Keywords

Introduction

Even though the USA keeps growing more diverse every day, there is still a deficit in completely understanding how ethnicity and culture influence psychological assessment and more specifically, the interview process as well as diagnostic and treatment decisions (Aklin and Turner 2006). Symptom patterns associated with mental disorders do not manifest identically across cultures or people of different ethnicities. For example, research suggests that African American (AA) and Hispanic patients tend to exhibit more somatic and physical complaints when diagnosed with depression compared to Caucasians (Brown et al. 1996; Myers et al. 2002). In addition, different kinds of hallucinations are manifested in people of different cultures (Bauer et al. 2011). Further, it is plausible that the phenomenology of disorders differs across racial/ethnic groups, resulting in some cultures viewing behaviors as pathological that others regard as normative (Lewis-Fernández et al. 2010). For example, some cultures that are far removed from the Western culture consider hallucinations and epileptic seizures as mystical gifts or higher powers of a religious nature.

The Diagnostic and Statistical Manual of Mental Disorders Fourth Edition Revised (DSM-IV-R) acknowledged cultural differences in diagnosis by including an outline for preparation of cultural formulations, increasing the description of the client’s individuality by multidimensional data collection and amalgamation of a clinical presentation format, and including a glossary of culturally bound syndromes (American Psychiatric Association 1994; Dana 2008). For example, among AAs “spell” refers to a trance state in which individuals can communicate with deceased relatives and which is associated with brief periods of personality change (American Psychiatric Association 1994). The Diagnostic and Statistical Manual of Mental Disorders Fifth Edition (DSM-5) also included a glossary of cultural concepts of distress. However, the glossary does not include cultural concepts of distress related to AAs. In addition, it updated the cultural formulation outline and added a Cultural Formulation Interview (CFI; American Psychiatric Association 2013).

These updates suggest that cultural issues now have more relevance for clinical diagnosis; moreover there is substantial research reporting the importance and influence of cultural and ethnic factors in psychological assessment, particularly in the clinical interview (e.g., Adebimpe 1981; Neighbors 1989; Paniagua 2001; Trierweiler et al. 2000; Aklin and Turner 2006; Grieger 2008). The principal goal of the clinical interview is to gather information related to the problem at hand. It represents an interpersonal interaction that aims to elucidate an individual’s feelings and attitudes about their symptoms and problem behaviors, while at the same time allowing the clinician to observe verbal and nonverbal behavior. It is the best tool to gather information that that leads to a diagnosis and establishes the stage needed for a fruitful therapeutic relationship (rapport) (Aklin and Turner 2006; Turner et al. 2003). When working with AAs it is also the best tool to gather and incorporate relevant cultural and ethnic factors that might influence the assessment process, case conceptualization, and diagnosis.

Types of Interviews

Clinical interviews remain one of the most commonly used methods to assist in the diagnoses of an individual. There are three main types of clinical/diagnostic interviews: open, structured, and semistructured.

Open Interviews

Open interviews permit clinicians to choose what questions to ask, the depth to which topics are explored, the length, or any other pertinent variable. They are advantageous in that the interview can be tailored to the specific concerns of the patient, they can be less time consuming, are generally easy to learn and administer, can be administered anywhere, the conversational nature aids rapport, and they allow for flexibility of depth of symptom exploration. Open interviews risk intrusion of theoretical or personal biases, omission of important questions, variation of wording and question tone, variability of the order in which the questions are asked, variation of the depth and style of recording, and often lack ratings to measure presence and severity of symptoms (Rogers 2001). These variables can affect how patients respond to diagnostic questions. In a seminal paper Ward et al. (1962) reached the conclusion that the majority of diagnostic variability is a result of the evaluations, not the patients. In his study, 62.5 % of diagnostic variability resulted from criterion variance, 32.5 % from information variance, and a scant 5.0 % from patient variance. Criterion variance refers to differences among clinicians in implementing standards for what is clinically important and when the diagnostic criteria are met. Information variance refers to differences among clinicians in the questions that are used, the observations made, and how the gathered information is organized. Patient variance refers to differences within the same patient that lead to significant discrepancies in clinical presentation and subsequent diagnosis. These sources of variance are likely to be present in open interviews. Adding variables such as culture, ethnicity or language barriers to the equation greatly risks further increased variability (Aklin and Turner 2006). In a study by Strawoski et al. (1997), when a patient was of minority status, information variance was the cause of diagnostic disagreements in 58 % of the cases. Criterion variance was the cause of disagreement in 42 % of the cases and was not associated with race.

Structured Interviews

Structured interviews allow for a systematic evaluation by standardizing the specific language of clinical questions, the order of these questions, and the quantification of responses. All questions must be asked verbatim as it is instructed. Standard questions and optional probes are usually utilized (Rogers 2001). Structured interviews may be advantageous because they reduce patient variance and capitalize on systematic evaluation to reduce misdiagnosis. Their comprehensiveness is likely to abate missed diagnosis by removing a priori hypotheses from the interview process, and a high level of standardization is ensured. In addition, they allow for systematic comparisons to be made (across collateral sources, time, clinicians, within the same patient, within settings, and within diagnosis; Rogers 2001). These qualities make structured interviews the instrument of choice for research purposes. Such instruments, however, are lengthy, require advanced training, and their rigid guidelines cannot account for all possible eventualities. Structured interviews enhance information variance, sometimes at the cost of criterion variance (Rogers 2001).

Semistructured Interviews

Semistructured interviews include guidelines that allow clinicians considerable flexibility while still assuring that a certain set of questions will be reviewed across clinicians and patients (Aklin and Turner 2006). They utilize standard questions, optional probes, and unstructured questions. Semistructured interviews tend to increase criterion variance, sometimes at the cost of information variance (Rogers 2001). Aklin to structured interviews, semistructured interviews have the advantage of standardization and reduced variability. In addition, examiners may ask their own questions when diagnostic issues remain unresolved. To improve patient understanding, they allow for a more spontaneous and conversational interview, and they can be modified according to a client’s specific needs (Aklin and Turner 2006; Rogers 2001). Disadvantages include long administration time, extensive training requirements, and reliance on protocols. Research shows that structured and semistructured interviews have better validity and reliability when assessing individuals that belong to an ethnic minority when compared to open interviews (Widiger 1997). Structured and semistructured interviews allow for systematic and comprehensive coverage of symptoms by ensuring that specific criteria are met for diagnoses, and decrease variance resulting from differences in ethnicity and culture (Turner et al. 2003).

Cultural Considerations when Interviewing AAs

The following are important cultural factors specific to AAs that are important to consider in the diagnostic interview process, regardless of the type of interview utilized.

Socioeconomic Status and Population Characteristics

Currently AAs represent 13 % (38.9 million) of the total US population. Their poverty rate (13 %) is close to being two times higher than that of all households (25.5 %) and their unemployment rate (13.6 %) is almost twice as high (7.4 %) (U.S. Census Bureau 2010). Nearly 13 % of AA adults do not have a high school diploma (Fry 2010). In addition, the median wealth of white households is 20 times that of AA households and nearly half of all prisoners in the USA are AAs. Prison inmates and people experiencing poverty are at high risk of developing a mental illness (NAMI 2004). These negative statistics are influenced by the disadvantaged position of AAs, racism , and poverty. However, these numbers do not represent the whole of the AA population in the USA. Most of the available literature is based on the economically disadvantaged portion of the AA population; therefore the diversity of this population is somewhat underreported (Holmes and Morin 2006; Sue 2013). For example, over 38 % of AA households are middle class vs. 44 % of all households (U.S. Census Bureau 2010). All of these facts and statistics should be taken into account when conducting a diagnostic interview with AAs; nevertheless they should not influence or bias the clinician’s judgment. Socioeconomic status (SES) is a particularly important variable to consider when assessing minorities. For example, many clinicians tend to perceive individuals of a lower SES as having more problems than individuals of higher SES (Bentacourt and Lopez 1993; Lindsey and Paul 1989; Robins and Regier 1991; Snowden and Cheung 1990). Using unstructured interviews increases the probability that these kinds of beliefs will negatively influence the clinician’s judgment and symptom interpretation (Garb 1997). Using structured and semistructured interviews could reduce this probability (Aklin and Turner 2003).

Racial Identity

Research suggests that members of cultural and ethnic minority groups go through a sequential process of racial identity (Sue 2013). For many AAs this process consists of an evolution from a non-Afrocentric identity to one that is Afrocentric. Cross (1995) presented a model of AA racial identity that involves the following stages: pre-encounter, encounter, immersion–emersion, and internalization. These stages are associated with differences in views related to the self and relationships with others, beginning with the acceptance of white culture and rejection or devaluation of black culture and ending with an appreciation and acceptance of both black culture and aspects of white culture (Sue 2013). AAs that are at the pre-encounter level are less prone to report racial discrimination, while those in the immersion stage are usually younger and least satisfied with societal conditions (Hyers 2001). In addition, AAs at the first stage tend to prefer a white counselor, whereas those in later stages prefer an AA counselor (Parham and Helms 1981). Discussing racial identity during the interview process and getting a picture of which stage the patient is in could help build the therapeutic relationship and reduce the possibility of misinterpreting answers to certain questions.

Racism, Discrimination, and Stereotypes

Clinicians, like every other person, could potentially be racist towards a minority group or hold stereotypes or prejudices regarding a particular minority group. Diagnostic errors could be the result of direct racism or indirect racism. Indirect racism refers to preconceived notions about a certain group (Aklin and Turner 2006). Further, clinicians should actively try to avoid engaging in microagressions. “Microaggressions are brief and commonplace daily verbal or behavioral indignities, whether intentional or unintentional, that communicate hostile, derogatory, or negative racial slights and insults that potentially have a harmful or unpleasant psychological impact on the target person or group” (Sue et al. 2007). Even counselors with a significant background in cross-cultural practices may engage in micro-aggressions against AA clients (Gushue 2004), and these micro-aggressions significantly impact the therapeutic working alliance (Constantine 2007).

Regarding discrimination, Approximately 50 % of whites believe blacks have equal societal opportunities and 81 % of blacks believe more change is necessary (Pew Research Center 2010). Moreover, many AAs believe that racial profiling occurs frequently (Carlson 2004) and 43 % report there is a great deal of anti-black discrimination in the USA, compared to 13 % of whites. This lack of trust and feelings of discrimination towards the mainstream culture can affect perceptions of social and health systems among AAs (Miller et al. 2001). Further, generalizations about a certain ethnic or cultural group and perceptions influenced by stereotypes can negatively affect the diagnostic process (Whaley 1998). Common stereotypes related to AAs include, but are not limited to: they are unmotivated for treatment, they are violent, hostile, and they are inferior or possess less intelligence (Devine et al. 1991; Monteith et al. 1996). Classic social psychology research regarding stereotypes, prejudice, and discrimination becomes relevant regarding the diagnostic interview of a minority group member. According to Nisbet and Wilson (1977) people are often unaware of the range of factors that can significantly influence their judgments. Intergroup bias seems to be an important factor in this kind of subtle influence. Stereotypes and prejudice are activated so rapidly and spontaneously that they can color initial reactions and potentially bias the processing of subsequently encountered evidence. This could have a negative impact on the outcome of a diagnostic interview, regardless of the type, if the clinician is unaware of the influence of his own stereotypes on his judgments. In addition to influencing people’s judgments, stereotypes can also influence people’s behavior. When stereotypes become activated they tend to elicit a corresponding behavioral response (automatic behavior; Dijksterhuis and Bargh 2001). For example, if a clinician consistently engages in less eye contact and hand movements with AA patients compared to whites, this could influence the AA patients responses to diagnostic questions and their level of engagement without the clinician realizing it. Another manifestation of stereotypes and prejudice at the behavioral level is the self-fulfilling prophecy (Darley and Fazio 1980). In an interracial interaction, this occurs when the biased expectations (driven by stereotypes) people have regarding the other person lead them to behave in a way that will provoke the expected kind of behavior. In a clinical interview a self-fulfilling prophecy could lead the interviewer to a certain diagnoses despite of evidence suggesting that it is false.

Minority group members also hold stereotypes and prejudices and this can include beliefs about their own group (internalized racism). Stereotype threat refers to apprehension experienced by members of a group that their behavior might confirm a stereotype. In classic study Steele and Aronson (1995) showed that when AAs were told that they were going to take an intelligence test they performed worse compared to whites. In contrast, when they were not told about the nature of the test no differences in performance were found. Mentioning that the test was diagnostic of intelligence triggered the stereotype that AAs are academically inferior to whites, which led to poorer performance. Clinicians have to be careful not to elicit this kind of reaction in their AA patients with the questions they ask. When assessing patients, vague and complicated symptoms and complaints, in addition to the clinician’s poor cultural competence enhances stereotypes in clinical settings. Because of this, mental health professionals tend to inappropriately link stereotypical beliefs and thoughts to AAs with mental illness characteristics, behaviors, and factors (Neighbors et al. 2003). Using structured and semistructured interviews diminishes the clinicians’ opportunities to include generalizations about a particular person or group and the influence of stereotypes when making diagnostic decisions (Aklin and Turner 2003). According to Devine (1989) automatic stereotypical reactions are likely to prevail unless they are suppressed by controlled process. Mental health professionals have the responsibility to evaluate themselves and realize what kind of stereotypes, prejudices, and racist beliefs they hold before attempting to serve a minority group such as AAs.

Language

It has been argued that the diagnostic bias inherent in unstructured interviews are to a large extent mediated by a clinician’s freedom to fall prey to stereotypes and prejudice. These biases are to some extent reined in by a more regimented style in structured and semistructured interviews and greater cognitive focus may be dedicated to objective evidence of diagnostic relevance. Clinicians are trained to pay close attention to language. This makes intrinsic sense, as linguistic fluidity, direction of speech, content of speech, poverty of speech, fluidity, and so forth are all significant cues to making diagnostic distinctions, many of which have great impact on the lives of a client. Given that language and culture are heavily intertwined, the importance of understanding the linguistic aspects of the client’s culture, and indeed unique subculture, cannot be overemphasized. Language is an important cultural factor that influences the outcome of a diagnostic interview. Language barriers and dialects used could affect how a patient answers diagnostic questions. For example, if a clinician is not familiar with common slang terms used by AAs, something that the patient says could be misinterpreted. Language capabilities, use, and preference of the patient are also important variables to consider. Research suggests that these variables largely depend on SES (Al-issa 1995). Structured and semistructured interviews seem to reduce communication errors because they provide the clinician with prompts that have to always be read the same way and they are commonly dichotomously coded, leaving the clinician less room to misinterpret verbal statements (Miller et al. 2001). Nevertheless, before using a structured or semistructured interview it should be considered whether the language that the instrument was developed with is appropriate for the particular patient being assessed.

Diagnostic Bias

Cultural bias in diagnosing psychiatric disorders, when assessing minorities, have been often reported in the literature (Adebimpe 1981; Bell and Mehta 1980; Neighbors et al. 1999; 2003; Strakowski et al. 2003). A common criticism is that current diagnostic systems (e.g., DSM-5) were developed primarily for individuals with a Western cultural background; therefore they fail to adequately include cultural differences that could impact diagnoses. Valid and reliable diagnoses are important for several reasons, including, but not limited to establishing prevalence rates, prescribing appropriate treatment, and identifying individuals at risk for developing psychiatric disorders. In addition, there is considerable stigma related to a psychiatric diagnosis, this may cause ethnic minority groups to be reluctant to access mental health services when more severe diagnoses are made (McGuire and Miranda 2008).

Regarding diagnostic bias towards AA patients there are several studies in the literature that report that AAs are more likely to be diagnosed with schizophrenia (SZ) compared to whites (Barnes 2008; Bresnahan et al. 2007; Minsky et al. 2003; Trierweiler et al. 2000; Strakowski et al. 1996), with some studies suggesting AAs are up to three times more likely to be diagnosed with SZ compared to whites (Bresnahan et al. 2007; Eack et al. 2012). In a classic review, Adebimpe (1981) suggested the high rates of SZ diagnoses in AA were due to the following variables: cultural distance between patient and provider (e.g., differences in language, values, and expressions of distress), stereotypes of AA psychopathology (e.g., hostility, reluctance to get treatment), false-positive symptoms (e.g., flat affect, paranoia), and biased diagnostic instruments (not culturally sensitive). DeCoux Hampton (2007) expanded on these factors and suggested that several client-level, care process-level, and system-level variables likely contribute to diagnostic bias, with the manner in which care is accessed (system level) being particularly important. AAs tend to use emergency and acute care services more than other racial and ethnic groups; therefore greater symptom severity and acute psychotic episodes may increase the possibility of a diagnosis of SZ (DeCoux Hampton 2007).

Strakoswki (1996) examined the relative importance placed on different symptom clusters when diagnosing SZ. AAs in this study had higher attributions of auditory hallucinations . Authors suggested that clinicians tend to diagnose SZ in AAs based in criterion A symptoms. In line with these results, Strakoswki (2003) examined diagnostic patterns in interviewers who were blinded and unblinded to race. Participants were 195 inpatients with psychotic symptoms. The Structured Clinical Interview for DSM-IV-TR Axis I Disorders (SCID) III was utilized for diagnostic purposes. Results showed unblinded interviewers endorsed criterion A or auditory hallucinations in AAs, while blinded clinicians found no ethnic differences . The symptoms identified by blinded and unblinded interviewers were similar, but unblinded clinicians reported higher rates of SZ among AAs (47 % compared to 17 % in whites). More recently, Eack et al. 2012 conducted a study that investigated whether the clinicians’ perception of the client’s honesty in reporting symptoms influenced diagnostic decisions. In this study the DSM-III-R Criteria Checklist (Janca and Helzer 1990) was used to determine psychiatric diagnosis. The DSM-III-R Criteria Checklist is a structured interview and checklist that obligates the clinicians to go through all Axis I diagnostic criteria before making a final diagnosis. The Brief Psychiatric Rating Scale (BPRS; Overall and Gorham 1962) was used to determine psychiatric symptomatology. In line with previous research AAs were diagnosed with SZ more than twice as frequently compared to Caucasians (45 vs. 19 %), in addition clinicians perceived AA clients to be less honest than whites. Results indicated that diagnostic biases were substantially reduced after controlling for perceived honesty. In line with these results, Trierweiler et al. (2000) examined clinicians’ attributions associated with diagnosing SZ. They used an open interview to determine diagnosis, and results showed that hallucinations and paranoid suspicious attitudes were more often attributed to AA patients, whereas elevated mood and the combination of negative symptoms and dysphoric mood were more often attributed to non-AA patients. They concluded clinicians tend to use different criteria to diagnose SZ depending on the client’s race (Trierweiler et al. 2006). Consistent with these results, Neighbors (2003) reported clinicians link symptom observations to diagnostic categories differently depending on the client’s race. In this study the sample was 665 psychiatric inpatient AAs and whites. A shortened version of the DSM-III-R Criteria Checklist was used to determine psychiatric diagnoses. Results showed AAs were somewhat more likely to be diagnosed with SZ while whites were much more likely to be diagnosed with bipolar disorder. In addition, loose associations, inappropriate affect, auditory hallucinations, and vague speech were more predictive of a SZ diagnosis in AAs than Caucasians. Even though, the frequency of these symptoms was the same in both groups, the importance the clinicians attributed to them was different depending on race. This influenced whether a diagnosis of SZ or bipolar disorder was given (Neighbors 2003). In a similar study, Neighbors (1999) administered open and semistructured interviews to a sample of AAs and Caucasians psychiatric inpatients (291 patients for phase 1/hospital and 665 patients for phase 2/research). Findings indicated that when patients were evaluated in a hospital setting using clinician-structured interviews (open) AAs were more likely to be diagnosed with SZ and less likely to be diagnosed with a mood disorder compared to whites. When the DSM-III-R Criteria Checklist was used to determine diagnosis the disparity in diagnosis across groups was still present, however it was largely reduced. These results suggest that the use of a semistructured diagnostic instrument can help reduce clinicians’ diagnostic bias due to cultural differences . Notably, in studies that used a semistructured interview to determine diagnosis (SCID III) to address clinicians’ subjectivity, AAs were still more likely to be diagnosed with SZ than whites, even though the disparity was sometimes reduced (Strakowski et al. 2003; Trierweiler et al. 2000). In contrast with these results, Jeste et al. (1996) examined racial differences in the prevalence of SZ in a community sample of AAs and whites and found no differences based on the Brief Psychiatric Rating Scale (BPRS). Providing support for these results, in another study with 226 AAs, Latinos, and Whites with SZ, also using the BPRS for diagnoses, no racial differences were found in symptomatology (Bae and Breke 2002).

The influence of the clinician race in diagnosing AAs with SZ has also been studied. Matthews et al. (2002) conducted a study with a large sample of inpatients with psychotic and mood disorders that compared the diagnoses of culturally matched and unmatched patients and clinicians. AAs were diagnosed with SZ and schizoaffective disorder more often compared to whites, regardless of the clinicians’ race. In another study with 234 inpatients, AA providers were significantly less likely to diagnose mood disorders in AA patients compared to non-AA providers (Trierweiler et al. 2005). In a similar study with 292 inpatients, AA clinicians were significantly more prone to diagnose SZ when hallucinations were present and non-AA clinicians were more likely to diagnose SZ when negative symptoms were present (Trierweiler et al. 2006).

In a recent brief review regarding diagnostic bias among AAs, Escobar (2012) suggested that the higher rates of SZ diagnosis among AAs in the USA may be due to clinicians overevaluating psychotic symptoms and minimizing affective disturbances when making diagnoses. Further, he suggested that in diagnosing SZ clinicians’ bias may include factors of discrimination and stigma. White clinicians could perceive black patients with suspicion and fail to understand cultural nuances that could give hints about other diagnoses (Escobar 2012).

There is considerable research literature suggesting that AAs tend to be diagnosed with mood disorders less often than whites (e.g., Bell and Mehta 1980; Rollman et al. 2002; Breslauet al. 2006; Jimenez et al. 2010; Williams et al. 2007). Jonas et al. 2003 examined racial differences in depression diagnoses in a sample that included over 7000 adults. Results indicated AAs were more likely to be diagnosed with dysthymia, while whites were more likely to be diagnosed with major depressive disorder (MDD). The Diagnostic Interview Schedule (DIS) imbedded in a larger battery was used for diagnostic purposes in this study. In line with these findings, Woodward et al. (2013) reported older whites and Caribbean Blacks had significantly higher lifetime prevalence rate of MDD than AAs. However, no racial differences were found in 12-month prevalence rate of MDD. This study used data form a large epidemiological study that utilized the Composite International Diagnostic Interview (CIDI) for diagnostic purposes. In contrast with these findings, studies conducted in primary care settings have found similar rates of depressive symptoms between AAs and whites (Brown et al. 1999; Diala et al. 2001; Oquendo et al. 2001). The difference in the rate of diagnosing depression between AAs and Caucasians could be related to several factors. One of them is the manifestation of depressive symptoms among AAs. AAs have been found to report more physical symptoms of depression instead of mood symptoms (Brown et al. 1996). Another factor is that there might be a higher prevalence of psychological protective resources among AAs, such as religion (Taylor et al. 2001; Chatters et al. 2008) and familial support (Chatters et al. 1985; Woodward et al. 2008). Further, research suggests that AA children are socialized to cultivate a high level of tolerance to unfair acts (e.g., racism, low SES, living in high-crime neighborhoods; Thornton et al. 1990). This type of upbringing could result in a protective psychological resource that buffers the manifestation of MDD.

Racial differences in the prevalence of posttraumatic stress disorder (PTSD) have also been reported, with minorities, particularly AAs and Hispanics having a higher prevalence of PTSD compared to whites (e.g., Lonigan et al. 1991; Norris 1992; Green et al. 1990; Kulka et al. 1990; Sutker et al. 1995). Graves et al. (2011) investigated the characteristics of PTSD in AAs in the primary care setting. They screened over 700 AAs for PTSD using the SCID IV and the Clinician Administered PTSD Scale for DSM-IV (CAPS) and concluded that most AA adult primary care patients in this sample were either undertreated or underdiagnosed. In a study that examined the influence of race in PTSD treatment, AAs were found to be significantly less likely to complete treatment compared to whites (Lester et al. 2010). In contrast with these findings Monnier et al. (2002) conducted a study that examined racial differences in outpatients seeking treatment for PTSD at a Veterans Affairs Medical Center. Participants were 71 Caucasians and 40 AAs and they were administered the CAPS in addition to an open clinical interview and other self-report measures for diagnostic purposes. Between racial groups no significant differences were found regarding anxiety , paranoia, dissociation, SZ, depression, and PTSD symptomatology. The researchers concluded that AAs and white veterans with combat-related PTSD did not differ in terms of severity and manifestation of symptoms. In a similar study using the CAPS that investigated symptom patterns and service use among AA and white veterans at a Veterans Affairs outpatient PTSD clinic, very few significant differences among groups were found. This suggests that white and AA veterans do not differ in the manifestation of PTSD and in the use of services (Frueh et al. 2004). In a critical review regarding racial differences in combat veterans with PTSD, Frueh (1998), suggested that the disparity in rates of PTSD between racial groups could be a function of distinct rates of traumatic stressors and other preexisting conditions. In addition, the general paucity and methodological limitations in the empirical data in this body of research considerably limits the conclusions that can be reached.

Cultural Mistrust

There is a documented trend of cultural mistrust among AAs (Nickerson et al. 1994; Terrell and Terrell 1984; Whaley 2001a). Whaley (1998) conducted a study that examined the experience of paranoia in AAs from a cross-cultural perspective. Participants included 96 depressed patients, 65 patients with SZ-like disorders, and 404 community members. Findings indicated AAs with and without a psychiatric diagnosis scored significantly higher on measures of distrust and perceived hostility by others. The author suggested that mild paranoia was a result of cultural mistrust. Cultural mistrust is defined as guardedness toward the dominant culture related to discrimination experienced by AAs. It stems from a long history of discrimination and racism towards AAs and the previously described long-held stereotypes and beliefs about this minority group. In relation with this cultural mistrust, AAs tend to underutilize mental health services compared to other ethnic and cultural populations (Whaley 2001b). AAs often might have negative feelings toward the mental health system and varying degrees of mistrust regarding Caucasians in the mental health system. Mistrust of mental health could influence current perceptions among AAs and subsequently affect the interviewing process (Aklin and Turner 2006). According to Whaley (1998) mild paranoia is higher in individuals of lower education and income and could be misinterpreted as psychopathy by clinicians. Further, misinterpretation of paranoid symptoms related to self-esteem protection plays a major role in misdiagnosing depressed individuals as schizophrenics. Among AAs cultural paranoia can be a nonpathological or normative behavior that functions as protection against racially based threats and discrimination (Ridley, 1984). Cultural mistrust being misinterpreted by clinicians as a symptom of psychosis instead of depression may contribute to the higher prevalence of SZ and lower prevalence of MDD diagnoses among AAs (Whaley 1998). This mistrust brings to light a significant assessment concern. Such lack of trust may mean that those who seek help are likely to be those in more acute distress. This makes it challenging to assess the actual base rates in a population of certain disorders thus affecting a clinician’s judgment of population base rates. Further, clinicians may misinterpret such mistrust as paranoia or noncompliance, even when structured or semistructured interviews are employed for diagnosis. Clinicians should be aware that many AAs may experience apprehension when going to receive services and not let this phenomenon influence their diagnostic decisions. Clinicians that recognize a distrustful attitude in their patients should be flexible in their approach (Whaley 1997). Engaging in an honest conversation with the patients regarding their cultural values, experiences of racism , and mistrust could be helpful in building the therapeutic relationship and avoiding diagnostic errors (Aklin and Turner 2006).

Brief Review of Commonly Used Structured and Semistructured Interviews

Structured and semistructured clinical interviews traditionally demonstrate good reliability in the diagnosis of mental disorders. Such interviews, however, require a great deal of clinical acumen to operationalize properly, and therein lies one of their weaknesses for cross-cultural diagnosis; clinicians have a long, well-documented history of overdiagnosing severe psychopathology, and notably psychotic disorders in AA clients (Bell and Mehta 1980; Neighbors et al. 1999; 2003; Strakowski et al. 2003). The importance of proper use and interpretation of psychodiagnostic interviews cannot be overstated. The following are commonly used diagnostic structured and semistructured interviews for Axis I disorders available for clinicians to utilize with AAs.

Structured Clinical Interview for the DSM-IV (SCID)

The SCID (First et al. 2002) is considered by many as the gold standard for Axis I psychiatric diagnoses. This semistructured interview is designed for use with adults who have an eighth grade or higher reading level. There are several versions of the SCID, with versions developed for use in research (SCID-I-RV) and clinical settings (SCID-CV), although the shortened clinical version limits the diagnostic scope. The SCID-I-RV is broadest in scope and allows subtype specification, severity, and course specifiers. There is a patient edition (SCID-I/P) designed specifically for individuals with psychiatric disorders in research settings, as well as a nonpatient version (SCID-I-N/P) for use with participants who are not identified as having a psychiatric disorder. In addition, a draft version of the SCID has also been developed to cover childhood disorders (KID-SCID).

The SCID is a long, in-depth interview designed to screen for the great majority of DSM-IV diagnoses and strives to be unbiased in assessing diagnostic criteria. Administration time is approximately 1–2 h, however subsections of the SCID may be used in isolation when specific diagnostic concerns warrant precise investigation. This being said, full SCID versions may be inefficient as a broad-range diagnostic tool outside the research environment. A thorough SCID interview by an experienced clinician will still take well over 1 h for most clinical cases. As we transition into the SCID DSM-5 version, a more in-depth screening module is used to reduce total administration time. This being said, it is unlikely that the SCID will become practical as a whole for administration outside the research setting for most clients.

In its many forms, the SCID is perhaps the most widely used diagnostic research instrument in the English language, and rightly so. It offers an efficient, yet thorough treatment of diagnosis for a multitude of Axis I disorders, is modularized and thus can be tailored to fit the scope of the assessment, and is easy to learn and use for someone well versed in psychopathology. The SCID has been compared to clinically derived diagnoses (Shear et al. 2000) as a metric for validity of clinician diagnoses of nonpsychotic disorders in a large catchment study, and very low kappa coefficients (0.15) for overall agreement between SCID and charted diagnoses were attributed to the SCID’s superiority as a diagnostic method. Similarly, Steiner et al. (1995) found poor reliability for the DSIII-R version of the SCID when compared to clinical diagnoses, also touting the merits of structured interviews. As with many structured interviews, the SCID has acceptable-to-excellent interrater reliability, with adequate administrator training (Lobbestael et al. 2011; Ventura et al. 1998).

Applications with AAs

Clinical judgment is a strong requisite when conducting semistructured interviews. The name “Structured Clinical Interview” belies the freedom a clinician has when making judgments. As such, one can be lulled into a false sense of security when making diagnostic distinctions using the SCID and fail to take into consideration specific and highly relevant cultural considerations that may prove important in the diagnosis, or rule-out, of any of a variety of mental disorders. Similarly, the structure of this diagnostic system is such that it cannot prevent the overinterpretation of the expression of psychotic symptomatology as a function of race; a phenomenon frequently observed in the diagnostic literature. Research on prevalence rates of eating pathology demonstrates significantly different attitudes toward body image and maladaptive eating patterns among AAs than among whites in the USA (Akan and Grilo 1995). Failing to consider the context in which a diagnostic criterion may or may not cause significant impairment in social or occupational functioning risks misdiagnosis.

Few studies report on validity or reliability measures of the SCID for AA clients. Those that do report a significant minority of AA clients also find (similarly to other studies of the SCID vs. other diagnostic strategies) abysmal concordance with clinical diagnoses given by open interviews (see Steiner et al. 1995). It is therefore challenging to tease apart the flaws in nonstructured clinical interviewing as opposed to those of a more structured interview format. The strong concordance between SCID diagnoses and those rendered by the MINI and CIDI (See Sheehan et al. 1998, 1997) ameliorate much concern with regard to the SCID’s validity as a diagnostic instrument, though such lack of in vivo clinical concordance is grounds for pause.

The Diagnostic Interview Schedule (DIS)

The Diagnostic Interview Schedule version IV (DIS-IV; Robins et al. 1995) is a completely structured diagnostic interview developed to assess current and lifetime presence of mental disorders included in the DSM-IV. It is organized in 19 diagnostic modules that cover a wide variety of Axis I disorders. Diagnoses include substance use disorders , SZ, mood disorders, anxiety disorders , and a small selection of other disorders and those originating in childhood. The DIS can be administered by professionals and nonprofessionals with 1 week of training. Administration time is approximately 2 h. Questions are read verbatim and no opportunity is given for unstructured questions. A training manual that describes how to reliably code the clinical ratings of specific items is included (Robins et al. 1991). Items are scored in a format that combines clinical relevance and possible etiology. In addition, interviewers have to make ratings about the onset, duration, and recency of symptoms. Some of the DIS’s fundamental characteristics are that it attempts to elucidate any organic etiology (exogenous substances, medical conditions), it includes the Mini Mental Status Examination (MMSE; Folstein et al. 1975), and it allows researchers to conduct comparisons across diagnostic systems through the inclusion of older DSM criteria, Feighner criteria, and Research Diagnostic Criteria (Rogers 2001).

The DIS was originally designed as a research tool for large epidemiological studies (Epidemiologic Catchment Area Program) to assess the prevalence and incidence of mental disorders in the USA (Regier et al. 1984). Several computerized versions of the DIS have been developed, although these have less diagnostic coverage. In addition, there is a shortened paper and pencil version that can be self-administered and covers depressive, anxiety , and alcohol-use disorders (Kovess and Fournier 1990). The DIS-IV has been translated into several languages, including Spanish and Chinese, which makes it a good tool for cross-cultural applications and research. The DIS has well-established validity and reliability for pre-DSM-IV diagnostic systems, making generalization to current diagnostic systems challenging. Overall, research shows moderate-to-good reliability and validity for the DIS (e.g., Helzer et al. 1985; Wells et al. 1988; Hesselbrock et al. 1982; North et al. 1997).

The DIS is limited in that it focuses more on etiology than symptom severity, it emphasizes diagnosis over symptom evaluation, and research suggests that it is vulnerable to response styles (Alterman et al. 1996, Cottler 1998; Rogers 2001). The DIS demonstrates utility in screening large samples of people for undetected mental disorders (epidemiological research), can be administered by nonprofessionals, and has been translated and validated in several languages. In addition, there is a children’s version of the DIS: the Diagnostic Interview Schedule for Children (DISC; National Institute of Mental Health 1991; Columbia DISC Development Group 1999).

Applications with AAs

Robins et al. (1984) conducted a large epidemiological study using the DIS. Results from 9543 participants indicated that AAs had a higher prevalence of mental disorders than those of other races across all three sites. Few other diagnostic differences were found inconsistently across sites. In most DIS studies racial differences in diagnoses or the utility of the tool with AAs was either not explored or was not the main focus of the study, even when AAs comprised a large part of the sample. North et al. (1997) conducted a study comparing the diagnostic utility of the DIS and an open clinical interview in two mental health clinics for homeless people; samples were 75 and 69 % AA, respectively. Compared to the clinical interviews, diagnosis made with the DIS underdiagnosed antisocial personality disorder and overdiagnosed MDD. Alcohol-use disorders and SZ showed only small discrepancies related to the method of diagnoses. Chantarujikapong et al. (1997) compared the diagnostic utility of the DIS III and the Alcohol Dependence Questionnaire (ADS) in a sample of 143 homeless substance-abusing women, most of whom were AAs. Overall, the ADS showed acceptable agreement with the DIS in this study across substance use, personality disorder , and PTSD diagnoses. More recently, Cook et al. (2010) estimated the 12-month prevalence of psychiatric disorders and frequency of treatment seeking in 744 low-income pregnant women. 57.5 % of the sample was AA and the DIS-IV was used to establish diagnoses. AA women were less likely than Caucasians to have at least one mental disorder, and Caucasians also had a higher prevalence of lifetime psychiatric disorders, particularly affective disorders, substance-abuse disorders, and attention deficit hyperactivity disorder (ADHD).

Schedule of Affective Disorders and Schizophrenia (SADS)

The Schedule for Affective Disorders and Schizophrenia (SADS; Endicott and Spitzer 1978) is a long-standing semistructured interview with a strong presence in the literature. The SADS is established as a gold standard for the assessment of its eponymous constructs. It contains modules for current and lifetime diagnoses. The SADS should be used by an experienced clinician with a strong background in diagnostic assessment, and may take from 45 to 75 min to assess for current symptomatology, and an additional 15 min to 1 h for additional assessment of lifetime mood and psychotic episodes (Rogers 2001). Note that these times reflect administration of the modules in the order discussed.

Applications with AAs

Few of the many studies using the SADS have focused on AA clients. Some have demonstrated striking differences in diagnostic prevalence. In one such early study, Vernon et al. (1982) found that AAs and Mexican Americans had a rate of bipolar II disorder twice that of Caucasians. Other research using this measure has found that for bipolar I disorder, no such differences exist between Caucasian and AA clients, and that suicidality is rated as being far lower by the SADS for AAs than for Caucasians (Dilsaver et al. 1994). Other work has pointed out that being AA is predictive of a diagnosis of SZ using the SADS (Pavkov et al. 1989). There is a stark absence of recent comparative diagnostic research using this measure in the assessment of AA clients.

Though this measure has demonstrated great utility in the diagnosis of affective disorders and SZ spectrum disorders, it is somewhat limited in scope. Further, given the time investment necessary, this measure may be best used when a thorough characterization of a client’s affective or psychotic symptoms is warranted.

Present State Examination (PSE)

The Present State Examination (PSE; Wing et al. 1967) is one of the most commonly used structured interviews within the International Classification of Diseases (ICD) framework; therefore it is more popular outside of the USA. The PSE focuses on the presence and description of any major disorder or symptoms and not on the association of specific symptoms with clinical diagnoses (Wing et al. 1967; Rogers 2001). For example, symptoms of depression may be subsumed in several diverse syndromes. The PSE is currently in its tenth edition and previous editions have varied significantly. This should be taken into account when reviewing its validity and reliability studies. The PSE 10th ed. (PSE-10; World Health Organization 1994) has total of 1224 items that evaluate from a descriptive perspective a wide variety of symptoms. Generally, the PSE has been shown to have moderately to high validity and reliability (e.g., Huxley et al. 1987; Mignolli et al. 1988; Lesage et al. 1991; Wilmink and Snijders 1989; Peveler and Faiburn 1990).

Applications with AAs

To our knowledge there are not any studies in the literature that investigate the validity of the PSE with the AA population or focuses on cultural aspects regarding AA; however the PSE has been widely used in cross-cultural research outside the USA (e.g., Hodiamont et al. 1987; Vasquez-Barquero et al. 1987). It has been translated into 40 different languages and it has been shown to be clinically useful in European countries (e.g., Garyfallos et al. 1991), English-speaking countries (e.g., Romans-Clarkson et al. 1990), and African countries (e.g., Katz et al. 1988). In a cross-national study of ten countries, results indicated that developing countries (e.g., India, Nigeria) had a higher incidence of brief psychoses than industrialized countries (e.g., Ireland, Japan, USA; Susser and Wanderling 1994). In Great Britain, people of African descent had a higher probability of being hospitalized or detained (Goater et al. 1999). In another study, Katz (1988) investigated cultural specific dimensions of psychotic disorders, in which Indian and Nigerian patients with SZ were assessed using the PSE. Indian patients showed more systematized delusions and olfactory hallucinations , whereas Nigerian patients showed more delusions of control, thought insertion, and visual hallucinations. Swartz et al. (1985) suggested that some symptoms were culturally bound and therefore difficult to interpret meaningfully. The application and validation of the PSE with diverse cultures is one of its major strengths. Nevertheless, the fact that it is based on the ICD and the lack of studies specifically focused on AAs limits the use of the PSE for US clinicians.

Schedules for Clinical Assessment of Neuropsychiatry (SCAN)

The Schedules for Clinical Assessment of Neuropsychiatry (SCAN; WHO 1994) was developed under the sponsorship of the World Health Organization (WHO) and its main component is the PSE 10th ed. The three main aims of the SCAN are rigorous clinical observation, common clinical language standardized across different diagnostic systems, and accumulation of clinical knowledge (Wing 1996). The SCAN is comprised of 27 sections. The PSE-10 (described above) is the first 25 sections. The other two sections are the Item Group Checklist (IGC) and the Clinical History Schedule (CHS). The IGC is composed of 59 ratings defined in PSE-10 terms and based on secondary sources such as records or informant. The CHS is an optional section of 88 items for the recording of childhood data, intellectual functioning, social relationships, adult personality, clinical diagnoses, and physical illness. The SCAN is a very detailed structured interview that contains standardized questions and optional probes.

Administration time is approximately 60–90 min. SCAN interviewers are required to undergo intensive training prior to administration. The SCAN was intended to be used by mental health professionals; however, it can also be administered by nonprofessionals that undergo extensive training and direct supervision (Rogers 2001). Considering that the PSE is the major component of the SCAN, their purposes are similar in that the focus is on symptomatology and the description of clinical phenomena more than linking symptoms to a diagnoses. The validity and reliability of the SCAN have been shown to be moderate to high (e.g., Easton et al. 1997; Wing et al. 1998; Brugha et al. 1999; Farmer et al. 1993, 1996; Hapke et al. 1998).

Applications with AAs

To our knowledge there are no studies in the research literature that focus on the effectiveness of the SCAN to evaluate AAs. Many of the reliability and validity studies included black people, but ethnic differences were not examined or not reported. More studies are needed to address the utility of the SCAN with AA.

Composite International Diagnostic Interview (CIDI)

The CIDI was developed under the sponsorship of the WHO. It was based on the DIS and expanded with questions from the PSE, but several items were modified to be more useful internationally. The main purpose of the CIDI was to facilitate cross-cultural epidemiologic and comparative studies. Further, it is a highly structured interview. Nonprofessionals can administer it after extensive training, and it is easily translated to different languages (Robins et al. 1988). In addition, the CIDI provides both DSM and ICD diagnoses. Administration time varies from 75 to 105 min depending on the experience of the interviewer. The current version of the CIDI (3.0) is composed of 42 sections that assess a wide variety of disorders included in the DSM-IV and ICD-10. There is a computer version available, the CIDI 3.0 Computer Assisted Personal Interview (CAPI V21.1.3). The CIDI has been translated to several languages and training materials are available in Arabic, Dutch, English, and Spanish (WHO, 2004). A CIDI Primary Health Care Version developed to address psychological problems frequently seen in medical settings (Janca et al. 1994) is also available. The reliability and validity of the CIDI has been shown to be moderate to high (Witchen et al. 1994; Peters et al. 1998; Cottler et al. 1997; Janca et al. 1992). However, these studies frequently report on data across language versions and diagnostic frameworks, so they should be interpreted with caution. Nevertheless, this is expected considering the cross-cultural focus of the CIDI and the multiple versions that are available.

Applications with AAs

Hickman III et al. (2010) conducted a study investigating past-year mental illness among 3411 adults identifying as Black Americans, with family having lived in the USA for three generations. A modified version of the computer version of the CIDI found lifetime prevalence of any mental disorder was 36.9 %, the prevalence for past-year mental disorders was 18.1 %, and the prevalence for past-month mental disorders was 4.7 %. low income and education were associated with high prevalence. Hickman III et al. also noted that divorced/separated or widowed status, and residence in the Northeast or Midwest predicted MDD for AA clients. Further studies on various samples of this database (Aranda et al. 2012; Woodward et al. 2013) have examined the prevalence of MDD in those 50–55 and older, and observed that older AAs had lower rates of MDD than whites or Caribbean black participants, that higher age and lower disability for all groups was associated with lower MDD than younger age and more disability, and that non-Hispanic whites and women had the highest prevalence of MDD across groups. In another study using the same database along with samples from two other large databases, Chou et al. (2012) found that perceived racial discrimination was associated with higher endorsement of psychiatric symptoms across racial minority groups. Notably, for AAs in this sample, higher rates of perceived racial discrimination were associated with elevated PTSD relative to Asian Americans, and had significantly higher rates of discrimination than other groups.

These studies suggest that the CIDI is a useful tool to assess AAs. However, there are very few studies that investigate the effectiveness of the CIDI compared to other interviews when assessing AAs. Green et al. (2012) conducted a study comparing DSM-IV diagnoses assessed by the CIDI and the Schedule for Affective Disorders and Schizophrenia for School Age Children (KSADS). Findings indicated that the sensitivity of the CIDI varied by race/ethnicity for ADHD, agoraphobia, panic disorder , and PTSD. Further, the specificity of the CIDI varied by race/ethnicity for agoraphobia. In line with these results Alegria et al. (2009) reported that the CIDI was problematic for accurate PTSD diagnosis in racial/ethnic minority groups. More studies like these are needed to further investigate measurement validity according to race and ethnicity.

Mini International Neuropsychiatric Interview (MINI)

As a diagnostic screening tool, the clinician-rated MINI (Sheehan et al. 1998) offers great utility and is indeed among the most widely used psychiatric structured diagnostic interviews. It is a quick, robust tool for the assessment of current, past, and lifetime diagnoses of 17 common ICD-10 and DSM disorders. Further, given its alignment with ICD-10 diagnostic categories, less may need to change for the MINI as the transition to the DSM-5 is made. In a series of validation studies (Sheehan et al. 1998), the MINI showed high negative predictive value (> .92) for all assessed diagnostic categories when compared with the SCID, and similarly (> .88) for the CIDI. This instrument also requires substantially less time to administer than either, has good-to-excellent kappa for the great majority of diagnostic criteria, and requires less training to properly administer than do many structured and semistructured interviews, such as the SCID. Two 120-min training sessions have been shown to be sufficient for raters with an undergraduate level education (Black et al. 2004; Gunter et al. 2008). It is a well-validated measure translated into more than 30 languages and is thus a good tool for the assessment of English-as-a-second-language clients.

Sheehan et al. (1997) assessed the validity of the MINI in relation to other gold standard structured diagnostic systems (SCID, CIDI), and for the majority of diagnostic categories found high agreement. This study, however, used an exceptionally homogeneous sample (96 % white), leaving much to the imagination for its validity within an AA sample.

Upon validation, the MINI demonstrated good interrater and test–retest reliability for the original sample of participants (Sheehan et al. 1998, 1997). Furthermore, good-to-great reliability coefficients were obtained in all studies noted above. Since robust demographic data were not presented, again, the reliability of this measure for the assessment of AA clients cannot be assumed.

Applications with AAs

The MINI has demonstrated success in the classification of depression and rule-out of severe psychiatric pathology in a small sample of low-income HIV-positive AA clients (Himelhoch et al. 2011). The MINI Kid has also been proven an effective screener for African child refugees in Austria (Huemer et al. 2011). There have been several studies demonstrating its effective use with Kurdish language speakers in Iraqi Kurdistan (Mitchell et al. 2011); Japanese language speakers in Japan (Otsubo et al. 2005); as a French and English socioculturally adapted version for asylum seekers in Geneva (Durieux-Paillard et al. 2006); to assess depression and depressive symptoms in a group of breast cancer patients in Lagos, Nigeria (Popoola and Adewuya 2012); and for the assessment of ADHD in an Iowa state prison sample (Westmoreland et al. 2010). The MINI is also used widely in clinical psychological and psychiatric practice around the world. Clearly then, this measure has demonstrated great generalizability across translations and cultures with at most minimal alterations (see Durieux-Paillard et al. 2006).

Though there is a good deal (comparatively) of research extolling the use of the MINI with a broad range of culturally diverse clients, there is far less information specifically validating its use with AAs. Even studies such as those addressing African refugees in Austria or Africans in Lagos are only tangentially related to such persons in the USA, and leave room for acculturative differences in the manifestation and interpretation of psychopathology.

Scale for the Assessment of Positive Symptoms (SAPS)

The Scale for the Assessment of Positive Symptoms (SAPS; Andreasen 1984) is considered a gold standard rubric for the assessment of positive symptoms in psychopathology, notably in persons with SZ. It is also used in other forms of mental illness for which positive symptoms are a prominent aspect, such as bipolar disorder. It provides four separate subscales addressing domains of positive symptomatology commonly found in persons with psychotic disorders. Given its long-standing use as a research and clinical tool, it is not hard to understand why this tool is considered by many to be a “gold standard.”

An excellent treatment of the reliability, validity, internal consistency, and score differentials as gained from well-trained administrators within a focused SZ setting is given in Chap. 16 of this volume, “Assessing Psychosis in African American Clients.”

Clinician Administered PTSD Scale for DSM-IV (CAPS)

The CAPS (Blake 1998) is a structured interview that aims to assess DMV-IV PTSD criteria in detail. Each inclusion criteria is evaluated according to frequency and intensity. Descriptive information is also inquired for relevant examples and to address inherent inaccuracy in self-reporting. Further, current and lifetime episodes are evaluated. Administration time is ranges from approximately 45 min to 1 h. The CAPS can be administered by nonprofessionals with a moderate level of training and extensive interviewing experience. The CAPS has been shown to have high reliability (Weathers et al. 1999; Fleming and Difede 1999) and validity (Davis et al. 2000; Weathers et al. 1999).

Applications with AAs

In two separate studies of AA and white service members (Frueh et al. 2004; Monnier et al. 2002), the CAPS did not demonstrate significant differences in anxiety , paranoia, dissociation, SZ, depression, and PTSD symptomatology between racial groups. Further, studies have assessed AAs with and without trauma exposure with the CAPS (Mellman 2009), have used the CAPS with the SCID to screen AA adults in primary care for PTSD (Graves et al. 2011), and have used the CAPS to diagnose participants with PTSD in a study examining the influence of race on CBT for PTSD (Lester 2010). The aim of the above entwined studies was not to evaluate the sensitivity of the CAPS to cultural factors or its validity in diagnosing AAs. However, compared to other interviews the CAPS has been used more often with AAs and seems to be a effective diagnostic tool for this population.

Other structured and semistructured interviews that are available for clinicians to use include the Royal Park Multidiagnostic Instrument for Psychosis (RPMIP; McGorry et al. 1988), the Diagnostic Interview for Genetic Studies (DIGS; Nurnberger et al. 1994), the Clinical Interview Schedule—Revised (CIS-R; Lewis et al. 1992), the Psychiatric Diagnostic Interview (PDI; Othmer et al. 1981), and the Polydiagnostic Interview (PODI; Phillp and Maier 1986). These clinical interviews are not as widely used as those described above and there is limited empirical information, especially concerning their use with AAs or in cross-cultural application. A review of these measures is beyond the scope of the current chapter. For a brief review of these tools and for a more extensive review of the most popular tools see Rogers (2001).

Conclusions and Recommendations

Currently there is a paucity of research regarding the validity and accuracy of the aforementioned structured and semistructured interviews when used with minority populations, in particular AAs. Some of the instruments have been used with AA populations; however, there are very few studies that focus specifically on the cross-cultural validity and reliability of the instrument with AAs. Nevertheless, the structured and semistructured interviews reviewed in this chapter could potentially be helpful in reducing diagnostic bias due to cultural factors (criterion and information variance). For example, structured and semistructured interviews ensure that a wide variety of symptoms are covered, standardize language, reduce the possibility of clinician stereotypes interfering because they restrict the range of questions, and also reduce the possibility of misinterpreting answers because many of the answers tend to be dichotomized. Several of the interviews reviewed seem to be very promising instruments that could be applied cross-culturally. Instruments such as the CIDI and the MINI were developed for cross-cultural applications. With the amount of research available it is not possible to conclude which interview would be the best to use with AAs. However, most of the interviews presented have been used in studies that included AAs. It is difficult to draw clear conclusions from these studies due to the substantial differences in the samples used, instruments used, and also the version of the instrument used. Nevertheless, some of the structured and semistructured interviews that have been used more often with AAs are the SCID, the DIS, the CAPS, the CIDI, and the SAPS. On the other hand, because some of these interviews are so standardized (fully structured), they cannot be tailored to the particular cultural needs of a population and questions cannot be modified accordingly. In the case of PTSD, it has been suggested that fully structured interviews (e.g., CIDI) are differentially biased towards minorities because they are less able than semistructured interviews to take the cultural context of trauma and trauma-related symptoms into account (Alarcon 1995). However, when using a semistructured instrument the clinician has to be culturally component enough to be able to interpret the cultural context correctly. Considering the lack of research in this area, clinicians’ awareness of possible diagnostic bias and cultural competence seem to be essential to avoid diagnostic errors. If a clinician is culturally incompetent a fully structured interview, such as the CIDI, would be more appropriate to use in order to reduce as much variance as possible. Nevertheless, a culturally incompetent clinician should attempt to refer a minority client to somebody that is an expert in that client’s culture. If a clinician is culturally competent a semistructured interview, such as the SCID, that allows for the incorporation of questions about cultural and ethnic values and factors would be recommended. For example, these questions could be added during the SCID overview. Another possibility is using both a structured interview and an interview that addresses cultural and ethnic factors. Grieger and Ponterotto (1995) developed a useful conceptual framework that can be integrated to the interview process or used as a complimentary assessment with minority group members, including AAs. Based on the constructs of worldview and acculturation they developed the following components as being fundamental in culturally situating the client and conceptualizing the client’s presenting problems within a relevant cultural context: the client’s level of psychological mindedness; the family’s level of psychological mindedness; the client’s and family’s attitudes towards helping and counseling; the client’s level of acculturation; the family’s level of acculturation; and the family’s attitudes toward acculturation. Grieger (2008) expanded this conceptual model by developing 11 categories: (1) Problem Conceptualization and Attitudes Towards Helping, (2) Cultural Identity, (3) Level of Acculturation, (4) Family Structure and Expectations, (5) Level of Racial/Cultural Identity Development, (6) Experiences with Bias, (7) Immigration Issues, (8) Existential/Spiritual Issues, (9) Counselor Characteristics and Behaviors, (10) Implications of Cultural Factors Between the Counselor and the Client, (11) Summary of Cultural Factors and Implications for Diagnosis, Case Conceptualization, and Treatment. Using a conceptual framework such as this can be helpful in acknowledging and identifying cultural and ethnic factors that can be influencing the diagnostic interview process with an AA client. Grieger (2008) presented a Cultural Assessment Interview Protocol, which contains a set of questions based on her conceptual framework. Questions such as these could potentially be integrated in a semistructured interview or asked separately (See Grieger 2008 for details) The DSM-5 also includes an interview to aid with the cultural formulation that contains several useful questions.

Though semistructured interviews do indeed provide a good scaffold to support empirically based diagnostic decision making, they are just that: a scaffold. They do not remove the burden of well-honed objective clinical skill from the equation. It is imperative that a clinician understands his or her biases, the background of a patient, and does not rely on the instrument to provide context for a unique person. Further, it is imperative that a clinician realize the extent to which many diagnostic considerations are indeed simply a measure of degree rather than always objective and self-evident. Considering the research reviewed in this chapter, when interviewing AAs cultural competence is essential. For AAs, the cultural sensitivity of the mental health provider is one of the most important characteristics. Culturally sensitive counselors recognize that race or culture might play role in the patient’s problem. Culturally blind counselors tend to focus on aspects other than race when dealing with the presenting problem. Counselors that are culturally sensitive are seen as more competent by AAs than those who are culture blind (Pomales et al. 1986; Want et al. 2004). Cultural competence involves more than just awareness and practice of the considerations reviewed in this chapter. It is an ongoing process that includes formal training, self-education, consultations with colleagues that are more knowledgeable, use of collateral information when conceptualizing cases, and attending workshops and continuing education programs. U.S. Department of Health and Human Services. (2001) In addition, experience working with AA clients and immersion to experience the AA culture firsthand are fundamental.