“How’s Your Mood”: Recorded Physician Mental Health Conversations with Chinese and Latino Patients in Routine Primary Care Visits

Abstract

Objective:

Patient-physician communication patterns may influence discussions around depressive symptoms and contribute to engagement in depression care among racial/ethnic minority adults. We examined patient-physician communication about depressive symptoms during routine primary care visits with Chinese and Latino patients with and without language barriers.

Methods:

We examined 17 audio-recorded conversations between primary care physicians and Chinese (N=7) and Latino (N=10) patients who discussed mental health during their visit and reported depressive symptoms on a post-visit survey. Conversations (in English, Cantonese, Mandarin, Hoisan-wa, Spanish) were transcribed and translated by bilingual/bicultural research assistants and analyzed using inductive and deductive thematic and discourse analysis.

Results:

Patients initiated mental health discussion in eleven visits. Physicians demonstrated care in word choice and sometimes avoided openly mentioning depression; this could contribute to miscommunication around symptoms and treatment goals. Interpreters had difficulty finding single words to convey terms used by either patients or physicians.

Conclusion:

Patients and doctors appeared willing to discuss mental health; however, variability in terminology presented challenges in mental health discussions in this culturally and linguistically diverse sample.

Practice Implications:

Further understanding patient preferred terminology about mental health symptoms and interpreter training in these terms could improve patient-physician communication about depressive symptoms and treatment preferences.

1. Introduction

Primary care settings play a key role for individuals with depression. Compared to those without depression, individuals with depression are more likely to have higher rates of complications from and poorer management of other chronic conditions(1–4) and increased risk of cardiovascular and all-cause mortality.(5–7) Most individuals present first for depression care in primary care settings.(8) While primary care physicians provide most depression care,(8) many physicians often fail to appropriately recognize depression(9) and lack training to effectively treat depression.(9–12)

In the U.S., Chinese-heritage (hereafter ‘Chinese’) and Latino individuals have a high burden of unrecognized and untreated depression.(13–21) These populations include the largest segments of the U.S. population with limited English proficiency (LEP),(22) which presents an additional challenge to receiving quality depression care.(20,21,23) Patients with language barriers are at increased risk of poor physician recognition of depressive symptoms and treatment initiation.(15,23–29) Furthermore, they are less likely to receive needed mental health services, with lower referrals and less treatment than English-speaking patients.(13,14,23,25,29–32). Prior studies have shown only 6-36% of Chinese and Latino patients with language barriers and depressive symptoms receive mental health care.(24,26,33–35) These disparities persist even when accounting for insurance coverage and access to care.(36)

In depression care, patient preferences, culture, stigma, family influences, social norms, and attitudes strongly influence help-seeking, discussions about depressive symptoms, treatment acceptance, and adherence.(25,27,28,37–39) Prior research has examined patient preferences for depression communication in primary care through focus groups and interviews, finding that although patients prefer to have the final decision regarding treatment, they also rely on a partnership with their physician to share information, preferences, and recommendations and to treat their depression as part of a larger patient-centered care approach.(40,41) Few studies, however, have directly examined primary care interactions for how physicians and patients discuss depressive symptoms, diagnoses, or treatment preferences(37) and whether these conversations demonstrate patient-centered care.(27,42–45) In this paper, we examine patient-physician communication regarding depressive symptoms in audio-recorded routine primary care visits for a sample of Chinese and Latino patients with and without language barriers.

2. Methods

2.1. Data Collection

Data came from the Language Access System Improvement (LASI) study which evaluated the effects of simultaneously increasing access to professional interpreters and certifying bilingual physicians’ language skills on communication and clinical outcomes in primary care.(46) This longitudinal study was conducted at an academic primary care practice serving an ethnically, linguistically, and socio-economically diverse population of over 25,000 adult patients in San Francisco. Patients were eligible to participate if they were age ≥ 40 years; self-identified as Chinese or Latino (the largest populations with LEP in this practice); preferred English, Cantonese, Mandarin, or Spanish; and received primary care in the practice. As part of the LASI study, visits for patients with LEP were categorized (22,47) as language concordant, discordant-professionally interpreted, and partially concordant (See Table 1 footnote).(20,48)

Table 1.

Characteristics of Study Participants in Visits with Mental Health Talk

Characteristic of Patients	N=17 (%)
Age, years
50-55	6 (35.3)
56-65	4 (23.5)
66-75	5 (29.4)
76+	2 (11.8)
Gender
Female	13 (76.5)
Male	4 (23.5)
Language Spoken with Physician
Spanish	7 (41.2)
Mandarin	4 (23.5)
Cantonese	1 (5.9)
Hoisan-wa	1 (5.9)
English	4 (23.5)
Education
Less Than High School	12 (70.6)
High School	2 (11.8)
College Degree or higher	3 (17.6)
Health Insurance Status
Medicare	12 (70.6)
Medicaid	4 (23.5)
Private Insurance	1 (5.9)
Visit with Primary Care Provider	9 (52.9)
Visit Language Concordant Statusa
Fully Concordant – English	4 (23.5)
Fully Concordant – Spanish	2 (11.8)
Partially Concordant – No Interpreter	3 (17.6)
Discordant – Professional Interpreter	8 (47.1)
Mental Health Need	11 (64.7)
Use of Mental Health Services	10 (58.8)
Characteristic of Clinicians	N=14 (%)
Age, years
50-55	6 (35.3)
56-65	4 (23.5)
66-75	5 (29.4)
76+	2 (11.8)
Gender
Female	13 (76.5)
Male	4 (23.5)
Language Spoken with Physician
Spanish	7 (41.2)
Mandarin	4 (23.5)
Cantonese	1 (5.9)
Hoisan-wa	1 (5.9)
English	4 (23.5)
Education
Less Than High School	12 (70.6)
High School	2 (11.8)
College Degree or higher	3 (17.6)
Health Insurance Status
Medicare	12 (70.6)
Medicaid	4 (23.5)
Private Insurance	1 (5.9)
Visit with Primary Care Provider	9 (52.9)
Clinician Type
Faculty Physician	7 (50)
Resident Physician	7 (50)
Clinician Gender
Female	6 (42.9)
Male	8 (57.1)

^aPhysician fluency and certification, as well as visit language concordant status, were determined as part of the parent study, the Language Access System Improvement (LASI) study. Fully concordant visits were those where the patient spoke in preferred language and physician was fluent or certified in that language. Partial concordance occurred if the patient was speaking in a dis-preferred language which the physician spoke fluently or physician spoke in a non-English language but was not certified in that language.

A subset of participants had primary care visits audio-recorded (N=191) between August 2016 and October 2017. For these participants, at the time of a regularly scheduled primary care visit, bilingual/bicultural research assistants obtained consent in the waiting room and provided patients with a digital recorder to take into the exam room. At the end of the visit, patients returned recorders to research assistants, who administered a short, in-person, post-visit survey.

We assessed current depressive symptoms during the post-visit survey using the Patient Health Questionnaire-2 (PHQ-2), a well-validated and widely used clinical measure. Participants were categorized as having current depressive symptoms if they had a PHQ-2 score ≥ 3 (sensitivity of 61-87% and specificity of 78-92% for major depression)(49–52). The PHQ-2 has been validated in both Chinese and Spanish with similar sensitivity and specificity among ethnically Chinese and Latino populations as among non-Hispanic whites (53–57).

We obtained additional participant demographics and characteristics from the electronic health record (EHR), including age, gender, and insurance type (private, Medicare, or MediCal); we collected participants’ self-reported preferred language and education in the post-visit survey. Participant clinical characteristics and health service utilization included prior history of depression (defined as depression listed in the EHR problem list prior to the visit or a diagnosis of depression entered at a previous outpatient encounter), whether the patient was seen by their own primary care provider (PCP) or another physician at the index visit, and patient-reported mental health service need and service use in the prior 12 months.(58)

Of the 191 audio-recorded visits, 35 participants reported current depressive symptoms on the post-visit research survey. We reviewed these 35 recordings for any discussion of mental health symptoms, diagnosis, or treatment during the visit (Appendix A). In this analysis, we included only the audio recordings for participants who reported current depressive symptoms on the post-visit survey and discussed mental health during their visit (N= 17). Mental health discussion was any discussion of emotional state (e.g., depressed or low mood, feelings of sadness, anxiety, etc.) and/or depression treatment, including references to medications or side effects, psychotherapy, and to mental health professionals. Physicians were unaware of the PHQ-2 results at the time of the visit as the measure was collected in the post-visit survey. This study was conducted prior to implementation of general depression screening of adults in the primary care practice.(59)

2.2. Data Analysis

We used qualitative thematic analysis and discourse analysis to investigate how patients with depressive symptoms communicate with physicians during routine primary care visits.(60) Discourse analysis has been used in healthcare to analyze both the taken-for-granted nature of everyday patient-physician interactions and the complex ways everyday talk constructs meaning in relational and health consequences.(61) Audio-recordings were professionally transcribed, verified against original audio-recordings to ensure accuracy, and translated to English if necessary. One of two investigators, including one primary care physician and one discourse analysis expert, and three research assistants first read each transcript using an iterative process,(62) marking all moments of possible mental health talk and who began the discussion. Obvious cases included mentions of depression, anxiety, or seeing a therapist/psychologist/psychiatrist. Less obvious cases included mentions of sadness, grief, or stress but all possibilities were included in initial coding. We then relied on the physician who reviewed all discussions categorized as not containing mental health talk to assess whether somatic symptoms that may be central to depressive symptomatology were being discussed. Next, we independently open-coded the excerpts identifying key concepts or interactional patterns. We read, reread, and relistened to the excerpts to understand how and by whom the talk about depressive symptoms was introduced, how it was reacted to, how the conversation progressed, and what purpose the mental health talk served for patient or doctor. From this discourse analysis-informed process, we solidified our coding book of concepts and patterns reaching consensus about the code structure before re-reading additional transcripts searching for these or other new themes or patterns of talk until reaching saturation. (See Appendix A and B for more information; contact authors for full code book). We used Dedoose to organize and share data and codes (Dedoose Version 8.0.35, 2018). To ensure rigor, we wrote memos and maintained an audit trail and used peer debriefing and consensus-building discussions to strengthen our analysis.(63) Once we agreed upon identified communication patterns, we listened to excerpts again to refine our transcripts and note any interactionally relevant micro-aspects of talk including pauses, restarts, wording, translation issues, etc. that participants seemed to attend to in their next turns of talk. To avoid influencing the qualitative analysis, the study team only accessed patient diagnoses of depression, reported perceived mental health need, and prior year mental health service use after developing patterns.

3. Results

Patient demographic, clinical, and visit characteristics are presented in Table 1. Among the 17 participants who reported current depressive symptoms on the post-visit survey and who discussed mental health at their visit, 77% were female; 41% Chinese, 59% Latino, and 80% reported having LEP. Twelve participants (71%) reported mental health need and eleven (65%) reported use of mental health services in the prior year. Half (53%) saw their own PCP at the audio recorded visit.

Mental health discussion was initiated by physicians in six visits and by patients in eleven visits. Visits reflected a wide array of purposes for mental health discussions, including discussing new depressive symptoms, initiating treatment (with a referral or medication), assessing previous treatment (including side effects and referrals to community resources), discussion of somatic symptoms, and raising concerns about social determinants of health (see Appendix B).

3.1. Interactional Impact of Varied Terminology to Discuss Mental Health Symptoms

The next four subsections explore patterns of terminology mismatch and their interactional consequences. Patients and physicians often varied in the terms used to initiate mental health discussions and describe symptoms, with physicians sometimes avoiding mentioning depression and using different terms than the patient. Patients engaged openly in mental health talk, even across languages. However, in interpreted visits, various terms were used for ‘mood’ and ‘psychologist’ adding layers of possible miscommunication. Besides language interpretation, these excerpts also demonstrate how participants negotiated the meanings of colloquial mental health terms versus their clinical or medical meanings. Finally, we provide an example with multiple forms of these misalignments.

3.1.1. Terminology Misalignment

In Excerpt 1a (Table 2), the patient speaks with the Cantonese-speaking research assistant (RA) who helps the patient complete the PHQ-9 questionnaire (additional seven questions asked if the PHQ-2 is positive) prior to the visit, as requested by that visit’s physician. Noteworthy in this exchange (besides the fact that the RA appears to answer for the patient in line 32) is how openly the patient declares in line 33 that she is “extremely depressed.” However, it is not clear if “depressed” means clinical depression, as colloquially this term can also express deep sadness and low emotional state (as in English). The patient does have a history of using mental health services and in line 111 (Excerpt 1b), expresses a desire to consult a psychologist. Despite this open mental health talk with the RA prior to the visit, this is not the case when the patient later discusses her “mood” with the doctor in Mandarin, her less preferred language. It is worth noting that the term for “mood” [心情 xinqing] was never mentioned when talking to the RA and the patient responds to queries about mood with references to her heart and mind [心腦 xinnao], a term not used by the PCP.

Table 2.

Quotations for Interactional Impact of Varied Terminology

Terminology Misalignment			Some Analytic Points
Excerpt 1a: Cantonese-preference Female patient; Cantonese-speaking Research Assistant.
Languag Used: Cantonese
32	RA:	Do you overly worry about different things? I think you do
33	Pt:	yes, I am extremely depressed [jauwat 憂鬱]	Patient openly uses ‘depression’
[Lines removed discussing other screening questions]
110	RA:	your family doctor is
111	Pt:	I can’t get an appointment with him, I want to get a consultation for a psychologist now	Patient states purpose openly – to see psychologist
Excerpt 1b: Cantonese-preference Female patient; Mandarin-speaking Physician. Language Used: Mandarin
124	Dr:	that’s okay. How is your mood [xinqing 心情] lately?	Doctor uses term ‘mood’
125	Pt:	I can’t sleep well at night, I always wake up during my sleep. Now, I also doze off. My heart/mind [xinnao 心腦 ] is not good, I always get chest tightness. My arms are not good, and now it’s even painful in my legs.	Chinese term in response to ‘mood’
		They are only looking at my arms now, but not my legs and waist yet. It is very tight, right here. Very uncomfortable, it’s so painful that I wake up in the middle of the night.	Move to somatic symptoms
126	Dr:	waking up because of pain…is that just at night, or you have other sleeping problems as well?
127	Pt:	pain, or sometimes just can’t sleep well
[50 lines removed: discussion about sleep and “medication for your mood”]			Doctor raises the point about the therapist – uses this term.
177	Dr:	you have seen a therapist in Chinatown before, is that correct?
178	Pt:	yes
179	Dr:	so did you like that therapist then? No? You would like to have another therapist?	Patient mirrors with ‘therapist’
180	Pt:	I would like to see a therapist from [this hospital]
Interpretation Misalignment			Some Analytic Points
Excerpt 2: Spanish-preference Latino patient; English-speaking Doctor; Professional Interpreter. Language spoken in (parentheses)
136	Dr:	(Eng) Mm-hm, ok. Ok, thank you for sharing that with me. Um, ok, and how, um, how is your — I mean a transition out of talking about your mood, cause I, I want to make sure that you’re uh depression is not getting worse. So can you tell me about your mood please?	Doctor starts with ‘mood’, says ‘depression’, but then reverts back to “mood”.
137	Int:	(Sp) Great sir, great, thank you very much for sharing. Now we are going to change topic, we are going to talk about another topic which is, is your mood [su humor]….how are you feeling [como se siente]. How are you feeling with all of this? We were talking about — we are going to talk about the depression [la depresión]. We have seen that the depression [la depresión] hasn’t gotten worse, but we would like to know how you’re feeling? What are you feeling?	‘Humor’ is not an accurate translation of ‘mood’, later interpreter uses ‘feeling’
138	Pt:	(Sp) Well, the depression [la depresión], mm. If I see a child cry, I cry. The depression [la depresión], is quite.. I have to be careful about what TV programs that I see. Nothing dramatic, because that, how can I say it? If I see a drama, it’s like I participate in the drama.	Of the various terms, the patient uses ‘depression’
[21 lines removed: talk about changing the TV channel to avoid sadness/upset]			Despite the patient’s term, which the doctor may not have recognized in Spanish, doctor uses ‘mood’ again
158	Dr:	(Eng) Yeah. So compared to uhm. Because I don’t know you that well, I’m not your primary care doctor, so um, I’m gonna ask you about how your mood now compares to how your mood was let’s say one to two months ago. So if you think back to one to two months ago, do you think your mood today is better, worse, or the same as a couple of months ago?
159	Pt:	(Sp) Yes, it’s better now
160	Int:	(Eng) It’s better now
161	Dr:	(Eng) Ok. And, and why do you think that is?
162	Int:	(Sp) And why do you think that is sir?
163	Pt:	(Sp) Mm, well now I can go to the bathroom, I can eat, and I can sleep early	Patient’s proof of improvement is based on physical things, less about feelings
164	Dr:	(Eng) Mm-hm
165	Int:	(Eng) Yeah, you know, it is because now I can, I go to the bathroom, I can eat, and I can sleep mostly early
Misalignments in Use of Medical Terminology			Some Analytic Points
Excerpt 3: English-preference Latina patient; English-speaking physician. Language used: English
16	Dr:	How are you?
17	Pt:	I’m fair.
18	Dr:	Yeah, what’s going on?
19	Pt:	I think I’m suffering from depression again	Patient openly uses ‘depression.’
20	Dr:	Yeah. Why is that?
21	Pt:	Oh, a series of things and then ( ) is just making me more depressed
22	Dr:	What’s happening with your work? Are you still working?	Doctor clarifies
23	Pt:	I am. It’s crazy.
24	Dr:	Yeah?
25	Pt:	I negotiated to have a flex schedule. I get three hours off in the middle of the day, so I go to the gym
26	Dr:	Okay. Good for you!
[165 Lines removed: discussion about exercise]
191	Dr:	You said that you’re feeling down?
192	Pt:	I am.	Doctor downgrades and defines patient’s experience away from ‘depression’ to “feeling down”
193	Dr:	Yeah?
194	Pt:	Because I don’t even want to stay here anymore. I mean, staying here in the United States.
195	Dr:	Why is that?
196	Pt:	Well, it’s because it’s not so quiet. It wasn’t for dog baby, I would’ve gone and my parents are no longer alive, so I remain here
197	Dr:	Where would you want to go?
198	Pt:	Back to Jamaica
199	Dr:	Uh-huh (affirmative) Uh-huh (affirmative)
200	Pt:	It’s just that health wise I’m not really sure how that would work because in terms of I do need to be monitored on a regular basis. I can’t believe people can be so xenophobic and so … Yeah.
201	Dr:	Yeah. Oh, I know. I would say at least a couple times a day I have patients who come here and say that they’re really anxious about what’s going on.	Doctor relates and describes similar other patients as ‘anxious’ and that it is ‘distressing’ but not ‘depressing’
202	Pt:	It makes me sick to my stomach. It really does
203	Dr:	Yeah. It’s distressing
Excerpt 4 Cantonese-preference female patient; Mandarin-speaking resident physician (Res); Cantonese-speaking attending/supervising physician (AP). Languages spoken: Cantonese with some English
510	AP:	hello! So you are using your husband’s sleeping pills right now?
511	Pt:	yes.
512	AP:	and it’s effective?
513	Pt:	yes, it’s been a while already
514	AP:	okay, got it, got it. He prescribed more for you. It doesn’t interact with your blood pressure medicine, and not much side effect, no dependency, so it’s good?
515	Pt:	yes
516	AP:	And for you had your gallbladder removed, right?
517	Pt:	yes
518	AP:	hoping there won’t be more problems, haha
519	Pt:	me too
520	AP:	haha, good good. (Eng directed to Res) she’s up to date on her immunizations and stuff? Okay.
521	Pt:	he said everything is (Eng) normal (Can) for me, but I feel like I’m foggy/not stable [wanwan zanzan 瘟瘟陣陣]	Patient differentiates between doctor’s diagnosis of normal and patient’s feeling of being foggy and unstable, a term the supervising physician doesn’t actually recognize. Patient explains that it’s a way to measure health, which she deems as ‘bad’-openly discussing mental health
522	AP:	foggy/not stable [wanwan zanzan 瘟瘟陣陣]? What do you mean by foggy/not stable [wanwan zanzan 瘟瘟陣陣]?
523	Pt:	it’s like, my health is very bad, kind of like that
524	AP:	mmm it’s not a proble:::m and now that things are taken care of. Like let’s hope that the problems will become stable[wanding 穩定] for a while don’t be too worried. (Eng) she’s saying everything looks normal, she’s just like she’s si::ck=
525	Res:	=(Eng) yeah haha=
526	AP:	=(Eng) I said yeah she’s been through a lot. (Can) I hope. With luck that you won’t have other problems.	Doctors continue to reassure patient Patient then suggests ‘stress’ but discloses stopping the psychologist
527	Pt:	maybe it’s just stress, maybe, I have to, I have a lot of stress. I even stopped my psychologist [samlei jisang 心理醫生], too. Ever since I got admitted to the hospital and did these things. I haven’t had much time, so stopped.
528	AP:	(Eng) she said she stopped seeing her psychiatrist. (Can) Do you feel your mood [samcing 心情] is still depressed?	Attending changes to ‘psychiatrist’, then asks specifically about the ‘mood being depressed’. Patient responds about ‘stress.’ Attending switches to ‘psychologist’. In talking to the Resident, switches to therapist.
529	Pt:	a lot of stress lately. A lot of things going on.
530	AP:	I think you should see the psychologist [samlei jisang 心理醫生] again.
531	Pt:	I will do so, too.
532	AP:	okay so you should go see him again. (Eng) she said she’s having a lot of stress, and pressure, so she’s going to see the therapist again.
533	Res:	(Eng) okay.
534	AP:	okay, so see you!
Complex Misalignment: Interpretation and Lay/Medical Terms			Some Analytic Points
Excerpt 5 Cantonese-preference female patient; Mandarin-speaking physician. Language used: Mandarin
65	Dr:	So you didn’t go anymore
66	Pt:	yes
67	Dr:	so you, may I ask after the passing of your daughter, how is you mood [jingshen 精神] now?	Doctor uses ‘mood’
68	Pt:	It’s a little better. Because what should I think, she passed away already
69	Dr:	yes
70	Pt:	she is not here already, it is useless for me to think. But don’t think ( ), I always think about it in the middle of the night as I am sleeping, about the process, when she was younger, studying, and what I did wrong. I think, would it be that I did something wrong, that’s why she left, sometimes I will blame myself.
[typing sounds throughout]
71	Dr:	Yes, this is normal. Especially someone in the family passed away, I will too, think about them once in a while. But I think that it is important that as long as your spirit [jingshen 精神] is not especially, you know, spirit [jingshen 精神] especially like, like, the dimensions of your heart [xinwei 心位] are not especially serious/extreme, as long as it’s not like this, then that’s considered a little better.	Despite the patient’s seriousness, doctor focuses on this being a normal response to grief. There are multiple Mandarin words to discuss ‘mood/spirit/heart’ which doctor includes
[14 lines removed discussing sleep and whether the patient communicated/talked to the psychiatrist]
85	Dr:	Yes. No, its, what I’m saying is, is, to, with your psychiatric doctor [jingshen 精神 doctor], does the you know the psychology [xinli 心裡], you know, like psychology[xinli 心裡] treatment. It’s the, you know, not the eat medicine, you know, to control your that, how should I put it, uh.	Many terms are included but the doctor seems to be trying to find the word for talk therapy or not-eating-medicine-treatment, which the patient understands and correctly identifies as psychologist. This is a concrete action (therapy) that doctor suggests to patient.
86	Pt:	He called me. He said if you don’t eat medicine, you can see us. We will find a psychologist [xinli 心裡 doctor] to talk with you
87	Dr:	yes just like this
88	Pt:	He asked me. Later I said, I said, because I have a family friend. She’s in the same situation. That person, her ( ) husband, later her son last time it was on the newspaper, that when a person went fishing, and a wave pulled him away that person. He is my sister’s husband’s brother.
89	Pt:	She said to me, her mother could get through it, why couldn’t you. She said it is testing me, if I could, I could get through this challenge, I think I should figure something to get over it.
90	Dr:	I think you should too, you should, there should be some ways to. But if you want to go back to you know and talk with a doctor
91	Pt:	yes
92	Dr:	talk thoroughly about this, and do a psychology plus this kind of therapy, it can help you in this situation.	Doctor suggests talking as a possible treatment, uses the terms psychology + therapy Patient explaining that the only option she is offered is medication, therefore, she’s not taking it Doctor shifts gears to social support
93	Pt:	He told me, he said, I told him this situation with taking medicine, and how I think. Later he said, he has to put it in a file, he said, we will keep a file in the hospital, if you need it then you will have to come back to us. If you don’t need it then no need to reach us, it’s been a month already
94	Dr:	already a month
95	Pt:	hence I am not going to him.
96	Dr:	Never mind then. That’s with, if, anyway in this important period of time, you need someone for support, if family and friends can support you, your ( ) will be the best.

When the physician asks about “mood,” the patient responds with a vast array of somatic complaints but nowhere mentions that she is feeling ‘extremely depressed,’ as she told the RA when completing the screening tool. The physician attends to each somatic complaint individually, first dealing with current medication, then family arguments, then sleep. In a number of instances, the physician uses the term mood rather than depression. For example, in the removed lines the physician says, “feeling weak, that could be related to your mood” and later “the medication for your mood… you started that 2 weeks ago, right?” It takes 50 more lines for the patient to be able to state her request – that she would like to see a therapist from this hospital – which is prompted finally by the physician in line 177. It could be that the physician’s shift of mental health talk away from specific medical diagnoses such as ‘depression’ toward a more general idea of ‘mood’ and the use of ‘therapist’ (rather than the more specific ‘psychologist’) allows for the patient to bring up a more holistic description of her current situation. However, it could also be that such terminology distracted the patient from her goal of a referral to a psychologist for what she declared to the RA as extreme depression.

3.1.2. Interpretation Misalignment

In Excerpt 2 (Table 2), a professionally-interpreted encounter, the doctor begins by agenda-setting around three topics: anemia, weight loss, and the patient’s “depressed mood.” The interpreter translates “depressed mood” as depresión [depression] in Spanish in the lines just before this excerpt. However, in line 136, the doctor’s mixed usage of the English words depression and mood become difficult to disentangle.

Throughout this visit, the doctor uses the term “mood” which is translated variably by the interpreter as humor, como se siente, depresión. The problem is that humor can also mean mood in the sense of temperament, whereas como se siente is typically used to ask ‘how are you?’ While mood may work in English, not having a specific translation in Spanish may affect how mental health gets discussed. In this case, in line 138, the patient answers specifically mentioning his depresión and then gives an example about emotions. On the other hand, by line 158 when the doctor asks about mood, the patient seems to focus on the improvement of somatic symptoms of appetite, sleep, and defecation. In this interpreted visit, the different terms may lead to a misalignment in patient and physician discussion of mental health symptoms since the conversation seems to focus primarily on somatic symptoms and potential treatments for these symptoms. While this focus may be intentional by the patient or physician, it may also be an unintended consequence of the variation in terminology and interpreted terms used by physicians and interpreters. Similar translational issues occurred in Chinese languages as well; when a physician suggested that a senior center might improve a patient’s mood, the interpreter used three different words in Hoisan including ‘heart’ (thlim 心), ‘feelings/mood’ (thlimtein 心情) and ‘spirit’ (deinthlin 精神).

3.1.3. Misalignment in Use of Medical Terminology

While previous excerpts demonstrate the patients’ use of the term depression, other patients used the term ‘depression’ in more colloquial ways to describe recent stressors or factors that were bothering them (e.g., “I get so depressed I just wanna cry” when referring to difficulties taking multiple medications). In some cases, physicians worked to disentangle stressors and low mood from symptoms meeting the diagnostic criteria for depression. For example, (Table 2, Excerpt 3) one patient who had multiple medical conditions (i.e., kidney stones, an autoimmune condition, chronic pain) and stressors (i.e., wanting to return to home country, political anxiety, response to social unrest) endorsed ‘depression,’ but when the physician probed, it seemed the patient’s situation was actually improving and she was feeling better, exercising, and recently returned to work (after taking extended time off for medical leave). The patient uses the expression “suffering from depression again” and “making me more depressed” but the physician does not reciprocate this language. Instead, the physician’s use of “feeling down” seems to function to remind the patient of the difference between “depression” which she has had before and the “distress” that she may be currently experiencing, which may not require escalation in medical treatment.

In the lines removed discussing exercise, the physician asks, “What do you feel?” to which the patient responds “feel upbeat,” and “in a better mood.” In line 191, the doctor again raises the issue stating, “you said that you’re feeling down.” The patient describes “upsetting” aspects of living in the U.S. The physician validates, sharing how other patients have similar anxieties and describing the climate as “distressing.” From a clinical standpoint, the doctor focuses on the patient’s ability to work, exercise, and interact socially with others as indications that these symptoms are not affecting function. The physician frames the concerns as anxieties and stressors, whereas the patient frames them as her depression recurring. In this example, both physician and patient are able to share important information and despite the differing terms, perhaps come to a shared understanding that differentiates clinical depression from the “depressing” state of current affairs.

In another instance, the lack of specificity in terms used for mental versus overall health leads the physician to focus on physical health throughout the visit. In Table 2, Excerpt 4, a female Cantonese patient first spoke in Mandarin to the resident physician, and later in Cantonese with the attending physician about using her husband’s sleeping pills. The patient had been admitted to the hospital for gallbladder removal and says in line 521 that even though her other doctor said everything is normal, she uses a colloquial term that can be translated as ‘foggy or not stable,’ a term the doctor doesn’t recognize and asks about. The patient explains using a more general statement, “my health is very bad.” In generalizing and moving away from mental health, it opens the door for the attending physician to reiterate to both the patient and to the resident (in English) that her (physical) health is now normal and hopefully stable. The patient again tries to bring up mental health via talk about stress and mentioning she has even stopped going to the psychologist (line 527), but it takes more conversational turns to get to this discussion and settle on the recommendation that she should return.

3.1.4. Complex Misalignment: Interpretation and Lay/Medical Terms

Variability in terminology used for mental health clinicians also led to misunderstandings between patients and physicians. For example, in a language concordant visit in Cantonese, a patient initially used the term ‘psychologist,’ but the physician then responded variably about the ‘psychiatrist’ and ‘therapist.’ This may be further complicated in partially concordant visits. In Table 2, Excerpt 5, a female Cantonese patient is speaking in Mandarin despite saying at the beginning of the visit that “My Mandarin is not so good, I never learned it.” With the Mandarin-speaking doctor, the patient discusses dealing with the death of her adult daughter at length. The patient had been referred to and met with a psychiatrist, took a single pill, experienced side effects of confusion and stopped both the medication and the visits.

The conversation about the patient’s management of her situation with the loss of her daughter continues throughout most of the visit. In this excerpt, the term the doctor uses for mood (精神) jingshen is actually the same term that is used in the term psychiatrist (a jingshen doctor 精神科醫生), which is notably different from the patient’s original term (xinli ke or psychology心理科). Just like in other English cases, neither the patient nor doctor necessarily differentiate between what a psychiatrist does and what a psychologist does, although it appears the patient here views the clinician (in her term – a psychologist) as being a supplier of medicine. This becomes even more apparent as the doctor tries repeatedly in line 85 to find the words for talk therapy or counseling since using the term psychology (treatment) doesn’t necessarily imply talking and isn’t clearly differentiating it from psychiatry for the patient. Starting in line 86 the patient seems to understand, explaining that she was offered talking as an optional therapy. She effectively rejects the suggestion that she needs to talk this over with anyone by using a story of her friend who suffered a loss as an example of how she just needs to get over it. It is unclear if she also agrees that she just needs to get over it, or if she is telling the doctor that ‘talking to people’ like her family friend is also unhelpful. Either way, the doctor moves on with ‘never mind’ in line 96.

4. Discussion and Conclusion

4.1. Discussion

Audio-recorded primary care visits presented important opportunities for both patients and physicians to check in regarding mental health topics and concerns in a trusted clinical setting. Primary care visits offer important moments for developing partnerships and patient centered care.(40) Patient-centered communication consists of a physician’s exploration of the patient perspective of health, emotion, and care expectations, leading to a mutual definition of the problem and goals for care.(64) Despite both patients’ and doctors’ willingness to discuss mental health during the observed visits, the mental health conversations in this culturally and linguistically diverse sample demonstrated challenges due to variation in language proficiency (in English and other languages) and terminology choice and knowledge. The misalignment in terminology used between patients, doctors, and interpreters to discuss depressive and mood symptoms in these audio-recorded primary care visits complicated discussions about mental health, with evidence of patient and physician misunderstandings or cross-talk about symptoms and treatment options.

In interpreted encounters, patients and physicians often used different terminology for mental health symptoms. When terms were either mistranslated or variably translated, this led to confusion or misunderstandings about symptoms or treatment options or preferences. Interestingly, physicians did not necessarily follow the patient’s use of terms (and vice versa), and interpreters also often switched between various terms. It is unclear if interpreters switched between terms because the patients or physicians were switching or because interpreters were not sure how to best translate a word that is not commonly used in Spanish or one of the Chinese languages.

More work is needed to determine Chinese and Latino patient preferred terminology to discuss mental health concerns (e.g., more specific terms such as ‘depression’ versus discussions of ‘mood’ or how one is feeling). While physicians may prefer terms such as ‘mood,’ which in English may feel less stigmatizing compared to ‘depression,’ these terms have the potential for misinterpretation when used in multilingual clinical settings and when patients are already using the term depression. This is particularly true, and potentially problematic, in interpreted encounters. Once preferred interpretations are established, training interpreters on mental health specific terminology could mitigate some of these misunderstandings.

Additionally, variability in terminology may be related to or contribute to poor patient mental health literacy. The conversations captured here demonstrate that some patients may not understand or believe in the relationship between somatic symptoms and mental illness. Furthermore, the variability in terms used to describe mental health professionals may reflect a lack of understanding or misunderstanding of their roles and what treatment modalities entail. The use of various terms could also reflect stigma about mental health or treatment options.(65) This may reflect a need to provide additional education to patients with depressive symptoms about what treatment modalities entail (to increase mental health literacy) and where they are available (including what can and cannot be achieved in primary care settings).

Our study has limitations. It was conducted in one primary care site and reflects a single clinical audio-recorded interaction between a patient and a doctor. Patients, and physicians, may have had competing demands for discussion in a time-limited primary care encounter. Furthermore, the physicians were not aware that the patient had screened positive for depressive symptoms since the screening was done by a research assistant immediately after the visit; discussions may be qualitatively different after implementation of routine adult depression screening. Additionally, while some patients had new depressive symptoms, others had established diagnoses of depression and thus physicians (and patients) may have chosen not to discuss a known or improving problem. Conversely, not all visits were with the patient’s own PCP, frequently leading to more focus on a non-mental health urgent purpose of the visit. We may also have missed non-verbal communication during visits regarding symptoms. Finally, while all patients had a PHQ score ≥ 3, not all patients may have been clinically depressed. However, high perceived mental health need, use of mental health services in the prior year, and previous diagnoses of depression indicates high likelihood of mental health symptoms in our sample.

4.2. Conclusion

Documented disparities in depression care for Chinese and Latino patients including those with language barriers, are likely multifactorial. Disparities may be driven by the patient-physician relationship and communication patterns, stigma, a focus on the somatic symptoms of depression, lack of culturally-congruent services, and unavailable treatment preferences.(13,15,23,26–28,32,34,38,39,66) Our findings demonstrate that patients and physicians are both willing to discuss mental health, but often have variable expectations for the purpose of mental health discussions. Furthermore, physician and, when present, interpreter, choice of mental health terminology has important interactional impacts that could possibly affect clinical care. Further understanding of culturally and linguistically preferred mental health terminology could improve depression care outcomes for Chinese and Latino patients in primary care.

4.3. Practice Implications

Our findings demonstrate a need to further understand Chinese and Latino patient preferred terms for discussion of mental health symptoms and treatment modalities. Once these terms are better understood, interpreters should be trained in the use of preferred terminology, to facilitate patient-physician communication about mental health symptoms. Meanwhile, physicians should try to keep terminology about mental health symptoms consistent and avoid terminology that could be easily misunderstood – such as “mood” in Spanish – and, when appropriate, mirror the terminology that patients themselves are using. Finally, our findings suggest a need for further education for patients about mental health conditions and treatment modalities to increase mental health literacy.

Highlights.

• Chinese and Latino patients initiate mental health discussion in primary care visits

• Physicians demonstrated care in word choice and hesitation mentioning depression

• Patients, physicians, and interpreters varied in terms used in mental health talk

• Physician and interpreter terms affect interaction and possibly clinical care

• Interpreters should be trained in the use of preferred mental health terminology

Appendix A

Interaction Definitions / Coding Decisions^*

	Definition	Example
Mental Health (MH) Talk*	Any discussion by one or both parties concerning depressive symptoms, mood/emotion, depression treatment (therapy/ medication/referrals), or somatic symptoms commonly present with depression as defined by DSM-V criteria (i.e. insomnia, low energy, etc.).	Patient: “I’m not in best mental status…. maybe it’s just stress… stopped seeing psychologist.”
MH Talk Initiation	The first initiation of the specific mention of mental health. This could be in the form of a statement or a question.	Patient Initiation: Doctor: “is there anything I can do to help you quit [drinking alcohol]” Patient: “I don’t know…” then a long story of being on probation, not seeing his son, his depression and medications. Doctor Initiation: “How has your mood been?”

^*All possible cases categorized by the study team as not containing MH talk were double-checked by the senior author who is a primary care physician with expertise in mental health. She reviewed all discussions to assess/identify whether somatic symptoms that may be central to depressive symptomatology were being discussed. Any close cases were included rather than excluded.

Appendix B

Additional Quotations Mental Health (MH) Talk Purposes^*

Purposes	Example(s)
Initiating treatment     • Referrals     • Medication	Doctor: “the whole point of asking you (about) these senior centers that speak Hoisan in Chinatown… to join and participate so that it is beneficial to you… to be more social.”
Checking in on previous treatment     • Effects of medications or referrals     • Refill requests     • Side effects of medications     • Referral to clinicians     • Referral to community resources	Patient: “I’m taking the pills for stress and for sleep. I sleep like 12 hours. I only take half of half, because the whole I’d sleep 24 hours.” Doctor: “Do you feel like your mood has gotten any better since you started to see (the therapist)?” Doctor: “So since we’re gonna make this change in your medicine for your mood today, then I don’t want to change your blood pressure pills today, cause I only wanna make one change at a time and I think we can have you change your blood pressure pills next time.” Doctor: “We can refer you to integrative medicine clinic. They sometimes do like massage therapy, sometimes acupuncture.” Patient: “I want to get a consultation for a psychologist now.”
Discussing new/recurrent depressive symptoms	Patient: “I think I’m suffering from depression again…I don’t even want to stay here (in the United States) anymore” Patient: “I feel like I’m foggy/not stable.”
Discussing somatic symptoms	Patients: “I can’t sleep well at night…My mind/heart is not good. I always get chest tightness. My arms are not good, and now it’s even painful in my legs and waist yet.” Doctor (in response to patient asking about heart palpitations): “Um. I suspect with the scores of this test we just completed (the anxiety screening), this is likely anxiety. But the thing is we should always make sure that it is nothing more serious.”
Raising concerns about social determinants of health	Patient (in response to the doctor stating, ‘Last time, the doctor thought you might be a little sad and depressed’): “I feel like these are related to problems with my family and our income. It’s very difficult for me to handle. Because now with my retirement checks and Social Security… you can see how now after rent is taken out, I don’t have much to buy food with”

^*MH Talk could have multiple purposes including different purposes for the patient versus the doctor. While we do not know what intent each party had, interactionally we could point out where the conversation headed and whether one party allowed the conversation to move to a logical next step or whether they continued to return to the topic. We identified all possible purposes and then categorized the common patterns here.