that contain affect-related items (e.g.
IBT) with scales that do not (e.g. IBS, ABS and SPB), the
items of which may artificially inflate the correlation be-
tween irrational beliefs and distress
[41] . In the same time,
these results could also be attributed to high reliabilities of
the used measures
[18] . In contrast to the chosen mea-
sures, the type of irrational beliefs being assessed did not
generally moderate the association between irrational be-
liefs and psychological distress significantly. These results
might suggest that irrational beliefs work as a transdiag-
nostic vulnerability and/or for some of them the self-re-
port measures might not be sensitive enough (e.g. de-
mandingness
[24] ). However, frustration intolerance
emerged to have a significantly higher correlation with all
distress types than all other irrational belief types taken
together. Further research is needed to clarify the ‘micro-
structure’ of this association, i.e. the associations of spe-
cific areas of frustration intolerance with certain distress-
ing problems
[42] . Overall, these findings are unexpected,
given that demandingness was postulated as the core ir-
rational belief in Ellis’ theory
[7] that leads to the arousal
of the other three (i.e. awfulizing, frustration intolerance,
and global evaluation). Therefore, either the centrality of
demandingness is retired as a viable hypothesis, or new
methods to assess demandingness as part of a tacit/im-
plicit process need to be developed in future research
[24] .
Regarding stressful events providing the context for
assessment, we obtained significant results for depression
and general distress. Whereas the association between ir-
rational beliefs and depression was higher when a stress-
ful event was present versus absent, a higher association
between irrational beliefs and general distress was report-
ed when the stressful event was experimentally induced
versus real/naturalistic. The latter finding was unexpect-
ed and may be explained by the novelty of the experimen-
tally induced stressful event, which can potentially in-
crease the reported level of distress. Several sample char-
acteristics emerged as significant moderators of the
irrational belief-distress association: the association be-
tween irrational beliefs and anger increases with age, it
was higher in clinical versus nonclinical samples, and the
association with anger and general distress was smaller in
students versus nonuniversity samples. Each of these as-
sociations needs to be replicated in larger samples before
further speculation about potential moderators.
There was no indication of publication bias, neither in
the overall sample nor in subsamples. In contrast to our
hypothesis, the correlation between irrational beliefs and
psychological distress was even smaller in studies con-
ducted by researchers who were developers/validators of
an irrational belief scale compared with studies con-
ducted by nondevelopers/nonvalidators. This somewhat
counterintuitive result may be explained by developers’
stricter adherence to scientific standards. It might be that
developers/validators of an irrational belief instrument
want to counter suspicions raised by what their develop-
ers/validators status might imply. At the same time, a sig-
nificantly smaller association between irrational beliefs
and psychological distress in studies conducted by re-
searchers who were developers/validators of an irrational
belief scale compared with studies conducted by nonde-
velopers/nonvalidators might be an indicator of research-
ers’ increased investment in the development of measures
Downloaded by:
Universitätsbibliothek Bern
198.143.58.33 - 11/27/2015 9:31:57 AM
Vîslă/Flückiger/grosse Holtforth/David
Psychother Psychosom 2016;85:8–15
DOI: 10.1159/000441231
14
of irrational beliefs that are minimally ‘contaminated’ by
emotional items.
Several limitations emerged in the present meta-anal-
ysis. First, despite the sporadic experimental evidence in-
cluded in this meta-analysis, the presented research does
not allow for drawing strong conclusions regarding the
causality of the relationship between irrational beliefs and
psychological distress. Irrational beliefs were manipulat-
ed before being measured in only 5 samples. As a conse-
quence, future research will need to conduct longitudinal
analyses which will allow for a better understanding of the
potentially etiopathogenetic nature of irrational beliefs
[24] . Second, the studies included in the meta-analysis
were conducted predominantly in a Western cultural
context. Therefore, future studies will need to test wheth-
er these results are also applicable to other cultural con-
texts. Third, nearly all studies included in the meta-anal-
ysis employed self-reported measures of the constructs;
future research may benefit by including measures based
on observer ratings and diagnostic interviews to further
control for possible self-report biases. Fourth, most stud-
ies were conducted in subclinical or nonclinical samples.
Fifth, this meta-analysis was focused on the associations
between irrational beliefs and psychological distress (in
our sample, we found a k of 5 studies that also reported
an association between rational beliefs and psychological
distress) and therefore excluded rational belief scales/
subscales. Rational beliefs were originally conceptualized
as low scores on irrational belief measures. Chronologi-
cally, scales that measure rational beliefs were developed
later; therefore, the available number of studies does not
yet allow for meta-analytic analyses.
The results of the present meta-analytic review [ref.
15 ,
17 , 19–23 , 25 , 26 , 41 have been included in the meta-anal-
ysis] have implications for future research. Future studies
may invest in constructing new methods for the assess-
ment of demandingness as a tacit/implicit process, may
investigate how specific components of frustration intol-
erance relate to certain distressing problems, may con-
duct more experimental studies that induce a (relevant)
stressful event and/or manipulate irrational beliefs, and
may conduct longitudinal analyses, which will allow for a
better understanding of the potentially etiopathogenetic
nature of irrational beliefs. Moreover, future research
should be conducted in different cultural contexts on dif-
ferent clinical samples and should measure more specific
irrational beliefs (e.g. cancer-related irrational beliefs).
Finally, future research will need to systematically inte-
grate other potentially relevant variables that may moder-
ate the relationship between irrational beliefs and psycho-
logical distress, such as marital status, occupational sta-
tus, and income of participants.
To conclude, this comprehensive meta-analysis inves-
tigated the robustness of the association between irratio-
nal beliefs and various types of psychological distress: a
test of one of the major basic associations in modern psy-
chology since the cognitive revolution
[2, 3] . The results
show that the overall strength of the relationship between
irrational beliefs and psychological distress is modest;
however, this relationship holds across different samples,
measurements, and study design.
Acknowledgments
This research was supported by the Scientific Exchange Pro-
gram Grant SCIEX-NMS-CH 12.319 awarded to Andreea Vîslă,
SNSF PZ00P1_136937 to Christoph Flückiger, and SNSF
PP00P1-144920/1-2 to Martin grosse Holtforth. The authors
would also like to thank Corinne Holzer for her valuable help with
the coding process.
References
1 World Health Organization (WHO): The
Global Burden of Disease – 2004 Update. Ge-
neva, WHO, 2008.
2 Cratsley K, Samuels R: Cognitive science and
explanations of psychopathology; in Fulford
K, Davies M, Gipps R, et al (eds): The Oxford
Handbook of Philosophy and Psychiatry, ed
1. New York, Oxford University, 2015, pp
413–433.
3 Chomsky N: Universals of human nature.
Psychother Psychosom 2005;
74: 263–268.
4 Ellis A: New approaches to psychotherapy
techniques. J Clin Psychol 1955;
11: 207–260.
5 Ellis A: Reason and Emotion in Psychothera-
py. Oxford, Stuart, 1962.
6 Ellis A, Bernard ME: What is rational-emotive
therapy (RET); in Ellis A, Bernard ME (eds):
Clinical Applications of Rational-Emotive
Therapy. New York, Plenum, 1985, pp 1–30.
7 Ellis A: Reason and Emotion in Psychothera-
py, revised. Secaucus, Birch Lane, 1994.
8 Beck AT: Cognitive Therapy for Emotional
Disorders. New York, International Univer-
sity, 1976.
9 Lazarus RS: Emotion and Adaptation. New
York, Oxford University, 1991.
10 Vîslă A, grosse Holtforth M, David D: De-
scriptive/inferential cognitive processes and
evaluative cognitive processes: relationships
among each other and with emotional dis-
tress. J Ration Emot Cogn Behav Ther 2015;
33: 148–159.
11 David D, Lynn SJ, Ellis A: Rational and Irra-
tional Beliefs. Research, Theory, and Clinical
Practice. New York, Oxford University, 2010.
12 Haaga D, Davison G: An appraisal of rational-
emotive therapy. J Consult Clin Psychol 1993;
61: 215–220.
Downloaded by:
Universitätsbibliothek Bern
198.143.58.33 - 11/27/2015 9:31:57 AM
Irrational Beliefs and Psychological
Distress
Psychother Psychosom 2016;85:8–15
DOI: 10.1159/000441231
15
13 Storch E, Roberti J, Roth D: Factor structure,
concurrent validity, and internal consistency
of the Beck Depression Inventory-Second
Edition in a sample of college students. De-
press Anxiety 2004;
19: 187–189.
14 Jones RG: A Factored Measure of Ellis’ Irra-
tional Belief Systems with Personality and
Maladjustment Correlated. Wichita, Test Sys-
tems, 1968.
15 Kassinove H: Self-reported affect and core ir-
rational thinking: a preliminary analysis. J Ra-
tion Emot Cogn Behav Ther 1986;
4: 119–130.
16 DiGiuseppe R, Leaf R, Exner T, Robin MW:
The development of a measure of irrational/
rational thinking. The World Congress of Be-
haviour Therapy, Edinburgh, 1988.
17 Malouff J, Schutte N: Development and vali-
dation of a measure of irrational belief. J Con-
sult Clin Psychol 1986;
54: 860–862.
18 Terjesen M, Salhany J, Sciutto M: A psycho-
metric review of measures of irrational beliefs:
implications for psychotherapy. J Ration
Emot Cogn Behav Ther 2009;
27: 83–96.
19 David D, Schnur J, Belloiu A: Another search
for the ‘hot’ cognitions: appraisal, irrational
beliefs, attributions, and their relation to
emotion. J Ration Emot Cogn Behav Ther
2002;
20: 93–131.
20 Szentagotai A, Freeman A: An analysis of the
relationship between irrational beliefs and au-
tomatic thought in predicting distress. J Cogn
Behav Psychother 2007;
7: 1–9.
21 DiLorenzo T, David D, Montgomery GH:
The impact of general and specific rational
and irrational beliefs on exam distress: a fur-
ther investigation of the binary model of dis-
tress as an emotional regulation model. J
Cogn Behav Psychother 2011;
11: 121–142.
22 Cramer D, Buckland N: Effect of rational and
irrational statements and demand character-
istics on task anxiety. J Psychol 1995;
129: 269–
275.
23 Szentagotai A, David D, Lupu V, Cosman D:
Rational emotive behavior therapy versus
cognitive therapy versus pharmacotherapy in
the treatment of major depressive disorder:
mechanisms of change analysis. Psychothera-
py (Chic) 2008;
45: 523–538.
24 David D, Freeman A, DiGiuseppe R: Rational
and irrational beliefs: implications for mecha-
nism of change and practice in psychothera-
py; in David D, Lynn SJ, Ellis A (eds): Rational
and Irrational Beliefs: Research, Theory, and
Clinical Practice. New York, Oxford Univer-
sity, 2010, pp 195–217.
25 Hayslip B Jr, Galt CP, Lopez FG, Nation PC:
Irrational beliefs and depressive symptoms
among younger and older adults: a cross-sec-
tional comparison. Int J Aging Hum Dev
1994;
38: 307–326.
26 Kassinove H, Eckhardt CI: Irrational beliefs
and self-reported affect in Russia and Ameri-
ca. Pers Individ Dif 1994;
16: 133–142.
27 Cooper H, Hedges LV, Valentine JC: The
Handbook of Research Synthesis and Meta-
Analysis. New York, Russell Sage Foundation,
2009.
28 Hunter JE, Schmidt FL: Methods of Meta-
Analysis: Correcting Error and Bias in Re-
search Findings, ed 2. Thousand Oaks, Sage,
2004.
29 Del Re AC, Hoyt WT: MAc: Meta-Analysis
with Correlations 2010. R Package Version
1.0.5. http://CRAN.R-project.org/package!
Mac.
30 Cohen J: Statistical Power Analysis for the So-
cial Sciences, ed 2. Hillsdale, Erlbaum, 1988.
31 Viechtbauer W: Conducting meta-analyses in
R with the metafor package. J Stat Softw 2010;
36: 1–48.
32 Higgins J, Thompson S: Quantifying hetero-
geneity in a meta-analysis. Stat Med 2002;
21:
1539–1558.
33 Begg C, Mazumdar M: Operating characteris-
tics of a rank correlation test for publication
bias. Biometrics 1994;
50: 1088–1101.
34 Egger M, Smith G, Schneider M, Minder C:
Bias in meta-analysis detected by a simple,
graphical test. BMJ 1997;
315: 629–634.
35 Duval S, Tweedie R: Trim and fill: a simple
funnel-plot-based method of testing and ad-
justing for publication bias in meta-analysis.
Biometrics 2000;
56: 455–463.
36 Beck AT, Steer RA, Brown GK: Beck Depres-
sion Inventory. San Antonio, Psychological
Corporation, 1996.
37 Spielberger CD, Gorsuch RL, Lushene R,
Vagg PR, Jacobs GA: Manual for the State-
Trait Anxiety Inventory (Form Y). Palo Alto,
Consulting Psychologists, 1983.
38 McNair DM, Lorr M, Droppleman LF: Profile
of Mood State. San Diego, Educational and
Industrial Testing Service, 1971.
39 Fava GA: Statistical alchemy for drug treat-
ment of generalized anxiety disorder: a com-
mentary on the meta-analysis by Baldwin et
al. [BMJ 2011; 342:d1199]. Psychother Psy-
chosom 2011;
80: 261–263.
40 Fava GA, Guidi J, Rafanelli C, Sonino N: The
clinical inadequacy of evidence-based medi-
cine and the need for a conceptual framework
based on clinical judgment. Psychother Psy-
chosom 2015;
84: 1–3.
41 Zurawski R, Smith T: Assessing irrational be-
liefs and emotional distress: evidence and im-
plications of limited discriminant validity. J
Couns Psychol 1987;
34: 224–227.
42 Harrington N: Frustration intolerance as a
multidimensional concept. J Ration Emot
Cogn Behav Ther 2007;
25: 191–211.
Downloaded by:
Universitätsbibliothek Bern
198.143.58.33 - 11/27/2015 9:31:57 AM
