3

Issues and Challenges

In the workshop’s second panel, Alicia Fernandez, professor of clinical medicine at the University of California, San Francisco, and an attending physician in the General Medical Clinic and the Medical Wards at San Francisco General Hospital, provided an overview presentation on the issues and challenges for integrating health literacy, cultural competence, and language access services. Following her presentation, Winston Wong, medical director for Community Benefit, Disparities Improvement, and Quality Initiatives in the National Program Office at Kaiser Permanente, moderated a discussion among Fernandez, and the members of the first panel, Dennis Andrulis, Michael Wolf, Guadalupe Pacheco, and Wilma Alvarado-Little.

AN OVERVIEW OF ISSUES AND CHALLENGES¹

Alicia Fernandez began her remarks by presenting four cases to provide a common reference for the later discussion. The first case, illustrating the challenge of interpretation versus communication, involved a 62-year-old, Spanish-speaking woman who was newly insured under the ACA and was seeking care at the university-affiliated medical center after a recent heart attack. Using a phone interpreter, her young, earnest, and wonderful physi-

___________________

¹ This section is based on the presentation by Alicia Fernandez, professor of clinical medicine at the University of California, San Francisco, and an attending physician in the General Medical Clinic and the Medical Wards at San Francisco General Hospital, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.

cian begins to tell her about the merits and drawbacks of combined anti-coagulation therapy using clopidogrel and aspirin, emphasizing the higher percentage bleed with combination therapy over single-agent treatment on the one hand, and the better protection conferred by combination therapy and low chance in absolute numbers of a major bleed on the other hand. After 4 minutes, he finishes by asking her which drug she wishes to take. Thoroughly confused, the patient looks at him and says, “Si?”

In the second case, illustrating the challenge of talking to the right people, an oncology team is conducting rounds at a teaching hospital and is happy to catch its elderly Cantonese-speaking male patient alone. He has had family continuously at the bedside, but the team needs to inform him of the pathology results showing lymphoma, and obtain his consent to initiate chemotherapy. A team member pulls up a speaker phone and calls for a professional interpreter. Impatient with the arrangements, the lead oncologist starts in: “I am glad we can talk with you now. We need to tell you about your cancer and have you decide on the treatment.” The patient turns away and barely responds.

The third case, illustrating how interpretation can go awry, involved the instructions that a member of the health care team was giving a mother regarding how to give her baby an antibiotic. “Please give the baby 1 teaspoon of the antibiotic three times a day for 7 days. Give it to her even if the fever is gone,” explained the well-meaning pediatrician. “Dele al bebe una cucharita de antibiotico tres veces al dia por siete dias, aunque ya no tenga fiebre,” the interpreter says. There is only one problem—what is a cucharita?

Case four, involving use of the wrong word, has a physician telling a patient, “You can enter hospice, and they will help your family take care of you.” The interpreter says, “Puede entrar al hospicio, y ellos ayudaran a su familia a cuidarlo.” The problem is that while hospice is a program for terminally ill patients focused on palliation of symptoms, a hospicio is a home for abandoned children or the poor elderly.

These cases, and the many others like it, are not egregious. “This is bread-and-butter clinical medicine in an extraordinarily and increasingly complex clinical panorama of care, and yet they illustrate the tip of the iceberg of how difficult it is for patients and clinicians to overcome the language, literacy, and cultural barriers to achieve effective, equitable patient-centered valuable care,” said Fernandez. To illustrate why this is the case, she started by providing some statistics. Using the strict definition of a low English proficiency—those who report speaking English “not well” or “not at all”—there are about 14 million people living in the United States, 11 million of whom speak Spanish. Seventy percent of the 14 million people with low English proficiency have less than a high school education, and 6.6 million have family incomes under the federal poverty line. In addi-

tion, many if not most of these individuals are less acculturated to U.S. health practices, including practices around informed consent and having a dialogue with their physicians. There is also a diversity of views in the low English proficiency population about how a patient and family should make decisions, some of which Fernandez said she was hinting at in her story about the elderly Chinese man with lymphoma.

Patients with low English proficiency, she explained, report more medication errors (Wilson et al., 2005), have a lower level of understanding of prescription labels (Masland et al., 2011), are less likely to know the medications they need to take when discharged from the hospital (Karliner et al., 2012a), and are less likely to know about follow-up appointments (Karliner et al., 2012b). Language barriers, she added, have been implicated in patient safety (Divi et al., 2007) and are associated with worse outcomes in diabetes care (Fernandez et al., 2011), though not with patients who have had a heart attack (Grubbs et al., 2008) or other conditions for which care is heavily regimented. “There is widespread agreement that things are not working,” said Fernandez.

Fernandez and Yael Schenker from the University of Pittsburgh Division of General Internal Medicine conducted a national survey of interpreters on their experiences with end-of-life discussions. These were experienced, certified interpreters, yet only half reported that the discussions usually went well. Some 80 percent of the interpreters surveyed said they would like additional training in end-of-life interpreting, and 81 percent reported that physicians need more training in working with interpreters in general, not just on end-of-life discussions (Schenker et al., 2012).

Discussing the common challenges to overcoming literacy, language, and cultural barriers, Fernandez said that determining the health literacy of a patient can be challenging, particularly when physicians do not take a social history that might tell them that the patient sitting in front of them has less than a high school education, is a highly trained engineer, or is someone who can truly understand risk percentages. As a result, information sharing can easily be too much or too little, she said, or can go on for too long when all the patient wants to know is when can he or she go back to work.

Another challenge is the belief that equitable care means the same care and the offer of similar information to all patients. As examples, Fernandez noted that some physicians might believe it is equitable to offer genetic counseling to all pregnant patients regardless of cultural background, or to let the patient choose between two different medication regimens regardless of their level of health literacy. On the other hand, she added, substituted judgment or paternalism can rob patients of agency and autonomy, yet an emphasis on shared decision making can leave patients feeling confused, left alone, or deprived of professional services (Penchaszadeh, 2001). A final

common challenge, said Fernandez, is that the cultural value of individualism is seen by Americans as universal, yet in many cultures decision making is never done alone but involves the family.

The current standard of care for patients with low English proficiency is to use a professional interpreter, whether that be in person, by phone or video link, or through the use of bilingual staff, but there is no data about how often conversations with hospitalized patients with low English proficiency include an interpreter. What is known is there is no mandate for physicians to be trained on how to work with an interpreter, and there is no testing required to demonstrate proficiency in working with an interpreter. Her institution, a recognized leader in medical education, did not have mandatory classes in interpreter training until 2015, she noted, and added that a few hospital systems are starting to test and certify bilingual physicians. “This is an important route if we want physicians to speak directly with their patients without the use of an interpreter,” said Fernandez.

The underuse of interpreters and the use of inappropriate interpreters, such as untrained staff or family members, is one of the biggest barriers, she noted, but so too is the fact that so few physicians and other members of the health care team are trained to work effectively with an interpreter. The fact that interpreters are trained to value a role as language conduit rather than as cultural advocate can be a barrier, and Fernandez wondered if interpreters should be trained to speak up when they sense that the patient and clinician are not understanding each other.

Another health system barrier is the lack of some means to facilitate family meetings with the clinical care team. She experienced a rather simple solution to this last barrier when she went to Argentina when her father was hospitalized there. When she asked to speak with her father’s doctors, she was told to come to “doctor time,” the twice-daily, hour-long periods set aside for families to meet with their loved ones’ physicians, ask questions, and even bring along a family friend who is a physician. “This kind of simple solution is not integrated within the U.S. health care system,” said Fernandez.

These barriers, she said, lead to the situation demonstrated in multiple studies that patients with low English proficiency have lower levels of comprehension, patient satisfaction, and trust in their physicians when they receive language-discordant care (Schenker et al., 2010) (see Table 3-1). Studies also show that while interpreters provide a valuable service, the best situation is to have physicians who speak the language of their patients. In one study Fernandez conducted using data from Kaiser Permanente in northern California (Wilson et al., 2005), she and her colleagues found that even patients who are treated by physicians who use trained interpreters and certified bilingual staff still feel as if they are treated poorly because the physician is not showing them respect by speaking their language, a

TABLE 3-1 Language-Discordant Care Leads to Suboptimal Communication

NOTE: LC = language concordant, LD = language discordant, LEP = low English proficiency.
SOURCES: Schenker et al., 2010, presented by Fernandez, October 19, 2015.

finding that Fernandez said did not vary according to language. Another study she cited (Rivadeneyra et al., 2000) found that communication is likely to be more patient centered when delivered by a language-concordant physician compared to an interpreter. Observational studies have found that interpreter-mediated conversations, compared to those involving a language-concordant physician, have fewer open-ended questions and less elicitation of questions and values, and one study has shown that patients cared for by language-concordant physicians can have better clinical outcomes (Fernandez et al., 2011).

With regard to what can be done, Fernandez offered several possible solutions to the challenges and issues she described. One solution is to structure health systems to require professional interpreters for patients with low English proficiency, and, at a minimum, every hospitalized patient with low English proficiency should have at least one conversation per day with the treating clinician mediated by an interpreter if the clinician does not speak the preferred language of the patient. This requirement, she said, could be integrated easily into a patient’s EHR. “This is not rocket science,” said Fernandez. “It does not require a single payer. It does not require a full reassessment of the U.S. health care system. This is an easy floor to establish.”

Two other obvious solutions, she said, are to do a better job training interpreters in high-risk communication scenarios such as end-of-life discussions and genetic counseling and to train not only physicians but all member of the health care team on how to work with an interpreter. Such training should be incorporated into medical and nursing school curricula, for example, and in all residencies. Clinical care team training should also incorporate skills in eliciting patient questions and concerns that reflect a

patient’s cultural background. Fernandez also called for better enforcement of the requirement that all translated materials be written at the fifth-grade level. Finally, she said, the workforce needs to be diversified. “At the end of the day, it is easier to ask people appropriate questions when they know from their own experiences and those of their family members how difficult it is to be a patient in poverty,” said Fernandez. She also recommended that common foreign language skills should be a plus or even a requirement for medical school admissions. She noted that some health systems, such as Kaiser Permanente in Southern California, are facilitating language matches for patients and providing differential pay for language skills. Some systems, she said, are investing in interpreter technology, but that technology, she cautioned, is only good if it is used with the recognition that patients are likely to have low English proficiency and health literacy.

Health systems are starting to experiment with ideas such as “teamlets” that match a primary care physician with language and culturally competent medical assistants or creating modules or grouped care. Health systems are also using language-appropriate health coaches and promotoras with some success. All of these approaches have to come to scale, however, and the requirement for language access services needs to be incorporated into the EHR. All of these ideas also need to be paid for, and while they certainly provide a return on investment in terms of equitable care, Fernandez noted that she performs many tasks as part of routine medical care for which she is not asked to justify a business case. “I think that there will be a return on investment in terms of a healthier community, but for now we need to hold the line around patient-centeredness, safety, equity and simply doing the right thing,” said Fernandez. She said she believes there will be a future in which the language barriers within hospitals will be broken down (see Table 3-2), but realizing that future will take a great deal of time. She noted in closing that she has the privilege of working at San Francisco General Hospital, where language, literacy, and culture are inexorably entwined, but that she sees the difficulty patients experience when they move to other systems.

MODERATED PANEL DISCUSSION

Winston Wong opened the discussion period by saying that he was struck by the fact that “we have tended to look at these cases from the perspective of being the provider, but maybe it is important for us to go to a place where we have felt marginalized in the discussion about something important, whether it had to do with a medical issue or a social family issue, and where we felt alone and disempowered. That might give us an understanding of the interventions that need to take place.” He then added a fifth example to the opening cases that Fernandez presented. In

TABLE 3-2 The Present and Future of Language Barriers in Hospitals

NOTE: VMI = Videoconferencing Medical Interpretation.
SOURCE: Presented by Fernandez, October 19, 2015.

this instance, he had a patient who was a Cantonese-speaking, first-time mother-to-be who was being evaluated for thalassemia, a genetic disorder that is more prevalent among Southeast Asians. Indeed, this woman, a recent immigrant from rural China with perhaps a sixth-grade education, did have a mutation that put her baby at risk for having serious sequelae, and Wong, who was the supervising physician, and the rest of the clinical staff were trying to communicate to her what the issues were with regard to carrying a child that might be seriously compromised. He recalled walking by the exam room and seeing the genetic counselor from Children’s Hospital in Oakland working with an interpreter to explain dominant and recessive genes using a pile of pegs and being impressed that these professionals were truly trying to enable this patient to be empowered. Wong noted that because this woman had recently emigrated from China and had a fairly low level of education, he perceived at the time that much of her value to her family and community was being a mother who could bring a first child into the family. “That was critical for her sense of being valued and being part of a bigger set of community propositions and social value,” explained Wong.

After telling this story, Wong asked each panelist to identify which one of the five cases they heard struck a chord, explain why, and talk about the critical issues that came into play in that case. Wolf said the scenario with the Cantonese man is something he saw frequently in a study he conducted some 15 years ago where he was trying to understand how to start a conversation with a patient when little is known about what was said previ-

ously or about what is going on with that patient. He then said that case and the one involving the discussion of different medication regimens and shared decision making both point to the difficulty of interpreting complex elements of medical jargon, pathology, and disease in the context of literacy, language, and cultural difference, and breaking them into “digestible units” for a family and patient to understand. This is particularly challenging given that each situation is unique, and what these two cases illustrate to him is the need to better support providers, not just through medical education and training, but with point-of-care support to help them navigate the information needs of each specific patient. Such an approach, he said, would be low-hanging fruit that health care systems could address while taking longer-term steps to increase workforce diversity and hire clinical care staff that is fluent in various languages. One possible way to provide support would be to leverage the EHR and use it to provide simple words and straightforward explanations that would be appropriate for a specific patient based on the information that Fernandez suggested should be included in a patient’s record.

Alvarado-Little, commenting on the case involving the Cantonese man being given a lymphoma diagnosis, said that too often information on a serious disease such as cancer is conveyed too rapidly, regardless of language, for a patient to assimilate and use to make a rational decision. With cancer in particular, patients usually stop listening after hearing the word cancer. She also noted that in some cultures, talking about cancer is taboo or at a minimum a very sensitive subject, and in some cultures it is important to consider the provider’s gender when speaking about anything relating to childbearing. “Providers are doing what they have been trained to do, and then here comes this health literacy, cultural, and linguistic piece being added to situations that are complex in and of themselves,” said Alvarado-Little. The five cases also made her think about the spiritual and nonverbal components of giving patients a serious and perhaps sensitive diagnosis.

Pacheco was struck by the limited time that physicians have to deliver a diagnosis and prescribe a treatment and how that time can be compromised by issues of literacy, language, and culture. He also commented on the crucial role that empathy plays in dealing with serious illness, particularly cancer, and how challenging it can be to convey empathy when language and cultural barriers exist. The cases presented also raised in his mind questions about how to navigate from diagnosis to treatment, how to provide patients with the information they need to understand all of the steps in their treatment, and how to involve in those discussions family members who may provide a strong support structure for the patient within that family’s culture. For Pacheco, the case Wong presented highlighted how the first birth in a family is such a milestone in so many cultures and how receiving news that something may not be right with the child can affect

not only the mother- and father-to-be, but also the extended family, which may believe that its members did something that caused this particular health outcome. In this particular case, said Wong, the mother-to-be could feel the baby moving inside her and asked him how he could tell her that her unborn child was going to have a bad outcome when she could feel its viability in her body.

Fernandez then asked the other panelists if they could talk about some structural solutions, separate from workforce diversification and training, that would require either payment reform or new regulations. One example, she said, could be certifying providers for demonstrating proficiency in working with an interpreter or being conversant in a second language. Wong added that it would be useful to consider structural solutions that go beyond the triad of doctor–interpreter–patient to include the entire multidisciplinary team that cares for patients.

Wolf suggested that the point of care might be the place to start looking at structural solutions because in his mind those solutions would take hold more quickly. He commented that when the discussion turns to provider training, the claim is made frequently that training does not seem to have a lasting impact on provider behavior, and one way to address that issue is to include provider behavior in meaningful quality indicators that have regulatory or payment consequences. He expressed the frustration that he said many in the health literacy field experience about how much longer it will take to convince regulators and payers that the evidence base is sufficient to mandate certain best practices. He reiterated earlier comments that health information technologies such as EHRs should be used to support these best practices and to demonstrate that implementing them does not take as big of an investment as many health systems believe. Returning to the analogy of nesting dolls he used in his earlier presentation, Wolf said he sees culture being the biggest of the three, then literacy, then language when it comes to cost. “Many health care systems view the language piece as the most burdensome in terms of cost, whether that is right or not,” he said, and he worries that efforts to integrate these three components may fail because of some “inappropriate thinking” about the investment required.

Pacheco said that implementing structural solutions requires a systems approach and that starting with point of care and continuum of care are key places to begin such an effort. To take a systems approach, quality indicators and data are essential, and he believes that each of the cases presented is a data point. He agreed that health information technologies can collect the necessary data to inform quality indicators and that execution is now the issue. Upcoming programs from the Centers for Medicare & Medicaid Services (CMS) offer the opportunity, he said, to embed the quality indicators that this community wants and tie them to reimbursement.

Alvarado-Little called for more education about and enforcement of existing policies, such as Title VI of the Civil Rights Act or New York State Executive Order 26, and to revisit those policies to see where they can be improved and strengthened. Pacheco noted that the EHR could be used to capture the data needed to determine whether health systems were meeting regulatory requirements. What is needed, he said, is to determine the new data elements required to develop patient-specific profiles relevant to health literacy, cultural competence, and language access. Once collected, these data could also help make a business case by demonstrating the value of providing patient-centered quality care.

Alvarado-Little also commented on the resistance she gets whenever she talks about cultural competence or interpreting services because some of the requirements are thought of as unfunded mandates. However, federal agencies take the position that anyone accepting federal funding agrees as part of the funding agreement to provide language access services and meet all regulatory requirements concerning literacy, language, and culture. If that is the case, she said, then this community should look hard at raising awareness of existing regulations, calling for better enforcement of those regulations, and advocating at the state level for state legislators to take advantage of the incentives that exist to move initiatives forward.

At the same time, she said, this community could educate and work to empower communities and providers to advocate for health systems to meet all existing requirements. She noted that medical students or residents often do not learn about these issues until they are put in a situation where they require an interpreter and at that moment are unsure of what to do. As an aside, she noted that it is important to teach students about body language—someone may not understand English or be able to hear, but they see someone sigh or roll their eyes without the need for an interpreter.

Wong then asked the panelists to put themselves in the position of a hospital administrator or director of patient services. What metrics would they need to put before their boards that captures the confluence of language access, health literacy, and cultural competence of the sort that arose in the five cases presented in this session? Fernandez thought that patient-reported metrics, such as comprehension, satisfaction, and trust in the provider and system on a per-encounter, per-hospitalization, or per-procedure level, might be the best bet. Wolf agreed that patient-reported outcomes would be key measures that would provide a more detailed picture of the quality of care being delivered than simply looking at readmissions and other currently collected information. He cited as an example, a study he conducted on medication adherence in which patients were asked to demonstrate how they understood they were supposed to take their medications. A high prevalence of those patients did not understand what they were supposed to do. Using simple measures such as readmissions does not provide a clinical signal

that this lack of understanding exists. “I would love to have these patient-reported metrics in place,” said Wolf. “The question is how do we that.”

Wolf also suggested that appropriate metrics could serve as the basis for a value-based model that bundles reimbursements for services related to literacy, language, and culture into payments for treatments. For example, Wolf said he is building a reimbursement model for the new hepatitis C treatments that includes measures on the proper use of these medications. Services supporting adherence to the medication regime could be embedded in that kind of bundled reimbursement.

In addition to using the metrics Fernandez and Wolf proposed, Alvarado-Little suggested involving risk management and patient relations as partners in these discussions. She noted that one of the tenets of the CLAS standards is to provide patients and health care systems with the ability to address grievances and complaints. The metrics Fernandez suggested could be used to determine if the CLAS standards are being met and if grievances and complaints are handled well. Other steps Alvarado-Little suggested were to ask the interpreters and translators how best to capture this information and ask them about the role they play in creating partnerships with other aspects of their organizations. “Once you open that door with interpreters and translators, many of them are happy to share that kind of information,” said Alvarado-Little.

Before opening the discussion to the workshop attendees, Wong made two points. First, use had not been discussed as a measure of how effective providers are at communicating what patients need and what the health care system asks them to do. Second, there was little discussion about waste, though there is certainly waste in the system because of poor communication, and reducing waste could be part of a business case.

DISCUSSION

Steven Rush from UnitedHealth Group remarked that the discussions about interpretation focused on the words being said, and he wondered to what extent interpreters can insert empathy and feelings into their work. Alvarado-Little responded by saying that when she interprets, she uses the same inflection in her voice as she heard in the provider’s voice in order to deliver a message in the spirit and with the meaning intended by the provider. For example, if a provider says, “I’m sure you don’t understand that,” she would interpret the message that way instead of “I’m sure you didn’t understand that.” However, if she senses the atmosphere in the room is getting spirited because not everyone in the room agrees with whatever message they are receiving she is going to use what is called the “luxury of two seconds,” the time she has to figure out how to interpret a message without tainting the outcome. If during that time she decides that a

phrase she is about to interpret is going to damage the provider–patient or provider–family relationship, she may decide to change the exact wording of her interpretation with the knowledge that she will be held accountable according to her code of ethics. Afterward, she will have to meet with the provider and explain why she changed the wording and altered the tone of that message. “There are some complexities that are nuanced, but what hits my ears is what I am obligated to interpret,” said Alvarado-Little.

Rush also said that he agreed that the first year of residency is too late for a young clinician to first learn to work with an interpreter. Other members of the clinical care team, including pharmacists and nurses, need to be involved in interpreter training as well. Rush then mentioned a metric called the net promoter score,² a measure of patient satisfaction, and wondered if that metric could be tied into cultural competence. The CAHPS measure, for example, includes a health literacy component, and he suggested looking at CAHPS and other measures, such as the Healthcare Effectiveness Data and Information Set (HEDIS) and net promoter scores, to help drive the business case.

Pacheco agreed with Rush’s comments about using existing metrics and data collection structures. “Why not adapt health literacy, cultural competency, and language access as key components of these metrics?” he asked.

Andrulis then asked Fernandez two questions about language concordance. First, he wondered if there were something interpreters could learn from working with language-concordant providers that would improve their interpretation skills. Second, he asked if she had any insights on what language-concordant physicians are missing that leads to less than perfect understanding by patients. Fernandez replied to the first question by saying there is some understanding of the difference between an interpreted conversation and a language-concordant conversation, the biggest of which is that concordant conversations are more likely to elicit a patient’s values than an interpreted conversation. She explained that she sees many patients with whom she is language concordant and many with whom she works with an interpreter. In the latter case, she focuses hard on making sure that the patient is understanding the specifics of the instructions that she is providing via the highly capable interpreters with which she works. “Somewhere in there, I forget to ask the patient what he or she thinks about taking all of these medicines, for example,” said Fernandez.

From watching video and listening to audio tapes of these kinds of interactions, she knows that her concordant conversations are more patient-centered than her interpreter-mediated conversations. Becoming more

___________________

² More information about net promoter scores can be found at http://www.peoplepulse.com/resources/useful-articles/net-promoter-score-nps-implement (accessed June 3, 2016).

patient centered in interpreter-mediated conversations is a skill that she can learn and practice. She and her colleagues have been teaching that skill and are slowly improving interpreter-mediated conversations. Another difference she has noticed when interacting with a language-concordant patient is the level of social noise is higher. Social noise, she explained, is the chatting she engages in with the patient or family members when she walks into a patient’s room, the “Hi, how are you? How’s your family? Did you watch the Giants game last night?” banter that engages people. Again, audio recordings show that the amount of social noise is directly related to the number of questions a patient asks and how much information a physician shares.

Wong then commented that it would be interesting to look at the effects of cultural and literacy concordance, in addition to language concordance. Fernandez said there are some data available to show that language concordance has the biggest effect on the provider–patient interaction. Her institution, she explained, has many providers who have learned to speak excellent Spanish. “For those things we have been able to measure, it is actually language concordance that makes a very big difference,” said Fernandez. She pointed to a recent study showing that 70 percent of the patients with low English proficiency seen in an outpatient setting were cared for by members of an ethnic minority, which she said is another great reason to diversify medical schools.

Rima Rudd asked if it would be appropriate to expand the notion of culture and language to include the millions of Americans who now have health insurance for the first time and are essentially new to the culture and language of health care. “Are we thinking about orientations for these new patients to tell them something about us and our culture?” asked Rudd. “Are we going to do that for cultural competency?” She also commented on the number of different accents in the United States and the challenges that can present even for those whose native language is English. As an illustration, she told of an encounter with her son’s first-grade reading teacher who thought he might have a hearing problem. When Rudd asked why, the teacher said that when she asked him about a cah, he did not know what she was talking about and did not know the letters in the word cah. Rudd apologized for the misunderstanding and told the teacher that his father comes from California and she is from New York, they both say car, and he knows how that word is spelled. Similarly, she said, there are many people from parts of the United States, but not from other regions, who greet everyone with “Hi, sugar.” To her, these examples suggests that the concepts of language and culture need to be expanded beyond just considering individuals who are new to this country. She worries, however, that such an expansion would lose something essential that the field is focused on and that it is working hard to achieve.

Pacheco thought such an expansion would be good, and while it would present challenges, he sees them as opportunities, much like the challenges and opportunities created by the ACA and the influx of newly insured Americans. He said that when he worked in the Office of Minority Health at HHS there was a big push to increase enrollment of Latinos and other minorities. Latino enrollments lagged because there was little understanding of what it meant to have health insurance. For reasons such as this, it is still a challenge to fully implement the ACA in a number of underserved communities, said Pacheco, but there are also opportunities to work in those communities to increase access and knowledge about health and quality health care.

Wolf remarked that these discussions raise the importance of recognizing there is waste in the way the health care system operates that creates challenges for communicating well. In terms of expanding how the field looks at culture and language, he raised the issue of preconceived notions—that even though someone speaks fluent English and “looks American,” their prior experience with and knowledge about the health care system may be limited. Becoming more productive, he said, means more time to have more meaningful experiences with patients; to find out about the literacy, language and cultural issues a patient might have; to learn about their preferences and prior experiences; and to form a real relationship with the patient that is not possible today in the typical 15-minute encounter. Wolf said that while he thinks of himself as an optimist, he is pessimistic about the near-term possibilities of health care systems trying to balance payment reforms with making a greater investment in the relationship with patients. He noted there is a substantial amount of data showing the benefits of integrating health literacy, cultural competence, and language access services, which is why it is important to make the case that spending more time with patients will pay off in terms of increased satisfaction and better care. “This is what patient centered means,” said Wolf.

Andrew Pleasant wondered why nobody had mentioned an objective indicator of health status as a measure of outcomes for addressing issues of literacy, language, and culture. Fernandez said that including such an indicator can be a good idea, but only if it measures those health outcomes that are sensitive to the process of care and not to those factors external to the process of care. Wolf agreed with Fernandez and noted, for example, how much of the work he does with medication adherence has to do with socioeconomic factors and not those related to the process of care. It is important to find measures that will demonstrate improvements in outcomes that are more proximal to the intervention. Pacheco voiced a similar concern with regard to the impact of social determinants of health that affect many high-risk patients and can negate any benefits that could come from improving communication between provider and patient. “If we do not address those

kinds of issues, we are still going to be talking about eliminating health disparities for the next 20 or 30 years,” he said.

As the last comment of the session, Alvarado-Little agreed that the negative effect of the social determinants of health, such as living in a food desert or a neighborhood plagued by violence, can be tremendous. The hope, though, is that because the patient feels respected and heard and treated with dignity, a better relationship is formed between patient and provider. The hope is also that by communicating information in a way that resonates with the patient, by impressing upon the patient the need to return for additional services, then health care systems can improve compliance and the overall health of the individual.

This page intentionally left blank.