Suicidality in obsessive compulsive disorder (OCD): a systematic review and meta-analysis.

Clinical Psychology Review 39 (2015) 1–15

Contents lists available at ScienceDirect

Clinical Psychology Review

Suicidality in obsessive compulsive disorder (OCD): A systematic review and meta-analysis Ioannis Angelakis a, Patricia Gooding b, Nichoas Tarrier c, Maria Panagioti a,⁎ a b c

Institute of Population Health, University of Manchester, Manchester, UK School of Psychological Sciences, University of Manchester, Manchester, UK Institute of Psychiatry, Kings College London, London, UK

H I G H L I G H T S • • • • •

The link between a diagnosis of OCD and suicidality was examined. A high, significant association between OCD and suicidality was established. Comorbid depression worsened the severity of suicidality in OCD. The methodological quality of the studies included in the review was low. Theory-driven studies are needed to examine the mechanisms of suicidality in OCD.

a r t i c l e

i n f o

Article history: Received 7 October 2014 Received in revised form 13 February 2015 Accepted 18 March 2015 Available online 25 March 2015 Keywords: Obsessive Compulsive disorder Suicidality Systematic review Meta-analysis Depression Comorbidities

a b s t r a c t Although a growing number of studies have examined the frequency of suicidal thoughts and behaviors in individuals with obsessive compulsive disorder (OCD), there is controversy about the frequency and burden of suicidality in OCD. This is the first systematic review aimed at examining the association between suicidality and OCD and at providing evidence about psychological mechanisms that may underlie suicidality in those with OCD. Five electronic bibliographic databases were searched up to April 2014: Medline, PsycINFO, Embase, Web of Science and CINAHL. Meta-analysis using random effects models was conducted. Forty-eight studies were included in the systematic review. The pooled effect size across 30 independent comparisons revealed a moderate to high, significant association between suicidality and OCD (Hedges’ g = 0.66, 95% confidence interval 0.49–0.82) which persisted across different types of suicidality including suicidal ideation and suicide attempts. Comorbid Axis I disorders, increased severity of comorbid depressive and anxiety symptoms, increased severity of obsessions, feelings of hopelessness and past history of suicide attempts were associated with worsening levels of suicidality in OCD. There was no indication for publication bias but the methodological quality of the studies was low. The theoretical, research and clinical implications of these findings are emphasized. © 2015 Elsevier Ltd. All rights reserved.

Contents 1. 2.

Introduction . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . 2.1. Eligibility criteria. . . . . . . . . 2.2. Data sources and search strategy . 2.3. Study selection . . . . . . . . . 2.4. Data extraction . . . . . . . . . 2.5. Appraisal of methodological quality 2.6. Data analyses . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

2 2 3 3 3 3 3 3

⁎ Corresponding author at: Centre for Primary Care, Institute of Population Health, Williamson Building, Oxford Road, University of Manchester, Manchester, M13 9PL. Tel.: +44 161 275 1971. E-mail address: [email protected] (M. Panagioti).

http://dx.doi.org/10.1016/j.cpr.2015.03.002 0272-7358/© 2015 Elsevier Ltd. All rights reserved.

2

I. Angelakis et al. / Clinical Psychology Review 39 (2015) 1–15

3.

Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Descriptive characteristics of the included studies . . . . . . . . . . . . . . . . . . . . 3.2. Rates of suicidal ideation in OCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Meta-analysis: suicidality and OCD . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.1. Overall association between suicidality and OCD . . . . . . . . . . . . . . . . 3.3.2. Publication bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.3. Subgroup analysis of the association between OCD and different types of suicidality 3.3.4. Sensitivity analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Potential mechanisms of suicidality in OCD . . . . . . . . . . . . . . . . . . . . . . . 3.4.1. Comorbid Axis I psychiatric disorders . . . . . . . . . . . . . . . . . . . . . 3.4.2. Severity of depressive symptoms . . . . . . . . . . . . . . . . . . . . . . . 3.4.3. Severity of anxiety symptoms . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.4. Hopelessness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.5. Past history of suicidality . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.6. OCD-specific factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.7. Other factors associated with suicidality in OCD . . . . . . . . . . . . . . . . . 4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Key research, theoretical and methodological implications . . . . . . . . . . . . . . . . 4.2. Limitations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Clinical implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Role of funding sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

7 7 7 7 7 7 7 9 9 9 9 9 10 11 11 11 11 11 13 13 13 13 13 13 13

Appendix A. Supplementary data . . . . . . . . . . . . . . . . 13 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

1. Introduction Obsessive compulsive disorder (OCD), together with four related disorders, forms a separate chapter of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), named as ObsessiveCompulsive and Related Disorders (APA, 2013). OCD is characterized by repetitive overt and covert acts which can be obsessions (i.e., intrusive recurrent thoughts), compulsions (i.e., repetitive behaviors in response to obsessions), or both (APA, 2004). The lifetime prevalence of OCD has been estimated to range between 1.5% and 3.5% with an equal gender distribution (Crino, Slade, & Andrews, 2005; Kessler, Chiu, Demler, & Walters, 2005; Ruscio, Stein, Chiu, & Kessler, 2010). OCD is a major mental health condition accompanied by severe distress, high levels of disability, and disruption of the person’s social and occupational functioning (Crino et al., 2005; Veale & Roberts, 2014). The World Health Organization has ranked OCD as one of the 10 most handicapping conditions by lost income and decreased quality of life (Veale & Roberts, 2014). OCD has been found to have a profound impact on mood, with feelings of depression and overwhelming anxiety states being common phenomena among those suffering from OCD (Marcks, Weisberg, Dyck, & Keller, 2011; Veale & Roberts, 2014). The prevalence rates of psychiatric comorbidities in those with OCD are strikingly high (up to 80%) and are considered as one of the most important factors that worsen disability in OCD (Angst et al., 2004; Levy, McLean, Yadin, & Foa, 2013; Lochner et al., 2014). Mood disorders, in particular, in those with OCD have been linked with more severe and persistent OCD symptoms, more incapacitating feelings of hopelessness and/or helplessness, and the experience of suicidal thoughts and behaviors (Angst et al., 2004; Levy et al., 2013; Marcks et al., 2011; Torres et al., 2013; Veale & Roberts, 2014). With regards to suicidality (e.g., this term will be used hereafter to describe behaviors ranging from suicidal thoughts/ideation to suicide attempts and completed suicides), the current evidence concerning OCD is largely unsystematic and inconclusive. A recent systematic review which examined the relationship between Anxiety Disorders

and suicidality failed to find evidence that the risk for suicide attempts is heightened in those with OCD (Kanwar et al., 2013). However, this review only included four OCD studies in the synthesis mainly because it restricted its focus to studies with pure anxiety samples. Studies with OCD individuals who also had depressive or non-anxiety comorbidities were excluded from the review (Kanwar et al., 2013). Given the high rates of comorbid mood disorders in OCD, only a minority of OCD individuals, potentially the milder cases do not experience concurrent nonanxiety mental health problems (Lochner et al., 2014). Additionally, in the light of recent advances in the diagnosis and classification of OCD (OCD not classified as an anxiety disorder in DSM-5) the rationale for the above restriction has become weak and might be misleading with respect to the actual magnitude of suicidality problems in patients suffering from OCD. Over the past decade, cumulative evidence suggests that suicidality in OCD is a significantly underrated phenomenon (Fawzy & Nashim, 2010; Kamath, Reddy, & Kandavel, 2007). We decided, therefore, to undertake the first compressive systematic review of the literature to explore the relationship between suicidality and OCD. Our overarching aim was to increase the understanding of the severity and the mechanisms of suicidality in OCD. To this end, this systematic review had two specific objectives: i. to systematically evaluate and quantify the association between OCD and suicidality, and ii. to identify factors that may worsen suicidality in OCD. Comorbidities (mainly depressive disorders but also other severe psychiatric disorders) were anticipated to comprise the core precipitating factors for suicidality in OCD. Disorder-specific (certain OCD symptoms) and emotio-cognitive factors (i.e., hopelessness) also were hypothesized to adversely affect suicidality in OCD.

2. Methods The presentation of this systematic review conforms to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Moher, Liberati, Tetzlaff, & Altman, 2009).


We included studies which: a) Employed a quantitative research design. Studies with any type of quantitative research design were included. b) Were conducted using adult samples (18 years and over). Although we focused on adults, two large community surveys which examined the association between OCD and suicidality recruited participants aging between 16 to 65 years (De Haan, Sterk, Wouters, & Linszen, 2013; Sareen et al., 2005). We decided to include those studies in this review because the vast majority of their sample (N 80%) were adults. c) Included a measure of OCD. d) Included a measure of suicidality (e.g., suicidal thoughts, suicidal plans, suicidal attempts and completed suicides). e) Either reported an outcome measure of the association between OCD and suicidality or examined factors that may underlie suicidality in OCD. f) Were written in English language and were published in a peerreviewed journal.

Studies written in other languages and grey literature were excluded from this review. 2.2. Data sources and search strategy Five electronic bibliographic databases were searched: Medline, PsycINFO, Embase, Web of Science and CINAHL. The search was conducted from inception to March 2014.Our search strategy included combinations of text-words and MeSH terms organized in two search blocks: suicide (suicide, suicid*, self*harm*) and OCD (OCD, obsessive*compulsive). Our search strategy is presented in appendix 1. We also checked the reference lists of the included studies for additional eligible studies.

Tool for Quantitative Studies (Thomas, Ciliska, Dobbins, & Micucci, 2004). Any disagreements were resolved by discussion. Six criteria were deemed essential for assessing the quality of the review. Each study was awarded one point for each criterion met. We did not exclude studies on the basis of their methodological quality but we used it as a framework for the synthesis of the results (Blakemore et al., 2014; Panagioti, Gooding, & Tarrier, 2012a,b). These five key criteria were: (i) methodological design (prospective = 1 crosssectional = 0), (ii) response rate at baseline (70% and over = 1, b70% or not reported = 0),(iii) response rate at follow-up (if applicable, 70% and over = 1, b 70% or not reported = 0), (iv)screening tool for OCD (psychometrically validated clinical interview = 1; self-report/not reported = 0), (v) screening tool for suicidality (psychometrically validated clinical interview/self-report = 1; other/not reported = 0), and (vi) control for confounding factors in the analysis (controlled = 1, not controlled/not reported = 0). 2.6. Data analyses We first examined the relationship between suicidality and an OCD diagnosis. Outcomes across studies were converted in Hedges’ g(the standardized mean difference, d, corrected for small sample bias) and pooled using Comprehensive Meta-Analysis version 2.2 (www.metaanalysis.com/) (Borenstein, Hedges, Higgins, & Rothstein, 2005). Subgroup analyses were performed to examine the association of OCD with different types of suicidality (i.e., suicide attempts and suicidal ideation). Sensitivity analysis was performed to examine whether the strength of the association between OCD and suicidality differed when retaining only studies with high methodological quality in the analyses. Where possible, the pooled effects of identified factors associated with

Identification

2.1. Eligibility criteria

3

Records identified through database searching (n = 627) • Duplicates removed (n =97)

Two reviewers (IA and MP) independently screened the titles and abstracts of the identified citations. Next, the full-texts of potentially eligible studies identified during title/abstract screening were accessed and assessed. High inter-rater agreement was achieved for title/abstract and full-text screening (kappa = .93 and .95, respectively). Disagreements were discussed until consensus was reached.

Screening

2.3. Study selection

428 records excluded following reading of title and abstract

The methodological quality of the studies was rated by two researchers (IA, MP) using criteria adapted from guidance on the assessment of observational studies (CRD, 2009) and the Quality Assessment

102 records were eligible for full-text screening

54 records excluded following reading the full-text: • 37 reported no relevant data on suicidality and OCD • 14werenon-empirical papers • 3 were literature reviews Included

2.5. Appraisal of methodological quality

Eligibility

2.4. Data extraction An electronic data extraction form was developed and piloted in five randomly selected studies and modified where necessary. The following descriptive information was extracted: study characteristics (country, design, and recruitment), participant characteristics (age, gender, primary and secondary diagnoses), suicidality outcome (type of suicidality including thoughts, attempted and completed suicides, screening tool for suicidality), OCD outcome (screening tool for OCD). Moreover, we extracted quantitative data of i. the association between OCD and suicidality, ii. factors associated with suicidality in OCD. Data extraction was performed by two authors (IA and MP) independently. High interrater agreement (kappa = .91) was revealed on 750 data points checked.

Abstracts from 530 records were screened for eligibility

Overall, 48 studies included in the review

Fig 1. Flowchart of study selection.

4

Table 1 Descriptive characteristics of the studies included in the review. Study

Country

Study Design

Alonso et al. (2010)

Spain

Prospective cohort

Anderson, Gehl, Marder, Beglinger, and Paulsen (2010)

USA

Angst et al. (2004)

Switzerland

Angst et al. (2005)

Switzerland

Turkey

Benvenuti et al. (2009)

Italy

Bolton et al. (2008)

Canada

Chakraborty et al. (2012)

India

Chen and Dilsaver (1995)

USA

Conceicao Costa et al. (2012)

Brazil

De Haan et al. (2013)

Dell'Osso et al. (2012)

Structured Clinical Interview for DSM-IV Axis I Disorders-Clinician Version (SCID-CV), Yale–Brown Obsessive–Compulsive Scale (Y-BOCS) Cross-sectional The behavioral section of the Unified Huntington’s Disease Rating Scale (UHDRS)

Prospective cohort

The Structured Psychopathological Interview, Rating of the Social Consequences for Epidemiology (SPIKE) Prospective The Structured Psychopathological cohort Interview, Rating of the Social Consequences for Epidemiology (SPIKE) Cross-sectional The Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) Cross-sectional The Structured Clinical Interview for DSM- IV Axis I Disorders, the Obsessive-Compulsive Spectrum Self-Report questionnaire, the Obsessive-Compulsive Spectrum Self-Report (OBS-SR) Prospective The National Institute of Mental Health cohort Diagnostic Interview Schedule (DIS) Cross-sectional The Y-BOCS checklist and severity rating scale, the Mini International Neuropsychiatry Interview plus, the Structured Clinical Interview for DSM-IV Axis I Disorders, the Global Assessment of Functioning (GAF) scale and the SI-R Cross-sectional The Diagnostic Interview Schedule based on the DSM-III Diagnostic criteria

Cross-sectional The Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), the Dimensional Yale–Brown Obsessive–Compulsive Scale (DY-BOCS), the Yale Obsessive-Compulsive Disorder Natural History Questionnaire The Cross-sectional DSM-IV diagnosis was based on the Netherlands SCID-I + open-ended questions, SCID-P, the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) Italy Cross-sectional The Obsessive-Compulsive Spectrum Self-Report (OBS-SR)

DeVylder et al. (2012)

USA

Cross-sectional The Yale-Brown Obsessive Compulsive Scale (Y-BOCS)

Fawzy and Nashim (2010)

Egypt

Cross-sectional Semi-structured psychiatric interview, the Yale–Brown Obsessive–Compulsive Scale (YBOCS)

Screening tool for suicidality

Mode of Suicidality

Target Population

Item 3 of the HDRS, the Beck Suicide Intent Scale (SIS)

Suicidal ideation & Suicide attempts

Individuals with OCD

225

57.34% M = 31.3 (SD =18.8), range: 18 - 65

The behavioral section of the Unified Huntington’s Disease Rating Scale (UHDRS) Open-ended questions as part of the clinical interview Open-ended questions as part of the clinical interview The Scale for Suicidal Ideation (SSI) The Mood Spectrum Self-Re- port questionnaire

Suicidal ideation

Individuals with Huntington disease

445

49.6%

n/r

Suicide attempts

General Population

430

44.14

M = 20

The National Institute of Mental Health Diagnostic Interview Schedule (DIS) n/r

n/r

Sample Men size N (%)

Age (years)

% Suicide attempts

General Population

274

44.14

M = 20

% Suicidal ideation & Suicide attempts Suicidality


23

45.5 %

M = 33.71 (SD = 11.51)

n/r

M = 38.93 (SD = 11)

Mixed sample of schizophrenia, psychotic mood disorders, non-psychotic mood disorders

221

Suicide attempts

Institutionalized and community participants

33

37 %

n/r

Suicidality & suicide attempts


20

51.5 %

n/r

Suicidal ideation & Suicide attempts

Individuals with Bipolar, Unipolar and any other Axis I disorder Individuals with OCD

5659

n/r

Range 18 - 65+

109

42.7 %

M = 34.4 (SD = 12.7)

16

85.8 %

Range 15 - 28

289

41.65

Range 18 - 60

Questions as part of clinical Suicidal ideation & interview Suicide attempts

n/r

Completed suicides

Individuals with schizophrenia or related disorders

The Mood Spectrum-Self Report (MOODS-SR) lifetime version 1 item from the Beck Depression Inventory II (BDI-II) The Beck’s Suicide Ideation Scale, lifetime suicide attempts were assessed by a direct question as part of the interview

Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), Suicidal ideation

Individuals with mood disorders, panic disorder or schizophrenia Individuals at high risk for psychosis

Suicidal ideation & suicide attempts

Psychiatric outpatients

% 12

60%

Range 14 - 27

21

61%

M = 29.1 (SD = 6.5), range 20 - 50


Balci and Sevincok (2010)

Screening tool for OCD

Fineberg et al. (2013)

UK

Prospective cohort

Interviews using the SPIKE

Brazil

Cross-sectional The SCID-I, Yale Obsessive–Compulsive Disorder Natural History Questionnaire, the DY-BOCS, the Yale–Brown Obsessive–Compulsive Scale (Y-BOCS)

Gentil et al. (2009)

Brazil


Goes et al. (2012)

USA

Goodwin, Koenen, Hellman, Guardino, and Struening (2002)

USA

Hagen, Hansen, Joa, and Larsen (2013) Hantouche et al. (2003)

Norway France

Cross-sectional Diagnoses were based on Diagnostic and Statistical Manual of Mental Disorders third edition revised (DSM-III-R) and fourth edition (DSM-IV) criteria Cross-sectional The screening questionnaire included two OCD items, one cognitive and one behavioral, with a 1-month reference period (Based on DSM-IV) Cross-sectional The SCID-I and the Positive and Negative Syndrome Scale (PANSS) Cross-sectional A questionnaire that included 250 items was prepared for self-assessment

Hung et al. (2010)

China

Cross-sectional The Y-BOCS

Kamath et al. (2007)

India

Koyuncu, Tukel, Ozyildirim, Meteris, and Yazici (2010) Kruger, Braunig, and Cooke (2000) Magalhaes, Kapczinski, and Kapczinski (2010) Mahasuar, Janardhan Reddy, and Math (2011)

Turkey

Cross-sectional The Structured Clinical Interview for DSM-IV Axis I and II disorders Cross-sectional The Y-BOCS & the SCID-I/CV

Germany Brazil Australia

Maina, Salvi, Tiezzi, Albert, and Bogetto (2007)

Italy

Mohammadi et al. (2005)

Iran

Moreira et al. (2013)

Brazil

Norton, Temple, and Pettit (2008) Peles et al. (2009)

USA Israel


General Population

191

49.41


917

% 43.05% n/r


M = 20

47

43.65% n/r

The Beck Depression and Anxiety Inventories (BDI & BAI), the Brown Assessment of Beliefs Scale (BABS) The Brown Assessment of Beliefs Scale (BABS), Beck Depression and Anxiety Inventories (respectively Beck-D and Beck-A) and the Global Clinical Impressions Scale (CGI) n/r



Suicide attempts

Individuals with bipolar disorder and major depression

1167

n/r

n/r

Suicidal ideation

General population

1241

32.92% Range 21 - 55+

Questions as part of the clinical interview n/r

Suicidal ideation & suicide attempts Suicidal ideation, suicide attempts, completed suicides Suicidal ideation

Psychiatric outpatients Patients with OCD

26

n/r

56.91% n/r

302

43.9%

n/r

Outpatients with OCD

58

64.1%

n/r


100

59%

n/r

35

35,7%

n/r

10

-

n/r

The Beck Scale for Suicide Ideation (BSS) The Scale for Suicide Ideation (SSI) n/r

Suicidal ideation & suicide attempts Suicide attempts

Cross-sectional The Structured Clinical Interview for DSM-III-R (SCID) Cross-sectional The SCID

n/r

Suicide attempts

n/r

Suicide attempts

Cross-sectional The Structured Clinical Interview for DSM-IV Axis I, the Y-BOCS and the Clinical Global Impression-severity (CGI-S) Cross-sectional The Structured Clinical Interview for the DSM-IV Axis I Disorders (SCID- I) and the Yale-Brown Obsessive Compulsive Scale (Y-BOCS) Cross-sectional The Schedule for Affective Disorders and Schizophrenia (SADS) based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) Cross-sectional The SCID-I, the Yale Obsessive–Compulsive Disorder Natural History Questionnaire, the DY-BOCS, the Yale–Brown Obsessive–Compulsive Scale (Y-BOCS) Cross-sectional A self-report version of the Y-BOCS was used. Cross-sectional The Yale-Brown Obsessive Compulsive Scale

The Scale for Suicidal Ideation (SSI)


The Item 3 of the Hamilton Rating Scale for Depression (HDRS)



n/r

Suicide attempts

General population

Individuals with Bipolar disorder Individuals with Bipolar disorder Individuals with Bipolar disorder Individuals with Bipolar disorder

259

29.34% n/r

34

72.53% n/r

58

64.66


Fontenelle et al. (2012)

It is not specified

n/r

%

362

43.65

Range 18 - 65

%

Item from Beck Depression Inventory II (BDI-II)



455

0%

n/r

The ASIQ is a 25-item self-report measure A structured questionnaire

Suicidal ideation

Undergraduate students

166

28.3 %

M = 21.68, (SD = 3.88)

Suicide attempts

Patients in Methadone maintenance treatment

67

76 %

Mean = 37.9 (SD = 9.1) (continued on next page) 5

6

Table 1 (continued) Study

Country

Study Design

Perugi et al. (1997)

Italy

Screening tool for OCD

USA

Phillips et al. (2007)

USA

Cross-sectional The SCID-I and Y-BOCS

Sallet et al. (2010)

Brazil


Sareen et al. (2005)

Canada

Prospective cohort

Sevincok, Akoglu, and Kokcu (2007) Tavares et al. (2012)

Turkey

Cross-sectional The SCID-I and Y-BOCS

Brazil

Cross-sectional Suicide risk and other mental disorders were assessed using the Mini International Neuropsychiatric Interview (MINI)

ten Have et al. (2009)

The DSM-III-R Axis I mental disorders were diagnosed with version 1.1 of the Composite International Diagnostic Inter- view (CIDI)

Torres et al. (2006)

Netherlands Prospective The CIDI, computerized version 1.1 cohort Brazil Cross-sectional The Clinical Interview Schedule-Revised


Brazil

Cross-sectional The Y-BOCS


Brazil

Cross-sectional The Y-BOCS & the DY-BOCS


Brazil

Viswanath et al. (2012)

India

Whitters, Cadoret, and Widmer (1985)

USA

Cross-sectional The Y-BOCS, the DY-BOCS & the Yale Obsessive–Compulsive Disorder Natural History Questionnaire Cross-sectional The diagnosis of OCD was made according to the DSM, Fourth Edition, criteria, the Y-BOCS, the MINI & CGI Cross-sectional All diagnoses based on DSM-III criteria

Zisook et al. (2011)

USA

Cross-sectional The Psychiatric Diagnostic Screening Questionnaire

Note: M = mean, SD = standard deviation, n/r = not reported.

Mode of Suicidality

Target Population

A question as part of the clinical interview

Suicide attempts

OCD

Suicide mortality was determined from the cause of death A semi-structured instrument to obtain information about clinical characteristics and suicidality Beck Depression and Anxiety Inventories (Beck-D and Beck-A) and the Clinical Global Impressions Scale (CGI) Items from the CIDI questionnaire were used to assess both suicidal ideation and suicide attempts n/r

Completed suicide

Patients with depression


Sample Men size N (%) 345

44.35

Age (years) n/r

% 1892

91.9%

Range 18 - 65

Patients with Body Dysmorphic Disorder (BDD) & OCD

250

49.6 %

n/r



739

48.17

n/r


General population

Suicidal ideation & suicide attempts Suicidality

Patients with OCD

Suicide risk and other mental disorders were assessed using the Mini International Neuropsychiatric Interview (MINI) The CIDI, computerized Suicide attempts version 1.1 n/r Suicidal ideation & suicide attempts Specifically designed Suicidal ideation & questionnaire suicide attempts Specifically designed Suicidal ideation & suicide attempts questionnaire Specifically designed Suicidal ideation & questionnaire suicide attempts

%

7074

57

n/r

n/r

54.39% n/r

Postpatrum women

106

0%

n/r

General population

7076

n/r

Range 18 - 64

General population

7185

51.04% Range 16 - 74


50

44%

Range 29 - 60


582

43.6%

M = 34.74 (SD = 0.52)


955

41.88% Mean = 35.8 (SD = 12.4)

545

52.48% Mean = 29.3 (SD = 10.6)

n/r


Patients with OCD

n/r


The Concise Health Risk Tracking Self- Report (CHRT-SR) scale

Suicidal ideation

Inpatients admitted to an acute alcoholic treatment program Outpatients with Major or other Depressive Disorder

31

110

73.68% n/r

32.03% Range 18 - 75


Pfeiffer et al. (2009)

Cross-sectional A specially constructed OCD Questionnaire and a semi-structured in-depth interview Cross-sectional It is not specified

Screening tool for suicidality


suicidality in OCD were calculated. Random effects models were applied to calculate pooled Hedge’s g because of anticipated heterogeneity. Heterogeneity was evaluated using the Q statistic and the I2 statistic. The Q statistic provides an estimate of whether differences between meta-analyzed studies are greater than would be expected by chance. Statistically significant results indicate the presence of heterogeneity. The I2 statistic provides a quantitative measure of the degree of between study differences caused by factors other than sampling error. Higher I2 values represent greater heterogeneity (Higgins, Thompson, Deeks, & Altman, 2003). Publication bias was examined using a test of funnel plot asymmetry (Egger’s test) (Egger, Smith, Schneider, & Minder, 1997)and the Rosenthal's fail safe N (FSN) (Rosenthal, 1979). The Egger’s test reveals whether or not the funnel plot is symmetric and indicates the existence/absence of a significant publication bias, and the FSN provides an estimate of the number of studies with statistically non-significant results are needed for a meta-analyzed finding to become statistically non-significant.

7

(Hantouche et al., 2003; Pfeiffer, Ganoczy, Ilgen, Zivin, & Valenstein, 2009). High variability was observed in the rates of suicidal ideation and suicidal attempts across different studies. However, we found no evidence that this variability can be attributed to variations such as type of population (clinical populations or community populations), sample size and severity of OCD and measurement of OCD and suicidality, suggesting that the sources of this variability are potentially more obscure. OCD is a heterogeneous disorder expressed by diverse symptoms and is often comorbid with other psychiatric disorders (Bloch, Landeros-Weisenberger, Rosario, Pittenger, & Leckman, 2008; Pallanti, Grassi, Sarrecchia, Cantisani, & Pellegrini, 2011). Although we extracted rates of suicidality from groups described as “pure OCD samples” by the individual studies, comorbidities might still be present in these groups and may account for these large variations. Moreover, in this review we included studies conducted across the globe (e.g., Egypt, India, Brasil, USA etc.) and therefore, ethnic, religious and socioeconomic differences might account for the high variability in the reported rates of suicidality.

3. Results The searches yielded 530 relevant citations. Of these, 428 citations were excluded following title and abstract inspection. The full texts for 102 citations were retrieved. Forty-eight studies met the inclusion criteria of the review following full-text inspection. The PRISMA flow chart for the studies is presented in Fig. 1. 3.1. Descriptive characteristics of the included studies Table 1 summarizes the characteristics of the 48 studies included in this review. As shown, the majority of studies were conducted in the US (22.9%) and Brazil (22.9%), followed by Europe (16.7%). Forty-two studies provided information about the gender characteristics of their sample. The gender distribution was equal across the studies. Twenty-five studies were predominately composed by females whereas 17 studies were predominately composed by males. The mean age of the study samples ranged from 21.7 to 38.9 years. The most widely used tools (n = 37) for assigning a diagnosis of OCD were the Structured Clinical Interview for DSM-IV or DSM-III-R (SCID) and/or the Yale-Brown Obsessive Compulsive Scale (Y-BOCS). Only two studies used self-reports as the primary diagnostic tool for OCD and one study provided no information about the tool used for the diagnosis of OCD. The most frequently reported types of suicidality were suicidal ideation (n = 36) and/or attempts (n = 40). A small portion of the studies reported data on suicidal plans (n = 6), a mixed/unspecified outcome of suicidality (n = 5), and completed suicides (n = 2). Twenty studies provided data on more than one type of suicidality. Suicidality was mainly assessed (n = 21) through validated self-report questionnaires (i.e., Beck Suicidal Ideation Scale or Scale for Suicidal Ideation) or selfreport items designed by the authors for the purposes of the studies. Additionally, 11 studies utilized open-ended questions as part of clinical interviews to obtain information about previous suicide attempts. Despite that, a considerable number of studies (n = 16) failed to provide any information about the tools used to assess suicidality (see Table 1). 3.2. Rates of suicidal ideation in OCD Twenty-five studies reported rates of suicidal ideation, attempts and completed suicides in those with OCD. Across 18 studies, the overall rates of suicidal ideation ranged from 10% to 63.5% (M = 30.57%, SD = 16.47) with current rates (past year suicidal ideation) ranging from 10% to 52.3%, (n = 13, M = 21.2%, SD = 12.04) and lifetime rates ranging from 31.7% to 63.5% (n = 9, M = 44.23%, SD = 11.99). In total, 22 studies reported rates of suicidal attempts in OCD ranging from 1% to 46.3% with a mean of 13.37% (SD = 9.79). Only two studies provided rates with regard to complete suicides (M = 5.73, SD = 1.03)

3.3. Meta-analysis: suicidality and OCD Among the 48 studies included in this review, 25 studies (providing 30 relevant comparisons) reported data on the association between suicidality and an OCD diagnosis. Across studies reporting more than one mode of suicidality (e.g., suicide attempts and suicidal ideation), the type of suicidality that was most proximal to completed suicide (suicide attempts) was entered in the analyses. The remaining 23 studies mainly compared suicidal OCD patients with non-suicidal OCD patients to identify potential mechanisms of suicidality in OCD. 3.3.1. Overall association between suicidality and OCD The pooled effect size across the 30 relevant comparisons indicated a moderate, significant association between suicidality and OCD but exhibited high heterogeneity (g = 0.66, 95% CI = 0.49-0.82, Q = 191.39, I2 = 84.8%, p b 0.0001).The individual effect sizes across 28 studies ranged from small to high and had a positive direction indicating that OCD was associated with higher levels of suicidality. Only two studies reported small, non-significant effect sizes with a negative direction indicating an inverse relationship between OCD and suicidality (see forest plot in Fig. 2). 3.3.2. Publication bias We found no evidence of publication bias. No funnel plot asymmetry was identified (Egger test p = 0.791; Fig. 3). In addition, the FSN test indicated that as many as 1056studies would be necessary to nullify the present significant effects. 3.3.3. Subgroup analysis of the association between OCD and different types of suicidality A subgroup analysis was undertaken to examine the association between OCD and different types of suicidality which were suicidal ideation, suicide attempts, and completed suicides. Six comparisons reported data on suicidal ideation, 22 comparisons reported data on suicide attempts and two comparisons reported data on completed suicides. The pooled effect sizes indicated a high, significant association between OCD and suicidal ideation (g = 0.86, 95% CI = 0.22-1.51), a moderate, significant association between OCD and suicide attempts (g = 0.64, 95% CI = 0.47-0.80), and a non-significant association between OCD and completed suicides (g = 0.21, 95% CI = −0.08-0.49). The heterogeneity was high for both suicidal ideation and suicide attempts (Q = 87.11, I2 = 94.3%, p b 0.0001 and Q = 94.89, I2 = 77.9%, p b 0.0001, respectively). No heterogeneity was observed for studies with completed suicides (Q = 0.15, I2 = 0%, p = 0.695) but this subgroup only included two studies (see forest plot in Fig. 4).

8


Study ID

ES (95% CI)

% Weight

Anderson 2010 Angst 2004 Angst 2005 Benvenuti 2009 Bolton 2007 Chen 1995a Chen 1995b Chen 1995c De Haan 2011 DeVylder 2012 Goes 2012a Goes 2012b Hagen 2013 Koyouncu 2010 Kruger 2000 Magalhães 2010a Magalhães 2010b Maina 2007 Mohammandi 2011 Peles 2009 Pfeiffer 2009 Phillips 2006 Sareen 2005 Sevincok 2007 Tavares 2012 ten Have 2009 Torres 2006a Torres 2006b Witters 1985 Zisook 2011 Overall (I-squared = 84.8%, p = 0.000)

0.32 (0.20, 0.44) 0.49 (0.02, 0.96) 1.13 (0.46, 1.80) 0.24 (-0.80, 1.28) 0.69 (0.02, 1.36) 0.45 (0.02, 0.88) 0.20 (-0.09, 0.49) 0.58 (0.36, 0.80) -0.03 (-1.25, 1.19) 0.89 (-0.42, 2.20) 0.44 (0.22, 0.66) 0.44 (0.20, 0.68) 0.69 (0.16, 1.22) -0.26 (-0.87, 0.35) 1.47 (0.31, 2.63) 0.81 (0.34, 1.28) 0.33 (0.09, 0.57) 0.57 (-0.10, 1.24) 0.99 (0.77, 1.21) 0.49 (0.14, 0.84) 0.22 (-0.07, 0.51) 0.81 (0.22, 1.40) 1.25 (0.56, 1.94) 1.08 (0.45, 1.71) 1.95 (1.62, 2.28) 0.89 (-0.25, 2.03) 0.39 (0.15, 0.63) 1.48 (1.23, 1.73) 0.71 (0.28, 1.14) 0.43 (0.14, 0.72) 0.66 (0.49, 0.82)

4.66 3.47 2.72 1.69 2.72 3.63 4.16 4.42 1.36 1.22 4.42 4.36 3.24 2.94 1.46 3.47 4.36 2.72 4.42 3.94 4.16 3.01 2.66 2.86 4.01 1.50 4.36 4.29 3.63 4.16 100.00

NOTE: Weights are from random effects analysis -2

-1

0

1

2

Fig. 2. Overall meta-analysis of the association between OCD and suicidality. Note: Meta-analysis of individual studies and pooled effects. Random effects model used. 95% CI = 95% confidence intervals; ES = Hedge’s g.

Funnel Plot of Standard Error by Hedges’ g 0.0

Standard Error

0.2

0.4

0.6

0.8 -2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Hedges’ g Fig. 3. Publication bias funnel plot showing standard error by the Hedges’ g for the 25 studies (30 relevant comparisons), which examined the association between suicidality and OCD. Egger’s regression intercept = 0.32 (SE = 1.20), p = 0.791.


Study ID Suicidal deaton Anderson 2010 Benvenuti 2009 DeVylder 2012 Sareen 2005 Tavares 2012 Zisook 2011 Subtotal (I-squared = 94.3%, p = 0.000) . Suicide attempts Angst 2004 Angst 2005 Bolton 2007 Chen 1995a Chen 1995b Chen 1995c Goes 2012a Goes 2012b Hagen 2013 Koyouncu 2010 Kruger 2000 Magalhães 2010a Magalhães 2010b Maina 2007 Mohammandi 2011 Peles 2009 Phillips 2006 Sevincok 2007 ten Have 2009 Torres 2006a Torres 2006b Witters 1985 Subtotal (I-squared = 77.9%, p = 0.000) . Completed suicides De Haan 2011 Pfeiffer 2009 Subtotal (I-squared = 0.0%, p = 0.695) . NOTE: Weights are from random effects analysis

-2

-1

0

1

9

ES (95% CI)

% Weight

0.32 (0.20, 0.44) 0.24 (-0.80, 1.28) 0.89 (-0.42, 2.20) 1.25 (0.56, 1.94) 1.95 (1.62, 2.28) 0.43 (0.14, 0.72) 0.86 (0.22, 1.51)

20.22 13.33 11.05 16.55 19.31 19.53 100.00

0.49 (0.02, 0.96) 1.13 (0.46, 1.80) 0.69 (0.02, 1.36) 0.45 (0.02, 0.88) 0.20 (-0.09, 0.49) 0.58 (0.36, 0.80) 0.44 (0.22, 0.66) 0.44 (0.20, 0.68) 0.69 (0.16, 1.22) -0.26 (-0.87, 0.35) 1.47 (0.31, 2.63) 0.81 (0.34, 1.28) 0.33 (0.09, 0.57) 0.57 (-0.10, 1.24) 0.99 (0.77, 1.21) 0.49 (0.14, 0.84) 0.81 (0.22, 1.40) 1.08 (0.45, 1.71) 0.89 (-0.25, 2.03) 0.39 (0.15, 0.63) 1.48 (1.23, 1.73) 0.71 (0.28, 1.14) 0.64 (0.47, 0.80)

4.46 3.31 3.31 4.73 5.67 6.16 6.16 6.04 4.09 3.62 1.62 4.46 6.04 3.31 6.16 5.27 3.73 3.51 1.66 6.04 5.92 4.73 100.00

-0.03 (-1.25, 1.19) 0.22 (-0.07, 0.51) 0.21 (-0.08, 0.49)

5.53 94.47 100.00

2

Fig. 4. Subgroup analysis examining the association between different types of suicidality (suicidal ideation, suicide attempts, completed suicides) and OCD. Random effects model used. 95% CI = 95% confidence intervals; ES = Hedge’s g.

3.3.4. Sensitivity analysis The methodological quality of the studies was generally low. Among the 25 studies included in the pooled analysis, only five studies met at least four of these quality criteria (Peles, Adelson, & Schreiber, 2009; Phillips et al., 2007; Sareen et al., 2005; Tavares et al., 2012; ten Have et al., 2009) (see Fig. 5). A sensitivity analysis based on these five studies showed that the magnitude of the pooled effect size increased considerably compared with the larger pooled analysis (g = 1.09, 95% CI = 0.391.80, Q = 37.01, I2 = 59.2%, p b 0.0001) indicating a stronger association between suicidality and OCD across studies with sound methodological quality (see forest plot in Fig. 5). 3.4. Potential mechanisms of suicidality in OCD In addition to examining the magnitude of the association between OCD and suicidality, potential drivers of suicidality in OCD were examined. A total of 30 studies examined factors that might underlie suicidality in OCD. The key factors discussed in the literature as potentially implicated in the development of suicidality in OCD include: 3.4.1. Comorbid Axis I psychiatric disorders Fourteen studies (providing 22 independent comparisons) were identified which compared the levels of suicidality in OCD groups with Axis I comorbidities with OCD groups without Axis I comorbidities.

The most common comorbidities were mood disorders (n = 10). The pooled effect size indicated moderate, significant increases in suicidality in the OCD groups with comorbidities, compared to OCD groups without comorbidities but exhibited moderate levels of heterogeneity (g = 0.51, 95% CI = 0.35-0.67, Q = 45.36, I2 = 53.7%, p = 0.002). With only one exemption, which reported a negative, non-significant effect size, all the effect sizes had a positive direction and ranged from small to high (See forest plot in Fig. 6). 3.4.2. Severity of depressive symptoms Ten studies reported data on the severity of depressive symptoms (measured using the Beck Depression Inventory) in the OCD groups with suicidality and the OCD groups without suicidality. The pooled effect size indicated high, significant increases in the severity of depressive symptoms in OCD groups with suicidality compared to OCD groups without suicidality (g = 0.72, 95% CI = 0.56-0.89). No substantial heterogeneity was observed (Q = 14.78, I2 = 39.1%, p =0.097). All the effect sizes had a positive direction and ranged from moderate to high (see forest plot in Fig. 7). 3.4.3. Severity of anxiety symptoms Three studies reported data on the severity of anxiety symptoms (measured using the Beck Anxiety Inventory) in the OCD groups with suicidality and the OCD groups without suicidality. The pooled effect size indicated moderate significant increases the severity of anxiety

10


Study

%

ID

ES (95% CI)

Peles 2009

0.49 (0.14, 0.84) 22.56

Phillips 2006

0.81 (0.22, 1.40) 20.50

Sareen 2005

1.25 (0.56, 1.94) 19.50

Tavares 2012

1.95 (1.62, 2.28) 22.70

ten Have 2009

0.89 (-0.25, 2.03) 14.75

Overall (I-squared = 89.2%, p = 0.000)

1.09 (0.39, 1.80) 100.00

Weight

NOTE: Weights are from random effects analysis -2

-1

0

1

2

Fig. 5. Sensitivity analysis examining the association between suicidality and OCD across studies with greater methodological quality. Random effects model used. 95% CI = 95% confidence intervals; ES = Hedge’s g.

symptoms in OCD groups with suicidality compared to OCD groups without suicidality while no heterogeneity was observed (g = 0.46, 95% CI = 0.27–0.66, Q = 0.46, I2 = 0%, p = 0.795) (see forest plot in Fig. 7).

3.4.4. Hopelessness Only two studies examined the levels of hopelessness (measured using the Beck Hopelessness Scale) in the OCD groups with suicidality and the OCD groups without suicidality. The pooled effect size indicated

Study ID

ES (95% CI)

% Weight

Conceico 2012 Fineberg 2013a Fineberg 2013b Gentil 2009 Mahasuar 2011 Maina 2007 Perugi 1997a Perugi 1997b Phillips 2006 Sallet 2010a Sallet 2010b Sallet 2010c Tores 2006 Angst 2005b Angst 2005c Angst 2005d Torres 2013 Viswanath 2012a Viswanath 2012b Fontenelle 2012b Hantouce 2013 Moreira 2013 Overall (I-squared = 53.7%, p = 0.002)

0.64 (0.37, 0.91) 1.29 (0.62, 1.96) 0.56 (0.07, 1.05) 0.62 (0.21, 1.03) 0.11 (-0.48, 0.70) -0.13 (-0.87, 0.61) 0.00 (-0.29, 0.29) 0.36 (0.12, 0.60) 0.70 (0.29, 1.11) 2.77 (-1.15, 6.69) 2.46 (-1.77, 6.69) 1.64 (-2.71, 5.99) 0.03 (-0.46, 0.52) 1.03 (-0.17, 2.23) 1.09 (-0.20, 2.38) 1.48 (0.30, 2.66) 1.61 (0.08, 3.14) 0.86 (0.45, 1.27) 0.09 (-0.36, 0.54) 0.66 (0.39, 0.93) 0.66 (0.25, 1.07) 0.35 (0.04, 0.66) 0.51 (0.35, 0.67)

8.22 3.77 5.38 6.32 4.41 3.24 7.94 8.78 6.32 0.16 0.14 0.13 5.38 1.54 1.34 1.58 1.00 6.32 5.83 8.22 6.32 7.66 100.00

NOTE: Weights are from random effects analysis -2 -1 0

1

2

Fig. 6. Differences in the levels of suicidality between OCD groups with Axis I comorbidities and OCD groups without Axis I comorbidities. Note: Meta-analysis of individual studies and pooled effects. Random effects model used. 95% CI = 95% confidence intervals; ES = Hedge’s g.


high, significant increases in the levels of hopelessness in OCD groups with suicidality compared to OCD groups without suicidality while no heterogeneity was observed (g = 0.70, 95% CI = 0.33-1.06; Q = 0.88, I2 = 0%, p = 0.345) (see forest plot in Fig. 7). 3.4.5. Past history of suicidality Three studies provided data on past history of suicidality (suicide attempts or persistent suicidal ideation) in the OCD groups with current suicidality and the OCD groups without current suicidality. The pooled effect size indicated moderate significant differences on the rates of a past history of suicidality in OCD groups with current suicidal ideation compared to OCD groups without current suicidal ideation while substantial heterogeneity is present (g = 0.64, 95% CI = 0.13-1.16, Q = 4.48 I2 = 55.1%, p =0.107)(see forest plot in Fig. 7). 3.4.6. OCD-specific factors Thirteen studies examined the role of OCD-specific factors in relation to suicidality including overall severity of OCD symptoms, OCD symptom clusters (i.e., obsessions and compulsions), and individual OCD symptoms. Nine studies compared the overall OCD symptom severity in the OCD groups with suicidality with the OCD groups without suicidality. The pooled effect size indicated medium significant differences on the overall OCD symptom severity in OCD groups with suicidality compared to OCD groups without suicidality, but the heterogeneity was high (g = 0.44, 95% CI = 0.13–0.75, Q = 67.25, I2 = 86.6%, p b 0.0001). With one exception which was a negative, non-significant effect size, all the effect sizes had a positive direction and ranged from small to high (see forest plot in Fig. 7). Five studies reported data on the severity of obsessions and compulsions in the OCD groups with suicidality and the OCD groups without suicidality. The pooled effect sizes indicated small and significant differences on the effects of severity of obsessions in the OCD groups with suicidality, compared OCD groups without suicidality (g = 0.32, 95% CI = 0.05-0.59), and small but non-significant differences on the severity of compulsions in OCD groups with suicidality, compared to OCD groups without suicidality (g = 0.25, 95% CI = −0.04–0.55). Moderate heterogeneity was observed (Q = 6.87, I2 = 41.8%, p = 0.03 and Q = 7.95, I2 = 49.7%, p = 0.02 for both obsessions and compulsions) (see forest plot in Fig. 7). Moreover, preliminary evidence suggests that suicidality in OCD is particularly heightened by the experience of aggressive and sexual obsessions (Dell'Osso et al., 2012). Other individual symptoms identified as predictors of suicidality in OCD include hoarding (Chakraborty et al., 2012) and symmetry/ordering symptoms (Alonso et al., 2010). Since each of these three studies focused on different OCD symptoms the results could not be pooled using meta-analysis. 3.4.7. Other factors associated with suicidality in OCD Other factors found to be implicated in the development of suicidality in OCD include adverse family relationships, and a number socio-demographic predictors including being male, unmarried and unemployed (Fawzy & Nashim, 2010). Additionally, Moreira et al. (2013) demonstrated that women with premenstrual worsening of OCD symptoms have significantly higher rates of suicidal ideation compared to women without premenstrual worsening OCD symptoms. The small number, and the heterogeneous populations/outcomes, of these studies do not allow any firm conclusions to be reached. 4. Discussion The vast majority of studies discussed in this review found that the incidence of suicidality in OCD is considerably higher than expected in the general population (median rates of suicidal ideation = 27.9%, median rates of suicide attempts = 10.3%). Consistent with this, the metaanalysis confirmed that an OCD diagnosis was associated with

11

substantially increased levels of suicidality. Evidence was also obtained that the association between OCD and suicidality pertains across different types of suicidality, including suicidal ideation, and attempts. The only exception to this trend was completed suicides. Two studies examined the impact of OCD on completed suicides among individuals diagnosed with psychoses and major depression, but neither of these studies evidenced a significant contribution of OCD on completed suicides) (Hantouche et al., 2003; Pfeiffer et al., 2009). Additionally, this review demonstrated that the elevated levels of suicidality in OCD tended to coexist with comorbidities (comorbid Axis I diagnoses and the severity of depressive and anxiety symptoms), generic risk factors for suicidality (past history of suicidality), emotion-cognitive risk factors for suicidality (hopelessness) and OCD-specific adversities (the severity of OCD symptoms and mainly obsession symptoms). 4.1. Key research, theoretical and methodological implications A major contribution of this systematic review to the extant literature is that it established a strong link between an OCD diagnosis and suicidality. This review confirms and expands the findings of a growing number of recent studies showing that suicidal thoughts and behaviors in OCD are far more frequent problems than suggested in the past (Peles et al., 2009; Phillips et al., 2007; Sareen et al., 2005; Tavares et al., 2012); patients with OCD have been historically viewed as being at low risk for suicide (Khan, Leventhal, Khan, & Brown, 2002; Torres et al., 2011). The present findings contrast the findings of a previous systematic review which examined the association between anxiety disorders and suicidality (Kanwar et al., 2013). The later review failed to demonstrate a statistically significant association between OCD and suicide risk, but it was based on a small number of OCD studies (four studies), focused exclusively on suicide attempts and excluded studies which comprised OCD participants with non-anxiety comorbidities (Kanwar et al., 2013). Another important contribution of this systematic review is that it provided evidence about the mechanisms which may underlie the association of OCD with suicidality. These mechanisms comprised an amalgam of comorbidity, generic/transdiagnostic and OCD-specific factors, with the evidence being stronger for comorbidity factors. In particular, we found that suicidal OCD individuals experienced a significantly greater severity of depressive and anxiety symptoms compared to non-suicidal OCD individuals. Similarly, the co-occurrence of OCD with other comorbid Axis I diagnoses (mainly depression, bipolar disorder and psychoses) was associated with substantially increased severity of suicidality compared to OCD alone. Thus, these findings suggest that depression and other Axis I diagnoses have a major impact on suicidality in those with OCD. It is uncertain however whether the concurrent presence of psychiatric comorbidities, especially depression, partly or fully account for the elevated levels of suicidality in those with OCD. To date, no study has applied mediation analyses to examine the potential mediating role of depression and other comorbidities in the link between OCD and suicidality. Generic factors such as a past history of suicidality and feelings of hopelessness which have been identified as core risk factors for suicidality across a range of psychiatric populations were also found to be strongly associated with suicidality in OCD (O’Connor & Nock, 2014). However, despite the consistency of the findings, only three studies have currently examined the impact of past history of suicidality and hopelessness on current suicidality in OCD (Alonso et al., 2010; Balci & Sevincok, 2010; Kamath et al., 2007). With regards to OCD specific factors, the findings were less consistent compared to comorbidity and generic factors. The overall severity of OCD symptoms was found to have a significant contribution to suicidality in our meta-analysis but there is also evidence that this contribution is rendered non-significant after accounting for comorbidities in the analysis (Phillips et al., 2007; Sareen et al., 2005). Despite this, obsessions have been found to be more strongly implicated in suicidality compared to compulsions, suggesting that severe obsessions could act as an OCDspecific mechanism leading to suicidality (Dell'Osso et al., 2012; Phillips

12


Study ID Depression severity Alonso 2010 Balci 2010 Hung 2010 Kamath 2007 Moreira 2013 Norton 2007 Sevincok 2007 Tores 2007 Torres 2011 Zisook 2011 Subtotal (I-squared = 39.1%, p = 0.097) . OCD severity Alonso 2010 Balci 2010 Fawzy 2011 Hung 2010 Kamath 2007 Moreira 2013 Norton 2007 Sevincok 2007 Tores 2007 Torres 2011 Subtotal (I-squared = 86.6%, p = 0.000) . Obsessions Alonso 2010 Balci 2010 Kamath 2007 Sevincok 2007 Torres 2011 Subtotal (I-squared = 41.8%, p = 0.143) . Compulsions Alonso 2010b Balci 2010b Kamath 2007b Sevincok 2007b Torres 2011b Subtotal (I-squared = 49.7%, p = 0.093) . Suicidality history Alonso 2010 Balci 2010 Kamath 2007 Subtotal (I-squared = 55.3%, p = 0.107) . Hopelessness Balci 2010 Kamath 2007 Subtotal (I-squared = 0.0%, p = 0.348) . Anxiety severity Balci 2010 Hung 2010 Torres 2011 Subtotal (I-squared = 0.0%, p = 0.795) .NOTE: Weights are from random effects analysis -2

-1

0

1

ES (95% CI)

% Weight

0.97 (0.40, 1.54) 1.39 (0.72, 2.06) 0.50 (0.15, 0.85) 0.70 (0.25, 1.15) 0.54 (0.27, 0.81) 0.86 (0.53, 1.19) 0.53 (-0.04, 1.10) 1.80 (-0.85, 4.45) 0.50 (0.25, 0.75) 0.96 (0.67, 1.25) 0.72 (0.56, 0.89)

6.45 5.00 12.26 9.03 15.78 13.05 6.45 0.38 16.80 14.81 100.00

-0.20 (-0.40, -0.00) 0.73 (0.12, 1.34) 2.11 (1.54, 2.68) 0.01 (-0.34, 0.36) 0.34 (-0.11, 0.79) 0.20 (-0.05, 0.45) 0.30 (-0.01, 0.61) 0.60 (0.03, 1.17) 0.73 (-0.21, 1.67) 0.20 (-0.05, 0.45) 0.44 (0.13, 0.75)

12.06 8.59 8.94 10.91 10.04 11.68 11.24 8.94 5.92 11.68 100.00

-0.19 (-0.76, 0.38) 0.73 (0.12, 1.34) 0.57 (0.12, 1.02) 0.40 (-0.15, 0.95) 0.20 (-0.05, 0.45) 0.32 (0.05, 0.59)

15.33 13.95 20.58 16.08 34.06 100.00

-0.25 (-0.82, 0.32) 0.60 (-0.01, 1.21) 0.16 (-0.29, 0.61) 0.76 (0.19, 1.33) 0.15 (-0.10, 0.40) 0.25 (-0.04, 0.55)

16.20 14.90 20.96 16.20 31.74 100.00

0.74 (-0.02, 1.50) 0.17 (-0.42, 0.76) 0.98 (0.51, 1.45) 0.64 (0.13, 1.16)

25.94 33.84 40.22 100.00

0.94 (0.31, 1.57) 0.57 (0.12, 1.02) 0.70 (0.33, 1.06)

34.06 65.94 100.00

0.64 (0.03, 0.49 (0.14, 0.42 (0.17, 0.46 (0.27,

10.36 30.73 58.91 100.00

1.25) 0.84) 0.67) 0.66)

2

Fig. 7. Differences in the levels of depression severity, OCD severity, obsessions, compulsions, past history of suicidality, hopelessness and anxiety severity between OCD groups with suicidality and OCD groups without suicidality. Note: Meta-analysis of individual studies and pooled effects. Random effects model used. 95% CI = 95% confidence intervals; ES = Hedge’s g.

et al., 2007). One important challenge while examining suicidal behavior in people with OCD is to differentiate the experience of suicidal thoughts and behaviors from being preoccupied with the idea of suicide in an obsessive manner, similar to other types of obsessions related to aggression or sexual issues. It is currently uncertain whether themecongruent obsessions (suicide, death or depression) affect the severity of suicidal behavior in OCD and its progression to severe representations such as suicide plans and attempts. The thematic content of obsessions in relation to suicidal behavior should be addressed by future studies because theme-congruent obsessions could mediate the link between OCD symptoms and suicidal behavior or could comprise a distinct phenomenon which requires a different research and clinical approach. Ultimately, more research is needed to investigate the mechanisms of suicidality in OCD, especially after accounting for the presence of comorbidities in the analyses (particularly depression). Future longitudinal studies applying mediational analyses would be a crucial step for disentangling the psychological architecture of suicidality in OCD. Another noteworthy observation is that none of studies included in this review was guided by a coherent theoretical model of suicide. The work of our research group has focused on examining the applications of two contemporary theories of suicide, namely Cry of Pain Model and Schematic Appraisals Model of Suicide, in a range of clinical groups (Johnson, Gooding, & Tarrier, 2008; Panagioti et al., 2012a; Panagioti,

Gooding, Taylor, & Tarrier, 2013; Panagioti, Gooding, & Tarrier, 2015; Taylor, Gooding, Wood, & Tarrier, 2011).As a result of this theorydriven approach, negative appraisals of defeat and entrapment have repeatedly found to be the main drivers of suicidality among individuals with PTSD, psychoses and para-suicidal individuals (Panagioti et al., 2012a; Taylor et al., 2011). The consistency of the findings across different research groups suggests that these processes could actually be generic and trans-diagnostic (Conceicao Costa et al., 2012) and therefore applicable to OCD. Additionally, other theoretical perspectives of suicide such as the interpersonal psychological theory (particularly the concepts of perceived burdensome and thwarted belongingness) (Joiner et al., 2009), hopelessness theories (Beck, Brown, Berchick, Stewart, & Steer, 1990) and the integrated motivational-volitional perspective of suicide are potentially useful frameworks for understanding suicidality in OCD (O'Connor, 2011). Overall, future studies are encouraged to follow a theory-guided research paradigm in the investigation of the mechanisms of suicidality in OCD. Methodologically, this review was predominately comprised of cross-sectional studies, conducted among diverse research populations using diverse methods for assessing suicidality. Although we endeavored to control for methodological constrains in the analyses, these constraints are indicative of the lack of well-articulated and systematic investigations in this research area. Moreover, the analytical approach


of this review was based on the argument that various forms of suicidality lie on a continuum (from mild suicidal thoughts to severe attempts with a willingness to die and completed suicides) and are underpinned by the same psychological mechanisms (van Os, Linscott, Myin-Germeys, Delespaul, & Krabbendam, 2009). We acknowledge, however, that there is also evidence that completed suicides might represent different phenomena than suicidal ideation and suicide attempts (Beautrais, 2001). To our knowledge, no previous study has examined the rates and mechanisms of completed suicides among individuals with a primary diagnosis of OCD. The examination of whether completed suicides in OCD comprise a clinically important problem is a fruitful research direction. 4.2. Limitations This is the first review which systematically evaluated the association between suicidality and OCD and also was conducted and reported in line with current guidance (Moher et al., 2009). Despite this, a number of limitations should be noted. First, the heterogeneity of the studies included in this review was high. Although random effect models were applied to adjust for between-study variations, uncontrolled factors might alter the strength of the association between OCD and suicidality (Higgins et al., 2003). A second and related limitation is that we only applied basic subgroup and sensitivity analyses to explore potential sources of heterogeneity. Conducting multiple subgroup analyses especially if not planned a priori increases considerably the risk of falsepositive. Cochrane guards against the use of sub-group analysis in the absence of clear theoretical basis or reason which ideally would be stated in the review protocol. A major weakness of sub-group analyses are that by splitting rather than pooling data between groups the comparisons are observational rather than based on randomized groups. Higher numbers of sub-group analyses along these lines can lead to false findings, either negative or positive. Cochrane further add that owing to these limitations that the number of sub-groups included in a review should be kept to a minimum and we have done this by only undertaking one subgroup analysis to examine the link between OCD and different types of suicidality (Deeks, Higgins, & Altman, 2011). We considered this analysis critical for the interpretation of the pooled effect size of the main analysis because there is a strong clinical imperative to understand the association between OCD and the risk of different suicidality. We acknowledge however, that there are a large number of other patient and study level factors along which studies might differ, and it is also possible, in theory, to use meta-regression techniques to explore these factors, but the small number and poor data reporting of the studies included in this review did not allow the use of this technique (Thompson & Higgins, 2002). Additionally we would add that there is a link between heterogeneity and size of meta-analysis and we would expect higher levels given the size of our review (Kontopantelis, Springate, & Reeves, 2013). There is a strong case to pool studies even in the presence of high levels of heterogeneity as long as there is sufficient similarity along clinical and methodological lines and we would suggest that this review meets this standard for meta-analysis (Gøtzsche, 2000). Third, the present review mainly comprised crosssectional studies which preclude any causal inferences to be made. A related point is that the methodological quality of the bulk of studies was low (Thomas et al., 2004). Although a sensitivity analysis based on a small number of higher-quality prospective studies replicated the overall findings, more methodologically succinct studies are needed to investigate suicidality in OCD. Moreover, a number of substantial revisions of the way that OCD is conceptualized have been incorporated in DSM-5 but these developments could not be taken into account in the current analyses since none of the included studies used DSM-5 to assign an OCD diagnosis (the majority of studies used DSM-IV-TR or DSM-IV criteria to diagnose OCD). Two key revisions that may affect future research on the association between OCD and suicidal behavior are that i. OCD is not conceptualized as an anxiety disorder anymore but

13

rather is included in a separate chapter of DSM-5 named as Obsessive Compulsive Related Disorders and ii. two new disorders have been proposed, namely hoarding and excoriation, which were viewed as symptoms of OCD in previous versions of the DSM (APA, 2013). It should be noted, however, that this is an inherent research difficulty in the area of OCD (Torres et al., 2011). 4.3. Clinical implications This review suggests that suicidality in OCD individuals should not be overlooked as a rare incident in clinical practice. Particularly the presence of comorbidities in those with OCD might be indicative of an increased vulnerability to suicidality. With regards to treatment of suicidality, a systematic review has shown that Cognitive Behavioral Therapy (CBT) reduced suicidality irrespective of diagnostic group (Tarrier, Taylor, & Gooding, 2008). More recently, a trial evaluated the efficacy of a suicide prevention intervention that has been intentionally derived from an empirically validated theoretical model of suicide in patients with psychoses who experienced suicidality. This intervention led to significant reductions in proxy measures of suicide including suicidal ideation and suicide probability (Tarrier et al., 2014). Although the suicide prevention intervention has the potential to be applied transdiagnostically and could be beneficial for OCD individuals experiencing suicidal ideation, any recommendations at this stage would be speculative since the evidence about the mechanisms of suicidality in OCD is rudimentary. 4.4. Conclusion In conclusion, this study presented the first comprehensive quantitative synthesis of the literature concerning suicidality and OCD. Our findings suggest that suicidality in those with OCD, especially in OCD subgroups with concurrent comorbidities, is a valid problem which has not yet received appropriate research attention. At present, the depth and quality of research evidence about the psychological drivers of suicidality in patients with OCD are low and hence, there is a dearth of theory-guided prospective studies to improve understanding of suicidality in OCD. Role of funding sources No financial support has been received for this piece of work. Conflict of interest All authors declare no conflict of interest. Contributors IA, MP, PG and NT developed the idea for the study. IA and MP reviewed articles, extracted the data and interpreted the results. IA and MP GR conducted the analysis. IA drafted the manuscript. All authors reviewed and substantively contributed to draft revisions, and approved the final manuscript. Acknowledgements We would to thank Dr. Peter Coventry, Senior Research Fellow in the Centre for Primary at the University of Manchester for his valuable expert advice about the meta-analytic approach. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.cpr.2015.03.002.

14


References Alonso, P., Segalas, C., Real, E., Pertusa, A., Labad, J., Jimenez-Murcia, S., et al. (2010). Suicide in patients treated for obsessive-compulsive disorder: A prospective follow-up study. Journal of Affective Disorders, 124, 300–308. Angst, J., Gamma, A., Neuenschwander, M., Ajdacic-Gross, V., Eich, D., Rössler, W., & Merikangas, K. R. (2005). Prevalence of mental disorders in the Zurich Cohort Study: a twenty year prospective stud. Epidemiologia e Psichiatria Sociale, 14, 68–76. American Psychiatric Association (2004). Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR). Washington, DC: American Psychiatric Association. American Psychiatric Association (2013). Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V). Washington DC: American Psychiatric Association. Anderson, K.E., Gehl, C.R., Marder, K.S., Beglinger, L.J., & Paulsen, J.S. (2010). Comorbidities of obsessive and compulsive symptoms in huntington's disease. Journal of Nervous and Mental Disease, 198, 334–338. Angst, J., Gamma, A., Endrass, J., Goodwin, R., Ajdacic, V., Eich, D., et al. (2004). Obsessivecompulsive severity spectrum in the community: prevalence, comorbidity, and course. European Archives of Psychiatry and Clinical Neuroscience, 254, 156–164. Balci, V., & Sevincok, L. (2010). Suicidal ideation in patients with obsessive-compulsive disorder. Psychiatry Research, 175, 104–108. Beautrais, A.L. (2001). Suicides and serious suicide attempts: two populations or one? Psychological Medicine, 31, 837–845. Beck, A.T., Brown, G., Berchick, R.J., Stewart, B.L., & Steer, R.A. (1990). Relationship between Hopelessness and Ultimate Suicide – a Replication with Psychiatric Outpatients. American Journal of Psychiatry, 147, 190–195. Benvenuti, A., Rucci, P., Sbrana, A., Ravani, L., Massei, G.J., Gonnelli, C., et al. (2009). Obsessive-compulsive spectrum and suicidality in schizophrenia and mood disorders. Italian Journal of Psychopathology, 15, 282–289. Blakemore, A., Dickens, C., Guthrie, E., Bower, P., Kontopantelis, E., Afzal, C., et al. (2014). Depression and anxiety predict health-related quality of life in chronic obstructive pulmonary disease: systematic review and meta-analysis. International Journal of Chronic Obstructive Pulmonary Disease, 9, 501–512. Bloch, M.H., Landeros-Weisenberger, A., Rosario, M.C., Pittenger, C., & Leckman, J.F. (2008). Meta-analysis of the symptom structure of obsessive-compulsive disorder. American Journal of Psychiatry, 165, 1532–1542. Bolton, J.M., Cox, B.J., Afifi, T.O., Enns, M.W., Bienvenu, O.J., & Sareen, J. (2008). Anxiety disorders and risk for suicide attempts: Findings from the Baltimore epidemiologic catchment area follow-up study. Depression and Anxiety, 25, 477–481. Borenstein, M., Hedges, L., Higgins, J., & Rothstein, H. (2005). Comprehensive meta-analysis. In (2.2.034 Ed.). Englewood, NJ: Biostat. Centre for Reviews and Dissemination University of York Systematic Reviews (2009). pp. CRD’s guidance for undertaking reviews in health care. Available from: http://www. york.ac.uk/inst/crd/pdf/Systematic_Reviews.pdf Chakraborty, V., Cherian, A.V., Math, S.B., Venkatasubramanian, G., Thennarasu, K., Mataix-Cols, D., et al. (2012). Clinically significant hoarding in obsessivecompulsive disorder: Results from an Indian study. Comprehensive Psychiatry, 53, 1153–1160. Chen, Y.W., & Dilsaver, S.C. (1995). Comorbidity for obsessive-compulsive disorder in bipolar and unipolar disorders. Psychiatry Research, 59, 57–64. Conceicao Costa, D.L., ChagasAssuncao, M., ArzenoFerrao, Y., ArchettiConrado, L., Hajaj Gonzalez, C., Franklin Fontenelle, L., et al. (2012). Body dysmorphic disorder in patients with obsessive-compulsive disorder: Prevalence and clinical correlates. Depression and Anxiety, 29, 966–975. Crino, R., Slade, T., & Andrews, G. (2005). The changing prevalence and severity of obsessive-compulsive disorder criteria from DSM-III to DSM-IV. American Journal of Psychiatry, 162, 876–882. De Haan, L., Sterk, B., Wouters, L., & Linszen, D.H. (2013). The 5-year course of obsessivecompulsive symptoms and obsessive-compulsive disorder in first-episode schizophrenia and related disorders. Schizophrenia Bulletin, 39, 151–160. Deeks, J.J., Higgins, J.P.T., & Altman, D.G. (2011). Chapter 9: Analysing data and undertaking meta-analyses. In J.P.T. Higgins, & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of InterventionsVersion 5.1.0 (updated March 2011). The Cochrane Collaboration (2011.Available from www.cochrane-handbook.org). Dell'Osso, L., Casu, G., Carlini, M., Conversano, C., Gremigni, P., & Carmassi, C. (2012). Sexual obsessions and suicidal behaviors in patients with mood disorders, panic disorder and schizophrenia. Annals of General Psychiatry, 11, 27. DeVylder, J.E., Oh, A.J., Ben-David, S., Azimov, N., Harkavy-Friedman, J.M., & Corcoran, C.M. (2012). Obsessive compulsive symptoms in individuals at clinical risk for psychosis: Association with depressive symptoms and suicidal ideation. Schizophrenia Research, 140, 110–113. Egger, M., Smith, G.D., Schneider, M., & Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. British Medical Journal, 315, 629–634. Fawzy, N., & Nashim, H. (2010). Prevalence and risk factors of suicide among patients with obsessive–compulsive disorder. Middle East Current Psychiatry, 18, 18–22. Fineberg, N.A., Hengartner, M.P., Bergbaum, C., Gale, T., Rossler, W., & Angst, J. (2013). Lifetime comorbidity of obsessive-compulsive disorder and sub-threshold obsessivecompulsive symptomatology in the community: Impact, prevalence, sociodemographic and clinical characteristics. International Journal of Psychiatry in Clinical Practice, 17, 188–196. Fontenelle, L.F., Cocchi, L., Harrison, B.J., Shavitt, R.G., do Rosario, M.C., Ferrao, Y.A., et al. (2012). Towards a post-traumatic subtype of obsessive-compulsive disorder. Journal of Anxiety Disorders, 26, 377–383. Gentil, A.F., Mathis, M.A. d., Torresan, R.C., Diniz, J.B., Alvarenga, P., do Rosario, M.C., et al. (2009). Alcohol use disorders in patients with obsessive-compulsive disorder: The

importance of appropriate dual-diagnosis. Drug and Alcohol Dependence, 100, 173–177. Goes, F.S., McCusker, M.G., Bienvenu, O.J., Mackinnon, D.F., Mondimore, F.M., Schweizer, B., et al. (2012). Co-morbid anxiety disorders in bipolar disorder and major depression: familial aggregation and clinical characteristics of co-morbid panic disorder, social phobia, specific phobia and obsessive-compulsive disorder. Psychological Medicine, 42, 1449–1459. Goodwin, R., Koenen, K.C., Hellman, F., Guardino, M., & Struening, E. (2002). Helpseeking and access to mental health treatment for obsessive-compulsive disorder. Acta Psychiatrica Scandinavica, 106, 143–149. Gøtzsche, P.C. (2000). Why we need a broad perspective on meta-analysis. British Medical Journal, 321, 585. Hagen, K., Hansen, B., Joa, I., & Larsen, T.K. (2013). Prevalence and clinical characteristics of patients with obsessive-compulsive disorder in first-episode psychosis. BMC Psychiatry, 13. Hantouche, E.G., Angst, J., Demonfaucon, C., Perugi, G., Lancrenon, S., & Akiskal, H.S. (2003). Cyclothymic OCD: A distinct form? Journal of Affective Disorders, 75, 1–10. Higgins, J.P.T., Thompson, S.G., Deeks, J.J., & Altman, D.G. (2003). Measuring inconsistency in meta-analyses. British Medical Journal, 327, 557–560. Hung, T. -C., Tang, H. -S., Chiu, C. -H., Chen, Y. -Y., Chou, K. -R., Chiou, H. -C., et al. (2010). Anxiety, depressive symptom and suicidal ideation of outpatients with obsessive compulsive disorders in Taiwan. Journal of Clinical Nursing, 19, 3092–3101. Johnson, J., Gooding, P., & Tarrier, N. (2008). Suicide risk in schizophrenia: Explanatory models and clinical implications, the schematic appraisal model of suicide (SAMS). Psychology and Psychotherapy-Theory Research and Practice, 81, 55–77. Joiner, T.E., Jr., Van Orden, K.A., Witte, T.K., Selby, E.A., Ribeiro, J.D., Lewis, R., et al. (2009). Main predictions of the interpersonal-psychological theory of suicidal behavior: empirical tests in two samples of young adults. Journal of Abnormal Psychology, 118, 634–646. Kamath, P., Reddy, Y.C.J., & Kandavel, T. (2007). Suicidal behavior in obsessive-compulsive disorder. Journal of Clinical Psychiatry, 68, 1741–1750. Kanwar, A., Malik, S., Prokop, L.J., Sim, L.A., Feldstein, D., Wang, Z., et al. (2013). The association between anxiety disorders and suicidal behaviors: A systematic review and meta-analysis. Depression and Anxiety, 30, 917–929. Kessler, R.C., Chiu, W., Demler, O., & Walters, E.E. (2005). Prevalence, severity, and comorbidity of 12-month dsm-iv disorders in the national comorbidity survey replication. Archives of General Psychiatry, 62, 617–627. Khan, A., Leventhal, R.M., Khan, S., & Brown, W.A. (2002). Suicide risk in patients with anxiety disorders: a meta-analysis of the FDA database. Journal of Affective Disorders, 68, 183–190. Kontopantelis, E., Springate, D.A., & Reeves, D. (2013). A Re-analysis of the Cochrane library data: the dangers of unobserved heterogeneity in meta-analyses. PLoS One, 8, e69930. Koyuncu, A., Tukel, R., Ozyildirim, I., Meteris, H., & Yazici, O. (2010). Impact of obsessivecompulsive disorder comorbidity on the sociodemographic and clinical features of patients with bipolar disorder. Comprehensive Psychiatry, 51, 293–297. Kruger, S., Braunig, P., & Cooke, R.G. (2000). Comorbidity of obsessive-compulsive disorder in recovered inpatients with bipolar disorder. Bipolar Disorders, 2, 71–74. Levy, H.C., McLean, C.P., Yadin, E., & Foa, E.B. (2013). Characteristics of individuals seeking treatment for obsessive-compulsive disorder. Behavior Therapy, 44, 408–416. Lochner, C., Fineberg, N.A., Zohar, J., van Ameringen, M., Juven-Wetzler, A., Altamura, A.C., et al. (2014). Comorbidity in obsessive–compulsive disorder (OCD): A report from the International College of Obsessive–Compulsive Spectrum Disorders (ICOCS). Comprehensive Psychiatry, 55(7), 1513–1519. Magalhaes, P.V.S., Kapczinski, N.S., & Kapczinski, F. (2010). Correlates and impact of obsessive-compulsive comorbidity in bipolar disorder. Comprehensive Psychiatry, 51, 353–356. Mahasuar, R., Janardhan Reddy, Y.C., & Math, S.B. (2011). Obsessive-compulsive disorder with and without bipolar disorder. Psychiatry and Clinical Neurosciences, 65, 423–433. Maina, G., Salvi, V., Tiezzi, M.N., Albert, U., & Bogetto, F. (2007). Is OCD atrisk for suicide? A case–control study. Clinical Neuropsychiatry, 4, 117–121. Marcks, B.A., Weisberg, R.B., Dyck, I., & Keller, M.B. (2011). Longitudinal course of obsessive-compulsive disorder in patients with anxiety disorders: a 15-year prospective follow-up study. Comprehensive Psychiatry, 52, 670–677. Mohammadi, M.R., Ghanizadeh, A., Rahgozart, M., Noorbala, A.A., Malekafzali, H., Davidian, H., et al. (2005). Suicidalattemptandpsychiatric disorders in Iran. Suicide and Life-Threatening Behavior, 35, 309–316. Moher, D., Liberati, A., Tetzlaff, J., & Altman, D.G. (2009). Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statementVol. 339, . Moreira, L., Bins, H., Toressan, R., Ferro, C., Harttmann, T., Petribu, K., et al. (2013). An exploratory dimensional approach to premenstrual manifestation of obsessivecompulsive disorder symptoms: A multicentre study. Journal of Psychosomatic Research, 74, 313–319. Norton, P.J., Temple, S.R., & Pettit, J.W. (2008). Suicidal ideation and anxiety disorders: Elevated risk or artifact of comorbid depression? Journal of Behavior Therapy and Experimental Psychiatry, 39, 515–525. O'Connor, R.C. (2011). Towards an integrated motivational–volitional model of suicidal behaviour. International handbook of suicide prevention (pp. 181–198). John Wiley & Sons, Ltd. O’Connor, R., & Nelson, M. K. (2014). The psychology of suicidal behaviour. Lancet Psychiatry, 1, 73–85. Pallanti, S., Grassi, G., Sarrecchia, E.D., Cantisani, A., & Pellegrini, M. (2011). Obsessive– compulsive disorder comorbidity: clinical assessment and therapeutic implications. Frontiers in Psychiatry, 2, 70. Panagioti, M., Gooding, P.A., & Tarrier, N. (2012a). Hopelessness, defeat, and entrapment in posttraumatic stress disorder: their association with suicidality and severity of depression. Journal of Nervous and Mental Disease, 200, 676–683.

I. Angelakis et al. / Clinical Psychology Review 39 (2015) 1–15 Panagioti, M., Gooding, P.A., & Tarrier, N. (2012b). A meta-analysis of the association between posttraumatic stress disorder and suicidality: the role of comorbid depression. Comprehensive Psychiatry, 53, 915–930. Panagioti, M., Gooding, P.A., & Tarrier, N. (2015). A prospective study of suicidal ideation in posttraumatic stress disorder: the role of perceptions of defeat and entrapment. Journal of Clinical Psychology, 71, 50–61. Panagioti, M., Gooding, P., Taylor, P.J., & Tarrier, N. (2013). A model of suicidality in posttraumatic stress disorder (PTSD): The mediating role of defeat and entrapment. Psychiatry Research, 209, 55–59. Peles, E., Adelson, M., & Schreiber, S. (2009). Association of OCD with a history of traumatic events among patients in methadone maintenance treatment. CNS Spectrums, 14, 547–554. Perugi, G., Akiskal, H.S., Pfanner, C., Presta, S., Gemignani, A., Milanfranchi, A., et al. (1997). The clinical impact of bipolar and unipolar affective comorbidity on obsessivecompulsive disorder. Journal of Affective Disorders, 46, 15–23. Pfeiffer, P.N., Ganoczy, D., Ilgen, M., Zivin, K., & Valenstein, M. (2009). Comorbid anxiety as a suicide risk factor among depressed veterans. Depression and Anxiety, 26, 752–757. Phillips, K.A., Pinto, A., Menard, W., Eisen, J.L., Mancebo, M., & Rasmussen, S.A. (2007). Obsessive-compulsive disorder versus body dysmorphic disorder: A comparison study of two possibly related disorders. Depression and Anxiety, 24, 399–409. Rosenthal, R. (1979). The file drawer problem and tolerance for null results. Psychological Bulletin, 86, 638–641. Ruscio, A.M., Stein, D.J., Chiu, W.T., & Kessler, R.C. (2010). The epidemiology of obsessivecompulsive disorder in the National Comorbidity Survey Replication. Molecular Psychiatry, 15, 53–63. Sallet, P.C., De Alvarenga, P.G., Ferrao, Y., De Mathis, M.A., Torres, A.R., Marques, A., et al. (2010). Eating disorders in patients with obsessive-compulsive disorder: Prevalence and clinical correlates. International Journal of Eating Disorders, 43, 315–325. Sareen, J., Cox, B.J., Afifi, T.O., De Graaf, R., Asmundson, G.J.G., Ten Have, M., et al. (2005). Anxiety disorders and risk for suicidal ideation and suicide attempts: A populationbased longitudinal study of adults. Archives of General Psychiatry, 62, 1249–1257. Sevincok, L., Akoglu, A., & Kokcu, F. (2007). Suicidality in schizophrenic patients with and without obsessive-compulsive disorder. Schizophrenia Research, 90, 198–202. Tarrier, N., Taylor, K., & Gooding, P. (2008). Cognitive-behavioral interventions to reduce suicide behavior: a systematic review and meta-analysis. Behavior Modification, 32, 77–108. Tarrier, N., Kelly, J., Maqsood, S., Snelson, N., Maxwell, J., Law, H., Dunn, G., & Gooding, P. (2014). The cognitive behavioural prevention of suicide in psychosis: a clinical trial. Schizophrenia Research, 156, 204–210. Tavares, D., Quevedo, L., Jansen, K., Souza, L., Pinheiro, R., & Silva, R. (2012). Prevalence of suicide risk and comorbidities in postpartum women in Pelotas. Revista Brasileira de Psiquiatria, 34, 270–276.

15

Taylor, P.J., Gooding, P., Wood, A.M., & Tarrier, N. (2011). The role of defeat and entrapment in depression, anxiety, and suicide. Psychological Bulletin, 137, 391–420. ten Have, M., de Graaf, R., van Dorsselaer, S., Verdurmen, J., van't Land, H., Vollebergh, W., et al. (2009). Incidence and course of suicidal ideation and suicide attempts in the general population. Canadian Journal of Psychiatry-Revue Canadienne De Psychiatrie, 54, 824–833. Thomas, B.H., Ciliska, D., Dobbins, M., & Micucci, S. (2004). A process for systematically reviewing the literature: providing the research evidence for public health nursing interventions. Worldviews on Evidence-Based Nursing, 1, 176–184. Thompson, S.G., & Higgins, J.P.T. (2002). How should meta-regression analyses be undertaken and interpreted? Statistics in Medicine, 21, 1559–1573. Torres, A.R., Prince, M.J., Bebbington, P.E., Bhugra, D., Brugha, T.S., Farrell, M., et al. (2006). Obsessive-compulsive disorder: Prevalence, comorbidity, impact, and help-seeking in the British National Psychiatric Morbidity Survey of 2000. American Journal of Psychiatry, 163, 1978–1985. Torres, A.R., Ramos-Cerqueira, A.T.A., Ferrao, Y.A., Fontenelle, L.F., Do Rosario, M.C., & Miguel, E.C. (2011). Suicidality in obsessive-compulsive disorder: Prevalence and relation to symptom dimensions and comorbid conditions. Journal of Clinical Psychiatry, 72, 17–26. Torres, A.R., Ramos-Cerqueira, A.T.D.A., Torresan, R.C., Domingues, M.D.S., Hercos, A.C.R., & Guimaraes, A.B.C. (2007). Prevalence and associated factors for suicidal ideation and behaviors in obsessive-compulsive disorder. CNS Spectrums, 12, 771–778. Torres, A.R., Shavitt, R.G., Torresan, R.C., Ferrao, Y.A., Miguel, E.C., & Fontenelle, L.F. (2013). Clinical features of pure obsessive-compulsive disorder. Comprehensive Psychiatry, 54, 1042–1052. van Os, J., Linscott, R.J., Myin-Germeys, I., Delespaul, P., & Krabbendam, L. (2009). A systematic review and meta-analysis of the psychosis continuum: evidence for a psychosis proneness-persistence-impairment model of psychotic disorder. Psychological Medicine, 39, 179–195. Veale, D., & Roberts, A. (2014). Obsessive-compulsive disorder. BMJ, 348, g2183. Viswanath, B., Narayanaswamy, J.C., Rajkumar, R.P., Cherian, A.V., Kandavel, T., Math, S.B., et al. (2012). Impact of depressive and anxiety disorder comorbidity on the clinical expression of obsessive-compulsive disorder. Comprehensive Psychiatry, 53, 775–782. Whitters, A.C., Cadoret, R.J., & Widmer, R.B. (1985). Factors associated with suicide attempts in alcohol abusers. Journal of Affective Disorders, 9, 19–23. Zisook, S., Lesser, I.M., Lebowitz, B., Rush, A.J., Kallenberg, G., Wisniewski, S.R., et al. (2011). Effect of antidepressant medication treatment on suicidal ideation and behavior in a randomized trial: An exploratory report from the combining medications to enhance depression outcomes study. Journal of Clinical Psychiatry, 72, 1322–1332.

Buspirone induced prolactin responses in obsessive-compulsive disorder (OCD): is OCD a 5-HT2 receptor disorder?

OCD: obsessive-compulsive … disgust? The role of disgust in obsessive-compulsive disorder.

Reported sleep patterns in obsessive compulsive disorder (OCD).

Fluoxetine: no association with suicidality in obsessive-compulsive disorder.

Diffusion tensor imaging (DTI) studies in patients with obsessive-compulsive disorder (OCD): a review.

Trauma-related obsessive-compulsive disorder: a review.

Neurobiology of obsessive compulsive disorder: a review.

Treatment of comorbid bipolar disorder and obsessive-compulsive disorder: a systematic review.

Genome Wide Association Study (GWAS) between Attention Deficit Hyperactivity Disorder (ADHD) and Obsessive Compulsive Disorder (OCD).

Pharmacological treatment strategies in obsessive compulsive disorder: A cross-sectional view in nine international OCD centers.

Group cognitive behaviour therapy in obsessive-compulsive disorder (OCD): a controlled study.

Cigarette smoking in obsessive-compulsive disorder and unaffected parents of OCD patients.

Atypical antipsychotic augmentation in SSRI treatment refractory obsessive-compulsive disorder: a systematic review and meta-analysis.

Tic-related obsessive-compulsive disorder (OCD): phenomenology and treatment outcome in the Pediatric OCD Treatment Study II.

The relationship between obsessive-compulsive personality disorder traits, obsessive-compulsive disorder and excessive exercise in patients with anorexia nervosa: a systematic review.

Obsessive-compulsive disorder (OCD): Practical strategies for pharmacological and somatic treatment in adults.

Obsessive-compulsive disorder and female reproductive cycle events: results from the OCD and reproduction collaborative study.

Integration of cortical brain stimulation and exposure and response prevention for obsessive-compulsive disorder (OCD).

Obsessive-Compulsive Personality Traits in Youth with Obsessive-Compulsive Disorder.

Assessment of Quality of Life (QOL) in Obsessive Compulsive Disorder (OCD) and Dysthymic Disorder (DD): A Comparative Study.

Neuropsychological investigations in obsessive-compulsive disorder: A systematic review of methodological challenges.

Increased decision thresholds enhance information gathering performance in juvenile Obsessive-Compulsive Disorder (OCD).

Obsessive-compulsive disorder.

Are 5-HT3 antagonists effective in obsessive-compulsive disorder? A systematic review of literature.