8   Graduate Student Journal of Psychology Copyright 2010 by the Department of Counseling & Clinical Psychology 2010, Vol. 12 Teachers College, Columbia University Race-Matching in Psychotherapy: Findings, Inconsistencies, and Future Directions Jerren C. Weekes University of Cincinnati Race-matching has been proposed as a viable approach for increasing mental health treatment utilization and enhancing treatment outcomes among ethnic minorities. However, consistent empirical support for race-matching has been lacking since the initial investigations in the 1960s. The variability in findings may result, in part, from methodological inconsistencies across race-matching studies. The purpose of this literature review was to examine the methodology and inconsistent findings among race/ethnicity-matching articles published between 1990 and 2009. It was determined that race-matching findings differed according to the outcome of interest and analytic strategies. Based on the findings and gaps in the literature, recommendations were provided in an effort to facilitate scientific progress and improve treatment outcomes for ethnic minorities. Race -matching in psychotherapy (the coupling of client and therapist based on self-reported race/ethnicity) 1983 and has continued to be an issue of interest. The controversy stems from the disagreement between supporters of intra-cultural and cross-cultural counseling. According to Atkinson (1983), the cross-cultural perspective posits that culturally sensitive therapists can provide effective care regardless of racial and ethnic differences with their clients, whereas supporters of intra- cultural counseling assert that counseling is most effective in racially matched psychotherapy dyads. Several studies have been conducted to assess the impact of racial dissimilarity on psychotherapy for ethnic minorities. Findings reveal that racial dissimilarity may serve as a barrier to treatment, as it counteracts the establishment of a strong therapeutic alliance thereby negatively influencing the client-therapist interaction (Terrell & Terrell, 1984; Watkins, Terrell, Miller, & Terrell, 1989). Sue and Sue (1977) note that it is not simply racial dissimilarity that serves as the barrier, but that cultural differences between the therapist and client can lead to poorer communication, understanding, and rapport. In an attempt to address the barrier of racial dissimilarity, race-matching has been proposed as a method to increase ethnic minority treatment use. However, contradictory research findings make it difficult to deduce the efficacy of race-matching, thereby contributing to the ongoing controversy. For example,                                                                                                   Correspondence concerning this article should be addressed to Jerren C. Weekes, University of Cincinnati, Department of Psychology, 4150 Edwards Building One, PO Box 210376, Cincinnati, OH 45221-0376, Fax: 513/556.1904, Email: weekesjc@mail.uc.edu. consider the first race-matching study conducted by Banks, Berenson, and Carkhuff (1967) in which eight Black college undergraduates were coupled with a Black or White therapist. Results indicated that Black clients were willing to return to treatment when racially matched with therapists, but Black clients in racially dissimilar dyads were unwilling to return to treatment. A subsequent study conducted by Cimbolic (1972), in which seventeen Black college freshmen participated in treatment with Black or White therapists, yielded different results. The results from this study did not indicate a client preference or that racial dissimilarity was a deterring factor for the Black clients involved. Several race-matching outcome studies were conducted after the investigation by Banks and colleagues (Atkinson, 1983); however, the findings of research on the impact of race-matching on treatment outcomes remain inconsistent to date. For example, in a study that examined race-matching in a sample of 302 drug users entering an outpatient substance abuse treatment facility (66% female, 43% African American, 30% White, 23% Latino, 4% Asian, M age = 34.1), race-matching was associated with greater ratings of counselor empathy, but was not found to be associated with abstinence from drug use (Fiorentine & Hillhouse, 1999). In contrast, a study investigating the impact of race-matching on substance use in a sample of 96 substance-abusing adolescents in family therapy (43 Hispanic and 43 White, M age = 15.7) found a significant effect such that matched Hispanic adolescents showed a significant decrease in their substance use (Flicker, Waldron, Turner, Brody, & Hops, 2008). The results discussed above are a few examples of the inconsistent findings regarding the impact of race-matching on treatment outcomes and service use. It is believed that methodological inconsistencies between race-matching WEEKES   9   studies may contribute to the contradictory findings. As such, the aim of this article is to (1) serve as a brief review of race-matching literature, (2) identify relevant inconsistencies in the race-matching literature that may have led to contradictory findings, and (3) provide recommendations for research in an effort to facilitate scientific progress. The paper is organized into four sections. The first section summarizes the search results and procedures used to gather relevant literature. The second section serves as a brief overview of the race- matching literature retrieved (aim 1). The third section describes factors that may account for, or contribute to, the contradictory findings found in the literature (aim 2). The last section provides specific recommendations for future research (aim 3). M ethod Search M ethod Pubmed and PsychInfo databases were searched for articles that assessed racial and ethnic match within mental health treatment. Search terms included psychotherapy and either race, racial similarity, race-matching, or ethnic match. The reference lists of retrieved articles were then reviewed to search for publications that did not appear in the initial database search. This analysis included a review of sample information, methods, statistical analysis, and results of the selected articles. Articles included in the current review met the following criteria: (1) assessed the utility of race-matching within mental health care, (2) were written in English, and (3) were published between 1990 and 2009. The third criterion was established to avoid potential overlap with a similar review conducted by Atkinson (1983), in which the prior race-matching literature was reviewed and assessed. Search Results Using these methods, 30 articles were selected and analyzed by the author. Of the 30 articles, 70% (n = 21) assessed the relationships between racial and ethnic match and treatment utilization, 50% (n = 15) assessed treatment outcomes (e.g., symptom improvement), and 20% (n = 6) assessed therapist-patient alliance and treatment satisfaction. Sixty percent (n = 18) of the studies utilized archival data, and 40% (n = 12) used clinical samples. Within these studies, data was primarily gathered at community mental health or university/college counseling centers. Summary of the L iterature Several general themes emerged after a review of the literature. First, race-matching is effective for a variety of ethnic minority groups (e.g., Farsimadan, Draghi-Lorenz, & Ellis, 2007). Second, empirical support has been found for race-matching in child, adolescent, and adult samples (e.g., Jerrell, 1998; Yeh, Eastman, & Cheung, 1994). Third, findings focus on the efficacy of race-matching with regard to increasing service utilization (Fujino, Okazaki, & Young, 1994; Halliday-Boykins, Schoenwald, & Letourneau, 2005), enhancing treatment retention (e.g., Jerrell, 1998), the working alliance, counselor effectiveness, and perceived benefit of therapy (e.g., Farsimadan, Draghi-Lorenz, & Ellis, 2007). Race-M atching in Various Racial and E thnic G roups Multiple studies support the efficacy of race-matching for Asians/Asian Americans (e.g., Fujino, Okazaki, & Young, 1994), African Americans (e.g., Thompson & Alexander, 2006), and Hispanics (e.g., Flicker, Waldron, Turner, Brody, & Hops, 2008). Flaskerud and Liu (1991) assessed race, gender, and language match in a diverse Asian sample and found that ethnic match related to increased treatment duration (i.e., number of sessions attended), and a significant decrease in the drop-out rate for Asian clients. In 2007, Farsimadan, Draghi-Lorenz, and Ellis investigated the utility of race-matching in a multi-ethnic adult sample (N = 100). Race-matching was found to be useful in increasing client-reported therapist credibility, bond/alliance, and treatment outcomes for South Asian, Middle Eastern, Black African and Black Caribbean clients. Race-M atching in Child, Adolescent, and Adult Populations Empirical support for the effect of race-matching for children, adolescents, and adults was also found. Though the number of race-matching investigations with child participants is fewer than those for adolescents and adults, studies (e.g., Jerrell, 1998) have found that children in racially-matched treatments remained in treatment longer than those who were not in racially-matched dyads. A race- matching study conducted by Yeh, Eastman, and Cheung (1994) found that minority adolescents, particularly Mexican and Asian Americans, who were racially matched with therapists were likely to attend more sessions, less likely to drop out of treatment, and more likely to have higher Global Assessment of Functioning (GAF) scores at discharge than adolescents who were not racially matched. Several studies found race-matching to have beneficial effects in adult samples (e.g., Farsimadan, Draghi-Lorenz, & Ellis, 2007; Thompson & Alexander, 2006). For example, a study of an ethnically diverse sample consisting of Asian American, African American, White, and Latino adults found that clients who were racially matched had lower odds of dropping out of therapy than unmatched clients, except for African Americans (Sue, Fujino, Hu, Takeuchi, & Zane, 1991). In the same study, ethnic match was associated with more sessions attended for matched clients than non- matched, and related to improved treatment outcomes for Mexican Americans. RACE-MATCHING IN PSYCHOTHERAPY   10   T reatment Utilization Empirical support for the impact of racial similarity between therapist and client on treatment use has been provided (e.g., Flaskerud & Liu, 1991). For example, a 2005 study investigating racial similarity between therapists and primary caregivers (N = 1711) in multisystemic therapy, a family based intervention s delinquent behaviors, found that racial similarity led to greater treatment enrollment, decreased symptoms, and higher likelihood of meeting treatment goals for the African American, Asian, Latino, and other ethnic minority child/adolescent clients involved (Halliday- Boykins, Schoenwald, & Letourneau, 2005). T reatment Satisfaction and Outcomes Racial similarity between client and therapist has been demonstrated to increase treatment satisfaction for minority clients. For example, race-matching was not found to affect the number of sessions attended, but did impact client perceived benefit of treatment and higher self-reported acceptance of therapist strategies (Thompson & Alexander, 2006). Therapist credibility and empathy ratings have also tended to be higher in matched than non- matched dyads (Farsimadan, Draghi-Lorenz, & Ellis, 2007; Fiorentine & Hillhouse, 1999). Although there is reasonable empirical support for the impact of race-matching on treatment utilization (i.e., increased sessions completed, lower dropout rates), there is a paucity of empirical support for the beneficial effects of race-matching on treatment outcomes. In addition, the use of race-matching has been questioned based on methodological flaws (e.g., measure of race vs. cultural match, which is the better proxy for similarity) and negligible effect sizes (Karlsson, 2005; Maramba & Hall, 2002). While client satisfaction, working alliance, and counselor effectiveness are of great importance, the purpose of psychotherapy is to facilitate change within the client that leads to better functioning and quality of life. To this point, the beneficial effect of race-matching on treatment outcomes has not been consistently supported. Given these equivocal findings, it is difficult to determine exactly what researchers and clinicians are to deduce from the assortment of empirical results surrounding the effects of race-matching. Inconsistencies in Research M ethodology: M ajor Contributing Factors Despite evidence for the positive effects of race matching, the overall findings, particularly those surrounding treatment outcomes, have been inconsistent. Variability in how previous race-matching studies have been conducted may account for these conflicted findings. Two specific factors may be related to the discrepant findings: 1) variability in outcome variable(s) and 2) analytic strategy. First, race-matching findings differ depending on whether the researcher is studying treatment utilization, treatment outcomes (e.g., substance use or externalizing behaviors), or therapeutic alliance/treatment satisfaction. It is clear that the aim of a study assessing the effect of race- matching on treatment use, retention, and client functioning (Yeh, Eastman, & Cheung, 1994) largely differs from the focus of a study assessing the impact of race-matching on patterns of diagnosis (Mathews, Glidden, Murray, Forster, & Hargreaves, 2002). The outcomes of these studies cannot be compared for multiple reasons, principally, dissimilar study aims and outcome variables. Researchers must take note of differing aims of race-matching investigations, especially those who will be conducting meta-analyses in the future. Second, race-matching findings differ according to the analytic strategy used. This occurs as a result of collapsing ethnic minority groups instead of conducting separate analyses for each racial and ethnic group represented in the sample. While the former is likely due to an insufficient number of subjects in each identified racial and ethnic group, combining different racial and ethnic minorities groups into one category is not best practice (Okazaki & Sue, 1995), as it does not yield detailed information about the impact of race-matching on treatment outcomes for independent ethnic minority groups. Where race-matching could be more effective for one racial minority group than another, combined (i.e., collapsed) analyses may not reveal that valuable information. Collapsing ethnic categories may be done to preserve statistical power, but it ignores the heterogeneity among ethnic minority groups (Burlew, Feaster, Brecht, & Hubbard, 2009). Thus, it is possible that observed empirical differences in treatment outcomes may be due to methodological strategy rather than actual group differences. Recommendations and Future Directions The inconsistencies in the existing race-matching literature have important implications for future research in this area. Additionally, several identified gaps in the literature should be addressed: 1. A meta-analyses examining effect sizes for differential outcome variables, client characteristics, and analytic strategies among race-matching studies would be beneficial to the field. Meta-analyses with inquiry focused upon evaluating the efficacy of race-matching based on specific outcome variables (e.g., the meta-analysis by Maramba and Hall in 2002) or client characteristics will assist the field in coming to an understanding of the efficacy of race-matching. 2. Combining individuals of different racial and ethnic backgrounds into one group for analysis should be WEEKES   11   avoided to steer clear of inaccuracy and overgeneralization. Separate analyses for each racial group represented may prove to be most beneficial. Qualitative interviews foc thoughts surrounding race-matching should be conducted with clients of varied racial and ethnic backgrounds, ages, and presenting mental health that have been overlooked thus far. 3. Future studies should assess the extent to which the effects of race-matching differ based on treatment modality. Mental health treatment modalities (e.g., case management, group/family therapy, and variants of individual therapy) differ in fundamental assumptions, objectives, and processes. As such, the findings may differ in certain treatment contexts due to the unique way in which therapists must interact with clients. For example, the duties and role of a family therapist versus an individual therapist differ. In the family treatment context the therapist must establish rapport with all members present, understand the family dynamics, process the family interactions, and enter into the family system in order to elicit change. The family unit may accept therapists more easily if racial similarity, and a deeper level of cultural understanding, exists. Since treatment modalities differ in fundamental processes, it is possible that racial similarity may be more important in certain treatment contexts than others. Research is needed to deduce more about the importance of treatment modality as it impacts the relationship between race-matching and treatment outcomes. 4. Future studies should also assess the extent to which the effects of race-matching are moderated by client age and social-cognitive development. Yeh, Eastman, and Cheung (1994) found race-matching to be effective in increasing session attendance, decreasing treatment drop-out, and increasing the Global Assessment of Functioning (GAF) scores for adolescent minority clients in their sample. However, race-matching was not found to be effective in any way for the children (ages 6- 11) within their study. Jerrell (1998) found that race- matching was effective in increase treatment use of ethnic minority children; however, treatment outcomes (e.g., symptom improvement, improvement in overall functioning, or a decrease in externalizing behaviors) were not included as variables of interest. Yeh, Eastman, and Cheung (1994) suggested that further research be conducted to determine if and why race- matching has differing effects on child outcomes than adolescent and adult outcomes. 5. The differential effects of race-matching may also be attributed to the varied social-cognitive development between children, adolescents, and adults regarding race perception, bias, and out-group (i.e., racially dissimilar groups) derogation (Cameron, Alvarez, Ruble, & Fuligni, 2001; Quintana, 1998). In addition, as a person progresses in his/her racial identity development the preference for a same-race therapist may become more salient. For example, Parham and Helms (1981) discuss the relation of same-race counselor preference in the context of Black racial identity development. Black/African American individuals in pre-encounter stage, a stage in which American culture/ideals, were more likely to have a strong preference for White therapists and non-acceptance for Black therapists. Preferences differed, however, for individuals in the latter stages of the racial identity model. Given the statements above, it is possible that racial similarity of a therapist may increase in importance as children advance in age and development. Future studies should seek to evaluate the importance of race-matching in child and adolescent populations, and conduct analyses to assess whether age and developmental stage act as moderator variables. 6. Studies examining the utility of race-matching within underrepresented racial and ethnic minority groups can provide valuable information and should be considered. As the United States continues to diversify, the issues of race become more than just Black and White. An increase in race-matching studies for other ethnic groups (such as Hispanic, Pacific Islander, Alaskan/Native American, and Middle Eastern) is needed. For example, only one reviewed study included Middle Eastern ethnic groups into race-matching analyses (Farsimadan, Draghi-Lorenz, & Ellis, 2007). Studies should seek to recruit individuals from the aforementioned racial groups in an effort to provide information on client preferences and effective components of psychotherapy for these underserved and increasingly populous groups. 7. Race-matching studies should include cultural similarity, acculturation, and racial identity measures in an effort to better assess client-therapist cultural similarity and understand the pathway(s) through which race-matching has its impact. To date, studies have not included those measures. This may be due to the fact that a good portion of race-matching analyses utilized archival data in which racial identity, acculturation, and cultural commitment measures were not a part of the initial study design. Including such measures will allow for greater understanding of the dynamics of cultural similarity and sensitivity that are particularly helpful in the therapeutic process. 8. Additional variables that relate to race-matching and treatment outcomes should be investigated. Extant race- matching literature has not fully described the factors associated with the success or limited effect of race- matching. There are several important variables that have not been investigated that will prove helpful in delineating RACE-MATCHING IN PSYCHOTHERAPY   12   possible mediator or moderator variables. Variables including, but not limited to, socioeconomic status, family history (e.g., parental or sibling drug use or legal problems) or family conflict, and the severity of client problems (e.g., severity of externalizing behaviors) are worthy of exploration. Race-related research has revealed a distrust of professional health services that can deter ethnic minorities from seeking psychological treatment (Terrell & Terrell, 1984; Whaley, 2001). Due to numerous factors (e.g., distrust, stigma), ethnic minorities have been noted to largely underutilize mental health treatments (U.S. Department of Health and Human Services, 2001). Race- matching has been proposed as a method to increase ethnic minority treatment use, treatment retention, and satisfaction with treatment, while associated research has been utilized to gain a better understanding of the effective factors in psychotherapy for ethnic minorities. While this line of research appears promising, the inconsistent findings make it difficult to glean a clear understanding on the efficacy of race-matching. Race-matching findings differed based on the outcome assessed (i.e., treatment satisfaction, treatment use, treatment outcomes) and analytic strategy used. It is believed that the noted inconsistencies account, in part, for the discrepancy in findings. The recommendations above are meant to facilitate scientific progress and advance knowledge regarding race-matching in psychotherapy. It should be noted that extreme race-matching practices (i.e., automatically race-matching without consulting the client first) is not advocated here. Racial identification and acculturation vary from individual to individual (Karlsson, 2005), and there are multiple factors that contribute to identity development racial and ethnic factors notwithstanding. As such, there are within-group differences regarding the facet of identity individuals and age may be more important than race for some minority clients. Race-matching should not be imposed on clients, but considered in an effort to increase minority client satisfaction, treatment engagement and retention, and to address mental health disparities that currently exist. Lastly, it is fully recognized that agencies across the United States may not have diverse representation in staff to meet the request for race-matching in psychotherapy. The lack of minority professionals may stem from the lack of minority representation in higher education, specifically clinical psychology programs (Maton, Kohout, Wicherski, Leary, & Vinokurov, 2006). While the field may not realistically be able to provide racially similar therapists to all clients who request one, it is important to reiterate that race-matching is not the end point. This line of research is a pathway to improve understanding of the therapeutic dynamics needed to provide better services to ethnic minorities. Research must continue, as it can inform the training of therapists from all racial and ethnic backgrounds on how to best work with racially and ethnically dissimilar clients. As the inconsistencies are clarified, and gaps in the literature filled in, significant gains will emerge in regards to the nature and quality of psychotherapeutic services provided to racial and ethnic minorities. References Atkinson, D. R. (1983). Ethnic similarity in counseling psychology: A review of research. The Counseling Psychologist, 11(3), 79-92. Banks, G., Berenson, B., & Carkhuff, R. (1967). The effects of counselor race and training upon counseling process with Negro clients in initial interviews. Journal of Clinical Psychology, 23, 70-72. Burlew, A. K., Feaster, D., Brecht, M., & Hubbard, R. (2009). Measurement and data analysis in research addressing health disparities in substance abuse. Journal of Substance Abuse Treatment, 36, 25-43. Cameron, J. A., Alvarez, J. M., Ruble, D. N., & Fuligni, A. outgroups: Reconceptualizing research on prejudice. Personality and Social Psychology Review, 5(2), 118-128. Cimbolic, P. (1972). Counselor race and experience effects on Black clients. Journal of Consulting and Clinical Psychology, 39, 328-332. Farsimadan, F., Draghi-Lorenz, R., & Ellis, J. (2007). Process and outcome of therapy in ethnically similar and dissimilar therapeutic dyads. Psychotherapy Research, 17, 567-575. Fiorentine, R., & Hillhouse, M. P. (1999). Drug treatment effectiveness and client-counselor empathy: Exploring the effects of gender and ethnic congruency. Journal of Drug Issues, 29(1), 59-74. Flaskerud, J. H., & Liu, P. Y. (1991). Effects of an Asian client therapist language, ethnicity and gender match on utilization and outcome of therapy. Community Mental Health Journal, 27, 31-42. Flicker, S. M., Waldron, H. B., Turner, C. W., Brody, J. L., & Hops, H. (2008). Ethnic matching and treatment outcome with Hispanic and Anglo substance-abusing adolescents in family therapy. Journal of Family Psychology, 22(3), 439-447. Fujino, D. C., Okazaki, S., & Young, K. (1994). Asian- American women in mental health system: An examination of ethnic and gender match between therapist and client. Journal of Community Psychology, 22, 164-176. Halliday-Boykins, C. A., Schoenwald, S. K., & Letourneau, E. J. (2005). Caregiver-therapist ethnic similarity predicts youth outcomes from an empirically based treatment. Journal of Consulting & Clinical Psychology, 73(5), 808-818. WEEKES   13   Jerrell, J. M. (1998). Effect of ethnic matching of young clients and mental health staff. Cultural Diversity and Mental Health, 4(4), 297-302. Karlsson, R. (2005). Ethnic matching between therapist and patient in psychotherapy: An overview of findings, together with methodological and conceptual issues. Cultural Diversity and E thnic Minority Psychology, 11(2), 113-129. Maramba, G. G., & Hall, G. C. N. (2002). Meta-analyses of ethnic match as a predictor of dropout, utilization, and level of functioning. Cultural Diversity and E thnic Minority Psychology, 8(3), 290-297. Mathews, C. A., Glidden, D., Murray, S., Forster, P., & Hargreaves, W. A. (2002). The effect on treatment outcomes of assigning patients to ethnically focused inpatient psychiatric units. Psychiatric Services, 53, 830-835. Maton, K. I., Kohout, J. L., Wicherski, M., Leary, G. E., & Vinokurov, A. (2006). Minority students of color and the psychology graduate pipeline: Disquieting and encouraging trends, 1989-2003.American Psychologist, 62(2), 117-131. Okazaki, S., & Sue. S. (1995). Methodological issues in assessment research with ethnic minorities. Psychological Assessment, 7, 367-375. Parham, T. A., & Helms, J. E. (1981). The influence of Journal of Counseling Psychology, 28(3), 250-257. Quintana, S. M. (1 understanding of ethnicity and race. Applied & Preventive Psychology, 7, 27-45. Sue, D. W., & Sue, D. (1977). Barriers to effective cross- cultural counseling. Journal of Counseling Psychology, 24(5), 420-429. Sue, S., Fujino, D. C., Hu, L., Takeuchi, D. T., & Zane, N. W. S. (1991). Community mental health services for ethnic minority groups: A test of cultural responsiveness hypothesis. Journal of Consulting and Clinical Psychology, 59, 533-540. Terrell, F., & Terrell, S. (1984). Race of counselor, client sex, cultural mistrust level, and premature termination from counseling among Black clients. Journal of Counseling Psychology, 31(3), 371-375. race and African American therapy. Psychotherapy: Theory, Research, Practice, Training, 43(1), 99-110. Watkins, C. E., Jr., Terrell, F., Miller, F. S., & Terrell, S. L. (1989). Cultural mistrust and its effects on expectational variables in Black client White counselor relationships. Journal of Counseling Psychology, 36(4), 447-450. Whaley, A. (2001). Cultural mistrust: An important psychological construct for diagnosis and treatment of African Americans. Professional Psychology: Research and Practice, 32(6), 555-562. U.S. Department of Health and Human Services. (2001). Mental health: Culture, race, and ethnicity A supplement to mental health: A report of the Surgeon General. Rockville, MD. Yeh, M., Eastman, K., & Cheung, M. K. (1994). Children and adolescents in community health centers: Does the ethnicity or the language of the therapist matter? Journal of Community Psychology, 22,153-163.