Language Barriers in the Clinic Could Harm Children's Health

Around 7,000 known languages exist across the world, each with a nuanced lexicon to describe sensations and anatomy, each with its own cultural context. When patients and doctors speak the same language, it stands to reason that they can communicate more effectively. But to what extent does that shared tongue — also known as language concordance — shape health outcomes?

“We know from numerous studies that people have better health and well-being when they receive health care, education, and other services in their native language,” says Ann Miller, principal associate in global health and social medicine in the Harvard Medical School Global Health Research Core.

A 2017 study in JAMA Internal Medicine, for example, found improved glycemic control among patients with diabetes who switched to a primary care provider who spoke their native language. A 2022 paper in the Canadian Medical Association Journal observed a lower risk of adverse events and in-hospital death, as well as shorter hospital stays among patients receiving language-concordant care. And after assessing thirty-three studies on language concordance, a 2019 systematic review in the Journal of General Internal Medicine found that more than three-quarters of the studies demonstrated improvement in at least one measured outcome among patients receiving care in their own language.  

Miller’s recent research offers clues about how language concordance may affect the health of children. Drawing from data on children across the globe who speak minority and Indigenous languages, Miller and colleagues found that speaking a language that’s different from the dominant language in their country can affect a child’s health and well-being because they are less likely to receive equitable access to education and health care.

Their study, published in Lancet Global Health in 2023, observed that children in families who did not speak the dominant language had lower scores on a global child-development measurement than children from families speaking a dominant language — even after controlling for social and economic factors. Notably, these data don’t suggest that speaking an Indigenous or minority language is responsible for developmental delays. Instead, they indicate that discrimination and lack of equitable access to health care and education for children and families in their home languages can come at a cost.

Harvard Medicine magazine caught up with Miller to learn more about how these communication gaps relate to the health of children — and how health systems can adapt to better serve them. An edited version of that conversation follows.

With all the other factors that might affect a child’s well-being and development, such as economic situation or access to educational resources, how big an influence does their home language have?

Lots of research has shown that education and health care in one’s home language is closely tied with well-being. We were interested in whether we could see an effect of language dominance on child development and well-being even when considering other influential factors like those you mention.

Our study looked at development measures based on the Early Child Development Index for more than 186,000 children, age 3 to 4 years, from forty countries.

We found that about 66 percent of children from a minority or Indigenous-language speaking household were overall on track with some key early childhood development measures, compared with 77 percent of children from households that speak a dominant language. We made sure to compare families that were similar in other ways — level of education, whether they sang with their children at home — to try to confirm that the difference we were tracking was related to home language.

Some of the biggest differences were in what parents said about how well their children use and understand words and numbers and in measures of how well children are developing the skills necessary to deal with emotions and develop meaningful relationships with other children and adults.

How did you determine which children spoke minority or Indigenous languages?

Most of the data we looked at reported on the languages respondents spoke at home, but not on whether the languages were dominant in their settings. So we had to use some new analytic techniques to answer that question.

We analyzed information from a big set of international population health databases alongside Ethnologue, an international database of languages, including measures of how widely used or how endangered a given language is.

Ours was the first study that we know of to combine big population and language datasets in this way.

What would you say to someone who looks at this study and sees it as an argument for assimilating these families into the mainstream language as quickly as possible so that children don’t get left behind?

We know that adult speakers of these languages aren’t deficient in math or reading, and that the communities aren’t lacking in rich social, cultural, and emotional characteristics. So that tells us that our study isn’t measuring the impact that speaking an Indigenous or other minority language has on a child’s development; it’s measuring the consequences of the discrimination and structural challenges that people in Indigenous and other language-minority communities face.

There’s also some concern about how the children in these datasets were assessed. Parents, not children, were asked the development questions. But regardless of who is being tested about the child’s development, if it’s in a language they don’t speak at home, or if the family is interacting with someone from an unfamiliar cultural group or community, the result may not reflect reality. The WHO Early Childhood Development Index tool has been recognized as limited. It’s being replaced by an expanded tool, but the same language challenges may affect the new one.

So what is the solution?

The gold standard of health care, education, and child development services should be to deliver care in the language of the people you’re serving. 

Several of my colleagues on this project work with Wuqu' Kawoq, a nongovernmental health organization that has provided health care in patients’ native Mayan languages in Guatemala for more than fifteen years. Peter Rohloff, an HMS assistant professor and senior author of our recent paper, co-founded the organization, which also conducts advocacy at the intersection of health care and Indigenous language revitalization. It’s important for patients to have doctors, nurses, midwives, and community health workers who speak their language.

Our study also shows that it’s not just health care workers who need to be fluent in the languages and cultures of the communities in which they work. To better serve these communities, we need more leaders, educators, child development specialists and biomedical, social science, and policy researchers to come from these communities. We also need policies supporting early education in nondominant languages.

That’s a lot of work to take on.

That’s why partnerships and collaborations are vital.

For this study, one small step we took was to have our article abstract translated into Spanish as well as Kaqchikel and K'iche', two of the Mayan languages spoken in the communities where my colleagues work.

Those of us in research and health care delivery have to consider our own communication, social, and emotional skills to make sure the work we do supports local communities and reduces disparities rather than entrenching them.

Jake Miller is a science writer in the HMS Office of Communications and External Relations.

Images: Courtesy of Maya Health Alliance | Wuqu' Kawoq (top); Gretchen Ertl (Miller)