A Qualitative Study of Clinicians’ Use of the Cultural Formulation Model in Assessing Posttraumatic Stress Disorder

doi:10.1177/1363461509342948

Transcult Psychiatry. Author manuscript; available in PMC 2010 September 1.

Published in final edited form as:

Transcult Psychiatry. 2009 September; 46(3): 429–450.

doi: 10.1177/1363461509342948

PMCID: PMC2879977

NIHMSID: NIHMS204647

A Qualitative Study of Clinicians’ Use of the Cultural Formulation Model in Assessing Posttraumatic Stress Disorder

Lisa R. Fortuna,¹ Michelle V. Porche,² and Margarita Alegría³

¹ University of Massachusetts Medical School

² Wellesley Centers for Women, Wellesley College

³ Center for Multicultural Mental Health Research, Cambridge Health Alliance and Harvard Medical School

The publisher's final edited version of this article is available at Transcult Psychiatry

Abstract

The Cultural Formulation of Diagnosis (CFD) Model of the Diagnostic and Statistical Manual (DSM) provides a potential framework for improving the diagnostic assessment of Posttraumatic Stress Disorder (PTSD) in culturally diverse patients. We analyzed data from the Patient-Provider Encounter Study, a multisite study that examines the process of diagnosis and clinical decision-making during an initial clinical intake session, in order to examine use of CFD for PTSD diagnosis. We find that the use of the CFD in routine community settings is inconsistently or underutilized in practice, but when employed may assist the formulation and interpretation of traumatic experiences. We discuss the implications for improving the assessment of PTSD in the time-limited setting of the clinical intake encounter and across race/ethnicity.

Keywords: Posttraumatic Stress Disorder, Cross-Cultural Diagnosis, Trauma, Clinical Decision Making, Cultural Formulation Model

BACKGROUND

Posttraumatic Stress Disorder (PTSD) refers to the cluster of psychological symptoms and reactions that interfere with the functioning of an individual as a result of extreme suffering including neglect and deprivation, physical abuse and injury, sexual abuse and/or exploitation, witnessing or surviving violence (American Psychiatric Association, 1994). Neurobiological reactions to such severe stressors are common throughout all cultures and based in the physiology of human beings. However, exposure and specific responses to trauma may differ significantly (Antai-Otong, 2002; Elsass, 2001; Hernandez, 2002; Hernandez, Gangsei, & Engstrom, 2007).

Common clinical assessment methods have been particularly problematic for clients from non-Western backgrounds. Researchers have found that ethnic-racial minorities tend to receive less accurate diagnoses (Alim, Charney, & Mellman, 2006; Dougherty, 2004; Antai-Otong, 2002; Carlson, 1994). Non-mainstream groups may not present classical symptoms of mental health disorders and may be diagnosed incorrectly. For example one study showed that African Americans with PTSD endorsed more items suggesting psychosis than Caucasians, which has led to misdiagnosis (Alim, Charney, & Mellman, 2006). In other cultures where open expression of emotional distress is not socially sanctioned traumatized clients may go undiagnosed (Antai-Otong, 2002; Schreiber, 1995).

Each person is uniquely tied to many cultural identities which are in turn influenced by their nationality, religion, income, age, education, sexual orientation and gender. These cultural identities interact and influence one another. For example, resource loss as the result of a natural disaster (such as a hurricane) is a trauma that is felt universally, but lower income people in developing countries may bear an inordinate amount of loss compared to their wealthier counterparts (De Girolamo & McFarlane, 1996; Norris & Alegria, 2005; Schreiber, 1995). Similarly, people with a lower socioeconomic status in the United States are liable to experience trauma and consequent post-traumatic stress symptoms because of the elevated levels of danger in their environment and the limited resources and supports to help them cope with the trauma (Alim, Charney, & Mellman, 2006; De Girolamo & McFarlane, 1996; Norris & Alegria, 2005; Ton-That, 1998). Females may be more prone to certain types of traumatic experiences like sexual abuse but also the emotional cost of trauma may differ by gender in regards to ongoing sense of safety and vulnerability which can exacerbate PTSD symptoms (Brave Heart, 1999; Breslau, Davis, Andreski, Peterson, & Schultz, 1997; De Wilde, Broekaert, Rosseel, Delespaul, & Soyez, 2007). Recent immigrants to the United States from countries engaged in political conflict are at risk for suffering from traumatic stress disorders because of their past experiences and may be unlikely to seek help due to their uncertain immigrant status (Antai-Otong, 2002; Fortuna, Porche & Alegria, in press).

The clinician needs to be able to consider the diagnostic symptom criteria within these cultural identities and contexts that influence risk and protection for the development of PTSD, its expression and its persistence. In this regard, it is important to consider how clinicians consider culturally informed models (Lim, 2006; Mezzich, 1995) which can assist with the diagnosis of PTSD and in understanding traumatic experiences across culture, gender and context in everyday clinical practice.

Cultural Formulation in Diagnostic Practice

This study uses qualitative methods to examine the extent to which clinicians in everyday clinical practice utilize the DSM-IV Cultural Formulation of Diagnosis (CFD) with patients who bring the sensitive and difficult material of traumatic experiences to the initial clinical encounter. The Cultural Formulation of Diagnosis, found in the Appendix IX of the Diagnostic and Statistical Manual-IV (DSM-IV; American Psychiatric Association, 1994), was developed by the National Institute of Mental Health (NIMH) committee on Culture and Diagnosis, a group of international cultural experts. The objective was to make the manual more culturally sensitive. The CFD model expands on the guidelines and diagnostic criteria published in DSM-IV to better incorporate cultural contexts and meanings in establishing psychiatric diagnoses for ethnic minorities and diverse populations (Lewis-Fernandez & Diaz, 2002). It consists of five components for formulating diagnosis and treatment needs by assessing the following about patients/clients: (1) cultural identity (e.g. group affiliation, acculturation); (2) cultural explanations of the illness; (3) cultural factors related to the psychosocial environment and levels of functioning (e.g. family and social support); (4) cultural elements of the clinician–patient relationship; and (5) the overall impact of culture on diagnosis and care. This model is known to the mental health field through the DSM and specific training of the model was not provided to the clinicians participating in this study, rather this a naturalistic examination of their use of the model in the context of assessing traumatic histories with culturally diverse patients.

In this paper we examine findings from the Patient-Provider Encounter Study (PPES) (Alegria et al., 2008). This parent study includes as its primary aim, understanding the process of psychiatric diagnosis with ethnic minority patients. The examination of PTSD diagnosis in this paper emerges from this aim. As part of the PPES study, each clinician (psychiatrist, psychologist, clinical social worker or psychiatric nurse) was interviewed by trained research staff after conducting a videotaped clinical intake with one of the ethnically diverse patient participants seeking care in one of eight clinics of the healthcare system in the Northeastern United States. Specifically, the focus of this paper is to explore: (1) to what degree clinicians in the PPES report using Cultural Formulation of Diagnosis components; and (2) how clinicians discuss their deliberation and inclusion of cultural factors in making a PTSD diagnosis, including understanding the impact of traumatic histories. Implications for evaluating trauma related symptoms and the diagnosis of PTSD across race/ethnicity in the initial clinical encounter are then elaborated on.

METHOD

Sample

Full study sample

The Patient-Provider Encounter Study sample includes clinicians and patients studied in naturalistic outpatient clinic setting during an initial clinical intake/evaluation session. All patient participants were seen for initial evaluation by clinicians who agreed to participate in the study. Forty-seven clinicians from eight safety-net clinics in the Northeast United States participated in the study. All eight clinics offer services for uninsured patients, working families, and recent immigrants with approximately 50% of the outpatient service volume attributed to uninsured patients. In this study the recruited clinicians were ones who offered mental health services to multicultural adult populations in general outpatient clinics as well as services that specialize specifically in substance abuse treatment.

One hundred and twenty-nine patients from adult mental health outpatient clinics at these hospital sites participated in the larger study. Patients who were less than 18 years old, and/or who were identified as psychotic or suicidal by clinicians and/or who required services of an interpreter during their intake appointment were excluded from the study. Use of an interpreter was excluded in order to remove the inherent confounding factor of third party involvement in relaying clinical information. An examination of the impact on diagnosis of using an interpreter is an important question beyond the scope of our study. The study included Spanish speaking and less acculturated clients being evaluated by Spanish speaking providers. Two patients who were negative on a capacity to consent screen to establish comprehension of informed consent (Zayas, 2005) were excluded from the study. Patient participants were from diverse backgrounds with approximately 51% Latino, 12% Black (African American/Afro-Caribbean); 37% non-Latino white. Patient participants ranged in age from 18 to over 65 with approximately 80% of participants between 18–49 years old. Sixty percent (60%) of the patient participants were female, 65% completed high school, approximately 63% of patients had an annual household income of less than $15,000, and 55% of the patient sample was unemployed. Patients were interviewed in English or Spanish based on their preferred language.

Analytic sample

In this report we limit our analyses to interviews with clinicians who assessed patients for which both the clinician and the independent research clinician designated PTSD as a diagnosis. This was done in order to limit the uncertainty regarding a diagnosis of PTSD and to focus analyses on the potential cultural considerations in making and contextualizing such a diagnosis. This yielded an analytic sample of 24 clinicians serving 31 patients. This sub-sample included a higher percentage of women and a higher percentage of Latino patients and fewer non-Latino white and Black patients than the larger study (see Table 1). Approximately two-thirds of patients and providers were matched on either race/ethnicity or gender.

Table 1

Provider and Patient Characteristics for 31 Cases Fulfilling Criteria for PTSD or PTSD NOS by both Clinician and Coder

Procedure

Recruitment of patient participants was conducted through direct person-to-person solicitation upon presentation for initial intake. All patient participants completed an assessment of their capacity to consent prior to their participation. The capacity to consent was established using a 10-item screening measure based on the four legal standards of demonstrating capacity (understanding, appreciation, reasoning, and voluntarism). The measure was developed and utilized in a study of psychiatric diagnosis in a different community clinic (Zayas, 2005). We adjusted the measure slightly to correspond to the description of the current project (e.g., adjusting the acceptable answers to the questions regarding the correct identification of study aims and specific risks and benefits). Clinician participants in the study were recruited through introductory information meetings. To ensure the diversity of the clinician sample, clinicians were invited to participate only up to five times in the study (e.g., five different intake sessions); no clinician in our analytic sample participated more than three times.

Participation in the PPES consisted of three components: (1) Videotaping of the initial clinical interview between patients and clinicians which is part of standard clinical care of patients in each clinic; (2) conducting a post diagnostic research interview separately with patient and clinician following the standard clinical intake; and (3) completing survey measures following the research interviews. To insure minimal intrusion on the intake work, videotaping and audio-taping of the intakes was conducted using a small camera which was positioned at the corner of the room, with no research personnel present at the time of intake. Intake clinicians were asked to provide their diagnosis for the patient. An independent research clinician coded the tapes in terms of the clinical symptom criteria covered and were also asked to give a diagnosis (blind to the diagnosis provided by the intake clinician) for the patient using only the information available in the videotaped intake. Clinical intakes lasted between 30–60 minutes (90% were closer to 60 minutes). Duration was under the discretion of the clinician and not the investigators.

The post-diagnostic research interview with clinicians is the focus of analysis for this paper. These interviews lasted approximately 30 minutes and were conducted in English or Spanish according to clinician preference. All research interviews were audio-taped and fully transcribed in their original language using a professional transcription service. For these analyses, coding and translation of Spanish transcripts were completed by the first author. Clinician interviews included questions regarding the understanding of patient presenting problem, process of clinical decision making, and rapport with patients. While questions about the explicit use of CFD components were not asked, clinicians did respond to direct questions about the role of socio-cultural factors in patient’s presenting problem and care offered. This more open-ended examination provides the opportunity for clinicians to report the use of any or all CFD components or other strategies for diagnosis. All interviews were conducted by trained research assistants. Training and supervision of interviews was provided throughout the data collection process by an expert in medical ethnography on a regular basis, and included review of interviews, with the aim of improving interviewing skills.

Analytical Method

Analyses were conducted using NVivo7 qualitative software (QSR International) to facilitate our process of creating interpretive codes (Miles & Huberman, 1994) based on the CFD model. Exploratory qualitative analyses were used to inform the primary research questions rather than to test specific hypotheses. Because our analytical focus is to increase our understanding of clinical diagnosis of PTSD, we read through each transcript and identified a topic-level or “bucket” code (Harry, 2005) of trauma and PTSD. These buckets were comprised of discrete turn-taking sections where clinicians used the terms “trauma,” “PTSD,” or equivalent words or phrases such as “abuse” or “domestic violence.” Secondly, we identified meaning-level codes within the trauma/PTSD sections and extracted themes that corresponded to the five components of the CFD model. Finally, we developed a code for alternative diagnostic strategies, which included reliance on patient records, emphasis on substance disorder within dual diagnosis cases, and systematic reliance on DSM-IV criteria. The first two authors read through the initial ten transcripts separately to first identity trauma/PTSD bucket codes and then to code for the six meaning-level categories within those buckets. In some cases, multiple CF model codes were appropriate for the trauma description; in other cases none of the six codes were applicable. We then met and compared codes, resolving any discrepancies through consensus. Once we established agreement on codes we divided the remainder of the interviews and completed coding. Because we used both deductive and inductive coding strategies, we also compiled frequency of codes across clinicians and patients to examine whether and in what circumstances CF model components were discussed.

RESULTS

Patient and Provider Characteristics

Table 1 shows the clinician and patient characteristic for the analytical sample. Out of the total study sample, these 31 patients who were diagnosed as having PTSD by both the provider and research clinician formed the analytic sample. Twenty-four clinicians were involved in assessing these PTSD cases. These 24 clinicians served 31 patients, of whom the largest percentage was Latino (68%), followed by non-Latino white (29%), and Black (Afro-Caribbean; 3%).

Below we provide examples of themes reflecting the way clinicians discussed strategies used for diagnosing PTSD. (Clinician and patient dyads are numerically identified to allow comparison across the examples. ID numbers used in this paper are not patient record numbers, but are solely for research classification; some identifying information is obscured to ensure confidentiality).

Theme 1: Assessing Cultural Identity

Cultural Identity of the CFD model was evident in 7 out of the total 31 patient cases (23%) and represents the reflections of seven different clinicians. Clinicians discussed this component of the CFD model in regards to identifying what country the patient comes from, or what larger ethnic group the patient identifies with, in order to help identify the potential mix of socio-cultural influences that has patterned the patient’s world and risk for traumatization. For instance, various providers note the different circumstances of recent immigration for Cubans who came to the US as “political refugees,” or Central American or Haitian immigrants who came to the US for economic reasons, or Italian Americans who have long established cultural enclaves. Some described broadly the risk for a particular ethnic group, rather than specific country of origin, as this Latina clinician described, “I consider culture. [My patient] is Latina, and childhood trauma is so common in Latino groups I need to consider [this possibility] (Clinician #10, Patient #138). In this case, the patient and provider shared gender and ethnic background. In another example, a Black clinician described familiarity with her patient’s cultural background related to trauma diagnosis because of the clincian’s time spent in the Middle East. Specifically, she focused her questioning on how the patient is treated as a woman in a Middle East context and how this may heighten her risk for trauma.

Study clinicians sometimes took into account that cultures are experienced differently by different members according to subgroup characteristics such as gender, class, religion, race, and sexual orientation, among other factors. In the following case, this white female psychiatrist considers how her Latina patient identifies and fits within her cultural origins and current cultural transitions as an immigrant. She considers what impact this has on the risk for PTSD and the consequences of trauma.

She described to me that she’s [from Central America]. And I think it’s a huge part of her presentation. Many of the struggles that she’s dealt with are related to her cultural background but also [in part] to a changing culture. This includes moving when she was [in her late teens] to a new country, and I think it’s a huge part of her story. Her depression and her PTSD feel cross-cultural. It’s independent, but I think her ethnicity and her life story are hugely a part of who she is and how she suffers. (Clinician #46, Patient #423)

Theme 2: Cultural Explanations of the Illness

The heart of the Cultural Formulation is this second section, which examines the cultural factors that affect the experience and interpretation of illness, as understood by the patient, the family, and the social network. Explanations of illness help to create the illness experience by affecting cognitive, bodily, and interpersonal aspects of disease. This includes helping to shape symptom presentations, perceived etiologies, severity attributions, treatment choices, and outcome expectations. This component was reported in only 7 cases with seven different clinicians.

Two examples represent challenges to clinicians regarding symptoms which could be interpreted as psychosis or dismissed as supernatural folklore. As one white male provider describes of his Latina patient:

Well, we talked a bit about her faith and she grew up Catholic and you know when she says: “Oh, I hear”— one of her grandchildren was murdered --, “Oh, I hear this toddler playing in the hallway, when in fact there’s no one there.” Well, is that psychotic or is that sort of culturally appropriate for Hispanics. It could go either way. (Clinician #45, Patient #409)

By entertaining the possibility that cultural beliefs (ethnicity and religiosity) are a lens through which to explore a PTSD diagnosis and its symptoms, this clinician is then subsequently able to extract extended information about the trauma history. Without extended exploration of contextual background the clinician might be drawn to assumptions of delusional thinking. However, this clinician relied on the patient’s “explanations of how traumatic experiences have affected and continue to affect her” in order to determine a diagnosis. Similarly, a white female psychiatrist described a Haitian patient who reported recurring “flashbacks of his mother’s death and that “the cause of his mother’s death was related to voodoo.” (Clinician #64, Patient #608) Although the clinician was unable to get more detailed information from the patient, she was primed to pay attention to “a big cultural component” of the patient’s expression of loss and trauma during the intake.

In addition, clinicians described cultural explanations for illness that were explicitly related to the cultural background of the patient and commonly understood as such in the patient’s community. For example, a Latina psychologist described “padece de nervios” (suffers from nerves) that she observed and heard from her Latina patient:

For example, I know that in El Salvador it’s important to ask whether “someone suffers from nerves? “Do you suffer from nerves?” This is a term that is used a lot diagnostically to describe a picture very similar to PTSD.] (Clinician #2, Patient #125—Translated from Spanish)

A more subtle example is that of a Latina psychologist describing her Latina patient’s presenting complaints: “The person came to me because she had [suffered] a long period of headaches. [These were] headaches that wouldn’t go away, and [she had] nightmares, and the nightmares gave her nausea” (Clinician #8, Patient #131—Translated from Spanish). This clinician emphasized the patient’s somatic expression of illness to identify that the patient was experiencing both emotional and physical distress related to her trauma history.

Theme 3: Cultural Factors Related to the Psychosocial Environment and Levels of Functioning

This section of the CFD model allows the clinician to examine how culture influences the stressors patients are exposed to and their reactions to these situations. The clinician can also consider the social supports available to patients and the contexts against which their functioning should be measured. This component of the CFD model was evident in 12 out of the total 31 patient cases (39%) and represents the reflections of 12 different clinicians during post-intake interviews.

In some cases, the evaluation of trauma is based on what the clinician believes is the “meaning” of a particular trauma for a patient given the developmental and social norms of his or her culture. This includes a consideration of the psychosocial supports, and competing stressors at the time of the experienced trauma, for instance a white female patient is “struggling financially” but the white male provider who evaluated her also pointed out the importance of exploring “religious background and current level of spiritual involvement the client uses as part of her recovery process” both for substance abuse and mental health (Clinician #30, Patient #301)

Another level of use for this CFD component is in reflecting how current psychosocial supports and stressors exacerbate preexisting PTSD symptoms. One Latino psychiatrist reflected on how immigration and risk of deportation brings to light other underlying PTSD symptoms and stressors in the area of psychosocial support for his Latina patient,

She has suffered a lot of trauma in her life. The incident that occurred a week ago (her friend telling her that she was going to report her to the immigration agency) was a trigger that brought her back to this state of trauma. She describes her problem as PTSD. She has been feeling well for some time, but this incident put her again in such a condition. (Clinician #7, Patient #120)

Consideration of psychosocial supports and social status (such as education and employment) in identifying risks for PTSD or how traumatic environments have affected patient psychological and social outcomes were evident.

Well with people coming from developing countries now, there is often neglect and abuse. Those countries have gone through civil war. There’s extreme poverty there, and also the parents, out of economic necessity, leave the children with other relatives and they have to go to the US to get jobs and make money and try and bring the kids over. But the people they leave them with, they don’t have control over how their kids are treated. And often they aren’t nurtured and are physically abused and are made to work, hence, I’d say that’s [a] cultural, socioeconomic [risk]. (Clinician 18, Patient 172).

However racial/ethnic considerations were not as primary among some clinicians, as this white female provider goes on to say that, “in terms of being Spanish, I didn’t really notice-I mean, I just know the deprivation that he’s coming from” (Clinician 18, Patient 172). In this example the socioeconomic vulnerabilities of trauma are considered although not articulated specifically in the context of culture.

Theme 4: Cultural Elements of the Clinician–Patient Relationship

This CFD component allows the clinician to consider how his/her own interaction has affected the disclosure of the patient’s illness experience, including symptom expression and treatment response (Lewis-Fernandez & Diaz, 2002). Cultural Elements of Clinician-Patient Relationship were described by 4 different clinicians across four different cases. One Latina provider emphasized the importance of having experience working with diverse cultures in order to better understand their experiences. This clinician also mentioned the additional gendered connection she had with her Latina patient who thus found it easier to reveal a history of sexual abuse to a Latina female provider in a shared language. (Clinician #8, Patient #131)

Just as important to noting shared connections was having the insight to realize one’s limitations. For instance, a Latina provider described her historical knowledge of her Latino patient’s country of origin and his exposure to political violence, but at the same time admits, “I’ve never lived in anything like that. I think that somebody who’s lived through that would understand it much better. If I had better firsthand knowledge I would have been able to “formulate question[s] in a different way thereby getting richer information about his trauma history” (Clinician 15, Patient 158).

In contrast to these deeper reflections on the cultural elements of the patient provider relationship, a white female provider discounted the importance of shared ethnic background of her white male patient: “Well, frankly, what difference would it make?” She goes on to say that while the patient might feel that “there’s a kinship” and maybe “feel more comfortable” it is not an important consideration for her (Clinician #36, Patient 332). Similarly, a Black female provider chuckles about the fact she was able to “make a little connection” with a Middle Eastern patient by speaking “some Arabic” (Clinician #76, Patient #726). Although the connection is identified as positive for the patient in each of these cases, there is little attention to how this enhancement to the patient-provider relationship might increase diagnostic capability or in the formulation of the clinical presentation. There was no mention across any of the clinicians regarding how patients from different cultures perceive authority figures (such as clinicians) when specifically discussing trauma.

Theme 5: The Overall Impact of Culture on Diagnosis and Care

The final component of the CFD model summarizes the information in the previous sections, focusing on cultural material that contributes to diagnosis and treatment. The role that cultural features have played in determining overall illness outcome is particularly emphasized. This CFD component was described by 10 different clinicians across ten different cases in discussing patients with trauma histories. Several examples centered on interpretation of the patient’s recollection of witnessing or experiencing domestic violence. For instance, providers had to assess patient reports that downplayed histories of violence, as in this example of a Latina provider describing her intake with a Latina patient: “There are cultural-gender roles that include machismo or male dominance but [what this patient] described [is] physical, verbal, and sexual abuse”. This clinician had to consider the context of a potentially traumatic experience by assessing the boundaries between culturally based gender role expectations and trauma (Clinician #2, Patient #125—Translated form Spanish).

In another case, a white female provider presented the patient’s recollection of traumatic childhood events (“when his uncles were physically abusing him”), as well as physical evidence (“he said he still has the scar he got when he was seven”), and then placed this in a larger cultural context: “I’ve worked with a lot of people from the Dominican Republic. And they had a terrible revolution and a lot of them at a certain age saw a lot of violent things, murders and terrible things.” (Clinician #18, Patient #172). The childhood abuse was a primary trauma but by taking into account the larger context of political violence the clinician feels she gains a clearer picture of how these experiences “interfere with relationship capacity” and can become cumulative. Delving into the patient’s cultural background, specific to this patient’s experience of living through a revolution, she notes, “that affects people too, if you ask them about it.”

For this theme, two of the clinicians stated that culture was not relevant to the diagnosis. When asked outright if social, cultural, or racial background influenced the diagnostic assessment of her Latina patient, a Black female clinician said that “It didn’t” (Clinician #76, Patient #721). In the second example, a Black clinician stated that she would not say her patient is “a white woman so I give her ‘X’ diagnosis – it’s really just about history of trauma and abuse” (Clinician #21, Patient #209).

Theme 6: Other Diagnostic Strategies and Concerns

The CFD components, when used, were usually used together with other strategies in assessing PTSD in this diverse patient population. In 25 out of the 31 patient cases (81%), when clinicians discussed PTSD they reported using other diagnostic strategies, which were not related to the CFD model. These strategies most commonly included: (1) previewing patient records to ascertain a history of trauma and previous clinical assessments; (2) prioritizing substance abuse as a focus in dual diagnosis cases; (3) delineating DSM-IV PTSD symptom criteria in the interview. The most common reported barriers to more fully assessing the context of trauma included needing to streamline inquiry about trauma because of time concerns and limiting discussions of trauma in the first meeting in order to limit distress in the patient.

In summary, the most frequently coded CFD model component related to trauma and PTSD was assessing cultural factors related to the psychosocial environment (39%). The least frequently coded CFD component was cultural elements of the patient-provider relationship (13%). In 30% of cases there was no reference to any of the CFD components, 37% of cases noted one component, 7% of cases noted two components, 23% of cases noted three components and in 1 case the clinician named four components. No clinician named the CFD or reported systematic use of the full CF. (However, we should note that CFD components may have been discussed in other sections of the interview not related to trauma and PTSD, and a more general analysis of cultural aspects of the patient-provider encounter is underway (Alegria et al., manuscript under preparation). Other diagnostic strategies (reviewing old charts, asking PTSD DSM-IV criteria, deciphering dual diagnoses or multiple diagnoses and time limitations for discussing multiple issues) were more often discussed by clinicians (81% of cases).

DISCUSSION

The results of this study suggest that almost two-thirds of the clinicians found it important and meaningful to consider culture and social context within a diagnostic formulation, through the positive use of one or more of the CFD components. Clinicians did vary in the ways they articulated, and how intensively they considered the Cultural Formulation Model of Diagnosis when examining patient’s history of trauma. For many of the clinicians in our study, understanding a patient’s psychosocial support system was of importance but not necessarily understanding patient’s cultural context, especially if the patient was seen as being part of a mainstream culture. Cultural elements of psychosocial supports appeared to be a particular area of interest for clinicians when assessing a minority patient and this was especially the case in considering issues of socioeconomic disadvantage including extreme poverty and immigration.

There are important clinical perceptions and issues regarding the use of the cultural formulation model for PTSD evident in the clinicians’ interviews presented here. One central issue that clinicians reflected is the deciphering of whether what is being presented to them by the patient should be noted as traumatic or culturally normalized experiences. There is an impression by clinicians that to some degree and for certain types of experiences (e.g., domestic and inter-marital violence), trauma is in the “eye of the beholder” and this is explicitly tied to the interaction of cultural and gender norms. For example, if it is acceptable culturally for a man to hit his wife, then that experience might not be considered traumatizing. However, some data refutes the notion that immigrants have more accepting attitudes towards partner violence (Sorenson, 2006), suggesting that this is more a stereotype or an incorrect assumption from the provider. Additionally, there are calls for placing greater attention to the full psychosocial context, including social class, in working with culturally diverse victims of domestic violence (Sokoloff & Dupont, 2005). The clinicians in our sample appeared to be considering and addressing poverty in their assessments where domestic violence history was a reality.

Secondly, clinicians reported exploring whether the cultural and psychosocial environment of the patient predisposes them to PTSD or perpetuates their symptoms. What are the ongoing stressors and how do these contribute to ongoing emotional distress, dysfunction or even substance addiction given an already previous and significant history of trauma? Or in another example, how are a patient’s fears about being reported to immigration by her fellow worker and the risk of being deported related to flashbacks and the return of other PTSD symptoms? Similarly, one clinician saw the process of acculturation as one additional stressor related to cultural identity, which has an impact on exacerbating PTSD symptoms.

A third issue presented by clinicians was the difficulty of disentangling cultural context, multiple diagnoses and clinician biases in the diagnostic process. Because people with PTSD can often have complex personal and interpersonal presentations coupled with high comorbidity, dealing with culture more explicitly adds another layer of complexity. Developing scientifically sound and clinically meaningful case formulations is so challenging that it verges on becoming a "lost art." Pressures (scientific, economic, and cultural) remain that can prevent psychiatrists and other mental health clinicians from getting a complete understanding of the patient. Thus, attempting to extract and consider data related to culture, psychosocial contexts and worldview might seem only to complicate an already arduous task. The clinician who includes these factors in treatment is faced with decisions of when to discuss these issues, how to discuss them and to what depth.

Diagnosing PTSD

The literature suggests that the systematic application of clinical and racial contexts assists in drawing out extended information about the trauma history and in evaluating competing diagnoses (Alim et al., 2006; Antai-Otong, 2002). For instance, research has shown that failing to assess for race-related stressors can results in missing as much as 20% of PTSD symptoms (Loo et al., 2001). Checklists and structured interviews also fail to address specific aspects of the immigrant experience. Specifically, grief, loss of, or separation from loved ones and insecure citizenship status have been demonstrated to mediate PTSD and depression (Loo et al., 2001; Schreiber, 1995; Suarez-Orozco, Todorova, & Louie, 2002). To add to this, there is additional data showing that cultural and race-specific stressors impact the development of PTSD and its chronicity, as well as its symptomatic presentation. This has been seen in the symptomotology and chronicity of PTSD in veterans of color who have experienced racial discrimination in the context of their military service (Loo, 1994; Loo, Fairbank, & Chemtob, 2005; Loo, Lim, Koff, Morton, & Kiang, 2007; Manson, 1996) and in higher rates of PTSD in Latino police officers as compared to their non-Latino counterparts (Pole et al., 2001). Similarly, posttraumatic symptom expression is influenced by culture as shown in higher rates of dissociative and somatic symptoms in less acculturated individuals (Carlson & Rosser-Hogan, 1994; G. N. Marshall & Orlando, 2002). Systematic assessment of these culturally latent expressions of illness presentation could prove to be invaluable to understanding and addressing PTSD.

The American Psychiatric Association Diagnostic and Statistical Manual provides diagnostic criteria that is helpful in identifying PTSD through the use of a common diagnostic language (Spitzer, First, & Wakefield, 2007), and has been found to support high rates of inter-rater reliability in studies of psychiatric diagnosis in differing contexts (Ruskin et al., 1998). However, by entertaining the possibility that cultural beliefs (ethnicity, social norms, and patient cultural explanations of illness) are a lens through which to explore a PTSD diagnosis, the clinician, in many cases is able to better extract extended information about the trauma history and to evaluate competing diagnoses (e.g., PTSD related perceptions and flashbacks vs. psychosis). For example, it was considered helpful by clinicians to examine where gender and ethnic-racial differences played a role in the patient-provider relationship therapeutically and diagnostically (e.g., How do I think about diagnosis in this patient? Am I underestimating or overestimating the symptoms because of my own biases in interpreting race and culture? How might it feel for this patient to disclose to me?). The importance of a culturally informed approach to understanding the therapeutic relationship has been documented in the literature (Blanchard, Nayar, & Lurie, 2007; Dedier, Penson, Williams, & Lynch, 1999; Harris, 2004) but less so how this information influences the relationship on diagnosis. Although, socio-economic status and other social circumstances are not always explicitly described as cultural, these were important considerations which clinicians in our study brought together with culture and DSM diagnostic criteria. In this way, it may be important to expand upon what we mean by “cross-cultural” encounters for intra-racial dyads, such as white clinicians who differ from their white patients in areas of religion or socio-economic privilege. Latino clinicians in our study already appear to include such considerations within assumed “intra-cultural’ encounters with Latino patients. That is, they include socio-economic concerns and related social adversity as an important component of what might be considered a socio-economic-cultural formulation model.

Limitations

The study results are limited in generalizability, given that they only include seven clinics in the Northeast. However, our findings begin to provide an understanding of the diagnostic process across diverse mental health trained clinicians. Although, our findings lack the depth of extensive case studies found in the existing literature on PTSD and cultural formulation (Barrett, 1997; Bucardo, Patterson, & Jeste, 2008; Fleming, 1996; Lewis-Fernandez, 1996b; Lim & Lin, 1996; Manson, 1996; Yilmaz & Weiss, 2000) our examples illustrate the patterns and extent of use of the CFD model in everyday practice given its inclusion within the DSM. The study included a heterogeneous group of clinicians as it relates to discipline and variations in diagnostic approach. However, our findings suggest some testable theories regarding ways to assist clinicians of varied mental health disciplines working with diverse patients in real life settings.

Conclusions

PTSD stands out as one of those few disorders in the DSM with diagnostic criteria which are inclusive of an event which is judged to be etiological. As we move on to revising the Diagnostic and Statistical Manual to its fifth edition in the near future, there are several suggestions for capturing the cultural aspects of PTSD diagnosis that may assist clinicians in systematically considering cultural contexts. The good news is that clinicians are willing to consider culture but the components of the Cultural Formulation Model of Diagnosis are inconsistently or underutilized in the initial evaluation.

Development of training and supervision opportunities for explicit use of the model is warranted. At a time when the value of delivering culturally congruent care is increasingly being recognized and requirements for cultural assessment are being incorporated into treatment guidelines and professional training curricula, the Cultural Formulation Model of Diagnosis represents one of the main existing methods for attaining and implementing a culturally valid approach to care (Lewis-Fernandez & Diaz, 2002). Regular use of the model potentially can assist clinicians not only in how to elicit culturally relevant clinical material in diagnosing PTSD, but also exposes them over time to the content of many cultural perspectives from diverse patients and their families regarding trauma, suffering, and the expressions and understanding of that suffering.

The CFD model should be a well-integrated aspect of DSM-IV criteria-based diagnoses. The next versions of DSM may well find it essential to further integrate the physical, emotional, and cultural assessment of PTSD as the mental health field has learned a great deal about the complex interaction of all of these for illness presentation across cultures and diagnoses (Breslau & Kessler, 2001; Fabrega, 2007; R. D. Marshall, Spitzer, & Liebowitz, 1999; Munir & Beardslee, 2001; Yilmaz & Weiss, 2000). Screening materials developed and tested in diverse cultures and structured brief “checklists” that elicit CF cultural components could help orient clinicians during the brief clinical intake. For example, physical manifestations of emotional distress are common among ethnic minority patients and patients suffering from PTSD as a group (Jenkins & Cofresi, 1998; Tamayo, Rovner, & Munoz, 2007) and would be an important way to screen for the consequences of trauma.

Models for ongoing training and supervision using refined CFD/DSM criteria for diagnosis and treatment planning would be especially helpful for clinicians working within communities with high risk for trauma and PTSD (e.g. inner city mental health, settings with large numbers of refugee and immigrant populations). Testing interventions for both PTSD identification and appropriate treatment planning in multicultural populations would benefit the field tremendously. Such studies could also provide invaluable information regarding the cultural implications of both traumatic suffering and opportunities for supporting healing.

Acknowledgments

This study was supported by NIH Research Grant # P50 MHO 73469 funded by the National Institute of Mental Health, # P60 MD02261 (NCMHD) funded by the National Center for Minority Health and Health Disparities, and Robert Wood Johnson Foundation, and the National Institute of Drug Abuse 1K23 DA018715-01A2.

References

Alegría M, Nakash O, Lapatin S, Oddo V, Gao S, Lin J, et al. How missing information in diagnosis can lead to disparities in the clinical encounter. Journal of Public Health Management Practice (Supplement) 2008;146(6):S26–S35. [PMC free article] [PubMed]
Alim TN, Charney DS, Mellman TA. An overview of posttraumatic stress disorder in African Americans. J Clin Psychol. 2006;62(7):801–813. [PubMed]
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4. American Psychiatric Association; 1994.
Antai-Otong D. Culture and traumatic events. Journal of the American Psychiatric Nurses Association. 2002;8(6):203–208.
Barrett RJ. Cultural formulation of psychiatric diagnosis. Death on a horse’s back: Adjustment disorder with panic attacks. Cult Med Psychiatry. 1997;21(4):481–496. [PubMed]
Blanchard J, Nayar S, Lurie N. Patient-provider and patient-staff racial concordance and perceptions of mistreatment in the health care setting. J Gen Intern Med. 2007;22(8):1184–1189. [PMC free article] [PubMed]
Brave Heart MY. Gender differences in the historical trauma response among the Lakota. J Health Soc Policy. 1999;10(4):1–21. [PubMed]
Breslau N, Davis GC, Andreski P, Peterson EL, Schultz LR. Sex differences in posttraumatic stress disorder. Arch Gen Psychiatry. 1997;54(11):1044–1048. [PubMed]
Breslau N, Kessler RC. The stressor criterion in DSM-IV posttraumatic stress disorder: an empirical investigation. Biol Psychiatry. 2001;50(9):699–704. [PubMed]
Bucardo JA, Patterson TL, Jeste DV. Cultural formulation with attention to language and cultural dynamics in a Mexican psychiatric patient treated in San Diego, California. Cult Med Psychiatry 2008 [PMC free article] [PubMed]
Carlson EB, Rosser-Hogan R. Cross-cultural response to trauma: A study of traumatic experiences and posttraumatic symptoms in Cambodian refugees. J Trauma Stress. 1994;7(1):43–58. [PubMed]
Dedier J, Penson R, Williams W, Lynch T., Jr Race, ethnicity and the patient- caregiver relationship. Oncologist. 1999;4(4):325–331. [PubMed]
De Girolamo G, McFarlane A. The epidemiology of PTSD: A comprehensive review of the international literature. In: Marsella A, Friedman M, Gerrity E, Scurfield R, editors. Ethnocultural aspects of post-traumatic stress disorder: Issues, research, and clinical applications. Washington, DC: American Psychological Association; 1996.
De Wilde J, Broekaert E, Rosseel Y, Delespaul P, Soyez V. The role of gender differences and other client characteristics in the prevalence of DSM-IV affective disorders among a European therapeutic community population. Psychiatr Q. 2007;78(1):39–51. [PubMed]
Dougherty RH. Reducing disparity in behavioral health services: a report from the American College of Mental Health Administration. Adm Policy Ment Health. 2004;31(3):253–263. [PubMed]
Elsass P. Individual and collective traumatic memories: A qualitative study of post-traumatic stress disorder symptoms in two Latin American localities. Transcult Psychiatry. 2001;38(3):306–316.
Fabrega H., Jr How psychiatric conditions were made. Psychiatry. 2007;70(2):130–153. [PubMed]
Fleming CM. Cultural formulation of psychiatric diagnosis. Case No. 01. An American Indian woman suffering from depression, alcoholism and childhood trauma. Cult Med Psychiatry. 1996;20(2):145–154. [PubMed]
Fortuna LR, Porche MV, Alegria M. Political violence, psychosocial trauma, and the context of mental health services use among immigrant Latinos in the United States. Ethnicity and Health. 2008;13(5):435–463. [PMC free article] [PubMed]
Harris PA. The impact of age, gender, race, and ethnicity on the diagnosis and treatment of depression. J Manag Care Pharm. 2004;10(2 Suppl):S2–7. [PubMed]
Harry B, Sturgis KM, Klinger JK. Mapping the process: An exemplar of process and challenge in grounded theory analysis. Educational Researcher. 2005;34(2):3.
Hernandez P. Trauma in war and political persecution: Expanding the concept. Am J Orthopsychiatry. 2002;72(1):16–25. [PubMed]
Hernandez P, Gangsei D, Engstrom D. Vicarious resilience: a new concept in work with those who survive trauma. Fam Process. 2007;46(2):229–241. [PubMed]
Jenkins JH, Cofresi N. The sociosomatic course of depression and trauma: A cultural analysis of suffering and resilience in the life of a Puerto Rican woman. Psychosom Med. 1998;60(4):439–447. [PubMed]
Lewis-Fernandez R. Cultural formulation of psychiatric diagnosis. Cult Med Psychiatry. 1996a;20(2):133–144. [PubMed]
Lewis-Fernandez R. Cultural formulation of psychiatric diagnosis. Case No. 02. Diagnosis and treatment of Nervios and Ataques in a female Puerto Rican migrant. Cult Med Psychiatry. 1996b;20(2):155–163. [PubMed]
Lewis-Fernandez R, Diaz N. The cultural formulation: a method for assessing cultural factors affecting the clinical encounter. Psychiatr Q. 2002;73(4):271–295. [PubMed]
Lim RF, editor. Clinical manual of cultural psychiatry. Washington, DC: American Psychiatric Publishing, Inc; 2006.
Lim RF, Lin KM. Cultural formulation of psychiatric diagnosis. Case no. 03. Psychosis following Qi-gong in a Chinese immigrant. Cult Med Psychiatry. 1996;20(3):369–378. [PubMed]
Loo CM. Race-related PTSD: the Asian American Vietnam veteran. J Trauma Stress. 1994;7(4):637–656. [PubMed]
Loo CM, Fairbank JA, Chemtob CM. Adverse race-related events as a risk factor for posttraumatic stress disorder in Asian American Vietnam veterans. J Nerv Ment Dis. 2005;193(7):455–463. [PubMed]
Loo CM, Fairbank JA, Scurfield RM, Ruch LO, King DW, Adams LJ, et al. Measuring exposure to racism: Development and validation of a Race-Related Stressor Scale (RRSS) for Asian American Vietnam veterans. Psychol Assess. 2001;13(4):503–520. [PubMed]
Loo CM, Lim BR, Koff G, Morton RK, Kiang PN. Ethnic-related stressors in the war zone: Case studies of Asian American Vietnam veterans. Mil Med. 2007;172(9):968–971. [PubMed]
Manson SM. The wounded spirit: A cultural formulation of post-traumatic stress disorder. Cult Med Psychiatry. 1996;20(4):489–498. [PubMed]
Marshall GN, Orlando M. Acculturation and peritraumatic dissociation in young adult Latino survivors of community violence. J Abnorm Psychol. 2002;111(1):166–174. [PubMed]
Marshall RD, Spitzer R, Liebowitz MR. Review and critique of the new DSM-IV diagnosis of acute stress disorder. Am J Psychiatry. 1999;156(11):1677–1685. [PubMed]
Mezzich JE. Cultural formulation and comprehensive diagnosis. Clinical and research perspectives. Psychiatr Clin North Am. 1995;18(3):649–657. [PubMed]
Miles MB, Huberman AM. Qualitative data analysis: An expanded sourcebook. 2. Thousand Oaks, CA, US: Sage Publications, Inc; 1994.
Munir KM, Beardslee WR. A developmental and psychobiologic framework for understanding the role of culture in child and adolescent psychiatry. Child Adolesc Psychiatr Clin N Am. 2001;10(4):667–677. [PubMed]
Norris FH, Alegria M. Mental health care for ethnic minority individuals and communities in the aftermath of disasters and mass violence. CNS Spectrum. 2005;10(2):132–140. [PubMed]
Pole N, Best SR, Weiss DS, Metzler T, Liberman AM, Fagan J, et al. Effects of gender and ethnicity on duty-related posttraumatic stress symptoms among urban police officers. J Nerv Ment Dis. 2001;189(7):442–448. [PubMed]
Ruskin PE, Reed S, Kumar R, Kling MA, Siegel E, Rosen M, et al. Reliability and acceptability of psychiatric diagnosis via telecommunication and audiovisual technology. Psychiatr Serv. 1998;49(8):1086–1088. [PubMed]
Schreiber S. Migration, traumatic bereavement and transcultural aspects of psychological healing: loss and grief of a refugee woman from Begameder county in Ethiopia. Br J Med Psychol. 1995;68(Pt 2):135–142. [PubMed]
Sokoloff NJ, Dupont I. Domestic violence at the intersections of race, class, and gender: Challenges and contributions to understanding violence against marginalized women in diverse communities. Violence Against Women. 2005;11(1):38–64. [PubMed]
Sorenson SB. Judgments about intimate partner violence: a statewide survey about immigrants. Public Health Rep. 2006;121(4):445–452. [PMC free article] [PubMed]
Spitzer RL, First MB, Wakefield JC. Saving PTSD from itself in DSM-V. J Anxiety Disord. 2007;21(2):233–241. [PubMed]
Suarez-Orozco C, Todorova IL, Louie J. Making up for lost time: the experience of separation and reunification among immigrant families. Fam Process. 2002;41(4):625–643. [PubMed]
Tamayo JM, Rovner J, Munoz R. The importance of detection and treatment of somatic symptoms in Latin American patients with major depression. Rev Bras Psiquiatr. 2007;29(2):182–187. [PubMed]
Ton-That N. Post-traumatic stress disorder in Asian refugees. Psychiatry Clin Neurosci. 1998;52(Suppl):S377–379. [PubMed]
Yilmaz AT, Weiss MG. Cultural formulation: Depression and back pain in a young male Turkish immigrant in Basel, Switzerland. Cult Med Psychiatry. 2000;24(2):259–272. [PubMed]
Zayas L, Cabassa L, Perez M. Capacity-to-consent in psychiatric research: Development and preliminary testing of a screening tool. Research on Social Work Practice. 2005;15(5):545–556.