Medical-AI trial: an AI app to make eating safer for older adults with swallowing difficulty (dysphagia) — sticking to the prescribed texture (RCT) (NCT07654088)
Swallowing difficulty (dysphagia), common in older adults, risks choking, pneumonia, and poor nutrition. Preparing food and drink at the correct modified texture at home is hard. This randomized controlled trial tests whether an AI smartphone app can help community-dwelling older adults eat more safely. The primary outcome is the proportion of meals adhering to the prescribed texture level. 332 participants.
Trial overview (primary data)
- StatusNot yet recruiting
- ConditionsDeglutition Disorders, Dysphagia
- InterventionsDEVICE: AI-Enabled Mobile Application plus Standard Care, BEHAVIORAL: Standard Dysphagia Education
- SponsorKAM Chi Shan Anna
- Target enrollment332 participants
- Period2028-01-01 〜 2030-12-01
Key points
- A randomized controlled trial of whether an AI smartphone app helps community-dwelling older adults with dysphagia eat more safely.
- Dysphagia risks aspiration, choking, aspiration pneumonia, poor nutrition, and reduced quality of life; the mainstay response is texture modification (thickness/softness).
- Preparing the correct texture at home is hard, and that is where the approach tends to break down.
- The primary outcome is the proportion of meals adhering to the prescribed texture level (adherence). 332 participants.
- It targets at-home adherence to a prescribed diet; a texture-adherence evaluation, not an establishment of aspiration or pneumonia prevention.
- A correct prescription does not work unless kept at home, so the endpoint is the share of meals meeting the prescribed level.
1Age-related swallowing difficulty and home practice
With age, swallowing can weaken, making dysphagia (difficulty swallowing) more likely. When swallowing goes wrong, food or liquid can enter the airway (aspiration), leading to choking, aspiration pneumonia, poor nutrition, and reduced quality of life. A mainstay response is texture modification, softening foods or thickening liquids.
But preparing the prescribed consistency day after day, especially at home, is hard, and that is where the approach tends to break down.
2An app that supports doing it at home
This study tests an AI smartphone app that supports that at-home execution, as a randomized controlled trial. Per the registry summary, against the risks of dysphagia (choking, aspiration pneumonia, poor nutrition) and the difficulty of texture modification, it examines whether an AI app can help community-dwelling older adults eat more safely.
The primary outcome is the proportion of meals adhering to the prescribed texture level (adherence). The size is 332, at the not-yet-recruiting stage.
3Where it breaks down
A dietary regimen does not work if what is correctly prescribed is not kept at home. What this study addresses is where it breaks.
- 1A specialist prescribesThe correct level of texture and thickness is settled
- 2It goes homeDaily preparation falls to the person or the family
- 3Keeping it consistent is hardThis is where the measure weakens
- 4An app supports itThe primary endpoint is the share of meals meeting the prescribed level
Failure to swallow properly sends food or fluid into the airway, leading to choking, aspiration pneumonia, malnutrition and loss of quality of life. Texture modification is the mainstay of prevention, and preparing the prescribed consistency day after day at home has been its weak point. The study covers 332 participants and is not yet recruiting.
Why it matters
It targets a real weak point that governs effectiveness: how well a hospital-prescribed diet is followed at home (adherence). If a familiar smartphone can help check texture, it could translate into a real reduction in aspiration risk, significant as populations age (this study is a texture-adherence evaluation, not an establishment of aspiration/pneumonia prevention).
FAQ
Why modify textures in dysphagia?
Can the app prevent aspiration or pneumonia?
Sources (primary)
Source: ClinicalTrials.gov (U.S. NIH/NLM, public domain). This site does not provide medical advice. Verify the latest and exact details with the official source. This site is not endorsed or certified by the NIH/NLM.
- ClinicalTrials.gov (study record, original)
- NCT ID: NCT07654088