WeĀ recentlyĀ had the pleasure of having anĀ in-depthĀ discussion with Debbie Lesser, Certified Healthcare Interpreter & Language Access Consultant. Her career spans interpreting, interpreting leadership, hospital operations, and national advocacy.Ā Ā
In our interview, she talked about a truth that many healthcare leaders overlook: safe patient care isnāt possible without effective language access.
Yet healthcare systems continue to unintentionally introduce risk with outdated language access practices, organizational silos, and use of unvalidated AI tools and unqualified interpreters.
Based on our conversation with Debbie and years of experience in healthcare environments, here are four signs your language access program may be putting patients in harm’s way and exactly how to refine them.
1. Clinicians rely on bilingual staff to interpret or treat patients without validating their competency
Multilingualism is an asset. But itās not the same thing as clinical fluency. Nor does every multilingual individual automatically understand interpreter protocol and ethics (aka the interpreterās toolkit).
Without assessing interpreter and/or target language competency, multilingual staff may be asked to fill roles they arenāt trained or prepared for. They may misunderstand medical terminology or feel deeply uncomfortable being put in positions beyond their skillset.
Interpreting or translating professionally requires a learned set of skills often gained from education, training, and years of experience.
How to fix it
- Talk to HR (or relevant parties) about implementing a formal validation program to assess interpreter ability and/or language proficiency for treating patients in the target language.
Debbieās hospital implemented a qualified bilingual staff (QBS) assessment program; CLI has a bilingual and interpreter assessment available. Both focus on medical environments.
- Build in any additional language and/or interpreting duties into the individualās job description.
- Make sure theyāre fairly compensated for this additional skillset.
| What to think about when validating language proficiency in bilingual staff | Why it matters |
|---|---|
| What competency level is needed? | Ensures bilingual staff arenāt communicating in language beyond their skillset. |
| Which roles needs which level? | Prevents staff from being asked to communicate in the target language without assessment. |
| What is clinical efficiency? | Helps set clear expectations and avoid ambiguity. |
| How often should staff be reassessed? | Ensures skills donāt diminish without regular use. |
2. Family members or friends are used as interpreters consistently
No one wants family members to help their loved ones more than us! But patient preference for their family members to act as interpreter does not eliminate clinical risk.
Family members may filter information, especially bad news. They may not understand medical terms or remain neutral. These factors can reduce clarity or accuracy and introduce risk to the patient (not reduce it).
How to fix it
- Establish a policy guiding staff and clinicians on the use of ad hoc interpreters. The policy should include:
- Exceptions to the rule (in emergencies, for example)
- Using minors as interpreters (never recommended)
- The importance of patient consent
- Documentation practices including both use of a qualified interpreter and patient refusal
- Train staff and clinicians on the policy and why itās unsafe for patients to use untrained interpreters.
- If the patient and family member agree to the arrangement, always have a professional interpreter sit in on the session. Remember: Working with a qualified interpreter is not just for the patient with NELP; providers need to understand their patients, too.
3. No guardrails around AI interpretation or translation technology exist
Weāve all seen whacky AI output, especially in healthcare contexts. Debbie shot off one example of a state agency used Google Translate to translate crucial vaccine information for Covid. The tech translated ānot requiredā to ānot necessaryā (thereās a major clinical difference).
Even though the accuracy of AI is improving, it carries bias and lacks situational awareness and ethical judgment. Only human judgement can understand if a patient truly understands.
How to fix it
- Use AI interpretation and translation tools where appropriate. For medical content and encounters, always advocate for human oversight.
- Be vocal about the need for an AI Governance Council in your healthcare system, focused on patient safety and civil rights. Route all AI tools through the council.
- Limit AI interpreting or translation tools to administrative or preātranslated content creation and not clinical decisions.
Read to choose the right fit for your organization: AI vs. Human Interpreters
4. Your language access program is siloed and not embedded in patient safety, compliance, or patient experience.
Language access can sometimes be placed in departments with little to no experience in patient communication, like IT, transportation, or nutrition.
As Debbie noted, this can make it easy for leaders with no interpreting experience to overlook the stakes until after a mistake has been made.
How to fix it
- Network internally with teams like compliance and patient experience.
“Compliance,ā states Debbie, “is just such a huge job. And language access is a small component of that. So although it’s top of mind for me because that’s what I do every day, it’s not what they do every day. So bringing a question to them or suggesting how you can partnerā is a great step.
- Build external networks with other language access leaders as well. Ask them how they approach challenges you or your system are facing. You can then bring those insights back to improve your own system. This relationship building can even extend to your IT team, for example. Consider setting up discussions with your IT group and your language access colleagueās IT team to solutions and approaches.
A Safer Future for Patients with NELP
Debbieās final advice: donāt give up!
Most organizations are already doing many, many things well and simply need to tweak a practice or two. Build cross-departmental partnerships, seek peer networks, put guardrails around AI, and keep patients at the center. Stronger language access is essential for safe, quality care.
Your language access colleagues keep up with what’s going on by subscribing to CLI’s newsletter, Open for Interpretation. You can too: https://resource.certifiedlanguages.com/newsletter
