Language barriers contribute substantially to inequities in U.S. surgical care. Spanish-speaking patients are especially affected by communication challenges and inconsistent interpreter access during complex or time-sensitive encounters.

Gezzer Ortega, MD, MPH

A study led by Gezzer Ortega, MD, MPH, a physician investigator from the Center for Surgery and Public Health at Brigham and Women’s Hospital, sought to understand how Spanish-speaking surgical patients perceive emerging interpreter technologies, specifically artificial intelligence (AI)–based interpretation and remote video interpretation (RVI), and how these modalities could be implemented in clinical practice.

The researchers conducted a concurrent mixed-methods study involving 23 adult Spanish-speaking surgical patients within a U.S. academic health system. Participants completed validated surveys assessing acculturation, patient activation, and digital health literacy, followed by qualitative interviews after viewing standardized simulated postoperative encounters using both AI-based and RVI interpreter modalities. This design enabled direct comparison of patient experiences under controlled conditions.

The study found that patients did not view AI and RVI as competing or mutually exclusive tools. Instead, preferences were strongly context-dependent. AI-based interpretation was valued for its speed, convenience, privacy, and direct communication, particularly in routine, time-sensitive, or low-emotional-intensity situations. In contrast, RVI was preferred for emotionally sensitive, complex, or high-stakes conversations due to its perceived empathy, cultural nuance, and interpersonal connection.

Among patients with limited English language proficiency, many participants showed moderate digital literacy and openness to AI tools. Concerns persisted about AI’s ability to capture dialects, emotional nuance, and cultural context, while RVI raised issues related to technical delays and message fidelity. Overall, the findings highlight that patient trust, autonomy, and perceived control drive interpreter preferences, supporting a patient-informed hybrid framework that integrates AI-based and human interpreter services based on clinical context and patient preference.

Published in NEJM Catalyst Innovations in Care Delivery in March 2026 | Read the paper: “Artificial Intelligence for Language Access in Surgical Care: Patient Preferences and an Implementation Framework”

Summary reviewed by: Gezzer Ortega, MD, MPH, senior author