In Cantonese, the word for mother — mā-mā — is one of the first sounds a baby in a Chinese-American family learns to make. In English, it's mama, which means the same thing and feels the same way. Almost everything else about caring for a Chinese-American mother who has stopped being able to live alone gets harder when the family speaks two languages.
This isn't a problem unique to Cantonese-speaking families. The U.S. Census counts more than 67 million Americans who speak a language other than English at home, and roughly 25 million who speak English less than "very well." A meaningful fraction of those people are aging. A meaningful fraction of their children speak English as their primary language. What happens at the intersection of those two facts has a name nobody uses: the translation tax.
Where the tax gets paid
It gets paid in clinics, mostly, where 14-minute appointments don't have room for a real-time linguistic handoff. Research on family-member interpretation in clinical settings has consistently found that adult children acting as informal interpreters omit a meaningful share of clinically important content — sometimes because they don't know the technical vocabulary, more often because they're trying to spare their parent from worrying.
It gets paid at discharge, when the after-visit summary comes in English and the parent reads in Korean. The summary contains the new medication name, the dosage, the time of day, the warning signs that would mean returning to the ER. Half the family understands it on first read. Half doesn't. The half that doesn't is often the patient.
It gets paid in the group text, where the sibling in Cleveland writes "we should look into Plavix versus the other one" and the parent in Queens cannot follow the thread. So the parent calls the daughter in Queens who lives ten minutes away, and the daughter relays, badly, and the conversation that needed three people ends up being held twice, in two languages, by one tired woman.
This is the part of American caregiving that doesn't show up in AARP's trillion-dollar valuation. The hours are counted; the translation tax is paid inside those hours, invisibly, by the person who knows both languages best.
Why professional interpreters don't fix it
The right answer to "who should translate in a clinical setting" is, in every clinical guideline, a trained medical interpreter. Title VI of the Civil Rights Act requires federally funded healthcare facilities to provide language services. The Joint Commission has standards for it. HHS-OCR's updated Section 1557 language access regulations, effective in 2024, added teeth.
In practice, in a 14-minute appointment, the interpreter is on a video cart in the corner, joining late, missing the first three minutes of the visit, and disconnected by the time the doctor walks the patient to the door. The family member is still in the room afterward. The family member is the one who explains, on the drive home, what really just happened.
This isn't a critique of interpreters or hospitals — both are doing what's possible inside a system that wasn't designed for the work. It's a description of where the gap stays open even after every policy lever has been pulled.
What the gap costs
The gap costs adult children promotions. It costs aging parents accurate symptom tracking. It costs entire families their first language as a working medium of family decision-making — because once English becomes the medium of every important conversation, the parent who reads only Spanish or Mandarin or Vietnamese stops being a participant in their own care. They become its subject.
That last cost is the one that gets under-counted, because it's not financial. There is no line item for a parent who used to make the family decisions and is now decided about. There is no productivity statistic for the daughter who knows her mother is being talked around at every appointment and cannot find a way to fix it.
What changes when the workspace itself is bilingual
When the shared notes, the appointment list, the after-visit summary, and the medication changes are visible in two languages side-by-side — the mother reading in her language and the children reading in theirs — three small things happen. The mother participates in her own care plan again. The sibling who lives far away can read in the language they actually read in. And the daughter who was carrying the translation alone gets to set the burden down.
This is a small mechanical change with a disproportionately large family effect. We see it in the families using Kintaria's bilingual workspace every week. A mother in Manhattan who hadn't been told her potassium was high — because nobody wanted to upset her — sees the lab trend in Mandarin and asks her cardiologist about it herself the next visit. A father in Houston whose discharge instructions came in English watches his children, in real time, post the Spanish version and ask him to read it back so they know he understood. These aren't dramatic moments. They are the moments that decide whether the next ten years of care happen with the parent or to the parent.
What the rest of the industry could do
The caregiving software market has spent the last five years figuring out how to do bills, medications, appointments, and documents. All four are real problems. None of them is the translation tax. Building bilingual into a caregiving product is more expensive than building monolingual — you need actual translators, not machine output, and you need to maintain the parity over time. Most companies, given the cost, opt to ship English first and add languages "later." Later rarely comes.
If you are building a product for American families caring for aging parents, the question is not "should we add another language someday." The question is whether you're willing to design the product, from the first screen, for the family that doesn't all read English. Forty percent of the families in this country don't. The rest of the industry is solving for the other sixty.
The trillion-dollar figure that AARP put on family caregiving in March is the number we use to argue for tax credits, paid leave, and Medicare reform. But the number that matters inside the families paying it is closer to the one nobody measures: how often, in a typical week of caring for an aging parent, does the family decision get made twice, in two languages, by the one person who can hold both.
That is the translation tax. It compounds quietly. The good news is that it doesn't have to.