VHA Tests AI-Powered Translation at Point-of-Care

Why It Matters

AI translation at the point of care can reduce language‑driven inequities and streamline communication for home‑based health services, setting a template for wider sector adoption.

Key Takeaways

• VHA piloted three AI translation tools with frontline caregivers
• Study covered top five client languages, later added four more
• Mabel chosen for medical-specific terminology and ease of use
• Next phase expands testing to 150 staff across diverse scenarios
• Findings aim to guide broader health sector AI translation adoption

Pulse Analysis

Language barriers have long hampered home‑care providers, forcing reliance on informal translators or family members. VHA Home HealthCare’s Innovation team tackled this gap by testing three AI translation platforms in real‑world settings, funded by the CABHI Discover & Adopt program. By focusing on the five most spoken languages among its clients—Cantonese, Mandarin, Hindi, Russian and Italian—the pilot mirrored the multicultural reality of Ontario’s home‑care market, while iterative feedback allowed the addition of four more languages, underscoring the importance of adaptability in health‑tech deployments.

The selection of Mabel as the preferred solution highlights a growing demand for domain‑specific AI tools. Unlike generic translators, Mabel is engineered for medical terminology, delivering higher accuracy for clinical conversations and reducing the cognitive load on nurses and personal support workers. Staff reported smoother workflows and greater confidence when discussing treatment plans, which translates into more equitable care and fewer miscommunication risks. This focus on usability and reliability is critical, as even minor translation errors can have serious clinical consequences.

VHA’s next testing phase, involving about 150 rehab and nursing professionals, will push the technology into more complex scenarios, providing data that could inform standards for AI translation in both community and hospital settings. As health systems worldwide grapple with diverse patient populations, VHA’s evidence‑based approach offers a replicable model for integrating AI language tools, balancing innovation with patient safety and regulatory compliance. The pilot’s outcomes may accelerate broader adoption, helping providers deliver culturally competent care while optimizing operational efficiency.