Patients’ Attitudes and Beliefs Toward Artificial Intelligence Use in Cancer Care: Cross-Sectional Survey Study

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Background: Artificial intelligence (AI) is being rapidly integrated into oncologic care, yet little is known about how patients perceive these applications. Understanding patient perceptions is critical to ensuring AI applications align with their needs and preferences. Objective: This study aimed to evaluate oncology patients’ attitudes and beliefs on the use of AI across clinical touchpoints in cancer care. Methods: We conducted a cross-sectional survey study with adult oncology patients from September to December 2024. The survey assessed patients’ comfort with AI use across 8 clinical touchpoints of cancer care (eg, screening, diagnosis, treatment) on a 5-point Likert scale (1=very uncomfortable to 5=very comfortable). Patients also rated their concerns about AI, including potential harms related to its use (eg, medical errors, privacy breaches), on a 3-point Likert scale (1=not concerned to 3=very concerned). Results: Of 383 patients approached, 330 (86.2% response rate) participated; 184 (55.9%) were male, 162 (49.4%) were aged 65 years or older, 35 (10.8%) were Black, 40 (12.1%) were Hispanic or Latino, and 233 (72.6%) were actively receiving cancer treatment. Patients were most comfortable with AI use in cancer screening (80.2%) and supportive care applications, including exercise (78.2%), diet (74.8%), and herbs/supplements (72.4%). Patients were least comfortable with AI use to assist with diagnosis (70.4%), symptom management (67.5%), treatment planning (64.8%), and prognosis (61.5%). Nonetheless, about half (49.7%) were at least somewhat concerned with the use of AI in cancer care, most commonly about the loss of human interaction and medical errors. Conclusions: Although the majority of oncology patients had a favorable view of AI in cancer care, nearly half had concerns about potential harms. Incorporating patient perspectives into AI development is essential for patient-centered and high-quality cancer care.

This publication has 14 references indexed in Scilit: