Bioethics Forum Essay
Medical Interpretation in the U.S. is Inadequate and Harming Patients
Over the past few decades, many major cities in the United States have become more diverse and gained more residents with limited English proficiency. Health care systems have attempted to accommodate these residents, but their medical interpreter services are inadequate and inefficient. The results can be delayed emergency care for children, an increase in medical errors and health care costs, and a decrease in care quality and patient satisfaction.
There is a shortage of in-person medical interpreters. To compensate, hospitals have turned to machine translation tools. However, these tools, such as Google Translate, are more accurate for some languages than for others. One study that compared the accuracy of Google Translate for Spanish and Chinese found that it was less accurate for Chinese and that Chinese mistranslations were more likely to pose clinically significant and life-threatening harm. Google Translate supports Mandarin, but does not include Cantonese, a Chinese language spoken by over 70 million people worldwide and 43.6% of LEP residents in San Francisco.
Limitations in the number of languages supported result in unequal access to accurate translation services and a lack of inclusivity for patients. Patients without accurate translation services can’t describe their symptoms effectively. And, when they are discharged from the hospital, LEP patients are more likely than English-speaking patients to be readmitted. The reason for their readmission is not illness, but rather poor communication.
A lack of cultural sensitivity can cause a further disconnect in patient-provider communication and trust. For example, a 2022 study suggests that Hmong patients with LEP prefer having family members, instead of a professional interpreter, translate for them, which would be important to consider in hospitals with high Hmong populations, such as those in the Minneapolis-St. Paul metro area. However, ad-hoc interpreters, such as family and friends, cannot always accurately translate medical terminology. By contrast, Spanish-speaking Latinx patients report a preference for professional interpreters since they sometimes note feeling uncomfortable sharing confidential information in the presence of family and friends.
Hospitals conduct Community Health Needs Assessments (CHNAs) to analyze data regarding both their patient population and their city’s population. These assessments assess the social determinants of health and partner with community organizations and residents to combat health inequity. This includes running focus groups and gathering data on race, ethnicity, language, and nativity of the surrounding area, with an emphasis on their priority neighborhoods.
But the improvements to interpreter services outlined in each hospital’s CHNAs are vague and lack structure. Merely suggesting “increased access to interpreter services” is a common but insufficient solution. Studies show that even when interpreter services are made accessible, clinicians still underuse them. One barrier cited is technical difficulty with interpretation technology.
Health systems can better utilize demographic data to allocate funding towards interpreter services that meet the needs of their patient populations. For example, Tufts Medical Center is the main health care provider for Boston’s Chinatown, where most of the non-English speaking population speaks either Cantonese or Mandarin. Therefore, it may be beneficial for this medical center to steer away from machine translation tools like Google Translate and have both family members and professional interpreters work together to communicate medical information.
However, even if a hospital employs a large team of professional interpreters, its interpreter services may not meet patients’ needs. That’s because, in addition to language concordance, effective communication requires cultural competence. Hospitals should promote cultural competence by investing in diversity training—and by making key improvements in that training. Diversity training in hospitals tends to be too general; the solutions suggested for addressing inequality are ambiguous and nonspecific. Including focus group responses from residents would provide more concrete recommendations to improve cultural competence. CHNA data should also be incorporated into new employee orientations, which are mandatory and already include diversity training components. Furthermore, these orientations offer an opportunity to train staff on how to best utilize available interpreter services.
Expecting patients to adhere to Western constructs of health communication is unfair and unjust. We must consider how linguistic differences and cultural values impact patients’ expectations and decision-making. By tailoring translation services to their LEP populations, health care systems can improve the quality of care for each of their unique and diverse patients.
Riya Dahima is a senior at Northeastern University majoring in behavioral neuroscience and research student at the Dhand Lab at the Brigham and Women’s Hospital.
Melinda Luo is a senior at Northeastern majoring in biology and psychology, and a research student at the Dhand Lab at the Brigham and Women’s Hospital.
Vrushali Dhongade, MBBS, MS, MBE, is the Project Manager in the Dhand Lab at the Brigham and Women’s Hospital. (@vrush_25)