Skip to content

Generative AI-Based Health Education for Older Adults With Sarcopenic Obesity

An Exploration of Health-Promoting Effects of Personalized AI-Generated Multimedia Health Education in Community-Dwelling Older Adults With Sarcopenic Obesity

Status
Not yet recruiting
Phases
Unknown
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT07782320
Enrollment
500
Registered
2026-08-24
Start date
2026-10-01
Completion date
2029-12-31
Last updated
2026-08-24

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Sarcopenic Obesity

Brief summary

Sarcopenic obesity is a major public health concern among community-dwelling older adult populations.Encouraging healthy behavior modification through health education emerges as an effective strategy for preventing and treating sarcopenic obesity. Generative Artificial Intelligence (AI) offers an innovative opportunity to tailor health education for aging populations. This study aims to explore the effectiveness of Personalized AI-generated Multimedia Health Education in community-dwelling older adults with sarcopenic obesity. The study comprises two phases, with 280 participants in Study 1 and 180 in Study 2.Study 1, a cluster randomized controlled trial, explores the feasibility, acceptability, and efficacy of different AI-generated multimedia, including images, sounds, and videos. Study 2 employs a three-armed,individually randomized controlled trial design, creating Personalized AI-generated Multimedia Health Education based on participant preferences. All materials focus on behavioral risk factors, delivered by social media-based AI chatbots. The intervention is conducted once a day, five days a week for 12 weeks.Structured questionnaires and objective instruments collect data before and after the intervention. The study outcome includes behavioral and psychological factors, quality of life, and sarcopenic obesity indicators. Statistical analyses include descriptive analyses, Chi-square tests, t-tests, One-way analysis of variance, path models, and generalized estimating equations. This study anticipates that Personalized AI-generated Multimedia Health Education will be a feasible, acceptable, and effective intervention for older adults. The study results are expected to demonstrate a significant improvement in study outcomes in the experimental group. Personalized AI-generated multimedia health education could be an easy-to-use,enjoyable, and effective strategy for health promotion and sarcopenic obesity prevention.

Interventions

BEHAVIORALGenerative AI-based text materials

Participants receive a text message regarding the health education topic.

BEHAVIORALGenerative AI-based video materials

Participants receive a video message , accompanied by AI-generated background music at the beginning and end. Male avatars are used for PA and SO education, and female avatars are used for HD education.

BEHAVIORALGenerative AI-based images materials

Participants receive one Gen-AI-generated image concerning the health education topic, formatted as a health poster.

BEHAVIORALGenerative AI-based voice materials

Participants receive a podcast-style voice message regarding the health education topic. The recording incorporates AI-generated background voice. Male voices are utilized for physical activity (PA) education, while female voices are used for healthy diet (HD) and sarcopenic obesity (SO) education.

BEHAVIORALNon-Personalized Gen-AI MHE

Participants receive non-personalized multimodal health education generated by generative AI.

BEHAVIORALPersonalized Gen-AI MHE

Participants receive personalized multimodal health education tailored through generative AI.

Sponsors

Taipei Medical University
Lead SponsorOTHER

Study design

Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
PREVENTION
Masking
NONE

Eligibility

Sex/Gender
ALL
Age
50 Years to 99 Years
Healthy volunteers
Yes

Inclusion criteria

1. . Age: 60 years or older 2. . Smartphone ownership with internet connectivity 3. . Appendicular fat-free mass (AFFM) calculated by the equation: AFFM = 14.529 + (17.989 \* height)+ (0.1307 \* fat mass). The cut-off value corresponds to a residual ≤ 3.4 in the equation 4. . Ability to read, listen, and understand health education materials with normal cognitive function (MMSE score ≥ 25) and normal sensory function.

Exclusion criteria

1. . Functional dependency 2. . Current residence in long-term care facilities or hospitals 3. . Presence of serious diagnosed diseases, disabilities, or mental health issues requiring medical treatment that might influence the study process.

Design outcomes

Primary

MeasureTime frameDescription
The Chinese version of the Physical Activity Scale for the ElderlyBaseline, midpoint (the 6th week), and post-intervention (the 13rd week)This questionnaire consists of 12 items assessing physical activity (PA) over the past 7 days, including leisure-time, household, and occupational activities. Total, light-, moderate-, and vigorous-intensity PA can be calculated in metabolic equivalents of task-minutes per week (MET-min/week).
General Dietary Behavior InventoryBaseline, midpoint (the 6th week), and post-intervention (the 13rd week)A total of 16 items measure participants' healthy behaviors using a 5-point bipolar scale.The score of each item is summed up to a total score, with a higher score representing healthier dietary behavior.
The skeletal muscle index (SMI)Baseline, midpoint (the 6th week), and post-intervention (the 13rd week)A body composition analyzer using bioelectrical impedance analysis technology is conducted. A higher score (kg/m2) of the skeletal muscle index (SMI) indicates greater muscle mass for sarcopenic indicators.
Muscle StrengthBaseline, midpoint (the 6th week), and post-intervention (the 13rd week)Muscle strength is assessed using a handheld grip device (SANKA, Japan). Participants are asked to stand, allow the wrist and arm of the dominant hand to hang straight down, and maintain maximum strength for more than 3 seconds. This measurement is repeated three times, and the maximum value (kg) is used as the muscle strength.

Secondary

MeasureTime frameDescription
The Health-Promoting Lifestyle Profile IIBaseline, midpoint (the 6th week), and post-intervention (the 13rd week)The Health-Promoting Lifestyle Profile II consists of 29 items measuring participants' engagement in healthy lifestyles. A 4-point Likert scale is used (1 never, 2 sometimes, 3 frequently, and 4 regularly).
World Health Organization (WHO)- Quality of Life ScaleBaseline, midpoint (the 6th week), and post-intervention (the 13rd week)A shorter version with 28 questions. The scale evaluates various aspects of life, including physical health, psychological state, social relationships, and environmental factors, scored between 1 and 5.

Countries

Taiwan

Contacts

CONTACTHsin-Yen Yen, PhD
kenji@tmu.edu.tw886-2-2736-1661

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Aug 25, 2026