Agentic data science with the Data Agent Kit
cloud.google.com·7h ago
AI Research
arXiv CS.AI·7d ago·5 min read
"I just assumed that it would translate": examining MT risk awareness among healthcare staff with abbreviations as a use case
TL;DR
Healthcare staff in the UK often rely on machine translation, particularly Google Translate, to communicate with patients. This study highlights the risks associated with using machine translation for medical abbreviations, which can lead to serious patient safety issues.
✦ Why It Matters
Healthcare organizations should implement training on the risks of machine translation for medical communication.
Key Takeaways
- Educate staff on the risks of using machine translation.
- Implement guidelines for translating medical abbreviations.
- Encourage verification of translations by bilingual staff.
- Monitor patient outcomes related to translation errors.
Full Summary
In the UK, healthcare professionals frequently use machine translation (MT) tools like Google Translate to bridge language gaps with patients. However, this reliance can jeopardize patient safety, especially when translating medical abbreviations, which are already known to pose risks even in a single language.
The study involved semi-structured interviews with 21 healthcare staff from various roles, who were presented with contextualized examples of French and Spanish medical abbreviations translated via Google Translate. Qualitative analysis of the interviews revealed significant gaps in staff awareness regarding the potential dangers of using MT in high-stakes medical contexts.
The findings underscore the need for better training and guidelines on the use of MT in healthcare settings to mitigate risks associated with miscommunication.