Neuropsychological predictors of response to randomized treatment in
obsessive
–
compulsive disorder
Carina C. D'Alcante
a,⁎, Juliana B. Diniz
a, Victor Fossaluza
a,b, Marcelo C. Batistuzzo
a, Antonio C. Lopes
a,
Roseli G. Shavitt
a, Thilo Deckersbach
c, Leandro Malloy-Diniz
d, Euripedes C. Miguel
a, Marcelo Q. Hoexter
a,eaObsessive
–Compulsive Disorder Spectrum Project Clinic, Department of Psychiatry, University of São Paulo Medical School, São Paulo, Brazil
bInstitute of Mathematics and Statistics, University of São Paulo, São Paulo, Brazil
cDepartment of Psychiatry, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA dGraduate Program in Neuroscience, Institute of Biological Sciences, Federal University of Minas Gerais, Brazil
eInterdisciplinary Laboratory for Clinical Neuroscience (LiNC), Department of Psychiatry, Federal University of São Paulo (UNIFESP), São Paulo, Brazil
a b s t r a c t
a r t i c l e
i n f o
Article history: Received 1 June 2012
Received in revised form 29 June 2012 Accepted 2 July 2012
Available online 10 July 2012
Keywords:
Cognitive–behavioral therapy Neuropsychology
Obsessive–compulsive disorder Serotonin reuptake inhibitors Treatment-naïve
Objective:To identify neuropsychological predictors of treatment response to cognitive–behavioral therapy (CBT) andfluoxetine in treatment-naïve adults with obsessive–compulsive disorder (OCD).
Method:Thirty-eight adult outpatients with OCD underwent neuropsychological assessment, including tasks of intellectual function, executive functioning and visual and verbal memory, before randomization to a 12-week clinical trial of either CBT orfluoxetine. Neuropsychological measures were used to identify predic-tors of treatment response in OCD.
Results:Neuropsychological measures that predicted a better treatment response to either CBT orfluoxetine were higher verbal IQ (Wechsler Abbreviated Scale of Intelligence) (p = 0.008); higher verbal memory on the California Verbal Learning Test (p = 0.710); shorter time to complete part D (Dots) (pb0.001), longer time to complete part W (Words) (p = 0.025) and less errors on part C (Colors) (pb0.001) in the Victoria Stroop Test (VST). Fewer perseverations on the California Verbal Learning Test, a measure of mentalflexibility, predicted better response to CBT, but worse response tofluoxetine (p = 0.002).
Conclusion:In general, OCD patients with better cognitive and executive abilities at baseline were more prone to respond to either CBT orfluoxetine. Ourfinding that neuropsychological measures of mentalflexibility predicted response to treatment in opposite directions for CBT andfluoxetine suggests that OCD patients with different neuropsychological profiles may respond preferentially to one type of treatment versus the other. Further studies with larger samples of OCD patients are necessary to investigate the heuristic value of suchfindings in a clinical context.
© 2012 Elsevier Inc.
1. Introduction
Obsessive–compulsive disorder (OCD) is a chronic mental illness as-sociated with significant distress and impaired functioning (Koran,
2000; Leon et al., 1995). Its lifetime prevalence is estimated to be 2–
3% (Ruscio et al., 2010a, 2010b; Weissman et al., 1994) and
accumulat-ing evidence supports the efficacy of two treatment interventions: cognitive–behavioral therapy (CBT), which includes exposure and re-sponse prevention (Foa et al., 2005) and pharmacotherapy, primarily serotonin reuptake inhibitors (Fineberg and Gale, 2005). Overall, up to 60% of patients receiving one of these treatments or a combination of the two will exhibit a clinically significant reduction in obsessive–
compulsive symptoms (OCS), although residual symptoms are often the rule rather than the exception (Pallanti et al., 2002).
Neuropsychological approaches have proven to be important tools in the investigation of cognitive functioning in OCD (Kuelz et al.,
2004; Olley et al., 2007). In previous studies, certain
neuropsycholog-ical domains, specially related to executive functioning, such as plan-ning, cognitiveflexibility, response inhibition, decision making and attentional bias/vigilance, were found to be impaired in OCD patients
(Bérdard et al., 2009; Bespalov et al., 2010; Bolton et al., 2000;
Cavedini et al., 2002; Chamberlain et al., 2005; Menzies et al., 2008;
Olley et al., 2007; Shin et al., 2010) although inconsistent results
have been reported (Kuelz et al., 2004). Possible reasons for dispar-ities in some neuropsychologicalfindings in OCD might be explained by methodological issues such as small and heterogeneous samples of OCD participants, matching criteria and comorbidities, and particular-ly previous exposure to medications (Kuelz et al., 2004; Simpson et al., 2006).
–
⁎ Corresponding author at: University of São Paulo Medical School at São Paulo, Rua Dr. Ovidio Pires de Campos, 785 054 53-010 São Paulo, SP, Brazil.
E-mail address:chaubetca@hcnet.usp.br(C.C. D'Alcante).
0278-5846 © 2012 Elsevier Inc. doi:10.1016/j.pnpbp.2012.07.002
Contents lists available atSciVerse ScienceDirect
Progress in Neuro-Psychopharmacology & Biological
Psychiatry
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / p n p
Open access under the Elsevier OA license.
There is also evidence that some neuropsychological deficits in OCD, such as those in nonverbal memory, set shifting, planning, organizational skills and problem-solving may improve with CBT and pharmacological treatments (Kuelz et al., 2006; Park et al., 2002). There are also otherfi nd-ings showing that some neuropsychological impairments, such as inhibi-tory control and verbal fluency, remain present even after successful treatment (Kim et al., 2002; Thienemann and Koran, 1995). Furthermore, impairments in executive function domains have been shown to be asso-ciated with treatment response to specific treatment modalities (Cavedini
et al., 2002; Flessner et al., 2010; Fontenelle et al., 2001). For example,
poorer response to CBT has been associated to lower scores in tests of se-mantic verbalfluency and learning of visual association pairs (Cavedini et
al., 2002), whereas impaired set-shifting ability, measured by the
Wisconsin Card Sorting Test (WCST), has been associated with better therapeutic response tofluoxetine (Fontenelle et al., 2001). Moreover, poorer decision-making performance predicted poorer outcome to phar-macological treatment (Cavedini et al., 2002). In youths with OCD, poorer performance on the Rey–Osterrieth Complex Figure Test predicted poorer response to treatment, particularly among those receiving CBT alone (Flessner et al., 2010). On the other hand, some trials have found no asso-ciation between pretreatment neuropsychological performance (visuo-spatial construction, visual memory, and set-shifting) and response to CBT (Bolton et al., 2000; Moritz et al., 2005).
Notwithstanding the studies described above, neuropsychological investigations designed to examine predictors of treatment response in clinical trials comparing simultaneously the two main modalities of intervention are lacking in OCD. This is unfortunate, given evidence in literature suggesting distinct neurobiological mechanisms of action of CBT and pharmacotherapy in OCD (Hoexter et al., 2012).
Therefore, to address these limitations, this study aimed to evalu-ate neuropsychological predictors of treatment response using a broad neuropsychological battery, focusing mainly on executive func-tions. The sample consisted of treatment-naïve adult OCD patients submitted to a 12-week randomized clinical trial of CBT andfl uoxe-tine. The selection of a treatment-naïve population aims to exclude the potential confounding bias of previous treatments on neuropsy-chological domains (Bolton et al., 2000; Kang et al., 2003; Kim et al.,
2002; Krishna et al., 2011). We hypothesized that better
neuropsy-chological performance in respect to executive function, i.e., inhibito-ry control; cognitiveflexibility and planning would predict greater response to both treatment modalities (CBT orfluoxetine). However, given that CBT is assumed to rely on executive functions, specifically cognitiveflexibility (Beck, 1995), patients with better performance on mentalflexibility and learning would be more likely to improve after CBT if compared tofluoxetine. Additionally, we explored wheth-er sociodemographic and clinical variables could influence treatment response.
2. Methods
The present study is a part of a research protocol evaluating multiple neurobiological measures in treatment-naïve OCD patients (Hoexter et al., 2009). Study subjects participated in a larger clinical trial designed to compare the effectiveness of group CBT versusfluoxetine (Belotto-Silva
et al., 2012) (registration athttp://clinicaltrials.gov—NCT00680602). A
detailed description of the assessment instruments (training, and reliabil-ity) can be found in the literature (Miguel et al., 2008). From 2006 to 2008, 369 subjects underwent clinical and psychiatric medical assess-ment in the OCD outpatient clinic at the Institute & Departassess-ment of Psychi-atry, University of São Paulo Medical School, São Paulo, Brazil. From these, 50 treatment-naïve OCD patients fulfilling inclusion criteria accepted participation in this study.
The required written informed consent was approved by the Re-search Ethics Committee of the University of Sao Paulo and signed by all participants.
2.1. Inclusion and exclusion criteria
Inclusion criteria were: age between 18 and 65 years; DSM-IV diagnosis for OCD; score≥16 on Y-BOCS; and, in the presence of other comorbid Axis I diagnoses, OCD as the primary and most severe diagnosis in terms of impairment. Patients were excluded if: they had prior exposure to any psychotropic medication or cognitive behavior therapy (at least 12 sessions); had a general medical condition (such as history of head injury with post-traumatic amnesia); or had severe mental illness other than OCD (i.e., psychotic disorder, active phase bipolar disorder, major depression with significant risk of suicide).
2.2. Clinical assessment
A broad range of clinical instruments was used, including the Struc-tured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) (First et al., 1995), the Y-BOCS, the Dimensional Y-BOCS (DY-BOCS) (
Rosário-Campos et al., 2006), the Beck Depression Inventory (BDI) (Beck et al.,
1961) and the Beck Anxiety Inventory (BAI) (Beck et al., 1988) (see
Hoexter et al., 2009; Belotto-Silva et al., 2012for more details).
2.3. Neuropsychological assessment
Prior to treatment initiation, patients underwent a neuropsycho-logical evaluation. Testing averaged approximately three hours. The application of neuropsychological tests followed a set presentation order. Instruments that required attention and memory were admin-istered early in the session when subjects were less susceptible to fatigue. The application of tests and scoring were performed by a trained neuropsychologist who was blinded to group assignment. Given that previous studies performed in OCD patients have empha-sized impairments in specific neuropsychological domains involving executive functions (planning, cognitiveflexibility, response inhibi-tion, decision making and attentional bias/vigilance) (Cavedini et al.,
2002; Kuelz et al., 2004; Menzies et al., 2008; Shin et al., 2010), we
se-lected the following tests:
2.3.1. Intellectual function
The Wechsler Abbreviated Scale of Intelligence(WASI) was used to estimate full IQ (Wechsler, 1999).
2.3.2. Attention
Digit Span(Forward and Backward) was used to assess auditory attention and short-term retention capacity (Wechsler, 1981).
Trail Making Test(TMT) was used to assess complex visual scan-ning, visual search speed, visual attention, mental flexibility (Part B), and inhibitory control (Wechsler, 1997a, 1997b).
2.3.3. Executive function
Object Alternation Task (OAT) (Freedman, 1990) was used as a measure of cognitiveflexibility.
Hayling and Brixton Test(Burgess and Shallice, 1997) evaluated initiation speed as well as response suppression, while theBrixton Spatial Anticipation(Burgess and Shallice, 1997) test was used as a rule attainment task.
TheWisconsin Card Sorting Test(WCST) (Heaton, 1981) assessed ab-stract behavior, set shifting, response inhibition and mentalflexibility.
Victoria Stroop Test(VST) (Regard, 1981) was used to determine selective attention, mentalflexibility and inhibitory control.
TheIowa Gambling Task (IGT) (Bechara et al., 1994) evaluated decision making.
2.3.4. Memory and learning
Logical Memory(Wechsler, 1997a, 1997b) assessed contextualized short- and long-term memory.
The California Verbal Learning Test(CVLT‐II) (Delis et al., 2000) measured verbal memory and learning.
The Brief Visual Motor Test—Revised(BVMT-R) (Benedict, 1997)
was used to measure visual memory and learning.
The Rey–Osterrieth Complex Figure (ROCF) (Osterrieth, 1944)
assessed visual–spatial constructional ability and visual memory.
2.3.5. Language
TheCOWA (Spreen and Strauss, 1998) assessed verbal fluency
under defined conditions.
2.4. Treatment allocation
After accepting to participate, patients were sequentially allocated to receive treatment between group CBT orfluoxetine. This procedure aimed to minimize possible differences between groups before treat-ment by adjusting for prognostic factors, such as initial Y-BOCS score, gender, and age (Fossaluza et al., 2009).
2.5. Treatment protocols
Patients allocated to CBT were divided into subgroups of 6–8 patients and attended weekly 2-hour therapy sessions for 12 weeks, with three additional follow-up sessions. The protocol emphasized exposure and response prevention exercises as well as cognitive tech-niques, such as correction of thoughts/beliefs and relapse prevention strategies. Patients allocated to pharmacological treatment received fluoxetine (up to 80 mg/day or maximum tolerated dosage) for 12 weeks, starting at 20 mg/day in thefirst week, with 20 mg weekly increases (Belotto-Silva et al., 2012).
2.6. Criteria for treatment response
The primary outcome measure for treatment response in this study was percent reduction in baseline Y-BOCS scores.
Treatment response was analyzed as a continuous variable, in terms of degrees of response, calculated as follows:
Y BOCSinitial−Y BOCSfinal
ð Þ
Y BOCSinitial
100:
2.7. Statistical analyses
Statistical analysis was performed using the computer program R, version 2.9.0 and the Statistical Package for Social Sciences (SPSS) for Windows, version 14.0.P. Statistic indicators inferior to 0.05 were considered statistically significant.
The Mann–Whitney or the chi-square tests were used to evaluate demographic and clinical variables. Correlation between neuropsycho-logical testing results and treatment response (defined as the reduction of initial Y-BOCS scores) was tested using Pearson's correlation coefficient.
Generalized linear models (Mccullagh and Nelder, 1989) were employed to determine the combined influence of neuropsychological parameters on treatment response. Only data from patients who com-pleted the 12-week protocol was used. The following variables were included: treatment type (i.e., CBT or fluoxetine), sociodemographic variables (i.e., gender, age and years of formal education), clinical vari-ables (i.e., Y-BOCS, Beck-A and Beck-D scores) and those neuropsycholog-ical variables which presented minimal correlation (p-valueb0.15) with
treatment response (i.e., CVLT-II perseverative responses, CVLT-II Total recall, IGT A+B, WCST learning to learning, Matrix Reasoning and Verbal IQ on the WASI, VST: part W—word, part C—colors, part C—colors errors, part D—dots, TMT B time and errors). We includedfirst-order interactions between sociodemographic, clinical and neuropsychological variables
and treatment type. Thefinal variable and test protocol model was chosen based on the AIC (Akaike Information Criterion) and BIC (Bayesian Infor-mation Criterion).
In order to test the power of our statistical model, we executed a “power analysis”using the G*Power v. 3.1.3. Including 38 subjects and 10 explanatory variables, we obtained a 0.9999983 “adequateness” index to validate this model.
3. Results
3.1. Sample
Of the 50 OCD patients initially enrolled in the trial, 12 (24%) dropped out in a very early stage of the clinical trial (5 participants from the CBT and 7 from thefluoxetine group). Reasons for abandonment in the CBT group were fast initial reduction of symptoms (n=2), treatment alloca-tion dissatisfacalloca-tion (n=2), and failure to comply with sessions (n=1). Reasons for abandonment in thefluoxetine group included failure to comply with consultations (n=4), treatment allocation dissatisfaction (n=2) and intolerable side effects (n=1). Therefore, only data from the completers (CBT=18 andfluoxetine=20) was analyzed.
3.2. Demographic and clinical characteristics in treatment groups (CBT andfluoxetine)
At baseline, patients assigned to the two treatment groups (CBT and fluoxetine) were similar in terms of demographic (Table 1) and clinical characteristics (Table 2). As can be observed, there was no statistical significant difference between groups in the major sociodemographic and clinical variables.
3.3. Treatment outcome
The mean reduction in Y-BOCS score after treatment was 36% (36.0% for the CBT group and 36.1% for thefluoxetine group, p=0.8). The effect size (Cohen's d) between initial andfinal YBOCS for CBT was 1.17, and forfluoxetine, 1.53, without significant statistical difference between treatments (p=0.08).
3.4. Neuropsychological performance at baseline
Intelligence and neuropsychological measures were similar be-tween treatment groups, with the exception of similarities (WASI), CVLT-II measures—total recall, short delay cued recall, long delay
Table 1
Demographic characteristics of obsessive–compulsive disorder (OCD) patients divided for groups of treatment.
Variable CBT
(n =18)
Fluoxetine (n=20)
p-value
Mean SE Mean SE
Age 32.7 9.7 31.1 10.8 0.65
freq % freq % p-value
Gender
Female 13 72.2 10 50.0 0.16
Socioeconomic status
Classes a/b (higher) 12 66.7 10 50.0 0.29
Classes c/d/e (lower) 6 33.3 10 50.0
Level of educationa
Higher 5 27.8 8 40.0 0.73
Middle 12 66.7 11 55.0
Lower 1 5.6 1 5.0
Right-handed 17 94.4 19 95.0 0.94
Abbreviations: CBT: cognitive behavior therapy.
aHigher: complete tertiary education; middle: incomplete tertiary and complete
cued recall, recognition hits, semantic clustering and OAT, in which patients allocated tofluoxetine showed better performance (Table 3).
3.5. Clinical predictors of treatment response
In our sample, age emerged as a predictor of treatment response to fluoxetine and had a small influence in predicting response to CBT. Elderly individuals responded better to fluoxetine, whereas, less extensively, younger individuals responded better to CBT (pb0.001). In terms of the
severity of OCS, higher initial Y-BOCS scores predicted a better response to both treatments (pb0.001).
3.6. Neuropsychological predictors of treatment response
The neuropsychological domains that were predictive of response to treatment in this study were as follows (Table 4):
Intelligence: a higher score on the verbal IQ section of the Wechsler Abbreviated Scale of Intelligence was found to be predictive of a better response to CBT and had a small influence on response to fluoxetine (p = 0.008).
Verbal memory and learning: better performance on the total recall portion (Trials 1–5 correct) of the CVLT-II translated to a better re-sponse tofluoxetine and presents a slight influence to CBT re-sponse (p = 0.013).
Mentalflexibility: fewer perseverations (CVLT-II) predicted a bet-ter response to CBT (p = 0.002) but worse response tofluoxetine (p = 0.003).
Table 2
Psychopathological features of OCD patients, by treatment group.
Measure Treatment Statistics
CBT (n=18)
Fluoxetine (n=20)
Mean SE Mean SE p-value
BDI 17.6 9.5 17.2 7.5 0.88
BAI 16.2 13.1 16.7 10.0 0.89
Age of onset of OCD 14.8 8.7 11.8 7.2 0.22
DY-BOCS I (aggression, violence) 6.9 4.7 11.0 21.3 0.42 DY-BOCS II (sexual, religious) 8.7 22.9 10.8 21.4 0.76
DY-BOCS III (arrangement) 13.1 21.8 9.0 3.0 0.40
DY-BOCS IV (washing) 6.4 5.0 15.8 28.9 0.18
DY-BOCS V (hoarding) 7.9 23.0 8.2 21.6 0.96
DY-BOCS global 21.2 0.8 22.5 1.0 0.32
Y-BOCS obsessions 13.2 2.4 12.1 3.6 0.26
Y-BOCS compulsions 13.0 3.4 13.0 2.5 0.99
Y-BOCS global 26.3 5.5 25.2 5.1 0.51
% Reduction in Y-BOCS global 36.1 6.0 36.0 6.5 0.99
Comorbidity freq % freq % p-value
Major depressive episode 5 27.7% 9 45.0 0.35
Dysthymic disorder 4 22.2% 2 10.0% 0.30
Bipolar disorder 1 5.6% 1 5.0% 0.93
Generalized anxiety disorder 1 5.6% 0 0.0% 0.82
Social phobia 6 33.3% 8 40.0% 0.75
Specific phobia 4 22.2% 4 20.0% 0.38
Skin picking 2 11.1% 3 15.5% 0.72
Trichotillomania 1 5.6% 2 10.0% 0.61
Body dysmorphic disorder 1 5.6% 1 5.0% 0.56
Pain disorder 1 5.6% 0 0.0% 0.28
Compulsive buying 1 5.6% 2.0 10.0% 0.87
PTSD 0 0.0% 3.0 15.0% 0.18
ADHD 1 5.6 2 10.0 0.74
Panic disorder with agoraphobia 0 0 2 10.0% 0.16
Intermittent explosive disorder 3 16.7% 3 15.0% 0.55
Abbreviations: DY-BOCS = Dimensional Yale–Brown Obsessive–Compulsive Scale. I—aggression, violence and natural disasters; DY-BOCS, II—sexual and religious; III— symmetry, ordering, arranging; IV—contamination and cleaning; V—hoarding; PTSD —post-traumatic stress disorder; ADHD—attention deficit hyperactivity disorder.
Table 3
Pre-treatment neuropsychological test results of 38 OCD patients.
Neuropsychological variables CBT Fluoxetine
Intelligence Mean SE Mean SE p-value
Full IQ 92.00 3.02 91.35 2.58 0.89
Vocabulary 47.33 2.51 47.10 1.93 0.23
Similarities 31.39 1.50 35.10 0.95 0.04
Verbal IQ 86.17 3.39 92.30 2.56 0.15
Block design 41.28 12.23 36.80 14.85 0.31
Matrix 20.83 1.76 22.55 1.56 0.47
Performance IQ 94.0 3.99 92.80 3.33 0.81
Visuospatial memory
ROCF Copy 32.58 0.89 33.02 0.78 0.71
Delayed recall 15.81 1.68 17.37 1.43 0.36
Planning score 3.67 0.41 3.80 0.40 0.81
BVMT Total 24.11 1.80 25.55 1.25 0.51
Learning 5.00 0.46 4.15 0.37 0.16
Delayed recall 10.06 0.61 10.35 0.39 0.69
Percent retained 96.50 3.26 97.00 1.27 0.88
Recognition hits 5.72 0.13 5.80 0.09 0.63
False alarms 0.06 0.56 0.00 0.00 0.33
Discrimination index 5.94 0.22 5.80 0.09 0.55
Response bias 0.39 0.01 0.48 0.02 0.04
Verbal memory
Logical memory (WMS-R) immediate 22.06 1.55 24.05 1.41 0.35
Delayed recall 18.88 1.74 18.80 1.37 0.97
CVLT Total recall 49.83 3.41 59.45 1.79 0.01
Intrusions 1.00 0.16 1.05 0.08 0.92
Perseverationsa 7.67 0.27 6.00 0.23 0.46
Learning 4.94 0.70 5.10 0.50 0.06
Delayed recall 11.06 0.71 12.70 0.59 0.08
Recognition hits 14.67 0.44 15.80 0.11 0.02
Recognition false alarms 0.50 0.20 0.50 0.25 0.02
Attention and inhibitory control
Trail A 36.72 2.25 40.20 3.39 0.40
Errors 0.06 0.56 0.05 0.50 0.00
Trail B 89.66 9.55 95.95 11.22 0.67
Errors 0.11 0.11 0.55 0.26 0.15
Letter number sequence 10.22 0.71 9.58 0.59 0.49
Digit span Total 16.22 1.04 15.10 0.64 0.89
Direct 9.44 0.51 9.25 0.38 0.76
Indirect 6.78 0.72 5.85 0.39 0.25
Stroop test
1st condition 14.29 1.23 14.10 0.73 0.89
Errors 1st condition 0.00 0.00 0.20 0.11 0.12
2nd condition 15.88 0.90 16.60 0.78 0.40
Errors 2nd condition 0.06 0.05 0.10 0.06 0.65
3rd condition 25.17 1.51 26.60 1.77 0.55
Errors 3rd condition 0.47 0.17 1.00 0.32 0.18
Stroop effect 14.29 1.35 14.10 1.61 0.45
Attention and inhibitory control (continuation)
COWA 36.93 2.66 39.56 3.00 0.35
OAT Hits 33.80 1.67 37.42 0.48 0.28
Errors 5.27 1.43 8.11 3.11 0.45
Go/No Go Omissions 6.25 3.42 9.57 4.43 0.62
Comissions 6.25 2.39 8.14 1.93 0.56
Cognitiveflexibility
WCST Completed categories 2.77 0.31 2.85 0.31 0.64
Perseverative responses 5.88 1.96 5.55 1.08 0.91 Perseverative errors 5.16 1.36 4.85 2.04 0.68 Perseverative error
percentage
8.79 2.47 5.91 1.69 0.33
Non‐perseverative errors 14.61 1.57 15.25 1.95 0.80
Loss of set 0.55 0.20 0.35 0.16 0.43
Trails to complete 1st category
15.33 2.81 12.15 0.48 0.24
Conceptual level responses 58.71 3.99 62.42 3.53 0.48 Learning to learning −4.44 2.61 −5.11 2.01 0.83
Inhibitory control: a shorter time to completion of part D (Dots) of the Victoria Stroop Test (VST) was predictive of a better response to both treatments (pb0.001). A longer time to completion of part
W (Words) was predictive of a better response to both treatments (p = 0.025). In part C (Colors), OCD patients who committed few errors showed a better response to both treatments, a result much more pronounced for CBT (pb0.001).
3.7. Hypothetical predictive model of treatment response
Considering that our generalized linear model was appropriate and had good treatment response predictive power, we hypothetical-ly and illustrativehypothetical-ly explored the Y-BOCS expected response rate of a given patient, based on his/her demographic, clinical characteristics and neuropsychological performance. For instance, as can be seen in
Table 5, an OCD patient with age 39, initial YBOCS 25, and the
follow-ing neuropsychological performance: 50 (raw score) on CVLT total recall, 114 (raw score) on verbal IQ, 20 s to complete the VST part D, 30 s to complete the VST part W, and 2 errors on VST part C,
would present a 37% improvement in Y-BOCS scores if submitted to CBT and 12% if submitted tofluoxetine. On the other hand, another patient with the same age 39, initial YBOCS 25, and the following neuropsychological performance: 40 (raw score) on CVLT total recall, 111 (raw score) on verbal IQ, 15 seconds to complete the VST part D, 25 s to complete the VST part W, and 1 error on VST part C, would present a 18% improvement in Y-BOCS scores if submitted to CBT and 57% if submitted tofluoxetine.
4. Discussion
To our knowledge, this is thefirst study to investigate neuropsy-chological predictors of treatment response in a sample of adult treatment-naïve OCD patients submitted to CBT orfluoxetine in a randomized design. Both treatments were shown to be effective in reducing obsessive–compulsive symptoms. The advantage of includ-ing OCD patients never submitted to treatment is the exclusion of potentially confounding effects of previous interventions on neuro-psychological measures.
Congruently with our initial hypothesis, higher baseline neuropsy-chological performance predicted better treatment response to both treatments, with the exception of mentalflexibility, in which fewer perseverations predicted better response to CBT, but worse response tofluoxetine.
4.1. Sociodemographic and clinical predictors of treatment response
In our sample, older patients had a better treatment response tofl uox-etine, whereas, less extensively, among those who were submitted to CBT, younger individuals responded better. In previous studies age has not been associated with treatment outcome in OCD (Ginsburg et al.,
2008; Maher et al., 2010). On the other hand, some studies have found
that longer periods without being treated were predictive of worse re-sponse to pharmacological treatments for OCD (Ravizza et al., 1995), suggesting that longer illness duration (which may be correlated to age) could be linked to the development of neurobiological abnormalities that make the disorder more resistant to pharmacological response.
Ourfindings, however, go in the opposite direction: older patients in this treatment naïve sample had a better treatment response to fluoxetine. We do not have a clear understanding for this discrepancy. We speculate that other clinical variables associated to treatment response may have played a role on this regard. Many clinical vari-ables, beyond age and illness duration, are associated with poor response to treatment in OCD, such as more severe OCD symptoms, anxiety and depression (Ferrão et al., 2006; Mataix-Cols et al., 1999), earlier onset of obsessive–compulsive symptoms (Fontenelle et al.,
2003; Ravizza et al., 1995; Rosário-Campos et al., 2001) and a greater
comorbidity profile (McDougle et al., 1990; Shavitt et al., 2006). Interestingly, in an exploratory analysis, dividing our patients in the fluoxetine-group (n=20) according to the median age (30 years), sub-jects with age≤30 presented significantly higher scores on measures of depression (BDI) (mean±SD: 21.4±5.5), anxiety (BAI) (mean±SD: 21.8±8.1) and had an earlier onset of obsessive–compulsive symptoms (mean±SD: 8.5±1.7) compared to subjects with age>30 (BDI mean±SD: 12±6.4; BAI mean±SD 10.3±8.7: and age of onset of OCS mean±SD: 15.5±9.5) (p≤0.05). These subgroups did not differ in terms of obsessive–compulsive symptom severity (Y-BOCS scores mean±SD: 25.7±5.0 for patients with age≤30 and Y-BOCS scores mean±SD: 24.6±5.5 for patients with age>30). Therefore, in our study, older patients may have responded better tofluoxetine because they were less depressed and anxious and had a later onset of OCS, in spite of longer illness duration.
In addition, ourfindings suggested that younger patients submitted to CBT presented a better treatment response than older patients. Given that CBT is assumed to rely on cognitive skills known as executive functions (Beck, 1995), one possible reason for this result may be the
Table 3(continued)
Neuropsychological variables CBT Fluoxetine
Intelligence Mean SE Mean SE p-value
Hayling test
A 3.21 0.28 3.31 0.28 0.80
B 3.92 0.69 3.75 0.51 0.83
C 5.35 0.65 6.00 0.50 0.43
Total 12.64 1.15 13.06 0.85 0.76
Brixton test
26.66 2.72 25.75 2.10 0.78
Decision making—IGT
Block 1 −1.62 1.34 −1.20 0.80 0.77
Block 2 −0.75 1.16 −2.10 0.94 0.37
Block 3 1.40 1.34 0.28 1.42 0.50
Block 4 −1.20 1.09 −2.14 1.16 0.79
Block 5 −2.40 2.12 0.57 1.40 0.30
General tendency −3.20 2.99 −2.71 5.03 0.94
Abbreviations: ROCF—the Rey–Osterrieth Complex Figure Test, BVMT—Brief Visual Memory Test, WMS-R—Wechsler Memory Scale Revised, CVLT—California Verbal Learning Test, COWA—the Controlled Oral Word Association, OAT—Object Alterna-tion Test, WCST—Wisconsin Card Sorting Test, and IGT—Iowa Gambling Test.
aNo difference is evident here because less perseveration predicted better response
to CBT treatment while more perseveration predicted better response tofluoxetine.
Table 4
Neuropsychological and clinical predictors of treatment responsea.
Independent variables Coefficient SE p-value
(Intercept) −214.791 58.821 0.0017
Type of treatment −9.893 64.937 0.8805
Age −0.399 0.395 0.325
Y-BOCS at baseline 3.717 0.614 b0.001
BDI 0.744 0.383 0.066
Verbal IQ (WASI) 2.379 0.809 0.008
CVLT-II total recall 0.216 0.573 0.710
CVLT-II perseverative responses −2.123 0.602 0.002
VST part D (Dots) time −5.037 0.847 b0.001
VST part W (Words) time 2.685 1.104 0.025
VST part C (Colors) errors −39.459 9.087 b0.001
Type of treatment: age 2.655 0.492 b0.001
Type of treatment: verbal IQ (WASI) −2.696 0.920 0.008 Type of treatment: CVLT-II total recall 1.988 0.728 0.013 Type of treatment: CVLT-II perseverative
responses
4.183 1.236 0.003
Type of treatment: VST part C (Colors) errors 30.820 10.366 0.007
Abbreviations: Y-BOCS—Yale Brown Obsessive–Compulsive Scale, BDI—Beck Depres-sion Inventory, WASI—Wechsler Abbreviated Scale of Intelligence, CVLT-II—California Verbal Learning Test, and VST—Victoria Stroop Test.
aLinear regression:dependent variable: % of response;independent or predictive variables:
type of treatment (fluoxetine or CBT); age; gender; years of education; age at onset of OCD; Y-BOCS score; BDI score, BAI score; and all neuropsychological measures.
higher age-related decrements in executive function observed in older subjects (Gorenstein and Papp, 2007). However, this interpretation should be taken cautiously. It should be acknowledged that in our statis-tical model, age exhibited a weak influence in predicting CBT response
(Table 4). Moreover, our OCD sample comprised patients with a
maxi-mum age of 65, precluding us to compare ourfindings to other studies that found cognitive rigidity in older subjects (Gorenstein and Papp, 2007).
Though the great majority of treatment response studies suggests that more severe OCD symptoms predict worse response to treatment
(Ferrão et al., 2006; Mataix-Cols et al., 1999), our findings have
shown that more severe OCS were predictive of better treatment re-sponse to CBT andfluoxetine. It should be highlighted that previous treatment response studies included patients with OCD that had al-ready been treated, whereas in our study patients were treatment-naïve. Therefore, one might argue that the likelihood offirst attempt treatment responses (measured by changes in Y-BOCS score percent-ages), is greater in patients with more severe OCS. However, limited literature investigating treatment response in treatment naïve pa-tients restrainsfirmer speculations.
4.2. Neuropsychological predictors of treatment response
4.2.1. Verbal IQ, verbal memory and learning
The association of higher verbal IQ and better response to CBT may be a consequence of the fact that patients with greater verbal repertoire may better understand CBT techniques. Therefore, patients with superior ver-bal skills are better candidates for cognitive restructuration (Kuelz et al., 2006). Ourfindings suggested that verbal IQ also predicted treatment response tofluoxetine in a much lesser extent. This supports the concept that IQ, in general, is a good predictive measure of treatment outcome
(Buitelaar et al., 1995). Likewise, the association between higher
perfor-mance on verbal learning memory tasks (CVLT-II on Trials 1–5 correct) and better response to both treatments reinforce a general concept that patients with greater cognitive abilities (a brain with more functional reserve) are candidates to better respond to treatment (Spreen and Strauss, 1998).
4.2.2. Executive functions
Fewer perseverations predicting a better response to CBT and worse fluoxetine response suggest that OCD patients with different mental flexibility profiles may preferentially respond to one type of treatment versus the other. In addition, our results corroborate previous studies showing that executive dysfunction is related to a worse prognosis in patients with OCD who are receiving CBT (Flessner et al., 2010). Fewer perseverations are a marker of better mentalflexibility (Cavedini et al.,
2002; Kuelz et al., 2004), which is certainly advantageous in the setting
of learning alternative behaviors as demanded in CBT. Conversely, more perseverations have been reported to be associated with a better response to fluoxetine (Fontenelle et al., 2001). There is evidence suggesting that altered serotonergic function, a neurotransmission sys-tem by which SRIs exert its action in the prefrontal cortex (El Mansari and Blier, 2006), relates to deficits in mentalflexibility in OCD patients
(Hollander and Wong, 1996). Therefore, our results suggest that more
impaired mentalflexibility should be better acknowledged as a tailoring variable for determiningfluoxetine treatment response.
Regarding VST, better inhibitory control was predictive of better treatment response to both treatments. In previous studies, failure in this domain has been postulated as an endophenotype in OCD
(Lennertz et al., 2012; Rajender et al., 2011) and has been associated
with treatment response to antidepressants in anxiety disorders
(Bespalov et al., 2010). In the CBT context, having a more efficient
in-hibitory control may help individuals with OCD to inhibit maladap-tive behaviors in order to generate newer adapmaladap-tive responses. In addition, given that antidepressants modulate serotonergic neuro-transmission in prefrontal regions (El Mansari and Blier, 2006), sub-jects with a more preserved cognition, evidentiated by a better performance in inhibitory control, are also prone to respond to SRIs.
5. Limitations
This study must be interpreted in the light of some limitations. First, the small sample size and the heterogeneous nature of our participants with respect to their psychiatric comorbidities may have added variability to ourfindings. Thus, our results should be considered exploratory and our findings deserve independent replication before generalization. Second, unexpectedly, some baseline performance measures, such as sim-ilarities in the WASI and total recall in the CVLT-II, differed between groups, which may limit the interpretation of our data. However, it must be acknowledged that the perseveration scores did not differ between groups, reinforcing that this characteristic (cognitive rigidity) indeed shows an opposite effect on treatment response across the two groups.
6. Conclusion
Our study showed that better cognitive performance (i.e., intelli-gence, verbal memory/learning and inhibitory control) predicts better response to treatment regardless of treatment modality in OCD. On the other hand, given the opposite influence of mentalflexibility in predicting treatment response to CBT andfluoxetine our results high-light that different neuropsychological performances may be involved in treatment response for each intervention. Therefore, depending on individual neuropsychological profile, treatment response to a given intervention may differ suggesting that particular treatment can be assigned according to specific neuropsychological features (e.g.: OCD patients with higher verbal IQ and less perseveration tend to respond better to CBT, whereas subjects with better inhibitory control, but more perseverations may respond better tofluoxetine). Further studies with larger samples of OCD patients are necessary to investigate the heuristic value of such measures in a clinical context.
Acknowledgment
This study receivedfinancial support in the form of grants provided by the following Brazilian governmental agencies: the National Council for Scientific and Technological Development (CNPq: Conselho Nacional de Desenvolvimento Científico e Tecnológico), grant numbers: 521369/ 96-7, 475919/2006-8 and 481791/2004-3; and the Foundation for the Support of Research in the State of São Paulo (FAPESP: Fundação de Amparo à Pesquisa do Estado de São Paulo), grant numbers: 05/ 55628-08, 05/04206-6, 06/61459-7 and 06/50273-0). Dr. Carina received
Table 5
Neuropsychological and clinical predictors of treatment response—an example in a clinical practice.
Type of treatment Age YBOCS initial CVLT total recall Verbal IQ VST part D Time
VST part W Time
VST part C Errors
Expected response %
Patient 1 CBT 39 25 50 114 20 30 2 37%
Fluoxetine 39 25 50 114 20 30 2 12%
Patient 2 CBT 39 25 40 111 15 25 1 18%
Fluoxetine 39 25 40 111 15 25 1 57%
research support from FAPESP (06/58286-3). All other authors report no competing interests.
References
Bechara A, Damásio AR, Damásio H, Anderson SW. Insensitivity to future consequences following damage to human prefrontal cortex. Cognition 1994;50(1–3):7-15. Beck JS. Cognitive therapy: basics and beyond. New York: Guilford; 1995.
Beck AT, Ward CH, Mendelson M, Mockj, Erbaugh G. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:561–71.
Beck AT, Epstein N, Brown G, Steer RA. An inventory for measuring clinical anxiety: psychometric properties. J Consult Clin Psychol 1988;56:893–7.
Belotto-Silva C, Diniz JB, Malavazzi DM, Valério C, Fossaluza V, Borcato S, et al. Group cognitive–behavioral therapyversusselective serotonin reuptake inhibitors for obsessive–compulsive disorder: a practical clinical trial. J Anxiety Disord 2012;26(1):25–31. [ Jan] [Epub 2011 Aug 19].
Benedict R. Brief visual memory test—revised: professional manual. Odessa: Psycho-logical Assessment Resources, Inc.; 1997.
Bérdard MJ, Joyal CC, Godbout L, Chantal S. Executive functions and the obsessive– compulsive disorder: on the importance of subclinical symptoms and other concomitant factors. Arch Clin Neuropsychol 2009;24(6):585–98. Se.
Bespalov AY, van Gaalen MM, Gross G. Antidepressant treatment in anxiety disorders. Curr Top Behav Neurosci 2010;2:361–90.
Bolton D, Raven P, Madronal-Luque R, Marks IM. Neurological and neuropsychological signs in obsessive compulsive disorder: interaction with behavioural treatment. Behav Res Ther 2000;38(7):695–708. (Jul).
Buitelaar JK, Van der Gaag RJ, Swaab-Barneveld H, Kuiper M. Prediction of clinical response to methylphenidate in children with attention-deficit hyperactivity disorder. J Am Acad Child Adolesc Psychiatry 1995;34(8):1025–32. (Aug). Burgess P, Shallice T. The Hayling and Brixton tests. Bury St Edmunds: Thames Valley
Test Company; 1997.
Cavedini P, Riboldi G, D'Annucci A, Belotti P, Cisima M, Bellodi L. Decision-making heterogeneity in obsessive–compulsive disorder: ventromedial prefrontal cortex function predicts different treatment outcomes. Neuropsychologia 2002;40:205–11. Chamberlain S, Blackwell A, Fineberg N, Robbins T, Sahakian B. The neuropsychology of obsessive compulsive disorder: the importance of failures in cognitive and behav-ioural inhibition as candidate endophenotypic markers. Neurosci Biobehav Rev 2005;29(3):399–419. (May).
Delis D, Kramer J, Kaplan E, Ober B. California verbal learning test. 2nd edition ed. San Antonio: The Psychological Corporation; 2000.
El Mansari M, Blier P. Mechanisms of action of current and potential pharmacother-apies of obsessive–compulsive disorder. Prog Neuropsychopharmacol Biol Psychi-atry 2006;30(3):362–73. May.
Ferrão YA, Shavitt RG, Bedin NR, de Mathis ME, Lopes A Carlos, Fontenelle LF, et al. Clinical features associated to refractory obsessive–compulsive disorder. J Affect Disord 2006;94(1–3):199–209. (Aug).
Fineberg N, Gale T. Evidence-based pharmacotherapy of obsessive–compulsive disor-der. Int J Neuropsychopharmacol 2005;8(1):107–29. (Mar).
First MB, Spitzer RL, Gibbon M, Williams JBW. Structured clinical interview for DSM-IV axis I disorders—patient edition (SCID-I/P, version 2.0). New York, NY: Biometrics Research Department, New York State Psychiatric Institute; 1995.
Flessner CA, Allgair A, Garcia A, Freeman J, Sapyta J, Franklin ME, et al. The impact of neuropsychological functioning on treatment outcome in pediatric obsessive– compulsive disorder. Depress Anxiety 2010;27(4):365. (Apr).
Foa E, Liebowitz M, Kozak M, Davies S, Campeas R, Franklin M, et al. Randomized, placebo-controlled trial of exposure and ritual prevention, clomipramine, and their combination in the treatment of obsessive–compulsive disorder. Am J Psychiatry 2005;162(1):151–61. (Jan).
Fontenelle L, Marques C, Engelhardt E, Versiani M. Impaired set-shifting ability and therapeutic response in obsessive–compulsive disorder. J Neuropsychiatry Clin Neurosci 2001;13(4):508–10.
Fontenelle LF, Mendlowic MV, Marques C, Versiani M. Early- and late-onset obses-sive–compulsive disorder in adult patients: exploratory clinical and therapeu-tic study. J Psychol Res 2003;37:127–33.
Fossaluza V, Diniz JB, Pereira BeB, Miguel EC, Pereira CA. Sequential allocation to balance prognostic factors in a psychiatric clinical trial. Clinics (Sao Paulo) 2009;64(6):511–8. Freedman M. Object alternation and orbitofrontal system dysfunction in Alzheimer's
and Parkinson's disease. Brain Cogn 1990;14(2):134–43. (Nov).
Ginsburg GS, Kingery JN, Drake KL, Grados MA. Predictors of treatment response in pediatric obsessive–compulsive disorder. J Am Acad Child Adolesc Psychiatry 2008;47(8):868–78. (Aug).
Gorenstein EE, Papp LA. Cognitive–behavioral therapy for anxiety in the elderly. Curr Psychiatry Rep 2007;9(1):20–5. (Feb).
Heaton R. Wisconsin card sorting test manual. Odessa: Psychological Assessment Re-sources, Inc.; 1981.
Hoexter MQ, Shavitt R, D'Alcante CC, Cecconi J, Diniz J, Belotto-Silva C, et al. The drug-naïve OCD patients imaging genetics, cognitive and treatment response study: methods and sample description. Rev Bras Psiquiatr 2009;31(4):349–53. (Dec).
Hoexter MQ, de Souza Duran FL, D'Alcante CC, Dougherty DD, Shavitt RG, Lopes AC, et al. Gray matter volumes in obsessive–compulsive disorder before and afterfluoxetine or cognitive–behavior therapy: a randomized clinical trial. Neuropsychopharmacology 2012;37(3):734–45. (Feb).
Hollander E, Wong CM. The relationship between executive function impairment and sero-tonergic sensitivity in obsessive–compulsive disorder. Neuropsychiatry Neuropsychol Behav Neurol 1996;9:230–3.
Kang DH, Kwon JS, Kim JJ, You T, Park HJ, Kim MS, et al. Brain glucose metabolic changes associated with neuropsychological improvements after 4 months of treatment in patients with obsessive–compulsive disorder. Acta Psychiatr Scand 2003;107:291–7.
Kim MS, Park SJ, Shin MS, Kwon JS. Neuropsychological profile in patients with obses-sive–compulsive disorder over a period of 4-month treatment. J Psychiatr Res 2002;36:257–65.
Koran L. Quality of life in obsessive–compulsive disorder. Psychiatr Clin North Am 2000;23(3):509–17. (Sep).
Krishna R, Udupa S, George CM, Kumar KJ, Viswanath B, Kandavel T, et al. Neuropsy-chological performance in OCD: a study in medication-naïve patients. Prog Neuropsychopharmacol Biol Psychiatry 2011;35(8):1969–76. (Dec 1).
Kuelz AK, Hohagen F, Voderholzer U. Neuropsychological performance in obsessive– compulsive disorder: a critical review. Biol Psychol 2004;65(3):185–236. (Feb). Kuelz AK, Riemann D, Halsband U, Vielhaber K, Unterrainer J, Kordon A, et al.
Neuropsycho-logical impairment in obsessive–compulsive disorder—improvement over the course of cognitive behavioral treatment. J Clin Exp Neuropsychol 2006;28(8):1273–87. (Nov). Lennertz L, Rampacher F, Vogeley A, Schulze-Rauschenbach S, Pukrop R, Ruhrmann S,
et al. Antisaccade performance in patients with obsessive–compulsive disorder and unaffected relatives: further evidence for impaired response inhibition as a candidate endophenotype. Eur Arch Psychiatry Clin Neurosci 2012;23. (Mar). Leon A, Portera L, Weissman M. The social costs of anxiety disorders. Br J Psychiatry
Suppl 1995;27:19–22. (Apr).
Maher MJ, Huppert JD, Chen H, Duan N, Foa EB, Liebowitz MR, et al. Moderators and predictors of response to cognitive–behavioral therapy augmentation of pharmaco-therapy in obsessive–compulsive disorder. Psychol Med 2010;40(12):2013–23. Dec. Mataix-Cols D, Rauch SL, Manzo PA, Jenike MA, Baer L. Use of factor-analyzed symptom
dimensions to predict outcome with serotonin reuptake inhibitors and placebo in the treatment of obsessive–compulsive disorder. Am J Psychiatry 1999;156(9): 1409–16. (Sep).
Mccullagh P, Nelder JA. Generalized linear models. London: Chapman and Hall; 1989. McDougle CJ, Goodman WK, Price LH. Neuroleptic addition influvoxamine-refractory
obsessive–compulsive disorder. Am J Psychol 1990;147:652–4. (19).
Menzies L, Chamberlain SR, Laird AR, Thelen SM, Sahakian BJ, Bullmore ET. Inte-grating evidence from neuroimaging and neuropsychological studies of obses-sive–compulsive disorder: the orbitofronto-striatal model revisited. Neurosci Biobehav Rev 2008;32(3):525–49.
Miguel EC, Ferrão YA, Rosário MC, Mathis MA, Torres AR, Fontenelle LF, et al. The Brazilian Research Consortium on Obsessive–Compulsive Spectrum Disorders: recruitment, assessment instruments, methods for the development of multicenter collaborative studies and preliminary results. Rev Bras Psiquiatr 2008;30(3):185–96. (Sep). Moritz S, Kloss M, Jacobsen D, Fricke S, Cutler C, Brassen S, et al. Neurocognitive
impair-ment does not predict treatimpair-ment outcome in obsessive–compulsive disorder. Behav Res Ther 2005;43(6):811–9. (Jun).
Olley G, Malhi P, Sachdev. Memory and executive functioning in obsessive–compulsive disorder: a selective review. J Affect Disord 2007;104:15–23.
Osterrieth P. Le test de copie d'unefigure complexe: contribution a l'e'tude de la perception et de la memoire (The test of copying a complexfigure: a contribution to the study of perception and memory). Arch Psychol 1944;30:286–350. Pallanti S, Hollander E, Bienstock C, Koran L, Leckman J, Marazziti D, et al.
Treat-ment non-response in OCD: methodological issues and operational defini-tions. Int J Neuropsychopharmacol 2002;5(2):181–91. (Jun).
Park S, Shin M, Kwon J. Neuropsychological profile in patients with obsessive–compul-sive disorder over a period of 4-month treatment. J Psychiatr Res 2002;36(4): 257–65. (Jul–Aug).
Rajender G, Bhatia MS, Kanwal K, Malhotra S, Singh TB, Chaudhary D. Study of neurocognitive endophenotypes in drug-naïve obsessive–compulsive disorder pa-tients, their first-degree relatives and healthy controls. Acta Psychiatr Scand 2011;124(2):152–61. http://dx.doi.org/10.1111/j.1600-0447.2011.01733.x. (Aug, Epub 2011 Jun 16).
Ravizza L, Barzega G, Bellino S, Bogetto F, Maina G. Predictors of drug treatment re-sponse in obsessive–compulsive disorder. J Clin Psychiatry 1995;56(8):368–73. (Aug).
Regard M. Cognitive rigidity andflexibility: a neuropsychological study. Victoria: University of Victoria; 1981.
Rosário-Campos MC, Leckman J, Mercadante M. Adults with early-onset obsessive– compulsive disorder. Am J Psychol 2001;158:1899–903.
Rosário-Campos MC, Miguel EC, Quatrano S, Chacon P, Ferrao Y, Findley D, et al. The Di-mensional Yale–Brown Obsessive–Compulsive Scale (DY-BOCS): an instrument for assessing obsessive–compulsive symptom dimensions. Mol Psychiatry 2006;11(5): 495–504.
Ruscio A, Stein D, Chiu W, Kessler R. The epidemiology of obsessive–compulsive disorder in the National Comorbidity Survey Replication. Mol Psychiatry 2010 Jana;15(1):53–63.
Ruscio AM, Stein DJ, Chiu WT, Kessler RC. The epidemiology of obsessive–compulsive disorder in the National Comorbidity Survey Replication. Mol Psychiatry 2010 Janb;15(1):53–63.
Shavitt RG, Belotto C, Curi M, Hounie AG, Rosário-Campos MC, Diniz JB, et al. Clinical features associated with treatment response in obsessive–compulsive disorder. Compr Psychiatry 2006;47(4):276–81. (Jul-Aug).
Shin NY, Kang DH, Choi JS, Jung MH, Jang JH, Kwon JS. Do organizational strategies mediate nonverbal memory impairment in drug-naïve patients with obses-sive–compulsive disorder? Neuropsychology 2010;24(4):527–33. (Jul). Simpson HB, Rosen W, Huppert JD, Lin SH, Foa EB, Liebowitz MR. Are there reliable
neuropsychological deficits in obsessive–compulsive disorder? J Psychiatr Res 2006;40:247–57.
Spreen O, Strauss E. A compendium of neuropsychological tests: administration, norms and commentary. 2nd edition. New York: Oxford University Press; 1998. Thienemann M, Koran MK. Do soft signs predict treatment outcome in obsessive–
compulsive disorder? J Neuropsychiatry Clin Neurosci 1995;7:218–22. Wechsler D. Wechsler adult intelligence scale. San Antonio: Psychological Corporation;
1997a.
Wechsler D. Wechsler memory scale. 3rd edition. San Antonio, TX: The Psychological Corporation; 1997b.
Wechsler D. The Wechsler abbreviated scale of IQ. San Antonio: Psychological Corpora-tion/ Harcourt Brace and Company; 1999.
Weissman MM, Bland RC, Canino GJ, Greenwald S, Hwu HG, Lee CK, et al. The cross national epidemiology of obsessive compulsive disorder. The Cross National Col-laborative Group. J Clin Psychiatry 1994;55 (Suppl.):5-10.
