The “Battle” of Managing Language Barriers in Health Care

Abstract

Providing safe and high-quality health care for children whose parents have limited English proficiency (LEP) remains challenging. Reports of parent perspectives on navigating language discordance in health care are limited. We analyzed portions of 48 interviews focused on language barriers from 2 qualitative interview studies of the pediatric health care experiences of LEP Latina mothers in 2 urban US cities. We found mothers experienced frustration with health care and reported suboptimal accommodation for language barriers. Six themes emerged relevant to health care across settings: the “battle” of managing language barriers, preference for bilingual providers, negative bias toward interpreted encounters, “getting by” with limited language skills, fear of being a burden, and stigma and discrimination experienced by LEP families. Parents’ insights highlight reasons why effective language accommodation in health care remains challenging. Partnering with families to address the management of language barriers is needed to improve health care quality and safety for LEP patients and families.

Latinos are the largest minority population in the United States and comprise the majority of the 25 million people in the United States with limited English proficiency (LEP).^1,2 Latino children experience disparities in the access to and quality and safety of medical care.³ Even greater health care disparities are experienced by Latino children in LEP families. Compared with Latino children with English-proficient parents, Latino children with LEP parents are less likely to have a medical home or timely medical care, more likely to have nonurgent ED visits and compromised medication safety, and their parents report worse communication with providers and greater dissatisfaction with health care.^3–11 Health care quality and outcomes improve for LEP patients and families, however, when professional interpreters are used or language-concordant providers are available.^12,13 Unfortunately, LEP patients and families often do not receive appropriate language services.^11,14,15

Based on Title VI of the Civil Rights Act, LEP patients and families must have meaningful access to language services.¹⁶ The National Standards for Culturally and Linguistically Appropriate Services in Health Care (CLAS standards) issued by the US Department of Health and Human Services provide guidance on Title VI compliance.¹⁷ Appropriate accommodations for LEP patients during health care encounters include use of language-concordant providers with sufficient target language proficiency and/or certified, professional interpreters.¹⁷ CLAS standards have been operationalized into the Joint Commission accreditation process, but many hospitals do not meet them, even though this poses a risk to their accreditation.^18.19 Optimally meeting the language needs of LEP patients and families remains challenging; the supply of bilingual physicians is low compared to the LEP patient population, and professional interpretation can present logistical and financial barriers for clinics and health systems.^{11,14,20–22}

To our knowledge, there is limited information on parent perspectives on the management of language barriers in health care. Understanding LEP parents’ perspectives on management of language barriers is necessary to increase use of appropriate and family-centered language services and reduce persistent health care disparities due to language. This study aims to describe the perspectives of LEP Latina mothers on their experiences with language services in pediatric health care to inform the development of more patient- and family-centered language services.

Methods

Study Design and Population

This study is a secondary data analysis of collated data from 2 semistructured Spanish-language interview studies conducted in urban settings with immigrant Latina mothers. Both studies included mothers of children with access to and use of pediatric primary care in the past year to focus on the experience of using care. Language barriers emerged as a prominent theme from both studies. Similarities in participant experiences and demographics in the 2 study populations facilitated conducting an in-depth analysis of this theme of language barriers in health care among urban Latina mothers.

The first study (n = 38) was conducted from September through December 2009 in the predominantly Latino community of Southwest Detroit, Michigan, in partnership with a local federally qualified community health center, The Community Health and Social Services (CHASS) Center.²³ Participants completed 1 semistructured interview about their pediatric health care experiences. The second study (n = 10) was conducted from October 2011 through July 2012 in Baltimore, Maryland, at an outpatient, general pediatrics practice serving predominantly immigrant Latino families.²⁴ Participants were family-member participants in the clinic’s inaugural year of the Latino Family Advisory Board (LFAB). Participants completed a total of 2 interviews to better understand their experiences as board members. A section of each interview was dedicated to exploring their pediatric health care experiences, more generally and were used in this study. Inclusion criteria did not require mothers to have LEP for either study. We used the US Census Bureau question “How well do you speak English?” to ascertain LEP status based on a response of speaking English less than very well.²⁵ All mothers except for one (Detroit study) had LEP. That mother still reported preferring Spanish for health care encounters, and her responses still reflected facing language barriers and so, as with the original analysis, the corresponding interview was retained.

Data Collection and Analysis

In both studies, the majority of interviews were conducted in the participants’ homes by a bilingual, bicultural experienced Latina interviewer with community knowledge. A minority of interviews in Detroit were conducted at the health center based on participant preference. Interviews lasted between 25 and 90 minutes. Participants signed informed consent after the consent form was orally read to them and received $25 for their participation. This study was approved by the institutional review boards at the University of Michigan, CHASS, and Johns Hopkins Medicine.

Interviews were digitally audio-recorded and transcribed verbatim into Spanish, for both studies, and then translated into English using a commercial transcription and translation services company. Names of people and places were assigned a pseudonym. The principal investigator for both studies (LRD) worked with bilingual and bicultural study staff at both sites to refine translations to better reflect meanings as understood by Spanish speakers in the study communities. Final interview transcripts included Spanish and English text to allow for data coding and analysis by study team members not proficient in Spanish and coding of interviews in the original language by those proficient in Spanish. Participants in both studies had opportunities to provide feedback on original study findings and interpretations. In Detroit, 2 study participants contributed to the data analysis through periodic meetings, and CHASS staff provided feedback on study findings and interpretation. In Baltimore, LFAB participants provided feedback on study findings and interpretation periodically during board meetings the year following study completion.

We extracted Spanish and corresponding English-language text segments from the original coding of both studies for inclusion in this study. Extracted text segments included those coded under themes of communication, language barriers, interpretation, English language proficiency, and discrimination; the coding structure was similar for each of the 2 original studies. We included the discrimination code as these experiences consistently related to discrimination based on language. A total of 281 unique text segments were extracted from the Detroit-based study, and 75 unique text segments were extracted from the Baltimore-based study.

After data extraction, all codes from the original studies were removed. The coding team (LRD, ES, DVA) then reviewed the extracted transcripts to identify preliminary themes. The research team developed a codebook based on these themes (eg, physician Spanish-language proficiency, interpreter access, ease of communication). An iterative consensus process determined that codes were clearly defined and could be consistently applied by all team members. During initial development and application of the codebook, 2 study team members, who would serve as primary coders, coded 6 transcripts (LRD, ES). All remaining transcripts were coded by 2 coders, with at least 1 of the 2 primary coders coding each transcript. Rather than employing a measure of intercoder reliability, we used previously established methods for addressing differences in coding due to multiple coders by addressing all coding discrepancies and reconciling them through discussion and consensus.^26,27

We used Atlas.ti V5.7.1 to apply codes to the transcripts and to organize text segments and relevant quotes abstracted during analysis.

Results

Interview data from 48 participants were included in this study. Most of the mothers were of Mexican descent and had 2 or 3 children, the majority of whom were US-born. Table 1 displays additional respondent characteristics. There only notable difference in demographic characteristics by study site was country of origin. In the Detroit study, 95% of mothers were of Mexican-origin, while 50% of the Baltimore study participants were of Mexican-origin. Non-Mexican-origin participants were from varied Latin countries in Central and South America and the Caribbean. Participants described receiving pediatric care at primary care practices, urgent care practices, the emergency room, and navigating referrals to specialty care or other child health services such as dental care or developmental services.

Table 1.

Characteristics of Sample Mothers (n = 48) and Their Children.

Maternal age (years), mean	32.5 (range = 20–44)
Country of origin—Mexico (%)	73
Length of stay in the United States (years), mean	9.5 (range = 2–21)
Maternal education
6th grade or less (%)	40
Greater than 6th grade to some high school (%)	30
High school graduate or more (%)	30
Annual household income
<$20 000 (%)	81
Mother lives with husband/partner (%)	79
Mother’s health status fair/poor (%)	23
Mean number of children	2.6 (range = 1–6)
Age of children (years), mean	6.6 (range = 2 months to 21 years)
US-born children (%)	81
Children’s health insurance status
Medicaid (%)	81
Uninsured (%)	19
Children’s health status fair/poor (%)	8

The majority of participants discussed their primary care experiences more positively than experiences in other health care settings. In general, participants reported more negative experiences overcoming language barriers in specialty care, emergency care, and other child health services. The majority of respondents identified that they had a primary care clinic for their child, but also commonly discussed seeking urgent/ emergent care for acute illnesses. Increased satisfaction with primary care compared with other health care was related to both better accommodation of language needs by primary care providers and more familiarity with the system of care in primary care on the part of parents. Specialty and emergency care were fraught with difficulties for mothers. They reported less access to language services, which then magnified the challenges they faced navigating unfamiliar health care settings. While some mothers had access to bilingual providers in primary care, most mothers did not report having bilingual specialty or emergency department providers.

Though mothers reported on experiences in varied health care settings, 6 themes emerged relevant to pediatric care across settings: the “battle” of managing language barriers, preference for bilingual providers, negative bias toward interpreted encounters, “getting by” with limited language skills, fear of being a burden, and stigma and discrimination due to language barriers. These themes were reflected in the choices mothers made in managing language barriers, reflected known limitations in the health care system for appropriately addressing language barriers, and the larger social context of language barriers.

The “Battle” of Managing Language Barriers

Many mothers characterized their health care encounters as a “batalla” [battle]. Some mothers used that exact term, while others described similar experiences, but did not apply the same language. When characterizing their experiences as a “battle,” participants were more often describing health care in non–primary care settings. Many mothers reported more consistent access to language accommodation either via bilingual provider or interpretation in primary care (Quote 1; see Table 2). Most mothers, however, reported a staff member, like a nurse, frequently served as an interpreter. When discussing non–primary health care, mothers also described “battling” to bring an interpreter and to make appointments (Quote 2).

Table 2.

Participant Experiences.

Theme 1: The “battle” of Managing Language Barriers

Mothers described frustration obtaining care and communicating during encounters due to language barriers and commonly employed the term “battle” to describe their experiences.

Quote 1

Te digo, donde son los problemas a veces es en el hospital, que, que batalla uno por, de hecho pues tienes que llevar un intérprete. … Ya en un problema grave pues sí en el hospital porque ahí tienes, es raro que haiga alguien que hable, casi nadie habla español, o sea.

I will tell you, usually, you will find problems at the hospital, that, you battle for, the fact that you have to bring an interpreter. … The bigger problem is in the hospital because you have, you hardly ever find someone, I mean, that can speak Spanish.

Quote 2

Pero lo que estoy batallando por la cita de mi hija porque, la recepcionista no hablan español y hay, no me dicen. … Me dicen, espérame voy a ir a hablar a tal línea para … ver si hay alguien que hable español. Y acá no, no más dicen que no, y que no, y que hable tal día, y no, no. Ya tengo toda la semana pasada que traté de … hacerle la cita a la niña y ya el lunes va a ser … no sé cómo lo hago.

But what I am battling with is the appointment for my child because the receptionist doesn’t speak Spanish, and they don’t tell me. … They ask me to wait while they try to find someone who can speak Spanish, and over here they just tell me no, and no, and call back such day. I tried all last week to … make that appointment for her, and it’s going to be Monday. … I don’t know how I do it.

Theme 2: Preference for Bilingual Providers

Mothers frequently described a strong preference for bilingual providers because it facilitated communication and improved understanding.

Quote 3

Sí, nos entendemos mejor de doctor a persona que de intérprete a persona, porque a veces cuando Ud. necesita que alguien diga algo, no lo dice como lo que Ud. siente, un doctor a lo mejor sí te lo va a entender como lo que tú le estás diciendo.

We understand each other better from person to doctor, than from person to interpreter, because sometimes when you say something you need, they don’t say it like I would say it, and the doctor would probably understand better how you are saying it.

Quote 4

El doctor creo que te … te explica, te dice mejor la… te habla mejor que cuando te están traduciendo…

I think the doctor … explains better and tells you better than when someone is translated …

Theme 3: Negative Bias Toward Interpreted Encounters

Mothers were negatively biased toward interpreters even if they had not experienced an interaction with poor quality interpretation.

Quote 5

Hazte cuenta que a veces también es que yo le digo una cosa a ella y ella lo dice de otra, o sea como que no, o que no le entienden I realized that sometimes I say something to her, and then she would say something else. It is like they don’t understand you.

Quote 6

Porque el doctor hablaba con ella, y le … duraba el rato platicando y a mí me decía cualquier dos, tres cosas, no, no le dijo no más eso …

Because the doctor would talk to her … for a long time, and she would only tell me two or three things, and she would say that was the only thing he said …

Quote 7

Oh, que a veces no le dicen todo, pienso yo. Para acortarlo pues a veces más.

Oh, I think that sometimes they don’t tell him everything. To make it shorter sometimes.

Quote 8

[La interprete] no me traducía bien… [la interprete] me decía dos o tres cosas, y el doctor hablaba mucho rato con ella, y no me, eso no me gustaba.

[The interpreter] wouldn’t translate right … [The interpreter] would tell me two or three things, but the doctor would talk to her for a long time, and I didn’t like that.

Quote 9

Lo podría decir directamente sin alguien que intervenga y que a lo mejor no entiende lo que tú estás preguntando o… o lo traduce de diferente manera, no sé, o sea, el hecho de que esté una persona en medio sí afecta.

I would be able to speak directly with him without having someone to intervene or that can’t understand what you are asking them or … or will interpret a different way, I don’t know, I guess, the fact that someone is in the middle, it affects you.

Quote 10

Yo decía, yo la llevo [la interprete de la comunidad] para que le explique mejor y a fin de cuentas medio le decía [al doctor] lo que yo le decía, entonces dije no, pues no. O sea realmente para llevar un intérprete, digo, pues …

I keep thinking, I bring her here so [the paid community member interpreter] can explain it better but she barely said to [the doctor] what I said. I say well in reality why take an interpreter …

Theme 4: Getting By

Many mothers reported Spanish-speaking providers, but with further elaboration described that the providers were actually “getting by” with minimal Spanish skills.

Quote 11

Yo le dije a la doctora que le hacían falta, a señas, a no señas yo le dije que le hacían falta sus vacunas.

I told the lady doctor that she needed some [immunizations], by hand gestures.

Quote 12

Él no habla español y yo no hablo inglés, pero sí lo entiendo.

He doesn’t speak Spanish and I don’t speak English, but I do understand him.

Quote 13

Siempre más o menos le entendí.

I always understood more or less.

Quote 14

No puedes entender … a lo mejor la recomendación que se está dando … o a lo mejor hasta el medicamento, cómo debe de dárselo, bueno viene la instrucción en la medicina pero …

You can’t understand … at best what he’s recommending … or even the medication, how to administer it, I know the instructions are listed on the medicine but …

Quote 15

Entonces cuando … lo iban a operar, que fue el día de la operación, una persona que hablaba español, ella nos dijo si, ya ves que tiene uno que firmar de lo que le van a hacer ese día, y me dijo le van a quitar anestesia y le van a poner los tubos en los oídos, y le dije yo, a mí no me dijeron nada de tubos, y ella se quedó así como, cómo no iban a saber, ¿verdad? … Ya estaba todo listo para la operación, pero yo nunca he sabido realmente por qué se los pusieron, ni … o sea no me explicaron.

And so … when they were going to operate him, on the day of the operation, a Spanish speaking person, asked if we knew, you know that you have sign before the procedure, and she said they were done with the anesthesia and now they are going to insert the tubes inside his ears, and I said, no one said anything about tubes, and she just stood there, how could she not know, right? Everything was ready for the operation, but I really never knew the reasons why they had to put them, and … in, in other words, they never explained it.

Theme 5: Fear of Being a Burden

Patients frequently settle for suboptimal language services due to fear of being a burden or bother.

Quote 16

Pero con ella puedes … a que … te diga … ¿cómo te diré? O sea que no estés molestando a otra persona, ¿sí me entiendes? A que ella te lo diga directamente.

But with her [bilingual provider] you can—she can tell you. … How can I tell you? You are not bothering other people, do you understand? Instead she can tell you directly.

Quote 17

Pos, cuando uno ve que si es que, cuando tiene una inquietud preguntarle, uno no sabe, pues como yo no sé decirlo, preguntarle algo—algo así. Le busca una intérprete pero uno sabe que ellas están ocupadas, pues su trabajo es otro y no …

Well, when I have a question, I don’t know, I don’t know how to say it—to ask about something or something like that. [The doctor] looks for an interpreter, but we know they are busy and that that is not their job …

Quote 18

Cuando voy sin cita, que no me toca el doctor que, habla español. Ya no lo menciono lo que, pos sí me quedé con la duda.

Yes, whenever I go without an appointment, and I don’t get the doctor that speaks Spanish. I don’t mention to him the questions that I had anymore and well I remain with doubts.

Theme 6: Stigma and Discrimination Due to Language Barriers

Mothers described episodes of discrimination or perceived racism that they attributed to their language and ethnicity.

Quote 19

Yo sé que son de aquí y yo no soy de aquí, pero no quiere decir que porque ellos son de aquí van a atender a sus hijos primero, no se trata así.

I know these people are Americans and I’m not from here, but that doesn’t mean that because they’re from here their children should be seen first. That’s wrong.

Quote 20

Tal vez con la gente latina porque [el médico] no se me acerco mucho ni nada.

Maybe [the doctor] only acts that way with Hispanics because he didn’t come near me or anything like that.

However, not all mothers had favorable primary care experiences. One mother stated, “It was a battle to say anything‥‥ I couldn’t ask why they had to get so many vaccines or why not. I battled for so long.” Other mothers described the need to switch primary care providers to maintain access to bilingual providers.

Preference for Bilingual Providers

Mothers also had a strong preference for bilingual providers compared with interpreted encounters. Mothers stated that access to bilingual providers eased communication, improved understanding, and strengthened rapport (Quotes 3 and 4). Some made sacrifices with regard to distance traveled to the clinic, or dealing with long wait times to be seen for visits, to attend clinics with bilingual providers. As one mother put it, “There’s nothing like asking in your own language.” A few mothers reported a specific desire for ethnically concordant providers, though this was less of a priority than a Spanish-speaking provider. One mother said, “If he is Latino, that would be a lot better.”

Negative Bias Toward Interpreted Encounters

Mothers’ discussion of their desire for bilingual providers was closely intertwined with their opinions about interpreted encounters. When bilingual providers were not available, mothers more commonly described using their own or a family members’ limited English-language skills or a doctor communicating using limited Spanish-language skills than interpreted encounters. Most mothers who discussed interpreted encounters reported use of nurses or other staff members serving as interpreters. Mothers reported health care encounters with interpreters could be of poor quality and could lack open communication (Quotes 5–7). One mother stated, “Oh, the information can be misinterpreted. You can’t understand … what he’s recommending … and not being able to ask questions … so you can understand what’s going on.” Mothers often described these situations as if their trust was being betrayed and the doctor-patient relationship was being undermined (Quotes 8 and 9).

Mothers expressed these expectations of interpreted encounters whether or not they had actually worked with interpreters. This mother described her fears about interpreted encounters, but admitted she had not actually experienced a negative encounter: “I wouldn’t really know what they were saying … [but] no, it has not happened to me.” Among mothers with interpreter experience, their biases seemed to relate to prevalent experiences with poorly interpreted encounters and lack of consistent availability of staff providing interpretation. Additionally, they described feelings of time pressures during office visits in which an interpreter was used. Lack of interpreter availability or poor experiences with interpreters provided by the health system led some mothers to pay an English-speaking community member to come with them to health care visits and translate for them. However, this solution was also often suboptimal, as some mothers reported poor-quality interpretations with these community members and significant cost burden (Quote 10).

“Getting By” With Limited Language Skills

In addition to difficulties with interpreted encounters, mothers also frequently described situations of providers “getting by” on limited language skills (Quotes 11–13). However, mothers did not always identify it as problematic. For example, one mother recounted, “No, I may not understand her too well … she can’t speak Spanish too well, but she tries to explain … and if you understand fine and if not, no … I don’t see anything wrong with her, I mean she’s fine.” One participant cited an experience where a doctor “would get so nervous that she would speak to me in Italian‥‥ And I didn’t want to make her feel bad‥‥ Because, she was very nice.” Other mothers did perceive problems with “getting by” and either had experienced adverse consequences from getting by or recognized the potential for them. As one mother said, “So, it is better to have an interpreter instead of trying to speak a little bit in Spanish and a little bit in English.” Another mother surmised that “getting by” could lead to misunderstanding about medications (Quote 14). Finally, one mother who had been “getting by” with a specialty care physician described an especially problematic situation when suboptimal communication had resulted in lack of knowledge about the surgical procedure to be performed that was only clarified on the day of surgery (Quote 15). Participants expressed willingness to accept “getting by” because of lack of access to certified interpreters and bilingual providers.

Fear of Being a Burden

Participants also frequently described settling for “getting by” due to fears of being a burden or being singled out because of their language need (Quote 16). One mother stated, “I understand a little, but not that much … [but] I don’t want to cause any trouble.” Common across many mothers were reports of not asking for clarification or bring up questions due to the perception it was burdensome. Mothers expressed concern about asking questions since they already felt that they had burdened providers with their language need that made communication more difficult or required the services of an interpreter (Quotes 17 and 18).

Stigma and Discrimination Due to Language Barriers

The desire to not be a burden, the negative bias toward interpreters, and the tolerance of “getting by” commonly were interwoven within discussions of stigma and discrimination due to language barriers. Mothers described several instances of humiliation and discrimination due to their language barriers (Quotes 19 and 20). One mother stated, “Honestly, I sort of have seen that there is a lot of racism at the hospitals.” She goes on to say, “Sometimes they don’t want to understand you because you are Hispanic.” Another mother stated, “They humiliate you‥‥ I feel they discriminate because they speak English.”

Discussion

Across pediatric health care settings in 2 urban areas, LEP Latina mothers experienced frustration with their health care experiences and reported suboptimal accommodation for language barriers. While mothers described some health care experiences that met their needs, this was not the norm. Mothers described managing language barriers and navigating the health care system as a “battle.” Their vivid descriptions of how frustrating the health care system could be underscore the great need to improve the care that the US health care system provides to LEP populations. Improving the health care quality, safety, and patient experience for LEP Latino populations requires improvements to language services, informed by a better understanding of the patient experiences of LEP Latino patients and families. The findings from this study highlight the complex web of health system, provider, and patient-level barriers to equitable care for LEP populations.

Mothers’ preferences for a bilingual provider demonstrate that appropriate language accommodation can result in improved patient experiences. Adequate language services have been associated with improved communication and patient experience, better health care outcomes, safer care, and more efficient resource utilization.^11–13,28 Unfortunately, the physician workforce does not match the diversity of patients in the US health care system.²⁹ In fact, in the past 30 years, there has been a decline in the number of Latino physicians.²⁹ Latino physicians are more likely than non-Latino physicians to have adequate Spanish-language skills for health care communication, and may also share a cultural background with LEP Latino patients, potentially further enhancing the health care experience.²⁹ Increasing the number of Latino physicians is critically important to better meeting LEP Latino patients’ needs.

In addition to programs that target increasing the number of bilingual Latino health care providers, care for LEP patients could also be improved by access to Spanish-speaking physicians, regardless of ethnic background. Mothers in this study did not indicate a clear priority for an ethnically concordant provider, but did strongly prefer a Spanish-speaking physician. Though research is limited, studies have not found a clear association between parent/provider ethnic concordance and improved pediatric primary care quality.^30,31 Nonnative Spanish speakers who wish to use their language skills during health care encounters, however, must have adequate proficiency. Language proficiency assessments of providers who wish to use their non-English language skills for health care communication are uncommon, and determining the threshold for adequate proficiency for safe and effective communication is challenging.^32,33 Additionally, among providers who do not meet standards for adequate proficiency, there is limited information on how health systems ensure interpreter use and how those providers can gain language skills to meet proficiency standards. Addressing the health care provider language gap may require comprehensive solutions, including providing additional incentives to providers who have adequate language skills and boosting language skills of those providers with intermediate proficiency.²⁸ Robust training programs for those medical providers who have intermediate foreign language skills may be an important way to increase the linguistic capacity of the health care workforce.^22,34

Despite efforts to improve cultural competency and access to interpreters, “getting by” on limited language skills by both the patients and the providers remains pervasive.^15,35,36 Our findings shed new light on why “getting by” remains so common and indicate that there is no quick, simple solution to avoid its occurrence in health care encounters. Mothers were inclined to “get by” even though it was not ideal. “Getting by” was appealing to mothers for many reasons. They believed it was more efficient and polite, it did not increase the burden on providers, and they perceived less stigma and discrimination if they were able to “get by.” The potential patient desire to “get-by” has implications for the way language services are offered to LEP patients and indicates that encouraging providers to avoid “getting by” may not be sufficient. Through research and quality improvement processes we need to explore how best to initiate use of interpreters. Patients may decline an interpreter if offered, but if one were present as a default, it is not known how patients may respond or if this would improve health care safety and quality.

Decreasing “getting-by” through use of interpreters requires addressing patient and family distrust of interpreted encounters. Mothers in our study had significant negative bias toward interpreted encounters, even if they had not ever used an interpreter during a health care encounter. This underscores the critical need to improve the patient and family member-interpreter relationship. Trust in the interpreter’s ability to translate information correctly has been shown to increase interpreter use among providers, but the concordance between provider trust in the interpreter and patient trust of the interpreter is unclear.^13,36 In our study, we had limited information on the type of interpreter used and providers’ assessments of the interpreter. Thus, we cannot assess what factors may have improved the patient/family experience with interpreted encounters among participants or whether participants’ lack of trust in interpreted encounters is based on only poor-quality interpretation experiences.

In addition increasing the frequency of interpreter use and improving the patient experience during interpreted encounters, we must also address the discrimination and stigma felt by LEP Latino patients and their families. There is evidence that perceived discrimination in health care among Latinos is associated with lower health care satisfaction and poor patient-provider communication, but the prevalence of this stigma is unclear.^37,38 Some prior research has found that foreign-born, Spanish-speaking Latinos are less likely to report discrimination in health care than nonimmigrants or English-speakers, but another study including immigrant Latino parents found they did perceive discrimination in the care of their child.^37–39 Since feelings of discrimination and stigma were frequently discussed among participants, more information is needed about perceived discrimination and health care quality among Latinos.

This study has certain limitations. First, we interviewed a small sample of Spanish-speaking only LEP mothers in urban communities, so our findings may not reflect the experiences of other LEP Latina mothers in the United States. Second, we only interviewed Latina mothers, so our findings may not reflect experiences of other LEP populations, though LEP populations speaking other languages report similar experiences.^40,41 Third, descriptions of some interpreted encounters involved presumed “ad hoc” interpreters (family members, friends, staff not trained in interpretation). The limited experience of participants with professional interpreters may have increased their negative bias toward interpreted encounters, but also highlights the undersupply and underuse of professional interpreters. Finally, the interview-based nature of this study may result in a negativity bias. When participants discuss their experiences, they may have more of a tendency to recall negative experiences, especially with probing, than they may otherwise.⁴²

Despite these limitations, this study contributes to filling an important gap in health services research by identifying and understanding patient perspectives on language barriers in health care. The need for this research is especially important given the large and growing LEP population in the United States. As the health care system endeavors to be more “patient-centered,” there is a critical need to better understand the patients’ perspectives on pervasive and problematic issues that contribute to health care disparities. Our findings demonstrate that decisions regarding how to manage language barriers are complex and involve choices on behalf of both the patient and provider. Current training on appropriate management of language needs primarily focuses on services and programs to be offered by health systems and choices that providers make during individual encounters. Our findings underscore the need to partner with patients and families in the process of managing language barriers from the first point of interaction with the health care system through the encounter and subsequent management. Dialogue with the family on how best to manage language barriers may be as critical as the dialogue on the health problem to be addressed during the encounter. Partnering with patients and their families may result in changes to language use policies and practices and generate needed improvements in the health care of LEP populations.