DOI: 10.1192/j.eurpsy.2026.10222 ISSN: 0924-9338

Artificial Intelligence to Meet the Psychosocial Needs of Cancer Patients: A Systematic Review of Randomized Controlled Trials

N. D. Kılınç, F. Oflaz

Introduction

Cancer diagnosis frequently leads to psychosocial distress, including anxiety, depression, and reduced quality of life. Although psychosocial support is essential, access barriers limit timely care. Artificial intelligence (AI) has emerged as a scalable and accessible means of delivering psychosocial interventions.

Objectives

This study systematically reviewed randomized controlled trials (RCTs) evaluating AI-based interventions addressing psychosocial needs in cancer patients, focusing on intervention types, effectiveness, and implications for supportive care.

Methods

Following PRISMA guidelines, a systematic review was conducted (PROSPERO: CRD420251134621). Database searches identified 996 records; 143 full texts were assessed, and 7 RCTs met inclusion criteria, encompassing 913 cancer patients. Eligible studies tested AI-based interventions designed to address unmet psychosocial needs. Risk of bias was assessed using the Cochrane RoB 2 tool.

Results

Interventions included chatbots, mobile apps, digital counseling, and app-based therapeutics. Most demonstrated positive psychosocial effects, with significant reductions in anxiety, depression, and distress, particularly in younger patients and those receiving chatbot or app support. Additional benefits included improved self-efficacy, social support, knowledge, and empowerment. Some studies also reported reductions in fatigue and improved attitudes toward AI. While certain trials showed no significant effects on distress or quality of life, feasibility, acceptability, and patient satisfaction were consistently high. Secondary benefits included reduced pain, fewer pain-related hospitalizations, and better symptom management (Image 1).

Image 1:

Image 1: Long description.

Conclusions

RCT evidence indicates that AI-based interventions can effectively reduce psychosocial distress and related symptoms in cancer patients, while enhancing empowerment and self-management. Despite mixed effects on quality of life, consistent psychological and symptomatic benefits highlight their potential as valuable adjuncts to oncology care. Further large-scale, culturally diverse, and long-term trials are needed to confirm generalizability and sustainability.

Disclosure of Interest

None Declared

More from our Archive