Mon, 22 Nov 2021 in Refuge: Canada's Journal on Refugees
Training and Integrating Public Service Interpreters in a Refugee Health Clinic: A Mixed-Method Approach to Evaluate an Innovative Program
Abstract
Language barriers can harm refugees’ health, and trained interpreters are a solution to overcome these barriers in all health consultations. This study trained interpreters and integrated them in a refugee clinic. Nepali-speaking migrants were recruited and underwent 50 hours of training to serve as interpreters for recently arrived Bhutanese refugees in Quebec City. To evaluate the project, mixed data were collected. At baseline and follow-up, patients’ health (as perceived by practitioners) and satisfaction were evaluated. Interpreters and practitioners were also interviewed and took part in joint discussion workshops. Patients’ health remained stable but, interestingly, patients were slightly less satisfied at follow-up. Practitioners and interpreters described both benefits and difficulties of the program. For example, integrating interpreters within the clinical team allowed for better collaboration and mutual knowledge of cultures. Challenges included work overload, conflicts between interpreters and practitioners, and role conflicts for interpreters. Overall, the full-time integration of trained interpreters in the clinic facilitated communication and case administration. This practice could be especially beneficial for refugee clients. In future interventions, interpreter roles should be better clarified to patients and practitioners, and particular attention should be paid to selection criteria for interpreters.
Resumé
Les barrières linguistiques peuvent avoir un impact négatif sur la santé des réfugiés et le recours à des interprètes qualifiés est un moyen de surmonter ces barrières lors de toute forme de consultation médicale. Dans le cadre de cette étude, des interprètes ont été formés et intégrés à une clinique pour réfugiés. Des migrants de langue népalaise ont été recrutés et ont suivi cinquante heures de formation afin d’agir comme interprètes auprès de réfugiés bhoutanais récemment arrivés dans la ville de Québec. À des fins d’évaluation du projet, des données mixtes ont été recueillies. Lors de l’étude initiale et de l’étude de suivi, la santé des patient.es (telle que perçue par les praticien.nes) et leur satisfaction ont été évaluées. Les interprètes et les praticien.nes ont aussi pris part à des entretiens et des ateliers de discussion. La santé des patient.es est demeurée stable, mais, curieusement, les patient.es étaient légèrement moins satisfait.es lors de l’étude de suivi. Les praticien.nes et les interprètes ont décrit tant les bénéfices que les difficultés liés au programme. Par exemple, l’intégration des interprètes au sein de l’équipe clinique a permis une meilleure collaboration et une plus grande connaissance mutuelle des cultures. Les difficultés rencontrées incluent une surcharge de travail, des conflits entre les interprètes et les praticien.nes ainsi que des conflits de rôle pour les interprètes. De façon générale, l’intégration à temps plein d’interprètes qualifiés au sein de la clinique a facilité la communication et la gestion des cas. Cette pratique pourrait être particulièrement bénéfique pour une clientèle de réfugié.es. Dans les interventions futures, le rôle des interprètes devrait être mieux défini auprès des patient.es et des praticien.nes et une attention particulière devrait être portée aux critères de sélection des interprètes.
Main Text
Introduction
Migrants’ health in Canada tends to be better than that of the general Canadian population upon their arrival to the country (Gushulak, Pottie, Roberts, Torres, & DesMeules, 2011). However, over time migrant health declines and eventually reaches a level similar to that of the Canadian population; this is known as the “healthy immigrant effect” (Newbold, 2005). This decline can be explained by many psychosocial factors, such as a poor understanding of the health-care system, institutional discrimination, and acculturation stress (Gkiouleka & Huijts, 2020; Khanlou, 2010). Refugees are a particularly vulnerable subgroup of migrants, and their health tends to decline faster than that of any other migrant group (Newbold, 2009; Reed & Barbosa, 2017).
Language barriers have also been identified as an important factor that could help to explain this health decline, as they engender many consequences for access to health care, understanding of care received, and treatment adherence or quality of care (Jacobs, Chen, Karliner, Agger-Gupta, & Mutha, 2006). Indeed, the lower the language proficiency in one of the host country’s official languages, the faster health declines (Pottie, Ng, Spitzer, Mohammed, & Glazier, 2008).
In order to alleviate the negative consequences ensuing from language barriers, benefits of collaborating with a trained interpreter have been clearly underlined in medical and psychosocial contexts of care (Brisset, Leanza, & Laforest, 2013; Gartley & Due, 2017; Karliner, Jacobs, Chen, & Mutha, 2007; Miller, Martell, Pazdirek, Caruth, & Lopez, 2005). Although trained interpreters are the gold standard in health care, the lack of interpretation services or the use of ad hoc interpreters (i.e., untrained interpreters) continues to be the norm (MacFarlane et al., 2009; Newbold, Cho, & McKeary, 2013; Yelland et al., 2016). Because they lack training, ad hoc interpreters are more at risk of jeopardizing consultations because of errors in translation, poor knowledge of medical terms, and role reversal issues in families (MacFarlane et al., 2009).
Barriers such as unawareness of interpreting resources, concerns about costs of services, complicated administrative procedures, and lack of training on working with interpreters hinder the use of trained interpreters (Brisset et al., 2014; MacFarlane et al., 2009). Barriers to access aside, training for community interpreters remains scarce and optional in Canada (Leanza, Brisset, Rocque, & Boilard, 2017). Therefore, actual access to trained interpreters is limited.
In the province of Quebec (where the present study took place), the Ministry of Health supports regional interpreter banks, which are established and managed by community organizations (except for the Montreal and City of Quebec areas, where the banks are part of the public health-care network). Interpreters enrolled in these banks are supposed to have some training, but in reality, it is limited or non-existent. This situation is not particular to the province of Quebec. It is observed in other Canadian provinces and other countries (Leanza et al., 2017; Vanderwielen et al., 2014).
Moreover, not only is this training scarce and unstandardized; community interpreters are also rarely integrated into the clinical team. This lack of integration has documented consequences, such as impeding continuity of care and compromising the administration and management of a patient’s case (Cheng, Vasi, Wahidi, & Russell, 2015). Continuity of care is perceived as increasingly important when treating certain types of patients, such as refugees, where developing a trusting relationship is especially crucial (Karliner & Mutha, 2010). Indeed, working with the same interpreter throughout patient follow-up is a frequent recommendation, and it has been urged that interpreters be integrated in interprofessional teams in order to enhance quality of care (Attard et al., 2015; Dubus, 2016).
Study Context
The present study is a collaborative action-research initiative between researchers from the Psychology and Cultures Lab at Université Laval, health practitioners from the Clinique de santé des réfugiés (Quebec Refugee Health Clinic), and community organizers from the Centre multi-ethnique de Québec (Quebec Multi-ethnic Centre). The Quebec Multi-ethnic Centre is an independent community organization, established in 1960, whose mission is to welcome migrants of any status to facilitate their establishment, to support their adaptation and integration into Quebec society, and to promote their access to better socio-economic conditions. The clinic, established in 2007, stems from the Ministry of Health’s program initiative on the accessibility of services to ethnocultural communities. Psychosocial and medical evaluations and care are offered to all refugees and refugee claimants newly arrived in the city.
At the time of the study, in the city of Quebec, interpreters were available to public institutions (e.g., health clinics, courts of law, etc.) through a community organization called the Interpreters Bank. However, health practitioners at the clinic reported serious and hindering difficulties with the Interpreters Bank in managing refugee patients’ follow-up, such as the inability to have the same interpreter every time and inadequate or lack of training. Therefore, in collaboration with these health practitioners and community organizers, an action-research project (Catroux, 2002) was established to address their needs and facilitate refugee patients’ follow-up at the clinic. This project focused on the Bhutanese community, since at the time of the study, Bhutanese refugees comprised the largest group of refugees who were expected to arrive in the province.
Study Objectives
The objectives of this study were twofold (see Figure 1). The first aim (action component) was to train interpreters and integrate them in the clinic on a full-time basis over a one-year period. A 50-hour training program for community interpreters and two training workshops for practitioners were developed by the research team (including researchers, health professionals from the clinic, and community organizers). To the best of our knowledge, no such training initiative had previously been undertaken in Quebec, or in other Canadian provinces. Interpreters’ integration into the clinical team (e.g., work schedule, finding a workspace for the interpreters in the clinic, joint workshops, etc.) was organized by the Quebec Multi-ethnic Centre and the clinic.
The second aim (research component) was to evaluate the effects of integrating trained interpreters in the clinic. Based on the literature, our assumption was that introducing trained interpreters into the clinic would enhance the quality of health care and of the relationship between practitioners and interpreters. The specific research aims were (1) to explore practitioners’ and interpreters’ points of view about their collaboration and the project, (2) to observe and measure the evolution of the relationship dynamic between practitioners and interpreters, (3) to assess patients’ health status evolution during the project, and (4) to measure patients’ satisfaction at the beginning and the end of the program. To address these aims, a mixed-method convergent parallel design was used (Creswell & Plano Clark2011). This type of design “occurs when the researcher collects and analyzes both quantitative and qualitative data during the same phase of the research process and then merges the two sets of results into an overall interpretation” (p. 77). Mixed methods gather a variety of data (quantitative and qualitative) and mix these data to capture the complex and multifaceted impacts of the interpreters’ integration into the clinic. Reporting on a mixed-method study was complex, so we followed Creswell and Plano Clark’s (2011) guidelines for mixed-methods research articles.
Methods
Training Program
Interpreters underwent 50 hours of training. The initial training, provided before integrating interpreters into the clinic, covered four major themes: interpreting techniques, terminology used in health care, knowledge of the Quebec health-care system, and issues ensuing from a three-way relationship (communicational, relational, and ethical issues, and interpreters’ roles). Training sessions were provided by members of the research team with experience in academic teaching in health sciences, an experienced university-trained interpreter, and a specialist of the Bhutanese community. Continuing training, during the integration of interpreters into the clinic, included six joint workshops with the practitioners and interpreters, held approximately every six weeks, with the third author acting as facilitator. Themes such as interpreters’ roles, home visits, and nutrition were discussed on the basis of real cases, in order to provide the opportunity for practitioners and interpreters to learn together. Interpreters were also offered group supervision sessions (six one-hour sessions) with a psychologist trained in intercultural interventions. Interpreters were invited to discuss their experiences during these sessions to further their skill development. The research team also arranged two interventions (45-minute lunch-time meetings) with the practitioners (before and during the integration of interpreters into the clinic), to provide an overview of the research in the field of interpreting and basic instructions on how to work with interpreters.
Participants
Refugee patients from Bhutan were selected for the current study, since Bhutanese refugees comprised the largest group of refugees who were expected to arrive in Quebec during the project. A total of 151 (T1) and 143 (T2) refugee patients took part in the study. Patients’ mean age was 34 (range 14–90, SD = 10.39), and 55% of participants were female. Ethics approval was obtained from research ethics committees from regional Health and Social Services Centres (Centres de santé et de services sociaux de la Vieille-Capitale, de Québec-Nord, de Portneuf), from the regional Public Health Office (Direction de Santé Publique de la Capitale-Nationale) and from the Jeffery Hale–St. Brigid’s health institution.
The Quebec Multi-ethnic Centre recruited seven Nepali-speaking migrants as interpreters on the basis of several criteria: interpreter’s maturity (e.g., sense of responsibility, previous professional experiences), previous experience in care (e.g., with children, adults, older adults), and the absence of apparent conflicts (e.g., reputation in the community). Mean age was 29 (range 21–44, SD = 8.10) and two were female. Only one had previous experience in community interpreting. In the first few weeks of the project, four interpreters dropped out of the program as the result of difficulties reconciling work schedules, mandatory French-language courses, and poor French-language proficiency. One interpreter was also evicted from the program halfway through the project in response to inappropriate behaviour at the clinic.
Nine health practitioners (the entire staff of the clinic) took part in the study (4 physicians, 2 nurses, 2 social workers, and 1 nutritionist), as well as the administrative agent of the clinic. Mean age was 39 (range 25–54, SD = 10.06) and 8 were female.
Data Gathering and Analysis
Different procedures (see Figure 2 for a summary) and analyses were used for the research part of the project. They are presented by specific research aim.
Practitioners and Interpreters’ Perspectives
Semi-structured individual interviews were conducted in two phases: at the beginning of the project with the initially recruited interpreters (n = 7), and at the end, with the remaining interpreters (n = 2), health practitioners (n = 9), and administrative agent (n = 1). An intermediate interview was also held with the interpreter who was evicted from the program, after the eviction. Pre-identified themes were explored during the interviews relating to participants’ general experiences with the program, representations of interpreter and practitioner roles, benefits and difficulties of the program, the interpreter–practitioner–patient relationship, as well as communicational and cultural challenges.
An inductive thematic analysis (Paillé & Mucchielli, 2008) was conducted in order to identify emerging themes in participants’ experiences with the program. To better understand the essence of interpreters’ and practitioners’ experiences, transcripts were analyzed separately for each group of actors, and different themes emerged in their experiences. In an inter-rater and iterative process, the first two authors and a research assistant coded the same interviews independently and met throughout the analyses to compare coding and resolve disagreements. Excerpts of interviews were translated from French to English.
Evolution of the Relationship between Practitioners and Interpreters
In addition to interviewing practitioners and interpreters, observational data during joint workshops between practitioners and interpreters were also collected to better understand how their relationships were built and evolved. All six workshops were audio recorded and transcribed verbatim. A conversation analysis was conducted to explore if, and how, interpreters were integrated in these discussions. Conversation analysis is based on “the atomization of actions, that is, their breakdown into small units and then their re-composition” (Vincent, 2001, p. 180, free translation) to obtain an overall picture of the integration of interpreters (Vincent, 2001). We consider the conversations that took place in the workshops to be social interactions “with risks and issues that can be decoded by conversational analysis” (p. 179). These risks and issues indicate the relationships between interpreters and practitioners that take place daily in the clinic. As recommended byHave (2007), common organizers of the conversations, such as sequences, turn takings, interruptions, and repairs (i.e., organized ways of dealing with interaction failures) were coded and described. These organizers allow to break down the conversation into smaller descriptive components. The presence of humour and negative statements (e.g., microaggressions) were also looked for as markers of the quality of relationships. Each workshop was analyzed independently; comparisons between workshops were then performed. As an objective measure of interpreters’ inclusion in the conversation, workshop facilitator, interpreter, and practitioner speaking times were recorded, and individual regression analysis with SPSS 18 was conducted to determine the evolution (or not) of participants’ speaking time.
Patients’ Health Status (as Perceived by Practitioners)
In order to preserve patient anonymity, the research team could not access patients’ files, and therefore practitioners’ perceptions about patients’ health were used as a proxy measure. Practitioners filled out a brief questionnaire at the end of each consultation, specifying whether they evaluated patients’ health as stable, better, or worse than the previous consultation. Descriptive statistics and frequency analyses were run using this information.
Patients’ Satisfaction
To evaluate patients’ satisfaction, patients were asked to complete a questionnaire at the beginning and at the end of the implementation of the program. A short version of the WHO scale for service satisfaction, as validated by St-Onge, Mercier, Perrault, and Harnois (1997), was used. It relies on a 5-point Likert scale (from 1: completely dissatisfied to 5: very satisfied). Items were translated into Nepali by the most experienced interpreter of the program and a Nepalese academic. Basic socio-demographic information was also obtained (age and gender). Questionnaires were completed with the help of the program’s interpreters. Quantitative analyses were conducted with SPSS 18, such as descriptive statistics and mean comparisons (with t-tests). Size effects (partial η2) were also calculated in order to grasp scientific significance as recommended byChampely and Verdot (2007). According toCohen (1988), a η2 greater than or equal to .01 refers to a small size effect, a η2 greater than or equal to .06 to a moderate size effect, and η2 greater than or equal to .14 to a large size effect.
Results
Practitioners’ Experiences during the Program
Practitioners raised benefits and difficulties ensuing from the project, and these were organized according to four themes relating to (1) the interpreter’s full-time presence at the clinic, (2) continuity of care, (3) selection of interpreters, and (4) perceptions about the pertinence and efficacy of the project. Benefits and difficulties are reported according to these four themes.
Benefits of the full-time integration of trained interpreters in the clinic included an increase in stability in their work due to a facilitated access to the same interpreters and an increase in ability to communicate with patients. Practitioners also appreciated having the opportunity to communicate with interpreters at all times, as this allowed them to access cultural information, debrief the interpreter before and after a consultation, and ensure a more efficient management of the clinic and better patient follow-up.
There were situations where we had to brief the interpreter before. The fact that he’s already here before the consultation, and that he knows how we work, that he knows us as practitioners, sometimes, it makes it easier.
(Practitioner 1)
Moreover, continuity of the relationship and regular contact with interpreters was much appreciated, since it facilitated collaboration with interpreters (e.g., allowed for health practitioners and interpreters to understand each other’s working habits) and was thought to have positive impacts on care.
I think that working often with the same person, it helps a lot as well. Not only can that person get to know the way I work, but I also get to know the way she works.
(Practitioner 1)
Practitioners also appreciated the better supervision of interpreters this program offered and the possibility to resolve issues more efficiently, and thus ameliorate collaboration. For instance, when an issue arose, there was an immediate follow-up with the interpreter, contrary to no follow-up when working with the Interpreter Bank. Practitioners also appreciated the discussion workshops, which allowed them to access interpreters’ opinions on different issues. In addition, having at least one female and one male interpreter was perceived very positively, as it provided sensitivity to patient preferences for the gender of the interpreter (e.g., female patients preferring a female interpreter for a gynecology consultation). Finally, practitioners wished the program could be extended to other linguistic groups at the clinic, since the project showed them how collaboration with the program’s interpreters was an improvement over their experience with the Interpreter Bank.
In spite of these benefits, the full-time presence of interpreters at the clinic also brought difficulties. For instance, confusion about interpreter roles emerged, as well as challenges in relationships with interpreters (i.e., the need to reframe the relationship with interpreters to preserve a strictly professional relationship). Moreover, implementation of the program with the full-time presence of an interpreter led to a work overload for health-care professionals. For instance, knowing that an interpreter was always present and able to help them, patients started showing up without appointments, and practitioners felt trapped by the situation, as ethically they could not refuse to treat their patients when required. This also resulted in cumbersome administrative procedures for practitioners (i.e., more paperwork). Finally, practitioners emphasized that it would be important, in the future, to select interpreters who do not belong to the same cultural community as patients, since there were problems related to caste systems.
Interpreters’ Experiences during the Program
Although interpreters also discussed positive and negative experiences, their experiences were regrouped in four themes that were different from those describing practitioners’ experiences: (1) representations of their role, (2) motivations to take part in the project, (3) relational issues, and (4) distress and lack of support.
When asked about the qualities of a good interpreter, most frequent answers consisted of being professional, respectful, and translating accurately. However, a few interpreters also described their role as cultural brokers who would provide cultural information to the health practitioner when judged necessary.
We have to explain our culture to the doctor if the doctor doesn’t understand what the client is saying.
(Interpreter 1)
Motivations to take part in the program included the possibility of helping others in their community, facilitating communication for Bhutanese patients, ameliorating their language skills in French, and acquiring knowledge about health in general, and the health-care system in Quebec. At the end of the program, interpreters reported feeling happy and fulfilled for having had the opportunity to engage in this role.
Positive experiences of collaboration with health practitioners involved working with a calm and patient practitioner who was respectful and open to understanding the patient’s culture. Interpreters also appreciated practitioners who were open to repeating, adapting their language (e.g., vulgarising when necessary), and using short sentences. Professionalism, respect, and the perception that practitioners viewed consultation as teamwork were highly valued and led to trusting relationships with practitioners.
Negative collaborations occurred when interpreters felt practitioners were impatient and disrespectful towards them. Interpreters also faced challenges such as having difficulty translating accurately when the physician used long sentences (i.e., forgetting all there was to translate), or spoke too fast, or used certain French words, especially French Quebecker expressions, that were difficult to understand.
Difficult collaboration … [occurs when] the client asks the doctor to repeat a few times, and if he asks because he didn’t understand a word, at that time, the doctor would get mad. “Why didn’t you understand? You must understand.”
(Interpreter 2)
Interpreters also experienced relational difficulties and challenges about their roles. For instance, patients seemed to believe that interpreters worked for them, so it was challenging to keep a professional distance with patients. Furthermore, patients tended to attribute a lot of power and prestige to the interpreter, and thus became exigent with interpreters. For instance, patients showed up with questions and needs (e.g., concerning not only health, but also administrative procedures for immigration, issues with bills, etc.) and expected interpreters to help out with these issues.
The client isn’t happy with me. “Why don’t you want to listen to me? You work for me.” The client doesn’t understand the difference between the interpreter and the practitioner. We really have to explain what it means to be a practitioner and what it means to be an interpreter. We have to clearly say that my role is only to translate. “If you have questions, it’s not for me, it’s for the practitioner.”
(Interpreter 3)
On health issues more specifically, patients arrived at the clinic and expected the interpreter to arrange for them to be seen, even without an appointment. Consequently, health professionals blamed the interpreter for the work overload.
As for relational difficulties, interpreters felt that health practitioners did not always trust them and saw them as siding with the patient. Consequently, interpreters felt that they were blamed when something went wrong (e.g. when patients didn’t show up on time or didn’t take their medicine correctly). In these difficult times, interpreters felt alone and misunderstood, and wished there had been someone to turn to for advice.
Interpreters also experienced distress after difficult consultations and did not have appropriate support. They needed someone available to meet with quickly on site. Finally, interpreters recounted having to navigate instances of discrimination, such as being careful with the translation they offered patients in order to avoid transmitting the potentially discriminating sentences to the patient.
Evolution of Interpreter–Practitioner Relationship (during Workshops)
Group discussion analyses showed that while some aspects of conversation organizers evolved throughout the project, others remain unchanged. Humour began to appear in the discussions and interpreters were included, while there was no change in interruptions, the need for repairs, and negative statements.
The first workshops (workshops 2–4) were organized mainly around the facilitator’s questions, addressed to all participants, but usually answered by practitioners, who then engaged in long interactions among themselves. Interpreters only spoke when they were questioned directly. They did ask questions, but only to confirm that what they did in consultations was appropriate or that they correctly understood a particular word or sentence. Only twice (in workshops 3 and 4) did an interpreter bring up an example (about the practice or an aspect of Bhutanese culture), which led to a larger group discussion that included interpreters. One interpreter was frequently interrupted when speaking, without possibilities of repairs, as shown in extract 1 (workshop 3). Interrupted turns are in bold:
Facilitator: What is a good diet for a Nepalese? When a Nepalese says, “I eat well,” what does it mean?
Practitioner 1: He eats rice (laughter).
Practitioner 2: He has enough to eat. Whatever he eats, he has enough to eat.
Interpreter 1: There is a lot of…
Facilitator: Lots of rice and bread?
Interpreter 1: There are many different kinds of diets. Many types of foods for the Nepalese. The foods that we eat often are rice with meat and lentils. But occasionally for celebrations, we cook differently.
Facilitator: Yes, like what, for example?
Interpreter 1: Like “momo”…
Practitioner 3: Oh yes, “momo” is really good.
Interpreter 1: Different kinds of…
Practitioner 3: It’s like dough with meat inside, I think, eh?
The third workshop, on nutrition, also contained many negative statements about the Bhutanese, such as, “They only eat rice,” “They don’t know what vegetables are,” “They cannot read,” “They didn’t receive any education on family economy,” etc. Such statements were also found in other workshops (workshops 2 to 4).
The facilitator-driven Q-A pattern was still present in conversation organization in the fifth workshop but had almost disappeared in the sixth. In these two workshops, practitioners also asked direct questions to interpreters, whose answers opened up other sequences of discussion, including interpreters. Humour and laughs were also more present and shared by all participants in the last workshops, as they did not target Bhutanese habits (such as eating rice).
While some aspects evolved during the group workshops, there were negative statements still made about the Bhutanese, and sometimes about Quebeckers, during the last workshops, as were interruptions, without repairs, of interpreters.
Figure 3 depicts the trend of interpreters gradually taking (or being given) more speaking time during group discussions. A similar trend was noticeable for practitioners, as their speaking time increased over time and remained stable during the last two workshops. In contrast, facilitator speaking decreased over time. However, individual regression coefficients for all roles were not significant.
Practitioner Perceptions of Patients’ Health
On average, patients had 2.25 consultations during the program (SD = 1.59; range 1 to 9). Frequency analyses showed that practitioners perceived that health remained stable for a majority (77.4%) of patients during the program, while 16.9% showed a perceived amelioration and 5.6% experienced a perceived deterioration. No differences in practitioners’ perceptions of patient’s health evolution were found according to patient gender (χ²(2) = 4.88; ns) and age (χ²(10) = 9.25; ns).
Patient Satisfaction with Care
T-test analyses showed that overall patient satisfaction with care decreased significantly at the end of the program (mT1 = 4.36 (SD = .34); mT2 = 4.10 (SD = .49); t(141) = 5.90; p < .001). Mean comparison and size effect were computed for each item. Table 1 presents these results organized according to size effect (from large to no effect).
There was no difference between men and women at T1 (t(148) = 1.67; ns) and T2 (t(141) = 1.12; ns), and both genders were significantly less satisfied in T2 (see Table 2), and size effect was large.
There were no differences in satisfaction according to age at T1 (F(144,5) = 1.36; ns). A difference appeared at T2 (F(137,5) = 4.08; p < .01), where the youngest patients (14–19 years of age) were less satisfied than the eldest (50 and over). While satisfaction mean scores for patients from 30 to 39 years, and from 50 and over showed no significant difference or a marginal one, from T1 to T2, size effect was small for the 30–39 and 50–59 ranges, and large for the 60 and over range. The other age ranges showed a significant decrease and a large size effect (see Table 3).
Discussion
The aim of this collaborative action-research project was training and integrating interpreters in a refugee health clinic, and evaluating the effects of such training and integration. Results will be discussed in the order of the research-specific aims: (1) practitioners’ and interpreters’ perspectives, (2) evolution of the relationship dynamic, (3) patients’ health, and (4) patients’ satisfaction.
Many positive repercussions were raised by practitioners and interpreters. Both groups appreciated the possibility to have time to discuss before and after seeing patients. Indeed, a study conducted in the United States showed that it is difficult for interpreters to work with refugee patients (who may have a difficult past), and it was recommended to take the time for a pre-briefing before the session (Dubus, 2016), which is exactly what the training program suggested. According to practitioners, it ensures quality in the follow-up and allows access to cultural information, which, in turn, has a positive impact on patients.
Interpreters appreciated when practitioners viewed consultations as teamwork and adapted their language to ease the interpretation. But at the same time, interpreters felt blamed for patients’ inappropriate behaviours, like non-compliance with treatment, failure to appear, or work overload due to spontaneous patient visits. Interpreters reported being in a double bind: on the one hand, they could be mistrusted and reproached by practitioners, and on the other hand, patients could be very demanding (e.g., appearing at the clinic without appointments) and sometimes directly addressed the interpreters with wide-ranging questions (not always about health or health care). In such situations, whatever they chose to do, the result always disappointed one party. This challenge echoes findings in other studies, which showed that although interpreters play a key role in the development of a therapeutic alliance, challenges arise due to role conflict and practitioners’ expectations (Brisset et al., 2013; Gartley et al., 2017; Krupic, Hellström, Biscevic, Sadic, & Fatahi, 2016).
The increase in interpreters’ speaking time in joint workshops could be an indicator of the interpreters progressively taking their place in the clinical team. However, an important question remains: Even if they speak more frequently, are they heard more? The workshops’ conversation organization showed a progressive change as interpreters began to ask questions and introduce new content in the last two workshops, thus transforming the passive facilitator-driven Q-A pattern into a more proactive one. Nonetheless, interpreters’ turn interruptions were present in all workshops, and prejudice about Bhutanese, in the form of negative comments about their “strange” or “inappropriate” (cultural) habits, were also present all along. This finding suggests the need for awareness and cultural humility training for practitioners (Tervalon & Murray-García, 1998), as they do not seem aware of their prejudice and microaggressions. This asymmetrical relationship between practitioners and interpreters, which tended to be perpetuated throughout the workshops, is the mark of a power imbalance that is difficult to transform. Research suggests that interpreters tend to remain with little capacity to act or to exert influence in the context of health care, despite having recognized professional training and competencies (Leanza, Angele, Cotret, Bouznah, & Larchanché, 2020).
This second set of results, along with ambivalent results from the interviews, mitigates the impression of a complete integration of interpreters in the team. But collaboration did improve over time, as practitioners expressed many positive effects and even asked for the expansion of such programs to include other languages. However, their trust in interpreters seemed to remain moderate. This is probably due to some characteristics of the selected interpreters. At the beginning of the program, most of them had low French proficiency and, despite having been trained, did not initially exhibit the “professional” competencies expected by practitioners. Yet this view changed over time, in part because as interpreters and practitioners interacted every day, their mutual understanding increased.
Integrating trained interpreters in the clinic seems to influence refugee patients’ health, as perceived by their health practitioners. The practitioners’ perception was that it remained stable or ameliorated, for most patients. These findings may be interpreted as a gain, since previous studies tend to show that refugee health decreases over time (Khanlou, 2010). Although the study methodology does not explore the causality between the program and patient health outcomes, this finding is consistent with previous evidence, which suggests that adapted care, such as the presence of trained interpreters, could explain patients’ stable health (Smith, Rodriguez, & Bernal, 2011). Indeed, Smith, Rodriguez, & Bernal showed that offering care in the patient’s preferred language and targeting a specific cultural group were among the main factors explaining the better outcome. It is therefore probable that the enhancement of clinical services by integrating trained interpreters for a specific cultural group of patients (along with all other positive ramifications of the program) partly explains this result. However, in the absence of previous data about the same clinic or other clinics with similar objectives, it is not possible to attribute this result uniquely to the program. Moreover, future studies should measure and consider the interval between visits, as this may also affect health changes.
Surprisingly, patients’ overall satisfaction decreased significantly over time. This drop in satisfaction is independent of gender and age. While significant, it needs to be nuanced by examining specific questionnaire items. For services satisfaction, the difference over time, for the “satisfied” and “very satisfied” answers, is only 4.2% (99.3% vs. 95.1%), which implies that more than 95% of patients are still (very) satisfied. This is true for most items, as mean scores fall between 4 and 5. Items with major statistical significance and size effect over time are related to the relationship with the clinic practitioners (attentiveness, kindness, helpfulness) and the way participants were treated (respect, dignity). Interestingly, the item showing the lowest satisfaction is infringement of personal rights. This item showed no difference over time, which implies the feeling did not improve or worsen, but remained poor. This is also true for difficulty in getting information: this difficulty was frequent or very frequent at both measuring times.
An explanation for this decrease in satisfaction might be the increased knowledge about the health-care system and familiarity with it that refugees gain over time. For example, they became more aware of the services/information they could be provided but did not obtain. As a result, they may become more critical about the health-care system. The generation difference in satisfaction at T2, as indicated by the post hoc test, could also be interpreted from the same perspective: the youngest who learn French more quickly and who go to school may have even higher expectations due to their understanding of the system and language. A similar generational difference in health-care use by migrants and interpretation was reported in a German study (Glaesmer et al., 2011).
The satisfaction drop may also be related to spontaneous patients’ visits, as patients had high expectations of interpreters (which interpreters could not meet). Although this increase in spontaneous visits was framed as a negative result by interpreters and practitioners, it may represent a positive outcome of this project. More precisely, previous research suggests that refugee patients may experience mistrust and/or misunderstanding of health-care services in the host country (Dubus & LeBoeuf, 2019), and failure to access care may partly explain the health decline in refugee populations. Our finding suggests that this program may have led to trust in the clinic and that refugee patients may not be as reluctant to access care.
Strengths and Limitations
To our knowledge, this action-research project is the first of its kind, at least in Canada. Its strengths include the collaborative approach to determine research objectives, in order to address an important need of health practitioners working with refugee patients. As for limitations, fewer Bhutanese refugees than expected arrived in the City of Quebec, due to geopolitics. This eventuality influenced both the generalization of the quantitative results and the interpreters’ integration into the clinic. Meeting with more refugees, especially at the start of the program, could have also resolved ambiguities in practitioners’ expectations of interpreters’ roles. Moreover, it was particularly challenging to find and recruit interpreters with the appropriate background and knowledge: most of them had limited French-language skills at the beginning of the project, low levels of education, and no previous training or experience in interpreting. In addition, only two interpreters remained at the end of the project, and this may limit our understanding of interpreters’ experiences. However, considering these limitations and the unforeseen difficulties encountered during the program, the research findings concerning patients’ health remaining stable might be considered stronger than if they would have been obtained in an ideal and controlled situation. Finally, patient satisfaction questionnaires were completed with the help of program interpreters, because there were no other resources, so this may have affected patients’ responses. However, no questions about interpreters were included in the questionnaire, and interpreters were trained to be impartial while completing questionnaires with patients.
Recommendations
While hiring full-time interpreters is not necessary in all health-care institutions, it probably is in clinics receiving refugees, for this population is especially vulnerable to health decline after they arrive in a new country (Newbold, 2009; Reed et al., 2017). Having full-time interpreters on site can facilitate communication and case administration (Leanza et al., 2015). In order to enhance practitioner-interpreter collaboration, initial training for both is necessary, as well as continuous education (e.g., workshops and discussions of specific cases).
One main issue that needs clarification is interpreters’ roles. They have to be clarified in order for practitioners and interpreters to reach a mutual understanding. Holding joint workshops with interpreters and practitioners could help avoid interpreters experiencing role tensions, as practitioners’ expectations can conflict with those of interpreters, as learned during their interpreter training. This issue also has to be clarified with patients, and this involves a communication effort to transmit a clear message about the work interpreters can and can’t do for patients. As some of the concerned populations are not very literate, the form of this communication needs to be creative (for example, information flyers will be of limited use).
It is necessary to establish criteria for the selection of interpreters. Being bilingual does not necessarily mean having the capacity to be a good interpreter. Beyond linguistic abilities, motivations for interpreting, human qualities (such as empathy), age, gender (have both male and female interpreters), and experience in health care should also be considered. As in our study, even if there is a limited population in which to find qualified people, an effort should be made in the training of these interpreters.
Conclusion
Health-care institutions around the globe should consider and embrace diversity (gender, sexual orientation, age, culture, language, etc.), as it is what constitutes humanity and cannot be ignored. Disregarding it can lead to discrimination and abuse. Interpreters are the embodiment of cultural and linguistic diversity, but also of the possibility of mediating meanings between different worlds. Their integration into institutions is a necessary development in a globalized world.
About the authors
Yvan Leanza is a full professor at the School of Psychology, Université Laval (Québec). He can be reached at yvan.leanza@psy.ulaval.ca
Rhéa Rocque is a postdoctoral fellow at the Prairie Climate Centre, University of Winnipeg. She can be reached at rhea.rocque@gmail.com
Camille Brisset is a senior lecturer at the University of Bordeaux, and member of the LabPsy. She can be reached at camille.brisset@u-bordeaux.fr
Suzanne Gagnon is a full professor at Université Laval. She can be reached at suzanne.gagnon@fmed.ulaval.ca
Abstract
Resumé
Main Text
Introduction
Study Context
Study Objectives
Methods
Training Program
Participants
Data Gathering and Analysis
Practitioners and Interpreters’ Perspectives
Evolution of the Relationship between Practitioners and Interpreters
Patients’ Health Status (as Perceived by Practitioners)
Patients’ Satisfaction
Results
Practitioners’ Experiences during the Program
Interpreters’ Experiences during the Program
Evolution of Interpreter–Practitioner Relationship (during Workshops)
Practitioner Perceptions of Patients’ Health
Patient Satisfaction with Care
Discussion
Strengths and Limitations
Recommendations
Conclusion
About the authors
