Partial Language Concordance in Primary Care Communication: What is Lost, What is Gained, and How to Optimize

Abstract

Objective:

In healthcare encounters, partial language concordance occurs when clinicians have some linguistic ability in the patient’s preferred non-English language or patients have some linguistic ability in English or another language (e.g., Mandarin when their preferred language is Cantonese). While the best practice is to work with qualified professional interpreters in these cases, oftentimes visits occur without qualified interpretation. This paper seeks to understand how these partially concordant visits are conducted and how they may be optimized to meet legal standards of non-discrimination.

Methods:

19 partially concordant primary care visits with Chinese speaking patients were analyzed using discourse analysis.

Results:

Findings illustrate expected communication challenges and unexpected benefits of partial language concordance. While partial language concordance can facilitate rapport building, vocabulary and fluency limitations can also open the door to miscommunication. We also observed how implicit and explicit request of a qualified interpreter to be on standby during the visit can be utilized by the physician to facilitate communication.

Conclusions:

Communication might be optimized by the inclusion of a qualified interpreter with an explicit discussion at the beginning defining interpreter roles.

Practice Implications:

Discussions with interpreters and how they can best engage throughout could retain the rapport created by a shared language, while minimizing miscommunication and errors.

1.0. Introduction

Approximately 22% of the US adult population speaks a language other than English at home, with more than a third of these speaking English less than very well, which the U.S. Census Bureau uses as the definition of limited English proficiency (LEP).[1] Populations with LEP, compared with English-speaking populations, consistently receive worse health care.[2–10] Poor communication also interferes with the quality of the patient-clinician relationship, including the development of trust and follow-up, which impedes patients’ ability to engage in joint decision-making and self-management.[11, 12] These challenges in communication and care delivery contribute to health disparities for LEP populations.

U.S. health care organizations receiving federal funds, which most do in the form of Medicaid or Medicare, must provide language access services for patients with LEP in order to meet legal standards of non-discrimination.[13, 14] Under federal guidance, language access in health care can take two main forms: professional interpreter services and language-concordant care by linguistically-qualified health care professionals.[15]

Language concordance, or when patients and clinicians communicate in a shared language, results in improved healthcare quality and outcomes, including better patient satisfaction with care, medication adherence, patient understanding of diagnoses and treatment, patient functioning for those with diabetes, patient-centeredness, more health education, fewer emergency department visits and fewer missed medications.[16] True language concordance occurs when physicians are proficient in the language their patient speaks.

However, partial language concordance occurs when clinicians have some linguistic ability in the patient’s preferred non-English language and/or patients have some linguistic ability in English or another non-preferred language (e.g., Mandarin when their preferred language is Cantonese). While the recommendation and expectation is to involve a qualified professional interpreter in these cases, clinicians often continue to utilize their non-English language skills and patients’ limited English skills to varying degrees,[17–20] raising concerns for communication errors leading to lower quality of care, patient safety risks, and legal liability.[21, 22] Reasons for this may include clinician or patient preference for direct communication, frustration with understanding much of what is said and interpreted, and time constraints.[23] Beyond recommendations to work with a professional interpreter, there are no guidelines, standards or training about how best to conduct these linguistically complex visits.

Sociolinguistic studies of recorded doctor-patient interactions have examined what actually happens in these interactions. Many of these studies have focused on the interpreter - their role and strategies,[24] how they manage the theoretical roles and front/back stage work,[25] and what they actually do amidst the multiplicity of voices that are engaged in the interactions, especially in “in-between” spaces of translation.[26, 27] As complex and documented these interactions are, language fluency is fluid and dynamic; what is less studied is when the doctor or patient has some degree of fluency in the patient/doctor’s language. Though these partial fluency situations may be less documented because of concerns of legal ramifications, these interactions still occur regularly in primary care settings.

Work on language and interactions among people with different cultural and linguistic backgrounds show how a transdisciplinary framework called translanguaging, or the use of one’s entire linguistic repertoire, is used to negotiate communicative contexts at hand.[28] Rather than attending to so-called named languages (e.g., “Spanish,” “Cantonese” or “English”), language users utilize their full range of linguistic resources - extensive or nascent - to “get the job done,” oftentimes utilized together in hybridized language practices and multimodal resources like gestures and contextualization cues. A translanguaging lens allows “others [to] come to the forefront”,[29] and in healthcare settings, help showcase how “non-native” speaking medical professionals and laypeople alike can carve out spaces to highlight language capacity and knowledge expertise.[30] It involves spotlighting the repertoires that are known (e.g., what the speaker is able to produce) as well as what the speaker does not know and thereby can ask for assistance and support for. Though an important goal of intercultural healthcare is to pass along information that is clear and comprehensible to the patients, communication can oftentimes fall short, even “when there is goodwill on both sides”.[31]

Through this translanguaging lens, the purpose of our study objective was to understand how partially concordant visits are navigated in clinical practice and to examine potential impacts on communication. Specifically, what are the expected and unexpected communication impacts when a patient and clinician have partial language concordance and when a professional interpreter is or is not present during the visit?

2.0. Methods

We identified 19 partially concordant primary care visits with Chinese speaking patients from a dataset of 189 audio recorded visits from a larger communication study.[32–34] Recordings were transcribed verbatim and translated into English by bilingual and bicultural research assistants, twice verified by another research assistant and a subject expert in Chinese sociolinguistics, who also served as a coder.

This study took place in a large, academic general internal medicine practice with three sites serving almost 25 000 diverse patients, approximately 12% of whom had a non-English preferred language for medical care. Professional interpreters were available remotely on demand via video-conferencing (and by telephone as a back-up), and on-site with prior scheduling. At the time of the audio-recordings, professional interpreters were included for the vast majority of language discordant visits.

Data analysis was done in Dedoose qualitative data analysis software (Dedoose Version 9.0.17) and done in iterative steps using discourse analysis,[35, 36] paying analytic attention to the interactional work of the talk. Two coders first coded one transcript together to discuss and calibrate codes and preliminary themes. Coders resolved any disagreements through further discussion. Once the coding protocol was clear and consistent, the remaining 10 transcripts were coded by one coder.

The two coders engaged in sessions of open and focused coding. Open coding involves labeling concepts, themes, and categories in the data, whereas focused coding analyzes the specific ways the themes play out line by line. [35, 36] Examining the data with an open mind, we were interested in the meso-level analysis of the data: what the words are doing interactively in the visit. Team members provided feedback on the themes during regularly scheduled meetings.

3.0. Results

Patient and clinician characteristics are presented in Table 1. Of the 19 patients, 13 reported Mandarin and 6 reported Cantonese was their preferred language for healthcare. Fourteen patients reported speaking English “not well” or “not at all”, while 5 reported that they spoke English “well.”

Table 1.

Patient and Clinician Characteristics During Partially Language Concordant Primary Care Visits (N=19)

Characteristic of Patients	N=19 (%)
Age, years
50-60	4 (21.1)
61-70	4 (21.1)
71-80	6 (31.6)
81+	5 (26.3)
Preferred Language
Cantonese	6 (31.6)
Mandarin	13 (68.4)
Characteristic of Clinicians	N=9 (%)
Self-Reported Mandarin Ability
None	5 (55.6)
Good	4 (44.4)
Language Test Results
Pass (Mandarin)	1 (11.1)
Marginal Pass (Mandarin)	1 (11.1)
Fail (Mandarin)	1 (11.1)
Not Tested in Chinese language	6 (66.7)

Among the 9 clinicians, none reported having any Cantonese ability. Four reported “good” Mandarin ability, while 5 reported none on the Interagency Language Roundtable for Healthcare (ILR-H) measure.[37, 38] The ILR-H is a validated self-report measure for clinician non-English language ability in healthcare settings, which allows the clinician to self-describe speaking and comprehension levels in detail. Six had not tested their language proficiency with the Clinician Cultural and Linguistic Assessment or CCLA[39] which was the health system’s preferred assessment tool; and, among the 3 who did test, 1 passed, 1 failed, and 1 marginally failed (and was offered a retest which they did not take).

Visit characteristics are presented in Table 2. Among the 19 visits, 5 did and 14 did not have qualified professional interpreters present. Of these, 11 visits were conducted by clinicians in Mandarin, with the remaining 3 conducted with patients accommodating in English. More details with the patient/provider combination, patient’s preferred language, visit language, and notes can be found in the appendix table.

Table 2.

Qualified Professional Interpreter Presence During Partially Language Concordant Primary Care Visits (N=19)

	N (%)
No Interpreter Present
Clinician Speaks Some Mandarin	11 (57.9)
Patient Speaks Some English	3 (15.8)
Professional Interpreter Present
Interpreter Actively Interprets	3 (15.8)
Interpreter on Explicit Standby*	1 (5.3)
Interpreter on Implicit Standby*	1 (5.3)

^*Standby defined as qualified interpreter jumps in to interpret when there is a breakdown in vocabulary or understanding of either patient or clinician. Explicit indicates that the clinician discussed the standby role with the interpreter at the beginning of the visit; implicit indicates that the standby role was assumed without discussion.

3.1. Negotiating Working with an Interpreter and Visit Language

Four of the 19 visits had explicit discussion of working with qualified interpreters, with two being initiated by patient request. A patient’s self-reported English language ability did not seem predictive of their desire for professional interpretation. For example, in one encounter the doctor begins two and a half minutes of opening talk in English asking about the patient’s recent and upcoming surgeries. When the doctor agenda-sets with the question “how about on your list, things you want to talk about?” the (Cantonese preferred, “well” English-speaking) patient responds with “ah today we don’t have a translator?” to which the doctor offers a telephone interpreter, which the patient accepts.

Similarly, the patient’s self-reported language ability may not always align with the choice of working with a professional interpreter. In the following example the doctor and patient have previously spoken in English even though the patient reports speaking English “not well.” The doctor offers a professional interpreter multiple times and ways, but the patient declines:

The doctor acknowledges that previous English use does not mean that the patient will also want to use English this visit. The doctor does not push the issue of working with an interpreter further and moves the visit along using English. However, as even this short excerpt demonstrates, the use of non-preferred language already leads to potential misunderstanding.

3.2. Working with a Non-Professional Interpreter

When patients and providers utilized non-professional interpretation (or with a professional interpreter absent), there were expected and unexpected communication impacts.

3.2.1. Expected Communication Impact: Vocabulary and Fluency Limitations

For non-professionally interpreted visits where one party is speaking in their non-preferred language, complications may arise due to a mismatch in fluency. These difficulties can present as linguistic compensation or imprecise diction, or in more extreme cases, miscommunication and misunderstanding. When there is no professional interpreter present, patients and providers negotiate language usage to communicate to the best of their abilities.

The following excerpt presents a situation where the physician encounters some confusion when inquiring about snoring, which they then attempt to overcome by acting out and emulating the sound.

In this scenario where English is the predominant language used, the physician (who speaks fluent Mandarin and has passed the certification test) initially attempts to overcome this confusion by first clarifying with Mandarin. The patient, however, speaks Cantonese, and instead advocates for their English-speaking ability (“just say it”). The physician reiterates in Mandarin and emulates the sound of snoring in an attempt to overcome the communication barrier. Without an interpreter present, patients and physicians rely on alternative, more informal methods of communication.

In the following excerpt, a more subtle, linguistic nuance is overlooked leading to a misunderstanding between patient and physician.

Here, a prolonged discussion regarding an advanced directive (in which the patient and physician both agree to defer further discussion) is concluded by the physician stating “好隨便你們講” [ok, whatever you say]. In this context, this specific phrase can have a negative connotation, similarly to “do whatever you want”. In response, the patient understands the physician’s reply as having underlying annoyance or anger. Although the physician clarifies that they are not angry, the patient continues to attempt to clarify their logic to alleviate the perceived anger from the physician. In this excerpt, we see how the limitation of partial proficiency extends beyond straightforward, correctable confusion in vocabulary, instead having deeper linguistic implications where nuanced aspects of communication, context and connotation can shift the direction and mood of a visit.

3.2.2. Unexpected Communication Impact: Rapport-building

The utilization of one’s language abilities or shared culture through small talk can provide a sense of familiarity and comfort that can help lay the foundation for rapport-building. Shared language can provide space for the patient to set the tone for a visit and build rapport in the form of praise.

In excerpt #5 (Table 4), the resident physician is seeing the patient for the last time prior to completing residency. Throughout this interaction, the patient expresses their gratitude for the care they have received from the physician (lines 2, 6, and 7) and praises them enthusiastically, repeatedly thanking the physician for their attentiveness. Although the physician’s conversation content initially revolves around the health of the patient and care plans, the patient’s content is focused on the emotional aspects of care and appreciation. The physician ultimately reflects this with reciprocal praise of the patient’s family (“your daughter and son also care about you so much”). Linguistically, we also see how in lines 3-5 the patient and physician naturally reiterate the other’s statements to reaffirm what was said and the conversation flows easily.

Table 4.

Quotations for Using Non-Professional Interpretation

Expected Communication Impact: Vocabulary and Fluency Limitations
Excerpt 2: 2095 Doctor: Languages: English, Mandarin (certified). Patient: Preferred Language: Cantonese; Speaks English: Not well. Visit Language: English 1. Doc: anyone ever told you that you snore? 2. Pt: excuse me? Pardon me? 3. Doc: no uhm打鼾 (snore) in Mandarin oh you don’t speak Mandarin 4. Pt: just say it 5. Doc: 打鼾 (snore) ((imitating snoring sounds)) Excerpt 3: 6004 Doctor Languages: English, Mandarin (not certified). Patient: Preferred Language: Mandarin; Speaks English: Not well. Visit Language: Mandarin 1. Doc: Ok 好 隨便你們講 ok 那六月份再見 (Ok, whatever you say, so, see you in June) 2. Pt: 你不要生氣啊 (Don’t get angry) 3. Doc: 我沒有 (I’m not) 4. Pt: 這個是我們 (This is our) 5. Doc: 我了解 我知道 (I understand, I know) 6. Pt: 我們中國人的心理 (This is how Chinese people think)
Unexpected Communication Impact: Rapport-building
Excerpt 4: 2315: Doctor: Languages: English, Mandarin (Marginal Pass). Patient: Preferred Language: Cantonese; Speaks English: Not well. Visit Languages: Mandarin 1. Doc [Mandarin]: one of them 2. Pt: you can write it out for me 3. Doc: triglycerides, that is a type of cholesterol, is a little elevated 4. Pt [Mandarin]: is it this one? 5. Doc: yes. This is one of the cholesterol, one of them is elevated. But it is still not a big problem for now 6. Pt: I think you said the same thing last time. When I checked that last time, you said, don’t eat too much butter! Ha ha Excerpt 5: 2463: Doctor Languages: English, Mandarin (Marginal Pass). Patient: Preferred Language: Mandarin; Speaks English: Not well. Visit Languages: Mandarin 1. Doc: Ok 那我今天給你那個 order 一下然後這個 (Ok, so I will order for you today and then) 2. C: 醫生你好仔細喔 (Doctor you are so attentive) 3. Doc: 因為其實一般人到八十六歲 我們一般不在檢查了 但是我覺得你還是比較健康 (Because normally when people get to 86 years old, we don’t check anymore, but I think you are still healthy) 4. Pt: 比較健康 (Still healthy) 5. Doc: 繼續檢查 (Keep checking) 6. Pt: 好的 好的 好的 你好仔細 很高興 真的 (Ok, ok, ok, you are so attentive, very happy, really) 7. C: 謝謝你照顧他 謝謝你照顧他 照顧的這麼好 (Thank you for taking care of her, thank you for taking care of her, for doing such a good job) 8. Doc: 你的女兒兒子都怎麼關心你 (Your daughter and son also care about you so much)

3.3. Working with Professional Interpreters

When professional interpreters were present, there were expected and unexpected communication impacts. Specifically, we highlight the ways explicit and implicit standby are employed in context. Notably, this was not a one-off interactional event; rather, it was a process undertaken throughout the whole visit.

3.3.1. Expected Communication Impact: Business as Usual Interpretation

In professionally interpreted visits, one expects a particular kind of turn-taking, with the interpreter providing the interpreted information after each turn by doctor and patient. In most cases, this occurred smoothly with very little commentary on the interpretation. In the following example, the interpreter’s presence is useful to clarify the location of the patient’s pain.

Prior to this excerpt, the physician asks about vaginal discharge, to which the patient notes she doesn’t “know how to say it” in English, and that is when the phone translator gets called. We present this “typical pattern” to demonstrate how deviations from this pattern - in the following sections - may represent noteworthy interactional moves, in other words, when speakers linguistically position themselves and take stances, create alignments, and construct personae (Bucholtz 2009).

3.3.2. Unexpected Communication Impact: Interpreter on ‘Standby’

In the two visits in which the interpreter was on standby, two different approaches were utilized by the same doctor who has “good” Mandarin language skills but who had failed the Mandarin certification test. In the first case, they explicitly request the interpreter to standby and encourage them to intervene when needed. In the second case, with a different interpreter, the doctor attempts to go “as far as possible” but without explicitly instructing the interpreter of their intentions. In moments where the doctor signaled a need for help or when they completely switched to English, the interpreter immediately chimed in with no hesitation and no instruction to do so.

3.3.2.1. Explicit Standby

Immediately after calling into interpreter services, the interpreter provides the usual introduction with their name and “how can I help you?” The doctor responds to the interpreter explicitly that they are actually able to speak Mandarin and would like to put the interpreter on what the doctor calls “standby”:

Not only does the interpreter agree to being put on “standby,” they also explicitly praise the doctor’s Mandarin. Through a translanguaging lens, the doctor and interpreter co-construct an interactional communicative event with the patient, wherein the doctor’s “little Mandarin” is able to do the work up until it can’t, and the interpreter, on standby, can step in to support.

A few lines later, when talking about medication options, the doctor and patient smoothly interact with many instances of code-switching, or switching between languages in the same utterance. Throughout this interaction, the doctor is typing while the patient talks.

The long pause and Mm? in line 33 signal to the interpreter that the doctor may not have understood. However, it’s actually not clear if the doctor does not understand or just didn’t hear the question because of the typing. When the interpreter starts to provide the patient’s utterance as far back as line 25, the doctor asserts their competency in Mandarin listening comprehension (“that part I all understood”) to interrupt the interpreter to direct the interpretation to the specific part of the interaction that was missed (line 31).

In the final excerpt near the end of the visit, it is clear how the doctor’s code-switching and bootstrapping, or utilizing the full range of existing linguistic resources to extend to new contexts, work to build rapport:

As seen in lines 182 and 185, the audible laughter in the exchange shows a lightheartedness that we did not notice in the other professionally interpreted encounters. The patient not only jokes in Mandarin but uses it to also ask a clinically relevant question. Although the doctor’s Mandarin ability is not officially up to par and the patient reports speaking English “not well,” they are both able to follow one another for much of the visit through code-switching. The doctor praises the patient in line 183 in English and the patient also responds in English. The doctor is able to deliver their medical recommendation (to worry less) in Mandarin. The exchange between both doctor and patient seems so smooth that the interpreter’s insertion at line 189 feels unnecessary.

3.3.2.2. Implicit Standby

Unlike the previous example, in the following visit, the doctor did not explicitly discuss their role with the interpreter. Instead, the doctor begins the visit, speaking a mix of English and Mandarin, going through pleasantries and catching up about how the patient is doing. Soon afterward, the doctor confirms that the preferred language is Mandarin and calls an interpreter. After the interpreter introduces themselves and interprets the first few lines, then the patient responds and the doctor bypasses the interpreter, demonstrating their understanding. However, in the example below, we see the limits of the doctor’s language abilities when an unfamiliar word (“medical record”) arises in the conversation.

It is not clear whether the doctor’s question in line three is addressing the interpreter or the patient, but the interpreter clarifies. Notably, the interpreter is able to augment the doctor’s language proficiency by smoothly stepping in at a key co-constructed translanguaging moment to provide a quick English interpretation so that the visit can continue. Though this is a fairly short exchange, it is clear that the patient, doctor, and interpreter are co-constructing the exchange together in an efficient manner that does not repeat words that the doctor or patient seem to understand. This excerpt exemplifies the multilingual and triadic nature of these partially concordant clinical visits and the interpreter’s flexibility in meeting the specific linguistic needs in each moment.

Throughout the visit, both doctor and patient offer conversational cues that signal to the interpreter that the previous utterance did not need interpretation. For example, in the following excerpt the patient is explaining that they won’t be tired if they get enough sleep:

In both English and Mandarin, the doctor and patient demonstrate their understanding of each other’s language - with the doctor latching, or following up with the next utterance without a beat of silence, the patient and responding both in Mandarin, “I understand,” in lines 69 and 71 and with shared laughter. Similarly, after the doctor’s question in English, the patient latches her response also in English, “Yes” in line 77. Unlike the latched speech from the previous two exchanges, the doctor’s long stretches of English in lines 76 and 78, coupled with a one-second pause, results in the interpreter providing a Mandarin interpretation. This extends to interpreting the patient’s “no,” even though the doctor likely understood that basic utterance. It is not always clear when an interpreter is needed in this implicit standby encounter; however, there are moments when it is clear that the interpreter is not needed.

4.0. Discussion and Conclusion

4.1. Discussion

In this novel qualitative study of primary care visits, we used the concept of translanguaging - the use of one’s entire linguistic repertoire - to understand how physicians, patients, and interpreters navigate partial language concordance. Among the visits that did not have a professional interpreter present, we found communication challenges related to vocabulary and fluency limitations of both the physician’s Chinese language and the patient’s English. Errors leading to miscommunication have been documented previously in the literature and thus were somewhat expected.[40, 41]

We also observed unexpected positive impacts of shared language and culture related to rapport-building. Rapport-building and trust are important as they have been demonstrated to have positive impact on health outcomes.[42–46] Often found in moments of small talk that blend the relational and transactional frames of the encounter,[26. 27] rapport-building and trust has been documented previously for fully language concordant visits,[16] but not for partially concordant visits. Our data show how these can also occur within the medical discussion and that, despite some complications that may arise from a lack of complete fluency, even partial shared language use can have important benefits that should be maximized while continuing to minimize the potential for harm and meeting the federal legal requirements for non-discrimination.[14]

In the majority of these partially concordant visits, a professional interpreter was present, and ‘business as usual’ interpretation was the primary vehicle for communication. As expected, these visits generally flowed well and were largely focused on informational exchange; however, there was less humor and rapport-building than seen in the visits without professional interpretation.

Notably, for two visits, the interpreter was incorporated into the visit in an unexpected way. In one case, the physician explained the partial language concordance situation to the interpreter at the beginning of the visit and asked the interpreter explicitly to be on standby to interpret when communication broke down due to a lack of vocabulary or understanding. In the other case, the physician launched into the visit without explicit explanation or instruction to the interpreter. In both cases, the interpreter was silent for most of the visit while the patient and physician communicated largely in Chinese with some English mixed in, yet the interpreter was clearly paying attention as they were able to jump in quickly to interpret when they noted possible confusion, lack of vocabulary, or a developing miscommunication. We believe this study illuminates the complexity of partially concordant visits and suggests the possibility of a new way to enhance communication in these moments of “in-between”-ness.[26] With further research involving professional interpreters, partially bilingual clinicians and patients, this approach could point to a more explicit manner of working fluidly with professional interpreters as part of the team while maintaining the potential benefits derived from moments of shared language usage.

Our study has limitations. This was a single site study in primary care and must be interpreted in that context. Additionally, all recordings were only audio without visual data, and were for Cantonese- and Mandarin-speaking patients; a more diverse language group may have allowed for additional insights. The only two examples of visits with an interpreter on standby were both with the same physician who had very strong Mandarin skills, likely not representing all of the existing variations of this type of communication during clinical encounters. However, even with this single physician, we did observe two different approaches – one explicit and one implicit – to integrating a professional interpreter on standby. In the implicit case, the interpreter was able to participate on standby despite no explicit discussion; however, it is not clear that all interpreters would be as adaptable in that situation. Challenges that we did not observe in our limited examples could include interpreters not jumping in easily due to a power differential between physician and interpreter, patient or physician missing or ignoring an interpreter’s efforts to speak, or interpreters disengaging from the conversation if they perceive they are not needed or valued during the interaction. This limitation demonstrates the need to further understand interpreter, patient, and physician viewpoints on the best approaches to incorporating a professional interpreter on standby for partially language concordant encounters.

4.2. Conclusion

In these data, we found that partial language concordance can facilitate dynamic moments of rapport-building, but vocabulary and fluency limitations also open the door to miscommunication. Interpreter use on standby may mitigate these disadvantages.

4.3. Practice Implications

Communication might be optimized by explicit discussion at the beginning of each partially concordant visit defining the role of the interpreter and how they can best engage throughout. Future research can help elucidate best practices for visits conducted in this manner.

Table 3.

Quotation for Negotiating Interpreter Use and Visit Language

Excerpt 1: 2095 Doctor Languages: English, Mandarin (certified). Patient: Preferred Language: Cantonese; Speaks English: Not well. Visit Language: English 1. Doc: we’ve done your visits in English before, do you want to do English? Or, in   Chinese? Up to you 2. Pt: okay 3. Doc: please sit. Mandarin and Cantonese? 4. Pt: Cantonese 5. Doc: Cantonese? Okay. Do you want to do this in Cantonese today or English 6. Pt: you English 7. Doc: yes I speak English, but I can have an interpreter in here, for Cantonese 8. Pt: I understand, I don’t know. Sorry I don’t mean 9. Doc: okay, you tell me if you feel like you want Cantonese, we will have someone   Cantonese okay. Alright, how are you feeling

Table 5.

Quotations for Using Professional Interpretation

Expected Communication Impact: Business as Usual Interpretation
Excerpt 6: 2608 Doctor Languages: English, Spanish (certified). Patient: Preferred Language: Cantonese; Speaks English: Well. Visit Languages: English and Cantonese 7. Dr: [Eng] Can you tell me about your abdominal pain? Tell me more about that! 8. Int: [Can] Hi Miss, my name is (name). I am your translator for today. The doctor   said you were telling her about your belly pain, and she wants to know   more about it 9. Pt: [Can] Hi! I do not have belly pain but stomach pain. Stomach pain 10. Int: [Eng] She just wants to clarify. It’s not exactly abdominal pain, but more of   stomach pain 11. Dr: [Eng] Okay! And how long has that been going on? 12. Int: [Can] How long has your stomach pain been going on? 13. Pt: [Can] Eh on and off for a couple months already. This month was more   serious! 14. Int: [Eng] It’s been going on and off for a few months and this month, it gets   worse
Unexpected Communication Impact: Interpreter on Explicit Standby
Excerpt 7A: 2668 Doctor Languages: English, Mandarin (not certified). Patient: Preferred Language: Mandarin; Speaks English: Not Well. Visit Languages: Mandarin and English 1. Int: thank you, how can I help you? 2. Doc: So I just want to put you on standby, because I do speak Mandarin. Uhm,   little so I’m gonna 如果我可以的話我就跟他 可是如果我沒辦法講或是我   聽懂的話 我就需要你的幫忙 (So I just want to put you on standby, because   do speak Mandarin a little so I gonna, if I can I will talk to him, but if I can’t   talk or I can’t understand, I would need your help) 3. Int: 可以啊 (Sure) 4. Doc: Ok 5. Int: 你國語很好啊 (Your Mandarin is really good)   Excerpt 7B   25. Pt: 還有一個老早用的ACE呢有一個現象就是咳嗽 (There’s also the ACE   [inhibitor] I have been using for a long time, the one issue is that I cough   [when using it]) 26. Doc: Ah ah huh ah huh (affirmative) 27. Pt: ((cough)) 28. Doc: Yeah 28. Pt: 所以呢 uh他呢 叫我用另外一個藥 ARB, mm   (So, he [the doctor], asks me to use another medicine, ARB) 29. Doc: ((begins typing)) 30. Pt: 還有甚麼其他措施嗎? (What other steps are there [besides ARB]?) 31. Doc: ((typing after patient stops talking))Mm? 32. Int: Ah. Uh. What he is saying is before when I was using ACE um which will   cause my 34. Doc: Oh yeah that part I all understood, it’s just the last part 35. Int: yeah the last part doctor said I should use the ARB but I’m wondering if there   is any other way you could do 36. Doc: Yeah, unfortunately 沒有別的 (there is no other way)   Excerpt 7C   180. Pt: 都是我在講今天 (I’m doing all the talking today) 181. Doc: Uh? 182. Pt: 哈哈哈哈 你有甚麼suggestion啊? (Ha ha ha ha, do you have any   suggestion?) 183. Doc: No nothing, I think you are doing well 184. Pt: No, no, no 185. Doc: 你是太擔心 (You just worry too much) thaťs the only thing ha ha ha ha 186. Pt: Because I. nothing to do I also read, read 187. Doc: Worrying is not healthy. Just know you are doing well 188. Pt: ((hearty laughter)) 189. Int: 你做的很好 不要太擔心 擔心本身就不是一件很好的事 (You are doing   well, don’t worry too much, worrying is not good.) 190. Doc: Ok thank you very much £interpreter£ 191. Pt: Ok thank you. thank you
Unexpected Communication Impact: Interpreter on Implicit Standby
Excerpt 8A: 6023 Doctor Languages: English, Mandarin (not certified). Patient: Preferred Language: Mandarin; Speaks English: Not Well. Visit Languages: Mandarin and English 1. Spouse: 你看看我的病歷 (You have a look of my medical records) 2. Pt: 等她看一看，你等她看一看，她自己會看的 (Let her check, you let   her check. She will check it herself) 3. Doc: 什麼是病歷 (What is medical record?) 4. Pt: 她會看的 (She will check) 5. Int: medical record 6. Doc: oh, ok Excerpt 8B   68. Pt: 我並不是因為晚上睡了早上起來還感覺那個疲倦，我不是那種情   况。我就是說需要很長時間，時間短了我就難受  (No, no. I am not feeling tired the next day when I go up, that’s not the case. I  mean I need long period of sleeping. If the sleeping time is too short I will feel uncomfortable) 69. Doc: .hhhh 我懂了 oh 我懂了我懂了 (I understand oh I understand I understand) 70. Pt: 時間長了我就不難受 (I won’t if the sleeping time is long.) 71. Doc: 我懂了，有一些人，我也是那樣子。我需要睡很多很多 (I understand,   some people, I am also like this. I need to sleep a lot, a lot) 72. Pt: [((laughter)) 73. Doc: 有一些人就是這個樣子，那没有關係, ok, ok  (It’s common for some people. That’s fine) 74. Pt: Uh huh huh huh (affirmative) 75. Doc: Ok  ((laughter from both)) 76. Doc: Ok. 你的血壓your blood pressure today is ok did you take your blood   pressuremedicine today? 77. Pt: Yes 78. Doc: Ok, good. Um. And then. do you have any issues of chest pain, palpitations,   difficulty breathing, headaches? 79. Int: 你有没有以下情况：頭痛啊，呼吸困難啊，或者是心悸的感覺啊，胸疼   啊？有没有這些症狀？(Do you have following issues: headache, difficulty   breathing, or palpitations, chest pain? Do you have them?) 80. Pt: 没有 (No) 81. Int: No 82. Doc: Good. Okay so I think that we can keep your blood pressure medicine as is,   and then I think that you are not due for any other health care tests, except for   colonoscopy

Appendix

Table A1.

Patient and Provider Combination, Patient Preferred Language, and Visit Language and Notes (N=19)

Patient ID	Provider ID	Patient Age	Patient Preferred Language	Visit Language and notes
2095	7	52	Cantonese	English*
2306	18	75	Cantonese	Mandarin
2338	18	76	Cantonese	Mandarin
2173	18	75	Mandarin	Mandarin
2269	18	83	Mandarin	Mandarin
2507	18	82	Mandarin	Mandarin
6030	18	85	Mandarin	Mandarin
2110	31^	51	Cantonese	Mandarin
2315	31	66	Cantonese	Mandarin
2463	31	86	Mandarin	Mandarin
6008	31	69	Mandarin	Mandarin
2668	32**	77	Mandarin	Mandarin, interpreter on standby
6004	32	81	Mandarin	Mandarin, interpreter offered, denied by patient
6023	32	74	Mandarin	Professionally interpreted, interspersed Mandarin use by physician
2271	5	61	Mandarin	Professionally interpreted
2608	16	58	Cantonese	Professionally interpreted
2719	30	51	Mandarin	Professionally interpreted
2288	35	62	Mandarin	English
2282	37	72	Mandarin	English

^*Patient explicitly requests visit to be done in English

^**Physician with self-identified “good” Chinese, failed test