Rational Emotive Behavior Therapy, Cognitive Therapy,

and Medication in the Treatment of Major Depressive

Disorder: A Randomized Clinical Trial, Posttreatment
Outcomes, and Six-Month Follow-Up
m

Daniel David and Aurora Szentagotai

Babes-Bolyai University, Romania
m

Viorel Lupu and Doina Cosman

Iuliu Hatieganu University of Medicine and Pharmacy,
Romania

A randomized clinical trial was undertaken to investigate the relative

efficacy of rational-emotive behavior therapy (REBT), cognitive
therapy (CT), and pharmacotherapy in the treatment of 170 out-
patients with nonpsychotic major depressive disorder. The patients
were randomly assigned to one of the following: 14 weeks of REBT,
14 weeks of CT, or 14 weeks of pharmacotherapy (fluoxetine). The
outcome measures used were the Hamilton Rating Scale for
Depression and the Beck Depression Inventory. No differences
among treatment conditions at posttest were observed. A larger
effect of REBT (significant) and CT (nonsignificant) over pharma-
cotherapy at 6 months follow-up was noted on the Hamilton Rating

A version of this article was presented at the European Congress of the European Association of
Behavioral and Cognitive Therapies, Thessaloniki, Greece in 2005. A manual and clinical guide based on
this research (cited herein) was published in Romania (David, 2006). Funding support was provided by the
Albert Ellis Institute, USA (Grant No. 113 to Dr. Daniel David; 20012004); the National Council for
Research (Grant No. 33374 to Dr. Daniel David; 20042006); the Romanian Center for Cognitive and
Behavioral Psychotherapies (Grant No. 9 to Dr. Daniel David). These organizations had no role in the
design and implementation of the study, data analysis, or the authorship of the manuscript.
Correspondence concerning this article should be addressed to: Daniel David, Department of Clinical
Psychology and Psychotherapy, Babes-Bolyai University, No. 37 Republicii Street 400015, Cluj-Napoca,
Cluj, Romania; e-mail: daniel.david@mssm.edu

JOURNAL OF CLINICAL PSYCHOLOGY, Vol. 64(6), 728--746 (2008) & 2008 Wiley Periodicals, Inc.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jclp.20487
 Cognitive Therapies, Medication, and Depression 729

Scale for Depression only. & 2008 Wiley Periodicals, Inc. J Clin
Psychol 64:728--746, 2008.

Keywords: major depressive disorder; rational emotive behavior

therapy; cognitive therapy; selective serotonin reuptake inhibitors;
efficacy/outcome study

Introduction
Among developed nations, major depressive disorder (MDD) is one of the leading
causes of disability. Major depressive disorder is also a tragic contributor to
mortality, with suicide representing one of the leading preventable causes of death
worldwide (U.S. Department of Health and Human Services, Substance Abuse and
Mental Health Services, 1999).
Antidepressant medication is the most widely used treatment for MDD (Olfson &
Klerman, 1993). Evidence supports the efcacy of antidepressant medication for
mildly to moderately and severely depressed patients (Thase & Kupfer, 1996). The
American Psychiatric Associations Practice Guidelines for Major Depressive
Disorder in Adults recommend the use of antidepressant medication as the rst
line therapy for patients with moderate to severe MDD (American Psychiatric
Association [APA], 2000). However, despite its success, about 3040% of patients
are nonresponsive to this therapy and therefore there is substantial room for
improvement in treatment outcomes (Antonuccio, Danton, & DeNelsky, 1995).
The eld of psychotherapy research is now at a stage where the efcacy (i.e., how
treatments work in controlled studies) and effectiveness (i.e., how treatments work in
real life) of psychological treatments have been demonstrated for a large spectrum of
disorders (see Barlow, 2001). Among psychological treatments, cognitivebehavior
therapies (CBT) are among the most empirically investigated (see Barlow, 2001;
Chambless & Hollon, 1998) although, in the case of depression, interpersonal
therapy and some forms of short-term dynamic therapy are also largely investigated
(David, 2006). Cognitivebehavior therapies are based on the premise that
psychological problems stem from dysfunctional cognitions (Beck, Rush, Shaw, &
Emery, 1979; Ellis, 1962). In CBT, the therapist works with the client to identify and
focus upon dysfunctional cognitions to modify them and remedy associated
emotional and/or behavioral consequences. Two of the most inuential and
widespread forms of CBT are cognitive therapy and rational emotive behavior
therapy (David, 2007; David & Szentagotai, 2006).

Cognitive Therapy
The cognitive model of depression as hypothesized by Beck (1976) states that
individuals have stable cognitive patterns (i.e., core beliefs), represented in our
cognitive system as schemas, that develop as a consequence of early learning. These
cognitive patterns predispose people towards negative, often distorted, descriptions
and interpretations of life events (i.e., automatic thoughts and intermediate beliefs),
depressed mood, and depressive behavior; whereas automatic thoughts are more
specically related to activating events, intermediate beliefs are more general (Haaga,
Dyck, & Ernst, 1991). Therapeutic intervention is focused on observable behavior,
automatic thoughts, intermediate and core beliefs (i.e., underlying schemas).
Journal of Clinical Psychology DOI: 10.1002/jclp
730 Journal of Clinical Psychology, June 2008

Treatment is conducted in a progressive manner so that the therapist rst focuses on

overt behavior change, then on the automatic thoughts, and nally on the
identication and modication of intermediate and core beliefs (Beck et al., 1979).
Numerous investigators have documented the clinical impact of cognitive theory and
cognitive therapy (CT) for MDD (for a review see Dobson, 1989; Hollon, Shelton, &
Davis, 1993; Hollon, Thase, & Markowitz, 2002). However, some have questioned
the state of this evidence (Hollon, Shelton, & Loosen, 1991) based on the results of
the Treatment of Depression Collaborative Research Program (TDCRP; Elkin et al.,
1989) and have criticized the theory of change as proposed by Beck et al. (1979) to
explain the efcacy of CT in the treatment of MDD (Jacobson et al., 1996).
Nevertheless, CT is still considered the gold standard for empirically validated forms
of psychotherapy in the treatment of MDD, showing short- and long-term effects
that are at least as strong as those of pharmacotherapy (medication) or other
therapies (i.e., interpersonal therapy; DeRubeis et al., 2005; Hollon et al., 2005; Shea
et al., 1992). One of the most important aspects of the efcacy of CT is that it shows
enduring effects in the prevention of relapse after treatment is terminated (Hollon et
al., 2005).
Despite the fact that there are many studies arguing for the efcacy of CT and
pharmacotherapy in the treatment of MDD, many patients are still nonresponsive
(approximately 3040%). Therefore, we believe that it would be productive to
explore new CBT treatments for MDD, with both theoretical and practical potential.

Rational Emotive Behavior Therapy

Rational Emotive Behavior Therapy (REBT) is hypothesized (Ellis, 1987) to exceed
the efcacy of other CBTs (e.g., CT) by virtue of promoting a deeper change through
(a) advocating unconditional self-acceptance; (b) focusing explicitly on reducing
secondary problems such as depression about depression (i.e., meta-emotions); and
(c) explicitly targeting demandingness (DEMimperative or absolutistic demands
on self, others, and life), which seems to be the core belief involved in MDD (e.g.,
Ellis, 1987; Solomon, Arnow, Gotlib, & Wind, 2003). Thus, in the treatment of
MDD, REBT is focused on this crucial component of depression (i.e., DEM) and its
main derivative (i.e., self-downingSD) that contribute to the understanding of the
causative factors in depression. In a reply to Ellis, Marzillier (1987) and then Brown
and Beck (1989) argue that DEM can be a component of depression, but it is not
specic and sometimes not even necessary. If DEM is identied, it is disputed/
restructured in cognitive therapy along with other schemas. In other words,
according to CT, DEM is one of the core beliefs that is sometimes readily
recognizable in depression (Beck et al., 1979). Thus, Beck et al. (1979) recognize that
cognitions collected as homework, as well as verbalizations during the therapy
session contain a high frequency of DEM. However, CT proponents also argue that
DEM is not always necessary for depression and that, when it is important, it is
identiable by standard CT strategies. On the other hand, Ellis (1987, 1994) insists
that DEM is always a component in depression and more effort than made in CT is
necessary to identify it. Sometimes, DEM is implicit or is not consciously accessible
(Dowd, 2006; Ellis, 2003). That is why in REBT, compared to CT, a special focus is
placed on identifying DEM. For example (see for details, Ellis, 2003), in REBT,
when the patient is presenting a SD belief (i.e., I am stupid because I did not pass
the exam) the therapist would advance the hypothesis that SD is a derivative of
DEM, and might infer the presence of DEM, So you think that you are stupid
Journal of Clinical Psychology DOI: 10.1002/jclp
 Cognitive Therapies, Medication, and Depression 731

because you did not pass the exam. Should you absolutely have passed it? However,
the restructuring of inferred DEM takes place only if the patient agrees with the
clinical conceptualization of the REBT therapist regarding the presence of inferred
DEM. Recently, Solomon et al. (2003), using individualized measures of irrational
beliefs (i.e., DEM) in nondepressed and formerly depressed people, have found that
DEM seems indeed to be the core belief in MDD. Rational emotive behavior
therapy has usually fared well compared to minimal treatment controls and
alternative interventions in the treatment of depression (Engels, Garnefsky, &
Diekstra, 1993; Lyons & Woods, 1991). It has typically proven superior to no
treatment or waiting list controls in the treatment of depression in college
populations (Fosterling, 1985), life problems-related depression in the normal
population (i.e., divorce-related depression; Malouff, Lanyon, & Schutte, 1988),
depressive symptoms in various subclinical and clinical contexts (depression in
individuals with the following other diagnoses: obsessivecompulsive disorder;
Emmelkamp, Visser, & Hoekstra, 1988; conduct disorder; Fava, Bless, Otto, Pava, &
Rosenbaum, 1994; Morse, Bernard, & Dennerstein, 1989; self-esteem-related
depression; Warren, McLellarn, & Ponzoha, 1988), Christian depressed clients
(Johnson, & Ridley, 1992; Johnson, Devries, & Ridley, 1994), multisymptomatic
patients (Lipsky, Kassinove, & Miller, 1980), retired clients (Caraway & Hayslip,
1985), and clinical depression (Kelly, 1982). Taken as a whole, these ndings appear
to provide a rather impressive support for the efcacy of REBT in the treatment of
depression. However, there are some aspects that need to be taken into account.
Most studies were conducted on depression measured by high scores on depression
scales in the normal population rather than by clinically relevant criteria (for a
review, see Engels et al., 1993). Moreover, even when clinically relevant populations
were used, the operationalization of depression by high scores on depression scales
rather than by clinically relevant criteria (First, Spitzer, Gibbon, & Williams, 1996)
may have yielded a mix of people with different kinds of depressive disorders. These
things undermine straightforward conclusions regarding the efcacy of REBT in the
treatment of depressive disorders in general, and MDD in particular. Given the large
number of clinical cases and previous studies of REBT (see the above review), it is
important to investigate its utility in the treatment of MDD.

Objectives
Overall, nothing is known about the efcacy of REBT alone compared to
pharmacotherapy and CT in the treatment of MDD, although its proponents
suggest that REBT could be more efcacious than CT because of its specic
techniques. Therefore, this is an important question to ask, for both theoretical and
practical implications. Given that antidepressant medication remains the current
standard treatment for depression (DeRubeis et al., 2005; Olfson & Klerman, 1993)
it is particularly important to also determine how REBT compares with clinical
pharmacotherapy. A recent study (Wang, Jia, Fang, Zhu, & Huang, 1999) suggests
that REBT in combination with amitriptyline is more effective than REBT or
amitriptyline alone in the treatment of dysthymic disorder. Macaskill and Macaskill
(1996) found that REBT in combination with Lofepramine produces more
improvement in unipolar depressed outpatients than pharmacotherapy alone. Also,
there are very few published studies comparing CT and the newer generation of
antidepressant medication (i.e., selective serotonin reuptake inhibitors; SSRIs) by
outcomes and hypothesized theory of change. Most of the previously published
Journal of Clinical Psychology DOI: 10.1002/jclp
732 Journal of Clinical Psychology, June 2008

randomized clinical trials use tricyclics and/or ignore the analysis of the mechanisms
of change. This affects the conclusions regarding the current status of CT compared
to pharmacotherapy in the treatment of MDD. Finally, there are very few published
studies investigating this topic among non-English speaking populations (and none
on a Romanian-speaking population); thus, such a study could contribute not only
to the advancement of knowledge in the eld, but also to the generalizability of the
results of research conducted on English-speaking populations.
Although we are aware that (a) in practice, clients often receive psychotherapy
while receiving concurrent pharmacotherapy, and that (b) research on the therapy
process is fundamental, we believe that the design of this study is highly informative
from the point of view of the current state of knowledge in the CBT eld. If the
efcacy of REBT and CT compared to pharmacotherapy is clinically meaningful, an
exploration of the therapy process and its cost-effectiveness will follow. To attempt
doing so in this initial study would have been premature however, given the lack of
prior empirical data regarding the above-mentioned issue. Moreover, before
combining the two treatments, it is important to know if both are active ingredients
and how they are related and compare to each other from an efcacy point of view
(see also, DeRubeis et al., 2005).
Our specic aim was to study the relative efcacy of REBT and CT for MDD in a
randomized clinical trial. Patients were randomly assigned to experimental and
control groups (i.e., the REBT group, the CT group, and the pharmacotherapy
group) to determine the efcacy of REBT in the treatment of MDD. It made sense to
examine the REBT intervention for MDD comparing its efcacy with that of the
rival intervention (i.e., CT) that has already been largely investigated in clinical
trials. Pharmacotherapy served as a reference condition in this study. These aspects
have not yet been addressed empirically in other studies.
There is growing consensus in the eld that a distinction needs to be made between
relapse and recurrence. Relapse refers to the return of symptoms associated with a
treated episode, whereas recurrence refers to the onset of a completely new episode.
Naturalistic studies indicate that a typical episode of depression can be expected to
last from 612 months with an intermorbid period of 34 years before the onset of
the next episode (Keller et al., 1984). Consequently, our study was designed to follow
participants at 6 months (see also Jacobson et al., 1996) to identify relapsed patients.

Method
Sample
A large number of patients were evaluated for the study from 2001 to 2004.
Diagnoses according to the Diagnostic and Statistical Manual of Mental Disorders,
Fourth Edition (DSM-IV; APA, 1994) were determined by use of the Structured
Clinical Interview for DSM-IV (SCID; First et al., 1996). Training and supervision
were provided by the rst author, an experienced clinical psychologist and expert
SCID-CV interviewer. Interviews were conducted by clinical psychologists, carefully
trained and supervised by Dr. David. Interrater reliability between Dr. David and
SCID-CV raters was .87, based on the percentage of times Dr. David and the rater
agreed on the primary diagnosis. The nal sample consisted of 170 participants who
met the criteria for MDD according to the DSM-IV, scored at least 20 on the Beck
Depression Inventory (BDI; Beck et al., 1979), and 14 or higher on the 17-items
Hamilton Rating Scale for Depression (HRSD; Hamilton, 1967; see also, Jacobson
et al., 1996). Patients not included in the study either did not meet the inclusion
Journal of Clinical Psychology DOI: 10.1002/jclp
 Cognitive Therapies, Medication, and Depression 733

criteria or met the exclusion criteria. Exclusion criteria (see also, Jacobson et al.,
1996) included a number of concurrent psychiatric disorders (i.e., bipolar or
psychotic subtypes of depression, panic disorder, current substance abuse, past or
present schizophrenia or schizophreniform disorder, organic brain syndrome, or
mental retardation). We also excluded participants who were in some concurrent
form of psychotherapy, who were receiving psychotropic medication, or who needed
to be hospitalized because of imminent suicide potential or psychosis.
Based on our previous experience with clinical trials as well as on the literature
(Jacobson et al., 1996), we anticipated an attrition rate of about 10%. Power
calculations for the study indicated that 52 subjects per group would maintain power
above the traditional .80 level, alpha 5 .05, to detect a medium effect size, relevant
from a clinical point of view. Therefore, the recruitment of 170 patients provided
sufcient statistical power to address the major research questions, even after
allowing for study attrition (10% of the patients).
Patients included in the study were directly referred from (a) the Romanian
Association of Cognitive and Behavioral PsychotherapiesRomanian Center for
Cognitive and Behavioral Psychotherapies, (b) the Romanian Association of
Cognitive ScienceCenter for Applied Psychology and Education, (c) the Interna-
tional Institute for the Advanced Studies of Psychotherapy and Applied Mental
Health, (d) Iuliu Hatieganu University of Medicine and Pharmacy, (e) public
service announcements, and (f) the private practice of professionals involved in the
study. The site of the research was Babes-Bolyai University, Cluj-Napoca, Romania.
The protocol was approved by the Institutional Review Board of the Babes-Bolyai
University (Psychology), and all patients provided written informed consent prior to
any research activity.
Participants qualifying for the study were randomly assigned to one of the three
study conditions (i.e., REBT, CT, Pharmacotherapy) after matching to ensure group
equivalence on the following variables, according to the established methodology
(see also, Jacobson et al., 1996): number of previous episodes of depression, presence
or absence of dystymia, gender, and marital status.
Table 1 shows the means and standard deviations for the demographic variables
and some of the associated disorders. Of the original 170 patients accepted in this
study, 113 were women and 57 were men. We rst correlated these variables with
measures of treatment outcome to determine whether they should be used as
covariates in the primary analyses. None of them (gender, education, marital status)
were signicantly correlated with either posttreatment BDI and HRSD scores, or
changes from pre- to posttreatment on these two measures of depression severity.
There were also no signicant differences among treatment conditions on any of
these variables (see Table 1), nor did the treatment groups differ in their
pretreatment BDI and HRSD scores.

Therapists
Eight experienced therapistssix psychologists and two psychiatrists(all certied
in CBT/REBT) provided treatment in the REBT and CT conditions (each in one
condition only). Their average age was 32 (range 5 2935 years), and they averaged
11 years of postdegree clinical experience (range 5 714 years). They had been
practicing psychotherapy (CT and/or REBT) for an average of 7 years since the
completion of formal training (meeting the standards of the European Association
for Behavioral and Cognitive Therapies; www.eabct.com), with a range of 3 to 10
Journal of Clinical Psychology DOI: 10.1002/jclp
734 Journal of Clinical Psychology, June 2008

Table 1
Demographics and Pretreatment Variables (Total N 5 170)

Variables REBT (N 5 57) CT (N 5 56) Pharmacotherapy (N 5 57)

Gender
Male 20 18 19
Female 37 38 38
Age, mean 7SD, (in years) 35713 39710 3772
Education
High School 27 28 26
University 30 28 31
Marital status
Never married 20 19 19
Married once 31 31 33
Divorced 5 4 3
Widowed 1 2 2
Ethnicity
Caucasian 53 54 55
Roma population 4 2 2
Mean (SD) no. of previous episodes 3.6 (4.9) 4 (4.3) 3.9 (4.8)
Dysthymia 10 8 7
Any Axis II comorbidity 15 18 13

Note. REBT 5 rational-emotive behavior therapy; CT 5 cognitive therapy.

years. Four psychiatrists provided treatment for the pharmacological group. Their
average age was 40 (range 5 3545 years), and they averaged 16 years of postdegree
clinical experience (range 5 1418 years). They had been practicing psychiatry for an
average of 10 years since the completion of their formal training, with a range of 6 to
14 years, and had participated in at least one previous clinical trial in which they
served as research consultants for the medical condition. Three manuals (they are
available upon request from Dr. David) were elaborated and published for this study
(David, 2006), one for each therapeutic condition (REBT, CT, and pharmacother-
apy, respectively) including guidelines for prescribed and proscribed interventions in
each treatment condition. A system for monitoring and calibrating for protocol
adherence was developed. The rst author listened randomly to 15% of all audio-
taped therapy sessions to check for protocol violation and for the quality of CT and
REBT interventions, using the adapted Romanian versions of the Cognitive Therapy
Rating Scale (Macavei, 2003) and the REBT Competency Scale (David, in press). In
addition, monthly meetings were held involving all therapists (including those
administering pharmacotherapy). Finally, as we describe later, adherence to
therapeutic protocol was systematically evaluated independently of these calibration
procedures.
Therapist allegiance. At the beginning of the treatment, all therapists were
required to assess (on a 5-point Likert-type scale: 1 5 not at all to 5 5 very much)
their condence in the success of all treatments. These scores were to be related to
efcacy in symptom reduction to check the impact of the therapists allegiance on the
outcomes.

Treatments
According to previous methodologies (e.g., Jacobson et al., 1996) 14 weeks clinical
trials involve a maximum of 20 individual 50-minutes therapy sessions or 20
Journal of Clinical Psychology DOI: 10.1002/jclp
 Cognitive Therapies, Medication, and Depression 735

individual 2050 minutes pharmacotherapy sessions. The REBT and CT treatment

delity was assessed by blind raters, according to the established methodology (see
Jacobson et al., 1996 for details). After the discontinuation of acute phase treatment,
patients were allowed up to three booster sessions during the 6-months follow-up
phase. These sessions could be scheduled at any time, with the provision that they be
scheduled at least one month apart. Patients and therapists were left to decide if and
when to have booster sessions. Some dyads scheduled booster sessions at regular
intervals (e.g., months 1, 3, and 6), whereas others saved their booster sessions to see
if they would be needed. Session content was left free to vary within the domain of
REBT and CT. All patients were asked not to pursue treatment for depression other
than provided in the research protocol during the 6-month follow-up phase.
Rational emotive behavior therapy condition. Treatment (20 sessions; for details,
see David, 2006) was based on the techniques and descriptions in REBT manuals
(e.g., Ellis, 1994; Walen, DiGiuseppe, & Dryden, 1992). After explaining the basic
rules of therapy (scheduling, condentiality, etc.), the rationale of REBT, the
ABDCDE model (i.e., cognitive conceptualization), the goals of REBT were
discussed with the patients. The REBT treatment was focused on the irrational
beliefs hypothesized to mediate depressive symptoms: demandingness (DEM) and
self-downing (SD); however, the other two core irrational beliefs hypothesized by
REBT to generate clinical symptomatology, awfulizing/catastrophazing (AWF) and
low frustration tolerance (LFT) were approached if identied. Cognitive, behavioral,
and emotive techniques were used to change target irrational beliefs. Automatic
thoughts, intermediate, and core beliefs (i.e., faulty inferences) were not the focus of
interventions. Also, distinctive REBT strategies were focused on (a) reducing
secondary and meta-emotional problems, (b) promoting unconditional self-
acceptance, and (c) the identication and modication of DEM as the central
irrational belief involved in depression. In REBT, if DEM is not readily recognizable
among cognitions collected as homework and verbalizations during therapy sessions,
its presence is inferred from its derivatives (i.e., self-downing, awfulizing, and low
frustration tolerance). The presence of DEM hypothesis was tested by directly asking
patients about it [e.g., patient: it is awful that I did not pass the exam (awfulizing);
therapist: it sounds like you absolutely had to pass that exam, right? (DEM)].
However, the disputation of inferred DEM was made only if the patient accepted the
clinical conceptualization including DEM. Treatment was conducted in a
progressive manner so that the therapist rst focused on overt behavior change,
then on specic irrational beliefs (e.g., it is awful that my wife is going to divorce me
and this should absolutely not happen) and then on the identication and
modication of core and general irrational beliefs (it is awful when people divorce,
and this should absolutely not happen).
Cognitive therapy condition. Treatment (20 sessions; for details, see David, 2006)
was based on the techniques and descriptions in the Beck et al. (1979) manual. The
CT treatment included behavioral activation and dysfunctional thoughts
modication, and also incorporated the identication and structural modication
of generalized core beliefs that are presumed to be the major causes of dysfunctional
thinking and depressive reactions. Treatment was conducted in a progressive manner
so that the therapist rst focused on overt behavior change, then on automatic
thoughts, and nally on the identication and modication of intermediate and core
beliefs. Demandingness was identied and disputed only if it could be revealed by
standard CT techniques. According to Beck et al. (1979), DEM is readily
Journal of Clinical Psychology DOI: 10.1002/jclp
736 Journal of Clinical Psychology, June 2008

recognizable in the cognitions collected as homework, as well as in verbalizations

during therapy sessions. No effort was made in the CT condition to infer the
presence of DEM if it was not made transparent by current CT techniques.
Pharmacotherapy condition. Patients assigned to pharmacotherapy (for details,
see David, 2006) attended about one weekly session with a psychiatrist. Initial
sessions typically lasted about 50 minutes, whereas subsequent sessions lasted at
the most 30 minutes. Treatment was focused on (a) pharmacotherapy management,
which involved educating patients about the medication, adjusting dosage and
dosage schedules, and inquiring about and dealing with side effects; and (b) clinical
management, which involved an assessment of the patients functioning in major
areas of life, brief supporting counseling, and limited advice giving. The medication
used was uoxetine, provided in exible daily dosage, typically taken in the
morning. Treatment protocol called for a beginning dose of 10 mg/d (rst day),
which was increased to 20 mg/d during week 1, and to 40 mg/d by weeks 214; the
maximum dosage of medication allowed during weeks 112 was 6080 mg/d.
During weeks 1214 the dosage was reduced to 20 mg/d, if the clinical conditions
allowed it (i.e., the patients symptoms improved). In the following 6 months,
pharmacotherapy continued, but face to face meetings were restricted to the
booster sessions. Booster sessions were focused on pharmacotherapy and clinical
management; the maximum dosage of medication allowed during the follow-up
phase was 6080 mg/d.

Measures
Masking was maintained for the measures by means of independent evaluators.
Except for the assessment phase, independent evaluators were physically isolated
from patients, data, and treating clinicians. Specic instructions were provided to the
patients, therapists, and the independent evaluators not to disclose treatment
assignments. All measures were adapted for the Romanian population and showed
good psychometric properties, comparable to those based on English-speaking
populations.
Outcome measures. All patients were evaluated before therapy, at the middle of
the treatment (week 7), at the time of termination, and at 6 months follow-up by
research assistants, blind to the status of the patient in the current study. To assess
the presence or absence of MDD, patients were examined using the following
measures:

* The Hamilton Rating Scale for Depression (HRSD; Hamilton, 1967), a 17-item,
interviewer-based measure of depression severity. Following the procedure of
Jacobson et al. (1996), the HRSD was administered as an adjunct to SCID-CV
and was inserted into a structured diagnostic interview. Similar to the English
version, the Romanian version of the HRSD has very good psychometric
properties (e.g., interrater reliability .89). Raters were not informed of the
treatment condition of the patient or of which tapes were being assessed for
reliability.
* The Beck Depression Inventory II (BDI; Beck et al., 1979) is a self-report measure
of depression that correlates highly with the HRSD. Similar to the English version
of the BDI, the Romanian version has very good psychometric properties (e.g.,
Cronbachs a 5 .84), and is sensitive to clinical change.
Journal of Clinical Psychology DOI: 10.1002/jclp
 Cognitive Therapies, Medication, and Depression 737

Safety Assessments
To ensure patient safety and evaluate the tolerability of treatment, we carefully
monitored adverse events. An adverse event was dened as any unfavorable medical
change occurring postrandomization that was accompanied by functional or clinical
impairment. This included harm and suicide-related adverse events, psychiatric
adverse events (e.g., mania, agitation), and other adverse events (e.g., headache).

Data Analysis and Statistics

Data entry and verication, data transfer, condentiality, and data analyses were
conducted under the direction of the principal investigator (Dr. David) and the
principal statistician (Dr. Radu).
First, at posttest all analyses were conducted using the intent-to-treat principle: the
analysis included all randomized patients in the treatment groups to which they had
been randomly assigned, regardless of their protocol adherence, and/or subsequent
withdrawal from treatment, assessments, or deviation from protocol. The last
available score on each outcome measure served as termination score for dropouts.
Posttest HRSD and BDI scores served as the primary measure of depression
severity. Analyses of covariance, with pretreatment scores on the dependent
measures used as covariates, were performed to compare the efcacy of the
experimental conditions. Follow-up analyses were conducted for patients who
completed treatment (dened as receiving at least 12 sessions of treatment).
Treatment response at posttest and follow-up was also analyzed categorically.
Participants were considered standard improved (response criterion) if they had a
HRSD score of less than 12; see also, Hollon et al., 2005). Recovery/(Remission) was
dened as no MDD, and HRSD scores less than 7. Relapse was dened as meeting
MDD criteria at follow-up, although standard improvement was observed at
posttreatment. Contingency table analyses were used to compare treatments in
improvement and recovery rates (see Table 4 for details). An alpha level of .05 was
used for all statistical tests.
Because we used a prospective design, we expected missing data, which could
negatively impact on the validity of our conclusions. Two questions were therefore
considered during data analyses: (a) Are the missing data mechanisms in this study
random or nonrandom, relevant or irrelevant? (b) How sensitive will the results of
the study be to different methods of analysis that make different assumptions about
missing data mechanisms? If the results would be consistent across different methods
of analysis, which make reasonable assumptions about missing data mechanisms, we
could be condent about the conclusions. If the results would be sensitive to the
assumptions, we would need to further explore them, by either obtaining additional
information or by indicating our uncertainty regarding the conclusions.
Univariate analyses showed that the data were suitable for further analyses,
presented as follows. The ow diagram bellow illustrates the progress through the
phases of the trial (see Figure 1).

Results
Adherence to Treatment Protocol
Following the procedure of Jacobson et al. (1996) the measure of treatment integrity
used in the present study was based on a modied version of the National Institute of
Mental Health Collaborative Study Psychotherapy Rating Scale (Hollon, Evans,
Journal of Clinical Psychology DOI: 10.1002/jclp
738 Journal of Clinical Psychology, June 2008

Figure 1. Flow diagram of the progress through the phases of the trial.

Elkin, & Lowery, 1984). Items included both techniques that were part of the
treatment manual, and techniques prohibited by it. Ideally, the main differences
between REBT and CT would be observed on items reecting interventions
addressing the modication of automatic thoughts, intermediate/core beliefs, and
irrational beliefs, as both conditions included behavioral activation. Our scale
contained 5 items measuring the use of interventions focused on behavioral
activation, 5 measuring focus on automatic thoughts, 5 measuring focus on
intermediate/core beliefs, 5 measuring focus on irrational beliefs, and 3 items
reecting interventions that are proscribed in both conditions. Raters listened to an
audio-taped therapy session, taking notes as they listened, and then rated each item
on a scale ranging from 0 (not al all) to 6 (extensively). Ten patients were selected for
adherence ratings from each condition. One early, one middle, and one late session
Journal of Clinical Psychology DOI: 10.1002/jclp
 Cognitive Therapies, Medication, and Depression 739

Table 2
The Adherence to Clinical Protocols Analysis
Behavioral Behavioral Automatic Automatic Core Core Irrational Irrational
activation activation thoughts thoughts beliefs beliefs beliefs beliefs Prohibited Prohibited
CT REBT CT REBT CT CT CT REBT CT REBT

M 5 4,9 M 5 5.1 M 5 5.1 M 5 1.1 M 5 4.9 M 5 1 M 5 2.5 M 5 5.5 M 5 .2 M 5 .2

SD 5 .73 SD 5 .99 SD 5 .73 SD 5 .73 SD 5 .73 SD 5 .66 SD 5 .70 SD 5 .70 SD 5 .42 SD 5 .41

Note. REBT 5 rational-emotive behavior therapy; CT 5 cognitive therapy.

were randomly selected for each of these patients, excluding sessions 1 and 20.
Treatment condition was kept masked to trained coders. Intraclass correlation
coefcients were used to determine interrater reliability. The mean intraclass
correlation was .87, coefcients ranging from .79 to. 89. across all scales. As Table 2
shows, therapists were successful at keeping the treatments distinct.
In addition, the rst author, who provided supervision for all therapists in the
project, randomly selected 15% of the audio-taped sessions in the CT and REBT
conditions and rated them for competence on the Cognitive Therapy Rating Scale
(CTS) and the Rational-Emotive Behavior Therapy Competency Scale (RCS). The
convention is to use a score of 40 as the cutoff for competence on the CTS and RCS.
The overall means were above 40, as were the means for each therapist.
Attrition. One hundred fty-one patients completed therapy (dened as receiving
at least 12 sessions of treatment). During the rst 7-week treatment period, the
overall attrition rate in the sample was 9%. Attrition was 12% in the medication
condition, 7% in the CT condition, and 7% in the REBT condition. Two patients in
the medication condition and one in the CT condition withdrew consent immediately
following randomization, while others (3 medication, 3 CT, and 4 REBT) withdrew
consent during treatment without stating a specic reason. Over the following 7
weeks of the trial, 3 patients in the medication condition dropped out because of the
adverse effects (i.e., sexual side effects and nausea) or because they were no longer
interested in treatment, whereas only 2 patients in the CT condition and 1 in the
REBT condition dropped out during this period. Thus, over the 14-weeks treatment
course, attrition rate was 14% for medication, 10% for the CT condition, and 9%
for the REBT condition.
Therapist allegience. For each experimental condition (i.e., CT, REBT,
pharmacotherapy) we conducted correlation analyses among therapists
expectations regarding its efcacy and BDI and HRSD posttest scores of the
patients treated in that condition. We found no signicant correlations (all ps4.05)
among these variables.
Missing data. We minimized the chance of missing data by avoiding unobserved
measurements as much as possible, and by encouraging the retrieval of data after the
patients dropout. The analysis of missing data showed that the proportion and the
time of appearance of missing values did not differ among treatment groups. Missing
or incomplete data were imputed with the average score of the completed items when
no more than four items were missing. Also, results of the sensitivity analysis (data
imputed with the average score of the completed items versus data imputed with the
Journal of Clinical Psychology DOI: 10.1002/jclp
740 Journal of Clinical Psychology, June 2008

Table 3
Means (and Standard Deviations) of the Primary Outcome Variables (Total N 5 170)

CT REBT Pharmacotherapy

Total sample Dropouts Total sample Dropouts Total sample Dropouts

N 5 56 N56 N 5 57 N55 N 5 57 N58

HRSD
Pre 22.9 (7.02) 21.5 (5) 23.1 (7.6) 22.4 (7) 21.4 (8.03) 21.5 (7)
7 weeks 11.6 (6.6) 12.3 (7.7) 16.8 (4) 12.5 (6.5)
Post 8.6 (7.3) 18.4 (3) 8.8 (7.4) 8.8 (6.7) 16.1 (4)
6 months 7 (6.6) 6.8 (6.4) 9.8 (5.5)
BDI
Pre 29.9 (9.47) 29 (8.8) 32.1 (11) 30.8 (9) 30.6 (11.3) 29.8 (8)
7 weeks 14.2 (6.3) 14 (6.3) 15 (6.3) 13.5 (6.5)
Post 10.4 (7.1) 18 (5.6) 10.4 (6.9) 10.9 (6.8) 16 (6)
6 months 9.6 (6.6) 9.2 (6.2) 10.8 (6.1)

Note. REBT 5 rational-emotive behavior therapy; CT 5 cognitive therapy; HRSD 5 Hamilton Rating
Scale for Depression; BDI 5 Beck Depression Inventory.

Table 4
Unimproved, Improved, Recovered, and Relapsed Patients at Posttreatment and 6-Month
Follow-Up

Posttreatment (PT) 6-Month Follow-Up (FWU)

Type of clinical response REBT CT Pharmacotherapy REBT CT Pharmacotherapy

N 5 57 N 5 56 N 5 57 N 5 48 N 5 49 N 5 47
Unimproved 20 21 24 12 16 15
Improved 37 35 33 36 33 32
Recovered (Remission) 25 28 27 25 24 19
1 3 5
Relapsed

Note. REBT 5 rational-emotive behavior therapy; CT 5 cognitive therapy. Standard improvement

(response criteria) is dened as a HRSD (Hamilton Rating Scale for Depression) score of less than 12.
Recovered (remission) is dened as no major depressive disorder (MDD), and a HRSD score of less
than 7. Relapsed is dened as meeting MDD criteria at follow-up, although standard improvement was
observed at posttreatment.

average score of the item across patients in the same treatment condition) were
consistent and led to reasonably similar estimates of treatment effects.

Medication dosage. The mean (7SD) daily uoxetine dosage during week 1 was
2074.1, and 50.174.5 during weeks 212. The dosage was reduced to 20 mg during
weeks 1214 for 53% of the patients, who tted the standard improvement criterion
(HRSDo12). At week 14, 59% of patients showed standard improvement
(HRSDo12). Patients who completed treatment continued the pharmacotherapy
during the 6-months follow-up period. The mean (7SD) daily uoxetine dose during
the 6-month follow-up period was 26.575.5.
Journal of Clinical Psychology DOI: 10.1002/jclp
 Cognitive Therapies, Medication, and Depression 741

Treatment Outcome: Middle of the Treatment (7 Weeks), Posttreatment and 6-Month

Follow-Up
The treatment outcome analyses considered the three treatment conditions, with
BDI and HRSD scores serving as dependent variables, and pretest scores on the
respective pretest measures serving as covariates. Table 3 presents the means and
standard deviations of our outcome analyses. Results are presented for the total
sample (including dropouts).
We also looked at the proportion of improved and recovered patients in each
condition, to assess the clinical signicance of each treatment condition at
posttreatment and at 6-month follow-up. Table 4 presents the improvement and
recovery rates for each treatment condition in each of the three samples. There are
no signicant differences at posttreatment between treatments on improvement or
recovery rates in any of the three samples (all ps4.05). However, at 6-month follow-
up, the number of relapsed patients was higher in the medication condition than in
the REBT or CT conditions.
Let us present these results in more detail.

Pretreatment
Pretreatment group differences were assessed by one-way analyses of variance
(ANOVAs) and we found no signicant pretreatment differences among conditions
(all ps4.05) on the BDI, F(2, 167) 5 0.60, p4.05, and HRSD, F(2, 167) 5 0.87,
p4.05.

Outcome at 7 Weeks
Continuous analyses. Continuous data analyses at 7 weeks failed to show any
signicant differences on the total sample among study conditions on the BDI, F(2,
167) 5 0.46, p4.05, and HRSD, F(2, 167) 5 0.56, p4.05.
Categorical analyses. Response rates at 7 weeks (HRSDo12) were 53% in the
pharmacotherapy condition, 59% in the REBT condition, and 59% in the CT
condition.

Outcome at 14 Weeks
Continuous analyses. At posttreatment (see Table 3), continuous data analyses
failed to show any signicant differences on the total sample among study conditions
(all ps4.05) on BDI, F(2, 167) 5 1.06, p4.05, and HRSD, F(2, 167) 5 0.98, p4.05].
Categorical analyses. Response rates at 14 weeks (HRSDo12) and recovery
(HRSDo7) were a 59% response rate and 50% recovery in the pharmacotherapy
condition, a 65% response rate and 45% recovery in REBT, and a 63% response
rate and 50% recovery in the CT condition.

Outcome at 6-Month Follow-Up

Continuous analyses. At 6-month follow-up (see Table 3), the level of depression
measured by the HRSD was higher in the pharmacotherapy condition than in the
REBT condition (po.05) and the CT condition (nonsignicant), F(2, 143) 5 3.57,
po.05; the effect size of the comparison of REBT with pharmachotherapy on the
HRSD was a medium one, d 5 0.54 (Cohens estimate). There was no difference
Journal of Clinical Psychology DOI: 10.1002/jclp
742 Journal of Clinical Psychology, June 2008

between the REBT condition and the CT condition at 6-month follow-up on the
BDI or HRSD (all ps4.05).

Categorical analyses. Response rates (HRSDo12) and recovery (HRSDo7) at

follow-up were a 68% response rate and 40% recovery in the pharmacotherapy
condition, a 75% response rate and 52% recovery in REBT, and a 67% response
rate and 51% recovery in the CT condition.

Adverse Events
As expected, adverse events were more common in patients receiving uoxetine (9/
49) compared to REBT (0/52) and CT (1/50). For patients receiving uoxetine, the
adverse events were of a psychiatric nature as follows: 1 patientpanic attacks; 2
patientsanxiety and insomnia; 1 patientcrying and anger; 2 patientsrest-
lessness; 3 patientsinsomnia. The adverse event of the patient receiving CT was
also of psychiatric type, namely, insomnia. All patients with reported adverse events
responded to modications accepted within the current protocols (e.g., dose
reduction, cognitive and behavioral techniques).

Discussion and Conclusions

The outcome analyses of this study suggest that REBT, CT, and pharmacotherapy
are equally efcient at posttreatment for patients suffering from major depressive
disorder. However, at 6-month follow-up, REBT and CT seem slightly more efcient
than pharmacotherapy (but only REBT was signicantly better than pharmacother-
apy, and only as assessed by the HRSD). These results are consistent with both
previous (see Beck et al., 1979) and more recent ndings in the eld (e.g., DeRubeis
et al., 2005), and we can extend them to non-English speaking patients. Thus, taking
into account that the sample included mild, moderate, and severe major depressive
disorder we can say that REBT and CT may be as efcient as medication, and
possibly even more efcient at 6-month follow-up, in the initial treatment of patients
suffering from nonpsychotic major depressive disorder. The allegiance effect seems
to have no impact on the outcome.
Having formulated this conclusion, several important issues need to be discussed.
First, both psychological treatments have an important common component:
behavioral activation. Jacobson et al. (1996) proved that behavioral activation is as
effective as the full cognitive therapy package in the treatment of MDD. Therefore, it
is possible that the similarity of the results of CT and REBT in the treatment of
MDD is related to this common component. Indeed, Dimidjian et al. (2006) have
found recently that behavioral activation is comparable to medication and even
more efcient than cognitive therapy among severely depressed patients. Future
studies should investigate the contribution of individual components of the
intervention (e.g., behavioral activation versus cognitive restructuring) to the
benecial effects of the intervention. Second, the standard pharmacotherapy
treatment includes brief supportive counseling and limited advice giving. This is a
large amount of clinical attention and the potential impact of this attention on the
outcome is important when comparing pharmacotherapy with psychotherapy.
Third, the fact that the results of REBT and CT are similar suggests that the
cognitions targeted in each therapy may be related; we have an article in preparation
that explores the mechanisms/theory of change involved in the results of this
outcome study.
Journal of Clinical Psychology DOI: 10.1002/jclp
 Cognitive Therapies, Medication, and Depression 743

The limitations of this study are inherent to the research questions, design, and
methods that were selected. Three are most important. First, our pharmacotherapy
protocol is based on the American Psychiatric Associations 2000 clinical guidelines,
which (a) do not straightforwardly stress the role of continuation/maintenance of
treatment in the case of uoxetine, (b) allow reacting to nonrespondent patients after
48 weeks of treatment with increased dosage, and (c) allow a reduction of the
dosage in the continuation phase with a careful evaluation of the patients condition.
Later updated guidelines (i.e., see American Psychiatric Associations guidelines at
www.psych.org) strongly encourage the continuation of treatment based on the
dosage of the acute treatment, no matter the antidepressive medication, and suggest
more diverse methods and strategies (e.g., with lithium carbonate augmentation) for
nonresponsive patients. Thus, our pharmacotherapy protocol, designed in 2001,
does not exactly t the current guidelines, and this could be an important limitation
of the study. However, the rate of improvement in our pharmacotherapy condition is
similar to that reported by using pharmacotherapy based on the new guidelines
(e.g., DeRubeis et al., 2005) and there are some studies showing that continued
treatment with low-dose uoxetine (i.e., 20 mg/d) may be as effective as an
increase in dosage in achieving a favorable clinical outcome (Schweizer et al., 1990).
These aspects allow us to interpret our ndings on reasonable grounds, despite
the limitation of the pharmacotherapy protocol. Second, although patients
exhibited the full range of mild to moderate and severe MDD, with a mean illness
severity indicating moderate toward severe MDD, the generalizability may not be as
high as desired because we excluded some of the comorbidity; however, this
increased the internal validity of the study, useful for theory of change analysis.
Third, some of our conclusions seem to be based on the null hypotheses; however,
taking into account the acceptable statistical power and that these conclusions are
fully consistent with those reported in other studies, their validity is enhanced by this
consistency.
The conclusions we can delineate corroborating the outcome analyses is that
REBT and CT have an enduring effect that extends beyond the end of treatment,
comparable to (at postacute treatment) or even slightly better (at 6-month follow-up)
than pharmacotherapy, and therefore they can be considered (also based on the
patients choice), a rst-line therapy for patients with nonpsychotic major depressive
disorder. Our future studies (in preparation) will explore the theory/mechanism of
change and cost-effectiveness related to these outcomes. Overall, we believe that the
integration of REBT and CT strategies into a theoretically coherent CBT package,
covering all types of cognitions, would be important to develop and test. This is
particularly needed as classic CT strategies do not pay enough attention to the
change of irrational beliefs, whose restructuring process seems to be enough to
change depressed mood. Newer CT strategies, based on more recent manuals (e.g.,
Beck, 1995), which more clearly incorporate the irrational beliefs component at the
intermediate and core beliefs levels, have proved more efcient than pharmacother-
apy in the relapse prevention of MDD at one year follow-up (Hollon et al., 2005).
Thus, a theoretically informed CBT package, focused on automatic thoughts,
intermediate/core beliefs, and irrational beliefs should be explored (see also
Szentagotai, 2006; Szentagotai & Freeman, 2007); we do this in our article in
preparation regarding the theory of change analysis based on the data of this study.
However, the CBT package should also be exible enough to allow the choice of a
specic CBT strategy (REBT and/or CT) for certain types of patients and
psychopathology.
Journal of Clinical Psychology DOI: 10.1002/jclp
744 Journal of Clinical Psychology, June 2008

References
Antonuccio, D.O., Danton, W.G., & DeNelsky, G.Y. (1995). Psychotherapy versus
medication for depression: Challenging the conventional wisdom with data. Professional
Psychology: Research and Practice, 6, 574585.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental
disorders (4th ed.). Washington, DC: Author.
American Psychiatric Association. (2000). APA practice guidelines for major depressive
disorder (2nd ed.). Washington, DC: American Psychiatric Press.
Barlow, D.H. (2001). Clinical handbook of psychological disorders. A step-by-step treatment
manual. New York: Guilford Press.
Beck, A.T. (1976). Cognitive therapy for emotional disorders. New York: International
University Press.
Beck, A.T., Rush, A.J., Shaw, B.F., & Emery, G. (1979). Cognitive therapy of depression.
New York: Guilford Press.
Beck, J.S. (1995). Cognitive therapy: Basic and beyond. New York: Guilford Press.
Brown, G.B., & Beck, A.T. (1989). The role of imperatives in psychopathology: A reply to
Ellis. Cognitive Therapy and Research, 13, 315321.
Caraway, M., & Hayslip, B. (1985). Facilitating rational thinking in older persons. Clinical
Gerontologist, 4, 4850.
Chambless, D.L., & Hollon, S.D. (1998). Dening empirically supported treatments. Journal
of Consulting and Clinical Psychology, 66, 719.
David, D. (in press). The REBT competency scale. Journal of Cognitive and Behavioral
Psychotherapies.
David, D. (2006). Evidence-based treatments for major depressive disorder. Bucharest:
Tritonic.
David, D. (2007). Quo vadis CBT? Trans-cultural perspectives on the past, present, and future
of cognitive-behavioral psychotherapies: Interviews with the current leadership in
cognitivebehavioral psychotherapies. Journal of Cognitive and Behavioral Psychothera-
pies, 7, 171219.
David, D., & Szentagotai, A. (2006). Cognition in cognitive-behavioral psychotherapies
(CBT): Toward an intergrative model. Clinical Psychology Review, 3, 284298.
DeRubeis, R.J., Hollon, S.D., Amsterdam, J.D., Shelton, R.C., Young, P.R., Salomon, R.M.,
et al. (2005). Cognitive therapy vs medications in the treatment of moderate to severe
depression. Archives of General Psychiatry, 62, 409416.
Dimidjian, S., Dobson, K.S., Kohlenberg, R.J., Schmaling, K.B., Kohlenberg, R.J., Addis,
M.E., et al. (2006). Randomized trial of behavioral activation, cognitive therapy, and
antidepressant medication in the acute treatment of adults with major depression. Journal
of Consulting and Clinical Psychology, 74, 658670.
Dobson, K.S. (1989). A meta-analysis of the efcacy of cognitive therapy for depression.
Journal of Consulting and Clinical Psychology, 57, 414419.
Dowd, T. (2006). What change in cognitive therapy? The role of tacit knowledge structures.
Journal of Cognitive and Behavioral Psychotherapies, 6, 141148.
Elkin, I., Shea, T., Watkins, J., Imber, S., Sotsky, S., Collins, J., et al. (1989). The National
Institute of Mental Health Treatment of Depression Collaborative Research Program:
General effectiveness of treatments. Archives of General Psychiatry, 46, 971982.
Ellis, A. (1962). Reason and emotion in psychotherapy. Secaucus, NJ: Citadel.
Ellis, A. (1987). A sadly neglected cognitive element in depression. Cognitive Therapy and
Research, 11, 121145.
Ellis, A. (1994). Reason and emotion in psychotherapy (Rev. ed.). Secaucus, NJ: Birch Lane.
Journal of Clinical Psychology DOI: 10.1002/jclp
 Cognitive Therapies, Medication, and Depression 745

Ellis, A. (2003). Similarities and differences between rational emotive behavior therapy and
cognitive therapy. Journal of Cognitive Psychotherapy, 17, 225240.
Emmelkamp, P.M., Visser, S., & Hoekstra, R.J. (1988). Cognitive therapy vs. exposure in vivo
in the treatment of obsessive-compulsives. Cognitive Therapy and Research, 12, 103114.
Engels, G.I., Garnefsky, N., & Diekstra, R.F.W. (1993). Efcacy of rational-emotive therapy:
A quantitative analysis. Journal of Consulting and Clinical Psychology, 61, 10831091.
Fava, M., Bless, E., Otto, M.W., Pava, J.A., & Rosenbaum, J.F. (1994). Dysfunctional
attitudes in major depression: Changes with pharmacotherapy. Journal of Nervous and
Mental Disease, 182, 4549.
First, M.B., Spitzer, R.L., Gibbon M., & Williams, J.B.W. (1996). Structured Clinical
Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV). Washington, DC:
American Psychiatric Press.
Fosterling, F. (1985). Rational emotive therapy and attribution theory: An investigation of the
cognitive detriments of emotions. British Journal of Cognitive Psychotherapy, 3, 1225.
Haaga, D.A.F., Dyck, M.J., & Ernst, D. (1991). Empirical status of cognitive theory of
depression. Psychological Bulletin, 110, 215236.
Hamilton, M. (1967). Development of a rating scale for primary depressive illness. Journal of
Social and Clinical Psychology, 6, 278296.
Hollon, S.D., DeRubeis, R.J., Shelton, R.C., Amsterdam, J.D., Salomon, R.M., OReardon,
J.P., et al. (2005). Prevention of relapse following cognitive therapy vs medications in
moderate to severe depression. Archives of General Psychiatry, 62, 417422.
Hollon, S.D., Thase, M.E., & Markowitz, J.C. (2002). Treatment of prevention of depression.
Psychological Science in the Publica Interest, 3, 3977.
Hollon, S.D., Evans, M.D., Elkin, I., & Lowery, A. (1984, August). System for rating
therapies for depression. Paper presented at the 92nd Annual Convention of the American
Psychological Association, Toronto, Ontario, Canada.
Hollon, S.D., Shelton, R.C., & Davis, D.D. (1993). Cognitive therapy for depression:
Conceptual issues and clinical efcacy. Journal of Consulting and Clinical Psychology, 61,
270275.
Hollon, S.D., Shelton, R.C., & Loosen, P.T. (1991). Cognitive therapy and pharmacotherapy
for depression. Journal of Consulting and Clinical Psychology, 59, 8899.
Jacobson, N.S., Dobson, K.S., Truax, P.A., Addis, M.E., Kowener, A.K., Gollan, J.K., et al.
(1996). A component analysis of cognitive-behavioral treatment for depression. Journal of
Consulting and Clinical Psychology, 64, 295304.
Johnson, W.B., & Ridley, C.R. (1992). Brief Christian and non-Christian rational-emotive
therapy with depressed Christian clients: An exploratory study. Counseling and Values, 36,
220229.
Johnson, W.B., Devries, R., & Ridley, C.R. (1994). The comparative efcacy of Christian and
secular rational-emotive therapy with Christian clients. Journal of Psychology and
Theology, 22, 130140.
Keller, M.G., Klerman, G.L., Lavori, P.W., Coryell, W., Endicott, J., & Taylor, J., et al.
(1984). Long-term outcome of episodes of major depression: Clinical and public health
signicance. Journal of the American Medical Association, 252, 788792.
Kelly, L.M. (1982). Rational emotive therapy vs Lewinsohnsion based approaches to
treatment of depression. Dissertation Abstracts International, 43, 1986.
Lipsky, M.J., Kassinove, H., & Miller, N.J. (1980). Effects of rational-emotive therapy,
rational role reversal, and rational-emotive imagery on the emotional adjustment of
community mental health center patients. Journal of Consulting and Clinical Psychology,
48, 366374.
Lyons, L.C., & Woods, P.J. (1991). The efcacy of rational-emotive therapy: A quantitative
review of the outcome research. Clinical Psychology Review, 11, 357369.
Journal of Clinical Psychology DOI: 10.1002/jclp
746 Journal of Clinical Psychology, June 2008

Macaskill, N.D., & Macaskill, A. (1996). Rational-emotive therapy plus pharmacotherapy

versus pharmacotherapy alone in the treatment of high cognitive dysfunction depression.
Cognitive Therapy & Research, 20, 575592.
Malouff, J.M., Lanyon, R.I., & Schutte, N.S. (1988). Effectiveness of a brief group RET
treatment for divorce-related dysphoria. Journal of Rational Emotive & Cognitive
Behavior Therapy, 6, 162171.
Marzillier, J. (1987). A sadly neglected cognitive element in depression: A reply to Ellis.
Cognitive Therapy and Research, 11, 147151.
Morse, C., Bernard, M.E., & Dennerstein, L. (1989). The effects of rational-emotive therapy &
relaxation training on premenstrual syndrome. Journal of Rational-Emotive & Cognitive-
Behavior Therapy, 7, 98110.
Olfson, M., & Klerman, G.L. (1993). Trends in the prescription of antidepressants by ofce-
based psychiatrists. American Journal of Psychiatru, 150, 571577.
Schweizer, E., Rickels, K., Amsterdam, J.D., Fox, I., Puzzuoli, G., & Weise, C. (1990). What
constitute an adequate antidepressant trial for uoxetine? Journal of Clinical Psychiatry,
51, 811.
Shea, M.T., Elkin, I., Imber, S.D., Sotsky, S.M., Watkins, J.T., Collins, J.F., et al. (1992).
Course of depressive symptoms over follow-up: Findings from the National Institute of
Mental Health Treatment of Depression Collaborative Research Program. Archives of
General Psychiatry, 49, 782787.
Solomon, A., Arnow, B.A., Gotlib, I.H., & Wind, B. (2003). Individualized measurement of
irrational beliefs in remitted depressives. Journal of Clinical Psychology, 59, 439455.
Szentagotai, A. (2006). Irrational beliefs, thought suppression and distress. Journal of
Cognitive and Behavioral Psychotherapies, 6, 119129.
Szentagotai, A., & Freeman, A. (2007). An analysis of the relationship between irrational
beliefs and automatic thoughts in predicting distress. Journal of Cognitive and Berhavioral
Psychotherapies, 7, 19.
Thase, M.E., & Kupfer, D.J. (1996). Recent developments in the pharmacotherapy of mood
disorders. Journal of Consulting and Clinical Psychology, 64, 646659.
U.S. Department of Health and Human Services, Substance Abuse and Mental Health
Services (USDHHS). (1999). Mental health: A report of the surgeon general-executive
summary. Rockville, MD: U.S. Department of Health and Human Services, Substance
Abuse and Mental Health Services, National Institute of Mental Health.
Walen, S.R., DiGiuseppe, R., & Dryden, W. (1992). A practitioners guide to rational-emotive
therapy (2nd ed.). New York: Oxford University Press.
Wang, C., Jia, F., Fang, R., Zhu, Y., & Huang, Y. (1999). Comparative study of rational-
emotive therapy for 95 patients with dysthymic disorder. Chinese Mental Health Journal,
13, 172173.
Warren, R., McLellarn, R.W., & Ponzoha, C. (1988). Rational-emotive therapy vs. general
cognitive behavior therapy in the treatment of low self-esteem and related emotional
disturbances. Cognitive Therapy and Research, 12, 2137.

Journal of Clinical Psychology DOI: 10.1002/jclp

The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publica