Obsessive belief and emotional appraisal correlates of symptom dimensions and impairment in obsessive–compulsive disorder

Background: Obsessive–compulsive disorder (OCD) is a heterogeneous and debilitating illness. Symptom dimensions of OCD lend homogeneous avenues for research. Variations in one's appraisal of thoughts and emotions can influence symptom dimensions and impairment. However, little is known about the combined influence of these appraisals in OCD. A clear understanding of these relationships has putative treatment implications.
Aim: The aim of the study is to examine the associations among obsessive beliefs, emotional appraisals, and OCD symptom dimensions in adults.
Materials and Methods: We examined 50 drug-naïve/drug-free adults with active OCD. Symptom dimensions and impairment were assessed using the Dimensional Yale–Brown Obsessive–Compulsive Severity Scale. Obsessive beliefs and emotional appraisals were studied using the Obsessive Beliefs Questionnaire-44 and Perception of Threat from Emotion Questionnaire.
Results: Tobit regression analysis showed the differential association of obsessive beliefs and symptom dimensions – perfectionism/certainty associated with contamination and responsibility/threat estimation associated with aggressive obsessions. Impairment was associated with dimensional symptom severities and with the perception of threat from anger. This association remained even after controlling for depression severity and obsessive beliefs.
Conclusions: OCD symptom dimensions are heterogeneous in underlying obsessive beliefs. Emotional appraisals contribute significantly to impairment alongside symptom severity. Emotion-focused interventions must be included in the psychotherapeutic interventions for OCD.

Keywords: Emotional appraisal, impairment, obsessive beliefs, obsessive–compulsive disorder, symptom dimensions

How to cite this article:
Sinha R, Mahour P, Sharma E, Mehta UM, Agarwal M. Obsessive belief and emotional appraisal correlates of symptom dimensions and impairment in obsessive–compulsive disorder. Indian J Psychiatry 2021;63:348-54
How to cite this URL:
Sinha R, Mahour P, Sharma E, Mehta UM, Agarwal M. Obsessive belief and emotional appraisal correlates of symptom dimensions and impairment in obsessive–compulsive disorder. Indian J Psychiatry [serial online] 2021 [cited 2021 Aug 7];63:348-54. Available from: https://www.indianjpsychiatry.org/text.asp?2021/63/4/348/323371

Introduction

Obsessive–compulsive disorder (OCD), with a lifetime prevalence of 0.3%–3%,[1] is a severely debilitating illness.[2] OCD symptoms aggregate in a dimensional structure, which has clinical, neurobiological, and therapeutic implications.[3] Most consistently replicated dimensions include contamination obsessions with washing/cleaning compulsions (contamination dimension); obsessions of causing harm or making mistakes with checking compulsions (aggressive dimension); obsessions about order and symmetry and ordering/arranging compulsions (symmetry dimension); and obsessional thoughts related to sex, religion, and violence with mental compulsive rituals and neutralizing behaviors (forbidden thoughts dimension).[4] Age at onset, comorbidities, and insight differ across symptom dimensions; for instance, early-onset OCD is associated with aggressive and symmetry dimensions, comorbidity with neurodevelopmental disorders, and poor insight.[5] Symptom dimensions differ in underlying cognitive biases (e.g, aggressive obsessions are accompanied by thought-action fusion bias)[6] and respond differently to cognitive-behavioral interventions such as exposure/response prevention;[7] contamination responds better, while forbidden thoughts are prone to treatment resistance.[8]

A putative translational utility of the dimensional structure of OCD is in designing personalized interventions. Cognitive models of OCD emphasize the role of dysfunctional cognitive appraisals, in the form of obsessive beliefs, in the development and maintenance of obsessions and compulsions.[9] Factor analytical studies have identified several obsessive belief patterns – responsibility, over-importance of thoughts, thought-action fusion-likelihood, thought-action fusion-morality, importance of thought control, overestimation of threat, and intolerance of uncertainty and perfectionism.[10] Of these, responsibility/threat estimation (RT), perfectionism/certainty (PC), and importance/control of thoughts (ICT) are identified by the Obsessive Beliefs Questionnaire (OBQ)[11] and have been widely studied in OCD. They are related to the overall severity of illness[12] and are differentially associated with individual symptom dimensions.[13] They also play a role in the treatment effects of cognitive-behavioral interventions; a reduction in dysfunctional belief scores accompanies clinical improvement.[14] Even though obsessive beliefs independently influence the illness, they could have a hierarchical sequential organization (e.g, RT influencing ICT).[15] It follows, therefore, that obsessive beliefs could shape the onset, course, severity, and treatment response in OCD.

Besides obsessive beliefs, emotional appraisals have been associated with OCD.[16] Emotional appraisal refers to the interpretation and response to emotion-provoking stimuli. The experience and “unacceptability” of certain emotional states exacerbate distress from OCD symptoms.[17] Compared to siblings and healthy controls, OCD patients show greater distress and amygdala activation in fear-provoking and visual OCD stimuli.[18] OCD patients perform poorer in emotion recognition, showing a negative bias in their performance compared to healthy controls.[19] Deficits in basic emotional processing contribute to illness severity of OCD, and indicate the need for emotion-focused treatment approaches.[20] Few studies have found a dimensional specificity in emotional appraisal difficulties in patients with OCD, e.g, symmetry was associated with “not just right feelings” in one study.[16]

To the best of our knowledge, there are no studies that have systematically examined the combined influence of obsessive beliefs and emotional appraisals on symptom dimensions and impairment in OCD. A deeper understanding of these associations, i.e, among symptom severity, impairment, cognition and emotional appraisals, could be useful in conceptualizing and designing targeted psychotherapeutic interventions.[21]

The primary aim of this study was to examine the associations between obsessive beliefs, emotional appraisals, and dimensional symptom severity in OCD. We hypothesized that obsessive beliefs and emotional appraisals would differentially associate with OCD dimensional symptom severity. The second aim was to examine the association of obsessive beliefs, emotional appraisals, and dimensional severity with impairment. We hypothesized that obsessive belief and emotional appraisal scores, in addition to symptom severity, would contribute to illness-related impairment.

Materials and Methods

Study design and sample

This was a cross-sectional study. The sample for the study was drawn from patients attending the adult psychiatry outpatient services in the department of psychiatry, at a tertiary care hospital in North India. Consecutive patients clinically diagnosed with OCD were screened for inclusion into the study as per the following criteria: (a) age between 18 and 60 years, (b) meeting ICD-10 DCR[22] diagnostic criteria for OCD, (c) drug-naïve or drug-free (off-treatment for at least one month), (a reduction in OCD symptoms and related depressive/anxiety symptoms, with treatment, may alter emotional and cognitive appraisals), (d) symptoms continuously present at least for the last 1 month, and (e) willing to give written informed consent. Patients were excluded if they (a) fulfilled current or lifetime ICD-10 DCR criteria for comorbid psychiatric disorders, other than depression, (b) had comorbid chronic medical or neurological disorders, or (c) had earlier undergone psychotherapeutic treatment. Psychotherapeutic interventions, especially cognitive-behavioral methods, impact obsessive beliefs and emotional appraisals of an obsessive–compulsive (OC) phenomenon and would have influenced study-related assessments. Given that comorbidities (mood disorders, anxiety disorders, and OC spectrum disorders) may themselves influence obsessive beliefs and emotional appraisals, we excluded them. While it was possible to do this for other disorders, we could not do this for depression. Both literature[23] and experience at our center suggest that comorbidity between OCD and depression is very high. The decision to not exclude patients with depression was therefore from a feasibility perspective. The institutional ethics committee approved the study protocol.

Assessments

(a) Mini International Neuropsychiatric Interview version 5.0 (MINI)[24] was used for diagnostic corroboration and evaluation of comorbidities. MINI is a semi-structured diagnostic interview, with high validity and reliability estimates for DSM-IV and ICD-10 psychiatric disorders. (b) Dimensional Yale–Brown Obsessive–Compulsive Severity Scale (DYBOCS)[25] assesses symptom severity across seven consistent and temporally stable dimensions of OCD – contamination/cleaning, symmetry/ordering/arranging/counting, sexual/religious, aggressive, somatic, hoarding/collecting, and miscellaneous. It computes dimensional ratings on three ordinal scales that measure symptom frequency (0–5), amount of distress they cause (0–5), and the degree to which they interfere with functioning (0–5). Global symptom severity (0–15) is measured by a combination of these three ordinal scales for each symptom dimension. Impairment (0 – “none” to 15 – “severe”) is measured separately. DYBOCS impairment scores do not include the symptom severity scores and are a reflection of the detrimental functional consequences of the illness overall. The total global score (0–30) is obtained by combining the global symptom severity with the impairment scores. Dimensional severity scores on the DYBOCS can be used to study clinical and phenomenological correlates of individual symptom dimensions in greater detail, beyond just the presence or absence of a given symptom. (c) OBQ-44[11] assesses dysfunctional beliefs – RT, PC, and ICT. (d) Perception of Threat from Emotions Questionnaire (PTEQ)[17] measures appraisals about happiness, sadness, anger, fear/anxiety, disgust, guilt, sexual desire/lust, and strong emotions in general. Higher scores indicate a higher “perception of threat” from emotions. Perception of threat from an emotion could modify the expression, experience, and resolution of an emotion. When emotional experiences accompany other psychopathological phenomena, perception of threat from emotions could also affect beliefs and cognitions and impact illness severity and functional impairment. PTEQ scores were used in the present study as measures of emotional appraisal. The OBQ and PTEQ were translated into Hindi by the primary investigator (RS) and back-translated by a colleague, not involved in this research, following the standard WHO translation-back translation procedures. Process of translation procedures (available at https://www.who.int/ substance_abuse/research_tools/translation/en/; accessed on March 2015). Expert validation of the translations was completed by research supervisors (ES and PM), who ensured face validity and conceptual accuracy of the translations. Study assessments were done in Hindi, most commonly.

Data analysis

Clinical characteristics, DYBOCS dimensional severity, impairment, OBQ dimensions, and PTEQ dimensions were summarized using descriptive statistics. The dimensional severity data were not normally distributed since a patient typically has only a few symptom dimensions, resulting in an over-representation of “zero” scores for the dimensional severities. In such instances, parameter estimates obtained by conventional linear regression methods are biased. Tobit regression is a suitable censored regression model to study such continuous dependent variables with insufficient range due to the detection limit of a measuring scale that results in a large number of observations clustering at one end.[26] We implemented separate Tobit regression models with contamination/cleaning, aggressive, and symmetry/ordering/arranging/counting (symmetry) dimensional scores as dependent variables, as well as each of the OBQ and PTEQ dimension scores as independent variables to understand the relationship between obsessive beliefs, emotional appraisals, and symptom dimensions. Data on other dimensions (sexual, hoarding, and somatic) were insufficient (n < 10) to run such modeling analyses. Next, we ran multiple linear regression models with impairment scores as the dependent variable and the symptom dimensional scores, OBQ and PTEQ scores, and clinical characteristics (age at onset, sex, and duration of illness, depression scores) as independent variables. All the analyses were performed using the R environment for statistical computing,[27] package AER.[28]

Results

Clinical characteristics

Selection criteria were met by 50 of the 85 patients screened for the study. Reasons for exclusion in 35 patients were ongoing pharmacological treatment (n = 17), consent refusal (n = 6), comorbid psychiatric disorders (n = 7), and comorbid chronic medical illness (n = 5). Sociodemographic and clinical characteristics of the sample are presented in [Table 1]. The sample largely had adult onset of illness, with a mean duration of illness around 5 years. The mean DYBOCS global severity score was 22 (range 10–30). 78% of the sample (n = 39) fulfilled the criteria for depression on MINI; the mean Beck depression inventory second edition scores were 19.08 ± 5.62, indicating an overall mild depression severity.[29]

Data on symptom dimensions, obsessive beliefs, and perception of threat from emotions are presented in [Table 2]. Contamination/cleaning was the most common (58%) symptom dimension, followed by aggressive (28%) and symmetry, ordering, arranging, and counting (20%). The contamination/cleaning dimension also had the highest severity rating, followed by sexual and religious obsessions. Among the various PTEQ dimensions, the highest scores were for anxiety, followed by anger; i.e, perception of threat was highest for emotions of anxiety and anger.

Table 2: Obsessive-compulsive disorder symptom dimensions, obsessive beliefs, and emotion appraisals

Click here to view

Determinants of symptom dimensions and impairment

Tobit regression analyses [Table 3] with the DYBOCS dimensional severity scores as the dependent variable and OBQ/PTEQ dimensions as the predictors revealed that higher levels of RT, anger, and strong emotions were significantly associated with higher scores on the aggressive dimension; higher levels of PC and lower levels of ICT were significantly associated with higher scores on the contamination dimension; and a higher threat from strong emotions and lower threat scores on disgust were significantly associated with symmetry. Among these, the overall model was significant for the contamination dimension (P = 0.008), but not for the aggressive (P = 0.072) or symmetry dimensions (P = 0.228), although the aggressive model reached trend-level significance (P < 0.1). On multiple linear regression analysis [Table 4], impairment was significantly associated (P < 0.001) with high severity scores of aggressive, contamination, and sexual dimensions, as well as anger. Variance inflation factors for all the independent variables were <2 in each of the models described, indicating limited multicollinearity.[30]

Table 3: Tobit regression analysis for determinants of symptom dimensions

Click here to view

Table 4: Multiple linear regression analysis for determinants of impairment

Click here to view

Discussion

This study examined the associations among obsessive beliefs, emotional appraisals, and OCD symptom dimensions, in a consecutive sample of currently symptomatic, drug-naïve/drug-free adult OCD patients. We found that contamination/cleaning dimension was associated with obsessive beliefs about PC, and aggressive dimension had a significant association with RT. Emotional appraisal featured among the significant predictors of impairment.

Contamination/cleaning dimension

Contamination/cleaning was associated with higher scores on PC and lower scores on ICT. The overall model that was statistically significant predicted 39.1% of the variance in contamination/cleaning dimensional severity. Our findings are contrary to previous studies that found RT associated with contamination/cleaning.[31] Some patients with contamination/cleaning report that their obsessions about contamination are related to the fear of some illness/harm befalling them or their family members. Underlying obsessive beliefs of RT are pertinent in this regard. Other patients do not endorse such fearful consequences; instead, their contamination obsessions are related to just a feeling of being unclean or dirty and how that might feel. They desire perfection in and certainty of cleanliness. Culturally, Indian households lay great emphasis on the need for one to be clean during daily chores, especially cooking and praying. Cultural beliefs may not always be related to the fearful consequences of noncompliance; they can just be a reflection of rules imbibed over generations. Given the cultural sanction for a heightened sense of cleanliness, these thoughts may evoke a lesser need to be controlled, reflected in the inverse association with ICT.

Other symptom dimensions

We carried out regression analyses for aggressive and symmetry dimensions. The remaining dimensions – somatic, hoarding, sexual/religious, and miscellaneous – had very low sample sizes (n ≤ 10), limiting meaningful interpretation of the analysis. The overall models were nonsignificant for both aggressive and symmetry dimensions. However, certain obsessive beliefs and emotional appraisals showed statistically significant estimates. Obsessive belief about RT and perception of threat from strong emotions (positive association) and anger (negative association) had statistically significant estimates for the aggressive dimension. For the symmetry dimension, significant estimates were found for the perception of threat from disgust (negative association) and strong emotions (positive association). An association between the obsessive belief of RT and aggressive obsessions has been previously reported.[16] Aggressive obsessions are characterized by underlying fears about causing harm to someone/oneself. Salkovskis' model of OCD describes how the mere presence of thought may be equated with the execution of that thought, i.e, thought-action fusion.[32] Beliefs about RT, which imply that one has the power to bring about or prevent subjectively critical negative outcomes, are understandable in this background. Obsessions can evoke a variety of emotional responses in a person – guilt, sadness, fear, anger, etc. This could underlie our findings on the association of both aggressive and symmetry dimensions with a higher perception of threat from strong emotions.

Impairment

Impairment scores correlated with aggressive, contamination, sexual symptom dimensions, and perception of threat from anger. The association of impairment with emotional appraisal, in the absence of association with obsessive beliefs, is an important finding of our study. It suggests that patients will be more impaired from their illness if they are not able to handle the emotional accompaniments of OC symptoms. Obsessive beliefs perhaps do not directly influence illness-related impairment. They play a role in the symptom dimensions, the severity of which in turn influences impairment. The association between perception of threat from anger and impairment persisted even after controlling for depression scores, suggesting that emotional appraisals play an important role even in the absence of depression. This finding is in line with previous research that found anger to be a strong predictor of obsessionality.[17] Difficulties with anger experience and control in OCD may contribute to the general negative effect in these patients.[33] Patients with OCD can experience various challenges vis-à -vis anger – they report a higher experience of anger, with greater difficulty in controlling it, and a tendency to suppress it inwardly.[33] Cognitive-behavioral interventions for OCD focus largely on cognitive restructuring methods and habituation of anxiety. It has been seen that even after symptom severity decreases, functional recovery is not complete.[34] The persisting difficulties with emotional appraisal resulting in impairment may underlie this observation. Our findings suggest that a focus on emotional appraisals, especially handling anxiety, is imperative for functional recovery. Building emotional awareness and emotional regulation training may be combined with cognitive-behavioral approaches.

We started with a mention of the heterogeneous nature of OCD and how symptom dimensional structures provide homogeneous groups. Our study validates this conceptualization in finding the differential obsessive belief underpinnings of symptom dimensions and importance of emotional appraisal in functional impairment. Psychotherapeutic strategies in OCD are largely based on cognitive-behavioral interventions such as cognitive restructuring and exposure and response prevention[35] that are effective in the short- and long-term treatment response. However, despite effective psychotherapeutic and pharmacological interventions, almost half of the patients with OCD do not achieve remission over long-term follow-up.[36] Clinical response to cognitive-behavioral interventions can be affected by underlying symptom dimensions,[37] as well as emotional appraisals. The “third wave” of cognitive-behavioral therapy recognizes the key role of emotions in illness origin and maintenance.[38]

Our study has several strengths. We examined obsessive beliefs and emotional appraisals together in a sample with a primary diagnosis of OCD. This facilitates a comprehensive picture of the obsessive beliefs and emotional architecture of OCD symptom dimensions. All the participants in this study were either drug-naïve or currently drug-free, and they had never received any psychotherapeutic interventions. The findings are therefore applicable to untreated OCD and can inform initial treatment planning. We used the Tobit method for regression analysis, given the left-censored data on symptom dimensions. We are aware of the limitations in our work. The sample under-represents some of the OCD symptom dimensions – somatic, hoarding, and sexual/religious. We may have missed statistically significant associations with obsessive beliefs due to a type II error. Our study is cross-sectional. Over the long term, symptom migration is seen in OCD. It is possible that obsessive beliefs and emotional appraisals also change over time or with treatment. A longitudinal design would better clarify the stability of beliefs and emotional appraisals. We have not assessed insight objectively. Insight may have an important role to play in cognitive and emotional appraisals. Patients with comorbid depression were not excluded from the study. This could not be done due to the feasibility of recruitment given the high comorbidity of these two conditions. The veridicality of our findings is higher given the naturalistically high comorbidity between OCD and depression. We have considered the severity of depressive symptoms in our analysis for impairment. We did not assess OC personality traits that could be associated with obsessive beliefs. Having YBOCS severity would have enhanced compatibility with other studies; however, given our interest in symptom dimensions, we chose the DYBOCS.

Conclusions

Our study on actively symptomatic adult OCD patients found that obsessive beliefs are differentially associated with OCD symptom dimensions and that impairment is associated with emotional appraisal (perception of threat from emotions), to the exclusion of obsessive beliefs and depression. These findings reiterate the heterogeneous nature of OCD symptom dimensions and suggest the need for the inclusion of emotion-focused psychotherapeutic for functional recovery in OCD.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.Williams MT, Steever A. Cultural manifestations of obsessive-compulsive disorder. In: Lack CW, editor. Obsessive-Compulsive Disorder: Etiology, Phenomenology, and Treatment. Hampshire, UK: Onus Books; 2015.

2.Subramaniam M, Soh P, Ong C, Esmond Seow LS, Picco L, Vaingankar JA, et al. Patient-reported outcomes in obsessive-compulsive disorder. Dialogues Clin Neurosci 2014;16:239-54.

3.Leckman JF, Bloch MH, King RA. Symptom dimensions and subtypes of obsessive-compulsive disorder: A developmental perspective. Dialogues Clin Neurosci 2009;11:21-33.

4.Abramowitz JS, Deacon BJ, Olatunji BO, Wheaton MG, Berman NC, Losardo D, et al. Assessment of obsessive-compulsive symptom dimensions: Development and evaluation of the dimensional obsessive-compulsive scale. Psychol Assess 2010;22:180-98.

5.Prabhu L, Cherian AV, Viswanath B, Kandavel T, Bada Math S, Janardhan Reddy YC. Symptom dimensions in OCD and their association with clinical characteristics and comorbid disorders. J Obsessive Compuls Relat Disord 2013;2:14-21.

6.Shafran R, Thordarson DS, Rachman S. Thought-action fusion in obsessive compulsive disorder. J Anxiety Disord 1996;10:379-91.

7.McLean PD, Whittal ML, Thordarson DS, Taylor S, Söchting I, Koch WJ, et al. Cognitive versus behavior therapy in the group treatment of obsessive-compulsive disorder. J Consult Clin Psychol 2001;69:205-14.

8.Alonso P, Menchon JM, Pifarre J, Mataix-Cols D, Torres L, Salgado P, et al. Long-term follow-up and predictors of clinical outcome in obsessive-compulsive patients treated with serotonin reuptake inhibitors and behavioral therapy. J Clin Psychiatry 2001;62:535-40.

9.Taylor S, Abramowitz JS, McKay D. Cognitive-behavioral models of obsessive-compulsive disorder. In: Antony MM, Purdon C, Summerfeldt LJ, editors. Psychological Treatment of Obsessive-Compulsive Disorder: Fundamentals and Beyond. Washington: American Psychological Association; 2007. p. 9-29. Available from: http://content.apa.org/books/11543-001. [Last accessed on 2018 Oct 22].

10.Belloch A, Morillo C, Luciano JV, García-Soriano G, Cabedo E, Carrió C. Dysfunctional belief domains related to obsessive-compulsive disorder: A further examination of their dimensionality and specificity. Span J Psychol 2010;13:376-88.

11.Obsessive Compulsive Cognitions Working Group. Cognitive assessment of obsessive-compulsive disorder. Behav Res Ther 1997;35:667-81.

12.Taylor S, Abramowitz JS, McKay D. Are there interactions among dysfunctional beliefs in obsessive compulsive disorder? Cogn Behav Ther 2005;34:89-98.

13.Myers SG, Fisher PL, Wells A. Belief domains of the Obsessive Beliefs Questionnaire-44 (OBQ-44) and their specific relationship with obsessive-compulsive symptoms. J Anxiety Disord 2008;22:475-84.

14.Polman A, Bouman TK, van Geert PL, de Jong PJ, den Boer JA. Dysfunctional beliefs in the process of change of cognitive treatment in obsessive compulsive checkers. Clin Psychol Psychother 2011;18:256-73.

15.Taylor S, Coles M, Abramowitz J, Wu KD, Olatunji BO, Timpano K, et al. How are dysfunctional beliefs related to obsessive-compulsive symptoms? J Cogn Psychother 2010;24:165-76.

16.Smith AH, Wetterneck CT, Hart JM, Short MB, Björgvinsson T. Differences in obsessional beliefs and emotion appraisal in obsessive compulsive symptom presentation. J Obsessive Compuls Relat Disord 2012;1:54-61.

17.McCubbin RA, Sampson MJ. The relationship between obsessive-compulsive symptoms and appraisals of emotional states. J Anxiety Disord 2006;20:42-57.

18.Thorsen AL, de Wit SJ, de Vries FE, Cath DC, Veltman DJ, ven der Werf YD, et al. Emotion regulation in obsessive-compulsive disorder, unaffected siblings, and unrelated healthy control participants. Biol Psychiatry 2019;4:352-60.

19.Daros AR, Zakzanis KK, Rector NA. A quantitative analysis of facial emotion recognition in obsessive-compulsive disorder. Psychiatry Res 2014;215:514-21.

20.Valeriani G, Bersani FS, Liberati D, Polli E, Girolami MT, Zullo D, et al. Generalized and specific emotion impairments as potential markers of severity in obsessive-compulsive disorder: A preliminary study using Facial Action Coding System (FACS). Psychiatr Danub 2015;27:159-67.

21.Robinson LJ, Freeston MH. Emotion and internal experience in obsessive compulsive disorder: Reviewing the role of alexithymia, anxiety sensitivity and distress tolerance. Clin Psychol Rev 2014;34:256-71.

22.World Health Organization, editor. The ICD-10 Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines. Geneva: World Health Organization; 1992.

23.Overbeek T, Schruers K, Vermetten E, Griez E. Comorbidity of obsessive-compulsive disorder and depression: Prevalence, symptom severity, and treatment effect. J Clin Psychiatry 2002;63:1106-12.

24.Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, 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 20:22-33.

25.Rosario-Campos MC, Miguel EC, Quatrano S, Chacon P, Ferrao Y, Findley D, et al. The Dimensional Yale-Brown Obsessive-Compulsive Scale (DY-BOCS): An instrument for assessing obsessive-compulsive symptom dimensions. Mol Psychiatry 2006;11:495-504.

26.Tobin J. Estimation of relationships for limited dependent variables. Econometrica 1958;26:24-36.

27.R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2014. Available from: https://www.R-project.org.

28.Kleiber C, Zeileis A. Applied Econometrics with R. New York: Springer-Verlag; 2008. Available from: https://www.springer.com/gp/book/9780387773162. [Last accessed on 2020 Sep 10].

29.Beck A, Steer R, Brown G. Beck Depression Inventory Second Edition Manual. San Antonio, Texas, USA: San Antonio TX Psychol Corp; 1996.

30.Allen MP, editor. The problem of multicollinearity. In: Understanding Regression Analysis. Boston, MA: Springer US; 1997. p. 176-80. Available from: https://doi.org/10.1007/978-0-585-25657-3_37. [Last accessed on 2020 Jan].

31.Wheaton MG, Abramowitz JS, Berman NC, Riemann BC, Hale LR. The relationship between obsessive beliefs and symptom dimensions in obsessive-compulsive disorder. Behav Res Ther 2010;48:949-54.

32.Salkovskis PM, Forrester E, Richards C. Cognitive-behavioural approach to understanding obsessional thinking. Br J Psychiatry 1998;173:53-63.

33.Whiteside SP, Abramowitz JS. Obsessive-compulsive symptoms and the expression of anger. Cogn Ther Res 2004;28:259-68.

34.Norberg MM, Calamari JE, Cohen RJ, Riemann BC. Quality of life in obsessive-compulsive disorder: An evaluation of impairment and a preliminary analysis of the ameliorating effects of treatment. Depress Anxiety 2008;25:248-59.

35.Hawton K, editor. Cognitive Behaviour Therapy for Psychiatric Problems: A Practical Guide. Oxford, New York: Oxford University Press; 1989.

36.Sharma E, Thennarasu K, Reddy YC. Long-term outcome of obsessive-compulsive disorder in adults: A meta-analysis. J Clin Psychiatry 2014;75:1019-27.

37.Williams MT, Farris SG, Turkheimer EN, Franklin ME, Simpson HB, Liebowitz M, et al. The impact of symptom dimensions on outcome for exposure and ritual prevention therapy in obsessive-compulsive disorder. J Anxiety Disord 2014;28:553-8.