AMA Journal of Ethics^®

Abstract

Patients with limited English proficiency (LEP) are among the most vulnerable populations. They experience high rates of medical errors with worse clinical outcomes than English-proficient patients and receive lower quality of care by other metrics. However, we have yet to take the issue of linguistic inequities seriously in the medical system and in medical education, tacitly accepting that substandard care is either unavoidable or not worth the cost to address. We argue that we have a moral imperative to provide high-quality care to patients with LEP and to teach our medical trainees that such care is both expected and feasible. Ultimately, to achieve linguistic equity will require creating effective systems for medical interpretation and a major culture shift not unlike what has happened in patient safety.

Case of a 56-Year-Old “Poor Historian” with Acute Renal Failure

The most memorable experiences of my third year as a medical student in the US were the ones in which I (the second author, CN) actually contributed something meaningful to the care of a patient. One such experience involved Mr. S, a 56-year-old Brazilian construction worker, who had recently undergone hip replacement surgery elsewhere and presented with several days of nausea, vomiting, food intolerance, and general malaise. He was found to have abnormal kidney function tests and elevated potassium. His English was fair at best, and his medical record was already thoroughly marked with the label of “poor historian.” I met Mr. S after he had been triaged by the emergency department (ED) physicians and seen by nephrology for his renal failure, with the result that a work-up was already in motion. It was a busy night, and no one had yet involved a medical interpreter in his case, so, as a medical student—despite not speaking Mr. S’s language—I thought I might contribute. I had low expectations, however. After all, multiple experienced clinicians had been unable to gain much from talking to Mr. S, and they seemed frustrated and doubted that the extra time would be worthwhile. I was not encouraged by my resident to call for an interpreter, but neither was I dissuaded, so I went ahead. During our conversation, I discovered that Mr. S had been taking high doses of meloxicam for his postsurgical pain. Not understanding what it was, he hadn’t mentioned it previously. It turned out that he had nephritis induced by nonsteroidal anti-inflammatory drugs (NSAIDS) and a bleeding gastric ulcer. I remember feeling partly triumphant and partly shocked that a student’s effort had prevented a missed (or at least delayed) diagnosis. Would this have happened had the patient been a native English speaker?

This case vignette illustrates the potentially serious consequences of language barriers for the care of patients with limited English proficiency (LEP). In this article, we discuss how it is that linguistic inequities in health care persist, how they are propagated by medical education, and what we can do about it.

Linguistic Inequities in Health Care and the Moral Imperative

People with LEP, defined by the US census as those who speak English less than “very well,” represented 8.7 percent of the US population five years and older in 2011 [1]. They have been consistently shown to receive lower quality care than English-proficient patients on various measures: understanding of treatment plans and disease processes, satisfaction, and incidence of medical errors resulting in physical harm [2-6]. These disparities are rooted in obvious communication barriers but also may reflect cultural differences, clinician biases, and ineffective systems (i.e., structural barriers) [7]. Medical interpreter services can help overcome some of these barriers, but they have associated costs—both financial and in terms of physician time [8, 9]. Title VI of the Civil Rights Act of 1964—as enforced by Executive Order 13166 [10]—requires that health care providers receiving federal funds implement Department of Justice guidance on providing competent interpreter services when needed [11]. However, this legislation has not been strongly enforced [12]. For example, one study found that 43 percent of hospitalized patients with LEP had communicated without an interpreter present during admission, and 40 percent had communicated without an interpreter present after admission [13]. In general, use of formal interpreter services (telephonic or live) is relatively low compared to the use of ad hoc interpreters (family members or other untrained individuals) [14, 15], a practice frought with potential for errors [16]. Although health systems are challenged by the volume of visits, diversity of languages, and lack of reimbursement for medical interpretation [17], physicians often have interpreter services available but choose not to use them [18]. This confluence of factors may have contributed to the suboptimal care in the case of Mr. S.

We believe that health care professionals, leaders, and medical educators have a moral imperative to address these persistent linguistic inequities (caused in part by communication barriers) by developing effective systems for medical interpretation, shifting the organizational culture, and educating health professionals.

Interpreter Services from the Physician’s Perspective

Most physicians would deem it unfair for a certain group of patients (those with LEP) to receive worse care; they might assume that unfairness is unavoidable and not take responsibility to change it, but they would not consider linguistic disparities in health care to be fair. Indeed, a large survey of residents found that 96 percent indicated a patient’s culture was an important consideration in providing care [19]. A smaller group of physicians might place responsibility to learn English on patients and point to their personal choice to come to the US and not learn the language. Regardless of one’s opinion, physicians’ standards of professionalism hold that we not base our care on judgments about patients [20]. A trauma surgeon, for example, has a moral imperative to treat the injuries of the drunken driver in an automobile accident just as he does the innocent victim.

How, then, do we prevent adverse events due to inadequate communication, such as missing a diagnosis of NSAID-induced nephritis and a bleeding gastric ulcer in the case of Mr. S or allowing a patient to take too much warfarin due to confusion about dosing [21]? Live interpreters, including professional interpreters and trained staff, are an option for large hospitals and smaller providers that serve a few predominantly immigrant populations. In the case of less common languages for which live interpreters are not a practical option, a wide variety of telephonic and video interpreter services are readily available in the more than 100 languages spoken in the US [8], but they are used relatively infrequently due in large part to a lack of organizational commitment and training of staff on the use of these services [22].

In practice, physicians may recognize that they are unintentionally providing worse care to patients with LEP but justify it in various ways [18]. In a qualitative interview study, internal medicine and emergency medicine residents in two hospitals where interpreter services were readily available blamed their failure to fully utilize these services on time constraints, the hassle of involving interpreters, and minimal incentives from supervisors, and they tended to shift responsibility to others [18]. One resident said:

I guess I feel like someone is talking to [the patient] and getting a good history at some point…. It’s okay if I kind of come in and do a little bit more of a perfunctory exam and history, knowing that I will at some point talk to them with an interpreter when time is available. But in some ways … it does not seem all that fair because English-speaking patients—I talk to them in their own language easily even on-call, even when I am busy [23].

One concerning aspect of this approach was the lack of standardization, which left decisions to individual discretion. Residents often assessed the cost-benefit ratio of using interpreter services for each interaction (the time it would take versus the perceived benefit to patient care) and reserved interpreters for particularly important conversations like goals-of-care discussions [18].

Mr. S was almost a casualty of this kind of rationalization. He had been seen by several physicians, all pressed for time and either deferring proper communication to someone else or writing it off altogether due to language barriers and the patient’s low health literacy. In the murky waters of clinical medicine and documentation, labeling the patient as a “poor historian” or having communication difficulties due to language barriers may be seen by some physicians as absolving them of responsibility for taking a complete history with a competent interpreter. In fact, it does not—the law is very clear on this point, as described above. However, it is also evident that placing responsibility solely on individual physicians, with no effective system in place to guide them, is a recipe for failure. Complex problems have complex solutions.

Medical Education and Implicit Messages about Care of LEP Patients

If residents accept and rationalize substandard care for patients with LEP, then clearly medical students, for whom residents are the most influential teachers [24-26], will learn the same attitude and approach. One of the authors (ARG) and colleagues carried out a qualitative study of medical students and nursing students to explore their experiences learning about the care of patients with LEP during their clinical rotations. Our findings described a powerful “hidden curriculum” (teaching that is not part of the formal curriculum but is transmitted through experiences and interactions) in which supervisors role-modeled an indifferent, and sometimes negative, attitude towards care of patients with LEP [27]. Supervisors often did not involve professional interpreter services or expect students to do so, nor did they teach how to effectively work with an interpreter. A hierarchy of values was conveyed to students whereby good communication, especially if it took extra time, was valued much lower than clinical knowledge and even fairly mundane tasks [27], which contravenes the “ethics of caring” [28]. One student said:

It just seemed to be an extra step or just take longer to use a phone or to call an interpreter up to the floor or something like that. It just seemed, “Oh, if we can get by with just speaking a little bit of the language or using some other form of communication, it would be easier than taking the time to make the phone calls and go through somebody else” [29].

Additionally, students’ efforts to work with interpreters were not appreciated by supervisors, and students blamed systems barriers in conjunction with time pressures for what they recognized as suboptimal care for patients with LEP. This hidden curriculum forces a moral conflict for many students whereby their own personal values (some of which were shaped during their formal curriculum) come in conflict with what is expected of them in practice.

We Can Do Better

In order to ensure high-quality, equitable care for patients with LEP, our health care system will either need to find a way to connect all patients to health care professionals who speak their preferred language or implement effective systems for medical interpretation. Numerous studies have demonstrated that language concordance improves the patient satisfaction, engagement, perceived understanding, utilization of preventative services, and outcomes of patients with LEP [30-34]. Improving the diversity of our health care workforce and expanding (and expecting) second language training would help optimize the care of patients with LEP. However, adopting this approach would be slow and unlikely to meet demand. Turning to interpreter systems, the expectations and resources for communicating with patients with LEP vary widely across health care organizations [7], and while some guidance for hospitals is available [7], we are a long way from standardized and effective systems.

We need to align our laws with our policies, and our policies with our expectations. Many physicians remember a time when hand washing before entering a patient’s hospital room was not expected or enforced, but the system has changed. Hand sanitizer is available and obvious in front of each room. Signs are posted everywhere and staff members monitor one another. The culture itself within the hospital has shifted. Something similar needs to happen to address language barriers. This will not be an easy process. It will require full commitment by leadership, consistent messaging and promotion, and policy setting and enforcement to change organizational culture. Some steps in this direction include: (1) investing in high-quality interpreter services using technologically advanced solutions (e.g., video remote interpreting) and automating the process to reduce the barriers to use, (2) training faculty and staff to use these services efficiently and effectively, (3) enforcing clear rules that remove ambiguity from the decision-making process and tracking enforcement openly, (4) creating a culture of equity in which excellent care is expected for patients with LEP as it is for all patients, and (5) sending a strong message to our medical trainees that good communication with patients with LEP is part of good clinical practice and that suboptimal care is unacceptable.

Looking ahead, as health care moves to a value-based model and patient-centered medical homes promote a team approach to care, medical interpreters may take on expanded roles. For example, they may serve also as patient navigators who help guide patients through the complexities of the health care system [35], cultural brokers [36, 37] who help to bridge the different perspectives of patient and clinician [38, 39], and safety checkers who ensure that dangerous miscommunications are caught before errors occur [7]. Interpreters will be valued not on a dollars-per-minute basis but as important team members who improve quality of care and outcomes, thereby generating revenue rather than adding cost. Ultimately, with all due respect to the moral imperative, it will be this financial equation that drives real change.