Preferences for Relational Style with Mental Health Clinicians: A Qualitative Comparison of African American, Latino and Non-Latino White Patients

Effective mental health treatment depends upon good communication and productive interactions between patients and mental health providers (Wills & Holmes-Rovner, 2006). Tailoring mental health services to address the needs and preferences of a diverse clientele is a central tenant of patient-centered care, and has been shown to be associated with engagement in care (Ryan & Lauver, 2002), increased adherence (Thompson & Scott, 1991) and improved outcomes of care (Lin et al., 2005). Good patient/provider communication and ease of interaction may help reduce differential health outcomes across racial and ethnic groups (Cooper et al., 2003b; Cooper, Beach, Johnson, & Inui, 2006). Identifying differences in communication patterns between clients and psychologists by race and ethnicity is a central recommendation made in the American Psychological Association’s (APA) Guidelines on Multicultural Education, Training, Research, Practice and Organizational Change for Psychologists (American Psychological Association, 2003). Understanding how diverse patients may differ in their expectations for care and preferences for relational style is a critical component of developing models for patient-centered partnerships that can reduce disparities in health and mental health service delivery (Tucker et al., 2007; Tucker et al., 2003).

What Do We Already Know About Differences in Relational Preferences?

Understanding and being responsive to subtle but important differences in relational preferences may be particularly important for some ethnic and racial minority patients in order to facilitate engagement in mental health care. Psychiatric epidemiologic studies of large nationally representative samples have demonstrated that, compared to non-Latino Whites, racial and ethnic minorities have lower rates of mental health service use by specialty mental health providers (defined as psychiatrists, psychologists, counselors and social workers seen in mental health settings) (Alegria et al., 2002; Snowden & Yamada, 2005). Identification of similarities and differences in relational preferences across groups could improve patient-provider interactions across race and ethnicity so that the communication problems that lead to differential access and lower quality of mental health care can be avoided (American Psychological Association, 2003).

Large studies using quantitative methods have identified several differences in preferences for delivery of care across groups, which may be driven by differences in the counseling and treatment styles practiced by varied professions. For example, a nationally representative study of 2107 African Americans found that this group utilized informal care from ministers more often than formal mental health care provided by psychologists and others within the specialty mental health sector (Neighbors, 1985). A telephone survey of 829 adult primary care patients found that African Americans were less likely and Latinos more likely than non-Latino Whites to find counseling an acceptable treatment for depression, and African Americans and Latinos were more likely to think that counseling brought up too many bad feelings than non-Latino Whites (Cooper et al., 2003a).

Clinical observations expand upon the findings from these large studies. For example, African Americans may prefer a more “interpersonal” approach focused on social interaction rather than an “instrumental” approach focused on problem-solving, particularly during the early stages of engagement in treatment (Jenkins, 1991). Latino culture stresses the importance of “personalismo”, where personal interactions are valued; hence, mental health providers who are interpersonally friendly and openly express how they value the patient as a person may be preferred to those with a more distant, clinical style (Gloria & Peregoy, 1996; Kennedy, 2003). A clinical case study of three patients, one each from a Latino, African American and Asian American background, concluded that preferences for particular communication styles in a mental health encounter varied (Sullivan, Harris, Collado, & Chen, 2006) based upon the different cultural experiences of the patients.

What is missing from the literature are in-depth qualitative research investigations that compares what matters to patients from different racial and ethnic backgrounds during a clinical encounter with a mental health provider. Although qualitative studies have examined relational qualities important in therapeutic relationships, these studies have not considered differences across race and ethnicity (Tunner & Salzer, 2006; Buck & Alexander, 2006; Johansson & Eklund, 2003; Kirsh & Tate, 2006; Ware, Tugenberg, & Dickey, 2004). In many cases, the broad themes identified in the studies are similar, but the analyses reveal variability in the content embedded within these thematic categories. For example, attention to the role definition of the practitioner or case manager was identified as an important factor in two studies of relational preferences among patients with schizophrenia, but the content varied. In Tunner and Salzer (2006), patients described professionalism in part as efficiency, being on time, and working well within the expectations of a clinical encounter. In contrast, Ware et al. (2004), found that patients valued the ways in which practitioners stepped outside of the professional constraints of their role (e.g., getting a cup of coffee, sharing a ride with the patients). Such differences may be due, in part, to variation in the demographic profiles of the respondents included in these studies (Ware et al., 2004). Hence, they underscore the importance of both identifying the broad theme of professionalism, but also looking at how this theme may differ across demographic characteristics of the sample, such as race and ethnicity.

Understanding the Meaning of Relational Preferences across Racial/Ethnic Groups Using Qualitative Methods

Qualitative methods allow us to identify less explicit components of preferences for styles of communication and describe subtle differences that may have real world relevance to practice but may not be apparent from data collected using structured surveys and quantitative methods of investigation (Cooper et al., 2003a). Quimby’s (2006) argument for the use of qualitative methods in reference to research and practice with African American women is also generalizable to others from diverse backgrounds:

Culturally competent clinical practices and research methods involve developing awareness that seemingly minor actions, such as voice tone, eye contact, and hand gestures, may impede or maximize willingness to accept treatment. If these components of culture are misunderstood, regarded as offensive, or simply not even recognized as communication techniques… consumers and clinicians may fail to establish rapport or sustain positive encounters (p. 864).

Qualitative methods enable us to examine these very factors, and to use this information to inform our understanding of patient-centered care across culture.

The objective of this particular study is to look closely at what African American, Latino and non-Latino White patients with depression say they want from a mental health provider in building a therapeutic relationship. We expand upon previous qualitative work on preferences for relational style by comparing emerging themes specifically across these three racial and ethnic groups. We also expand upon the methods utilized in other qualitative studies by adapting a constructivist qualitative approach (Ponterotto, 2005; Charmaz, 2003) in the second stage of our analysis to identify differences across the groups and to highlight the unique characteristics of relational preferences that emerge from the data. By analyzing a series of in-depth interviews conducted with patients seeking care in public specialty mental health clinics, the overarching objective is to identify differences in the meaning of the themes and constructs of what patients from diverse groups prefer in a relationship with a provider.

Method

Sample Characteristics

This sample was drawn from a larger convenience sample consisting of patients interviewed at eight outpatient mental health clinics in the Northeast. The original study was designed to analyze the process of clinical intake in safety net hospital institutions across race and ethnicity (Alegria et al., 2008). Safety net clinics are health care settings that primarily serve Medicaid or uninsured populations. The clinics used for this study predominantly served low-income respondents from diverse racial and ethnic backgrounds, with approximately 30% of the outpatient service volume attributed to uninsured patients. Both the patients (n=129) and the providers (n=47) were interviewed following the clinical intake. The providers represented a combination of psychologists, psychiatrists, social workers and psychiatric nurses. However, for this analysis, we use only the data from the patient interviews.

To delineate differences across race and ethnicity in what patients said they wanted from their care provider, we chose to hold the mental health diagnosis constant, selecting only cases with a diagnosis of depression (n=51). The subsample included respondents who reported a primary diagnosis of depression from the three ethnic/racial groups studied – 6 African Americans, 23 Latinos and 22 non-Latino Whites. About one third of the subsample was male and 61% were U.S. born. The respondents reported low socioeconomic status, with 61% either unemployed or out of the work force, and 71% reporting an income of less than $15,000 per year. Twelve percent of the subsample had a college education, and 31% reported not finishing high school. Chi-square comparisons found few differences across race and ethnicity (see Table 1), with roughly similar proportions by sex, employment status, education, income and insurance status across the three groups. Significant differences were evident by immigrant status and treatment experience; the Latino group was almost exclusively immigrant (87%) while the African American and non-Latino White groups were all U.S. born. Sixty percent of Latinos reported prior experience in mental health treatment, compared to 100% of African Americans and non-Latino Whites.

Table 1.

Sample Characteristics for Total Sample and by Race/Ethnicity

	African American n=6	Latino n=23	Non-Latino White N=22	Total N=51	p*
Sex
Male	.33	.21	.46	.33	0.241
Female	.67	.79	.54	.67
Nativity
Immigrant	.00	.83	.00	.39	p < .001
US Born	1.00	.17	1.00	.61
Employment Status
Employed	.33	.43	.36	.39	0.214
Unemployed	.50	.09	.27	.22
Out of Labor Force	.17	.48	.37	.39
Income
Income <15K	.83	.74	.64	.71	0.143
15K – 34K	.00	.22	.18	.18
35K – 74K	.00	.04	.14	.08
>=75	.00	.00	.04	.02
Missing	.17	.00	.00	.02
Insurance Status
Private	.00	.09	.09	.08	0.773
Medicare	.00	.26	.18	.20
Medicaid	.67	.39	.50	.47
Free Care/Other	.33	.26	.23	.25
Education
<=11 years education	.33	.48	.14	.31	0.281
12 years	.33	.13	.27	.22
13-15 years	.17	.30	.45	.35
>=16 years	.17	.09	.14	.12
Treatment History
Prior Treatment	1.00	.70	1.00	.86	p < .01
No Prior Treatment	.00	.30	.00	.14

^*p value from Chi-square test of differences across groups

Data Collection

Research interviews were conducted separately with patients using a semi-structured interview guide, immediately following the clinical intake session, lasting an average of about thirty minutes. Interviews were conducted in English or Spanish according to participant preference, and took place in a private office in the clinic where the intake had taken place. These interviews focused on understanding patients’ experience during the initial clinical interview¹ and were conducted by trained research assistants. Training was provided by an expert in medical ethnography on a regular basis. All study procedures were approved by the Institutional Review Boards of the clinics where the patients were interviewed.

Analytic Strategy

We conducted a content analysis of the responses to the qualitative in-depth interview guide that focused on answering the following questions: “What advice would you give to providers who want to help people with your problem/concerns? What advice would you give the provider on how to build a relationship with you?” Although the interview was conducted following the clinical intake, the respondent was asked to describe what advice he or she would give a provider in general. Hence, the questions were designed to elicit open-ended responses that reflected the preferences of the patients for the relational interaction they might desire in an ideal and hypothetical situation.

The first phase of the analysis utilized an open coding strategy based on inductive approaches which involve looking closely at the data first via a broad coding process, and allowing these data to inform the identification of themes in the text (Ryan and Bernard, 2003). The lead author of this study, a non-Latino white researcher trained in mental health services research, coded all interviews, identifying words and phrases that were used to describe relational preferences. Forty-eight initial codes were grouped together under primary codes that represented the broader patterns emerging in the discrete codes. For example, some common descriptions that occurred at this early stage of coding included being reassuring, not over-talking or interrupting, and paying close attention. Once these discrete codes were examined more closely, they all grouped under the first descriptive code in the series of “listening,” as the descriptors referenced what the respondent said made them feel listened to. Six primary codes were identified: listening, understanding, managing differences between patient and provider, spending time with the patient, connecting to the patient and helping the patient.

Next, we conducted a contextualized comparative analysis looking at the six primary themes to identify how each of the three racial and ethnic groups assigned meaning to the theme through its descriptive use in the context of preference elicitation (Ryan & Bernard, 2003). The uncoded transcripts were separated by the African American, Latinos and non-Latino White patient groups and coded separately by group. There were two objectives for this second stage of coding. First, we conducted a cross-check of the trustworthiness of the original primary themes through this recoding process by incorporating three unique coders in the analysis and assessing the credibility of the findings through reflexivity within our group on the assumptions and content of the original themes (Morrow, 2005). Through this process, four of the original primary themes were confirmed as robust and credible representations of what the patients wanted from the providers: listening, understanding, managing difference and spending time. The last two primary themes – connecting to the patient and helping the patient - were judged to be theoretically embedded in all of the first four themes, not unique representations of relational preferences, and were not included in the final analytic stage. New themes did not emerge in this process, indicating that we had reached saturation of the constructs in the content in this text.

Secondly, we identified differences in how the respondents represented the meaning of these themes within their particular context (e.g., across race and ethnicity). To do this, an African American researcher and social worker analyzed transcripts from the African American respondents, and a Latino researcher and social work student analyzed the untranslated Spanish transcripts from the Latino respondents, as well as any transcripts from Latino patients that were conducted in English. The lead author, who is non-Latino White, analyzed only the non-Latino White transcripts. In this stage of the analysis we applied a constructivist qualitative approach (Ponterotto, 2005; Charmaz, 2003) which conceptualizes qualitative research from the perspective of a mutual creation of knowledge through the inherent subjectivity of the research participant and the analyst; hence, the individual experiences and perspectives stemming from the unique racial/ethnic background of our coders was instrumental to the design of this stage of the analysis (Sandelowski, 1998). Our objective was to uncover a richness of detail through the individual viewpoints of the coders, confirmed through dialogue among the collaborators (Graneheim & Lundman, 2004). Distinctions in how the primary themes were described across the three racial and ethnic groups were identified, summarized and discussed by the analysts in multiple meetings until consensus about these interpretations were reached. In addition, the findings were discussed with four clinicians who work with multicultural populations, as well as two qualitative researchers. In this way, we aimed to enhance a deep understanding of cross-ethnic variation in meaning attributed to the shared themes of listening, understanding, spending time and managing difference.

Overall, the analysts maintained the original phrasing that the respondents used in their descriptions, with the interpretation tied as closely as possible to the primary text. However, the theme of “managing difference” represents a group of codes that described how patients wanted providers to behave, given cultural, ethnic, racial and professional differences; the attribution of these codes to the theme of “managing difference” was an interpretive decision made through discussion amongst the analysts.

Results

Throughout the transcripts the four primary thematic categories - listening, understanding, managing difference between patient and provider, and spending time with the patient - were overlapping but distinct. In most cases, these themes emerged across all racial and ethnic groups, with the single exception of the “spending time” theme for African American patients which did not emerge. This thematic consistency illustrates how the basic yearning for authentic connection with a provider transcends racial categories. Nonetheless, how these yearnings were expressed across race and ethnicity showed marked differences in the way the respondents described the experience of connection with and help from the provider that was facilitated by these qualities (see Table 2).

Table 2.

Summary of Themes by Race/Ethnicity

Themes	Contextual subthemes by racial and ethnic group
	African American patients	Latino patients	Non-Latino White patients
Listening	To listen to who the patient really is; to recognize the patient is the expert on him or herself, “knows” him or herself best.	To listen as a way to communicate “attention”; as an expression of the quality of the relationship.	To listen so that the patient is “comfortable” enough to talk and express feelings.
Understanding	To understand the patient beyond immediate impressions; to understand aspects of the patient that cannot be seen.	To understand the feelings of the patient.	To understand the complexity of the patient’s choices and life circumstances.
Spending Time	This subtheme did not emerge for the African American patients.	To take enough time to connect more deeply.	To take enough time so that feelings are not forced and emerge at their own pace.
Managing Difference	To actively work against assumption of difference; provider should “lower” self to where the patient sits, at his/her level.	To embrace authoritative provider role; but, provider should connect first and then offer concrete advice, feedback and “solutions.”	To not “judge” due to social difference; to maintain professional distance but still be human.

Listening

For the African American patients, the initial work of a mental health provider was described as the willingness to really know the patient on his or her own terms. Listening was seen as a vehicle for connecting across differences, and engaging with patients as they see themselves. It was described as the responsibility to not “override or over-talk,” and discussed in terms of recognizing that the patients are the experts on themselves:

African American female (406): You know, just listen to me and listen to where I’m coming from. Because I know what I’m talking about, I’m not crazy, you know what I mean? Just listen to me, have trust in me, have faith in me, because I know what I’m talking about.²

For the African American patients, the emphasis was on the responsibility of the provider to listen in a way that showed the patients they trusted their own knowledge about themselves.

For the Latino patients, the way the respondents described how they knew that the provider was listening was different from the African American patients. The Latino patients, in general, described the subtle components of what made them feel listened to in greater detail than the other racial and ethnic groups. One female respondent (139) described: The doctor “sat there all comfortable, crossed his leg and was ready there for me… I think that’s the key as well, a relaxed provider… to listen to the problems of others and forget about their own problems.” Another described someone who knew how to make them feel listened to as a person who looks them in the eyes, and criticized the lack of attention she felt from a previous therapist who she had stopped going to see:

Latino female (125): I mean, to be honest I never felt that I was listened to, on the contrary at times I felt bored because I was talking with her and she was like “Yeah, hmm-mmm.” She almost never had questions for me, but she would just write, and she never told what she was writing; I only saw her writing but she never looked me in the eye.

Here, the nonverbal messages communicated by not looking at the respondent in the eye contributed to the sense that this respondent was not really being listened to. Listening was often described in terms of how providers communicated “attention” to patients in the session. These qualitative, non-verbal components of listening were of great importance to the Latino patients in the process of developing a strong relationship with their providers.

For non-Latino Whites, the descriptions of how providers helped them through listening were less detailed. The words used by these patients to describe the listening process were communicated with more emotional detachment than the African American and Latino patients. Throughout the transcripts, the word that non-Latino Whites used most frequently to describe how providers showed they were listening was through their ability to make the patient feel “comfortable”. As one non-Latino white female (125) said: “You know, listen to them… continue doing the things I feel comfortable with, like listening to me… if someone listens to me and, I see that they’re sincere and that they can give me advice, I’m comfortable.” For non-Latino White patients, the process of connecting through listening was described in a calm way, where the desired goal was not a detailed expression of the quality of the relationship as observed with the Latino patients but instead a light and casual process of slowly getting to know a provider: as one non-Latino White describes (601): “[Providers should engage in] making-easing conversation, like keeping things on a conversational level.” For these patients, providers demonstrated listening through creating a space that was comfortable, safe, and almost casual.

Understanding

For all three groups, the desire to be understood by providers was strong, and frequently tied closely to the experience of being listened to. For the African American patients, this theme meant understanding where the person was coming from beyond immediate impressions, such as assumptions based on the color of their skin. One African American female respondent described how important it was for providers to try to understand less obvious personal qualities:

African American female (208): It would be helpful to ask a person, what is your religious belief?… because I think once the clinician has a view of you, they see that you are African American or they see that you’re Asian or they hear that you have an accent so you’re from a different culture. But spirituality is something you can’t see.

She continued to discuss how important it was to know about her spirituality and education, the things that were not immediately seen on the surface, in order to be understood. She talked about how even though she was unemployed, she was using this time to her benefit to volunteer and learn more about herself. She described how important it was for the provider to understand the context of these descriptive facts about her. For the African American patients, understanding by providers was intimately linked with their ability to see beneath immediate appearances, and patients felt that providers had the responsibility to seek to understand them at that level.

What was important for providers to understand was described differently by the Latino patients. Latinos discussed the importance of the providers going in depth to truly understand their feelings:

Latino female (139): That they not only present an issue just to want to know what is happening, but that they go in depth a little bit more; that they try to understand how the patient feels, not only what the patient is saying about what is happening to them…

The focus of understanding was going beyond just the words that were being said or the facts of the situation to touch deeply on the patient’s emotions. The Latino patients used the phrase “to deepen” or “to go more in depth” (profundizar) multiple times to describe what it means to really understand and connect with a provider. Understanding was synonymous with relating to how the person feels.

For non-Latino White patients, the desire to be understood was expressed in terms of providers’ understanding the complexity of the patients’ particular situation and not judging them based on the circumstances of their life. For the African Americans, being understood had meant that the provider knew them beneath surface appearances. In contrast, non-Latino Whites described being understood in terms of accurately perceiving differences in life experiences, not racial differences. One female respondent (209) discussed wanting the provider to understand her experience of domestic violence and to help her not to be ashamed of it. Another non-Latino White male who was involved in criminal activity described wanting a provider to understand him:

Non-Latino White male (332): Just take everything with a grain of salt because you know I know it is easy for some to say “Why don’t you just walk away?” but it’s not that easy, you know? There’s a lot of things that you got to take into consideration. People who are going to be left down the food chain that ain’t gonna survive because you’re not there.

Here, it was important to the patient for the provider to appreciate the complexity of this patient’s situation, as leaving a life of crime would have repercussions on his ability to support his family. Whereas African Americans’ concern that the provider would assume difference and not really understand them was reinforced by racial differences, the differences for non-Latino Whites were less explicit. Hence, the non-Latino Whites were more concerned that the provider might assume similar circumstances, not understanding the complexity of the choices they had been forced to make in their lives.

Managing Difference

Throughout the interviews, African American respondents assumed difference between themselves and their providers, and expressed that providers needed to bridge these differences in order to connect. In contrast to the previous theme of understanding, where patients described what was important for a provider to understand about them, “managing difference” represented the patients’ views on how the provider should actually behave in light of the differences between them. In one case, an African American male (412) explicitly named that a provider should not “judge someone on the basis of their origin, skin color, or cultural background” as the first advice he would give to a provider. These patients were explicit that building the therapeutic relationship required that the provider manage the socio-cultural and racial differences between them. The respondent above continued, “Most doctors are arrogant, stuck-up, self-centered and have attitudes of wanting to be above their patients.” Providers were thought to see themselves as better than their patients, and the African American patients felt that it was the responsibility of providers to adjust their position in relationship to the patients:

African American male (324): The most important thing is for them to understand the patient. It is important for them to get to the patient’s level, meaning they need to explain things according to their educational level. The patient can’t go up to their level, but they can lower theirs by explaining things in terms the patient would understand.

Here, the expectation was that a provider needed to actively “lower” him- or herself, if necessary, in order to really understand and connect with African Americans seeking mental health treatment.

For the Latino respondents, we found little concern for social difference between patient and providers; in fact, many of the descriptions welcomed a directive, and more authoritative approach on the part of the provider. Latino patients focused on how important it was for providers to engage with their problems, ask questions and suggest concrete, proactive solutions. The descriptions were often framed in interactive terms where the role of a good provider was how a provider asked questions and gave the patient direction: “That they don’t leave you ‘hanging’… don’t leave anyone without solutions” (Latino male, 136). These respondents viewed the provider as knowing more than them, and part of what they wanted out of the patient/provider relationship was for the provider to help problem-solve and suggest concrete advice.

However, there was always the sense that good providers listened first and then provided the more directive advice, feedback or solutions: “To listen and, and to give him/her instructions how to organize his/her, their disequilibrium, if you will” (female Latino, 130). Although some expressed a desire for providers who would not judge, the relationship was always central, and respect for the connection was primary: “always first listen to the person, and not judge the person… not make a determination about something unless the provider is sure” (female Latino, 102). For most Latino respondents, the sense they gave was that as long as the quality of the relationship was strong, they welcomed feedback and input from the provider, whether it was perceived as “judgment” or not.

For the non-Latino White respondents, building a successful therapeutic relationship with a provider meant navigating a situation where the patient was often similar ethnically or racially but different culturally or professionally from the provider. Having a provider who understood these differences and did not judge was a strong criterion for developing a viable therapeutic relationship. How the provider managed the development of a comfortable therapeutic relationship despite professional and cultural differences was of paramount concern for the non-Latino White patients:

Non-Latino White female (417): Don’t be like really judgmental because not all rape victims or abuse victims are the same… like my last therapist, he just really belittled me like put me down because I was homeless and stuff like that… Which made me feel worse than I already was.

Non-Latino White patients also appeared very sensitive to the nature of the relationship with a provider as a professional, and not a friend, and at times appeared to struggle with the professional boundaries and the power dynamics of the patient/provider relationship. One male respondent (427) when asked what advice he would give to a provider on how to build the relationship replied: “I don’t know if that would come up. I don’t know, uh, besides they’re not supposed to anyway… [not supposed to] have too much of a relationship.” Other respondents mentioned how important it was for a provider to “be human” and not too much of a professional in order to connect with them. Still others expressed their desire for a provider who didn’t pry for details or ask too many invading questions, seeming to imply that appropriate distance is important. Wanting a provider who is able to make them comfortable, while also maintaining professional boundaries and managing the social and cultural differences between them, was a consistent theme for the non-Latino White patients.

Spending the Time

For the Latinos, in order to effectively reach a deep level of connection, several patients discussed how important it was for the provider to spend enough time with them. Respondents bemoaned the lack of time in a session for being able to really delve deeper with the provider. According to one Latino female (115), “I would at least like more time, to go deeper, for him to go deeper, that I explain everything… get everything out, you know, and all my anxieties…” The quality of the relationship with a provider was closely tied to his or her ability to fully encourage this expression of feelings and a willingness on the part of the provider to take the time to allow for a deep connection to develop. Some respondents expressed a desire for the provider to ask about their family, and gave the sense that a good provider was one who was able to connect with the personal details of the patient’s life outside of the clinical encounter. One Latino male (711) jokingly said the best way to build a relationship was for the provider to “take me out to Friendly’s and out to drink…”

Spending enough time with the provider also emerged as a theme for the non-Latino Whites; however, the emphasis was on having enough time to access emotions slowly. It was important to spend enough time connecting with the provider but also to maintain the clinical boundaries to contain deep feelings:

Non-Latino White female (204): I think really taking the time and… being able to touch a little bit upon everything, making sure that you are as thorough as possible but then also letting the patient know that there’s time in the future to go deeper into those issues…

Some non-Latino White respondents discussed this dynamic in terms of taking time to tell the provider what they want to tell them and not forcing things out. Others talked about taking the time to feel comfortable with the provider and not be ashamed of their need to get help. Throughout the transcripts for the non-Latino Whites, the narrative used by the respondents when discussing the need to spend enough time with the provider tended towards a desire for a comfortable and safe connection where feelings were slowly allowed to emerge.

Discussion

This analysis did not find differences in the primary themes of what patients wanted in their interactions with a mental health provider. All respondents wanted a good relationship in which they felt listened to and understood. However, how they described these relational qualities, the expressions they used and what they meant by similar words and phrases differed qualitatively across the groups. The findings from this study are consistent with recent studies of communication patterns across race and ethnicity in general health. One study of orthopedic surgeons and patients found no differences in the content of conversations across race and ethnicity but differences in the process of relationship building (Levinson et al., 2008). They are also consistent with a qualitative analysis in primary care that found many similar themes in preferences for care across patients from diverse racial and ethnic groups, but found differences across groups in the specifics of the content of these themes (Tucker et al., 2003). Our study suggests that patients may react to subtle, qualitative cues that communicate the quality of the engagement they are experiencing in a clinical encounter, and attention to these nuanced differences in preferences may be an important component of reducing service disparities for these populations.

The findings from this study also support the recommendations of the APA Guidelines on Multicultural Education, Training, Research, Practice, and Organizational Change for Psychologists (American Psychological Association, 2003), in particular Guideline One which advocates commitment to cultural awareness and knowledge of self and other and Guideline Two which encourages psychologists to recognize the importance of multicultural sensitivity. In both areas, the APA highlights the importance of sensitively taking into account group differences, while at the same time being conscious of one’s own cultural background to avoid unconscious assumptions that could lead to stereotypes. The findings expand upon Guideline Five by carefully explicating how the specific characteristics of communication patterns and variability in preferences for relational style may vary across groups, and inform the way that such an approach could enhance the clinical skill set for clinicians working diverse cultural groups. These results could be replicated and incorporated into training programs for clinicians in order to facilitate discussion and awareness of relational preferences across diverse racial and ethnic backgrounds.

It is important to note that the communication patterns identified in this study can be adapted by clinicians no matter what the race or ethnicity. Significant attention has been paid to the impact of racial/ethnic concordance of patient and clinician for effective care. Cooper and colleagues (2003b) found race concordant visits in general health care were longer and had higher ratings of patient positive affect, suggesting that concordant race may facilitate quality in health care encounters. However, it may also be the case that racial and ethnic concordance is beneficial due to a match in implicit communication patterns shared between individuals from the same backgrounds. By making differences in these implicit communication patterns across race and ethnicity explicit, we inform efforts to enhance training across clinician background and enable better communication matches as a skill set that can be achieved whether one is in a concordant patient/clinician relationship or not (Smith et al., 2007; Beach, Rosner, Cooper, Duggan, & Shatzer, 2007). Recent studies that test the impact of communication skill interventions in health care (Cooper et al., 2009) should inform this debate in mental health care as well.

Many of our study findings are consistent with other descriptive work on differences in relational style across race and ethnicity in health care. The importance for African American patients of being understood as they really are, and having the clinicians trust that they know their situation best, suggests that clinicians may need to proactively work against the assumption of difference in their clinical encounters with this population. These findings are important to place within the context of a history of distrust in the African American population of the mainstream medical establishment (Suite, La Bril, Primm, & Harrison-Ross, 2007; Wasserman, Flannery, & Clair, 2007). Racial biases have been identified whereby physicians judge African Americans as less intelligent and more dangerous than Whites (Bhui, 2001; van Ryn & Burke, 2000). African Americans have been found to view mental health services with suspicion and fear, which can negatively impact engagement in care (Keating & Robertson, 2004).

For Latinos, some have noted a cultural tendency for individuals from this ethnic group to be more interested in knowing others as a person, and less interested in judging someone on external qualities such as socioeconomic status or occupation (Gloria & Peregoy, 1996; Triandis, Marin, & Betancourt, 1984). Our findings are consistent with this cultural approach whereby Latinos seemed less concerned about feeling judged by the provider, and less aware of patient/provider hierarchies. It is important to note, however, that most of the Latino respondents were not U.S. born, and thus it would be important to conduct further research to determine whether these findings are also apparent with more acculturated Latino groups.

One finding from the study that is particularly salient is how aware the African American and non-Latino White respondents felt about the social position of the therapist, and how important it was to the respondent that the provider managed the dynamic so they felt listened to and understood. For the non-Latino Whites, in particular, this sensitivity to the position of the therapist and the nature of the relationship that was possible within this therapeutic context was a source of confusion at times. There exists almost no attention in the literature to the interpersonal barriers that could exist for non-Latino Whites from low socioeconomic backgrounds in engaging with other non-Latino White clinicians who have higher socioeconomic status. The experience of being both like the clinician (in terms of race) but unlike the clinician (in terms of class), and the impact this difference may have on developing a working alliance, is an important area for future qualitative research. Furthermore, given the predominantly low income population included in this study, future research should consider the degree to which the findings on managing difference across the groups may be due to cultural factors determined by socioeconomic status, racial/ethnic identity, or some interaction between the two.

Our findings also provide direct support for the concern raised in the APA Guidelines that traditional therapeutic models may not apply across all cultures. By including non-Latino Whites in the analysis, we demonstrate how preferences for relational style vary across multiple groups and suggest ways that clinicians can adjust communication patterns to respond effectively depending upon these preferences. In particular, the need to maintain boundaries with the clinician as non-Latino Whites became comfortable enough to express their feelings was an important subtheme for this group, but not evident in the African American and Latino patients. Not only does this finding imply that mental health providers should be sensitive to this pattern for non-Latino Whites, but it also highlights the importance that clinicians not assume a similar emotional and relational paradigm for racial and ethnic minority patients. The African American and Latino patients did not express the same reticence towards sharing intense emotions in a clinical setting early in a therapeutic relationship, and mental health providers may need to adjust their practice to allow varying degrees of emotional disclosure early in a clinical encounter, as this could be a factor in engagement in mental health care across the populations.

This study has several limitations. All of these interviews were conducted in health care institutions that mostly serve low-income persons; hence some of our findings may not reflect the attitudes of racial and ethnic groups from higher socioeconomic backgrounds. Most of the Latino respondents were women, were immigrants, and were seen by Latino providers who spoke Spanish with them, so our findings may not apply to more acculturated Latino men and women. The African American sample was much smaller than the other two samples, and thus comparisons with this group should be made with caution. In particular, the finding that the “spending time” theme did not emerge for African Americans should be further investigated and replicated in future studies that include larger numbers of African Americans.

Lastly, it is possible that some patients’ descriptions of what they preferred in a relationship with a mental health provider were potentially influenced by the intake session they had just completed. Given the study design, we were not able to determine the extent to which the intake session had an effect on responses to the preference questions. However, since 86% of the respondents in this sample reported prior mental health treatment, patient preferences for relational style likely developed in response to multiple experiences with the mental health treatment system. We also cannot determine whether attitudes about relational preferences were driven by previous or current experiences with racially concordant or discordant providers. Future qualitative work is necessary in order to examine more closely the provider experiences that drive particular preferences towards and attitudes about relational styles of the provider.

Our findings provide information that mental health clinicians can use to increase their own sensitivity for different styles of engagement in diverse populations, to foster productive patient-centered interactions across race and ethnicity. These findings are not meant to represent rigid prescriptions for clinical behavior, but rather to bring attention to the variability in preferences for relational style across all racial/ethnic groups. For example, given that Latinos often expressed the desire for a deep and expressive quality of relationship, close attention to the range of emotional expression desired in a clinical encounter may help improve engagement in the beginning stages of treatment. Particularly when working with African American patients, clinicians may want to actively seek out situations where they can affirm their understanding of who the patient is beneath surface appearance, and be aware of instances where power differentials might be unintentionally reinforced. Awareness of subtle differences and attention to discomfort due to professional and class distinctions may help facilitate productive interchanges regarding the desire to connect across these differences, especially for some lower income non-Latino Whites. Attention to these differences may help us better understand the sources of mental health service disparities and enhance patient-centered care to reduce them.