A nurse hands you a diet sheet on the way out. It says lean protein and whole grains, and the sample dinner is a turkey sandwich on wheat bread.
You have never made a sandwich for dinner in your life. Neither has your mother, who does most of the cooking. You fold the sheet, put it in your bag, and nothing changes.
That is a piece of paper that was written for somebody else, not a patient who failed to comply.
The federal government has a written answer to this, and most people have never heard of it. The U.S. Department of Health and Human Services, through its Office of Minority Health, publishes fifteen National Standards for Culturally and Linguistically Appropriate Services in Health and Health Care, known as the CLAS Standards. They describe what a clinic, hospital or health plan is supposed to do so that care fits the person in front of it.
The fifteen standards your clinic is measured against
Start with what they are not. The CLAS Standards are not a law. The Office of Minority Health calls them a blueprint for health and health care organizations, which means a hospital that ignores them is not thereby breaking a statute. The enforceable obligations come from elsewhere, and that distinction matters when you are asking for something.
The first standard, the one the other fourteen serve, is the Principal Standard. The Office of Minority Health states that an organization should “Provide effective, understandable, and respectful quality care and services that respond to cultural health beliefs and practices, languages, health literacy, and other communication needs.”
Read that slowly and notice the structure. It is not a request that clinicians be nice. It names specific things care must respond to, and it puts the responsibility on the organization rather than on you.
The remaining fourteen fall into three practical categories, which is a more useful way to hold them than a numbered list.
What the organization owes you when you walk in. This is the language group, and it is the most concrete. Interpretation and translation are to be offered free of charge to people with limited English proficiency, patients are to be told that help exists in a language they read, written materials and signs are to be available in the languages their community speaks, and the people doing the interpreting are supposed to be trained for it.
How the organization is built. A second group addresses leadership, hiring and training: that governance reflects the community served, that staff are trained on this work regularly rather than once at orientation, and that somebody with budget authority owns it. These sound like internal housekeeping. They are the reason the first group either happens or does not.
Whether any of it works. The last and largest group is about measurement and accountability: collecting accurate demographic data on who is served, folding cultural and language measures into routine quality improvement, assessing what a community needs rather than guessing, building complaint processes people can use, and reporting publicly on progress.
Three of these repay a closer look.
The demographic data standard sounds bureaucratic and is not. An organization that does not record which patients speak which languages cannot staff an interpreter line, cannot find out whether its Spanish-speaking patients are getting worse blood pressure control, and cannot be held to account for either. When a registration clerk asks your preferred language, that question is the entry point for this entire system.
The complaint standard matters because the usual grievance process is the thing most likely to fail the people it is for. A complaint form in English, online only, with a 30-day deadline, is not available to a patient who could not understand the appointment in the first place.
The community partnership standard separates real work from a poster in the lobby. Designing a diabetes program for a neighborhood without asking anyone who lives there produces a program that is technically correct and empty on Tuesday nights.
One note on sourcing: the Office of Minority Health page carrying the standards shows no publication or revision date, though the PDF it links is filed under a name indicating a June 2025 revision.
Language help is free, and federal law requires it
Here is the part with legal teeth.
The HHS Office for Civil Rights states that two federal laws require programs receiving federal health funding to provide language assistance services at no cost: Title VI of the Civil Rights Act of 1964 and Section 1557 of the Affordable Care Act, along with the regulations that implement them. OCR names the ordinary places this applies: the pharmacy, the nursing home, the emergency room, health insurance enrollment, and a routine visit to a doctor or dentist.
Two parts of that do the work. Required means it is not a favor. At no cost means you cannot be billed for it, and that you do not have to apologize for asking.
In practice, asking still takes effort, because the default is for nobody to offer. Which brings us to the most common workaround in American medicine, and the most dangerous.
Why the bilingual nephew is the wrong person in the room
Families fill the gap. Someone’s daughter, cousin or neighbor comes along and translates. It feels efficient, it feels kind, and the research on what it costs is unusually clear.
In a 2003 study in Pediatrics, Glenn Flores and colleagues recorded and transcribed pediatric clinic visits conducted through interpreters. Across thirteen encounters they counted 396 interpreting errors, an average of 31 per visit. Sixty-three percent had potential clinical consequences. Errors by untrained interpreters were significantly more likely to carry clinical consequences than errors by hospital interpreters, 77% against 53%. The untrained group in that study included nurses, social workers and an 11-year-old sibling.
The errors were not exotic. The most common was simple omission: something was said and did not arrive. In the cases the authors described, that included leaving out the question about drug allergies and leaving out how often and for how long to give an antibiotic.
A larger 2012 study by the same lead author in Annals of Emergency Medicine compared professional interpreters, untrained interpreters and no interpreter across 57 emergency department visits in Massachusetts and logged 1,884 errors. Errors with potential consequences made up 12% of those by professional interpreters, 22% of those by untrained interpreters, and 20% when no interpreter was used.
That study also found something a patient can use. Among professionals, what predicted accuracy was training hours, not years on the job. Interpreters with at least 100 hours of training made a median of 12 errors per encounter against 33 for those with less, and their share of consequential errors dropped to 2%.
This is why the standard on interpreter competence includes a line worth quoting. The Office of Minority Health writes that organizations should ensure interpreter competence, “recognizing that the use of untrained individuals and/or minors as interpreters should be avoided and discouraged.”
Minors, plainly: a child should never interpret a medical conversation. Not for a parent, not for a grandparent, not “just this once” in the waiting room. A child cannot be expected to carry vocabulary about cancer staging or medication dosing, and no child should be the one who has to say out loud that a parent has a serious diagnosis. The burden is unfair and the accuracy is poor.
Adult relatives are a gentler problem with the same shape. They edit. They soften bad news, skip the embarrassing question, answer on the patient’s behalf because they already know, and sometimes leave out the symptom the patient did not want them to hear.
Getting the professional one is still not routine. A 2011 study in the Journal of General Internal Medicine by Yael Schenker and colleagues surveyed 234 hospitalized patients with limited English proficiency and found professional interpreters were used in 17% of admission encounters with physicians and 4% of encounters with nurses, while family or friends interpreted at admission for 28%. A 2023 analysis of vascular surgery cases in Hawai’i by Uliana Kostareva and colleagues found that on the day of surgery or of signing consent, 16% of patients with limited English proficiency had documented interpreter access, 25% had no documentation at all, and 59% had a mix of family, staff and interpreters.
So ask early, ask in writing if you can, and ask again at consent.
Food advice that ignores your kitchen gets ignored back
Start with the thing that has to be said directly: no cuisine is one thing, and no community eats one way. A Nigerian household in Houston and a Nigerian household in Minneapolis may share almost nothing on the stove. Two Mexican American families on the same street cook differently, and the same family cooks differently on a Tuesday than at a holiday. Treating “Asian food” or “Latino food” or “Southern food” as a single pattern with a single nutritional profile is not cultural awareness. It is a stereotype with a nutrition label on it.
The useful fact underneath is simple. The nutrients exist in every cuisine. Fiber is in beans, lentils, plantain, collards, bulgur, barley, nopales, okra and oats. Calcium is in dairy, in small bony fish, in tofu set with calcium, in sesame paste and in cooked leafy greens. Protein shows up in everything from goat to chickpeas to eggs to queso fresco. Eating well does not require a different grocery store.
Current federal guidance acknowledges flexibility. The Dietary Guidelines for Americans, 2025-2030, issued by the U.S. Department of Agriculture and the Department of Health and Human Services, describes its framework as supporting cultural traditions, personal preferences and different lifestyles rather than dictating exact meals. What specific nutrients people need and in what amounts belongs to article 670.
What this means in an exam room is a question you can ask: “Can you show me how to do this with the food we cook?” A clinician who cannot answer may be able to refer you to a registered dietitian who can. A handout you cannot use is not neutral. It is a missed appointment that already happened.
Traditional practices deserve respect and honest evidence
Many families use something alongside conventional care: herbs, teas, cupping, massage, dietary rules tied to religious observance, a practitioner trusted in the community, a remedy a grandmother taught. According to the National Center for Complementary and Integrative Health, a 2012 national survey found more than 30% of U.S. adults and about 12% of children used approaches outside conventional care, and most of those people used conventional care too.
Two things are true at once here, and most writing on this subject manages only one.
The first: long use is history, not evidence. Saying a practice has been used for centuries tells you it is meaningful and durable. It does not tell you whether it works for a given condition or whether it is safe with a given medication. That is not a judgment about any particular culture. It is the same standard applied to any treatment, including the ones sold in pharmacies.
The second: dismissing what a family does is both disrespectful and a clinical mistake, because it guarantees the clinician stops being told. This article makes no health claims for any traditional remedy, and it does not suggest any of them replace care that works.
What is reasonably well supported is that tailoring care to the person helps somewhat. A 2021 meta-analysis in Drug and Alcohol Dependence led by Audrey Hang Hai pooled 22 trials of culturally adapted treatments for substance use among adults of color and found an overall effect size of 0.23, which the authors called promising while asking for more high-quality studies. That is honest preliminary evidence of a modest benefit, not proof that any single adapted program works.
The practical instruction is short. Tell your clinician and your pharmacist everything you take and everything you do, including teas, powders, tinctures and practices you think are too minor to mention. Some botanicals interact with prescription medicines, and some raise bleeding risk around surgery. Article 677 covers those interactions and is the place to go before you combine anything.
If you are choosing a practitioner, the National Center for Complementary and Integrative Health notes that credentials vary enormously from state to state and between disciplines. It suggests checking training and licensing with the relevant state board, asking whether the practitioner will coordinate with your regular providers, and checking coverage with your insurer first.
Trust has a documented history
Some communities approach the medical system with less trust than others, and that is not a misunderstanding to be corrected with a brochure. There is a record.
The clearest documented example is the U.S. Public Health Service Untreated Syphilis Study at Tuskegee. According to CDC, the study ran from 1932 to 1972, enrolled 399 Black men who already had late-latent syphilis along with 201 men in a control group, and did not obtain informed consent. Participants were not offered treatment even after penicillin became widely available, and many in the community believed they were in a special government health program. The study ended after news coverage prompted an advisory panel to recommend stopping it. President Clinton later issued a formal apology, and the federal investment announced with it established what became the National Center for Bioethics in Research and Health Care at Tuskegee University.
That is one study. It is not the only one, and it is not the only reason. Within living memory, people have been refused admission, given different pain treatment, talked past in their own appointments, and told their symptoms were imagined.
The effect persists and it is measurable. In a 2021 study in the Journal of Acquired Immune Deficiency Syndromes, Laura Bogart and colleagues interviewed 101 Black Americans living with HIV in Los Angeles County and found that nearly all, 97%, endorsed at least one COVID-19-related mistrust belief, and more than half endorsed at least one belief associated with vaccine or treatment hesitancy. The authors describe medical mistrust as a result of systemic racism, and they found health care providers were among the most trusted sources of information for the same group.
That last detail is the useful one. Mistrust of a system is not the same as mistrust of your own clinician. A relationship with one person who listens is often the thing that carries a patient back into care.
What to ask for, by name
Asking for something by its correct name changes the answer you get, because it tells the person at the desk this is a service they provide rather than a favor you want.
Before the appointment. “I need a professional medical interpreter in [language] for this visit. Please note it in my chart.” Say it when you book, not when you arrive, and ask whether it will be in person, by video or by phone.
At the desk. If no interpreter was arranged, ask again and ask for the language line. Say plainly that you do not want a family member to interpret.
Before anything gets signed. Consent forms, procedure explanations and discharge instructions are the highest-stakes moments. Ask for the interpreter specifically for those, even if you got through the rest of the visit on your own English.
About food and daily life. “Show me how this works with what we cook at home.” And: “What is the one change that matters most, if I can only make one?”
About everything else you use. Bring the actual bottles, boxes and packets, including anything herbal, in one bag. Say what you take, how often, and who recommended it.
If something goes wrong. Ask for the patient advocate or patient relations office by name, and for the complaint process in your language.
When to stop managing it yourself
Call the clinic back before your next dose if you left an appointment unsure what a medicine is for, how much to take, or when to stop. That is not a small loose end. Dosing and duration were among the instructions most often lost in the interpreting studies above.
Call before you start anything new if you use a traditional remedy and have just been prescribed a medication, or have surgery scheduled.
Call the same day if symptoms are getting worse while you wait for a clearer explanation, and call 911 for chest pain, trouble breathing, sudden weakness or confusion, or uncontrolled bleeding. No communication problem is worth waiting out in an emergency.
The useful part
If you take one thing from this, take the interpreter.
Everything else in this article is a preference that improves care at the margins. The interpreter is the one that changes whether the right drug is taken at the right dose, and it is the one federal law already says must be free. The research says the margin between a trained interpreter and a willing relative is not small, and the willing relative is the arrangement most families default to.
Ask when you book. Ask again at the desk. Ask a third time before you sign anything. And never let a child do it.
Sources
- U.S. Department of Health and Human Services, Office of Minority Health. “National Standards for Culturally and Linguistically Appropriate Services (CLAS) in Health and Health Care: The Standards.” thinkculturalhealth.hhs.gov
- U.S. Department of Health and Human Services, Office of Minority Health. “CLAS: What Is CLAS?.” thinkculturalhealth.hhs.gov
- U.S. Department of Health and Human Services, Office for Civil Rights. “Limited English Proficiency (LEP).” hhs.gov
- Agency for Healthcare Research and Quality. “Improving Patient Safety Systems for Patients With Limited English Proficiency.” ahrq.gov
- Centers for Disease Control and Prevention. “The U.S. Public Health Service Untreated Syphilis Study at Tuskegee.” cdc.gov
- National Center for Complementary and Integrative Health. “Complementary, Alternative, or Integrative Health: What's In a Name?.” nccih.nih.gov
- National Center for Complementary and Integrative Health. “Things to Know When Selecting a Complementary Health Practitioner.” nccih.nih.gov
- U.S. Department of Agriculture and U.S. Department of Health and Human Services. “Dietary Guidelines for Americans, 2025-2030.” realfood.gov
- Pediatrics (via PubMed). “Flores G, Laws MB, Mayo SJ, Zuckerman B, Abreu M, Medina L, Hardt EJ. Errors in medical interpretation and their potential clinical consequences in pediatric encounters. 2003. PubMed. DOI: 10.1542/peds.111.1.6.” doi.org
- Annals of Emergency Medicine (via PubMed). “Flores G, Abreu M, Barone CP, Bachur R, Lin H. Errors of medical interpretation and their potential clinical consequences: a comparison of professional versus ad hoc versus no interpreters. 2012. PubMed. DOI: 10.1016/j.annemergmed.2012.01.025.” doi.org
- Journal of General Internal Medicine (via PubMed). “Schenker Y, Perez-Stable EJ, Nickleach D, Karliner LS. Patterns of interpreter use for hospitalized patients with limited English proficiency. 2011. PubMed. DOI: 10.1007/s11606-010-1619-z.” doi.org
- Hawai'i Journal of Health & Social Welfare (via PubMed). “Kostareva U, Pe'a Varik K, Siriwardhana C, Liu M, Qureshi K. Limited English Proficiency, Postoperative Complications, and Interpreter Use in Vascular Surgery Patients in Hawai'i. 2023. PubMed. PMID: 36779005 (no DOI assigned).” pmc.ncbi.nlm.nih.gov
- Journal of Acquired Immune Deficiency Syndromes (via PubMed). “Bogart LM, Ojikutu BO, Tyagi K, Klein DJ, Mutchler MG, Dong L, Lawrence SJ, Thomas DR, Kellman S. COVID-19 Related Medical Mistrust, Health Impacts, and Potential Vaccine Hesitancy Among Black Americans Living With HIV. 2021. PubMed. DOI: 10.1097/QAI.0000000000002570.” doi.org
- Drug and Alcohol Dependence (via PubMed). “Hai AH, Lee CS, Abbas BT, Bo A, Morgan H, Delva J. Culturally adapted evidence-based treatments for adults with substance use problems: A systematic review and meta-analysis. 2021. PubMed. DOI: 10.1016/j.drugalcdep.2021.108856.” doi.org