Lost in Translation

Published on
March 18, 2009

Imagine if your doctor asked your 12-year-old son to explain to you that you had just been diagnosed with cancer, or asked the hospital custodian who was walking by to weigh in on your CAT scan.

Does this sound like good medical care? For the 24 million people in United States who report speaking English less than "very well," this is unfortunately a common occurrence. Even in 2009, most physicians' offices and hospitals do not have reliable systems in place to provide medical interpreting services for their patients who speak limited English. As a result, children and untrained staff are routinely asked to serve as medical interpreters - because they are readily available and don't cost anything for the healthcare provider.

But what are the hidden costs? Research shows that the impact of language barriers is pervasive and profound. When compared to English speakers, limited-English-speaking patients are less likely to be insured and less likely to have a family doctor. Once in care, they are less likely to feel engaged in their own care, less likely to understand their diagnoses and prescribed medications, less likely to be satisfied with their providers and the care they receive. On the whole, studies show that they receive poorer quality of care, and when hospitalized, stay longer and are at higher risk of serious adverse outcomes.

Some of the costs of language barriers can be quantified, particularly in the ER. One study done in a Chicago pediatric ER found that when compared to patients who could communicate directly with their physician, patients who faced a language barrier stayed an average of 28 minutes longer and required on average an additional $41 worth of testing. This makes sense if you consider that between 75 and 80 percent of medical cases are diagnosed on the basis of patient history alone. Without a clearidea of a patient's symptoms and past medical problems, physicians have to rely on additional diagnostic tests.

Language barriers are not a trivial issue. According to the U.S. Census Bureau's 2006 American Community Survey, nearly 20 percent of U.S. residents speak a language other than English at home, and 9 percent report speaking English less than very well. In certain states, the numbers are even more dramatic: 43 percent of Californians speak a language other than English at home, and one in five are considered limited English proficient, suggesting that they would benefit from interpreter services when seeking healthcare.

There are some who would say, "Well, patients should learn English." Patients should learn English - not only does it help them in the health care system, but studies show that being able to speak English results in as much as a 25 percent increase in income for workers. And it turns out that our immigrant patients are learning English, and at a faster rate than ever before in our country. But as a 2007 New York Times article noted, the demand for public and low-cost English classes far outstrips supply, with some waiting lists as long as two years. The main problem with this "solution" is that patients can't predict, postpone, or plan when they will have appendicitis, come down with pneumonia, or be involved in an accident. The health care system needs to have something in place for patients who haven't yet mastered English when they need healthcare.

There are others who advocate that physicians should learn another language. Based on the proliferation of books and classes on medical Spanish, it appears that many health professionals agree. Studies show that limited-English-speaking patients who have access to a bilingual clinician are more satisfied with and better understand their care. As a result, many medical schools now offer courses in medical Spanish. But connecting and communicating with patients takes more than a rudimentary understanding of their language. Most physicians I work with know how to say "¿Donde le duele?" (where do you hurt), but may be baffled at the answer they receive. Even if all physicians were fluent in Spanish, there would still need to be a system for the other 300+ languages encountered in the United States.

And that is where interpreters come in: someone who speaks both English and a second language, who can assist a physician and her patient in communicating. However, all too often because of convenience and cost, the medical interpreter is a child, a janitor, or even another patient who happens to speak the same language as the patient. Studies that audiotape and analyze such "ad hoc" encounters show that these interpreters are reliably unreliable: they routinely omit, add, condense, and alter the content in interpreting, so that what the physician says is not what the patient hears, and vice versa.

This should not be surprising to anyone who has ever tried to be an interpreter. Not only does interpreting require fluency in two languages (and in health care settings, additional familiarity with medical terminology and concepts in two languages and cultures), but also the ability to convert language bi-directionally and manage a three-way conversation. This is a complex skill set that requires training and practice. Over the past decade, many health care providers and organizations have become aware of the pitfalls of using untrained interpreters. The question is, who will pay for trained interpreters?

There is no easy answer to this question. In a 2002 study, the federal Office of Management and Budget estimated that the additional cost of providing trained interpreter services for all ER, hospital, physician and dental visits across the country would only add an additional $4.04 to the average cost of $856 for a health care visit. But that $4.04 has to come from somewhere. Ironically, public hospitals and community clinics - whose patients are largely uninsured or poor - have been national leaders in developing robust interpreter services programs for their patients, while many for-profit health care organizations have lagged behind. San Francisco General Hospital, where I work, spends more than $1.3 million annually on interpreter services for our patients.

The cultural and linguistic diversity of our country continues to grow. Without investing in interpreter services, we cannot ensure clear communication between physicians and their patients. And without communication, we might as well be practicing veterinary medicine.

A few ideas worth pursuing

Here are some stories I would like to see written:

• The impact on patients of SB853, which became effective on January 1, 2009 and requires interpreter and translation services for insured patients in California.

• How pharmacies deal with patients with limited English proficiency. Walgreens can print prescription labels in multiple languages. Why can't everyone else? A group in Sacramento is trying to pass a law that would require standardized and translated labels on medication bottles.

• How new technologies can enhance access to interpreter services. As an example, San Francisco General Hospital is using videoconferencing to provide medical interpretation services.

• How children of non-English speakers and other ad hoc interpreters feel about their experience as interpreters.

• How bilingual are bilingual physicians? What efforts are being made to train and test physician language skills?

Alice Huan-mei Chen, M.D., M.P.H., is the medical director of the Adult Medical Center at San Francisco General Hospital and an assistant clinical professor of medicine at the University of California at San Francisco.