High rates of OCD symptom misidentification by mental health professionals
Ferkauf Graduate School of Psychology, Yeshiva University, Bronx, NY, USA
Ferkauf Graduate School of Psychology, Yeshiva University, Bronx, NY, USA
Ferkauf Graduate School of Psychology, Yeshiva University, Bronx, NY, USAAnthony Pinto, PhD
Columbia University/New York State, Psychiatric Institute, New York, NY, USA
BACKGROUND: More than a decade may pass between the onset of obsessive-compulsive disorder (OCD) symptoms and initiation of treatment. One explanation may be health care professionals’ limited awareness of OCD symptom presentations. We assessed mental health care providers’ ability to identify taboo thoughts as manifestations of OCD.
METHODS: A random sample of 2,550 American Psychological Association members were asked to give diagnostic impressions based on 1 of 5 OCD vignettes: 4 about taboo thoughts and 1 about contamination obsessions.
RESULTS: Three-hundred sixty (14.1%) providers completed the survey. The overall misidentification rate across all vignettes was 38.9%. Rates of incorrect (non-OCD) responses were significantly higher for the taboo thoughts vignettes (obsessions about homosexuality, 77.0%; sexual obsessions about children, 42.9%; aggressive obsessions, 31.5%; and religious obsessions, 28.8%) vs the contamination obsessions vignette (15.8%).
CONCLUSIONS: Mental health professionals commonly misidentify OCD symptom presentations, particularly sexual obsessions, highlighting a need for education and training.
KEYWORDS: obsessive-compulsive disorder, misidentification, taboo thoughts, symptom presentation
ANNALS OF CLINICAL PSYCHIATRY 2013;25(3):201-209
The delay between the onset of obsessive-compulsive disorder (OCD) symptoms and the inception of treatment typically is ≥10 years.1-3 This gap is of significant concern because OCD can be a debilitating condition; for example, the World Health Organization reported that OCD is the 11th leading cause of nonfatal burden in the world.4 Because highly effective, empirically based treatments exist,5 decreasing the time from onset of symptoms to treatment engagement may reduce the symptom severity, distress, and impairment associated with OCD.
One explanation for the delay between OCD symptom onset and treatment may be health care professionals’ lack of awareness regarding the broad range of OCD symptoms, which may result in misdiagnosis or nondiagnosis of OCD. As many as 26% of individuals who meet OCD criteria are initially not correctly identified as having OCD, with the most common impressions being generalized anxiety disorder, depression, family problems, and personality disorder.2 Misdiagnosis of OCD can result in improper treatment, higher treatment-related costs, and poorer outcomes, including clinical worsening and treatment dropout.2,6,7
OCD is a heterogeneous disorder with varying symptom presentations both across and within individuals over time.8 Besides hoarding (now a distinct disorder in the DSM-59), factor analytic studies have reported the following major OCD symptom dimensions: symmetry/ordering, doubt/checking, contamination/cleaning, and taboo thoughts.10,11 Because no study to date has assessed mental health professionals’ knowledge of OCD symptom presentations, the impact of the heterogeneity of OCD symptomatology on misdiagnosis is not known. We hypothesize that the taboo thoughts dimension of OCD, including intrusive thoughts about harming others, morality, incest, pedophilia, sacrilege, sexuality, and violent images,12 is more likely to be misidentified because of limited awareness of these presentations among clinicians. Taboo thoughts are common, with reported prevalence rates ranging between 14% and 44% among individuals with OCD,13-15 but these obsessions are often overlooked in educational materials for psychiatric clinicians.16-18
We conducted a vignette-based survey study in a random sample of mental health providers who were members of the American Psychological Association (APA). Because the contamination-symptom presentation of OCD is commonly described in educational materials for psychiatric clinicians, we decided to compare participants’ ability to correctly identify taboo thoughts vignettes (covering aggressive, religious, and sexual obsessions) vs a contamination vignette. We expected that vignettes about taboo thoughts would be more likely to be misidentified, compared with vignettes about contamination obsessions.
The APA online membership directory (http://memforms.apa.org/apa/cli/mbdirsearch/index.cfm) was used to locate potential participants for this e-mail survey. Fifty APA members were randomly selected from each state and the District of Columbia. When no e-mail address was provided for a selected individual, the next member who provided his/her contact information was selected instead. Results from a power analysis with a medium effect size (f = .25), α = .05, and 80% power supported a sample size of 200. Based on the response rate of a comparable online survey (14.8%), we predicted a response rate of 15%.19 Therefore, our assessment was that e-mailing 2,550 APA members should result in approximately 380 completed surveys, well above the sample size estimated by our power analysis.
Of the 2,550 APA members e-mailed, 360 (14.1%) completed the survey (57.5% female; mean age, 51.8 years [SD = 12.6]). Most participants held a degree in clinical psychology, had a PhD, worked in a clinical setting predominantly with adults, and reported cognitive-behavioral therapy as their main theoretical orientation. It should be noted that the participants’ training and experience in treating OCD was not assessed. This was done in attempt to keep the participants blind to the study’s aim. See FIGURE 1 for response rates by vignette type and TABLE 1 for sample demographics.
Description of total sample
|Age, years (SD)
||Top 5 types of professional degree (%)
|| Clinical psychology
|| Counseling psychology
|| Child psychology
| Yes, %
|| School psychology
||Top 5 specialty areas
| African American
|| Anxiety disorders
|| Mood disorders
|| Adjustment disorders
| Native American
|Years since highest degree awarded, mean (range)
||20 (0 to 50)
|| Family therapy
|Professional setting (%)a
||Predominant patient population
||Clients seen per day, mean (SD)
||New clients per month, mean, (SD)
| Currently licensed (%)
||Top 5 main theoretical orientations (%)
|Top 5 degrees/licenses (%)a
|| Cognitive-behavioral therapy
|| Family relational
||Primary location description (%)
FIGURE 1: Response rates by vignette type
a50 US-based APA members per state and the District of Columbia.
APA: American Psychological Association.
Five vignettes were created to assess the ability of mental health professionals to accurately identify specific symptom presentations of OCD. A vignette about contamination obsessions was selected as the study control because mental health professionals are more readily exposed to this presentation through educational materials and the media. The 4 experimental vignettes each focused on a common symptom presentation of taboo thoughts OCD: aggressive obsessions, religious obsessions, obsessions about homosexuality, and sexual obsessions about children. To reduce content bias, the demographic information of the patient described remained constant across all 5 vignettes. Furthermore, in accordance with a review article of clinical vignette–based studies that emphasized that “clarity and brevity [of the vignettes] are imperative,”20 the length of this study’s vignettes ranged from 4 to 5 sentences (word count range, 64 to 80). According to the Coleman-Liau Index, the vignettes were written at an average grade level of 10.80 (range, 9.80 to 12.10).21 Also, similar to previous vignette-based studies in which the content of the vignettes was validated by specialists in the field,22-24 the content of the 5 vignettes was approved by 5 members of the Center for OCD and Related Disorders at the New York State Psychiatric Institute, Columbia University. The OCD specialists who validated the vignettes consisted of a combination of researchers and clinicians, each of whom had at least 5 years of experience working with OCD. The 5 vignettes appear in FIGURE 2.
FIGURE 2: Study vignettesa
aFor participants who work primarily with children/adolescents:
1. “Jack” was presented as a “teenaged boy” in each vignette.
2. In the vignette on sexual obsessions about children, “uncle” and “nieces and nephews” were changed to “cousin” and “cousins,” respectively.
The study was approved by the institutional review board at Albert Einstein College of Medicine/Yeshiva University, and data collection took place from June 2011 through September 2011. The survey was piloted on 10 individuals and the average completion time was 5 minutes, 37 seconds (SD=59.7 seconds).
One of the 5 vignettes was randomly assigned to each participant. To ensure equal distribution of vignettes, 10 of each vignette were assigned per state. Participants were e-mailed a description of the study and a link to the survey site. At least 2 weeks after the first e-mail, a follow-up reminder e-mail was sent. The first page of the survey included consent information; participants who elected not to provide informed consent were not permitted to continue with the survey. Upon completion of the study, participants were invited to enter a raffle drawing for a $100 gift card.
Participants were asked whether they primarily work with children/adolescents or adults. Based on their response, they were presented with a vignette that described “Jack, a teenaged boy” or “Jack, a middle-aged man.” After the vignette, participants were asked to give their diagnostic impressions of “Jack” by selecting from a list of 36 psychiatric and nonclinical diagnoses (“Other” was also an option) (FIGURE 3). If participants selected >1 condition, they were asked to rank the order of likelihood for each disorder chosen. Participants were considered to have provided a correct response as long as OCD was selected as one of the possible conditions, regardless of where OCD was listed on their ranking.
FIGURE 3: Survey participants’ diagnostic impressions of patient in vignettea
aBased on a list of 36 psychiatric diagnoses and nonclinical symptom presentation and “Other.”
SPSS (IBM) and SAS (SAS Institute) statistical software were used for descriptive and logistic regression analyses. All analyses were 2 tailed, and statistical significance was determined by α=.05. Rates of OCD misidentification were examined for each vignette. A Wald chi-square test within the context of a logistic regression analysis was conducted to compare rates of OCD misidentification for each vignette type vs the contamination obsessions vignette (control condition). Frequencies were also calculated for the most prevalent non-OCD response for each vignette.
Point biserial correlations between OCD identification and each demographic variable were examined. To test for collinearity between demographic variables, point biserial correlations were conducted between those demographic variables that had a significant correlation with OCD identification. It was decided a priori that if a pair of demographic variables was correlated at ≥0.6, guidelines for addressing multicollinearity would be followed.25
A multivariate regression was conducted to determine the best predictors of OCD identification among 1) each of the demographic variables significantly correlated with an OCD response, and 2) vignette type.
OCD misidentification by vignette type
The response rate did not differ by vignette type (12.9% to 14.9%; Wald χ2 =0.806; P=.938). Across all 5 vignettes, 38.9% of participants provided an incorrect (non-OCD) response. The contamination obsessions vignette resulted in the lowest misidentification rate, with 15.8% of participants providing a non-OCD response. In contrast, the 4 taboo thoughts vignettes were incorrectly identified by 44.7% of mental health professionals (Wald χ2 =17.91; P < .001). APA members who reviewed one of the taboo thoughts vignettes were 99.7% less likely to identify OCD than those who reviewed the contamination obsessions vignette. See TABLE 2 for rates of OCD misidentification for each vignette type and the odds ratio of misidentification for each of the taboo thoughts vignettes vs the contamination obsessions vignette. When comparing specific taboo thoughts vignettes with the contamination vignette, OCD misidentification was significantly higher for both the obsessions about homosexuality and sexual obsessions about children vignettes.
Comparison of rates of incorrect OCD identification between the contamination obsessions vignette and each of the taboo thoughts vignettes (n=360)
|Obsessions about homosexuality
|Sexual obsessions about children
|Any taboo thoughts vignette
The most common clinical impressions assigned to “Jack” by participants who did not select OCD were as follows (listed by vignette type): sexual identity confusion (65%; obsessions about homosexuality), pedophilia (37%; sexual obsessions about children), impulse control disorder (38%; aggressive obsessions), strong religious values (30%; religious obsessions), and specific phobia (63%; contamination obsessions).
Predictors of OCD identification
Bivariate correlations and chi-square analyses identified the following variables as significantly associated with an OCD response: cognitive-behavioral therapy (CBT) theoretical orientation (r=.121; P = .024), licensure (r=.129; P=.017), location of practice, ie, urban or suburban vs rural (χ2 =7.531; P=.023), clinical psychology degree (r=.109; P = .044), and mood disorder specialist (r=.112; P=.039). Significant correlations were found between the following demographic variables: licensed and clinical psychology (r=.298; P < .001), licensed and mood disorder specialist (r=.249; P < .001), clinical psychology degree and mood disorder specialist (r=.154; P < .004), clinical psychology degree and CBT orientation (r=.125; P=.020), and licensed and CBT orientation (r=.125; P=.020). Based on these correlations, there was no evidence of significant multicollinearity among the demographic variables.
When the 5 variables correlated with OCD identification were entered in a multivariate logistic regression along with vignette type, only vignette type (Wald χ2 =57.17; P < .001) and CBT orientation (Wald χ2 =3.92; P = .048) remained significant. However, even though CBT-oriented practitioners were significantly more likely to provide an OCD response, their rate of OCD misidentification was still 31.5%.
The significant delay typically seen between onset of OCD symptoms and treatment initiation may be impacted by limited awareness among mental health professionals about the variety of OCD symptom presentations. Our findings provide preliminary evidence to support this argument: the rate of OCD misidentification in this study was surprisingly high for a sample of mental health professionals, with more than one-third failing to identify OCD based on a vignette, consistent with prior research on rates of misdiagnosis (26%) in OCD.2 As a group, the taboo thoughts vignettes were misidentified at a significantly higher rate (44.7%) than the contamination vignette (15.8%). In fact, participants were 99.7% less likely to provide a correct OCD response when randomized to one of the taboo thoughts vignettes compared with the contamination vignette. Vignettes about sexuality were most commonly misidentified, with more than 75% of participants randomized to the obsessions about homosexuality vignette and more than 40% of participants randomized to the obsessions about children vignette providing an incorrect response. Approximately one-third of participants randomized to the aggressive and religious obsessions vignettes provided a non-OCD diagnosis.
Approximately two-thirds of participants who did not provide an OCD response selected sexual identity confusion (based on the obsessions about homosexuality vignette) and specific phobia (based on the contamination obsessions vignette) as the primary diagnosis. More than one-third of participants who gave a non-OCD response reported impulse control disorder (based on the aggressive obsessions vignette), pedophilia (based on the sexual obsessions about children vignette), and strong religious values (based on the religious obsessions vignette) as the primary condition. Our data suggest confusion among providers between the OCD symptom presentations and these commonly endorsed alternative diagnoses, providing areas to target in continuing education and training programs.
Vignette type was by far the strongest predictor of correct OCD identification. The only demographic variable that remained significant in the regression model was whether or not the participant identified CBT as his/her main theoretical orientation. However, even among those with a CBT orientation, one-third misidentified OCD.
Misdiagnosis of OCD may occur for various reasons. Individuals with OCD may be hesitant to reveal certain symptoms to their providers due to the “embarrassing” or “disturbing” content of their obsessions and/or compulsions. However, as found in this study, even with disclosure of one’s symptoms, misidentification by mental health professionals is still common. The elevated misidentification rates suggest that lack of awareness regarding the range of OCD symptom presentations, particularly those that center on taboo thoughts, may be key.
A misdiagnosis of OCD may have severe implications. Individuals given an incorrect diagnosis may receive treatment that does not target the actual obsessions and compulsions; this delays symptom reduction, wastes health care resources, and may result in clinical worsening or treatment dropout.2,6,7 For example, individuals who seek treatment for obsessions about homosexuality but are considered to be struggling with sexual identity confusion (which occurred 65% of the time in this study when clinicians were presented with an obsessions about homosexuality vignette) may receive treatment that targets sexual orientation ambivalence instead of exposures that target the obsessive thoughts. Attempting to resolve sexual orientation confusion among individuals with obsessions about homosexuality has been shown to have deleterious effects, whereas exposures have been shown to have beneficial outcomes.26
Furthermore, an incorrect diagnosis may have significant societal consequences for individuals. For example, the most commonly selected diagnosis by professionals who received the sexual obsessions about children vignette was pedophilia (identified as primary diagnosis by 37%). An incorrect “pedophile” label brings on intense adverse societal reactions toward the individual as well as detrimental psychological consequences. Individuals with OCD who have intrusive sexual thoughts about children experience significant distress from the thoughts; to have their worst fears of being a pedophile incorrectly confirmed by a mental health professional may induce greater impairment in functioning and depression. A further complication is that in some states clinicians are mandated to report individuals who they believe may harm an identifiable victim.27 Clinicians who are not aware of intrusive aggressive and sexual thoughts as symptoms of OCD may incorrectly report individuals with these types of obsessions to the authorities.
To increase participation, the study was designed to take place over the Internet and to have an average completion time of approximately 5 minutes. Although all the vignettes were validated as illustrating OCD, having to decide on an individual’s condition based on 4 to 5 sentences has significant limitations. We attempted to mitigate the impact of this limitation by allowing participants to select as many conditions as they deemed appropriate and giving credit for OCD identification whether or not it was named as the primary disorder. Another study limitation was that the study only included members of the APA, resulting in an oversampling of clinicians with a PhD or PsyD. Furthermore, the low response rate (14.1%) suggests possible limitations to the representativeness of the findings to the sample population. In addition, the lower than expected response rate may be due to receiving the e-mail solicitation from a graduate student, mental health professionals’ busy schedules, and the low financial incentive.
Future studies could assess mental health providers’ ability to correctly identify OCD across a broader range of OCD symptom presentations, for example, including vignettes on ordering/arranging, checking, and hoarding, in addition to the vignettes on contamination and taboo thoughts. Furthermore, future research could include non-OCD vignettes to provide a stronger control group. In addition, assessing the ability of primary care physicians (PCPs) to correctly identify common OCD presentations is recommended. Research shows that from 20% to 62% of individuals first present their psychiatric symptoms to PCPs.28,29 A focus group study found the vast majority of patients reported feeling more comfortable speaking with their PCP regarding psychiatric issues as opposed to a mental health professional.30 However, 38% of OCD patients who reported being misdiagnosed were given an incorrect diagnosis by their general practitioner.2 Proper OCD detection at the primary care stage facilitates appropriate mental health referrals.
Intervention studies that focus on increasing awareness of the various OCD symptom presentations may help improve identification of the disorder and reduce health care costs. Targeting graduate and medical school training programs may serve as an appropriate entry point; increasing OCD education among individuals in training may have broad implications for increasing awareness of OCD in clinical practice.
DISCLOSURES: The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.
ACKNOWLEDGMENTS: This study is supported by National Institute of Mental Health grant K23 MH080221 (Pinto). We would like to thank Gary Winkel, PhD, for his help with the statistical analyses and members of the Center for OCD and Related Disorders at Columbia/New York State Psychiatric Institute for approving the OCD content of the vignettes. We would also like to thank Shafou Chen, MD, PhD, Jerome Wakefield, PhD, DSW, and Sonia Suchday, PhD, for their input and guidance.
- Cullen B, Samuels JF, Pinto A, et al. Demographic and clinical characteristics associated with treatment status in family members with obsessive-compulsive disorder. Depress Anxiety. 2008;2:218–224.
- Hollander E, Stein DJ, Kwon JH, et al. Psychosocial function and economic costs of obsessive-compulsive disorder. CNS Spectrum. 1998;3:48–58.
- Pinto A, Mancebo MC, Eisen JL, et al. The Brown Longitudinal Obsessive Compulsive Study: clinical features and symptoms of the sample at intake. J Clin Psychiatry. 2006;67:703–711.
- Ayuso-Mateos JL. World Health Organization. Global burden of obsessive-compulsive disorder in the year 2000. http://www.who.int/healthinfo/statistics/bod_obsessivecompulsive.pdf. Accessed January 14, 2013.
- American Psychiatric Association Practice Guidelines. http://www.psychiatryonline.com/pracGuide/pracGuideTopic_10.aspx. Accessed January 14, 2013.
- Gordon WM. Sexual obsessions and OCD. Sexual and Relationship Therapy. 2002;17:343–354.
- Williams M. Homosexuality anxiety: a misunderstood form of OCD. In: Sebeki LV ed. Leading-edge health education issues. New York, NY: Nova Science Publishers, Inc; 2008:198-205.
- Pinto A, Grados MA, Simpson HB. Challenges in OCD research: overcoming heterogeneity. In: Simpson HB Neria Y, Lewis-Fernandez R, et al, eds. Anxiety disorders: theory, research, and clinical perspectives. Cambridge, UK: Cambridge University Press; 2010:69-79.
- Diagnostic and statistical manual of mental disorders, 5th ed. Arlington VA: American Psychiatric Association; 2013.
- Pinto A, Eisen JL, Mancebo MC, et al. Taboo thoughts and doubt/checking: a refinement of the factor structure for obsessive–compulsive disorder symptoms. Psychiatry Res. 2007;151:255–258.
- Pinto A, Greenberg BD, Grados MA, et al. Further development of YBOCS dimensions in the OCD Collaborative Genetics study: symptoms vs categories. Psychiatry Res. 2008;160:83–93.
- Goodman WK, Rasmussen SA, Price LH, et al. Yale-Brown obsessive compulsive scale (Y-BOCS). http://www.stlocd.org/handouts/YBOC-Symptom-Checklist.pdf. Accessed January 14, 2013.
- Grant JE, Pinto A, Gunnip M, et al. Sexual obsessions and clinical correlates in adults with obsessive-compulsive disorder. Compr Psychiatry. 2006;47:325–329.
- Mataix-Cols D, Rauch SL, Manzo PA, et al. Use of factor-analyzed symptom dimensions to predict outcome with serotonin reuptake inhibitors and placebo in the treatment of obsessive-compulsive disorder. Am J Psychiatry. 1999;156:1409–1416.
- Tek C, Ulug B. Religiosity and religious obsessions in obsessive-compulsive disorder. Psychiatry Res. 2001;104:99–108.
- Carlat DJ. Practical guides in psychiatry: the psychiatric interview. Philadelphia PA: Lippincott Williams & Wilkins; 2005.
- Morrison J. The first interview. 3rd ed. New York NY: Guildford Press; 2008.
- Walker JI. Complete mental health: the go-to guide for clinicians and patients. New York NY: W.W. Norton & Company; 2010.
- Porter SR, Whitcomb ME. The impact of contact type on web survey response rates. Public Opinion Quarterly. 2003;67:579–588.
- Veloski J, Tai S, Evans AS, Nash DB. Clinical vignette-based surveys: a tool for assessing physician practice variation. Am J Med Qual. 2005;20:151–157.
- Tests document readability: readability calculator. http://www.online-utility.org/english/readability_test_and_improve.jsp. Accessed January 13 2013.
- Bruchmüller K, Margraf J, Schneider S. Is ADHD diagnosed in accord with diagnostic criteria? Overdiagnosis and influence of client gender on diagnosis. J Consult Clin Psychol. 2012;80:128–138.
- Gude T, Dammen T, Friis S. Clinical vignettes in quality assurance: an instrument for evaluation therapists’ diagnostic competence in personality disorders. Nord J Psychiatry. 1997;51:207–212.
- Lanza ML, Carifio J. Use of a panel of experts to establish validity for patient assault vignettes. Evaluation Review. 1992;16:82–92.
- Grewal R, Cote JA, Baumgartner H. Multicollinearity and measurement error in structural equation models: implications for theory and testing. Marketing Science. 2004;23:519–529.
- Penzel F. How do I know I’m not really gay? International OCD Foundation. http://www.ocfoundation.org/EO_HO.aspx. Accessed January 14 2013.
- Tarasoff v Regents of the University of California. In: Gostin LO, ed. Public health law and ethics: a reader. Berkeley, CA: University of California Press. http://www.publichealthlaw.net/Reader/docs/Tarasoff.pdf. Accessed January 14, 2013.
- Kessler RC, Zhao S, Katz SJ, et al. Past-year use of outpatient services for psychiatric problems in the National Comorbidity Survey. Am J Psychiatry. 1999;156:115–123.
- Torres AR, Prince MJ, Bebbington PE, et al. Treatment seeking by individuals with obsessive-compulsive disorder from the British psychiatric morbidity survey of 2000. Psychiatr Serv. 2007;58:977–982.
- Lester H, Tritter JQ, Sorohan H. Patients’ and health professionals’ views on primary care for people with serious mental illness: focus group study. BMJ. 2005;330:1122.
CORRESPONDENCE: Kimberly Glazier, MA Ferkauf Graduate School of Psychology Yeshiva University 1165 Morris Park Avenue Bronx, NY 10461 USA E-MAIL: firstname.lastname@example.org
Annals of Clinical Psychiatry ©2013 Frontline Medical Communications Inc.