Social Identities of Clients and Therapists During the Mental Health Intake Predict Diagnostic Accuracy

Abstract

Across countries, common mental disorders are often more prevalent and/or more persistent among disadvantaged members (e.g., ethnic minorities) compared with advantaged group members. Although these disparities constitute a heavy challenge to national health organizations, there is little empirical evidence to help account for the mechanism underlying them. In this study, conducted in clinics across Israel, we investigated processes, rooted in the clinical encounter that may contribute to mental health disparities. We focused on the accuracy of diagnostic decisions, which are likely to substantially impact the client’s prognosis. Therapists’ diagnostic decisions following the initial intake with their client were compared with independent structured diagnostic interview of the client. Results revealed that therapists were twice as likely to misdiagnose mental illness when their client was a member of a disadvantaged (relative to advantaged) group. Implications for the quality of mental health services that members of disadvantaged groups receive are discussed.

Keywords

social identities mental health disparities Israel mental health intake misdiagnosis working alliance

A frequent finding in the mental health literature is that common mental disorders are often more prevalent and/or more persistent among members of disadvantaged groups compared with members of advantaged groups. Social advantage refers to the relatively favorable sociodemographic, economic, or political conditions that some groups systematically experience based on their relative position in society (Braveman, Egerter, & Williams, 2011; Saguy, Tropp, & Hawi, 2013). For example, in Israel, the prevalence rate of common mental disorders among Mizrahim (Jews of Asian/African descent who are socially disadvantaged relative to Ashkenazim) is double the size of the rate among Ashkenazim (Jews of European/American descent; Nakash, Levav, & Gal, 2013). Although more scarce, research in Europe shows consistent findings, for example, with higher prevalence rates of common mental disorders among Pakistani and Indian immigrants in England compared to Native Whites (Weich et al., 2004). In the United States, although disadvantaged ethnic groups (Hispanics and non-Hispanic Blacks) have lower risk for mood and anxiety disorders relative to advantaged ethnic groups (non-Hispanic Whites), they show higher persistence of these disorders (Breslau, Kendler, Su, Gaxiola-Aguilar, & Kessler, 2005).

Social causation models attribute poor mental health status to chronic adversity and stress among socially disadvantaged groups (Aneshensel, 2009). For example, lower socioeconomic status is associated with exposure to chronic stress, to limited access to health care, and to lower rates of health literacy (Sentell & Halpin, 2006)—which are all predictors of poorer mental health (Braveman et al., 2011). Beyond these structural elements, an additional set of factors that has been identified as contributing to mental health disparities is rooted in the clinical encounter between therapists and ethnically/racially diverse clients (Van Ryn, Burgess, Malat, & Griffin, 2006). These factors center on processes in the client–therapist interaction that can shape the course and outcome of the clinical encounter, and consequently, the client’s mental health. In the current work, we focused on one such process which has received very little empirical attention, that is, the accuracy of the diagnostic decisions of therapists.

Research that considered elements in the clinical encounter as underlining health disparities focused primarily on client-related outcomes. For example, relative to advantaged group members, members of disadvantaged groups in the United States (Hispanics and Blacks) tend to terminate their treatment prematurely and to underutilize mental health services (Cook, McGuire, & Miranda, 2007). Similar findings were documented in countries with national health care laws such as Israel, indicating that Mizrahim underutilize services relative to Ashkenazim (Nakash et al., 2013). Moreover, disadvantaged clients often delay seeking treatment resulting in overutilization of psychiatric emergency services (Snowden, Masland, Libby, Wallace, & Fawley, 2008). Such delay is typically ascribed to mistrust, shown to be based on the expectation of being stereotyped and receiving lower quality care (Burgess, Fu, & Van Ryn, 2004).

Despite growing attention to variables related to clients, little research exists on psychological processes among therapists, and their potential role in the development and persistence of health disparities. The possible role of these processes was highlighted by the U.S. Institute of Medicine’s report (2003), which posited that prejudice and stereotyping play a key role in the development of health disparities. Support for this proposition comes mainly from research in the general medical field. Physicians were shown to be less client centered (Johnson, Roter, Powe, & Cooper, 2004) to engage in less joint decision making (Gordon, Street, Sharf, Kelly, & Souchek, 2006) and to have shorter medical encounters (Siminoff, Graham, & Gordon, 2006) when interacting with disadvantaged, relative to advantaged clients. Furthermore, primary care doctors rated African American clients as less intelligent and less educated than non-Latino White clients (Van Ryn et al., 2006). Tapping into more behavioral outcomes, related studies have shown that racial/ethnic match between therapists (which are typically White) and clients was associated with increased compliance, less dropout rates, and longer visits (Johnson et al., 2004).

Although limited, some literature linked the above outcomes to biases associated with the “mismatch” of social identities during the encounter (Burgess, Van Ryn, Dovidio, & Saha, 2007; Dovidio, Gaertner, Kawakami, & Hodson, 2002; Williams & Mohammed, 2009). In particular, implicit bias, which refers to unconscious negative feelings and attitudes that operate automatically during social interactions that involve discordant identities (e.g., a White physician and a minority client; Greenwald et al., 2002), was theorized to influence the way physicians approach service-users belonging to out-groups (Van Ryn et al., 2006; Van Ryn & Williams, 2003). For example, Green et al. (2007) found that physicians higher in implicit bias were less likely to recommend appropriate treatment for alleged African American patients with myocardial infraction compared with alleged White patients.

None of these studies, however, has considered behavioral outcomes on part of therapists which can potentially impact the course of treatment and ultimately client’s prognosis. In this study, we investigated how the behavior of therapists who belong to advantaged social groups is shaped by the social identity of their client (i.e., either advantaged or disadvantaged). We chose to focus on a behavior with much consequence to a client’s mental health treatment: the diagnostic decision following the initial meeting with the client (i.e., mental health intake). To reach a correct diagnosis, therapists need to conduct an initial interview with the client and inquire into diagnostic criteria necessary for each disorder (Nakash, Rosen, & Alegría, 2009).

Different lines of research suggest that the collection of clinical information might differ as a function of the social identities of therapist and client (Nakash, Dargouth, Oddo, Gao, & Alegría, 2009; Nakash, Saguy, & Levav, 2012; Rosen, Miller, Nakash, Halpern, & Alegria, 2012). When the identities of therapists and clients do not match, processes pertaining to different social categorization may impact the collection and processing of information on part of therapists. First, the different group identities might cause difficulties in understanding and correctly interpreting client’s information (a cultural dysfluency effect). Second, given that such discordant encounters, for the most part, involve a therapist from an advantaged group and a client from a disadvantaged group (and not vice versa¹), processes pertaining to intergroup bias and intergroup power might further come into play. Processes of categorization and associated intergroup biases, which are prevalent in a variety of contexts, might also play a role in such discordant encounters (Green et al., 2007). For example, if a member of a minority group is stigmatized as incompetent, such thinking might drive the processing of information in the intake leading clinicians to overdiagnose the minority group member. Beyond these processes of out-group biases, discordant clinical encounters are also characterized by greater asymmetry in social power, being the product of both the power as a clinician and as a majority group member. The psychological experience of power has been shown to impair perspective taking and accuracy in detecting others’ emotions (Galinsky, Magee, Inesi, & Gruenfeld, 2006), as well as to give rise to heuristics when making attributions of others’ behavior (Fiske, 1998). Such processes may take place in discordant encounters involving greater power asymmetry, potentially resulting in impaired attention leading to underdiagnosis of a minority client.

Taken together, we predicted that socially advantaged therapists will be less accurate when treating clients from a disadvantaged (relative to advantaged) group, and as a result they will have higher rates of misdiagnosis (either overdiagnosis or underdiagnosis). The study was conducted in the field within community clinics across Israel. All measures were based on the initial stage of treatment: the mental health intake. Despite the obvious importance of early sessions in client retention and outcome and in forming diagnostic decisions, a dearth of research exists examining the quality of clinical interaction during the initial intake. The challenges of the intake, and particularly reaching an accurate diagnosis and establishing rapport, are likely to be amplified when cultural differences exist given the unfamiliarity and discomfort often associated with such encounters (Nakash et al., 2009). We, therefore, set out to test our predictions in this setting.

Method

Setting

The current investigation includes a subsample from a larger study on the mechanism contributing to mental health service disparities in Israel (for description of the full study, see Nakash, Nagar, & Levav, 2015). The study was conducted in four public mental health clinics in three large cities in Israel offering mental health services to an ethnically and socioeconomically diverse adult client population. Access to care in these clinics does not necessitate medical referral. At each of the clinics, clients were consecutively allocated to therapists based on therapist availability.

We focused on differences between encounters involving a Mizrahi and Ashkenazi clients. These ethnic groups make up the majority of the Jewish population in Israel and both mostly migrated during the early years of the foundation of the State of Israel. However, consistent evidence from research in public health and the social sciences (Israel Central Bureau of Statistics, 2012; Nakash et al., 2013; Nakash, Gerber, Goldbourt, Benyamini, & Drory, 2013; Saguy, Dovidio, & Pratto, 2008) shows that inequality between these groups is apparent in levels of education (Ashkenazim have three additional years of education on average compared with their counterparts), average income (Mizrahi families earn 85% of the income of their counterparts), and mental health status (prevalence rates of mood and anxiety disorders are twice higher among first and second generation Mizrahi respondents). Also, Mizrahi group members tend to experience more implicit prejudice and negative stereotyping (Nakash et al., 2012; Saguy et al., 2008).

Participants

A convenience sample of therapists and clients participated in the study. We recruited the therapist participants at the clinics through introductory informational meetings. Here, we included only therapists who self-identified as Ashkenazi (n = 20) or as “mixed” ethnic origin (n = 4; a group which traditionally tends to identify more with Ashkenazi group) who saw either a Mizrahi or Ashkenazi self-identified client. Each therapist in the sample saw between one to five clients who were either Mizrahi or Ashkenazi. The majority of therapists were female (70%), ages ranged from 28 to 64 (M = 44, SD = 9.88). Fifty percent were psychologists, 10% psychiatrists, and 40% social workers, with the majority of therapists (75%) having more than 5 years of clinical practice (M = 11.79, SD = 11.31).

Client participants were recruited through direct person-to-person solicitation which took place at the clinic as they waited for their intake visit. Client inclusion criteria were adults (18 years and older), who did not require interpreter services. Exclusion criteria included people whom the therapists identified as psychotic or actively suicidal. Of the clients who were invited to participate in the study, 122 agreed to participate (31 clients declined: 21 were not able to stay for additional time following their intake; 3 did not feel well enough to participate; and 7 did not want to have the session recorded). Here, we included only clients who went through a full length intake session (i.e., we excluded 41 sessions which lasted less than 45 minutes) and who self-identified as either Mizrahi (n = 33) or Ashkenazi (n = 25).

The majority of clients were female (76%) and ages ranged from 21 to 78 (M = 42.30, SD = 16.02). About half of the sample (52%) had 12 years of education or less and 63% were unemployed. There were significant differences between Mizrahim and Ashkenazim in years of education, M = 11.98, SD = 2.67 for Mizrahim and M = 15.10, SD = 2.98 for Ashkenazim, F(1, 48) = 14.06, p < .001, and marginally significant differences in unemployment status, 57% for Mizrahim versus 33% for Ashkenazim, χ²(1) = 2.92, p = .08. There were no significant differences in gender, age, average income (measured on a 1–5 scale from much below average to much above average; M = 1.97, SD = 1.09 for Mizrahim and M = 2.44, SD = 1.34 for Ashkenazim), and importantly, in emotional distress at entry into treatment as measured by self-reported General Health Questionnaire (GHQ; M = 31.3, SD = 7.84 for Mizrahim and M = 31.54, SD = 7.25 for Ashkenazim).

Procedure

After complete description of the study to the participants, written informed consent was obtained. The presenting problems clients described were diverse and included familial and other interpersonal problems as well as symptoms of mental health disorders (e.g., depression, anxiety, and eating disorders; Nakash, Nagar, & Levav, 2014). Intake visits ranged between 45 and 99 min (M = 62.76, SD = 13.68).

Participation in the study included the following three parts: (a) clients completed survey measures prior to intake which included demographic information and a measure assessing emotional distress (for full list of measures, see Nakash et al., 2013), (b) audio-recording of the intake session, and (c) immediately following the intake session clients completed structured diagnostic interview (Mini International Neuropsychiatric Interview [MINI]) with an independent interviewer, and a measure assessing the quality of the rapport during the intake session, while therapists completed a form detailing the client’s diagnosis according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) and a measure assessing the quality of the rapport and attitudes toward the client they saw in the intake session. Clients and therapists received an honorarium of US$25 for their participation. The appropriate Institutional Ethics Committees at each participating clinic approved all aspects of the study and data collection complied with all human subject protocols. Subsequently, we provide information only on the measures we included in this study.

Measures

Demographic questionnaire

This included gender, age, years of education, employment status, and income for clients. Therapists’ measure included gender, age, discipline, and years in clinical practice.

The GHQ-12

This 12-item scale screens for common mental disorders and measures emotional distress. Items are rated on a 4-point-Likert-type scale and scores range from 12 to 48, where higher scores indicate increased emotional distress. The internal consistency of the scale for both Mizrahim and Ashkenazim was good (αs = .87 and .86, respectively; Goldberg, 1978).

Working Alliance Inventory (WAI) Short, client and therapist versions

This 12-item self-report scale measures the client’s and therapist’s view of the quality of the working alliance (Horvath & Greenberg, 1989). It has been widely used in studies on therapeutic outcomes to assess therapeutic alliance in treatment and during a single therapy session (Nakash et al., 2015). The measure has corresponding versions for clients and therapists and both versions have shown good reliability and validity (Horvath & Symonds, 1991). The measure includes the following three original scales: (a) task (e.g., “the therapist/client and I agree about the steps to be taken to improve my/their situation”), (b) goals (e.g., “We are working towards mutually agreed upon goals”), and (c) bond (e.g., “I believe the therapist/client likes me”) that are considered elements of good alliance. Each item was rated on 7-point scale ranging from 1 (not at all) to 7 (very much). Final score included mean score for all items with higher score reflecting better therapeutic alliance. The internal consistency was good for clients and therapists (α = .84, α = .91, respectively; Tracey & Kokotovic, 1989).

MINI

To assess the accuracy of the therapist’s diagnoses, their diagnostic decision was compared with assessment of the MINI conducted by an independent rater. The MINI is a structured diagnostic interview for primary and comorbid diagnoses based on DSM-IV (Sheehan et al., 1998). The MINI involves asking clients direct questions about specific diagnostic criteria while relying primarily on the examinee’s explicit endorsement of each diagnostic criterion. Structured questions follow directly from the DSM-IV and correspond to symptoms for each diagnosis. These interviews are considered the gold standard for all psychiatric diagnoses, as they increase reliability of diagnostic assessment and minimize clinical judgment that could lead to unreliable diagnoses (Farmer & Chapman, 2002; Rogers, 2003). Further validation was obtained from a study in which the diagnosis of general practitioners using the MINI after short 2- to 3-hour training sessions was compared with a specialized interviewer, with results yielding high concordance rates (Eytan, Durieux-Paillard, Whitaker-Clinch, Loutan, & Bovier, 2007).

Six advanced graduate students in clinical psychology (all self-identified as Ashkenazi) served as independent interviewers. Training interviewers was similar to procedures established in previous studies (Baune et al., 2010) and included three 6-hour training sessions, led by a licensed clinical psychologist prior to commencement of the study. Supervision was provided throughout the data collection period on a weekly basis to assure adherence.

Perceptions of client’s warmth

This were assessed through 4 items, adapted from Van Ryn et al. (2006) asking therapists to indicate the extent to which the client is warm, caring, friendly, and hostile (r). For each item, ratings ranged from 1 (not at all) to 5 (very much). Internal consistency was good (α = .76), and all items were averaged to form an index of warmth perceptions.

Diagnoses

Therapists were asked to detail their decision about the clients’ main problem, by listing the Axis I major psychiatric disorders (e.g., anxiety, mood disorder, eating disorder, and substance abuse) and Axis II personality disorders. These diagnoses were listed according to the DSM-IV for each client immediately following the intake. If therapists thought that the client had more than one disorder (comorbidity), they were asked to list all disorders for each client. For this article, we only considered Axis I disorders as the structured interview (MINI) does not include an assessment of Axis II personality disorders.

Results

Diagnostic Accuracy

To assess diagnostic accuracy for each client, we compared their therapist’s diagnoses and the diagnoses established through independent structured clinical interview (MINI), which serves as a gold standard for diagnostic accuracy of psychiatric disorders. As indicated earlier, a therapist could diagnose their client with several disorders, and the MINI assessment could also include several disorders for the same client. The comparison of MINI and therapist diagnoses, therefore, yielded four groups for each cluster of disorders: clients who were diagnosed by both MINI and therapist (true positive); clients who were diagnosed by MINI but not by therapist (false negative); clients who were not diagnosed by MINI but were diagnosed by therapist (false positive); and clients who were neither diagnosed by MINI nor by therapist (true negative). Table 1 presents the descriptives for diagnostic accuracy for each cluster of disorders separately (mood, anxiety, alcohol and substance abuse, eating, and psychotic disorders).

Table 1.

Diagnostic Accuracy of Major Clusters of Psychiatric Disorders Assessed During the Mental Health Intake (N = 58).

		Mizrahim (n = 33)	Ashkenazim (n = 25)
Mood disorders	True positive	9 (27.3%)	11 (44%)
	False positive	4 (12.1%)	0 (0%)
	True negative	7 (21.2%)	8 (32%)
	False negative	13 (39.4%)	6 (24%)
Anxiety disorders	True positive	7 (21.2%)	4 (16%)
	False positive	4 (12.1%)	4 (16%)
	True negative	7 (21.2%)	8 (32%)
	False negative	15 (45.5%)	9 (36%)
Alcohol and substance abuse disorders	True positive	2 (6.1%)	1 (4%)
	False positive	0 (0%)	0 (0%)
	True negative	29 (87.9%)	20 (80%)
	False negative	2 (6.1%)	4 (16%)
Eating disorders	True positive	1 (3%)	0 (0%)
	False positive	0 (0%)	1 (4%)
	True negative	29 (87.9%)	23 (92%)
	False negative	3 (9.1%)	1 (4%)
Psychosis	True positive	1 (3%)	0 (0%)
	False positive	1 (3%)	0 (0%)
	True negative	28 (84.8%)	18 (72%)
	False negative	3 (9.1%)	7 (28%)

Note. Numbers within diagnostic accuracy categories (true positive, false positive, true negative, and false negative) represent therapists’ unstructured diagnostic assessments in regular practice compared to independent structured clinical interview (MINI), which serves as a gold standard for accuracy of psychiatric diagnosis. There were no significant differences between Mizrahi and Ashkenazi clients in accuracy of diagnostic decisions for all examined disorders.

Given the high variability and comorbidity of diagnostic decisions for each client (approximately 80% of clients had more than one diagnosis), we combined the categories such that 1 = accurate diagnosis (at least one true positive) and 0 = misdiagnosis (indicating all diagnoses were either false negative or false positive). Because all clients received at least one diagnosis by the therapist and by the MINI interviewer, we had no case of true negative (i.e., MINI and therapist agreed about the absence of a disorder). Thus, given that even a single match in the therapist’s and the MINI’s diagnosis was considered a “true” diagnosis, our criteria for misdiagnosis can be considered quite liberal. Misdiagnosis was unrelated to any of the other outcome measures (WAI scales were positively correlated (r = .51, p < .01) and the warmth ratings were positively correlated with therapist’s WAI (r = .63, p < .01) and with client’s WAI (r = .43, p < .01).

To test our key hypothesis, we conducted a χ² test which assessed whether diagnostic accuracy (0 or 1) was associated with the client’s ethnicity (Ashkenazi vs. Mizrahi). The analysis revealed that, as predicted, Ashkenazi therapists had significantly more misdiagnosis in intakes involving Mizrahi clients (52%) relative to intakes including Ashkenazi clients (24%), such that with Mizrahi clients, therapists were twice as likely to misdiagnose mental disorder, χ²(1) = 4.5, p = .03.

To examine whether differences between Ashkenazi and Mizrahi clients in misdiagnosis were due to false positives or false negatives, we conducted a chi-square test. We compared four combinations of possible misdiagnosis for comorbid conditions: only false positive (Mizrahi: n = 4, 12.1%; Ashkenazi: n = 0), only false negative (Mizrahi: n = 13, 39.4%; Ashkenazi: n = 6, 24%), neither false positive nor false negative (Mizrahi: n = 16, 48.5%; Ashkenazi: n = 19, 76%). None of the clients were diagnosed with both false positive and false negative. The analysis revealed that although false negatives were more prevalent in general for all clients, Mizrahi clients were more likely to get both false positive and false negative misdiagnoses, χ²(2) = 5.8, p = .054.

We further analyzed diagnostic efficiency by assessing sensitivity (test to correctly identify those patients with the disorder) and specificity (test to correctly identify those patients without the disorder). We focused on mood disorders that were most prevalent in the current sample. Sensitivity (calculated as number of true positive divided by [number of true positive + false negative]) was higher for Ashkenazi clients as compared to Mizrahim (64.7% and 40.9%, respectively). Specificity (calculated as number of true negative divided by [number of true negative + false positive]) was notably higher for Ashkenazim compared to Mizrahim (100% and 63.6%, respectively).

Quality of Working Alliance (WAI)

An analysis of variance (ANOVA) considering the client’s ethnicity as an independent variable and the therapist’s WAI as an outcome variable revealed that Ashkenazi therapists reported significantly more positive working alliance with Ashkenazi clients (M = 5.06, SD = .95) relative to the quality of working alliance reported with Mizrahi clients (M = 4.52, SD = 1.04), F(1, 55) = 3.90, p = .05, η² = .07. Given the nonsignificant association between WAI and accuracy (r = −.04 controlling for client’s ethnicity), these differences could not account for the effect of client’s ethnicity on diagnostic accuracy. Interestingly, an ANOVA on the client’s WAI revealed no significant differences between Mizrahi (M = 5.19, SD = 1.15) and Ashkenazi clients (M = 5.29, SD = .85), F < 1. Indeed, in intakes with Mizrahim there was a significant difference between the therapist’s WAI and clients’ WAI such that clients reported more positive alliance during the intake than did their therapists, F(1, 31) = 15.39, p < .000; while in intakes with Ashkenazim, no such difference was observed, F(1, 23) = 1.2, p > .28.

Warmth Ratings

An ANOVA considering the client’s ethnicity as an independent variable and the therapists’ assessment of the client’s warmth revealed that Ashkenazi therapists rated Ashkenazi clients more positively (M = 3.72, SD = .67) than Mizrahi clients (M = 3.29, SD = .89), F(1,55) = 3.87, p = .05, η² = .07. As with WAI, given the nonsignificant association between warmth and accuracy (r = .02 controlling for client’s ethnicity), these differences could not account for the effect of client’s ethnicity on diagnostic accuracy.

Discussion

In this research, we focused on a critical behavior on part of therapists: the accuracy of their diagnostic decisions during the mental health intake. To our knowledge, no prior research has considered diagnostic accuracy in the context of divergent social identities during early stages of the mental health treatment. This outcome, which has substantial impact on the planning and thus the success of treatment, is critical for understanding what may underlie ethnic mental health service disparities. Indeed, if members of disadvantaged groups are more frequently misdiagnosed relative to advantaged group members as indicated by our findings, it is no surprise that the quality of the mental health services they receive, and their mental health outcomes, are worse.

Our findings further showed that, while the clients’ accounts of the quality of the working alliance did not differ by ethnicity, socially advantaged therapists reported worse working alliance with clients belonging to disadvantaged (relative to advantaged) groups. They further found disadvantaged clients as less warm. These findings are significant, as considerable research showed that the quality of the therapeutic alliance is the single best predictor of positive clinical outcomes of psychotherapy (Horvath & Symonds, 1991; Zuroff & Blatt, 2006). Recent research further indicated that therapists tend to conduct implicit judgments, primarily based on nonverbal cues and affective communication, in appraising the quality of working alliance with their clients (Nakash & Alegría, 2013).

Several processes may account for the difference in therapist’s attitudes and diagnostic decisions, as a function of the client’s identity. The effects may be explained by cross-cultural difficulties that are present in discordant encounters (cultural dysfluency). If this is indeed the reason, then we would expect similar results in (the relatively rare) situations involving a therapist from a disadvantaged group and a client from an advantaged group. Other processes, more directly pertaining to intergroup bias against minorities, and to asymmetrical power, are less likely to take place when the therapist belongs to a disadvantaged group.

Given that in this study we only included Ashkenazi therapists, we cannot determine whether the bias they reported is a result of cultural discordance of simply automatic favoritism for in-group over out-group members (Fiske, 1998), or a bias against disadvantaged clients, which is more likely to come into play when power asymmetry is greater. Therefore, exploration of outcomes of interactions that are characterized by reversed power dynamics (socially advantaged clients and disadvantage therapists), though much less common, can be of great value for informing this question. Such encounters may result in better outcomes than concordant encounters because power differences are more balanced. A contrasting possibility is that processes of constraint and behavioral inhibition will debilitate the therapist’s clinical decision making (Keltner, Gruenfeld, & Anderson, 2003).

Our findings suggest that although mental health disparities are a multifaceted phenomenon, processes related to the social identities of clients and the extent to which these identities match the ones of the treating therapist can affect the quality of the clinical interactions. Research on this “matching” hypothesis in the contexts of mental health care has focused on clients’ related outcomes (e.g., service utilization; Cabral & Smith, 2011). Here, we extended this line of research in critical ways by showing the effect of the racial/ethnic match on behaviors and attitudes on the part of therapists in their everyday practice.

This study has several limitations. First, the data were collected before the publication of the DSM-5. However, since the categorical basis for the diagnostic system has not changed in this recent version, we believe our findings can be extrapolated to the current version. Second, due to limited statistical power, we were unable to examine possible differences that may exist in the assessment process of therapists from different disciplines. Third, the study was conducted among a convenience sample which may be subject to selection bias. Fourth, due to clinic procedures and ethical committee requirements, we were not able to collect reliability data for the diagnoses according to the MINI as well as to counterbalance the administration of the unstructured assessment by therapist and the structured diagnostic interview by the independent rater.

Our findings have important implications to clinical practice and training as they show that processes of categorization and possible associated biases, which are prevalent in a variety of contexts, also take place in the mental health field (see Dovidio et al., 2008; Van Ryn, 2002). Our findings highlight the importance of using structured diagnostic assessments to reduce bias and increase reliability of diagnostic process. However, since clinical determinations in community mental health clinics must be made in severely resource-constrained environments, clinical use of any diagnostic system needs to deal with the issue of missing information. One possible approach to increase diagnostic efficiency in this context is to examine the use of best probes for correct diagnosis for specific disorders (Nakash, Nagar, & Kanat Maymon, in press) or use structured measures to complement the information collected during the mental health intake (Nakash et al., 2009).

In addition, extensive social psychological research has identified both individual-level factors and contextual influences, which impact processes of intergroup biases. This body of work can be effectively used to develop relevant interventions. For example, diversity training can be implemented to educate therapists about their potential biases and consequences. Raising awareness to the nature of stereotypical thinking was found to reduce cognitive biases particularly unconscious ones (e.g., Rudman, Ashmore, & Gary, 2001). This study sets the stage for implementing such intervention, which can assist in raising awareness to therapists’ processes that contribute to mental health disparities.

Footnotes

Acknowledgments

The authors gratefully acknowledge Prof. Itzhak Levav, Dr. Eli Danilovich, Dr. Daphne Bentov-Gofrit, Dr. Ido Lurie, Dr. Henry Szor, Dr. Evelyn Stiener, and Dr. Shiri Sadeh-Sharvit, for their support as well as all participating clients and therapists.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Israeli National Institute for Health Policy and Health Services Research (2006/6/א to Nakash).

Note

References

Aneshensel

C. S.

(2009). Toward explaining mental health disparities. Journal of Health and Social Behavior, 50, 377–394.

Baune

B. T.

Miller

McAfoose

Johnson

Quirk

Mitchell

(2010). The role of cognitive impairment in general functioning in major depression. Psychiatry Research, 176, 183–189.

Braveman

Egerter

Williams

D. R.

(2011). The social determinants of health: Coming of age. Annual Review of Public Health, 32, 381–398.

Breslau

Kendler

K. S.

Gaxiola-Aguilar

Kessler

R. C.

(2005). Lifetime risk and persistence of psychiatric disorders across ethnic groups in the United States. Psychological Medicine, 35, 317–327.

Burgess

D. J.

S. S.

Van Ryn

(2004). Why do providers contribute to disparities and what can be done about it? Journal of General Internal Medicine, 19, 1154–1159.

Burgess

D. J.

Van Ryn

Dovidio

Saha

(2007). Reducing racial bias among health care providers: Lessons from social-cognitive psychology. Journal of General Internal Medicine, 22, 882–887.

Cabral

Smith

T. B.

(2011). Racial/ethnic matching of clients and therapists in mental health services: A meta-analytic review of preferences, perceptions, and outcomes. Journal of Counseling Psychology, 58, 537–554.

Cook

McGuire

Miranda

(2007). Measuring trends in mental health care disparities, 2000–2004. Psychiatric Services, 58, 1533–1540.

Dovidio

J. F.

Gaertner

S. L.

Kawakami

Hodson

(2002). Why can’t we just get along? Interpersonal biases and interracial distrust. Cultural Diversity and Ethnic Minority Psychology, 8, 88–102.

10.

Dovidio

J. F.

Penner

L. A.

Albrecht

T. L.

Norton

W. E.

Gaertner

S. L.

Shelton

J. N.

(2008). Disparities and distrust: The implications of psychological processes for understanding racial disparities in health and health care. Social Science and Medicine, 67, 478–486.

11.

Eytan

Durieux-Paillard

Whitaker-Clinch

Loutan

Bovier

P. A.

(2007). Transcultural validity of a structured diagnostic interview to screen for major depression and posttraumatic stress disorder among refugees. Journal of Nervous and Mental Disease, 195, 723–728.

12.

Farmer

R. F.

Chapman

A. L.

(2002). Evaluation of DSM-IV personality disorder criteria as assessed by the structured clinical interview for DSM-IV personality disorders. Comprehensive Psychiatry, 43, 285–300.

13.

Fiske

(1998). Stereotyping, prejudice, and discrimination. In Gilbert

Fiske

(Eds.), The handbook of social psychology (4th ed., pp. 357–411). New York, NY: McGraw-Hill.

14.

Galinsky

A. D.

Magee

J. C.

Inesi

M. E.

Gruenfeld

D. H.

(2006). Power and perspectives not taken. Psychological Science, 17, 1068–1074.

15.

Goldberg

D. P.

(1978). Manual of the general health questionnaire. Windsor, England: NFER.

16.

Gordon

H. S.

Street

R. L.

Sharf

B. F.

Kelly

P. A.

Souchek

(2006). Racial differences in trust and lung cancer patients’ perceptions of physician communication. Journal of Clinical Oncology, 24, 904–909.

17.

Green

A. R.

Carney

D. R.

Pallin

D. J.

Ngo

L. H.

Raymond

K. L.

Iezzoni

L. I.

Banaji

M. R.

(2007). Implicit bias among physicians and its prediction of thrombolysis decisions for black and white patients. Journal of General Internal Medicine, 22, 1231–1238.

18.

Greenwald

A. G.

Banaji

M. R.

Rudman

L. A.

Farnham

S. D.

Nosek

B. A.

Mellott

D. S.

(2002). A unified theory of implicit attitudes, stereotypes, self-esteem, and self-concept. Psychological Review, 109, 3–25.

19.

Horvath

A. O.

Greenberg

L. S.

(1989). Development and validation of the working alliance inventory. Journal of Counseling Psychology, 36, 223–233.

20.

Horvath

A. O.

Symonds

B. D.

(1991). Relation between working alliance and outcome in psychotherapy: A meta-analysis. Journal of Counseling Psychology, 38, 139–149.

21.

Institute of Medicine. (2003). Unequal treatment: Confronting racial and ethnic disparities in health care. Retrieved June 28, 2011, from http://www.iom.edu

22.

Israel Central Bureau of Statistics. (2012). Demographics. Retrieved April 1, 2012, from www.cbs.gov.il

23.

Johnson

R. L.

Roter

Powe

N. R.

Cooper

L. A.

(2004). Patient race/ethnicity and quality of patient-physician communication during medical visits. American Journal of Public Health, 94, 2084–2090.

24.

Keltner

Gruenfeld

D. H.

Anderson

(2003). Power, approach, and inhibition. Psychological Review, 110, 265–284.

25.

Nakash

Alegría

(2013). Examination of the role of implicit clinical judgments during the mental health intake. Qualitative Health Research, 23, 645–654.

26.

Nakash

Dargouth

Oddo

Gao

Alegría

(2009). Patient initiation of information: Exploring its role during the mental health intake visit. Patient Education and Counseling, 75, 220–226.

27.

Nakash

Gerber

Goldbourt

Benyamini

Drory

(2013). Ethnicity and long-term prognosis after myocardial infarction: A population-based cohort study. Medical Care, 51, 137–143.

28.

Nakash

Levav

Gal

(2013). Ethnic-based intra- and inter-generational disparities in mental health: Results from the Israel-world mental health survey. The International Journal of Social Psychiatry, 59, 508–515.

29.

Nakash

Nagar

Danilovich

Bentov-Gofrit

Lurie

Steiner

… Levav

(2013). Ethnic disparities in mental health treatment gap in a community-based survey and in access to care in psychiatric clinics. International Journal of Social Psychiatry, 60, 575–583.

30.

Nakash

Nagar

Kanat-Maymon

(in press). The clinical utility of DSM categorical diagnostic system during the mental health intake. Journal of Clinical Psychiatry.

31.

Nakash

Nagar

Levav

(2014). Presenting problems and treatment expectations among service users accessing psychiatric outpatient care: Are there gender differences? Israeli Journal of Psychiatry, 51, 212–218.

32.

Nakash

Nagar

Levav

(2015). Predictors of mental health care stigma and its association with the therapeutic alliance during the initial intake session. Psychotherapy Research, 25, 214–221.

33.

Nakash

Rosen

Alegría

(2009). The culturally sensitive evaluation. In Ruiz

Primm

(Eds.), Disparities in psychiatric care: Clinical and cross-cultural perspectives (pp. 225–235). Bethesda, MD: Wolters Kluwer Lippincott Williams and Wilk.

34.

Nakash

Saguy

Levav

(2012). The effect of social identities of service-users and providers on mental health disparities: A review of theory and facts. Israel Journal of Psychiatry, 49, 202–212.

35.

Rogers

(2003). Standardizing DSM-IV diagnoses: The clinical applications of structured interviews. Journal of Personality Assessment, 81, 220–225.

36.

Rosen

Miller

Nakash

Halpern

Alegria

(2012). Interpersonal complementarity in the mental health intake: A mixed-methods study. Journal of Counseling Psychology, 59, 185–196.

37.

Rudman

L. A.

Ashmore

R. D.

Gary

M. L.

(2001). “Unlearning” automatic biases: The malleability of implicit prejudice and stereotypes. Journal of Personality and Social Psychology, 81, 856–868.

38.

Saguy

Dovidio

J. F.

Pratto

(2008). Beyond contact: Intergroup contact in the context of power relations. Personality and Social Psychology Bulletin, 34, 432–445.

39.

Saguy

Tropp

L. R.

Hawi

D. R.

(2013). The role of group power in intergroup contact. In Hodson

Hewstone

(Eds.), Advances in intergroup contact (pp. 113–132). New York, NY: Psychology Press.

40.

Sentell

T. L.

Halpin

H. A.

(2006). Importance of adult literacy in understanding health disparities. Journal of General Internal Medicine, 21, 862–866.

41.

Sheehan

D. V.

Lecrubier

Sheehan

K. H.

Amorim

Janavs

Weiller

Dunbar

G. C.

(1998). The Mini-International Neuropsychiatric Interview (MINI): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry, 59, 22–33.

42.

Siminoff

L. A.

Graham

G. C.

Gordon

N. H.

(2006). Cancer communication patterns and the influence of patient characteristics: Disparities in information-giving and affective behaviors. Patient Education and Counseling, 62, 355–360.

43.

Snowden

L. R.

Masland

M. C.

Libby

A. M.

Wallace

Fawley

(2008). Racial/ethnic minority children’s use of psychiatric emergency care in California’s public mental health system. American Journal of Public Health, 98, 118–124.

44.

Tracey

T. J.

Kokotovic

A. M.

(1989). Factor structure of the working alliance inventory. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 1, 207–210.

45.

Van Ryn

(2002). Research on the provider contribution to race/ethnicity disparities in medical care. Medical Care, 40, 140–151.

46.

Van Ryn

Burgess

Malat

Griffin

(2006). Physicians’ perceptions of patients’ social and behavioral characteristics and race disparities in treatment recommendations for men with coronary artery disease. American Journal of Public Health, 96, 351–357.

47.

Van Ryn

Williams

(2003). Commentary on racial disparities in health care. Medical Care Research Review, 60, 496–508.

48.

Weich

Nazroo

Sproston

McManus

Blanchard

Erens

Stansfeld

(2004). Common mental disorders and ethnicity in England: The EMPIRIC study. Psychological Medicine, 34, 1543–1551.

49.

Williams

D. R.

Mohammed

S. A.

(2009). Discrimination and racial disparities in health: Evidence and needed research. Journal of Behavioral Medicine, 32, 20–47.

50.

Zuroff

D. C.

Blatt

S. J.

(2006). The therapeutic relationship in the brief treatment of depression: Contributions to clinical improvement and enhanced adaptive capacities. Journal of Consulting and Clinical Psychology, 74, 130–140.