February 2013  << Back  

  Can't open the PDF? Click here for help.



Correlates of insight into different symptom dimensions in obsessive-compulsive disorder

Júlia M. Fontenelle, PhD

Anxiety and Depression Research Program, Institute of Psychiatry, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil

Ben J. Harrison, PhD

Melbourne Neuropsychiatry Centre, University of Melbourne, Melbourne, Victoria, Australia

Lívia  Santana, PhD

Anxiety and Depression Research Program, Institute of Psychiatry, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil

Maria Conceição  do Rosário, MD, PhD

Department of Psychiatry, Federal University of São Paulo, São Paulo, Brazil

Marcio  Versiani, MD, PhD

Anxiety and Depression Research Program, Institute of Psychiatry, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil

Leonardo F. Fontenelle, MD, PhD

Anxiety and Depression Research Program, Institute of Psychiatry, Federal University of Rio de Janeiro, Department of Psychiatry and Mental Health, Institute of Community Health, Fluminense Federal University, D’Or Institute for Research and Education, Rio de Janeiro, Brazil

BACKGROUND: In this study, we evaluated insight into different obsessive-compulsive disorder (OCD) symptom dimensions and their impact on clinical and sociodemographic features of patients with OCD.

METHODS: Sixty OCD patients were assessed with the Brown Assessment of Beliefs Scale (BABS), the Dimensional Yale-Brown Obsessive-Compulsive Scale–Short Version, the Beck Depression Inventory, and the Sheehan Disability Scale. Two methods of using BABS were employed: 1) a traditional approach, which considers a composite of the insight into existing OCD symptoms, and 2) an alternative approach, which includes assessments of insight into each OCD symptom dimension separately.

RESULTS: Composite BABS scores correlated with global severity of OCD and depressive symptoms, and degree of interference on social life/leisure activities and family life/home responsibilities. Dimension-specific correlations between severity of symptoms and insight ranged from very high (P = .87, for hoarding) to moderate (P = .61, for miscellaneous symptoms). Greater severity of depression and concomitant generalized anxiety disorder were independently associated with lower levels of insight into aggressive/checking symptoms. While earlier-onset OCD was associated with lower insight into sexual/religious and symmetry symptoms, later-onset OCD displayed lower insight into hoarding.

CONCLUSIONS: Assessing insight into dimension-specific OCD symptoms may challenge the existence of clear-cut OCD with fair or poor insight.

KEYWORDS: obsessive-compulsive disorder, obsessions, hoarding, awareness, psychopathology



Insight is the human ability to critically evaluate one’s own mental state with the “mind’s eye.”1 Classically, patients with obsessive-compulsive disorder (OCD) have been described as possessing a good level of awareness in relation to their own obsessions or compulsions.2 However, there is a growing recognition that a certain percentage of patients with OCD may be partially or totally unable to recognize the irrationality of their symptoms, which suggests that they have poor or no insight.3 In such cases, the symptoms have been best classified as overvalued ideas or delusions, respectively.4-6

Different methods to evaluate insight with regard to OCD have been developed. Perhaps one of the most often used instruments is the Brown Assessment of Beliefs Scale (BABS), a 7-item, semistructured, clinician-administered scale with specific probes and anchors designed to assess insight into a dominant belief that has preoccupied the patient during the past week.7 The dimensions covered by the scale are conviction, perception of others’ views of beliefs, explanation of differing views, fixity of ideas, attempt to disprove beliefs, insight, and ideas of reference.7 For each item, there are specific probes with 5 anchors ranging from 0 (nondelusional/least pathologic) to 4 (delusional/most pathologic). The seventh item is not included in the total score. BABS ratings represent an average score for the past week.7

When examining insight in OCD, one must consider that there are various possible aspects of the illness that patients may or may not be able to recognize (“objects of insight”), including particular obsessive beliefs or compulsive behaviors, the disorder as a whole, or even its consequences.2,8 On the basis of considerable recent work, studies of insight in OCD should at least consider its potential association with specific symptom dimensions (eg, symmetry/ordering, contamination/cleaning, obsessions/checking, and hoarding),9 given that these dimensions have been associated with partially distinct patterns of comorbidity, genetic transmission, neural substrates, and treatment response.9

In early studies that have reported on the link between insight and different OCD symptom dimensions, poor insight has been emphasized as particularly common in relation to hoarding.10-12 This finding has been extended in a number of controlled studies comparing OCD hoarders vs nonhoarders13-19 and frequently cited as evidence supporting the proposal that hoarding may be a specific subtype of OCD or even an independent nosologic entity.20

However, contrary to the above evidence, it is possible that hoarding, or any specific obsessive-compulsive dimension, may not be easily linked with measures of overall insight. For instance, in OCD samples, 1) hoarding frequently is associated with other OCD symptoms (in a study of 1,001 OCD patients, approximately one-half of the sample [52.7%] presented hoarding symptoms, but only 4 patients solely presented this dimension [0.4%])21; 2) hoarding is generally the mildest OCD symptom among OCD samples21; and 3) hoarding represents a clinically significant problem in <5% of patients with OCD.22,23 Because the BABS recommends that raters assess insight into the most clinically significant OCD symptoms, insight into OCD symptoms in many OCD hoarders may reflect insight into other nonhoarding OCD symptoms. In fact, other symptoms have been associated with poor insight, including aggressive,24 religious,24 somatic,13-25 and miscellaneous obsessions,19 and checking and washing compulsions.26

In the present study, we performed a systematic assessment of insight in patients with OCD under treatment. Based on existing studies using a single composite level of insight with regard to OCD symptoms (“traditional approach”), we predicted that lower insight into OCD symptoms would be associated with earlier-onset OCD,19 greater severity of OCD,27-29 increased prevalence of hoarding,13 comorbid major depressive disorder (MDD),19-28 greater severity of depressive symptoms,13,28-30 lower intelligence quotient (IQ),31 and increased levels of disability.32

Conversely, considering that an overall or composite measure of insight may not be reliably linked to hoarding or to any other specific obsessive-compulsive dimension, we measured insight with regard to dimension-specific OCD symptoms and sought to describe their correlates (“dimension-specific approach”). Specifically, we hypothesized that different sociodemographic factors, age at onset, OCD symptoms, and comorbid disorders would influence the relationship between insight and different OCD symptom dimensions.

Based on previous research, we predicted that poor insight into OCD symptom dimensions would be influenced by sex (men would show lower insight into sexual/religious, symmetry, and hoarding, whereas women would show lower insight into washing),32 age at OCD onset (early-onset cases would show lower insight into aggressive, sexual/religious, symmetry, and hoarding),34 symptom severity (severe symptoms within a dimension would be associated with lower insight regarding the same dimension), and comorbid conditions.35-40

Regarding the latter, we predicted that insight into aggressive/checking and sexual/religious dimensions would be associated with occurrence of depressive and anxiety disorders,35,36 insight into symmetry would be related to comorbid eating and bipolar disorders,36,37 and insight into hoarding would be associated with comorbid bipolar,38,39 social anxiety,40 and posttraumatic stress disorders.38 Finally, we anticipated that greater disability and lower IQ would be related to lower insight in all dimensions.31,32


In a cross-sectional study, 60 patients with OCD under treatment in a specialized university OCD clinic were consecutively recruited according to the following criteria: 1) primary diagnosis of DSM-IV OCD, both in terms of severity and age at onset; 2) age 18 to 65; 3) ability to read and fill out forms; and 4) no neurologic, endocrine, or systemic medical illness that could interfere with patterns of response. The research protocol complied with the Declaration of Helsinki and was approved by the local institutional review board. Informed consent was obtained for participants after the nature of the procedures was explained. Participants were evaluated with the following instruments: the Mini International Neuropsychiatric Interview 6.0,41 the Dimensional Yale-Brown Obsessive-Compulsive Scale–Short Version (DY-BOCS–SV),35 the BABS,7 the Beck Depression Scale,42 the Sheehan Disability Scale,43 and the Wechsler Abbreviated Scale of Intelligence.44

The BABS was used to rate insight whenever an OCD symptom was considered present and of sufficient severity (ie, score ≥3) according to the DY-BOCS-SV (range, 0=never to 6=all the time, very upsetting). To evaluate insight with regard to OCD symptoms and its relationship with other relevant clinical and sociodemographic variables, 2 methods of using BABS were employed: 1) a traditional approach, recommended by developers of the scale, which rates insight into OCD symptoms as a composite of the general average insight7 and 2) a dimension-specific strategy, which assesses insight into all clinically significant OCD symptoms separately. In both cases, the main OCD belief from each DY-BOCS-SV dimension was identified. Although some previous studies using the BABS in OCD samples (eg, Storch et al32) have based their evaluation of insight on the basis of 1 or 2 specific obsession fears identified by the patient and the clinician, this strategy is not specifically recommended in the BABS administration guidelines.

Statistical analyses

Continuous variables (eg, BABS scores) exhibited by patients with different categorical features (eg, sex, religion, range of age at onset, comorbid condition) were compared using Mann-Whitney or Kruskal-Wallis tests followed by Bonferroni correction. Conversely, Spearman correlation coefficient analyses were performed to assess relationships between continuous variables (eg, BABS and DY-BOCS scores) along with partial correlations whenever deemed adequate.


Traditional approach to BABS

Composite insight did not differ according to sex (z = 0.0; P = 1.0) and religion (z = –.82; P = .40) and did not correlate with age (ρ = 0.17; P = .19) and educational levels (ρ = –.12; P = .36). However, patients whose onset of OCD occurred before age 10 (z = –2.47; P = .01) or after age 18 (z = –1.91; P = .05) had lower levels of composite insight than patients whose OCD began between age 10 and 18. Further, composite insight correlated positively with global severity of OCD symptoms (ρ = 0.46; P < .001), severity of depression (ρ = 0.38; P = .003), and degree of interference on social life/leisure activities (ρ = .39; P = .002) and on family life/home responsibilities (ρ = 0.34; P = .08). Finally, although the presence of MDD was associated with a trend toward higher BABS scores (z = –1.86; P = .06), the concomitant occurrence of any mood disorder increased BABS scores to a significant level (z = –2.37; P = .01).

Dimension-specific approach to BABS

The alternative approach showed that insight into different OCD symptoms did not differ according to sex and religion (data not shown), but that insight into hoarding correlated negatively with current age (ρ = 0.27; P = .03). This approach also identified very high (ρ = 0.87; P < .001) to moderate (ρ = 0.61; P < .001) symptom-specific correlations between severity of symptom dimensions and insight (TABLE). Although composite BABS scores did correlate with global severity of OCD symptoms (ρ = 0.46; P < .001), maximum scores on BABS did not correlate with global severity of OCD symptoms (ρ = 0.17; P = .17).

The presence of concurrent generalized anxiety disorder (GAD) was associated with lower levels of insight into aggressive/checking symptoms (z = –2.61; P = .009) but not with greater severity of aggressive/checking symptoms (z = –1.63; P = .10). We also found correlations between insight into symptoms belonging to aggressive/checking symptoms and 1) the severity of depression (ρ = 0.38; P = .002); 2) the degree of interference on home/school (ρ = 0.36; P = .046) and social life/leisure activities (ρ = .33; P = .008); 3) the performance on block design (ρ = –.30; P = .02); and 4) IQ level (ρ = –.31; P = .01). However, after covarying for the severity of aggressive/checking symptoms, only the correlation between insight into aggressive/checking symptoms and severity of depression remained significant (ρ = 0.33; P = .01).

Insight into symptoms from the symmetry dimensions correlated with the degree of interference on family life/home responsibilities (ρ = 0.29; P = .02), but this correlation did not remain statistically significant after the severity of symmetry dimensions was partialled out. Similarly, insight into symptoms from the contamination/washing dimensions correlated with educational levels (ρ = –0.26; P = .04) and with the degree of interference on social life/leisure activities (ρ = 0.28; P = .03). Nevertheless, this correlation did not remain statistically significant after the severity of contamination/washing dimensions was partialled out. Further, as reported above, insight into symptoms from the hoarding dimensions correlated with current age (ρ = 0.27; P = .03) but, similarly, this correlation did not remain statistically significant after the severity of hoarding dimension was taken into account.

Patients whose OCD symptoms began before age 10, between age 10 and 18, and after age 18 differed in terms of their insight into sexual/religious (χ2 = 8.4; df = 2; P = .01), symmetry (χ2 = 7.01; df = 2; P = .03) and hoarding symptoms (χ2 = 8.05; df = 2; P = .02). Patients with OCD onset before age 10 had lower insight into symmetry symptoms as compared with patients with onset between age 10 and 18 (z = –2.54; P = .01). Patients whose OCD began between age 10 and 18 had lower insight into sexual/religious symptoms as compared with patients whose OCD began after age 18 (z = –2.94; P = .003). Patients with OCD onset after age 18 had lower insight into hoarding compared with patients who had OCD onset between age 10 and 18 (z = –2.87; P = .004).


In this study, we found that composite insight into OCD and dimension-specific insights are each associated with a range of distinct clinical and sociodemographic correlates. However, instead of suggesting that there is a superior way to conceptualize insight in OCD, we believe that both approaches provide relevant information and should be considered largely complementary. For instance, the traditional approach to insight in OCD showed associations between poor overall insight and earlier-onset OCD,19 greater severity of obsessive-compulsive27-29 and depressive symptoms,13,28-30 comorbid MDD,19-28 and higher levels of disability.32 Broadly speaking, these findings suggest that assessing the overall level of insight in OCD may be clinically relevant, as this construct may provide information on the global severity level exhibited by an OCD patient.

The dimension-specific approach to insight generated a number of interesting correlates. For instance, we found that insight into aggressive/checking symptoms correlated negatively with depressive symptoms and was lower among patients with comorbid anxiety, particularly GAD. These results are largely consistent with previous research showing that severity of aggressive/checking symptoms correlated positively with depression and anxiety symptoms.35,36 These findings also suggest that there might be a vulnerability trait associated with aggressive/checking symptoms—eg, inflated responsibility45 or guilt46—that makes them particularly vulnerable to the effects of comorbid depressive symptoms, resulting in decreased levels of insight. Further, because GAD preoccupations are typically egosyntonic,47,48 one might argue that egosyntonicity contaminates other symptoms (eg, obsessions) or, alternatively, that raters have assessed insight into GAD preoccupations and OCD checking symptoms instead of OCD aggressive/checking symptoms.

Dimension-specific correlations between severity of symptoms and insight ranged from very high (ρ = 0.87, for hoarding) to moderate (ρ = 0.60, for miscellaneous symptoms). Although severity of some OCD symptom dimensions correlated with insight into different OCD symptom dimensions, correlation values were remarkably lower than the ones reported above (TABLE). These findings indicate that decreased insight into a particular dimension may reflect increased severity within that particular dimension. They also suggest that it may be interesting to incorporate an evaluation of insight into different dimensions of symptoms in future versions of multidimensional assessment instruments of OCD symptoms, such as the DY-BOCS35 or the dimensional obsessive-compulsive scale.49

Although earlier-onset OCD was associated with lower insight into sexual/religious and symmetry symptoms, later onset of OCD was associated with lower insight into hoarding. The fact that early-onset OCD was related to lower insight into sexual/religious and symmetry symptoms was largely in agreement with previous studies showing greater severity of these symptoms among individuals with precocious onset.34 Lower insight into hoarding in association with later-onset OCD was more counterintuitive. However, because late-onset hoarding was reported to be rare,50 patients who described OCD onset after age 18 and lower insight into hoarding may be early-onset cases who deny precocious onset of their symptoms. In fact, substantial recognition of hoarding typically begins a decade or more after the onset of symptoms.51

Our study has a number of limitations. First, it comprised a relatively low number of participants. Second, because patients with OCD can have multiple symptoms within a single dimension, our decision to select and rate insight into the main belief from each OCD dimension can be considered somewhat arbitrary. Finally, in order to adopt a fully dimensional approach to insight in OCD, insight into an absent or mild OCD symptom was rated as present (ie, as “zero”), and not as “missing,” thus possibly inflating some correlation values. While these methodological shortcomings may lead to some degree of uncertainty about the significance of our results, they should not detract researchers from investigating correlates of different types of insight in OCD.


Correlation between severity of different obsessive-compulsive symptom dimensions and insight into OCD symptoms

DY-BOCS subscores - Severity BABS - Insight into OCD symptoms
OC symptom dimension Aggressive/ checking Sexual/ religious Symmetry/ ordering Contamination/ washing Hoarding Miscellaneous
Aggressive/checking ρ = 0.73
P < .001
ρ = –0.56
P = .67
ρ = 0.03
P = .79
ρ = 0.20
P = .11
ρ = –0.03
P = .80
ρ = –0.01
P = .94
Sexual/religious ρ =–0.16
P = .22
ρ = 0.84
P < .001
ρ = –0.33
P = .01
ρ = 0.07
P = .57
ρ = –0.12
P = .33
ρ = 0.11
P = .94
Symmetry/ordering ρ = 0.18
P = .16
ρ = –0.24
P = .06
ρ = 0.76
P < .001
ρ = 0.15
P = .23
ρ = 0.22
P = .09
ρ = 0.005
P = .97
Contamination/washing ρ = 0.15
P = .25
ρ = 0.12
P = .35
ρ = 0.07
P = .55
ρ = 0.78
P < .001
ρ = –0.07
P = .57
ρ = 0.15
P = .26
Hoarding ρ = 0.02
P = .86
ρ = –0.18
P = .17
ρ = 0.27
P = .03
ρ = 0.20
P = .88
ρ = 0.87
P < .001
ρ = –0.05
P = .67
Miscellaneous ρ = –0.04
P = .72
ρ = –0.01
P = .94
ρ = 0.24
P = .85
ρ = 0.29
P = .02
ρ = 0.02
P = .85
ρ = 0.61
P < .001
BABS: Brown Assessment of Beliefs Scale; DY-BOCS: Dimensional Yale-Brown Obsessive-Compulsive Scale; OCD: obsessive-compulsive disorder.


Our results suggest that assessing insight into dimension-specific OCD symptoms may meaningfully add to the descriptive psychopathology of OCD. Future studies along these lines may ultimately prove to challenge the existence of clear-cut OCD with fair or poor insight.

DISCLOSURES: Dr. Leonardo F. Fontenelle receives grant or research support from Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (ID E-26/111.176/2011 & E-26/103.252/2011), Conselho Nacional de Desenvolvimento Científico e Tecnológico (303846/2008-9), and D’Or Institute for Research and Education. Drs. Júlia M. Fontenelle, Harrison, Santana, Conceição do Rosário, and Versiani report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.


  1. Oyebode F. Sims’ symptoms in the mind: an introduction to descriptive psychopathology. Philadelphia PA: Elsevier Health; 2008.
  2. Marková IS, Jaafari N, Berrios GE. Insight and obsessive-compulsive disorder: a conceptual analysis. Psychopathology. 2009;42:277–282.
  3. Melca IA, Rodrigues CL, Serra-Pinheiro MA, et al. Delusional misidentification syndromes in obsessive-compulsive disorder [published online August 26, 2012]. Psychiatr Q. doi: 10.1007/s11126-012-9237-z.
  4. Solyom L, DiNicola VF, Phil M, et al. Is there an obsessive psychosis? Aetiological and prognostic factors of an atypical form of obsessive-compulsive neurosis. Can J Psychiatry. 1985;30:372–380.
  5. Ballerini A, Stanghellini G. Phenomenological questions about obsession and delusion. Psychopathology. 1989;22:315–319.
  6. Kozak MJ, Foa EB. Obsessions overvalued ideas, and delusions in obsessive-compulsive disorder. Behav Res Ther. 1994;32:343–353.
  7. Eisen JL, Phillips KA, Baer L, et al. The Brown Assessment of Beliefs Scale: reliability and validity. Am J Psychiatry. 1998;155:102–108.
  8. Frost RO, Tolin DF, Maltby N. Insight-related challenges in the treatment of hoarding. Cogn Behav Pract. 2010;17:404–413.
  9. Mataix-Cols D, Rosario-Campos MC, Leckman JF. A multidimensional model of obsessive-compulsive disorder. Am J Psychiatry. 2005;162:228–238.
  10. Greenberg D. Compulsive hoarding. Am J Psychother. 1987;41:409–416.
  11. Fitzgerald PB. ‘The bowerbird symptom’: a case of severe hoarding of possessions. Aust N Z J Psychiatry. 1997;31:597–600.
  12. Damecour CL, Charron M. Hoarding: a symptom not a syndrome. J Clin Psychiatry. 1998;59:267–272.
  13. De Berardis D, Campanella D, Gambi F, et al. Insight and alexithymia in adult outpatients with obsessive-compulsive disorder. Eur Arch Psychiatry Clin Neurosci. 2005;255:350–358.
  14. Storch EA, Lack CW, Merlo LJ, et al. Clinical features of children and adolescents with obsessive-compulsive disorder and hoarding symptoms. Compr Psychiatry. 2007;48:313–318.
  15. Jakubovski E, Pittenger C, Torres AR, et al. Dimensional correlates of poor insight in obsessive-compulsive disorder. Prog Neuropsychopharmacol Biol Psychiatry. 2011;35:1677–1681.
  16. Matsunaga H, Hayashida K, Kiriike N, et al. The clinical utility of symptom dimensions in obsessive-compulsive disorder. Psychiatry Res. 2010;180:25–29.
  17. Matsunaga H, Hayashida K, Kiriike N, et al. Clinical features and treatment characteristics of compulsive hoarding in Japanese patients with obsessive-compulsive disorder. CNS Spectr. 2010;15:258–265.
  18. Samuels JF, Bienvenu OJ 3rd, Pinto A, et al. Hoarding in obsessive-compulsive disorder: results from the OCD Collaborative Genetics Study. Behav Res Ther. 2007;45:673–686.
  19. Ravi Kishore V, Samar R, Janardhan Reddy YC, et al. Clinical characteristics and treatment response in poor and good insight obsessive-compulsive disorder. Eur Psychiatry. 2004;19:202–208.
  20. Mataix-Cols D, Frost RO, Pertusa A, et al. Hoarding disorder: a new diagnosis for DSM-V? Depress Anxiety. 2010;27:556–572.
  21. Torres AR, Fontenelle LF, Ferrão YA, et al. Clinical features of obsessive-compulsive disorder with hoarding symptoms: a multicenter study. J Psychiatr Res. 2012;46:724–732.
  22. Foa EB, Kozak MJ, Goodman WK, et al. DSM-IV field trial: obsessive-compulsive disorder. Am J Psychiatry. 1995;152:90–96.
  23. Mataix-Cols D, Rauch SL, Manzo PA, et al. 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;156:1409–1416.
  24. Tolin DF, Abramowitz JS, Kozak MJ, et al. Fixity of belief, perceptual aberration, and magical ideation in obsessive-compulsive disorder. J Anxiety Disord. 2001;15:501–510.
  25. Marazziti D, Dell’Osso L, Di Nasso E, et al. Insight in obsessive-compulsive disorder: a study of an Italian sample. Eur Psychiatry. 2002;17:407–410.
  26. Matsunaga H, Kiriike N, Matsui T, et al. A comparative study of clinical features between pure checkers and pure washers categorized using a lifetime symptom rating method. Psychiatry Res. 2001;105:221–229.
  27. Catapano F, Sperandeo R, Perris F, et al. Insight and resistance in patients with obsessive-compulsive disorder. Psychopathology. 2001;34:62–68.
  28. Türksoy N, Tükel R, Ozdemir O, et al. Comparison of clinical characteristics in good and poor insight obsessive-compulsive disorder. J Anxiety Disord. 2002; 16:413–423.
  29. De Berardis D, Campanella D, Serront N, et al. Insight and perceived expressed emotion among adult outpatients with obsessive-compulsive disorder. J Psychiatr Pract. 2008;14:154–159.
  30. Alonso P, Menchón JM, Segalàs C, et al. Clinical implications of insight assessment in obsessive-compulsive disorder. Compr Psychiatry. 2008;49:305–312.
  31. Grisham JR, Fullana MA, Mataix-Cols D, et al. Risk factors prospectively associated with adult obsessive-compulsive symptom dimensions and obsessive-compulsive disorder. Psychol Med. 2011;41:2495–2506.
  32. Storch EA, Abramowitz JS, Keeley M. Correlates and mediators of functional disability in obsessive-compulsive disorder. Depress Anxiety. 2009;26:806–813.
  33. de Mathis MA, de Alvarenga P, Funaro G, et al. Gender differences in obsessive-compulsive disorder: a literature review. Rev Bras Psiquiatr. 2011;33:390–399.
  34. Taylor S. Early versus late onset obsessive-compulsive disorder: evidence for distinct subtypes. Clin Psychol Rev. 2011;31:1083–1100.
  35. Rosario-Campos MC, Miguel EC, Quatrano S, 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.
  36. Hasler G, LaSalle-Ricci VH, Ronquillo JG, et al. Obsessive-compulsive disorder symptom dimensions show specific relationships to psychiatric comorbidity. Psychiatry Res. 2005;135:121–132.
  37. Halmi KA, Sunday SR, Klump KL, et al. Obsessions and compulsions in anorexia nervosa subtypes. Int J Eat Disord. 2003;33:308–319.
  38. LaSalle-Ricci VH, Arnkoff DB, Glass CR, et al. The hoarding dimension of OCD: psychological comorbidity and the five-factor personality model. Behav Res Ther. 2006;44:1503–1512.
  39. Fontenelle LF, Mendlowicz MV, Soares ID, et al. Patients with obsessive-compulsive disorder and hoarding symptoms: a distinctive clinical subtype? Compr Psychiatry. 2004;45:375–383.
  40. Samuels J, Bienvenu OJ 3rd, Riddle MA, et al. Hoarding in obsessive compulsive disorder: results from a case-control study. Behav Res Ther. 2002;40:517–528.
  41. Amorim P. Mini International Neuropsychiatric Interview (MINI): validation of brief interview to diagnose mental disorders. Revista Brasileira de Psiquiatria. 2000;22:106–115.
  42. Cunha JA. Manual for the Portuguese Version of Beck Scales. São Paulo Brazil: The Psychological Corporation; 2001.
  43. Sheehan DV, Harnett-Sheehan K, Raj BA. The measurement of disability. Int Clin Psychopharmacol. 1996;11:89–95.
  44. Yates DB, Trentini CM, Tosi SD, et al. Presentation of the Wechsler Abbreviated Scale of Intelligence (WASI). Avaliação Psicológica. 2006;5:227–233.
  45. Wheaton MG, Abramowitz JS, Berman NC, et al. The relationship between obsessive beliefs and symptom dimensions in obsessive-compulsive disorder. Behav Res Ther. 2010;48:949–954.
  46. Shapiro LJ, Stewart ES. Pathological guilt: a persistent yet overlooked treatment factor in obsessive-compulsive disorder. Ann Clin Psychiatry. 2011;23:63–70.
  47. Langlois F, Freeston MH, Ladouceur R. Differences and similarities between obsessive intrusive thoughts and worry in a non-clinical population: study 1. Behav Res Ther. 2000;38:157–173.
  48. Langlois F, Freeston MH, Ladouceur R. Differences and similarities between obsessive intrusive thoughts and worry in a non-clinical population: study 2. Behav Res Ther. 2000;38:175–189.
  49. Abramowitz JS, Deacon BJ, Olatunji BO, et al. Assessment of obsessive-compulsive symptom dimensions: development and evaluation of the Dimensional Obsessive-Compulsive Scale. Psychol Assess. 2010;22: 180–198.
  50. Ayers CR, Saxena S, Golshan S, et al. Age at onset and clinical features of late life compulsive hoarding. Int J Geriatr Psychiatry. 2010;25:142–149.
  51. Grisham JR, Frost RO, Steketee G, et al. Age of onset of compulsive hoarding. J Anxiety Disord. 2006;20:675–686.

CORRESPONDENCE: Leonardo F. Fontenelle, MD, PhD, Rua Visconde de Pirajá, 547 Sala 719, Ipanema, Rio de Janeiro 22410-003, Brazil E-MAIL: lfontenelle@gmail.com