Update on the Cultural Formulation Interview

The CFI field trials conducted for DSM-5 launched several ongoing research efforts. The field trial aimed to recruit 30 patients from each of several sites in the United States, Canada, the Netherlands, India, Kenya, and Peru and to apply a standard methodology to assess implementation (7). An initial, 14-item version of the CFI was tested; based on field trial results, it was then revised to produce the final 16-item version included in DSM-5.

In the United States, Aggarwal and his team (8) analyzed the content of 32 CFI interviews with patients and 32 debriefing interviews with clinicians at the New York site of the field trial. Common themes included how the CFI benefited medical communication by enhancing clinician-patient rapport, allowing the clinician to obtain new data in a relatively short period of time, eliciting the patients’ perspectives on what caused their symptoms, and collecting data that helped patients to become aware of their problems in more insightful ways.

Another U.S.-based study developed and pilot-tested a fidelity instrument to determine how reliably clinicians implemented the CFI and documented their perception of challenges during its use (9). The instrument was designed to capture adherence to method (i.e., whether clinicians adhered to the CFI topics) and clinical competence items (i.e., ratings of qualities such as empathy and patient centeredness). The CFI–Fidelity Instrument represents a first step toward quantifying fidelity in implementation of the CFI.

An international group of researchers led by Lewis-Fernández (10) reported data from the DSM-5 international field trials on feasibility, acceptability, and clinical utility of the CFI. A total of 75 clinicians from participating outpatient clinics received 2 hours of training prior to conducting an initial mental health evaluation of 318 patients at their clinics. Depending on the site, new or existing patients were enrolled, and each clinician assessed at least three patients over the course of the trial. Mixed-methods evaluation found the CFI to be feasible, acceptable, and useful. Clinicians’ initial concerns about feasibility were significantly reduced after the second time they used the CFI, suggesting a rapid learning curve. Results demonstrated the value of investing about 20 minutes of an initial evaluation on a cultural assessment in terms of possible effects on communication, engagement, diagnostic accuracy, and treatment participation. However, direct assessment of these potential effects is only now under way (11).

Bäärnhielm and colleagues (12) discussed experience with the CFI field trials in India, Kenya, and the Netherlands and reviewed work in Nordic countries with the OCF and CFI. The experience of the CFI field trial at the Indian site was noteworthy for how patients appreciated the extra time doctors devoted to them and to the cultural aspects of their problems. However, given the high volume of patients seen at the Indian sites (200–250 outpatients treated per day and 10,000 new patients per year at the New Delhi site alone) and the severe shortage of mental health professionals in India, the CFI may need to be shortened for routine use. Details of the CFI field trial in Kenya indicated that the CFI provided sufficient context, on at least one occasion, for the interviewer to reassess apparent psychosis as depressive disorder rather than schizophrenia. Experience with the CFI field trial at the Dutch site revealed that some respondents of Dutch origin appreciated the opportunity to discuss aspects of their cultural identity beyond simply being from the Netherlands, such as being a veteran or being born in the Dutch East Indies. Bäärnhielm and colleagues (12) emphasized the need to translate the CFI into other languages and to produce local training materials. They also noted that the CFI requires a flexible approach rather than simply repeating the standard questions to patients of diverse backgrounds. They suggested that initial CFI training needs to be augmented by careful support and follow-up to maintain what has been learned and to reinforce how to appropriately implement the interview. Modifications to the CFI itself, such as condensing it for use in busy clinical practices, are important issues and are being considered; ongoing research with the current version of the CFI is needed to establish a firm basis for amending it.

A team in Pune, India, administered semistructured debriefing interviews following audio-recorded CFIs with eight clinicians, 36 patients, and 12 relatives in a psychiatric clinic of a general hospital (13). The authors found that the CFI was acceptable to family members accompanying patients but that patients and clinicians rated the CFI lower when relatives were present. The role of families is important everywhere, but in India this is even more the case because family members often accompany patients to the doctor and expect to be included in the process of evaluation and treatment. However, the authors suggested that the presence of families might have limited how much patients could discuss family issues related to their mental health problems. In addition, relatives tended to raise additional aspects of the presenting problem, which tended to make the clinical assessment more complicated. Nevertheless, the study concluded that the CFI needed to be adapted to include families in the evaluation process. The study also raised important questions about respecting patient confidentiality and negotiating cultural norms in settings in which the focus of clinical attention is relatively more on families than on the individual.

Ramírez Stege and Yarris (14) translated the CFI into Spanish and used it routinely during outpatient follow-up in a regional psychiatric hospital in central Mexico. Participating clinicians did not receive training on how to use the CFI, beyond reading the written CFI guidelines in DSM-5. Furthermore, the interviews were regular follow-up appointments rather than intake evaluations. The authors found that the CFI was clinically useful for diagnosis and treatment planning, decreased mistrust during the session, and helped to elicit information about patients’ social networks and support. Another clinical benefit was the contribution of cultural identity to the person’s understanding of the illness and options of care that came from using the CFI; for example, information about a person’s age, gender, educational background, sexual orientation, religion, occupation, and language became part of the clinical discussion. However, patients and clinicians frequently misunderstood question 8, which is intended to elicit clinically relevant aspects of the patient’s cultural identity. Only two clinicians with social science training appreciated the intent of the question as written, which is to assess for broad aspects of cultural background and identity, not just national or ethnic origin. Clinicians without social science training tended to conflate “culture” with “indigeneity,” rejecting answers that did not relate to this narrow understanding of cultural identity. This misunderstanding demonstrates the need to train clinicians adequately before they use the CFI, including ensuring that local views of culture and identity are considered in framing the interview questions. More research is needed to address the impact of training and the need for local adaptation. More broadly, the CFI could eventually be revised to add assessments of specific social determinants of mental health; however, to ground changes in evidence, further research on the current version of the CFI is needed to establish which aspects are most important for clinical care.

In a study at a U.S. Department of Veterans Affairs clinic, Muralidharan and colleagues (15) conducted the core CFI with 14 patients with chronic psychosis, most of whom were male and African American, to assess the extent to which their diagnoses interfered with CFI implementation. They found that the CFI validated and deepened patients’ realization of their clinical experience and their process of recovery. The authors concluded that the CFI usefully enhances rapport and yields meaningful clinical information when used with individuals with chronic psychosis. However, this finding may not apply to individuals with acute psychoses.

Yet another research team found that the CFI elicited data that augmented culturally responsive care by bringing to attention what mattered most to Hispanic patients in an ethnic-focused clinic in New Haven, Connecticut: establishing relationships of trust with caregivers, addressing the stigma of mental illness, and paying attention to family- and church-related matters (16). The authors judged that some of this material would not have been uncovered without the CFI, which would have undermined the care of patients. Importantly, the New Haven experience suggested that use of the CFI not only sensitizes clinicians to the cultural issues of individual patients but can also contribute to the evaluation of clinical programs by highlighting problem areas and adapting the delivery of interventions to the cultural needs of patients and their families.

In addition to local cultural issues that complicate implementation of the CFI, general barriers to implementation have also been described. Aggarwal and colleagues (17) evaluated data from the New York site of the CFI field trial, which used the 14-item version of the CFI. Results were based on the responses of 32 patients and seven clinicians. The authors conducted a qualitative study evaluating barriers to CFI implementation based on debriefing interviews with patients and clinicians after administration of the CFI. For patients, the most common concerns were lack of differentiation from other interviews they had received, discomfort with discussing the past or talking about religion, and difficulty understanding some CFI questions. For clinicians, the most common concerns included lack of clarity about the relevance of the CFI to the presenting problem, diagnostic assessment, and intervention planning; doubts about using the full CFI at the beginning of an evaluation; repetitiveness; difficulty engaging patients with severe illness; and lack of clinician buy-in, which may mostly reflect lack of familiarity with the CFI. Patients with severe symptoms, including acute psychosis, suicidal behavior, and aggression, and those with acute disabilities, such as cognitive impairment, may not be able to complete the CFI (7).

The Cultural Consultation Service at the Jewish General Hospital in Montreal has been using a cultural formulation approach to evaluate patients from diverse backgrounds since 1999. This experience suggests strategies to address some of the challenges reported by practitioners implementing the CFI, especially when patients are recent migrants, acutely psychotic, suicidal, uncooperative, or otherwise difficult to engage. First, the quality of information obtained by the CFI and its ability to increase clinical rapport depend on good clinical communication. An adequate linguistic assessment is necessary for all migrant patients to determine the need for an interpreter. Even patients who strive to speak the mainstream language during the evaluation may have limited capacity to discuss intimate matters in their second (or third) language. Routinely offering the help of an interpreter, even to those without obvious language barriers, greatly increases the quantity and quality of the information obtained. An interview with an interpreter present can suggest diagnostic changes or can bring relief to patients who have been unable to communicate for prolonged periods. Patients who have been deemed mute, intellectually impaired, or unresponsive may brighten considerably and have much to say that bears directly on the diagnosis and treatment, such as details of the trauma they have experienced, family they hope to contact, and their history of migration.

Second, when facing communication difficulties, an educational assessment can help clarify the context of clinical findings. Patients with little formal schooling may have limited literacy and find it difficult to answer overly broad, abstract, or open-ended questions. An educational assessment need not be time consuming but should establish the number of years of schooling and whether the patient learned to read and write in the mother tongue and in the language(s) of the new country.

Third, although many patients appreciate the CFI’s focus on their personal experience, some patients may feel uncertain about the purpose of the interview. They may find some questions difficult because of shame, stigma or loss of face, or family honor. Careful explanation of the purpose of the evaluation and the fact that it is confidential, that the patient is in control of its duration and content, and that the aim is to clarify what is important for patient care is essential. The CFI may need to be used with caution in some kinds of evaluations, such as those conducted for forensic or insurance determination purposes, or in some delicate refugee evaluations, when the patient is forced to participate and is not in full control of the interview and its products. Use of the CFI in these situations will need to be carefully considered.

Fourth, for most patients, information from other family or community members is needed to clarify the social context and to assess family and community networks. The CFI has a key informant version that can be used with others in the patient’s entourage. This can identify sources of stress and resilience factors as well as caregiver issues that need to be addressed. Clinicians should check with managers in the organizations where they practice as well as local laws to ensure that collateral information from others is balanced with maintaining the patient’s confidentiality.

The core CFI is framed in a way that allows it to be applied to assessments in any clinical setting by mental health practitioners (18). Studies of the CFI in diverse training and clinical settings are starting to appear. For example, Alarcón and his team (19) discussed how the CFI improved communication in several clinical settings: the emergency department, consultation-liaison psychiatry, community health centers, and outpatient settings. Drawing from the CFI field trial data from Lima, Peru, the authors noted that the main positive outcomes had to do with improved clinical communication: patients felt listened to and better understood by their doctors; the CFI enabled honest discussions about prejudice among ethnic groups and about religious beliefs, such as bewitchment, as a cause of illness; and the CFI gave permission to patients to frankly express their views about illness and treatment. Other researchers reported how the CFI contributed to sensitively communicating a psychiatric diagnosis to a patient for whom mental health problems carried a heavy burden of stigma (20). These examples make it clear that the CFI can improve intercultural clinical communication. Improving how patients and clinicians speak to one another by bridging cultural divides is a fundamental building block of developing culturally appropriate services.

Other examples of the CFI in action include case studies of patients from specific origins, such as Ethiopians in Israel and immigrants to Italy from Morocco and Sierra Leone, and applications of the CFI with children and families. Clinicians from Israel evaluated two young Ethiopian women with apparent eating disorders (21). The CFI led the clinical team to appreciate cultural and familial meanings of stomachache in one case and led to active involvement of the mother in the second—both of these applications led to breakthroughs in the treatment, drew appropriate attention to potential predicaments of Ethiopians in Israeli society, brought to awareness clinical information that was previously neglected, and caused the treating team to question diagnoses of eating disorders given new cultural information. The authors considered that these factors facilitated recovery in these patients who had previously been difficult to help. Italian clinicians found that the CFI was helpful for clarifying the diagnosis, strengthening the therapeutic relationship, facilitating patient communication, and fostering adherence to treatment in their cases of immigrants to Italy from Morocco and Sierra Leone (22).

Although the CFI has a Supplementary Module for School-Age Children and Adolescents (SACA) (23), little research has been published on this module to date. In a case study, La Roche and Bloom (24) found that the components of the CFI were too dependent on adult communication capabilities to be used with children. In their research, the authors found that the CFI raised issues of mistrust of whites in the interview of a Somali-American child and enabled the development of a more nuanced treatment plan that incorporated religious practice and family involvement. However, the authors found that the CFI and SACA depended on verbal questions and cognitive capacity at the formal operations level and thus were not well suited to interviewing children younger than 11 years. They proposed a supplementary module for young children that would focus on drawing, building or sculpting, and acting through puppets or role playing as ways to access their perspectives, rather than relying on verbal questioning.

The role of families in cultural assessment is another area in need of further work. Investigators found that of the 321 patient interviews in the DSM-5 field trial, 86 at four of the 12 sites included family members, who generally found the CFI acceptable, although some found the questions too time consuming, intrusive, or personal (25). Of interest was the pattern of family involvement by site. Family members accompanied all patients in Kenya, whereas no relatives went to interviews in Canada and the United States, perhaps reflecting differences in family- versus individual-oriented care and support in these settings. Adapting the CFI for use with families may require changing the order of questions and remaining alert and responsive to diverse perspectives within the family.

Given the significant challenges working in cross-cultural mental health, training clinicians to use the CFI is of critical importance. Aggarwal and collaborators (26) used mixed methods to analyze field trial interviews with 75 clinicians from five continents about their training preferences for the CFI. Most of the clinicians preferred active behavioral simulations, such as role-play mock interviews. Respondents felt that mock interviews helped them to learn how to ask the CFI questions and resolve their doubts about the interview. Video presentations were deemed to be less helpful because they did not demonstrate how to work with uncooperative patients. The authors concluded that CFI training might best be accomplished through a combination of written guideline review, video demonstration, and behavioral simulations but that older, more experienced clinicians tended to prefer active rather than passive learning techniques. The New York State Center of Excellence for Cultural Competence developed an online training module (https://nyculturalcompetence.org/cfionlinemodule/) on how to use the CFI and reported its use by 423 service providers in the state, mostly social workers and other nonmedical counselors (27). Feedback was favorable, with an overall mean evaluation score of 4.13 (range 1–5, with 1=strongly disagree and 5=strongly agree that the module was helpful in various ways) and with most participants expressing the belief that the module would change their clinical practice.

In other work, researchers developed a novel curriculum for training psychiatry residents to be culturally sensitive (28). Four 90-minute training sessions were delivered in the second residency year, with CFI-related segments inserted into the second and fourth sessions. Generally, the residents appreciated the opportunity to practice the CFI questions, and the investigators found that the CFI questions acted as scaffolds for the residents to build better understanding of health disparities, biases in clinical work, and culture-related attitudes that may affect patient outcomes. These educational achievements resulted from residents practicing how to deliver the CFI questions in light of information they had previously received about disparities, bias, and other culture-related issues.

Yet other research found that 1-hour education sessions on the CFI improved resident cultural competence scores throughout six residency programs in the United States and Canada (29). Importantly, the amount of previous training in cultural competence did not influence the outcome. In another study, psychiatry residents who were given a 1-hour training session on the CFI showed significant pre- and posttest changes in overall scores on an adapted version of the Cultural Competence Assessment Tool, as reflected on the nonverbal communications and cultural knowledge subscales (30). These findings suggested that residents receiving even brief training on the CFI improve their understanding of how members of various ethnic groups differ in their use of space and physical contact as well as in health beliefs and practices and use of health services.

Another team described the use of small-group objective structured clinical examinations for teaching use of the CFI and found that this approach enhanced trainees’ comfort in using the CFI, improved their knowledge of cultural syndromes, and refined their diagnostic skills by helping them differentiate between acceptable religious practices and religious behaviors associated with psychosis (31).

An important question concerns whether and how training with the CFI can contribute to broader cultural competence. Lim and colleagues (32) discussed the use of the CFI in training medical students and psychiatry residents and in providing continuing medical education. The authors found that medical students might benefit from learning to conduct a CFI-enhanced medical interview that embeds core CFI questions into standard medical interviews. For psychiatry residents, the CFI can be used to foster the development of skills in clinical interviewing and patient-centered care. The CFI–Informant Version and supplementary modules, which could be introduced in core training sessions or in elective cultural psychiatry rotations, can provide residents with tools to elicit cultural material throughout their clinical practice. The CFI focus on individual experience in context may counteract the tendency to present cultural information in terms of stereotypes and may help students develop a systemic perspective on patients’ circumstances. Ideally, the CFI and related materials should be introduced early in residency training, and trainees should be encouraged to adapt the CFI according to clinical needs in different settings throughout their training. Developing a flexible, culturally informed approach to assessment is an important component of cultural competence and person-centered medical practice. University-based training on the CFI is just the beginning. Future efforts should also address the educational needs of practicing clinicians.

The CFI was designed to provide a simple and systematic way to collect the kind of information listed in the OCF and thus to provide a basis for clinically relevant case formulation. Other approaches have been developed to assess key dimensions and domains of illness experience (33). A Dutch team piloted the Brief Cultural Interview with the aim of reducing the time required to conduct a culturally competent evaluation while incorporating the basic components of the OCF (34). The CFI may not always stand alone but may be used as one component of a broader approach to cultural formulation. Multicultural assessment (35), an approach developed specifically for clinical psychologists, includes the CFI along with other instruments and procedures.

Some researchers have called for caution and warned of the risk of stereotyping and simplifying cultural material when using the CFI (36). They proposed the use of interdisciplinary case discussions (ICDs) as a way to guard against cultural formulations that draw overly general conclusions from idiosyncratic cultural and personal data. More specifically, ICDs broadened the scope of clinical data collection, eschewed simplistic assignments of cultural identity, and brought to the discussion structural issues that may perpetuate stigma, inequity, and “othering” of immigrants and refugees. ICDs represent a check on possible misuse of the CFI by those who lack experience or training. However, conducting an ICD requires staff members with high levels of cultural and clinical experience, as well as the time to hold an intensive case discussion. The number of staff and clinical hours needed for such an approach may not be feasible or acceptable in most practice contexts. This suggests the need for studies that test various components of a cultural formulation assessment to balance comprehensive information gathering with the aim of making cultural evaluation systematic yet practical in clinical practice.

Since the publication of the CFI as part of DSM-5 in 2013, the literature on its use in training and clinical work has grown. The instrument is widely available and relatively easy to use. Most investigators report favorable evaluations of its clinical utility, feasibility, and acceptability to patients and clinicians alike. Training is straightforward, and even one hour of training conveys measurable benefits to students. The time required for its completion shrinks to about 20 minutes after one administration. However, studies on the effectiveness of the CFI to improve clinical outcomes are lacking, and this paucity of data points to an urgent research priority. To date, most studies of the CFI have been authored by a small group of researchers and academics, mostly from North America, who were involved in the DSM-5 field trials.

Research on the CFI is needed to guide its further refinement and implementation in diverse settings. Given that the current version of the CFI needs further testing and evaluation, it may be more efficient at this time to adapt the CFI guidelines to address specific issues while retaining core elements. Areas requiring attention include translation of the CFI to other languages; cultural adaptation to nonacademic settings worldwide; and best training practices for optimal clinical implementation, including its use with interpreters, culture brokers, and allied health professionals.

Growing attention to structural competency (37) raises questions about the extent to which the CFI sufficiently addresses this important aspect of culturally competent assessment and care. Structural competency refers to focusing attention not just on the local cultures of the patient and clinician but also on the social structures, institutions, inequities, dynamics of power, and exclusion experienced in the societies in which we live and work (38). This includes the historical legacies of oppression and racism that persist as part of the social fabric of everyday life as well as within professional education and clinical practice. Addressing the health needs of marginalized people is crucial for health equity. Responding to issues of structural violence and marginalization is essential to achieve health equity (39). Research should examine how best to ensure that the CFI elicits these aspects of patients’ lives and experience.

Use of the CFI is an important step toward person-centered care. Future research may suggest additional refinements, but in its current form it provides a simple way to begin the process of cultural assessment, and its systematic use can foster a reflective stance and promote systemic thinking about the patient’s life world, which otherwise may be hard to establish in routine clinical practice. Widespread use of the CFI can contribute to more respectful clinical interaction and more inclusive care for all—in psychiatry and the mental health professions, as well as in general health care and social services.