Language Equity in Health Technology
19:54 Share

I'm Yulin Xun, Associate Editor of JAMA and JAMA Plus AI, and you're listening to JAMA Plus AI Conversations. In this episode, we explore the importance of equipping clinicians with cultural and linguistic competencies to enhance patient care.

Our guest today is Dr. Pilar Ortega, a multilingual Latina physician leader. Dr. Ortega serves as a clinical associate professor of medical education and emergency medicine at the University of Illinois College of Medicine. Her research focuses on improving healthcare access and quality for Hispanic and Latino patients. Welcome. Thank you. Mucho gusto and thank you so much for including me. Wonderful.

Can you tell me about your career inspiration? What inspired you to focus your career on improving healthcare access and outcomes for Hispanic and Latino communities?

Well, you know, the focus of my scholarly work is really centered on this idea of language equity in healthcare, which is the idea that individuals of all and any language preferences can access the same quality of healthcare and communication. And the way that I got there is really through my own lived experience.

I am the youngest child of an immigrant family whose language was Spanish. And as the youngest, I quickly became, you know, the most proficient multilingual person in my family, able to communicate both in Spanish and in English. And so...

that made me a child interpreter in medical and other situations as well. I was also a cultural navigator for my family, understanding the system, the educational system, the healthcare system better than anybody else in my family. And interestingly, and today we're also talking a little bit about AI, being also the youngest in my family kind of made me the person with the greatest access to technology and advancements because of the earliest exposure. So

As I went through my education and career, I couldn't help but notice gaps in how certain resources and healthcare were designed to make people like my family, who had a non-English language preference,

able to access them. So they weren't designed for them, in other words. And so I couldn't help but notice those gaps. And so it became very clear, especially as I progressed through my career and had more positions of leadership, that I needed to make this a priority.

I love that you have a personal connection to this. So can you remember your earliest memory about being the family navigator and a barrier that you faced when you were dealing with the healthcare system?

Yeah, I remember there's a lot of potential examples. You know, I remember vividly learning English myself the first day that I went to kindergarten. And I remember the first person who was that navigator for me, who was one of my peers. And so that's the first thing, you know, that kind of comes to mind as I'm thinking about

later being that for my own family is another five-year-old child who took me under her wing and on the school bus every day would tell me some words that I would start to learn. And, you know, children being sponges, of course, very quickly, I was able to navigate that on my own. But I remember whether it was going to my own doctor's appointments myself, accompanied by my mom, I

that I would be that person who would sometimes, unless we happen to have a Spanish-speaking physician, which we did sometimes but not all the time, I would be that person who would be communicating the information back to my mom so that we could make whatever decision had to be made. And then the same when somebody else in my family had to seek health care and then I was often helping to navigate that were complete school forms or complete health forms.

Now, as a physician, have you seen a change where you have seen more linguistic competency and more access compared to when you were growing up? Well, I think there are definitely advancements, but I think we're still far behind. So I think some of the things that have changed is that we recognize that this is not something that this experience that I had growing up is not unique.

You know, there are 350 languages spoken in the United States. The most common non-English language, of course, is Spanish. But in different areas of the country, there are diverse needs and populations that come from different national origins that might have different immigration stories, different experiences in general, socially, culturally, religious, and other intersecting aspects of their identity that affect the

how they can access care. And so I think we have made some strides in recognizing that. We've made some strides in also coming together and saying, you know, hey, this is not a unique experience. We need to do things that actually on a system level make it possible for individuals of any language preference to actually not only access care, but then once they're accessing care,

that the healthcare quality that they receive is comparable and equitable so that they can be enrolled in clinical trials, so that they can actually access all of the kind of different levels of care that they might need in different situations.

And I think part of what is important to keep in mind is that it takes a lot of people with a lot of different areas of expertise to achieve this on a system level. So I think we are doing better at kind of coming together and collaborating, but that's definitely one of the things that I currently work on considerably. Well, let's talk more about that. You recently had an article in JAMA Network Open, Language Equity and Health Technology for Patients with Non-English Language Preference.

So tell me a bit about this study and how you see it could help.

Yeah, so our paper was really kind of putting forward a framework to think about how do we address issues as technology advances, as we see advancements, for example, in artificial intelligence and other aspects of technology, such as video visits. How do we make sure that populations who prefer non-English languages are able to also reap the benefits of

of these advancements in healthcare. And I think with the rapid growth in this space, in particular of AI and other digitalization of healthcare, I think there are many opportunities for these communities to actually benefit. However, what we've noticed is that unless you explicitly think about

these communities and the specific barriers that they face and their specific realities, what might happen is we might actually widen the gap and worsen disparities if we're not keeping in mind how to make those advancements accessible to these groups.

So this paper is really focused on sharing a framework to think through what are the pieces, what are the guidelines that we should have in mind as we're thinking about populations with non-English language preference to make sure that they don't get left behind. So systems are obviously hard to change and so forth. And so does this framework talk about how to incentivize and support these environments and prioritize this type of culturally sensitive care?

Yes. Incentives are a big part of it because we know that in order for systems to work and move and for people who are in the healthcare system, they oftentimes will just, there are some priorities that one has to make, right? As you are going through as a clinician taking care of patients, right?

Whatever your role is in that healthcare system, you do prioritize what things you are tackling along the way, depending on what is the accountability that is being expected of you from the system. So in the framework, we do think about what are those pieces that we need to keep in mind. So to give you an example, oftentimes when people think about populations with non-English language preference,

in accessing healthcare, they often think about calling an interpreter and that's it. And interpreting is definitely a critical piece. Interpreters are some of my closest colleagues and one of the disciplines with whom I absolutely work extremely closely. However, it's not the only aspect that needs to be considered. And there are some strategies that are more effective than others in specific situations.

So when you think about, for example, video visits, which is one of the areas that we examine a little bit more deeply in this paper, this is the type of technology use that if you don't consider the needs and realities of the population that we're serving in particular, it's

and the skills of the clinicians who are going to be seeing the patient in the video visit, then how can you really address what resources you might need for an effective communication between a clinician and a patient with non-English language preference? In other words, if you are

able to match the patient with a clinician who speaks their language, then you don't need an interpreter at all, right? If you have that confirmed proficiency from the clinician. On the other hand, if you don't have that, then you do need an interpreter and that's something you can pre-plan for. So that's one example of...

something where thinking about it ahead of time so that you can do the appropriate resource allocation and you're not disincentivizing language appropriate care, but rather you're thinking about it ahead of time, putting the resources that you need, that's actually going to save costs and save time and be a more efficient and productive workflow if you have pre-planned for it. And how can digital technologies and AI help solve some of those issues?

Yeah, that's a great question because there are many ways in which I think AI and digital technologies can definitely help in these situations. One example that I think a lot of clinicians already think about using is things that they can use for direct translation.

So for example, I'm an emergency medicine physician. At the end of every visit, if I'm discharging a patient, you provide instructions, right, for the patient in terms of their self-care at home or instructions regarding their diagnoses. A lot of hospital systems and a lot of electronic health records are

already have that information available in some of the local languages, most prevalent languages of the local population. That's a great example of something that can be done. One of the things that I would caution people about and that is in the TEKI framework that we describe in the paper is these digital advancements to actually provide rapid translations of discharge materials do need to have oversight.

Translation is not something where, you know, word for word, things are equivalent from one language to another. And studies have shown that depending on the language, the degree of accuracy actually varies quite a bit.

And so remember that AI and machine translation relies on information that is out there in the public domain. And not all of that is necessarily reliable, and it may vary in a medical context. So a word can mean something in plain language, can mean something different in a medical context. And so when you think about those potential inaccuracies, you know, I always go back and think about, well, what part of this is

being inaccurate, would I be okay with in terms of giving my patient those discharge instructions? So it's really important that any of these mechanisms that we might implement as fast ways to provide information to patients in their preferred language, we have to have that oversight to make sure that the translations are actually accurate and not making, again, disparities potentially worse for those populations.

The TEKI framework, can you explain what that is? Yes. So TEKI stands for Translation, Education, Concordant Care, Community Outreach, and Interpretation. And it basically means that these are the areas that we need to pay attention to when we're thinking about creating or designing tools appropriate for non-English language preference communities.

Thank you. And so in this techie framework, how do you think that can be integrated into medical education? You know, for instance, I think that there's a lot of discussion about how the changing population and so forth and the multitude of languages and going forward, how do you think we should best start to train our new physicians in terms of ensuring that they are fully competent in these areas?

Well, I'm an educator, so I love this question. And this is really, really key. And the E in techie stands for education. And I think the number one thing I would say is language appropriate communication is a learnable and teachable skill.

And it should come with accountability. So there are two pieces of that. One is what are we doing in education? And two, what are we doing in assessment and confirming those skills? Because currently in medical education in the United States broadly, we really are only teaching and assessing clinical skills in English and online.

with as diverse linguistically as our patient population is, and the fact that there are actually a lot of medical students, resident physicians, and practicing physicians that come in with multilingual skills. So they may have learned a language either as a heritage language, meaning that they learned it at home growing up like I did,

Or perhaps they grew up in another country and they were native speakers of another language. Or maybe they learned it in school, right, as they were growing up and they have another language besides English as a second language. So a lot of times people who have those skills, they want to use them in patient care. They tend to be motivated actually to put those skills to use with patients.

And actually, regardless of what their proficiency level is, studies have shown that clinicians end up using their skills with patients because they feel that necessity to do so. So it's absolutely critical that we teach in medical education how to...

self-assess your skills in a reliable way, how to know when it's time to request a medical interpreter, how to know how to access those resources. And then if you are eligible, then have those skills confirmed and certified so that you can be a language concordant clinician, meaning that you are directly using your language skills with patients, not that you are going to be an interpreter for someone else.

that you are directly going to be communicating in that language with the patient who has that language preference. So those pieces of education are one of the core parts that I work on. And I actually founded a nonprofit called the National Association of Medical Spanish that really focuses on that for Spanish in particular, so that we can increase the number of Spanish-speaking physicians.

Now there's been discussion of AI replacing potentially physicians in that type of way, where you have the ability for AI to do direct translation and so forth and potentially speak in all of these languages. What are your thoughts on that possibility?

Yeah, well, I think there is a role for additional research for sure on using AI to augment the accessibility of interpreting and translation. So when I say interpreting, I'm talking about that oral communication piece. When I say translation, I'm talking about the written communication piece.

Both of those are important parts of the TEKI framework. And I think importantly, both of those need additional study and both of those need that human oversight because of how contextualized medical communication is. And research to date shows that these resources, while they are advancing, are

are imperfect in terms of accuracy. And so again, I have the lived experience of having been a child in this kind of situation. But I think if one just thinks about what would I want for my parent? What would I want for my child? What would I want for my grandparent?

That is what I think about when I think about the possibility of having AI replacing that linguistic aspect of communication. You know, I think we're simply not there yet. And I think that there is definitely opportunity to explore how technology can help us

scale up the things that we do have? How can we make interpretation more accessible and more successful while also having that oversight to making sure that the accuracy is there? And I think the other aspect of it that we have to keep in mind is that as we look at doing more studies on

various ways in which technology can benefit healthcare overall, that we don't assume that non-English language preference populations are not interested in technology. And I think that's one of those assumptions that sometimes underlies the reason why we'll say, oh, this population doesn't necessarily need to be in this study and we're kind of okay with that because

we assume that this is not a population that is very comfortable engaging in technology. While it may be true that there have been historically less access to technology for certain populations, that's not the same as saying that this population is not interested in or could not benefit from certain advancements.

So I think one of the ways that we need to make sure that as technology advances, that it is equitably accessible to these populations is by engaging these populations themselves as participants, as researchers, as part of the decision makers in how the technology gets integrated. I couldn't agree with you more.

Well, thank you very much, Dr. Ortega. It was a wonderful conversation and I'm very excited to see the future where we can really have competent communication that is fully culturally together.

I completely agree. Thank you so much for having me. Muchas gracias. Thank you. I am Yulin Xuan, Associate Editor at JAMA and JAMAplus AI, and I've been speaking with Dr. Pilar Ortega about the critical role of cultural and linguistic competency in empowering clinicians to deliver best practice care.

You can find a link to the article in this episode's description. And for more content like this, please visit our new JAMA Plus AI channel at jamaai.org. To follow this and other JAMA Network podcasts, please visit us online at jamanetworkaudio.com or search for JAMA Network wherever you get your podcasts. This episode was produced by Daniel Moreau at JAMA Network. Thanks for listening.

This content is protected by copyright by the American Medical Association with all rights reserved, including those for text and data mining, AI training, and similar technologies.