1

British Journal of Clinical Psychology (2019), 58, 1–18

© 2018 The British Psychological Society
www.wileyonlinelibrary.com

Does the addition of cognitive therapy to exposure

and response prevention for obsessive compulsive
disorder enhance clinical efficacy? A randomized
controlled trial in a community setting
Neil A. Rector1,2* , Margaret A. Richter1,2, Danielle Katz1 and
Michelle Leybman3
1
Frederick W. Thompson Anxiety Disorders Centre, Sunnybrook Health Sciences
Centre, Toronto, Ontario, Canada
2
Department of Psychiatry, University of Toronto, Ontario, Canada
3
Centre for Addiction and Mental Health, Toronto, Ontario, Canada

Objectives. Exposure and response prevention (ERP) remains the most empirically
supported psychological treatment for obsessive compulsive disorder (OCD). Clinical
guidelines recommend the addition of cognitive approaches to ERP although the
presumed additive benefits have not been directly tested. The aim of this was to compare
a treatment that integrated cognitive therapy with ERP (ERP + CT) to traditional,
manualized ERP to test the additive benefits.
Design. A longitudinal, randomized control trial design was used.
Methods. Participants (N = 127) with OCD were randomly assigned to receive
individual outpatient ERP or ERP + CT. Obsessive-compulsive symptom severity
measures were completed pre- and post-treatment and at 6-month follow-up.
Results. While both conditions led to significant symptom and obsessive belief
reduction, ERP + CT led to significantly greater symptom and belief reduction as
compared to ERP across all main symptom presentations of OCD. Based on a priori
definitions of effectiveness, more patients in ERP + CT compared to the ERP group were
also deemed treatment responders.
Conclusions. The results of this study suggest that cognitive therapy can be readily
integrated with ERP to improve clinical outcomes beyond ERP alone.

Practitioner points
 Both ERP and ERP + CT were effective, however a course of ERP + CT was significantly more
effective at reducing symptoms of OCD than the ERP treatment condition.
 Significantly more participants who received ERP + CT experienced clinically significant change in
OCD symptoms compared to those who received ERP.
 OCD symptom dimension did not significantly impact response to either ERP or ERP + CT
treatments.

*Correspondence should be addressed to Neil A. Rector, Sunnybrook Research Institute & Department of Psychiatry, Sunnybrook
Health Sciences Centre, 2075 Bayview Ave., Toronto, ON M4N 3M5, Canada (email: neil.rector@sunnybrook.ca).

DOI:10.1111/bjc.12188
2 Neil A. Rector et al.

Exposure and response prevention (ERP) treatment is the most empirically supported
psychotherapeutic intervention for obsessive compulsive disorder (OCD) and is
considered a frontline treatment in international clinical guidelines (American
Psychiatric Association, 2007; Katzman et al., 2014; NICE, 2005). Studies demonstrate
that ERP leads to clinically significant and lasting change in OCD symptoms
irrespective of symptom presentation and severity, timing of delivery (intensive vs.
outpatient), and comorbidity status (Eddy, Dutra, Bradley, & Westen, 2004; Olatunji,
Davis, Powers, & Smits, 2013). Cognitive therapy (CT) is less frequently examined and
is less empirically established as a treatment for OCD, although it has also been found
to be an effective treatment for OCD (e.g., Anholt et al., 2007; Cottraux et al., 2001;
Whittal, Thordarson, & McLean, 2005; Whittal, Woody, McLean, Rachman, &
Robichaud, 2010). While the aim of ERP is to reduce anxiety through prolonged
exposure to anxiety-inducing stimuli while preventing overt or covert rituals, the aim
of CT for OCD is to reduce obsessions and compulsions by targeting the maladaptive
appraisals and dysfunctional beliefs that are implicated in the development and
maintenance of obsessive-compulsive (O-C) symptoms. In the past two decades,
cognitive features have been demonstrated to be associated with the maintenance of
obsessions and compulsions (Obsessive Compulsive Cognitions Working Group, 1997,
2001; Rachman, 1997; Salkovskis, 1985) and have been directly targeted in theoretically
based manualized CT approaches for OCD (Clark, 2004; Steketee, Frost, Rh
eaume, &
Wilhelm, 1998).
The majority of controlled trials and meta-analytic summaries have examined ERP
versus cognitive therapy as two distinct treatments for OCD. When compared, no
differences have been found, with both treatments demonstrating significant and
equivalent efficacy (Eddy et al., 2004; Olatunji et al., 2013; Rosa-Alc
azar, S
anchez-Meca,
G
omez-Conesa, & Mar
ın-Mart
ınez, 2008). However, an alternative approach that has been
proposed in prominent treatment guidelines (NICE, 2005) is to integrate CT and ERP into a
comprehensive integrated cognitive behavioural therapy (CBT) approach. With varying
degrees of CT/ERP integration (Cordioli et al., 2003; O’Connor et al., 2006), clinical
effectiveness of the approach has been demonstrated (Anderson & Rees, 2007; Braga
et al., 2016; Cordioli et al., 2003; Freeston et al., 1997; Gomes et al., 2016; Jaurrieta
et al., 2008; O’Connor, Todorov, Robillard, Borgeat, & Brault, 1999; O’Connor et al.,
2006). As well, a recent mega-analysis combining information from both controlled
trials and clinical settings found that integrated CBT, CT, and ERP all had large effect sizes
(Steketee, Siev, Yovel, Lit, & Wilhelm, 2018). As of yet, no randomized control study has
directly tested whether integrated, manualized CBT produces enhanced effectiveness
beyond ERP alone.
The aim of this study was to examine whether the addition of manualized cognitive
therapy assessment, conceptualization, and intervention strategies to manual-based ERP
results in additional treatment benefits beyond ERP alone to the extent that would justify
adding new elements to an already effective therapy. It was hypothesized that the
integrated ERP + CT condition would result in better treatment outcomes than ERP alone
in terms of symptom improvement and reduction in cognitive markers of OCD
vulnerability. A second purpose of the study was to test distinction versus uniformity in
treatment effects across different obsessive-compulsive presentations. Past research has
demonstrated (Abramowitz, Franklin, Schwartz, & Furr, 2003) differential treatment
effects in ERP based on O-C symptom presentation although no published research has
examined whether adding CT to ERP produces uniform or more symptom dimension-
specific treatment outcomes.
 Exposure Response Prevention and Cognitive Therapy 3

Methods
Participants and study design
Recruitment
Participants were recruited from a large community-based OCD and Anxiety Disorders
Clinic with appropriate institutional review board approval. Once informed consent was
obtained, participants were administered the Structured Clinical Interview for Axis 1
Disorders (SCID-1/P version 2.0) and a YBOCS clinician-rated interview. Participants were
considered for this study according to the following criteria. Inclusion criteria were as
follows: (1) meeting DSM-IV-TR (American Psychiatric Association, 2000) criteria for OCD
based on the SCID, (2) between the ages of 18 and 65, (3) experiencing clinically
significant obsessive-compulsive symptoms as defined by a Y-BOCS score >16, (4) able to
provide informed consent, and (5) on stable medications, as defined by no change in
medication type or dose during 8 weeks before treatment. Exclusion criteria were as
follows: (1) past or present treatment with behaviour therapy or cognitive behaviour
therapy, (2) concurrent diagnosis of a mood disorder, schizophrenia or other psychotic
disorders, (3) active substance abuse/dependence within 6-months of study entry, (4)
suspected organic pathology, (5) primary hoarding obsessions/compulsions. All inclusion
and exclusion criteria were outlined in the advertisements of the study.

Procedure
Ethical approval for this study was provided by a university-affiliated Research Ethics
Board (Ref: MOP-44081). Stratified random sampling was used to assign eligible
participants to one of two conditions: ERP or ERP + CT. Randomization was conducted
by computer-generated blocks of random numbers. An independent assessor not involved
in treatment or in post-treatment assessments performed the randomization. Assessment
teams at each phase of the study did not overlap and all attempts to ensure participant
condition masking was undertaken. Two hundred and twelve participants were assessed
for eligibility for this study. Of these, 35 were excluded at the phone screen phase, 30
were excluded after the SCID-interview phase, and 20 were excluded after the clinician-
administered Y-BOCS. The remaining 127 participants were accepted into the study and
were randomized into the two treatment conditions: ERP (n = 62) and ERP + CT
(n = 65). Of the participants who completed the assessment phase and were successfully
randomized the dropout rate was not significantly different by treatment condition,
(X2 = 3.17, p > .05) and the overall number of treatment sessions received in the ERP
(M = 14.6, SD = 2.2) and ERP + CT (M = 15.3, SD = 1.5) conditions were equivalent, F
(1, 94) = 2.87, p > .05. A detailed overview of recruitment and retention for the study is
illustrated in Figure 1. The study resulted in a final intent-to-treat (ITT) sample of 124 and a
final completer sample of 94 participants, 42 who were in the ERP group and 52 who were
in the ERP + CT group. Demographic features for the completers sample are presented in
Table 1. Additionally, 31.5% of the sample had a secondary SCID diagnosis at the time of
the study with the most commonly occurring comorbidities being: social anxiety disorder
(8.9%); specific phobia (5.6%); and generalized anxiety disorder (2.4%).

Medication use
Participants taking psychotropic medications were advised not to make any changes
during the active phase of their treatment although all changes to medications were
4 Neil A. Rector et al.

Assessed for eligibility (N = 212)

Excluded (n = 85)
Not meeting inclusion/exclusion criteria (n = 51)
Refused to participate (n = 8)
Requested group therapy (n = 5)
Contact/scheduling problems (n = 8)
Other reasons (n = 13)

Randomized (N = 127)

Allocated to intervention ERP (n = 62) Allocated to intervention ERP+CT (n = 65)

Received allocated intervention (n = 62) Received allocated intervention (n = 65)
Did not receive allocated intervention (give Did not receive allocated intervention (give reasons-
reasons-dropouts) (n = 0) dropouts) (n = 0)

Discontinued intervention (give reasons) (n = 16) Discontinued intervention (give reasons) (n = 9)

Too busy/lost interest (n = 11)
Not ready/suitable for treatment (n = 2)
Developed depression during treatment (n = 2)
Language difficulties (n = 1)
Did not complete post-treatment YBOCS (n = 3)
Too busy/lost interest (n = 7)
Not ready/suitable for treatment (n = 2)
Did not complete post-treatment YBOCS (n = 2)

Analyzed (ITT n = 61; completer n = 42) Analyzed (ITT n = 63; completer n = 52)
Excluded from analysis (give reasons) (n = 1) Excluded from analysis (give reasons) (n = 2)
Refused to complete any questionnaires (n = 1) Misdiagnosed enuresis (n = 1)
Misdiagnosed paraphilia (n = 1)

Figure 1. Flow diagram of participant progress through randomized trial.

monitored and recorded by team therapists and verified by research staff across the
study duration. A series of ANOVAs and Chi squares were computed to test for group
differences on the type of medication, current medication usage, dosage levels at the
time of randomization, changes in medication while in treatment, and number of past
SSRIs, other antidepressants and anxiolytics and antipsychotics. Twenty-seven
(64.29%) participants in the ERP condition and 36 (69.23%) in the ERP + CT
treatment condition were on medications. No differences were observed between the
two groups on any of the medication variables (p’s > .05). A summary of these
analyses is shown in Table 2.

Treatment conditions
Exposure and response prevention
Exposure and response prevention was conducted according to the manual developed,
widely tested and disseminated by Kozak and Foa (1997) and consistent with Foa,
Yadin, and Lichner (2012). The intervention consisted of 16 60-min weekly sessions in
individual outpatient format. ERP entailed prolonged exposure to obsessional cues that
induced discomfort and strict abstinence of ritualizing behaviour until the discomfort
abated (leading to habituation of fears). Exposure proceeded in a hierarchical fashion.
 Exposure Response Prevention and Cognitive Therapy 5

Table 1. Demographic characteristics of treatment completers

ERP (n = 42) ERP + CT (n = 52)

Variable M SD M SD F/Χ2 p

Age (years) 32.81 8.04 31.58 9.50 0.45 .51 ns

Onset age of OCD 17.00 9.65 18.25 7.95 0.47 .50 ns
Duration 15.02 9.39 13.27 10.07 0.74 .39 ns
Gender (females) 69.05 50.00 3.47 .06 ns
Education 3.58 .61 ns
Graduate school 19.05 11.54
College/university 71.43 73.08
High school 9.52 15.38
Ethnicity 1.70 .89 ns
Caucasian 78.57 76.92
Asian 9.52 11.54
Black 4.76 3.85
Hispanic 2.38 3.85
Arabic 2.38 3.85
Other 2.38 –
Occupational status 9.85 .20 ns
Employed 61.90 50.00
Student 16.67 25.00
Homemaker 4.76 3.85
Disabled 2.38 9.62
Unemployed 14.29 11.54
Relationship status 3.11 .38 ns
Single 47.62 57.69
Married/Cohabiting 47.62 38.46
Divorced/Separated 4.76 3.85

Table 2. Summary of medication use for treatment completers

ERP (n = 42) ERP + CT (n = 52)

Variables M SD M SD F/Χ2 p

Currently on medications (Yes) 64.29 69.23 0.26 .61 ns

Types of medication 0.46 .50 ns
SSRIs 45.24 50.00
Anxiolytic 2.38 –
Neuroleptic/Novel antipsychotic 2.38 –
Antidepressants and anxiolytics 9.52 17.31
Non-psychotropic medications 4.76 1.92
Prozac equivalency dosage level 51.36 31.63 47.27 28.78 0.34 .56 ns
Change in medication while in treatment (% Yes) – 3.85 0.85 .36 ns

Although informal challenging of cognitions can be a component of ERP, such as by

guiding participants towards noticing whether their hypotheses related to the
exposure were supported, formal and structured cognitive restructuring techniques
6 Neil A. Rector et al.

as described below were not a component of the ERP group. Participants were also
instructed to practice exposure exercises between sessions, upwards of 1–2 hrs per
day for a minimum of four times per week. The final phase of treatment focused on
relapse prevention.

Integrated treatment: exposure and response prevention plus cognitive therapy

The integrated ERP + CT was conducted according to the same treatment manual based
on Kozak and Foa (1997) described above. It was based on providing psychoeducation,
monitoring obsessional triggers, developing hierarchies, teaching anxiety (SUDS)
ratings, and monitoring SUDS ratings during exposure tasks in-session and as homework.
Exposure tasks included exposure to obsessional cues and ritual prevention. However,
the theoretical framework for all exposure tasks in-session and as part of homework was
not on habituation but rather using the exposure task to activate cognitive appraisals and
obsessional beliefs for further questioning, evidence-gathering and reframing including
appraisals pertaining to the ability to cope with anxious arousal when refraining from
rituals. The delivery of in-session and between-session exposures in the early phase of
treatment was then followed by an increasing integration of cognitive principles and
approaches from sessions 4 to 14 based on approaches outlined by Clark (2004),
Rachman (1998, 2003), Salkovskis (1999) and Steketee et al. (1998) targeting appraisals
and dysfunctional beliefs related to the six domains outlined by the OCCWG (1997,
2001). Specific cognitive strategies were included to normalize intrusions, target
obsessional appraisals of intrusions, identify and change cognitive distortions, and test
alternative appraisals/interpretations of intrusions during in-session and between-
session exposures. Further, obsessive beliefs pertaining to inflated responsibility,
overestimation of threat, perfectionism, need for certainty, and overimportance and
need to control thoughts were targeted as previously outlined (Clark, 2004; Rachman,
1998, 2003; Salkovskis, 1999; Steketee et al., 1998). The final stage of treatment
(Sessions 15 and 16) focused on relapse prevention including ERP and CT features. Like
the ERP condition, this intervention consisted of 16 60-min sessions in individual format
on a weekly basis.

Therapists
To control for allegiance and cross-contamination of protocols within the same site, two
senior M.D./Ph.D. therapists with allegiance to ERP directed the ERP condition and two
senior M.D./Ph.D. level therapists with expertise in CT and integrated ERP + CT for OCD
directed the integrated ERP + CT condition. Other therapists (n = 4) were matched for
their level of expertise and familiarity with ERP and ERP + CT in the two conditions.
Amongst the eight trained community psychiatrists and psychologists responsible for
delivering treatment, the mean age was 38 years (SD = 7.39 years) with 7.68 years
(SD = 4.69 years) of specialized ERP/ERP + CT experience.

Treatment adherence and integrity

To examine adherence to the treatment protocols, items for the adherence rating scale
were generated from the treatment manuals. Raters who were separate from the study
team and blind to participant condition, therapist protocol, and the study hypotheses
evaluated 200 randomly selected audiotapes of ERP (n = 100) and ERP + CT (n = 100)
 Exposure Response Prevention and Cognitive Therapy 7

treatment sessions stratified by phase of treatment (early, middle, late). A reliability check
on tape ratings was achieved by having a second rater trained in the ERP and ERP + CT
protocols rate 20 of the 200 tapes (10 per condition) and the inter-rater reliability was
determined to be acceptable (r = .97). Overall, adherence ratings across the 16-session
protocol was determined to be excellent and equivalent between the ERP (M = 87.89,
SE = 1.467) and ERP + CT conditions (M = 87.61, SE = 1.47), F(1, 198) = 0.02, p > .05.
A final evaluation of treatment fidelity was the determination of whether the ERP
condition included formal cognitive restructuring techniques at the appraisal or belief
level. Because the ERP manual of Kozak and Foa (1997) allows for cognitive challenging of
estimates of threat during in-session and between-session exposure tasks, independent
raters were requested to identify all sessions that included more formal cognitive therapy
with structured interventions for negative appraisals of intrusions and obsessive beliefs.
None of the ERP tapes (n = 100) were identified as having formal cognitive therapy
elements. Finally, there were no differences in outcomes between therapists in ERP or
ERP + CT (p > .05).

Measures
Structured Clinical Interview for DSM-IV Disorders (SCID-I/Pl; First, Spitzer, Gibbon, & Williams, 1996,
2002)
The SCID-I interviews were administered by research staff whom had received extensive
formal training in the administration and scoring of the interview protocol with OCD and
anxiety disordered patient populations, and who had completed a rigorous inter-rater
reliability training programme prior to administration. Additionally, the SCID-I assessors
attended weekly clinical case conference meetings with senior psychologists who
specialized in the assessment and treatment of anxiety disorders, to establish consensus
principal and secondary psychiatric diagnoses.

Yale-Brown Obsessive Compulsive Scale (Y-BOCS; Goodman, Price, Rasmussen, Mazure, Delgado,
et al., 1989; Goodman, Price, Rasmussen, Mazure, Fleischmann, et al., 1989)
The Y-BOCS is standardized rating scale designed to identify frequency, interfer-
ence, and distress associated with obsessions and compulsions. The Y-BOCS
consists of 10 items pertaining to obsessions and compulsions that are rated on a 5-
point Likert scale ranging from 0 (no symptoms) to 4 (severe symptoms). The Y-
BOCS symptom checklist was also used for subtyping analysis. The Y-BOCS has
been shown to possess high internal consistency and validity (Goodman, Price,
Rasmussen, Mazure, Delgado, et al., 1989; Goodman, Price, Rasmussen, Mazure,
Fleischmann, et al., 1989). Prior to the initiation of the study, an extensive training
programme was conducted in the administration and coding of audiotaped Y-BOCS
interviews. Inter-rater reliability based on tape ratings of the Y-BOCS total was
excellent (r = .98).
The pre-treatment Y-BOCS was administered prior to randomization and so blinding
was not necessary for this phase. However, assessors administering the post-treatment Y-
BOCS were blind to participant condition and were unaware of the study hypotheses. In
an effort to maintain the blind, the participants were requested not to reveal the name of
their therapist or the nature of their clinical care during the Y-BOCS interview. Blind
assessors were not members of the study team and had no contact with team therapists or
study participants.
8 Neil A. Rector et al.

The Obsessional Belief Questionnaire (OBQ-87; OCCWG, 1997, 2001, 2003)

The Obsessional Belief Questionnaire (OBQ) is an 87-item self-report questionnaire
designed to assess the extent to which respondents agree or disagree with various
attitudes and beliefs related to obsessional thoughts. The OBQ-87 comprises three factors:
Responsibility/Threat Estimation (e.g., ‘Harmful events will happen unless I’m careful’),
Perfectionism/Certainty (e.g., ‘Things are not right if they are not perfect’), and
Importance/Control of Thoughts (e.g., ‘A bad thought is morally no different than a bad
deed’). Respondents indicate their level of agreement with items on a 7-point rating scale.
Higher scores indicate a greater strength of beliefs. The OBQ is internally consistent and
evidences good test–retest reliability, convergent validity, and discriminant validity
(OCCWG, 2005). In this study, internal consistency (Cronbach a) for the total OBQ of .97
was deemed acceptable.

Beck Depression Inventory-II (Beck & Steer, 1987)

The Beck Depression Inventory-II (BDI-II) is a 21-item self-report instrument designed to
assess the severity of depressive symptoms over the preceding week. The BDI-II has been
shown to be a reliable and well-validated measure of depressive symptomatology (Beck,
Steer, & Carbin, 1988).

Beck Anxiety Inventory (Beck & Steer, 1990)

The Beck Anxiety Inventory (BAI) consists of 21 self-reported items employed to assess
the intensity of physical and cognitive anxiety symptoms during the past week. A variety
of studies have supported the reliability and validity of the BAI as a measure of anxiety in
both clinical and non-clinical samples (Beck, Epstein, Brown, & Steer, 1988; Borden,
Peterson, & Jackson, 1991).

Psychotherapy Expectancies Inventory – Revised (Rickers-Ovsiankina, Geller, Berzins, & Rogers, 1971)
The Psychotherapy Expectancies Inventory – Revised (PEI-R) is a 30-item self-report
questionnaire consisting of 24 keyed items and six filler items. It measures client
expectations about the likelihood that certain interactions will take place during therapy.

Statistical analyses
A power analysis revealed that in order to detect a potential medium-to-large effect size
difference between ERP and ERP + CT at post-treatment on the principal outcome
measure with power of .80 and Alpha set at .05, 32 participants per condition were
required. The final ITT sample (n = 127) and the final sample of treatment completers
(n = 94) provided excellent power to test differences in the sample as a whole, as well as
in distinct symptom groups (e.g., contamination/washing vs. doubting/checking etc.). In
the ITT analyses, participants who withdrew prior to completing the treatment were
retained by carrying forward the last obtained score. For the analyses on treatment
completers, no data were missing for the clinician-rated Y-BOCS, which was their
principal outcome measure. However, where data were missing on self-report measures,
the last observation was carried forward. To test for between-group treatment differences
at outcome, repeated-measures multivariate analyses of variance (MANOVAs)
were conducted with treatment modality as the between-group variable and symptom
 Exposure Response Prevention and Cognitive Therapy 9

(Y-BOCS) and cognitive (OBQ) change scores as the repeated within-group variables. Chi-
square tests were conducted to determine whether the number of participants being
deemed ‘treatment responders’ differed by condition. As a confirmation of blinding,
assessors did not guess participant treatment condition beyond chance and there were no
significant differences in correct guesses between treatment conditions (p’s > .05)

Results
Treatment credibility
Treatment credibility was measured using the PEI-R (Rickers-Ovsiankina et al., 1971).
There were no differences between the ERP (M = 104.2, SD = 23.7) and ERP + CT
(M = 105.4, SD = 20.5) conditions for the ITT sample, F(1, 124) = 0.10, p > .05, or
between ERP (M = 103.6, SD = 22.4) and ERP + CT (M = 104.5, SD = 21.2) conditions
for treatment completers, F(1, 94) = 0.04, p > .05 on pre-treatment ratings of expecta-
tions to benefit from treatment.

Treatment outcomes: Y-BOCS change and responder status

The study measures pre- and post-treatment means, standard deviations, and effect sizes
for study completers according to treatment condition are presented in Table 3. There
were no significant differences between treatment conditions on any of the study
measures (all p’s > .05) within the ITT or completers samples at baseline.

Intent-to-treat
A repeated-measures MANOVA for the ITT sample revealed a significant multivariate
effect for Y-BOCS change, Wilks’ k = .63, F(1, 122) = 133.89, p < .001 and a significant Y-
BOCS change by treatment modality interaction, Wilks’ k = .93, F(1, 122) = 9.41,
p < .005, with the ERP + CT group showing superior outcomes to the ERP group
(d = 0.61). Subsequent analyses on the obsession and compulsion subscales of the Y-
BOCS demonstrated significant differences in favour of ERP + CT compared to ERP on
obsession scores, Wilks’ k = .94, F(1, 122) = 8.12, p < .005, d = 0.53, and compulsion
scores, Wilks’ k = .94, F(1, 122) = 8.06, p < .005, d = 0.63.

Treatment completers
A repeated-measures MANOVA for completers similarly demonstrated a significant
multivariate effect for Y-BOCS change, Wilks’ k = .33, F(1, 92) = 187.34, p < .001 and a
significant Y-BOCS change by treatment modality interaction, Wilks’ k = .95, F(1,
92) = 5.37, p < .05, d = 0.67, with better outcomes in ERP + CT compared to ERP. As
before, effects were observed on obsession scores, Wilks’ k = .95, F(1, 92) = 4.92,
p < .05, d = 0.64, and compulsion scores, Wilks’ k = .95, F(1, 92) = 5.37, p < .05,
d = 0.64.

Outcomes by OCD symptom subtype group

Scores on the Y-BOCS symptom checklist were quantified in a similar manner to Baer
(1994) and Calamari, Wiegartz, and Janeck (1999) into 15 derived categories. Responses
 10
Neil A. Rector et al.
Table 3. Pre- and post-treatment means and standard deviations for symptom and cognition measures for treatment completers

ERP (n = 42) ERP + CT (n = 52)

Pre-treatment Post-treatment Pre-treatment Post-treatment

Scale M SD M SD d M SD M SD d

Symptom measures
Y-BOCS-clinician total 24.43 4.33 17.31 6.61 1.27 23.15 4.27 12.96 6.34 1.89
Obsession subscale 12.07 2.32 8.81 3.72 1.05 11.48 2.79 6.61 3.14 1.64
Compulsion subscale 12.29 2.61 8.50 3.35 1.26 11.65 2.57 6.35 3.54 1.71
BDI-II 17.43 12.33 9.39 8.92 0.75 16.04 9.67 10.71 8.87 0.57
BAI 13.74 9.55 8.38 10.97 0.52 13.08 8.57 8.35 8.02 0.57
Cognition measures
OBQ total 336.36 99.59 272.03 92.31 0.67 326.12 105.14 237.92 99.04 0.86
Responsibility/Threat estimation subscale 62.88 22.83 52.64 23.10 0.45 62.23 23.85 45.76 21.09 0.73
Perfectionism-intolerance for uncertainty subscale 67.14 21.67 55.22 19.93 0.57 63.87 23.73 51.12 24.19 0.53
Control/importance of thoughts subscale 38.95 16.81 30.79 14.01 0.53 38.87 18.11 25.29 13.80 0.84
 Exposure Response Prevention and Cognitive Therapy 11

on the 15 symptom categories of the Y-BOCS were scored ‘0’ (not endorsed), ‘1’ (the
patient endorsed at least one symptom, but the category was not considered primary) and
‘2’ (symptoms were endorsed and the category was identified as the principal problem in
treatment). A symptom category was considered principal if was rated as ‘2’ and was the
principal focus in treatment. This procedure resulted in the following subtype
classification: contamination/washing (n = 40), doubting-harming/checking (n = 45),
symmetry/ordering (n = 20), pure obsessions/’bad thoughts’ (n = 13), and miscella-
neous (n = 6). No patients had primary hoarding symptoms. Analyses were conducted on
the four principal symptom dimensions excluding the miscellaneous group.
MANOVAs were repeated with Y-BOCS scores at pre-treatment and post-treatment as
the within-subject repeated measures and treatment condition and symptom subtype
group as the between-group variables. In the ITT, Wilks’ k = .93, F(4, 114) = 8.43,
p < .005 and treatment completer samples, Wilks’ k = .93, F(4, 84) = 9.43, p < .005 the
original Y-BOCS change by treatment condition multivariate interactions were retained,
although there were no Y-BOCS change by symptom subtype multivariate effects
(p’s > .05) or Y-BOCS by symptom subtype by treatment condition multivariate effects
(p’s > .05).

Treatment responder status by condition

Treatment responder status was determined using per cent reductions as a measure of
outcome. Per cent reductions used were based on a signal detection study using ‘gold
standard’ criteria from the Clinical Global Impressions scale, which found that 30%
reduction is optimal for determining clinical improvement and 50% reduction provides
indication of ‘mild illness’ at post-treatment (Tolin, Abramowitz, & Diefenbach, 2005).

Percentage reductions
For treatment completers, the rates of significant response were 65.4% for the ERP + CT
group and 45.5% for the ERP group, and based on logistic regression this rate was not
significant, OR = 2.27, p = 0.052. In terms of those reaching mild illness status, more
occurred in the ERP + CT group (44.2%) compared to ERP (22.7%), OR = 2.70, p = .03.

Effects on obsessive beliefs

Intent-to-treat
A repeated-measures MANOVA on the OBQ scores for the ITT sample indicated that both
treatments produced large and significant effects on OBQ scores, Wilks’ k = .69, F(1,
121) = 53.98, p < .001 although the ERP + CT produced greater change on obsessive
beliefs compared to ERP as demonstrated in a OBQ change by treatment condition
multivariate interaction, Wilks’ k = .95, F(1, 117) = 6.28, p < .01.

Treatment completers
The same MANOVA with treatment completers similarly revealed a main effect for OBQ
change, Wilks’ k = .62, F(1, 92) = 57.01, p < .001 and a two-way OBQ change by
treatment condition interaction, Wilks’ k = .95, F(1, 92) = 4.86, p < .05 with the
ERP + CT group producing greater change on obsessive beliefs than the ERP group.
12 Neil A. Rector et al.

Effects on anxiety and depression

Intent-to-treat
A repeated-measures MANOVA demonstrated that both treatment conditions produced
significant change on anxiety, Wilks’ k = .84, F(1, 122) = 23.04, p < .001 although
anxiety outcomes did not differ by treatment condition, Wilks’ k = 1.00, F(1, 122) = 0.67,
p > .05. Similarly, depression improved across treatment, Wilks’ k = .76, F(1,
122) = 39.67, p < .001 although again, there was no effect for treatment condition,
Wilks’ k = 1.00, F(1, 122) = 0.25, p > .05.

Treatment completers
A repeated-measures MANOVA demonstrated a significant multivariate effect on anxiety,
Wilks’ k = .80, F(1, 92) = 22.38, p < .001 and depression Wilks’ k = .68, F(1,
92) = 44.32, p < .001 although there were differences in anxiety or depression reduction
by treatment condition (p’s > .05).

Treatment follow-up effects

Amongst treatment completers, 75% of participants (ERP + CT: N = 39; ERP; N = 31)
were retained for their 6-month follow-up assessment. A MANOVA with treatment
modality as the between-group independent variable and Y-BOCS at baseline and 6-month
follow-up as the within-group repeated measures demonstrated a main effect for Y-BOCS
change, Wilks’ k = .34, F(1, 68) = 134.77, p < .001 and a two-way Y-BOCS change by
treatment condition interaction, Wilks’ k = .94, F(1, 68) = 4.47, p < .05 with the
ERP + CT group producing greater reduction in Y-BOCS scores from baseline to 6-month
follow-up than the ERP group.1

Discussion
Main findings
While large gains have been made in recent decades towards improving treatment for
OCD, a substantial proportion of clients fail to show clinically significant change (Fisher &
Wells, 2005), suggesting that further improvements can yet be made to existing
treatments. Consistent with previous research, both ERP and ERP + CT were found to be
highly efficacious treatments with significant and statistically large reductions in O-C
symptoms at post-treatment. There were also significant differences between treatment
conditions: Integrated ERP + CT led to comparatively greater reductions in O-C
symptoms at post-treatment compared with ERP alone. This additive benefit was
equivalent to a medium to large treatment effect. This pattern of treatment effects was
observed regardless of whether the treatment groups were compared in the ITT or the
treatment completer samples. Employing a rigorous 50% symptom reduction cut-off to

1
As an alternate method to examine treatment effects while accounting for missing data, multilevel longitudinal modeling was
also used to examine the effect of treatment group on Y-BOCS symptom severity over the course of treatment and 6-month follow-
up. Restricted maximum likelihood estimation was used for all analyses. Analyses were conducted using R (R Core Team, 2017)
and the package nlme (Pinheiro, Bates, DebRoy, Sarkar, & R Core Team, 2017). A model including random intercept and slope
failed to converge. Using a random intercept model, there was a significant interaction between treatment group and time,
B = 1.73, p = 0.02, 95% CI (0.26, 3.21), such that ERP + CT resulted in significantly greater rate of change in Y-BOCS scores
over time compared to ERP.
 Exposure Response Prevention and Cognitive Therapy 13

determine treatment response status and achieve a ‘mild illness’ designation (Tolin et al.,
2005), a significant difference was observed with more participants deemed recovered in
ERP + CT than ERP. The additive benefits that were found were shown to be efficacious
across the main OCD dimensions addressed in the study: contamination/washing,
doubting-harming/checking, order-symmetry/repeating and pure obsession (harming,
sexual, somatic, and religious). Overall, these results suggest that the integration of
cognitive therapy into the well-validated ERP approach to OCD may lead to clinically
significant additive benefits beyond ERP alone.
In the ITT and treatment completer analyses, both ERP and ERP + CT produced
medium-to-large and equivalent effects on general symptoms of anxiety and depression.
However, ERP + CT was found to produce greater change on the underlying obsessive
beliefs that are seen to constitute vulnerability to the development and persistence of the
disorder within current cognitive frameworks (Clark, 2004; OCCWG, 1997, 2001, 2003;
Rachman, 1997, 1998; Salkovskis, 1985, 1999). These results suggest that the additive
benefits of ERP + CT compared to ERP appear to be specific to the symptomatic and
cognitive aspects of OCD and cannot be explained in terms of non-specific therapy
factors. In further support of this interpretation, the differential treatment effects in favour
of ERP + CT compared with ERP were not due to other therapeutic factors such as
treatment expectations or differences in adherence to protocols. Moreover, differences in
outcome were not due to medication use which was rigorously monitored and controlled.
The significantly larger effect of ERP + CT on obsessive beliefs is somewhat of a
departure from previous research findings comparing CT to ERP, in which the treatments
were found to have equivalent effects on obsessive beliefs at post-treatment (Whittal
et al., 2005). As this study differed from previous research in that CT and ERP were
integrated and then compared to ERP, it is possible that the addition of the two
interventions creates a stronger synergistic treatment for obsessive beliefs than either
intervention alone. On the one hand, ERP may be enhanced through the direct treatment
targeting of obsessive appraisals and beliefs during in- and between-session exposure
tasks and as part of modular targeting of obsessive beliefs. On the other hand, CT may be
enhanced through the greater experiential learning that follows from frequent anxiety-
inducing exposure tasks compared to the previous reliance of CT on behavioural
experiments that focused on data collection pertaining to testing threat appraisals.

Strengths and limitations

The current study has several limitations. First, the study did not employ a waitlist
condition to examine the differential effects of ERP or ERP + CT versus another form of
psychotherapy. However, past research has demonstrated only modest treatment gains
in non-CBT psychotherapy interventions for OCD (e.g., d < 0.20; Eddy et al., 2004) and
this study did permit the direct examination of two active psychological treatments
while controlling for the influence of medication use. Second, because the examination
of OCD subtype outcomes was exploratory, the subtype analysis was based on
previously established categories using the Y-BOCS rather than on a validated
dimensional measure. These results will require replication with more reliable and
valid dimensional O-C measures and with more larger sample sizes for each symptom
dimension (Abramowitz et al., 2010). However, these preliminary subtyping results do
suggest promise for the beneficial effects of integrated ERP + CT for the range of O-C
symptom dimensions opposed to a single symptom dimension (i.e., ‘bad thoughts’).
Third, participants were permitted to have dimensionally elevated depressive
14 Neil A. Rector et al.

symptoms, but the presence of a diagnosable mood disorder was the basis for exclusion
which may limit the external validity of the study given the high rates of depression
comorbidity in community and clinical samples (see Rector, Wilde, & Richter, 2017 for
review of epidemiological, clinical, and treatment aspects of depression/OCD comor-
bidity). Past research with ERP (Abramowitz & Foa, 2000; Abramowitz, Franklin, Street,
Kozak, & Foa, 2000) has demonstrated that depressive symptoms in the mild-to-
moderate range do not adversely impact on the efficacy of ERP whereas the presence of
severe depressive symptoms or diagnosable MDD does predict worse treatment
outcomes at post-treatment and follow-up. Less is known about the impact of comorbid
depression on response to CT although a study by McLean et al. (2001) found that
‘treatment responders’ to both ERP and CT for OCD were found to have lower
depression scores than non-responders. The current study was unable to test the
moderating influence of MDD on integrated ERP + CT treatment outcomes, and this will
require future examination. Fourth, the pre–post treatment effect size for the ERP group
(d = 1.27) was found to be somewhat smaller than observed in some other ERP
treatment studies, though within the range of ERP treatment effects reported in meta-
analytic reviews (e.g., d = 1.53; d = 1.13; Eddy et al., 2004; Rosa-Alc
azar et al., 2008).
One possible explanation for the lower ERP effect size compared to previous findings is
that the ERP treatment in this study consisted of once-weekly 60 min sessions, rather
than more intensive ERP of two or more weekly sessions that have been frequently
described in the literature (Foa et al., 2005). More intensive ERP generally results in
higher effect sizes at post-treatment compared to less intensive (J

onsson, Kristensen, &
Arendt, 2015). Importantly, the once-weekly 60-min session in the current study may
represent the more typical outpatient delivery of ERP in the community. Moreover, the
pre–post treatment effect (d = 1.89) for the combined treatment in the current study
exceeds the typically reported pre–post effects for ERP alone. Despite these limitations,
the findings of the current study provide evidence supporting NICE (2005) treatment
guidelines that recommend the integration of ERP and CT for optimal treatment
outcome in OCD.

Acknowledgements
We would like to thank all patients who participated in this research. We are also grateful to all
of the therapists in the study.

Funding
This work was funded by the Canadian Institutes of Health Research.

References
Abramowitz, J. S., Deacon, B. J., Olatunji, B. O., Wheaton, M. G., Berman, N. C., Losardo, D., . . . Hale,
L. R. (2010). Assessment of obsessive-compulsive symptom dimensions: Development and
evaluation of the Dimensional Obsessive-Compulsive Scale. Psychological Assessment, 22, 180–
198. https://doi.org/10.1037/a0018260
Abramowitz, J. S., & Foa, E. B. (2000). Does comorbid major depressive disorder influence outcome
of exposure and response prevention for OCD? Behavior Therapy, 31, 795–800. https://doi.
org/10.1016/S0005-7894(00)80045-3
 Exposure Response Prevention and Cognitive Therapy 15

Abramowitz, J. S., Franklin, M. E., Schwartz, S. A., & Furr, J. M. (2003). Symptom presentation and
outcome of cognitive-behavioral therapy for obsessive-compulsive disorder. Journal of
Consulting and Clinical Psychology, 71, 1049–1057. https://doi.org/10.1037/0022-006X.71.
6.1049
Abramowitz, J. S., Franklin, M. E., Street, G. P., Kozak, M. J., & Foa, E. B. (2000). Effects of comorbid
depression on response to treatment for obsessive-compulsive disorder. Behavior Therapy, 31,
517–528. https://doi.org/10.1016/S0005-7894(00)80028-3
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders
(4th ed., text rev.). Washington, DC: Author.
American Psychiatric Association. (2007). Practice guideline for the treatment of patients with
obsessive-compulsive disorder. The American Journal of Psychiatry, 164(7 Suppl), 5–53.
Anderson, R. A., & Rees, C. S. (2007). Group versus individual cognitive-behavioural treatment for
obsessive-compulsive disorder: A controlled trial. Behaviour Research and Therapy, 45, 123–
137. https://doi.org/10.1016/j.brat.2006.01.016
Anholt, G. E., Kempe, P., de Haan, E., van Oppen, P., Cath, D. C., Smit, J. H., & van Balkom, A. J. L. M.
(2007). Cognitive versus behavior therapy: Processes of change in the treatment of obsessive-
compulsive disorder. Psychotherapy and Psychosomatics, 77, 38–42. https://doi.org/10.1159/
000110058
Baer, L. (1994). Factor analysis of symptom subtypes of obsessive compulsive disorder and
their relation to personality and tic disorders. The Journal of Clinical Psychiatry, 55(3,
Suppl), 18–23.
Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety:
Psychometric properties. Journal of Consulting and Clinical Psychology, 56, 893–897.
https://doi.org/10.1037/0022-006X.56.6.893
Beck, A. T., & Steer, R. A. (1987). Manual for the Revised Beck Depression Inventory. San Antonio,
TX: Psychological Corporation.
Beck, A. T., & Steer, R. A. (1990). Manual for the Beck Anxiety Inventory. San Antonio, TX:
Psychological Corporation.
Beck, A. T., Steer, R. A., & Carbin, M. G. (1988). Psychometric properties of the Beck Depression
Inventory: Twenty-five years of evaluation. Clinical Psychology Review, 8, 77–100. https://
doi.org/10.1016/0272-7358(88)90050-5
Borden, J. W., Peterson, D. R., & Jackson, E. A. (1991). The Beck Anxiety Inventory in nonclinical
samples: Initial psychometric properties. Journal of Psychopathology and Behavioral
Assessment, 13, 345–356. https://doi.org/10.1007/BF00960446
Braga, D. T., Abramovitch, A., Fontenelle, L. F., Ferr~ao, Y. A., Gomes, J. B., Vivan, A. S., . . . Cordioli, A.
V. (2016). Neuropsychological predictors of treatment response to cognitive behavioral group
therapy in obsessive-compulsive disorder. Depression and Anxiety, 33, 848–861. https://doi.
org/10.1002/da.22509
Calamari, J. E., Wiegartz, P. S., & Janeck, A. S. (1999). Obsessive–compulsive disorder subgroups: A
symptom-based clustering approach. Behaviour Research and Therapy, 37, 113–125. https://
doi.org/10.1016/S0005-7967(98)00135-1
Clark, D. A. (2004). Cognitive-behavioral therapy for OCD. New York, NY: Guilford Press.
Cordioli, A. V., Heldt, E., Bochi, D. B., Margis, R., Basso de Sousa, M., Fonseca Tonello, J., . . .
Kapczinski, F. (2003). Cognitive-behavioral group therapy in obsessive-compulsive disorder: A
randomized clinical trial. Psychotherapy and Psychosomatics; Basel, 72, 211–216. https://doi.
org/10.1159/000070785
Cottraux, J., Note, I., Yao, S. N., Lafont, S., Note, B., Mollard, E., . . . Dartigues, J.-F. (2001). A
randomized controlled trial of cognitive therapy versus intensive behavior therapy in obsessive
compulsive disorder. Psychotherapy and Psychosomatics, 70, 288–297. https://doi.org/10.
1159/000056269
Eddy, K. T., Dutra, L., Bradley, R., & Westen, D. (2004). A multidimensional meta-analysis of
psychotherapy and pharmacotherapy for obsessive-compulsive disorder. Clinical Psychology
Review, 24, 1011–1030. https://doi.org/10.1016/j.cpr.2004.08.004
16 Neil A. Rector et al.

First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1996). Structured clinical interview for
DSM-IV axis I disorders – Patient edition (SCID-1/P, Version 2.0). New York, NY: Biometrics
Research Department, New York State Psychiatric Institute.
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (2002). Structured Clinical Interview for
DSM-IV-TR Axis I Disorders, research version, patient edition (SCID-I/P). New York, NY:
Biometrics Research Department, New York State Psychiatric Institute.
Fisher, P. L., & Wells, A. (2005). How effective are cognitive and behavioral treatments for
obsessive–compulsive disorder? A clinical significance analysis. Behaviour Research and
Therapy, 43, 1543–1558. https://doi.org/10.1016/j.brat.2004.11.007
Foa, E. B., Liebowitz, M. R., Kozak, M. J., Davies, S., Campeas, R., Franklin, M. E., . . . Tu, X. (2005).
Randomized, placebo-controlled trial of exposure and ritual prevention, clomipramine, and
their combination in the treatment of obsessive-compulsive disorder. The American Journal of
Psychiatry; Washington, 162, 151–161. https://doi.org/10.1176/appi.ajp.162.1.151
Foa, E. B., Yadin, E., & Lichner, T. K. (2012). Exposure and response (ritual) prevention for
obsessive-compulsive disorder: Therapist guide. Oxford, UK; New York, NY: Oxford University
Press. https://doi.org/10.1093/med:psych/9780195335286.001.0001
Freeston, M. H., Ladouceur, R., Gagnon, F., Thibodeau, N., Rh
eaume, J., Letarte, H., & Bujold, A.
(1997). Cognitive—behavioral treatment of obsessive thoughts: A controlled study. Journal of
Consulting and Clinical Psychology, 65, 405–413. https://doi.org/10.1037/0022-006x.65.3.
405
Gomes, J. B., Cordioli, A. V., Bortoncello, C. F., Braga, D. T., Goncßalves, F., & Heldt, E. (2016). Impact
of cognitive-behavioral group therapy for obsessive-compulsive disorder on family
accommodation: A randomized clinical trial. Psychiatry Research, 246, 70–76. https://doi.
org/10.1016/j.psychres.2016.09.019
Goodman, W. K., Price, L. H., Rasmussen, S. A., Mazure, C., Delgado, P., Heninger, G. R., & Charney,
D. S. (1989). The Yale-Brown Obsessive Compulsive Scale: II. Validity. Archives of General
Psychiatry, 46, 1012–1016. https://doi.org/10.1001/archpsyc.1989.01810110054008
Goodman, W. K., Price, L. H., Rasmussen, S. A., Mazure, C., Fleischmann, R. L., Hill, C. L., . . .
Charney, D. S. (1989). The Yale-Brown Obsessive Compulsive Scale: I. Development, use, and
reliability. Archives of General Psychiatry, 46, 1006–1011. https://doi.org/10.1001/archpsyc.
1989.01810110048007
Jaurrieta, N., Jim
enez-Murcia, S., Alonso, P., Granero, R., Segalas, C., Labad, J., & Mench
on, J. M.
(2008). Individual versus group cognitive behavioral treatment for obsessive-compulsive
disorder: Follow up. Psychiatry and Clinical Neurosciences, 62, 697–704. https://doi.org/10.
1111/j.1440-1819.2008.01873.x
J

onsson, H., Kristensen, M., & Arendt, M. (2015). Intensive cognitive behavioural therapy for
obsessive-compulsive disorder: A systematic review and meta-analysis. Journal of Obsessive-
Compulsive and Related Disorders, 6, 83–96. https://doi.org/10.1016/j.jocrd.2015.04.004
Katzman, M. A., Bleau, P., Blier, P., Chokka, P., Kjernisted, K., & Van Ameringen, M. (2014). Canadian
clinical practice guidelines for the management of anxiety, posttraumatic stress and obsessive-
compulsive disorders. BMC Psychiatry, 14(1), S1. https://doi.org/10.1186/1471-244X-14-S1-S1
Kozak, M. J., & Foa, E. B. (1997). Mastery of obsessive-compulsive disorder. A cognitive
behavioural approach. Therapist guide. New York, NY: Graywind Publications.
McLean, P. D., Whittal, M. L., Thordarson, D. S., Taylor, S., S€ ochting, I., Koch, W. J., . . . Anderson,
K. W. (2001). Cognitive versus behavior therapy in the group treatment of obsessive-compulsive
disorder. Journal of Consulting and Clinical Psychology, 69(2), 205–214. https://doi.org/10.
1037//0022-006x.69.2.205
NICE. (2005). Obsessive-compulsive disorder and body dysmorphic disorder: Treatment. Nice
guideline (CG31). Retrieved from https://www.nice.org.uk/guidance/cg31
Obsessive Compulsive Cognitions Working Group. (1997). Cognitive assessment of obsessive-
compulsive disorder. Behaviour Research and Therapy, 35, 667–681. https://doi.org/10.1016/
S0005-7967(97)00017-X
 Exposure Response Prevention and Cognitive Therapy 17

Obsessive Compulsive Cognitions Working Group. (2001). Development and initial validation of the
obsessive beliefs questionnaire and the interpretation of intrusions inventory. Behaviour
Research and Therapy, 39, 987–1006. https://doi.org/10.1016/S0005-7967(00)00085-1
Obsessive Compulsive Cognitions Working Group. (2003). Psychometric validation of the Obsessive
Beliefs Questionnaire and the Interpretation of Intrusions Inventory: Part I. Behaviour Research
and Therapy, 41, 863–878. https://doi.org/10.1016/S0005-7967(02)00099-2
Obsessive Compulsive Cognitions Working Group. (2005). Psychometric validation of the obsessive
belief questionnaire and interpretation of intrusions inventory–Part 2: Factor analyses and
testing of a brief version. Behaviour Research and Therapy, 43, 1527–1542. https://doi.org/10.
1016/j.brat.2004.07.010
O’Connor, K. P., Aardema, F., Robillard, S., Guay, S., Pelissier, M.-C., Todorov, C., . . . Doucet, P.
(2006). Cognitive behaviour therapy and medication in the treatment of obsessive–compulsive
disorder. Acta Psychiatrica Scandinavica, 113, 408–419. https://doi.org/10.1111/j.1600-0447.
2006.00767.x
O’Connor, K., Todorov, C., Robillard, S., Borgeat, F., & Brault, M. (1999). Cognitive-behaviour
therapy and medication in the treatment of obsessive-compulsive disorder: A controlled
study. The Canadian Journal of Psychiatry, 44, 64–71. https://doi.org/10.1177/0706743
79904400108
Olatunji, B. O., Davis, M. L., Powers, M. B., & Smits, J. A. J. (2013). Cognitive-behavioral therapy for
obsessive-compulsive disorder: A meta-analysis of treatment outcome and moderators. Journal
of Psychiatric Research, 47, 33–41. https://doi.org/10.1016/j.jpsychires.2012.08.020
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., & R Core Team. (2017). nlme: Linear and Nonlinear
Mixed Effects Models (Version R package version 3.1-131.1). Retrieved from https://CRAN.
R-project.org/package=nlme
R Core Team. (2017). R: A language and environment for statistical computing. Vienna, Austria: R
Foundation for Statistical Computing. Retrieved from https://www.R-project.org/
Rachman, S. (1997). A cognitive theory of obsessions. Behaviour Research and Therapy, 35, 793–
802. https://doi.org/10.1016/S0005-7967(97)00040-5
Rachman, S. (1998). A cognitive theory of obsessions: Elaborations. Behaviour Research and
Therapy, 36, 385–401. https://doi.org/10.1016/S0005-7967(97)10041-9
Rachman, S. (2003). The treatment of obsessions. Oxford, UK: Oxford University Press. https://doi.
org/10.1093/med:psych/9780198515371.001.0001
Rector, N. A., Wilde, J. L., & Richter, M. A. (2017). Obsessive compulsive disorder and comorbidity. :
Rates, models and treatment approaches. In J. S. Abramowitz, D. McKay & E. A. Storch (Eds.), The
Wiley handbook of obsessive compulsive disorders (pp. 697–725). Chichester, UK: John Wiley
& Sons, Ltd.
Rickers-Ovsiankina, M. A., Geller, J. D., Berzins, J. I., & Rogers, G. W. (1971). Patient’s role-
expectancies in psychotherapy: A theoretical and measurement approach. Psychotherapy:
Theory, Research & Practice, 8, 124–126. https://doi.org/10.1037/h0086637
Rosa-Alc
azar, A. I., S
anchez-Meca, J., G

omez-Conesa, A., & Mar
ın-Mart
ınez, F. (2008). Psychological
treatment of obsessive–compulsive disorder: A meta-analysis. Clinical Psychology Review, 28,
1310–1325. https://doi.org/10.1016/j.cpr.2008.07.001
Salkovskis, P. M. (1985). Obsessional-compulsive problems: A cognitive-behavioural analysis.
Behaviour Research and Therapy, 23, 571–583. https://doi.org/10.1016/0005-7967(85)
90105-6
Salkovskis, P. M. (1999). Understanding and treating obsessive—compulsive disorder. Behaviour
Research and Therapy, 37. Behaviour Research and Therapy, 1 (Suppl. 1), S29–S52. https://
doi.org/10.1016/S0005-7967(99)00049-2
Steketee, G. S., Frost, R. A., Rh
eaume, J., & Wilhelm, S. (1998). Cognitive theory and treatment of
obsessive compulsive disorder. In M. A. Jenike, L. Baer & W. E. Minichiello (Eds.), Obsessive
compulsive disorders: Practical management (3rd ed., pp. 368–399). St. Louis, MO: Mosby.
18 Neil A. Rector et al.

Steketee, G., Siev, J., Yovel, I., Lit, K., & Wilhelm, S. (2018). Predictors and moderators of cognitive
and behavioral therapy outcomes for OCD: A patient-level mega-analysis of eight sites. Behavior
Therapy. https://doi.org/10.1016/j.beth.2018.04.004
Tolin, D. F., Abramowitz, J. S., & Diefenbach, G. J. (2005). Defining response in clinical trials for
obsessive-compulsive disorder: A signal detection analysis of the Yale-Brown Obsessive
Compulsive Scale. The Journal of Clinical Psychiatry, 66, 1549–1557. https://doi.org/10.
4088/jcp.v66n1209
Whittal, M. L., Thordarson, D. S., & McLean, P. D. (2005). Treatment of obsessive–compulsive
disorder: Cognitive behavior therapy vs. exposure and response prevention. Behaviour
Research and Therapy, 43, 1559–1576. https://doi.org/10.1016/j.brat.2004.11.012
Whittal, M. L., Woody, S. R., McLean, P. D., Rachman, S. J., & Robichaud, M. (2010). Treatment of
obsessions: A randomized controlled trial. Behaviour Research and Therapy, 48, 295–303.
https://doi.org/10.1016/j.brat.2009.11.010

Received 20 April 2018