Long-term Outcomes of Intensive Inpatient Care for Severe,Resistant Obsessive-Compulsive Disorder: Résultats à long terme de soins intensifs à des patients hospitalisés pour un trouble obsessionnel-compulsif grave et résistant

Abstract

French

Objective:

A substantial proportion of severely ill patients with obsessive-compulsive disorder (OCD) do not respond to serotonin reuptake inhibitors (SRIs) and are unable to practice cognitive behavioral therapy (CBT) on an out-patient basis. We report the short-term (at discharge) and long-term (up to 2 years) outcome of a multimodal inpatient treatment program that included therapist-assisted intensive CBT with adjunctive pharmacotherapy for severely ill OCD patients who are often resistant to SRIs and are either unresponsive or unable to practice outpatient CBT.

Methods:

A total of 420 patients, admitted between January 2012 and December 2017 were eligible for the analysis. They were evaluated using the Mini International Neuropsychiatric Interview, the Yale-Brown Obsessive Compulsive Scale (YBOCS), and the Clinical Global Impression (CGI) scale. All patients received 4 to 5 therapist-assisted CBT sessions per week along with standard pharmacotherapy. Naturalistic follow-up information at 3, 6, 12, and 24 months were recorded.

Results:

At baseline, patients were mostly severely ill (YBOCS = 29.9 ± 4.5) and nonresponsive to ≥2 SRIs (83%). Mean duration of inpatient stay was 42.7 ± 25.3 days. At discharge, there was a significant decline in the mean YBOCS score (29.9 ± 4.5 vs. 18.1 ± 7.7, P < .001, Cohen’s d = 1.64); 211/420 (50%) were responders (≥35% YBOCS reduction and CGI-I≤2) and an additional 86/420 (21%) were partial responders (25% to 35% YBOCS reduction and CGI-I≤3). Using latent class growth modeling of the follow-up data, 4 distinct classes were identified, which include “remitters” (14.5%), “responders” (36.5%), “minimal responders” (34.7%), and “nonresponders” (14.6%). Shorter duration of illness, better insight, and lesser contamination/washing symptoms predicted better response in both short- and long-term follow-up.

Conclusion:

Intensive, inpatient-based care for OCD may be an effective option for patients with severe OCD and should be considered routinely in those who do not respond with outpatient treatment.

Keywords

obsessive-compulsive resistant treatment outcome predictors residential treatment combined/multimodal treatment

Introduction

Obsessive-compulsive disorder (OCD) is a common^1,2 and disabling mental illness.³ While outpatient treatment is the standard strategy for OCD, inpatient/residential programs that involve a combination of intensive cognitive behavioral therapy (CBT) with exposure and response prevention (EX/RP) and medications are being explored as potential treatment approaches for severe and often treatment-resistant patients.^4–6 CBT is effective, both as a first-line treatment for OCD^7,8 and as an augmenter in patients with poor response to serotonin reuptake inhibitors (SRIs).^9,10 However, many patients may not be able to participate in outpatient CBT, mainly due to intolerable anxiety.^11,12 About 20% to 30% of patients are known to drop out of EX/RP prematurely.^13,14 Inpatient care offers a unique opportunity to carry out intensive CBT that is perhaps more effective than less intense approaches, at least in the short-term.^15

–18

A meta-analysis of 19 studies of residential/inpatient/day-care programs involving 2,306 patients reported a significant reduction in the severity of illness with a large effect size.⁴ However, long-term benefits of such an approach are unclear. Moreover, previous studies have failed to identify consistent predictors of outcomes mostly due to their small sample sizes. Given the high costs of these programs, it is important to examine long-term outcomes and their predictors in large samples.

A careful review of the residential/inpatient programs show that family members do not stay with patients during the period of hospitalization although involving them in treatment is vital for improving outcomes. Family accommodation plays a key role in the outcome and long-term prognosis of OCD.^19,20 Treatment programs that are “family-inclusive” are superior to those that are not.^21,22 The inpatient treatment program of the OCD Clinic at the National Institute of Mental Health & Neurosciences (NIMHANS), Bangalore, India, is unique in one way. The immediate family members (mostly a first-degree relative or the spouse) stay with patients during the entire course of hospital stay. This is a usual practice in all inpatient medical settings in India, which thus gives us an opportunity to involve family members in dealing with family accommodation, expressed emotions and proxy compulsions. This is especially relevant in our cultural context, where family accommodation is high and contributes to poor outcome.²³

We report the naturalistic short- (at discharge) and long-term (up to 2 years of follow-up) outcome and its predictors in a large sample of patients from the inpatient treatment program of the NIMHANS OCD Clinic in India. We also sought to look at clinical and sociodemographic characteristics that might predict response at discharge and at long-term follow-up.

Method

We reviewed the clinical charts of all inpatients with a primary diagnosis of DSM-IV-TR OCD from January 2012 to December 2017. The patients were admitted primarily for adjunctive intensive CBT treatment, in view of severe OCD which had little or no response to outpatient management. It is not the usual practice at our center to admit patients with a primary diagnosis of OCD for purposes other than adjunctive intensive CBT. For those patients who were admitted more than once for inpatient CBT, only the first admission was considered for analysis. As this is a chart review, no informed consent was obtained from the patients, but Institute Ethics Committee approved the study for ethical aspects. Out of a total of 451 unique cases, 31 (5.8%) were excluded because of very brief (<2 weeks) admission (16/31), inability to trace records (9/31), and inadequate clinical data (6/31). Hence, a total of 420 patients were included for the analysis.

Assessments

All patients underwent detailed clinical evaluation followed by a structured interview using the Mini International Neuropsychiatric Interview 6.0.0 (until 2016) and 7.0.0 (2017 onward).²⁴ In addition, patients were evaluated at the time of admission, discharge, and during the follow-up using the Yale-Brown Obsessive Compulsive Scale (YBOCS)^25,26 and the Clinical Global Impression (CGI) scale.²⁷ We administered the Structured Clinical Interview for DSM-IV for Personality Disorders²⁸ or the International Personality Disorder Examination-II²⁹ upon a clinical suspicion of the presence of personality disorder. All the assessments were performed by either postgraduate junior residents in psychiatry or postgraduate/doctoral students in clinical psychology or psychiatric social work. All raters received training sessions in administering the above-mentioned instruments every 3 months and were supervised by the senior residents or the consultants of the OCD clinic. The internal consistency (Cronbach α) of the YBOCS severity ratings at admission and discharge were found to be fairly high at 0.84 (obsessions) and 0.87 (compulsions), having a high convergent validity of the total score with the CGI-S ratings (r = 0.89).

Inpatient Care

All patients continued to receive SRIs along with intensive CBT (see Table 1) as per the existing standards.^30,31 Decisions to change SRI/add augmentation/adjust dosages were made after evaluating the adequacy of and response to the ongoing medication trial. These were made in the weekly grand rounds headed by one of the consultants of the OCD clinic.

Table 1.

Pharmacological Treatment Details of the Sample.

Treatment Characteristic			Mean (SD)/n (%)
Treatment prior to inpatient care
No. of failed adequate SRI trials^a			2.59 (1.29)
No. of failed pharmacological augmentation trials			1.21 (1.03)
SRI resistance (failed ≥2 adequate SRI trials)			348 (82.85%)
Failed 2 augmentation trials (at least 1 antipsychotic)			134 (31.90%)
History of poor response to outpatient CBT^b			131 (31.19%)
Treatment during inpatient care
Number of days of inpatient care			42.74 (25.31)
SRI (mean dose ± SD in mg)	Fluoxetine (71.51 ± 11.97 mg)		79 (18.57%)
	Sertraline (229.11 ± 49.75 mg)		78 (18.81%)
	Escitalopram (26.03 ± 5.18 mg)		78 (18.57%)
	Fluvoxamine (280.65 ± 43.75 mg)		62 (14.76%)
	Clomipramine (183.56 ± 37.92 mg)		54 (12.86%)
	Venlafaxine (232.69 ± 23.05 mg)		39 (9.29%)
	Paroxetine (70.24 ± 10.06 mg)		21 (5%)
	Desvenlafaxine (116.67 ± 28.87 mg)		3 (0.71%)
	Vilazodone (60 ± 20 mg)		3 (0.71%)
	No SRI		3 (0.71%)
Augmentation (mean dose + SD in mg)	Risperidone (2.72 ± 1.24 mg)		78 (18.57%)
	Aripiprazole (13.37 ± 4.51 mg)		52 (12.38%)
	Quetiapine (209.38 ± 118.63 mg)		16 (3.81%)
	Haloperidol (15.75 ± 8.5 mg)		4 (0.95%)
	Clomipramine (75 mg)		6 (1.43%)
	Buspirone (30 mg)		3 (0.71%)
	Ondansetron (4.5 ± 2.33 mg)		8 (1.9%)
	Granisetron (2 mg)		2 (0.48%)
	Mirtazapine (15 mg)		2 (0.48%)
	Memantine (20 mg)		5 (1.19%)
	N-Acetyl cysteine (2200 ± 346.41 mg)		3 (0.71%)
	Lamotrigene (200 mg)		2 (0.48%)
	Any augmenting agent		175 (41.47%)
Concomitant Anxiolytic/Sedatives (mean dose + SD in mg)	Clonazepam (0.98 ± 0.60 mg)		133 (31.67%)
	Pregabalin (235 ± 117.73 mg)		31 (7.38%)
	Zolpidem (10 mg)		16 (3.81%)
	Lorazepam (2 mg)		10 (2.38%)
	Trazodone (125 ± 50 mg)		8 (1.9%)
Management during follow-up (n = 330)
Length of follow-up in months			25.12 (17.06); median = 21
Readmission			56 (13.33%)
Outpatient CBT (“booster CBT”)			42 (10%)
Medication management summary		No change	125 (37.88%)
		Change to another SRI/clomipramine	45 (13.64%)
		Change/addition of augmenting agent only	77 (23.33%)
		Both SRI change and subsequent augmentation	83 (25.15%)

Note. N = 420. SRI = serotonin reuptake inhibitor.

^a Defined as completed 8 to 12 weeks at adequate dosing in accordance with APA/IJP Clinical Practice Guidelines.

^b Completed at least 10 sessions of outpatient-based cognitive behavioral therapy (CBT).

Intensive CBT (4 to 5 sessions/week, each session for 60 to 90 minutes) included psychoeducation, EX/RP with cognitive restructuring, psychoeducation, and relapse prevention.³² Trained mental health professionals (psychiatry residents and doctoral students in clinical psychology or psychiatric social work) delivered CBT under the close supervision of a therapist with expertise in administering CBT for OCD. Following initial assessment and deciding upon the exposure hierarchy/goals, therapist-assisted EX/RP sessions were held daily for 1 hour, self-EX/RP homework tasks for 2 to 3 hours were included as part of a daily activity schedule.

Nearly all patients (409/420; 97%) stayed with a family member who was most often a parent (67%), spouse (23%), sibling (5%), or child/other relative (2%). The family member staying with the patient in the hospital was actively involved in the therapy. They were educated about OCD, theoretical rationale behind CBT (EX/RP in particular), and role of family accommodation in maintenance of symptoms. They were also frequently enlisted as co-therapists to aid in carrying out additional assisted EX/RP sessions and help in monitoring homework tasks. Progress in CBT was also reviewed in detail during the weekly grand rounds, in the presence of the therapist, patient, and family member.

Outcome Measures

Response was defined as ≥35% reduction in the YBOCS total score with CGI-I ≤2; partial response as 25% to 35% reduction in YBOCS total with CGI-I ≤3.³³ A YBOCS score of ≤12 with CGI-S ≤2 defined remission.³³

Follow-up

After discharge from the hospital, patients continued to receive treatment in a naturalistic setting (Table 1). Follow-up data were available for varying number of patients at 3 (n = 321, 76.4%), 6 (n = 262, 62.4%), 12 (n = 232, 55.2%), and 24 (n = 169, 40.2%) months. A total of 330 (78.57%) patients had at least 1 follow-up (median duration of follow-up = 20.5 [inter-quartile range 8 to 33] months).

Statistical Analysis

Chi-square/Fisher exact test and t test were used to compare categorical and continuous variables, respectively. We performed factor analysis of the main categories of the YBOCS symptom checklist (except miscellaneous) by principal component analysis with “varimax” rotation. Those components with eigenvalue greater than 1 were included, and items with factor loadings >0.5 were considered as representative of that particular factor. Factor scores (regression method) were generated for each subject for use in the subsequent analyses. For predictors of response at the time of discharge, binary logistic regression (enter method) was used. Independent variables were chosen based on extant literature on predictors of outcome of OCD³⁴ and of outcome of intensive residential treatment.⁴ Additionally, variables with P < 0.2 in the univariate analysis were also included as predictors.

In order to assess longitudinal course of illness following inpatient treatment, we performed a censored-normal latent class growth analysis (LCGA) using MPlus (version 8.1).³⁵ LGCA is an approach to model classes of longitudinal treatment response trajectories based on the available data.³⁶ This method helps to identify whether the response patterns of individual patients fall under subclasses with distinct, homogenous trajectories. The YBOCS total scores measured at baseline, at the time of discharge, and at each follow-up were used for modeling the latent growth trajectories. The modeling procedure assumed a quadratic growth pattern underlying the response trajectories. Statistical measures like Bayesian information criterion (BIC), Akaike information criterion, sample size–adjusted BIC, Lo–Mendell–Rubin likelihood ratio test, and the bootstrap likelihood ratio test in conjunction with clinical meaningfulness of the results were used to determine the optimal number of latent classes. LCGA employs the expectation–maximization (EM) algorithm for determination of latent classes and their corresponding growth factors. This algorithm takes into account all available information on individual subjects at each time point; hence, all patients are included in the final model solution regardless of missing data. Therefore, there was no requirement for using any imputation procedure to account for the missing data. Baseline characteristics that were found to predict response at discharge were used as covariates in the subsequent conditional LCGA model, which uses a multinomial logistic regression to look for those factors that could predict “class-membership,” that is, long-term outcome trajectories.

Results

Treatment details and clinical characteristics of the sample are described in Tables 1 and 2, respectively. Principal component analysis of the YBOCS checklist items (excluding miscellaneous symptoms) resulted in a 5-component solution: PC-1 (symmetry/arranging): with robust loadings on symmetry obsessions (0.908) and arranging/ordering compulsions (0.924), PC-2 (contamination/washing): contamination obsessions (0.952) and cleaning/washing compulsions (0.947), PC-3 (hoarding/collecting): hoarding (0.946) and collecting (0.938), PC-4 (doubts/checking): pathological doubts (0.809) and checking (0.823), PC-5 (forbidden thoughts): religious (0.780), sexual (0.698), and aggressive obsessions (0.534). The 5 factors accounted for 68% of the total variance.

Table 2.

Clinical and Demographic Characteristics of the Sample at the Time of Admission.

Clinical/Demographic Characteristics		Mean (SD) or n (%)
Age (years)		29.54 (10.77)
Age at onset of OCD (years)		19.28 (8.54)
Duration of illness (years)		10.18 (7.57)
Sex	Male	251 (59.76%)
Marital status	Single	258 (61.43%)
	Married	143 (34.05%)
	Separated/divorced	16 (3.81%)
	Widowed	3 (0.71%)
Domicile	Urban	325 (77.38%)
Domicile	Rural	96 (22.86%)
Years of education		12.78 (2.83)
YBOCS checklist obsessions (current)	Contamination^a	259 (61.67%)
	Somatic	63 (15%)
	Aggressive	144 (34.29%)
	Sexual	118 (28.1%)
	Religious	123 (29.29%)
	Symmetrical	127 (30.24%)
	Hoarding	42 (10%)
	Pathological doubt	200 (47.62%)
	Miscellaneous	139 (33.1%)
YBOCS checklist compulsions (current)	Washing^a	261 (62.14%)
	Arranging	107 (25.48%)
	Checking	223 (53.1%)
	Repeating	197 (46.9%)
	Counting current	51 (12.14%)
	Collecting	37 (8.81%)
	Cognitive (mental rituals)	187 (44.52%)
	Miscellaneous	247 (58.81%)
	Proxy compulsions	186 (44.29%)
YBOCS Severity Scale	Obsession subscore	15.72 (9.38)
	Compulsion subscore	14.67 (2.74)
	YBOCS total	29.94 (4.5)
	Insight (YBOCS-11)	1.61 (0.87)
	Avoidance	2.19 (1.02)
	CGI-S	5.55 (0.71)
Insight^b (dichotomous)	Poor	69 (16.43%)
Insight^b (dichotomous)	Good	351 (83.57%)
Major depressive disorder (current)		174 (41.43%)
Dysthymia (current)		31 (7.38%)
Mania/hypomania (lifetime)		33 (7.86%)
Psychosis (lifetime)		35 (8.33%)
Generalized anxiety disorder (current)		27 (6.43%)
Social anxiety disorder (current)		24 (5.71%)
Panic disorder (current)		8 (1.9%)
Agoraphobia (current)		4 (0.95%)
Tourette’s/tic disorder (current)		18 (4.29%)
Alcohol use disorder (current)		6 (1.43%)
Other obsessive-compulsive related disorders (current)	Body dysmorphic disorder	17 (4.05%)
	Hoarding disorder	1 (0.24%)
	Olfactory reference syndrome	1 (0.24%)
	Skin picking disorder	2 (0.48%)
	Trichotillomania	4 (0.95%)
Personality disorder (evaluated using IPDE/SCID-II-PD if clinically suspected)	None	302 (71.9%)
	Schizoid	1 (0.24%)
	Schizotypal	16 (3.81%)
	Borderline	36 (8.57%)
	Histrionic	2 (0.48%)
	Antisocial	2 (0.48%)
	Narcissistic	1 (0.24%)
	Anankastic	31 (7.38%)
	Dependent	6 (1.43%)
	Anxious avoidant	17 (4.05%)
	Mixed	7 (1.67%)

Note. N = 420. YBOCS = Yale Brown Obsessive-Compulsive Scale; CGI-S = Clinical Global Impression–Severity; OCRD = Obsessive-Compulsive related disorders; IPDE = International Personality Disorder Examination; SCID-II-PD = Structured Clinical Interview for DSM-IV-Personality Disorders.

^a216/420 (51.42%) had contamination obsessions and washing compulsions as the principal symptoms.

^b Good insight= excellent, good or fair insight on item 11 of the YBOCS, poor insight = poor or lack of insight on item 11 of the YBOCS.

Outcome at the Time of Discharge

The total YBOCS score declined significantly by the time of discharge, along with commensurate improvements in insight, and reduction of avoidance and global severity of illness (Table 3). More than half of the sample were “responders” at the time of discharge. Additionally, about a fifth of the sample was “partial responders.” In essence, about 70% of the sample was either “responders” or “partial responders” (297/420). In our study, about a third of our patients (n = 131) had a history of poor response to outpatient CBT. Among these, 65/131 (49.6%) were “responders,” and an additional 26/131 (19.4%) were “partial responders” at the time of discharge, similar to those who were CBT naïve (χ2 = 0.011, df=1, P = 0.961).

Table 3.

Outcomes at the Time of Discharge.

Outcome		Admission (n = 420)	Discharge (n = 420)	t/χ²/Z	P
YBOCS total		29.94 (4.50)	18.13 (7.73)	−33.76^a	<0.001^b
Insight	Poor	69 (16.43%)	14 (3.33%)	53.02^c	<0.001
Insight	Good	351 (83.57%)	406 (96.67%)	53.02^c	<0.001
Avoidance		2.19 (1.02) (median = 2)	1.02 (1.03) (median = 1)	−19.83^a/−14.53^d	<0.001
CGI-S		5.55 (0.71) (median = 6)	3.83 (1.07) (median = 4)	−34.8^a/−17.14^d	<0.001
CGI-I		—	2.45 (0.71) (median = 2)	na
Mean decline in YBOCS		—	−11.81 (7.17)	na
% YBOCS total reduction		—	39.70 (23.16)	na
Responders		—	211 (50.24%)	na
Partial responders		—	86 (20.48%)	na
Nonresponders		—	123 (29.29%)	na
Remitters		—	38 (9.05%)	na

Note. N = 420. YBOCS = Yale Brown Obsessive-Compulsive Scale; CGI-S = Clinical Global Impression–Severity; CGI-I = Clinical Global Impression–Improvement; na = not applicable.

^a Paired t.

^b Cohen d = 1.64.

^c McNemar χ².

^d Wilcoxon Z.

We compared baseline clinical characteristics (i.e., at the time of admission) of “responders” (includes “remitters”) and “nonresponders.” Compared to “responders,” “nonresponders” were overrepresented by males (138 [66%] vs. 112 [53%], χ2 = 7.306, df=1, P = 0.007), had an earlier onset of illness (18.31 [8.26] vs. 20.26 [8.73], t = 2.34, P = .020) along with a longer duration of illness (11.30 [8.01] vs. 9.11[6.97], t = 3.00, P = 0.003), poor insight (51 [24%] vs. 18 [9%], χ2=19.26, df=1, P < 0.001), more of contamination/washing (139 [67%] vs. 120 [57%], χ2=4.12, df=1, P=0.042), hoarding (25 [12%] vs. 12 [6%], χ2=5.146, df=1, P=0.023), less of current major depressive disorder (75 [36%] vs. 99 [47%], χ2 = 5.268, df=1, P = 0.021), greater lifetime psychosis (25 [12%] vs. 10 [5%], χ2 = 7.17, df=1, P=0.007), and a higher number of failed SRI (2.72 [1.46] vs. 2.46 [1.08], t = 2.063, P=0.039) and augmentation trials (1.33 [1.09] vs. 1.08 [0.95], t = 2.538, P = 0.01). In essence, they were perhaps more severely ill at baseline, also reflected in the global severity rating (5.67 [0.74] vs. 5.45 [0.67], t =3.18, P=0.002). Logistic regression analysis was then carried out using these variables and other clinically relevant factors (Table 4). The overall model had a prediction accuracy of 67.4% (Nagelkerke R2 = 0.183, P < 0.001). Significant predictors of better response at the time of discharge were female gender, shorter duration of illness, less of contamination and washing, and better insight.

Table 4.

Short-term and Long-term Predictors of Outcome following Inpatient Treatment.

Variable	β (SE)	Odds Ratio (95% CIs)	P Value
Predictors of outcomes at discharge using logistic regression (responders 211 [50.2%]; nonresponders 209 [49.8%])
Duration of illness (years)	−0.042 (0.015)	0.959 (0.93to 0.988)	0.006
Sex (0 = male; 1 = female)	0.624 (0.235)	1.866 (1.178 to 2.956)	0.008
PC-1 (symmetry/arranging)	−0.003 (0.106)	0.997 (0.81 to 1.227)	0.977
PC-2 (contamination/washing)	−0.253 (0.113)	0.776 (0.623 to 0.968)	0.025
PC-3 (hoarding/collecting)	−0.062 (0.113)	0.94 (0.753 to 1.174)	0.587
PC-4 (pathological doubts/checking)	0.041 (0.107)	1.041 (0.844 to 1.285)	0.705
PC-5 (forbidden thoughts)	0.054 (0.11)	1.055 (0.851 to 1.308)	0.625
YBOCS total score	−0.01 (0.024)	0.99 (0.944 to 1.039)	0.688
Insight (0 = good, 1 = poor)	−1.13 (0.329)	0.323 (0.17 to 0.615)	0.001
Number of failed adequate SRI trials	−0.129 (0.105)	0.879 (0.716 to 1.08)	0.219
Past CBT failed (0 = no, 1 = yes)	−0.343 (0.254)	0.709 (0.431 to 1.167)	0.176
Number of failed augmentation trials	−0.12 (0.126)	0.887 (0.693 to 1.136)	0.342
Comorbid major depressive disorder (current)	0.233 (0.221)	0.792 (0.513 to 1.221)	0.291
Comorbid psychosis (lifetime)	−0.488 (0.438)	1.63 (0.69 to 3.846)	0.265
Comorbid personality disorder	−0.394 (0.236)	1.483 (0.933 to 2.356)	0.095
Constant	0.769 (0.916)	—	0.401
Predictors of long-term outcome trajectory classes using conditional latent class growth analysis
Class 1: “Nonresponders” (14.6%): Reference class
Class 2: “Minimal responders” (34.7%)
Duration of illness (years)	−0.08 (0.039)	0.923 (0.855 to 0.996)	0.038
Sex (0 = male, 1 = female)	0.297 (0.571)	1.346 (0.439 to 4.121)	0.603
PC-2 (contamination/washing)	−0.611 (0.385)	0.543 (0.255 to 1.154)	0.112
Insight (0 = poor, 1 = good)	−0.368 (0.592)	0.692 (0.217 to 2.209)	0.534
Class 3: “Remitters” (14.5%)
Duration of illness (years)	−0.341 (0.319)	0.711 (0.381 to 1.329)	0.285
Sex (0 = male, 1 = female)	0.558 (0.601)	1.747 (0.538 to 5.674)	0.353
PC-2 (contamination/washing)	−0.979 (0.416)	0.376 (0.166 to 0.849)	0.018
Insight (0 = poor, 1 = good)	1.747 (1.161)	5.737 (0.589 to 55.844)	0.133
Class 4: “Responders” (36.3%)
Duration of illness (years)	−0.133 (0.038)	0.875 (0.813 to 0.943)	0.001
Sex (0 = male, 1 = female)	0.024 (0.502)	1.024 (0.383 to 2.74)	0.962
PC-2 (contamination/washing)	−0.897 (0.328)	0.408 (0.214 to 0.776)	0.006
Insight (0 = poor, 1 = good)	0.992 (0.622)	2.697 (0.797 to 9.126)	0.11

Note. N = 420. YBOCS = Yale Brown Obsessive-Compulsive Scale; CBT = cognitive behavioral therapy.

Outcomes at Long-term Follow-up

We examined whether those who were followed up (n = 330) differed from those who were not (n = 90) with respect to clinical characteristics and key outcome variables. A significant difference was found only in terms of age (younger age among those who did follow-up, P = .027). There were also no differences between the outcomes (response) at discharge in terms of YBOCS total score (t = .573, P = .567), percentage reduction in the YBOCS score (t = 1.005, P =0.315) or in the response rates (χ2 = 0.477, df=1, P = 0.515). Additionally, Little’s “Missing Completely at Random” test using EM algorithm was applied to the YBOCS total score at each time point and was not found to be significant (χ² = 65.15, df = 49, P = 0.061).

Using the YBOCS total scores data at follow-up, latent class growth modeling was done to look for specific long-term outcome trajectories. A 4-class solution was found to have the best fit indices, with an entropy value of 0.76 which indicates a moderate level of fit.³⁶ As shown in Figure 1, 14.6% were classified under latent class 1 with a nearly flat trajectory (“LC nonresponders”), 34.7% in latent class 2 (“LC minimal responders”), 14.5% in latent class 3 (“LC remitters”), and 36.3% in latent class 4 (“LC responders”).

Figure 1.

Growth trajectories of YBOCS total scores at admission, discharge, and long-term follow-up. YBOCS = Yale Brown Obsessive-Compulsive Scale. For the purpose of latent class growth curve modeling, the value of the time point “admission” was considered to be 0, and “Discharge” as 1.5 which was the mean duration of admission. Each subsequent time point—3, 6, 12, and 24 months were given the values 4.5, 7.5, 13.5, and 25.5, respectively.

Using latent class 1 (“LC nonresponders”) as the reference class, we examined whether predictors of short-term outcome (gender, duration of illness, contamination factor score, and insight) predicted long-term trajectories (Table 4). A shorter duration of illness and lesser contamination/washing (PC-2) were found to be associated with latent class 4 (“LC responders”). Lesser contamination was also found be associated with latent class 3 (“LC remitters”), and a slightly lesser duration of illness was associated with latent class 2 (“LC minimal responders”).

Since contamination emerged as one of the predictors of outcome both at discharge and at follow-up, we examined whether those with “principal contamination and washing” symptoms (n = 216, 51.4%) differed from those in whom contamination was not a principal symptom (n = 204, 48.6%). Those with principal contamination and washing were more severely ill in terms of the baseline YBOCS score (30.98 [4.55] vs. 28.84 [4.19], t = 5.013, P < 0.001), the CGI-S (5.7 [0.74] vs. 5.4 [0.65], t = 4.497, P < 0.001), the avoidance (2.47 [0.93] vs. 1.89 [1.03], t = 3.83, P < 0.001), the presence of proxy compulsions (136/216 [63%] vs. 50/204 [24.5%]; χ2 = 62.87, df=1, P < 0.001).

Discussion

Main findings of our study are as follows: (1) At discharge, about 70% had shown some response to inpatient care with a mean decline of about 12 points in the YBOCS total score. (2) During the long-term follow-up, latent class growth modeling showed that over 50% continued to retain treatment gains. (3) A shorter duration of illness, better insight, and less of contamination symptoms predict better response in the short- and long-term follow-up.

To the best of our knowledge, only two other centers have reported outcomes of such intensive residential/inpatient care in large samples, one being the McLean Hospital at Harvard in United States^5,37

–41 and the other being the National Health Service (NHS) in United Kingdom.⁶ To put our findings in perspective, the similarities and differences in clinical characteristics of these two studies with those of our study have to be kept in mind. The sample characteristics in our study are somewhat comparable to those in the two large studies in that patients were severely ill and resistant to SRIs. However, the rates of comorbid illnesses, especially substance use, were much lesser in our sample similar to other studies published from our center.^42
–44 The characteristics of the interventions offered during inpatient management, that is, pharmacotherapy and intensive CBT also seem to be similar. The mean duration of inpatient stay was shorter in our study (42.7 days or 6.1 weeks) compared to that of other studies (10.4 weeks).⁶ Another recent study from Norway⁴⁵ looked at the outcomes of a 3-week intensive inpatient program in a sample of 187 subjects, but this was a sample in which only 54.7% were described as “being on antidepressants,” with no other mention of the degree of their resistance to treatment. On the other hand, majority (83%) of patients in our sample were SRI-resistant. As mentioned previously, at our center, immediate family member(s) stayed with the patient in the hospital and participated in the treatment.

Outcomes

The outcome of inpatient care at discharge was highly robust (mean YBOCS reduction = 11.8 points; Cohen d = 1.64; Table 3, comparable to the finding of the meta-analysis (Mean YBOCS reduction of 10 points, and an effect size [Hegde g] of 1.87).⁴ In terms of percentage of responders, the two other studies with large sample sizes also found similar rates of response, that is, 59.3%³⁷ and 43.2%⁶ but used a 25% YBOCS reduction cut-off to classify responders,⁴⁶ whereas we employed ≥35% reduction in the YBOCS and a CGI-I score of 1 (very much improved) or 2 (much improved).³³ The recent study from Norway⁴⁵ found a much higher (79.9%) response rate using ≥35% YBOCS reduction criteria with a much shorter, 3-week intervention. This possibly could be due to lesser “treatment resistance” in their sample as mentioned earlier. Additionally, we have demonstrated that, in spite of the attrition, improvement seems to persist postdischarge from inpatient facility. The LCGA shows that about 50% of the patients were either “remitters” or “responders” during the follow-up period.

Encouraging findings of our study offer further support to routinely recommend intensive inpatient care to patients who are severely ill, resistant to SRIs, and unable to practice CBT on an outpatient basis. In our study, even those who had a history of poor response to outpatient CBT responded (responders = 49.6%, partial responders = 19.4%) following inpatient care, suggesting that inpatient care may be useful for even those who had a history of poor response to CBT in the past. It is also a common clinical experience that many patients with severe illness cannot practice EX/RP on an outpatient basis. It is in this context that findings of our study assume special significance. OCD is an illness that is typically managed on an outpatient basis, and several guidelines^47
–49 do not discuss inpatient/residential treatment explicitly as a potential alternative for severely ill patients. Only the National Institute of Clinical Excellence guidelines⁵⁰ and the Indian Psychiatry Society guidelines³¹ recommend inpatient treatment before considering ablative surgery and deep brain stimulation.

Predictors of Outcome and Trajectories of Improvement

The strongest predictor of outcome both at discharge and at long-term follow-up was found to be insight. This is an important predictor frequently replicated both in naturalistic follow-up studies^51

–54 and in intervention studies.^34,55 Despite the fact that poor insight is associated with treatment challenges⁵⁶ and poorer outcome,⁵⁴ there is no controlled data that specifically address efficacy of various treatments or their combinations. Our finding further emphasizes the need to study treatment of poor insight OCD systematically.

None of the earlier studies of inpatient treatment have reported on the duration of illness being a predictor of response. However, several recent reports have found that duration of illness may be an important prognostic factor in OCD and that longer duration of illness may portend poorer outcomes.^57
–59 Hence, it is plausible that the duration of illness also has some bearing on the outcome of intensive inpatient treatment. Longer duration of illness also is known to correlate with greater illness severity, poorer insight, and worse psychosocial factors such as greater degree of family accommodation,⁵⁷ all of which are in turn known to predict poor outcomes in other studies.⁶⁰

Greater contamination and washing symptoms were also found to predict poorer response in the current study. Reasons for this is unclear as contamination/washing is known to be the “prototypical” OCD symptom and is the most common principal symptom in our sample. One possible explanation for this could be that in many patients with contamination OCD, due to the change in milieu, skills obtained in inpatient setting may not be generalizable to the patient’s own surroundings.⁶¹ Washers in our study were also more severely ill and hence possibly gravitated toward inpatient care. This group of patients also reported severe avoidance and more proxy compulsions, the kind of symptoms that lead to more family accommodation and treatment resistance.⁶²

The other studies found that living alone, comorbid alcohol misuse, and higher hoarding symptoms were found to predict poor response.⁴ As these factors were hardly represented in our sample, their impact on the outcomes could not be analyzed.

Strengths and Limitations

The strengths of the study include its large representative sample, systematic assessments using standard measures, and a relatively long duration of follow-up. Some of the limitations include its naturalistic design with the use of retrospective data, attrition at follow-up, uncontrolled concomitant pharmacological treatment and lack of measures of quality of life, and functioning. Even though the treatment program included participation of family members, there were no objective measures of family accommodation and functioning done. We assessed insight using a single item (YBOCS 11); a comprehensive assessment of insight using instruments such as the Brown Assessment of Beliefs Scale⁶³ could have been more useful.

Although it is reasonable to assume that intensive CBT played a major role since they were already resistant to SRIs and could not practice CBT on outpatient basis, one cannot attribute the outcome to only CBT because all patients received concurrent drug treatment as per the existing standards. One could even argue that the study demonstrates the efficacy of a multimodal intensive program that is a combination of rigorous CBT and optimal pharmacotherapy. Attrition rate was high, but less than in other studies,⁶ and there were no significant differences in baseline clinical characteristics and time-of-discharge outcomes of those who dropped out and those who did not. Trained clinicians administered key outcome measures such as the YBOCS and the CGI, but no formal psychometric exercises were carried out.

Conclusion

Our study demonstrates that an intensive inpatient care that includes CBT and pharmacotherapy is an effective strategy in severely ill SRI-resistant patients, both in the short- and long-term. Even those who previously failed outpatient CBT seemed to benefit from inpatient care. Better response was found to be predicted by a shorter duration of illness, better insight, and less of contamination/washing symptoms. Intensive inpatient/residential treatment should be considered in resistant patients who do not improve with treatment on outpatient basis and also be a part of standard treatment guidelines. The findings of this study may have to be replicated in controlled studies.

Supplemental Material

Supplement_material - Long-term Outcomes of Intensive Inpatient Care for Severe, Resistant Obsessive-Compulsive Disorder: Résultats à long terme de soins intensifs à des patients hospitalisés pour un trouble obsessionnel-compulsif grave et résistant

Supplement_material for Long-term Outcomes of Intensive Inpatient Care for Severe, Resistant Obsessive-Compulsive Disorder: Résultats à long terme de soins intensifs à des patients hospitalisés pour un trouble obsessionnel-compulsif grave et résistant by Srinivas Balachander, Aakash Bajaj, Nandita Hazari, Ajay Kumar, Nitin Anand, M. Manjula, Paulomi M. Sudhir, Anish V. Cherian, Janardhanan C. Narayanaswamy, T. S. Jaisoorya, Suresh Bada Math, Thennarasu Kandavel, Shyam Sundar Arumugham and Y. C. Janardhan Reddy in The Canadian Journal of Psychiatry

Footnotes

Acknowledgments

The authors thank all the staff at the Medical Records Department at the institute for facilitating the data collection for this study.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Srinivas Balachander

Supplemental Material

The supplemental material for this article is available online.

References

Fontenelle

Mendlowicz

Versiani

. The descriptive epidemiology of obsessive-compulsive disorder. Prog Neuropsychopharmacol Biol Psychiatry. 2006;30(3):327–337.

Ruscio

Stein

Chiu

Kessler

. The epidemiology of obsessive-compulsive disorder in the national comorbidity survey replication. Mol Psychiatry. 2010;15(1):53–63.

Hollander

Doernberg

Shavitt

, et al. The cost and impact of compulsivity: a research perspective. Eur Neuropsychopharmacol 2016;26(5):800–809.

Veale

Naismith

Miles

Gledhill

Stewart

Hodsoll

. Outcomes for residential or inpatient intensive treatment of obsessive–compulsive disorder: a systematic review and meta-analysis. J Obsessive Compulsive Related Disord. 2016:838–849.

Stewart

Stack

Farrell

Pauls

Jenike

. Effectiveness of intensive residential treatment (IRT) for severe, refractory obsessive-compulsive disorder. J Psychiatr Res. 2005;39(6):603–609.

Veale

Naismith

Miles

, et al. Outcome of intensive cognitive behaviour therapy in a residential setting for people with severe obsessive compulsive disorder: a large open case series. Behav Cognit Psychother. 2016;44(3):331–346.

Skapinakis

Caldwell

Hollingworth

, et al. Pharmacological and psychotherapeutic interventions for management of obsessive-compulsive disorder in adults: a systematic review and network meta-analysis. Lancet Psych. 2016;3(8):730–739.

McKay

Sookman

Neziroglu

, et al. Efficacy of cognitive-behavioral therapy for obsessive-compulsive disorder. Psychiatry Res. 2015;225(3):236–246.

Simpson

Foa

Liebowitz

, et al. Cognitive-behavioral therapy vs risperidone for augmenting serotonin reuptake inhibitors in obsessive-compulsive disorder: a randomized clinical trial. JAMA Psychiatry. 2013;70(11):1190–1199.

10.

Anand

Sudhir

Math

Thennarasu

Janardhan Reddy

. Cognitive behavior therapy in medication non-responders with obsessive-compulsive disorder: a prospective 1-year follow-up study. J Anxiety Disord. 2011;25(7):939–945.

11.

Simpson

Maher

Wang

Bao

Foa

Franklin

. Patient adherence predicts outcome from cognitive behavioral therapy in obsessive-compulsive disorder. J Consult Clin Psychol. 2011;79(2):247–252.

12.

Wheaton

Galfalvy

Steinman

Wall

Foa

Simpson

. Patient adherence and treatment outcome with exposure and response prevention for OCD: Which components of adherence matter and who becomes well? Behav Res Ther. 2016;85:6–12.

13.

Abramowitz

. The psychological treatment of obsessive-compulsive disorder. Can J Psychiatry. 2006;51(7):407–416.

14.

Ong

Clyde

Bluett

Levin

Twohig

. Dropout rates in exposure with response prevention for obsessive-compulsive disorder: what do the data really say? J Anxiety Disord. 2016:408–417.

15.

Abramowitz

Foa

Franklin

. Exposure and ritual prevention for obsessive-compulsive disorder: effects of intensive versus twice-weekly sessions. J Consult Clin Psychol. 2003;71(2):394–398.

16.

Storch

Geffken

Merlo

, et al. Family-based cognitive-behavioral therapy for pediatric obsessive-compulsive disorder: comparison of intensive and weekly approaches. J Am Acad Child Adolesc Psychiatry. 2007;46(4):469–478.

17.

Storch

Merlo

Lehmkuhl

, et al. Cognitive-behavioral therapy for obsessive-compulsive disorder: a non-randomized comparison of intensive and weekly approaches. J Anxiety Disord. 2008;22(7):1146–1158.

18.

Jónsson

Kristensen

Arendt

. Intensive cognitive behavioural therapy for obsessive-compulsive disorder: a systematic review and meta-analysis. J Obsessive-Compuls Related Disord. 2015:6:83–96.

19.

McGuire

Martino

Phares

Selles

Storch

. A meta-analysis of family accommodation and OCD symptom severity. Clin Psychol Rev. 2016;45(1):34–44.

20.

Lebowitz

Panza

Bloch

. Family accommodation in obsessive-compulsive and anxiety disorders: a five-year update. Expert Rev Neurother. 2016;16(1):45–53.

21.

Grunes

Neziroglu

McKay

Family involvement in the behavioral treatment of obsessive-compulsive disorder: a preliminary investigation. Behavior Therapy. 2001;32(4):803–820.

22.

Thompson-Hollands

Edson

Tompson

Comer

. Family involvement in the psychological treatment of obsessive-compulsive disorder: a meta-analysis. J Fam Psychol. 2014;28(3):287–298.

23.

Cherian

Pandian

Bada Math

Kandavel

Janardhan Reddy

. Family accommodation of obsessional symptoms and naturalistic outcome of obsessive-compulsive disorder. Psychiatry Res. 2014;215(2):372–378.

24.

Sheehan

Lecrubier

Sheehan

, et al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry. 1998;59(Suppl):2022–2033; quiz 34–57.

25.

Goodman

Price

Rasmussen

, et al. The Yale-Brown Obsessive Compulsive Scale. I. development, use, and reliability. Arch Gen Psychiatry. 1989;46(11):1006–1011.

26.

Goodman

Price

Rasmussen

, et al. The Yale-Brown Obsessive Compulsive Scale. II. validity. Arch Gen Psychiatry. 1989;46(11):1012–1016.

27.

Guy

. ECDEU assessment manual for psychopharmacology. In: ECDEU assessment manual for psychopharmacology. Rockville (MD): U.S. Dept. of Health, Education, and Welfare, Public Health Service, Alcohol, Drug Abuse, and Mental Health Administration, National Institute of Mental Health, Psychopharmacology Research Branch, Division of Extramural Research Programs; 1976.

28.

First

Gibbon

Spitzer

, et al. Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II). Washington (DC): American Psychiatric Press; 1997.

29.

Loranger

Sartorius

Andreoli

, et al. The International Personality Disorder Examination: the World Health Organization/alcohol, drug abuse, and mental health administration international pilot study of personality disorders. Arch Gen Psychiatry. 1994;51(3):215–224.

30.

Koran

Hanna

Hollander

, et al. Practice guideline for the treatment of patients with obsessive-compulsive disorder. Arlington (VA): American Psychiatric Association; 2007. [accessed August 21, 2019]. http://www.psych.org/psych_pract/treatg/pg/prac_guide.cfm

31.

Janardhan Reddy

Sundar

Narayanaswamy

Bada Math

. Clinical practice guidelines for obsessive-compulsive disorder. Indian J Psychiatry 2017;59(Suppl 1):S74–S90.

32.

Foa

Yadin

Lichner

. Exposure and response (ritual) prevention for obsessive compulsive disorder: therapist guide (treatments that work). 2nd ed. New York (NY): Oxford University Press; 2012.

33.

Mataix-Cols

De la Cruz

Nordsletten

Lenhard

Isomura

Simpson

. Towards an international expert consensus for defining treatment response, remission, recovery and relapse in obsessive-compulsive disorder. World Psychiatry. 2016;15(1):80–81.

34.

Hazari

Narayanaswamy

Arumugham

. Predictors of response to serotonin reuptake inhibitors in obsessive-compulsive disorder. Expert Rev Neurother. 2016;16(10):1175–1191.

35.

Muthén

. MPlus user’s guide. 6th ed. Los Angeles (CA): Muthén & Muthén; 2011.

36.

Jung

Wickrama

KAS

. An introduction to latent class growth analysis and growth mixture modeling. Soci Person Psychol Compass. 2007:11–19.

37.

Stewart

Yen

C-H

Stack

Jenike

. Outcome predictors for severe obsessive-compulsive patients in intensive residential treatment. J Psychiatr Res. 2006;40(6):511–519.

38.

Stewart

Stack

Tsilker

, et al. Long-term outcome following intensive residential treatment of obsessive-compulsive disorder. J Psychiatr Res 2009;43(13):1118–1123.

39.

Brennan

Lee

Elias

, et al. Intensive residential treatment for severe obsessive-compulsive disorder: characterizing treatment course and predictors of response. J Psychiatr Res. 2014:56:98–105.

40.

Krompinger

Monaghan

Gironda

, et al. Early response is predictive of outcome in intensive behavioral treatment for obsessive compulsive disorder. J Obsess Compuls Related Disord. 2017:15:57–1563.

41.

Falkenstein

Nota

Krompinger

, et al. Empirically-derived response trajectories of intensive residential treatment in obsessive-compulsive disorder: a growth mixture modeling approach. J Affect Disord. 2019;24(5):827–833.

42.

Jaisoorya

Reddy

YCJ

Srinath

. The relationship of obsessive-compulsive disorder to putative spectrum disorders: results from an Indian study. Compr Psychiatry. 2003;44(4):317–323.

43.

Viswanath

Narayanaswamy

Rajkumar

, et al. Impact of depressive and anxiety disorder comorbidity on the clinical expression of obsessive-compulsive disorder. Compr Psychiatry. 2012;53(6):775–782.

44.

Kashyap

Kumar

Kandavel

Reddy

. Neuropsychological correlates of insight in obsessive-compulsive disorder. Acta Psychiatr Scand. 2012;126(2):106–114.

45.

Grøtte

Hansen

Haseth

Patrick

Ismail

Stian

. Three-week inpatient treatment of obsessive-compulsive disorder: a 6-month follow-up study. Front Psychol. 2018;9:620. doi: 10.3389/fpsyg.2018.00620.

46.

Pallanti

Hollander

Bienstock

, et al. Treatment non-response in OCD: methodological issues and operational definitions. Int J Neuropsychopharmacol. 2002;5(2):181–191.

47.

Katzman

Bleau

Blier

, et al. Canadian clinical practice guidelines for the management of anxiety, posttraumatic stress and obsessive-compulsive disorders. BMC Psychiatry. 2014;14(Suppl 1):S1.

48.

Baldwin

Anderson

Nutt

, et al. Evidence-based pharmacological treatment of anxiety disorders, post-traumatic stress disorder and obsessive-compulsive disorder: a revision of the 2005 guidelines from the British association for psychopharmacology. J Psychopharmacol (Oxford). 2014;28(5):403–439.

49.

Bandelow

Zohar

Hollander

, et al. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for the pharmacological treatment of anxiety, obsessive-compulsive and post-traumatic stress disorders—first revision. World J Biol Psychiatry 2008;9(4):248–312.

50.

National Institute for Health and Clinical Excellence. Obsessive-compulsive disorder: core interventions in the treatment of obsessive-compulsive disorder and body dysmorphic disorder (CG 31). London, UK: NICE; 2005.

51.

Eisen

Rasmussen

Phillips

, et al. Insight and treatment outcome in obsessive-compulsive disorder. Compr Psychiatry 2001;42(6):494–497.

52.

Ravi Kishore

Samar

Janardhan Reddy

Chandrasekhar

Thennarasu

. Clinical characteristics and treatment response in poor and good insight obsessive-compulsive disorder. Eur Psychiatry. 2004;19(4):202–208.

53.

Visser

Megen

Oppen

, et al. The impact of poor insight on the course of obsessive-compulsive disorder in patients receiving naturalistic treatment. J Obsess Comp Related Disord. 2017:1342–1348.

54.

Catapano

Perris

Fabrazzo

, et al. Obsessive-compulsive disorder with poor insight: a three-year prospective study. Prog Neuropsychopharmacol Biol Psychiatry. 2010;34(2):323–330.

55.

Foa

Abramowitz

Franklin

Kozak

. Feared consequences, fixity of belief, and treatment outcome in patients with obsessive-compulsive disorder. Behav Therap. 1999;30(4):717–724.

56.

Santana

Fontenelle

Yücel

Leonardo

. Rates and correlates of nonadherence to treatment in obsessive-compulsive disorder. J Psychiatr Pract. 2013;19(1):42–53.

57.

Fineberg

Dell’Osso

Albert

, et al. Early intervention for obsessive compulsive disorder: an expert consensus statement. Eur Neuropsychopharmacol. 2019;29(4):549–565.doi: 10.1016/j.euroneuro.2019.02.002.

58.

Albert

Barbaro

Bramante

Rosso

De Ronchi

Maina

. Duration of untreated illness and response to SRI treatment in obsessive-compulsive disorder. Eur Psychiatry. 2019;58(1):19–26.

59.

Dell’Osso

Buoli

Hollander

Altamura

. Duration of untreated illness as a predictor of treatment response and remission in obsessive-compulsive disorder. World J Biol Psychiatry. 2010;11(1):59–65.

60.

Sharma

Thennarasu

Reddy

YCJ

. Long-term outcome of obsessive-compulsive disorder in adults: a meta-analysis. J Clin Psychiatry. 2014;75(9):1019–1027.

61.

Craske

Kircanski

Zelikowsky

Ayson

Najwa

Aaron

. Optimizing inhibitory learning during exposure therapy. Behav Res Ther. 2008;46(1):5–27.

62.

Albert

Bogetto

Maina

Paola

Cinthia

David Mataix

. Family accommodation in obsessive-compulsive disorder: relation to symptom dimensions, clinical and family characteristics. Psychiatry Res. 2010;179(2):204–211.

63.

Eisen

Phillips

Baer

Beer

Atala

Rasmussen

. The Brown Assessment of Beliefs Scale: reliability and validity. Am J Psychiatry. 1998;155(1):102–108.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.13 MB