AI-Enabled Mobile App for Safe Eating in Older Adults With Dysphagia

NCT07654088 | Not Yet Recruiting

• 1 other identifier: 2025-2026-0777
• Study Type: interventional
• Target: 332
• Locations: 1 country
• Sites: 1

Brief Summary

Difficulty swallowing (called dysphagia) is common in older adults and can make eating and drinking unsafe. It may lead to serious problems such as choking, lung infections, poor nutrition, and reduced quality of life. One common way to reduce these risks is to modify food and drink textures (for example, making foods softer or liquids thicker). However, people often find it difficult to prepare food at the correct texture level in everyday life, especially at home, which may reduce the effectiveness of this approach. This study aims to test whether a smartphone application powered by artificial intelligence (AI) can help older adults with swallowing difficulties eat more safely. The app allows users (or their caregivers) to take a photo of food or drinks, and the app then estimates the texture level and provides guidance to help ensure it is safe to swallow. It also gives simple prompts to double-check food texture when needed. In this clinical trial, community-dwelling adults aged 60 years or above with swallowing difficulties will be randomly assigned to one of two groups. One group will receive usual care, which includes education about safe swallowing and written instructions on appropriate food textures. The other group will receive the same usual care plus access to the AI-enabled mobile app for 16 weeks. Participants will continue their daily eating routines at home. The main question this study is trying to answer is: Does using the AI-enabled mobile app improve how often people eat foods that match their recommended safe texture level compared with usual care alone? The study will also examine whether the app helps reduce swallowing-related problems (such as choking), improves quality of life, and supports better overall eating ability. In addition, the study will evaluate how easy the app is to use and whether it places any burden on users. Hypothesis: The researchers hypothesize that participants who use the AI-enabled app, in addition to usual care, will more consistently follow recommended food texture guidelines and experience safer eating compared with those who receive usual care alone.

Conditions

Deglutition Disorders; Dysphagia

Keywords

Dysphagia; Deglutition disorders; Swallowing disorders; Texture modified diet; Food texture assessment; Artificial intelligence in health care; Clinical decision support systems; Digital health intervention; Older adults; Aging population; Mobile health applications

Outcome Measures

Primary Outcomes (1)

• Proportion of Meals Adhering to Prescribed Texture Level

  The participant-level proportion of meals that correctly match the prescribed food and liquid texture level, based on standardized dysphagia diet guidelines, during a defined assessment period. In the intervention group, classification is supported by the mobile application and verification procedures; in the control group, adherence is determined using structured dietary logs with verification. Measure Type / Units: Proportion (0 to 1, or percentage 0-100%) Interpretation: Higher values indicate better adherence to prescribed diet texture

  Week 16

Secondary Outcomes (5)

• Swallowing-Related Quality of Life

  Baseline, Week 16, Week 24

• Functional Oral Intake

  Baseline, Week 16, Week 24

• Incidence of Dysphagia-Related Adverse Events

  Baseline to Week 16; Week 16 to Week 24

• mHealth App Usability Questionnaire (MAUQ) Score

  Week 16

• NASA Task Load Index (NASA-TLX) Global Score

  Week 16

Other Outcomes (4)

• Dynamic Imaging Grade of Swallowing Toxicity (DIGEST) Score

  Baseline, Week 16

• Mobile Application Engagement (Usage Frequency)

  During the intervention period (Weeks 1-16)

• Response to Application Prompts

  Weeks 1-16

Study Arms (2)

AI-Enabled Mobile Application plus Standard Care

EXPERIMENTAL

Participants receive access to an artificial intelligence-enabled mobile application designed to support real-time classification of food and liquid textures according to standardized dysphagia diet levels. The application provides safety-oriented guidance and prompts for verification of food texture during daily meal preparation and consumption. Participants also receive standard dysphagia education materials and training. The intervention is used in a home setting over the study period to support adherence to prescribed dietary recommendations. Intervention used: AI-Enabled Mobile Application plus Standard Care

Device: AI-Enabled Mobile Application plus Standard Care

Standard Care Education

ACTIVE COMPARATOR

Participants receive standard dysphagia education, including guidance on safe swallowing practices and instructions for preparing texture-modified foods and liquids. Educational materials and training are provided, reflecting usual community care. No digital or application-based decision support is provided.

Behavioral: Standard Dysphagia Education

Interventions

AI-Enabled Mobile Application plus Standard Care DEVICE

A smartphone-based application that uses artificial intelligence to classify food and liquid textures from images captured by the user. The application provides real-time guidance aligned with standardized dysphagia diet levels and delivers safety-focused prompts to verify texture using simple methods when needed. The tool is designed to support safe meal preparation and improve adherence to prescribed texture-modified diets in daily home settings. Participants receive onboarding and use the application during meals throughout the intervention period.

AI-Enabled Mobile Application plus Standard Care

Standard Dysphagia Education BEHAVIORAL

A structured education session providing guidance on safe swallowing practices and preparation of texture-modified foods and liquids. Participants receive printed educational materials describing appropriate food textures and simple methods for checking consistency. This reflects usual care in community dysphagia management and does not include digital or automated decision support.

Standard Care Education

Eligibility Criteria

• Age: 60 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Adults aged 60 years or above
• Community-dwelling (living in a home or community setting)
• Suspected or clinically identified oropharyngeal dysphagia
• Currently consuming food or liquids orally at texture-modified levels
• Able to provide informed consent, or with caregiver support if mild cognitive impairment is present
• Access to a smartphone or tablet, either independently or with caregiver assistance
• Willing and able to participate in study procedures and follow-up assessments

You may not qualify if:

• Exclusive dependence on non-oral feeding (e.g., tube feeding)
• Severe cognitive impairment or severe visual impairment that prevents meaningful participation
• Medical conditions or circumstances that make participation unsafe
• Life expectancy less than 6 months
• Current participation in another dysphagia-related interventional study
• Use of other digital tools specifically designed for food texture classification during the study period

Sponsors & Collaborators

• KAM Chi Shan Anna: lead

Study Sites (1)

The Education University of Hong Kong

Hong Kong, Hong Kong

Location

MeSH Terms

Conditions

Deglutition Disorders

Condition Hierarchy (Ancestors)

Esophageal Diseases; Gastrointestinal Diseases; Digestive System Diseases; Pharyngeal Diseases; Otorhinolaryngologic Diseases

Study Officials

• Anna Kam, AuD

  The Education University of Hong Kong

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Anna Kam, AuD

CONTACT

852-29488194; annakam@eduhk.hk

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: OUTCOMES ASSESSOR
• Purpose: TREATMENT
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR INVESTIGATOR
• PI Title: Associate Professor

Model Details: This is a two-arm, parallel-group, randomized controlled trial. Participants are individually assigned in a 1:1 ratio to either (1) an intervention group receiving access to an artificial intelligence-enabled mobile application in addition to standard dysphagia education, or (2) a control group receiving standard dysphagia education alone. Participants remain in their assigned group for the entire study duration, including the intervention and follow-up periods. Randomization is stratified by baseline functional status, cognitive status, and living arrangement to ensure balance between groups. Outcome assessors are blinded to group allocation to minimize bias, while participants and intervention staff are not blinded due to the nature of the intervention.

Study Record Dates

• First Submitted: June 5, 2026
• First Posted: June 17, 2026
• Study Start (Estimated): January 1, 2028
• Primary Completion (Estimated): December 1, 2030
• Study Completion (Estimated): December 1, 2030
• Last Updated: June 17, 2026

Record last verified: 2026-06

Data Sharing

• IPD Sharing: Will not share

Locations

HK (1)