Barriers to Treatment Among African Americans with Obsessive-Compulsive Disorder

Abstract

African Americans are underrepresented in OCD treatment centers and less likely to experience a remission of symptoms. This study examines the barriers that prevent African Americans with OCD from receiving treatment. Seventy-one adult African Americans with OCD were recruited and administered the modified Barriers to Treatment Participation Scale (BTPS) and the Barriers to Treatment Questionnaire (BTQ). Comparing the BTQ between a European American Internet sample (N=108) and the African American OCD sample (N=71) revealed barriers unique to African Americans, including not knowing where to find help and concerns about discrimination. A Mokken Scale Analysis of the BTPS in the African American participants identified seven major barriers, including the cost of treatment, stigma, fears of therapy, believing that the clinician will be unable to help, feeling no need for treatment, and treatment logistics (being too busy or treatment being too inconvenient). Pearson and point-biserial correlations of the scales and demographic and psychological variables were conducted. Significant relationships emerged between age, gender, income, education, insurance status, and ethnic affirmation/belonging among several of the Mokken scales. A one-way ANOVA demonstrated that concerns about cost were significantly greater for those without insurance, versus those with public or private plans. Suggestions for overcoming barriers are presented, including community education, affordable treatment options, and increasing cultural competence among mental health providers.

1.Introduction

1.1 OCD and African Americans

Significant progress has been made in developing effective treatments for people suffering from obsessive-compulsive disorder (OCD), but not all segments of our society have benefited. Epidemiological studies indicate that African-Americans appear to experience OCD at similar rates as the general population (White 2.6% vs. Black 2.3%, Zhang & Snowden, 1999; White 1.6% vs. Black 1.6%, Kessler et al., 2005; Himle et al., 2008), but are less likely to receive treatment. Among those with severe OCD, 93% of Americans receive some type of treatment (NCS-R; Ruscio et al., 2010), whereas this is true for only 60% for African Americans with severe OCD (NSAL; Himle et al., 2008), indicating a significant health disparity. Even among those who are able to access mental health care, few African-Americans with OCD receive specialized treatment and only 20% are using an SRI medication (Himle et al., 2008).

African Americans are notably underrepresented in the OCD literature (Himle et al., 2008; Williams et al., 2010). The absence of research on OCD in the African American population brings into question whether African Americans are optimally treated in the same ways that have been demonstrated to be beneficial to European Americans. Presently, only a handful of studies have attempted to examine treatments for African Americans with OCD, and none of these have been randomized treatment outcome studies (Hatch et al., 1996; Williams et al., 1998; Friedman et al., 2003). African Americans are also underrepresented in OCD treatment clinics. For instance, in the OCD Field Trial conducted for DSM-IV (Foa et al., 1995), participants were recruited from OCD treatment centers in Atlanta, Boston, New York, New Haven, Philadelphia, Providence, and Toronto, but less than 1% of the 450 patients were African American. One must then hypothesize that certain, specific barriers may be preventing African Americans from accessing treatment.

1.2 General barriers to receiving OCD treatment

There are a number of barriers that prevent people with OCD from receiving treatment. Research done by Goodwin et al. (2002) determined that gender, age, and race were some of the strongest predictors of having received any form of mental health treatment. The demographic attributes determined to have the highest correlation with treatment-seeking behaviors belonged to older European American females.

Shapiro et al. (1984) determined that less than 35% of those with OCD received specialized mental health treatment among adults in a community-based sample. Individuals with OCD and a comorbid disorder are four times more likely to seek treatment than those with OCD alone (56% vs. 14% over a 3 month period; Torres et al., 2007). This suggests that comorbidity is a strong predictor of increased treatment-seeking behaviors; however comorbid disorders may also obfuscate the diagnosis of OCD.

A study by Mayerovitch et al. (2003) found that only 37% of their clinical sample sought help for OCD symptoms. A study by Goodwin et al. (2002) found that among those who did not seek treatment, reasons included not knowing where to get help (40.0%), wanting to handle it on their own (28.4%), being unable to afford it (24.9%), or being afraid of the stigma attached to seeking help (5.7%), with minority racial status as a barrier to receiving care [OR = 0.7(0.5–0.9)]. Approximately 20% of this sample endorsed all of these barriers, however only 5.7% believed that treatment would not help at all (Goodwin et al., 2002). Torres (2007) also determined that avoidance because of fears of contamination or other obsessions has the potential to prevent some people from seeking services.

An Internet study complied by Marques et al. (2010) uncovered similar barriers to treatment-seeking among individuals with OCD. They found that logistic and financial concerns, stigma, shame and discrimination, as well as treatment perception, and satisfaction barriers were all significant obstacles to receiving treatment. As a result of these barriers, specifically stigma and shame, most OCD sufferers wait 8–10 years before seeking treatment (Marques et al., 2010; Pallanti, 2008; Rasmussen & Tsuang, 1986), and as few as 29% of the afflicted population obtain treatment specifically for their OCD (NCS-R; Ruscio, et al., 2010). This treatment implementation gap suggests that clinicians are either missing or misdiagnosing OCD due to its high comorbidity with other Axis I disorders, or perhaps that OCD patients are not describing the typical symptoms of the disorder to their health care providers (Grabill et al., 2008; Sussman, 2003). Combine these issues with the fact that cognitive behavioral therapy (CBT) for OCD is very specialized, expensive, and difficult to find, and the delay for treatment seems inevitable. Although research on barriers to treatment afflicting many sectors of society have been examined, it is not clear that we can extrapolate such findings to an underrepresented minority group, such as African Americans.

1.3 Cultural differences in barriers to OCD treatment

In addition to the aforementioned barriers to OCD treatment, there may be additional barriers experienced uniquely by African Americans. For example, research has noted that many African Americans prefer a counselor of the same race (Malat, Purcell, & van Ryn, 2010), which may present a barrier to treatment as African Americans are under represented among clinical psychologists. This preference may protect African Americans against the psychological effects of racial discrimination, but prevent the initiation of needed services (Whaley, 2001). Many African Americans have concerns about receiving inadequate treatment or misdiagnosis due to discrimination. African Americans have a cultural memory of abuses such as the US Public Health Service Syphilis Study at Tuskegee, which continues to affect medical decision-making (Gamble, 1997). In addition, Snowden et al. (2009) found that African Americans and Caribbean Blacks born in the U.S. have the highest risk of psychiatric hospitalization, which demonstrates reasons for apprehension when African Americans do need help. In a special report of the US Surgeon General, it was noted that African Americans have reduced access to medical insurance, are less likely to have needed mental health services in their communities, and are more likely to be misdiagnosed than European Americans (USDHHS, 2001).

These issues, combined with lower incomes and higher unemployment rates among African Americans, may pose an additional barrier to treatment among those with OCD (Himle et al., 2008). However, this does not explain the fact that many African Americans belong to the upper and middle classes, yet still do not receive treatment. Thus it is clear that a better understanding of barriers to OCD treatment among African Americans is critically needed. The purpose of this study is to determine what barriers prevent African Americans from receiving treatment for OCD through a comprehensive assessment of symptoms and self-reported concerns about treatment issues.

2. Methods

2.1. Participants

Recruitment occurred over a nine-month period during 2009–2010 at the Center for the Treatment and Study of Anxiety (CTSA) at the University of Pennsylvania in Philadelphia. Following IRB approval, eighty-three African American participants were recruited for the study. Of these, 71 met criteria for a current diagnosis of OCD, 4 had past OCD, and the reminder did not have OCD.

On average, participants with lifetime OCD were 41.3 years old with ages ranging from 20 to 65. Participants had a median household income of approximately $20–40K, which is comparable to the African American median household income estimate of $33,578 (U.S. Census Bureau, 2010). African American group membership was determined based on self-report, however, participants were also required to have lived most of their lives in the US since age 5. Demographics for the sample can be found in Table 1.

Table 1.

Participant Demographics

Demographics		n.	Percentage (%)
Gender	Male	31	43.7
	Female	40	56.3
Marital Status	Married	13	18.3
	Single/ Never Married	41	57.7
	Divorced/ Annulled	17	24.0
Highest Education	Primary	21	29.6
Completed	Secondary	43	60.6
	Tertiary	7	9.9
Employment	Employed	28	39.0
	Unemployed	22	31.0
	Disabled	11	16.0
	Students	9	13.0
	Retired	1	1.0
Yearly Income	<$9,999	18	26.0
	$10,000 – $39,999	33	47.0
	$40,000 – $79,999	11	16.0
	>$80,000	9	11.0
Insurance Status	None	17	23.9
	State or Federal Plan	28	39.4
	Private or Group Plan	12	16.9
	Unknown	14	197

2.2. Procedure

The CTSA research assistant pre-screened potential participants for eligibility by phone. Once the participant passed the initial phone screen, s/he met with masters or doctoral-level study evaluator to review procedures and give informed consent. The research assistant then provided self-report assessments. Once all the self-assessments were completed, the evaluator administered a psychiatric diagnostic interview to the participant, followed by measures of psychopathology and barriers to treatment. Participants were diagnosed by primarily African American master’s or doctoral-level evaluators who were trained for the study by the first author. To ensure accurate diagnosis, 22.5% of OCD assessments were checked via a live second interview with a doctoral-level expert OCD clinician, for a proportion of agreement of 94.4% and a kappa coefficient of .77 (p=.001), indicating substantial agreement. Additional details about the methodology are reported elsewhere (Williams Proetto, Casiano, & Franklin, in press).

2.3. Measures

Participants received both self-report and clinician-administered measures as part of the comprehensive assessment process.

2.3.1 Barriers to treatment questionnaire

The Barriers to Treatment Questionnaire (BTQ; Goodwin et al., 2002; Marques et al., 2010) measures participants’ perceived barriers to seeking OCD treatment, based on similar questionnaires from the broader literature. The BTQ assesses barriers to treatment in the following domains: logistic and financial, stigma, shame and discrimination barriers, and treatment perception and satisfaction barriers. Participants were asked if any of these possible barriers influenced them to not seek treatment. Cronbach’s alpha (α) for the BTQ was .82 for our African American sample and.71 for the European American sample.

2.3.2 Multigroup ethnic identity measure

The Multigroup Ethnic Identity Measure (MEIM; Phinney, 1992) is suitable for use with any ethnic group to quantify ethnic identity. Included are subscales for determining feelings of affirmation and belonging, ethnic identity achievement, and ethnic behaviors. Ethnic identity was determined using the 12-item revision by Roberts et al. (1999) validated in adolescents of various ethnic groups. The MEIM has also been validated in a national sample of Black and White adults, with good reliability (α = 0.91; Duque et al., 2011). Items were scored from 0 to 3, with higher numbers corresponding to greater agreement. For this measure we examined Ethnic Identity (α = .86) and the subscale for Ethnic Affirmation and Belonging (α = .85).

2.3.3 Structured clinical interview for DSM-IV axis I disorders

The Structured Clinical Interview for DSM-IV Axis I Disorders (SCID; First et al., 1995), is a widely used semi-structured clinical interview, designed to aid clinicians and researchers in making reliable DSM-IV psychiatric diagnoses. This measure provided information about the participant’s treatment history, age of onset, OCD symptoms, and comorbid conditions.

2.3.4 Yale-Brown obsessive-compulsive checklist and severity scale

The Yale-Brown Obsessive-Compulsive Checklist and Severity Scale (YBOCS; Goodman et al., 1989), was administered to assess the phenomenology and severity of obsessive-compulsive symptoms. The YBOCS is considered the gold-standard for the assessment of OCD in adults and has been validated in many ethnic groups, including African Americans (α = .92; Washington, Norton, & Temple, 2008).

2.3.5 Barriers to treatment participation scale

Following the YBOCS, the Barriers to Treatment Participation Scale (BTPS; Kazdin et al., 1997), a 44-item scale that has been used in prior studies to identify factors that prevent patients from obtaining mental health services, was administered. It originally was developed to assess children; however a modified and expanded version (Simon, 2005) was used to ascertain which factors are important barriers to treatment among African American adults.

2.3.6 Post assessment checklist

Once all assessments were completed, a Post Assessment Checklist was used to summarize the results of the assessment. This checklist was devised specifically for this study to aid the clinician in reviewing diagnostic impressions with the participant and answering any questions. The checklist records whether participants with a DSM-IV-TR diagnosis of any anxiety disorder, who are not already receiving treatment, would like to be referred for treatment opportunities, and the reasons for their decisions. Thus, participants’ responses to this assessment served as data to aid in our understanding of actual barriers to treatment. All participants were provided with a list of local treatment providers, including low-cost options.

2.4.Statistical procedures

Two-tailed Fisher’s Exact Tests were used to make comparisons between groups on responses to true-false items. Pearson and point-biserial correlations were used to explore the relationship between scales and demographic variables. One-way Analysis of Variance (ANOVA) was used for comparisons between three groups. These analyses were conducted using SPSS, version 19.

Mokken scale analysis (MSA) was used to identify meaningful and reliable subscales of the BTPS. MSA is a technique for polytomous test item data, used for scaling test and questionnaire responses. Based on nonparametric Item Response Theory (IRT) models, MSA includes an automated selection algorithm, which partitions the set of polytomous variables (items) into scales (called Mokken scales) that optimize scale properties, and potentially leaves some items unselected (Mokken, 1971). The scales are optimized in terms of a scalability coefficient, called Hij, which describes optimal scale properties. We used a Mokken module designed for the statistical software application R (van der Ark, 2007). We specified a minimum scalability coefficient of .45 in order for an item to be included on a scale.

3. Results

3.1. BTQ comparison

The BTQ served as a preliminary overview of the perceived barriers, with responses collected from participants before receiving the more in-depth exploration via the clinical interview. To examine differences in barriers to mental healthcare between European American and African American individuals with OCD symptoms, we compared endorsement of the individual BTQ items between the present sample and a European American subset of an OCD Internet sample collected by Marques et al. (2010). Participants were only included in the comparison sample if they met a cut-off score of ≥ 16 on the Yale-Brown Obsessive-Compulsive Scale for OCD – Self Report version (YBOCS-SR; Baer, 1992).

The comparison sample (N=108) was majority female (79.6%) with M =32.7 years (SD=11.8) for age and a M= 21.4 (SD = 4.7) for the YBOCS-SR score. Fisher’s Exact Tests of sample differences in endorsement of the 19 BTQ items were conducted using a Bonferroni adjusted alpha level of 0.0026 (0.05/19). Results indicated that significantly more participants in the European American sample endorsed BTQ item 4 (e.g., I received treatment before and it didn’t work) as a barrier to treatment. Conversely, more participants in the African American samples endorsed BTQ item 11 (e.g., I was unsure about who to go see or where to go) as a barrier that had caused them to delay or avoid seeking treatment for their symptoms. Greater concerns by African Americans about discrimination (BTQ item 16) demonstrated a trend toward significance. There were no other significant differences between European Americans and African Americans for the remaining barriers examined at the adjusted alpha level. These results are congruent with past research indicating that European Americans are more likely to obtain treatment than African Americans (Himle et al., 2008).

To better understand the cause of these differences, the three items were correlated to income, education, and insurance status in the African American sample using a two-tailed Pearson/point-biserial correlation. Neither education nor insurance status was correlated to affirmative responses for any items, but participants who had tried treatment before tended to have a higher income (r=.23, p=.05). Thus it appears that knowing where to find help and concerns about discrimination are not a function of socioeconomic status in the African American sample.

3.2. Mokken scale analysis

In this study, 82 modified items (including those on the original BTPS plus additional items;Kazdin et al., 1997; Simon, 2005) made up our Mokken Scale Analysis. Prior to using MSA, however, ten items were removed due to lack of endorsement by participants. Among these, 8 questions were removed by the MSA analysis and six others were manually removed due to poor fit in their respective scales. This left a total of 58 items and 12 scales, to which we gave the following names: Cost of Treatment, Stigma and Judgment, Forced Change (including fears of involuntary hospitalization), No Need for Treatment, Afraid of Therapy, Clinician Unable to Help, Too Busy for Treatment (treatment too inconvenient and logistical issues), Does Not Trust White Establishment, Health Concerns, Family and Familiarity Concerns, In a Hurry (treatment will take too long), and Partner Problems.

We computed Cronbach's alpha for each of these new scales, and Table 3 summarizes our results. Our subsequent analyses and discussion include only the top seven Mokken scales, due to their high endorsement and reliability. Sample items from each of the seven scales are shown in Table 4.

Table 3.

Mokken Scale Analysis

Scales	Alpha	No. of Items	Scale Mean	SD	Mean Item
1. Cost of treatment	0.842	4	8.76	4.87	2.19
2. Stigma and judgment	0.700	3	5.34	3.01	1.78
3. Forced change	0.736	3	5.20	2.96	1.73
4. No need for treatment	0.696	3	5.20	2.75	1.73
5. Afraid of therapy	0.892	8	13.24	6.62	1.66
6. Clinician unable to help	0.926	12	19.61	10.01	1.63
7. Too busy for treatment	0.922	10	15.50	8.43	1.55
8. Does not trust White establishment	0.651	2	2.96	1.82	1.48
9. Health concerns	0.755	4	5.64	2.77	1.41
10. Family and familiarity concerns	0.737	5	6.96	3.19	1.39
11. In a hurry	0.536	2	2.74	1.51	1.37
12. Partner problems	0.660	2	2.29	1.05	1.15

N=70 due to 1 missing BTPS.

Items 1–7 were used in subsequent analyses due to their high endorsement and reliability.

The Mean Item value (far right column) is the scale mean divided by the number of items in the scale.

Table 4.

Sample Items from Seven Largest Barriers and Means

Scales	Example Questionnaire Items	Mean	SD
1. Cost of Treatment	Cost is fair but cannot afford the fees	2.44	1.59
	Treatment costs too much and should be less expensive	2.41	1.62
	Treatment will cost more than expected (hidden fees)	1.97	1.33
2. Stigma and Judgment	Worries about other people finding out	2.14	1.54
	Worries about people judging him/her	1.96	1.39
	Has negative feelings about those who obtain mental health care	1.24	0.71
3. Forced change	Worried about being given unwanted treatments	1.96	1.31
	Worried about being involuntarily hospitalized	1.76	1.27
	Concerned that family not accept changes in him/her	1.49	1.06
4. No need for treatment	Symptoms don't interfere with life or cause problems	1.89	1.23
	Treatment does not seem necessary	1.63	1.17
	Symptoms will improve on their own	1.69	1.08
5. Afraid of Therapy	Does not know where to find right kind of help	2.40	1.46
	Clinician unable to relate (will not understand participant’s world, will be disrespected or talked down to)	1.87	1.28
	Does not know what to expect	1.73	1.15
	Worries that treatment more work than expected	1.61	1.09
6. Clinician Unable to Help	Treatment will not work for him/her	1.91	1.39
	Clinician will be unable to help	1.79	1.22
	Significant others do not agree should have treatment	1.70	1.22
	Treatment will not work at all	1.66	1.18
7. Too Busy for Treatment	Difficult to find transportation for treatment	1.84	1.28
	Difficult to schedule appointment	1.81	1.12
	Other commitments get in the way of treatment	1.60	1.20
	Treatment is another life stressor	1.49	1.05

Questions are in order by mean response across all participants on a 1–5 scale.

3.3.Correlational analysis

To further understand the concerns of our sample, we conducted Pearson and point-biserial correlations of the seven scales to various demographics and psychological variables. The variables included gender, age, income, education level, insurance status, OCD severity (YBOCS), global psychopathology severity (GAF), total ethnic identity (MEIM) and the ethnic belonging/affirmation subscale. There were significant relationships between age, income, education year, insurance, and ethnic affirmation/belonging, for four of the Mokken scales. These findings are shown in Table 5. There was no significant relationship between total ethnic identity or psychopathology severity (GAF or YBOCS) and any of the Mokken scales, so these do not appear in the table.

Table 5.

Pearson/Point-Biserial Correlations between Demographics, Ethnic Belonging, and Four Mokken Scales

Mokken Scale	Age	Gender	Income	Insurance Coverage	Education Level	Ethnic Belonging
Cost of treatment	−.244*	−.032	−.066	−.387**	.030	−.320**
Stigma and judgment	−.218	.283*	.093	−.073	.264*	−.038
Clinician unable to help	−.288*	.208	.068	−.062	.284*	−.242*
Too busy for treatment	−.314**	.028	.251*	.032	.226	−.128

N ranges from 57–70. Insurance status coded: 0=uninsured, 1=insured. Gender coded: 0=male, 1=female. Higher values on Mokken scales represent a greater barrier to treatment. Higher values for ethnic belonging/affirmation represent stronger and more positive feelings about one’s ethnic group.

^*Correlation is significant at the .05 level (2-tailed).

^**Correlation is significant at the .01 level (2-tailed).

3.4. ANOVA for insurance coverage concerns

To better understand the significance of insurance coverage and concerns about the cost of treatment we conducted a one-way ANOVA. We used three categories; those with No Insurance, those with a State or Federally Funded Plan, or Private/Employer-Provided Insurance, and compared means on the Cost of Treatment scale. Results were significant, [F(2, 54)=4.91, p=.011], indicating that those without any insurance (M=12.00, SD=4.39; Median Income = $30,000 – $39,999), were significantly more likely to be concerned about cost of treatment than their poorer State/Federal (M=7.54, SD=4.92; Median Income = $20000 – $29,999) or wealthier private insurance plan counterparts (M=8.33, SD=5.04; Median Income=$60,000 – $69,999).

3.5. Qualitative data

Although our study did not formally analyze qualitative data, we did ask participants for open-ended comments. When asked to write-in a response about how OCD had impacted their lives at the end of the BTQ, themes such as being unaware of a problem, fear, and helplessness are apparent. One person wrote, “I was just embarrassed. Getting this type of help has, and continues to be, like a sore thumb in the African American community. Unfortunately, I don’t have insurance, so my fear was that if I sought help, it would not be good because I couldn’t afford it.” Other comments included remarks such as, “I was afraid,” and “I didn't realize I had a problem.” After the clinical assessment, participants were asked what prevented them from getting treatment. Among those who provided a response (N=51), the most prevalent answer (47.1%) was that the participant did not realize s/he had a disorder or that there was a treatment for it.

4. Discussion

4.1 Black-White comparisons

The results of our study suggest that many similar barriers afflict both ethnic/racial groups. Concerns about insurance coverage or whether treatment would be successful are endorsed relatively evenly, and stigma and shame attached to seeking mental help appears to be particularly distressing. This includes feeling ashamed of having a problem at all and wanting to conceal it, which is hypothesized to be a symptom of OCD that maintains the disorder (Newth & Rachman, 2001).

African Americans have less knowledge about where to go for help than their European American counterparts. In contrast, European Americans were more likely to report having tried treatment before and being disappointed with the results. Notably, almost a quarter of African Americans were concerned with racial inequality in treatment whereas European Americans were much less concerned with such a barrier. Major barriers to treatment among African Americans

This study’s goal was to determine the major barriers to OCD treatment facing the African Americans to ultimately increase treatment receipt in this group. The primary barriers identified are described below.

4.2.1. Cost of treatment and lack of insurance coverage

The financial burden of mental health treatment is a barrier that disproportionally affects African Americans. The wealthiest and poorest African Americans were less concerned about cost, as they believed they could get help through their public or private insurance providers. The participants most concerned about cost were more likely to be uninsured; their incomes were not low enough to qualify for publically provided services but not high enough to afford a private insurance plan.

4.2.2 Stigma and judgment

African Americans are often exposed to stigmatizing beliefs about mental illness in their families and communities (Alvidrez et al., 2008), and the current study also found such concerns among our sample, in women more than men. Many participants expressed negative attitudes about people who obtain mental health care, and understandably did not want to identify with this group. Additionally, African Americans are often resistant to seeking treatment because it may reflect badly on the family, exposing the family’s failure to handle problems internally or because the stigma of seeking help would transfer from the individual to the entire family (Alvidrez et al., 2008). Thus, concerns about stigma may result in a failure to obtain needed care due to worries about the social repercussions of exposure.

4.2.3. Concerns about therapist or treatment process

Participants reported serious concerns about treatment effectiveness, including beliefs that treatment will fail to meet expectations, and that he or she will lose interest in following through. For these reasons treatment seems to be too difficult to be worthwhile, and potentially just another life stressor.

Interestingly, having a lower sense of ethnic affirmation and belonging was significantly correlated with the belief the clinician would be unable to help. This is potentially due to the fact that feeling uncertain about one’s place in the world results in an exacerbated sense of worry. This may be especially true for the many consumers who believe that mental health treatment was designed for White people rather than Black people (Alvidrez et al., 2008). African Americans view the typical psychologist as an older White male, who would be insensitive to the social and economic realities of their lives (Thompson et al., 2004). A major concern embedded in this belief is that the participant will be unable to relate to the clinician. As a consequence these African Americans are more likely to mistrust their clinician or believe that treatment will be unsuccessful (Linnehan, Weer, & Uhl, 2005). Younger age and greater education are associated with this barrier.

Worries of being vulnerable add to the lack of motivation to seek help. Many participants voiced fears about being involuntarily hospitalized. Although African Americans did not endorse this fear in significantly greater numbers than their European American counterparts, a lower social status makes African Americans more vulnerable to abuse in a medical setting, and subsequently more likely to experience a psychiatric hospitalization (Snowden et al., 2009; Suite et al., 2007).

4.2.4 No need for treatment

It may be difficult to understand why a person with a disorder as disabling and distressing as OCD may feel they have no need for treatment, however this was a common response from our sample. Many participants indicated that their symptoms did not interfere or cause problems for them, yet this attitude was not correlated to OCD severity. One explanation could be that certain symptoms may be viewed as normal differences rather than psychopathology in this group. For example, greater washing is a cultural norm among African Americans, which may result in symptoms of contamination-themed OCD being unrecognized (Williams, Abramowitz, & Olatunji, 2012). It is possible that the greater acceptance of such differences among African Americans, may make it easier to miss or deny serious problems.

Another reason that many with significant OCD symptoms may believe they have no need for mental health treatment is the prevalent view among African Americans that the church is the only appropriate source of support outside of the family (Alvidrez et al., 2008). Research shows that religion is an important coping strategy for African Americans with anxiety, and they are more likely to use positive religious strategies (e.g., trusting God, search for personal meaning) than their anxious European American counterparts (Chapman & Steger, 2010). However, having OCD has been found to interfere with church attendance, thus African Americans with OCD seem to more heavily rely on prayer to cope with symptoms (Himle et al., this issue).

4.2.6 Lack of awareness about the disorder and treatment options

Almost half of our participants were unaware that they had OCD or did not know that there was a treatment for their problem. The taboo surrounding talking about mental illness openly, results in African Americans often possessing little knowledge of specific mental health problems and their treatments (Alvidrez et al., 2008).

4.2.5 Too busy for treatment

Another barrier focused on treatment logistics and conflicting activities. It appears that despite the level of functioning, many in our sample did not hold treatment among their priorities. Work, commitments, and transportation issues all overshadowed the need for therapy. This barrier was more likely to be endorsed by younger participants and those with lower incomes. This troublesome reality suggests that after struggling for years with their disorder, many African Americans with OCD are reluctant to take time for themselves to get better, and consider treatment logistics an insurmountable hurdle.

4.3 Increasing treatment participation

The barriers described can create a difficult uphill battle when trying to increase OCD treatment participation within the African American community. The following recommendations are made to increase the numbers of those obtaining treatment and thereby reduce this health disparity.

4.3.1 Making treatment more affordable

The cost of treatment may be prohibitive for many, especially among those without insurance coverage. Within our sample, approximately a quarter of the participants did not have any type of insurance plan. Many others who were low income were a part of the community mental health system and had access to psychiatric services. However, the community health system may suffer from a lack of clinicians able to treat OCD using CBT, and without insurance, securing help can be almost impossible. Within our study it was difficult to secure community referrals for our participants due to lack of insurance, instability of the participants’ financial and living situations, and the inability to make appointments due to overcrowding. Thus, effective CBT treatments must be disseminated into the community health system to make treatment affordable and accessible (Gunter & Whittal, 2010). On a more national scale, more CBT providers on insurance panels may also help with the diffusion of effective treatments for those with insurance.

In terms of individual practices, treatment centers can publicize effective low cost treatment options (i.e. practicum students, sliding scale slots, etc.) Another cost-effective option is stepped-care. For example, one model proposed by Tolin et al., (2005) involves bibliotherapy, followed by clinician-guided self-treatment if needed, followed by traditional EX/RP for non-responders. The self-help component of bibliotherapy may make it more acceptable and less stigmatizing to patients, while also resulting in a lower cost. Nonetheless, it is important to be aware that the cost of treatment is a larger problem that involves many aspects of our society, including the politics of health care and racial/social inequalities.

4.3.3 Increasing awareness of OCD and treatment options

The National Institute for Health and Clinical Excellence (NICE, 2006) reports that the outlook for OCD has improved dramatically since the 1980s with the introduction of better pharmacological and psychological interventions. Current evidence is strong on the effectiveness of SSRIs. Additionally, exposure and ritual prevention is a highly effective psychotherapeutic technique for battling OCD.

Education about the disorder and treatment process is a vital to reduce barriers to treatment (Ayalon et al., 2007). Research with college students suggests that minorities are less likely to successfully recognize typical OCD symptoms as part of a psychological disorder (Coles & Coleman, 2010). Suggestions for overcoming this barrier include public education campaigns (e.g., mass media), open information sessions at local mental health clinics, and educational presentations at community venues (i.e., Black churches). Many Black currently churches offer counseling services to their congregations (Blank et al., 2002). Incorporating Christian principles into the treatment protocol may make treatment more palatable for this population (e.g., Queener & Martin, 2001).

For those who do start conventional treatment, the initial clinical encounter presents an important opportunity to address skepticism about the utility of treatment. It remains the clinician’s responsibility to demystify the process and explain the benefits of staying motivated. Without this knowledge the participant may only assume what he or she may encounter and base their decision to initiate treatment on incorrect assumptions. If the expected outcomes, number of sessions, and potential goals are clearly outlined in advance, this reduces the risk of feeling misled or not in control. The client-patient bond may be strengthened with this extra attempt at transparency.

4.3.4 Making mental health a priority and the family

Treatment has the potential to conflict with many daily activities or commitments. Whether the individual feels treatment is a necessary priority to achieving good mental health despite prior engagements, transportation, or scheduling issues is an important positive step. Moreover, incorporating the family is another crucial measure in overcoming barriers to treatment. By gaining familial support the patient may gain peace of mind as well as lose the fear of being stigmatized by those he or she cares about. In addition, with the family’s acceptance, making time for treatment becomes easier and priorities may be put into perspective. For instance, in concert with the clinician, the family may help tally the number of hours spent on OC behaviors, to highlight time-savings once treatment is completed. The YBOCS is an excellent tool to facilitate this discussion as it includes several items about hours spent on OC behaviors. Furthermore, utilizing the family to emphasize the importance of good mental health creates more allies to emphasize the relationship between improved functioning and greater success at home and work.

4.3.5 Reducing fears about therapy and stigma

Making psychotherapy less intimidating may be one of the most important ways of improving help-seeking attitudes. Careful use of language can help to reduce discomfort surrounding mental health care. For example, African Americans are generally more comfortable with the term “counseling” over “psychotherapy” (Thompson et al., 2004), so clinicians may adapt their language accordingly in advertising and conversations. One practical way to reduce fears is to offer free assessments and phone consultations, which will help familiarize potential patients with the clinic, clinician, and treatment. Clinicians might also use initial contacts to address fears of being involuntarily hospitalized by explaining the difference between having OCD and “being crazy,” including the role of insight and self-efficacy.

However, it is important to note that reservations against treatment may be rooted in actual experiences of racial inequality and encounters with medical professionals lacking cultural awareness (Whaley, 2001). Both the patient and clinician must remember that treatment is a partnership that can only be successful with mutual-respect, and for this to occur it remains the duty of the therapist to become culturally competent and sensitive to disparities, and in turn, communicate support and understanding to the patient. African Americans look for subtle cues to determine if a therapist holds racist attitudes (Thompson et al., 2004), and fully a quarter of study participants were afraid of being mistreated due to their race or ethnicity. Clinicians must be willing to undertake an honest self-examination of their own conscious and unconscious attitudes about race (Suite et al., 2007), including preconceived notions about who would be a good client, as such biases may be detected by client and also impact care. By increasing the cultural competence and social awareness of all clinicians, the mental health system can begin to shed its bias against ethnic minorities. This would result in greater understanding and empathy for the patient’s experience and improved treatment outcomes (Fortier & Bishop, 2003). A supportive environment and a genuine effort by the clinician have been found to facilitate continued use of mental health services in African Americans (Ayalon et al., 2007).

4.4 Study limitations

Some considerations may limit the conclusions that can be drawn from our results. Our comparison of participant-endorsed responses of barriers to treatment may not be a perfect match due to demographic differences in the Internet sample. Thus, these comparisons should be interpreted with caution. In addition, a single discrete geographical location may prevent the generalizability of our data to the entire African American population, especially given potential geographic differences in OCD symptoms (Williams et al., 2008). Finally, the recommendations advanced herein must be empirically tested to determine if they result in greater treatment participation.

4.5 Future directions

This study underscores the need for research in additional areas of mental health disparities. Presently, it remains unclear whether barriers to treatment are in fact the same across different psychological impairments for African Americans, and by continuing this research we will be able to better understand obstacles faced by this underserved population.

It is also important to understand the experience of other ethnic groups. As previously mentioned, the United States Department of Health and Human Services (2001) has stated that many minority groups are plagued with unequal and less effective health care treatments. Some questions that warrant further study include the role of culture, prejudice, income, and insurance status in this problem.

5. Conclusion

Prior to the completion of this study, it was not clear which barriers to treatment were most salient to African Americans with OCD. The major barriers can be summarized as cost/insurance coverage, fears about the therapist or therapeutic process, stigma and judgment, feeling too busy for treatment, and a belief that treatment is unnecessary. Additionally, many participants worried that they would be treated unfairly due to their race or ethnicity, but for many, no one barrier will be the direct of cause of not receiving treatment for OCD.

To reduce barriers to treatment among African Americans, there are many steps to be taken. Our society must determine better ways of increasing the affordability of treatment. We must educate the public about the treatment process so to demystify how it works, and work to remove the fears of treatment through cultural sensitivity. People with OCD may then be better able to make their own mental health a priority rather than believing there are no options. It remains the mental health community’s responsibility to enact many of these recommendations as they are reasonable interventions that have a great potential to aid this invisible population.

Table 2.

BTQ Comparisons by Ethnic Group Using Fisher’s Exact Test

Question	White	Black	p
1. Health insurance would not cover	41.67%	46.48%	.541
2. Wanted to handle it on own	51.85%	64.79%	.093
3. Didn't think treatment would work	44.44%	49.30%	.543
4. Received treatment before that didn't work	38.89%	12.68%	.000**
5. Worried about cost	54.63%	53.52%	1.00
6. Worried about what people would think	38.89%	50.70%	.126
7. Afraid of criticism by family if sought psychiatric help	36.11%	40.85%	.533
8. Felt ashamed of problem	50.93%	47.89%	.760
9. Felt ashamed of needing help	57.41%	56.34%	1.00
10. Problems with transportation or scheduling	23.15%	36.62%	.063
11. Unsure about who to go see or where to go	50.00%	76.06%	.001**
12. Too inconvenient or will take too much time	25.93%	38.03%	.099
13. Could not get an appointment	15.74%	14.08%	.833
14. Scared about being involuntarily hospitalized	23.15%	29.58%	.383
15. Not satisfied with services that were available	27.78%	18.31%	.158
16. Would be treated unfairly because of race or ethnicity	7.41%	22.54%	.006*
17. Communication concerns because of language barriers	3.70%	8.45%	.198
18. Could not choose provider wanted to see	22.22%	15.49%	.336
19. Uncomfortable discussing problems with professional	25.93%	38.03%	.099

^*Significant at the p<.05 level

^**Significant at the p<0.0026 level (Bonferroni-adjusted)

White= European American participants from Marques et al. (2010)

Black = African American participants from the current study