Mental Health

The impact of an artificial intelligence (AI)-assisted education program on mental health, social support and quality of life in older individuals with head and neck cancer: study protocol of a randomized controlled trial.

TL;DR

This paper describes the protocol for a randomized controlled trial evaluating whether an AI-driven personalized health education program ('Kangkang' AI assistant) can improve mental health, social support, and quality of life in older patients (aged ≥60 years) following head and neck cancer surgery over 12 months.

Key Findings

Head and neck cancer treatment leads to functional impairments that severely affect quality of life, with older patients facing heightened mental health challenges due to combined stress of aging and disease.

  • HNC treatment often causes functional impairments in speech, swallowing, and appearance
  • Older individuals face compounded stress from both aging and disease processes
  • The study specifically targets patients aged ≥60 years as a vulnerable population
  • Mental health, social support, and quality of life are all identified as affected domains

The trial is designed as a single-center, two-group randomized controlled trial with 100 postoperative HNC patients randomly assigned to intervention or control groups.

  • Total sample size of 100 patients: intervention group (n=50) and control group (n=50)
  • Inclusion criterion requires patients to be aged ≥60 years and postoperative
  • Randomization is used to assign participants to groups
  • Single-center design is acknowledged as a limitation to generalizability

The intervention group will receive 12 months of personalized and phased health education through the 'Kangkang' AI assistant, including video/graphic content and real-time AI Q&A.

  • Intervention duration is 12 months
  • The AI platform is named the 'Kangkang' AI assistant
  • Education modalities include video/graphic materials and real-time AI question-and-answer functionality
  • Education is described as both 'personalized' and 'phased'
  • The control group receives standardized SMS health education at the same frequency

The study protocol includes 13 validated outcome measures spanning mental health, physical function, nutrition, pain, sleep, loneliness, and quality of life domains.

  • Outcome measures include: Perceived Stress Scale, Barthel Index, Patient Assessment of Constipation Quality of Life, Nutritional Risk Screening 2002, Numerical Rating Scale for Pain, Rosenberg Self-Esteem Scale, Pittsburgh Sleep Quality Index, UCLA Loneliness Scale (3rd edition), Fear of Cancer Progression Questionnaire-Short Form, World Health Organization Quality of Life Assessment for Older Adults, Generalized Anxiety Disorder-7, Patient Health Questionnaire-9, and Morse Fall Scale
  • Assessments will be performed 5 times: at baseline (preoperative) and at 1, 3, 6, and 12 months postoperatively
  • Both mental health outcomes (anxiety, depression, stress) and physical outcomes (falls, nutrition, pain) are included

Statistical analysis will use intention-to-treat analysis with linear mixed models and maximum likelihood estimation to handle continuous variables and missing data.

  • Intention-to-treat analysis is the primary analytical approach
  • Linear mixed models with maximum likelihood estimation will be used for continuous variables
  • Maximum likelihood estimation is specifically cited as the method for managing missing data
  • The analytical approach is described prospectively as part of the protocol

The trial was prospectively registered in the China Clinical Trials Registry on December 5, 2025, with registration number ChiCTR2500114052.

  • Registration date: December 5, 2025
  • Registry: China Clinical Trials Registry
  • Registration number: ChiCTR2500114052
  • The registration is described as 'prospective'

The single-center design and reliance on self-reported outcomes are identified as limitations, with future multicenter studies recommended.

  • Single-center design may limit generalizability of findings
  • Reliance on self-reported outcomes is acknowledged as a methodological limitation
  • Authors explicitly call for future multicenter studies to address these limitations

What This Means

This paper describes the design of a clinical trial—not its results—that will test whether an artificial intelligence (AI) health education tool can help older cancer patients feel better mentally and physically after surgery. The study focuses on people aged 60 and older who have had surgery for head and neck cancer, a group that faces especially difficult challenges because cancer treatment can impair their ability to speak, swallow, and maintain their appearance, while aging adds additional physical and emotional burdens. Half of the 100 enrolled patients will use an AI assistant called 'Kangkang' that provides personalized videos, graphics, and answers to health questions over 12 months, while the other half will receive standard text message health tips. Researchers will measure a wide range of outcomes including anxiety, depression, stress, self-esteem, sleep quality, loneliness, fear of cancer returning, nutritional status, pain, fall risk, and overall quality of life. These measurements will be taken five times over a year—before surgery and at 1, 3, 6, and 12 months after surgery—to track how patients change over time. The study uses rigorous statistical methods designed to account for patients who may drop out before the study ends. This research suggests that AI-driven, personalized health education could be a scalable way to support older cancer patients beyond the hospital setting, potentially addressing gaps in mental health and quality-of-life support that standard care may not fully meet. However, because this is a single-center study relying largely on patients' own reports, the authors note that the results may not apply broadly to all populations, and they call for larger, multi-site studies in the future. As of publication, this is a protocol paper only—no outcome data have yet been collected or reported.

Have a question about this study?

Citation

Sun Y, Ma H, Zhong W, Peng Y, Fan X, Liang Y, et al.. (2026). The impact of an artificial intelligence (AI)-assisted education program on mental health, social support and quality of life in older individuals with head and neck cancer: study protocol of a randomized controlled trial.. BMC geriatrics. https://doi.org/10.1186/s12877-026-07666-6