Faculty Affiliations and Disclosures
Dr. Torres is associate professor in the Department of Neurology and Psychiatry at Botucatu Medical School–Universidade Estadual Paulista “Julio de Mesquita Filho”–São Paulo State University (UNESP) in Botucatu, Brazil. Dr. Ramos-Cerqueira is associate professor in the Department of Neurology and Psychiatry at Botucatu Medical School–UNESP. Dr. Torresan is psychiatrist and postgraduate student in the Department of Public Health at Botucatu Medical School–UNESP. Dr. Domingues is psychiatrist and former resident in the Department of Neurology and Psychiatry at Botucatu Medical School. Ms. Hercos and Mr. Guimarães are undergraduate students at Botucatu Medical School–UNESP.
Disclosures: The authors do not have an affiliation with or financial interest in any organization that might pose a conflict of interest.
Acknowledgment: The authors thank Ana Paula M. Negreiros, Caroline N. Vitorino, and Anna Paola V. Chiarelli, MD, for their assistance during the interview process.
Submitted for publication: July 23, 2007; Accepted for publication: September 14, 2007.
Please direct all correspondence to: Albina Rodrigues Torres, MD, PhD, Departamento de Neurologia e Psiquiatria, Faculdade de Medicina de Botucatu–UNESP, Distrito de Rubião Jr., 18618-970, Botucatu, SP, Brazil; Tel: 55-14-3811-6260/3811-6089, Fax: 55-4-3815-5965; E-mail: email@example.com.
Introduction: Patients with obsessive-compulsive disorder (OCD) have historically been considered at low risk for suicide, but recent studies are controversial.
Objective: To study the prevalence of suicidal thoughts and attempts in OCD patients and to compare those with and without suicidality according to demographic and clinical variables.
Methods: Fifty outpatients with primary OCD (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition) from a Brazilian public university were evaluated. The Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) was used to assess OCD severity, the Beck Depression Inventory to evaluate depressive symptoms and the Alcohol Use Disorders Identification Test to assess alcohol problems.
Results: All patients had obsessions and compulsions, 64% a chronic fluctuating course and 62% a minimum Y-BOCS score of 16. Half of the patients presented relevant depressive symptoms, but only three had a history of alcohol problems. Seventy percent reported having already thought that life was not worth living, 56% had wished to be dead, 46% had suicidal ideation, 20% had made suicidal plans, and 10% had already attempted suicide. Current suicidal ideation occurred in 14% of the sample and was significantly associated with a Y-BOCS score >16. Previous suicidal thoughts were associated with a Beck Depression Inventory score >19.
Conclusion: Suicidality has been underestimated in OCD and should be investigated in every patient, so that appropriate preventive measures can be taken.
Obsessive-compulsive disorder (OCD) has a lifetime prevalence of ~2%1,2 and is characterized by the occurrence of obsessions and/or compulsions. These symptoms are time consuming and have a negative impact in the individual’s daily activities and in family and social relationships. 3
Historically, OCD patients have been considered at low risk for suicide, due to the small number of occurrences described in clinical samples. Most studies have shown that <1% of these patients committed suicide,4 even though suicide is a common preoccupation among individuals with OCD, usually as an egodystonic fear of self-destructive impulses. Hence, suicide has been considered to be infrequent in OCD because these individuals are often vigilant on their aggressive obsessions and harm avoidant.5 In a study conducted by Coryell,6 there were no cases of suicide, but most studies had small samples and did not aim specifically at studying this question.
Recently, a meta-analysis7 on suicide risk in individuals with anxiety disorders found a suicide attempt rate of only 1.5% (14/954) for OCD patients. Koran and colleagues8 found 3% of suicide attempters among 60 OCD outpatients, a rate that did not differ substantially from the 2% rate of the general United States population. However, Hollander and colleagues9 reported that >90% of OCD patients of the Obsessive Compulsive Foundation had low self-esteem, 57% had previously experienced suicidal ideation and 12.2% had already attempted suicide. Approximately 18% were alcohol abusers and 13% drug users, both problems considered to be risk factors for suicide.10 In a secondary analysis of data from the Epidemiological Catchment Area study, Hollander and colleagues5 found that participants with uncomplicated OCD had significantly higher rate of suicide attempts than those without any psychiatric disorder (3.6% vs 0.9%), also that individuals with comorbid OCD had a higher rate than participants with other psychiatric disorders (15% vs 7%).
The most common comorbid condition in OCD is depressive disorder,11,12 which may worsen the sufferers’ limitations and quality of life,13 but the impact of this comorbidity on suicidal risk is unclear. An epidemiological study conducted in Great Britain in 200014 found that 26% of OCD patients in the community had at least one lifetime suicidal attempt and that the prevalence did not differ significantly among those with and without other comorbid neuroses, indicating an independent association between obsessive-compulsive (OC) symptoms and suicidal behaviors.
Another population-based study in the Nertherlands15 failed to find an independent association between OCD and suicidal ideation and attempts after adjusting for sociodemographic factors and other comorbid mental disorders; even though it demonstrated that a pre-existing anxiety disorder, in general, is an independent risk factor for suicidality. Furthermore, a clinical study with adolescent inpatients16 found a lower rate of suicidal attempts in OCD patients compared with individuals with other mental disorders (10% and 35%, respectively) and, curiously, an inverse relationship between suicidal behaviors and depression in OCD patients.
Besides these controversial results, studies about suicidality in OCD are scant in the international literature and nonexistent in developing countries, including Brazil. Therefore, the aims of the present study were to study the prevalence of suicidal ideation and attempts in adults with OCD from a public university outpatient clinic in Brazil and to compare OCD patients with and without suicidal ideation and/or attempts according to their sociodemographic and clinical characteristics.
A cross-sectional study was conducted, comparing OCD patients with and without suicidal thoughts or behaviors according to variables that are possible risk factors for these outcomes.
Fifty adult outpatients (>18 years of age) from the University Clinical Hospital of Botucatu Medical School–Universidade Estadual Paulista “Julio de Mesquita Filho” who met lifetime Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria3 for OCD as the main psychiatric diagnosis. The only exclusion criterion was having any serious psychiatric comorbid condition, such as psychotic disorders, dementia, or mental retardation. All participants were using anti-obsessive medications. All subjects voluntarily agreed to participate in the study, which was approved by the university’s ethical research committee, after being fully informed about its purposes and methods, and signed a written informed consent of participation.
Trained research assistants applied the instruments of assessment below. Sociodemographic and clinical variables, including suicidal thoughts, plans, and attempts were assessed using a specifically created questionnaire (available upon request).
The severity of obsessive-compulsive (OC) symptoms was accessed using the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS).17,18 This instrument has five items about obsessions and five about compulsions, each with a score ranging from 0 (no symptom) to 4 (very severe symptoms) and a maximum total score of 40. The cut-off score for clinically significant symptoms is >16.
The presence and severity of depressive symptoms were measured with the Beck Depression Inventory (BDI),19 which consists of 21 items, each ranging from 0–3. The cut-off points for depressive symptoms are <10 (inexistent or minimal); 10–18 (mild to moderate); 19–29 (moderate to severe), and 30–63 (severe). The threshold to differentiate cases and non-cases of clinically significant depression was 18/19.
Hazardous alcohol use or dependence was assessed through The Alcohol Use Disorders Identification Test,20 which has 10 questions about the use of alcoholic beverages in the previous year (maximum score of 40). The threshold that indicates hazardous alcohol use is >8.
The statistical analysis was performed using the STATA 8.0 software.21 A descriptive analysis was conducted, followed by a univariate analysis of the associations between the independent variables and the presence of suicidal thoughts or attempts (outcomes), using the χ2 or Fisher exact tests as appropriate for categorical variables and the t-test for continuous variables. A standard significance level of P<.05 was adopted.
Demographics and Clinical Characteristics
Most of the patients’ demographic characteristics are described in Table 1. Clinical characteristics are shown in Table 2.
Sixty-four percent of patients had qualified occupations and a personal income of at least one minimum wage per month (66%), and the most frequent religion was Roman Catholic (54%). All subjects had obsessions and compulsions, 68% had been previously treated in other health services, but just three had been previously hospitalized in a psychiatric ward for treatment (two of them in partial hospitalization services). At the time of interview, 62% of the sample still had clinically significant OC symptoms (Y-BOCS total score >16), while 38% were quite improved (Y-BOCS score <15). Drinking problems in the previous year occurred in only three participants, while current relevant depressive symptoms (BDI score >19) were present in 52% of the sample (Table 2). Most patients (74%) with Y-BOCS score >16 had also clinically relevant depressive symptoms (BDI score >19). In other words, there was a highly significant association (P<.001) between OCD severity and depressive symptoms severity, considered either as categorical or continuous variables.
Prevalence of Lifetime Suicidal Thoughts and Behaviors and Family History of Suicidal Acts
Thirty-five patients (70%) described that had already thought that life was not worth living, 56% had wished to be dead, 46% had suicidal thoughts, 20% had made suicidal plans, and five (10%) had attempted suicide. Two individuals had attempted just once and one had attempted three times. All five patients received medical treatment after the attempts, but just two of them required hospitalization. Current suicidal ideation was present in 14% (n=7) of the patients. Family history of suicidal attempts and completed suicide occurred in 14% and 12% of the sample, respectively.
Risk Factors for Suicidality
Gender was not significantly associated with any of the investigated suicidal aspects. Lifetime suicidal ideation in particular was not associated with marital status, family arrangements, levels of education, occupation or income, religious practice, self-evaluation of health, OCD clinical course (either episodic/chronic fluctuating or chronic stable/deteriorating), age at OCD symptoms onset (<12, 13–17, and >18 years of age) or with family history of suicidal attempts or completed suicide. In fact, previous suicidal ideation was only associated with severity of current depressive symptoms (BDI score >19), while current suicidal ideation was associated with severity of OC symptoms (Y-BOCS score >16) (Table 3). All five patients who had previous attempts had a worse self-evaluation of health (regular, bad, or very bad). This was the only variable significantly associated (P=.01) with suicidal attempts.
The severity of OC symptoms (Y-BOCS score >16) was also significantly associated with having had thoughts that life was not worthwhile (P=.036) and having already wished to be dead (P=.033). The total Y-BOCS score (as a continuous variable) was also significantly associated with having had thoughts that life was not worth living (P=.042), and with past and current suicidal ideation (P=.012 and P=.006, respectively) but not with previous suicidal plans or attempts.
The Y-BOCS obsessions subscale severity was significantly associated with lifetime and current suicidal ideation (P=.009 and P=.004, respectively) and tended to be associated with having thought that life was not worth living (P=.053). The compulsions subscale severity, on the other hand, was associated with previous thoughts that life was not worthwhile (P=.043) and current suicidal ideation (P=.001) but only tended to be associated with previous suicidal thoughts (P=.052).
Depressive symptoms severity (BDI total score) was also associated with all investigated aspects of suicidality (having thought that life was not worthwhile, having wished to be dead, having had suicidal thoughts, having made suicidal plans and attempts, and having had current suicidal ideation). Only the association with “having already wished to be dead” was at a trend level.
The severity of OC symptoms (Y-BOCS score considered as a dichotomous and continuous variable) was highly associated with depressive symptoms severity (P<.001), which was also investigated both categorically and continuously.
This is the first Brazilian study to systematically investigate “suicidality” in patients with DSM-IV-defined OCD, taking into account sociodemographic characteristics and clinical features of the disorder, such as age of symptoms onset, clinical course, and severity of symptoms.
Some limitations of the present study should also be highlighted. The small sample size may have limited the study power for some of the analyses and the specific content of OC symptoms, which could have a differential impact on suicidal thoughts and acts, was not assessed. It is known that OCD is heterogeneous and maybe different dimensions of symptoms23 have different suicidality risk; however, this aspect was not investigated. Yet, comorbidity evaluation was restricted to depressive symptoms and alcohol problems, which are known to be associated with suicidality, but other comorbid diagnoses were not evaluated. Tic-related OCD, for example, which is a possible subtype that overlaps with early onset and has the poorest prognosis, was not assessed.23 It is possible that the number and types of comorbid Axis I and II disorders influence past and present suicidality. Our sample is from a university treatment program, which usually includes more severe cases, and cannot be generalized to other clinical settings or to community samples.
It is noteworthy that 70% of our sample had previously thought that life was not worth living, 56% had wished to be dead, 46% had suicidal ideation, 20% had made suicidal plans, and 10% had attempted suicide. These aspects are distinct from obsessive fears and doubts about possible self-destructive behaviors, which have an egodystonic nature, usually leading to more vigilance over oneself and not to higher suicidal risk.22 This distinction between “suicidality” and self-aggressive obsessions was carefully investigated in the present study. Considering that past behavior predicts future behavior and that previous suicidal thoughts, emotions and actions predict future “suicidality”, including ideation, intent, attempts, and completions,24 our findings highlight that suicidal risk in OCD is probably not as low as it has been considered thus far in clinical practice. Interestingly, a recent study25 comparing schizophrenia patients with and without comorbid OCD found that the former group was more likely to have a previous history of suicidal ideation and attempts, suggesting that OC symptoms may account for the emergence of suicidality in patients with schizophrenia and OCD.
Our findings show a much higher lifetime prevalence of suicidality in OCD sufferers compared with a recent community survey that reported 17.1% of suicidal thoughts, 4.8% of suicidal plans, and 2.8% of suicidal attempts in the general population of Campinas, a city also located in the Brazilian state of São Paulo.26,27
The fact that seven individuals had current suicidal ideation is noticeable, especially if we take into account that one third of the sample was quite improved with treatment, as shown by a Y-BOCS score <16. Also, current suicidal thoughts were only associated with OC symptoms severity (P=.035), while previous suicidal thoughts were only associated with relevant depressive symptoms (P=.02). However, considering the small difference in number of individuals with past and current suicidal ideation associated with relevant depressive or obsessive symptoms (Table 3), one can suppose that these statistically significant differences are due to the sample size and do not necessarily reflect a clinically significant difference. An important aspect is the high association between OC and depressive symptoms severity. Although the study design does not allow any inferences about causal relationships, depression severity may be just an indicator of OCD severity, situated in the causal pathway of suicidal thoughts, a consequence of OCD and not a confounding variable. Further studies with larger samples are needed in order to clarify whether depression is an independent variable regarding suicidality in OCD or not.
All other demographic and clinical variables of the patients were not significantly associated with either suicidal ideas or acts but possibly our sample size was too small to detect significant differences (type 2 error).
Our study confirms the high prevalence of clinically significant depressive symptoms in OCD patients11,28-30 and also the rarity of alcohol problems in clinical samples.31,32 It should be noted that only one of the three individuals with alcohol hazardous use in our sample had previous suicidal ideation and none had experienced suicide attempts. Interestingly, a recent community-based study14 found that 14.5% of OCD participants were hazardous alcohol users and 20% were alcohol dependents. This prevalence of 34.5% of problem drinking, although investigated in a different culture, may indicate that some OCD sufferers in the community are using alcohol to cope with symptoms, instead of seeking treatment. The same study14 has shown that 26% of OCD individuals in the community, 40% of whom were in treatment, had a lifetime history of suicidal attempts. Therefore, it is also possible that community samples, even including less severe OCD cases, have higher rates of suicidal acts because those who seek treatment probably may have more understanding about the disorder and more hope in their improvement.
Suicide prevention is one of the most important aspects of psychiatric practice and its importance has probably been underestimated regarding OCD patients. The role of comorbidity with other mental disorders, particularly depression, which is the rule rather than the exception in OCD, needs more studies concerning its impact on suicidal thoughts and acts. Further research on suicidal aspects in OCD should include larger and ideally multicenter samples, systematic evaluation of OCD possible subtypes and symptoms content, including a dimensional approach, as well as presence of other comorbid conditions, besides depression and alcohol problems.
1. Weissman MM, Bland RC, Canino GJ, et al. The cross national epidemiology of obsessive-compulsive disorder. J Clin Psychiatry. 1994;55(suppl):5-10.
2. Torres AR, Lima MC. Epidemiology of obsessive-compulsive disorder: a review. Rev Bras Psiquiatria. 2005;27:237-242.
3. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994.
4. Goodwin DW, Guze SB, Robins E. Follow-up studies in obsessional neurosis. Arch Gen Psychiatry. 1969;20:182-187.
5. Hollander E, Greenwald S, Neville D, Johnson J, Hornig CD, Weissman MM. Uncomplicated and comorbid obsessive-compulsive disorder in an epidemiologic sample. Depress Anxiety. 1996-1997;4:111-119.
6. Coryell W. Obsessive-compulsive disorder and primary unipolar depression: comparisons of background, family history, course and mortality. J Nerv Ment Dis. 1981;169:220-224.
7. Khan A, Leventhal RM, Khan S, Brown WA. Suicide risk in patients with anxiety disorders: a meta-analysis of the FDA database. J Affect Disord. 2002;68:183-190.
8. Koran LM, Thienemann ML, Davenport R. Quality of life for patients with obsessive-compulsive disorder. Am J Psychiatry. 1996;153:783-788.
9. Hollander E, Stein DJ, Kwon JH, et al. Psychosocial function and economic costs of obsessive-compulsive disorder. CNS Spectr. 1998;3(suppl):48-58.
10. Suicide Prevention: A Manual for Health Primary Care Professionals. Geneva, Switzerland: World Health Organization; 2000.
11. Tukel R, Polat A, Ozdemir O, Aksut D, Turksoy N. Comorbid conditions in obsessive-compulsive disorder. Compr Psychiatry. 2002;43:204-209.
12. Denys D, Tenney N, van Megen HJ, de Geus F, Westenberg HG. Axis I and II comorbidity in a large sample of patients with obsessive-compulsive disorder. J Affect Disord. 2004;80:155-162.
13. Overbeeck T, Schrues K, Vermetten E, Griez E. Comorbidity of obsessive-compulsive disorder and depression: prevalence, symptom severity, and treatment effect. J Clin Psychiatry. 2002;63:1106-1112.
14. Torres AR, Prince MJ, Bebbington PE, et al. Obsessive-compulsive disorder: prevalence, comorbidity, impact and help-seeking in the British National Psychiatric Morbidity Survey of 2000. Am J Psychiatry. 2006;163:1978-1985.
15. Sareen J, Cox BJ, Afifi TO, et al. Anxiety disorders and risk for suicidal ideation and suicide attempts: a population-based longitudinal study of adults. Arch Gen Psychiatry. 2005;62:1249-1257.
16. Apter A, Horesh N, Gothelf D, et al. Depression and suicidal behavior in adolescent inpatients with obsessive compulsive disorder. J Affect Disord. 2003;75:181-189.
17. Goodman WK, Price LH, Rasmussen SA, et al. The Yale-Brown Obsessive Compulsive Scale. I. Development, use, and reliability. Arch Gen Psychiatry. 1989;41:1006-1011.
18. Goodman WK, Price LH, Rasmussen SA, et al. The Yale-Brown Obsessive Compulsive Scale. II. Validity. Arch Gen Psychiatry. 1989;41:1012-1016.
19. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry. 1961;4:53-63.
20. Babor TF, Higgins-Biddle JC, Saunders JB, Monteiro MG. AUDIT: The Alcohol Use Disorders Identification Test: Guidelines for Use in Primary Care. 2nd ed. Geneva, Switzerland: World Helath Organization; 2001.
21. Stata Statistical Software. release 8.0. College Station, Tex: Stata Corporation; 2003.
22. Wetzler AJ, Elias R, Fostick L, Zohar J. Suicidal ideation versus suicidal obsession: a case report. CNS Spectr. 2007;12:553-556.
23. Mataix-Cols D, Pertusa A, Leckman JF. Issues for DSM-V: how should obsessive compulsive and related disorders be classified. Am J Psychiatry. 2007;164:1313-1314.
24. Silverman MM. Preventing suicide: a call to action. World Psychiatry. 2004;3:152-153.
25. Sevincok L, Akoglu A, Kokcu F. Suicidality in schizophrenic patients with and without obsessive-compulsive disorder. Schizophr Res. 2007;90:198-202.
26. Botega NJ, Garcia LSL. Brazil: the need for violence (including suicide) prevention. World Psychiatry. 2004;3:157-158.
27. Botega NJ, Barros MBA, Oliveira HB, Dalgalarrondo P, Marin-León L. Comportamento suicida na comunidade: prevalência e fatores associados à ideação suicida. Rev Bras Psiquiatria. 2005;27:45-53.
28. Rasmussen SA, Eisen JL. The epidemiology and clinical features of obsessive-compulsive disorder. Psychiatr Clin N Am. 1992;15:743-758.
29. Milanfranchi A, Marazziti D, Pfanner C, et al. Comorbidity in obsessive-compulsive disorder: focus on depression. Eur Psychiatry. 1995;10:379-382.
30. Crino RD, Andrews G. Obsessive-compulsive disorder and axis I comorbidity. J Anxiety Disord. 1996;10:37-46.
31. Riemann BC, McNally RJ, Cox WM. The comorbidity of obsessive-compulsive disorder and alcoholism. J Affect Disord. 1992;6:105-110.
32. Yaryura-Tobias JA, Grunes MS, Todaro J, McKay D, Neziroglu FA, Stockman R. Nosological insertion of axis I disorders in the etiology of obsessive-compulsive disorder. J Anxiety Disord. 2000;14:19-30.