Chatbot-Delivered Stage of Change–Tailored Web-Based Intervention to Promote Physical Activity Among Inactive Community-Dwelling People Aged 65 years or More: Protocol for a Randomized Controlled Trial
Bibliographic record
Abstract
BACKGROUND: Physical activity (PA) has significant health benefits for older adults. However, many older adults in Hong Kong remain physically inactive. Interventions tailored to one's current stage of change (SOC) are more effective than non-SOC-tailored ones in facilitating behavioral changes. Chatbots are potentially useful to deliver SOC-tailored interventions to promote PA among older adults. OBJECTIVE: This randomized controlled trial (RCT) will compare the efficacy of an SOC- versus a non-SOC-tailored intervention in increasing the prevalence of meeting World Health Organization (WHO)-recommended PA levels 6 months after completion of the intervention among inactive community-dwelling individuals aged ≥65 years. METHODS: This is a partially blinded (outcome assessors and data analysts) and parallel-group RCT. A total of 278 inactive community-dwelling people aged 65 years or more will be randomized evenly into either an intervention group or a control group. In the intervention group, a fully automated chatbot with natural language processing (NLP) functions will measure participants' SOC related to PA and deliver web-based interventions tailored to their current SOC every week for 12 weeks. In the control group, the chatbot will not measure participants' SOC but will deliver a non-SOC-tailored web-based intervention every week for 12 weeks. Participants will be interviewed at baseline (T0), after completion of the intervention (T1), and 6 months after T1 (T2). The primary outcome is the prevalence of meeting WHO-recommended PA levels (ie, at least 150 minutes of moderate-intensity aerobic PA, at least 75 minutes of vigorous-intensity aerobic PA, or an equivalent combination of moderate-to-vigorous physical activity [MVPA] every week). PA will be measured using the Chinese version of the International Physical Activity Questionnaire Short Form (IPAQ-SF) and accelerometers at T0, T1, and T2. Secondary outcomes include (1) minutes of MPVA, low-intensity PA, and sedentary time in the past week; (2) step counts in the past week; (3) SOC levels, perceived pros, perceived cons, and perceived self-efficacy related to PA; (4) compliance to the web-based interventions; and (5) cognitive status measured at T0, T1, and T2. Intention-to-treat analysis will be used for data analysis. RESULTS: Recruitment started in November 2024. By February 2025, a total of 185 participants completed the baseline assessment and were randomly assigned to either the intervention group (n=93, 50.3%) or the control group (n=92, 49.7%). Recruitment will be completed by the end of June 2025. The follow-up assessment at T1 started in March 2025. Data collection is expected to be concluded in February 2026. CONCLUSIONS: The findings will extend the application of SOC and contribute to the evidence of the effectiveness of SOC-tailored and chatbot-delivered interventions. If the chatbot-delivered SOC-tailored intervention is proven effective to increase PA levels, it will require relative less resources to implement and maintain. It can be integrated into the existing WhatsApp groups operated by organizations providing services to older adults in Hong Kong and create public health impacts. TRIAL REGISTRATION: ClinicalTrial.gov: NCT06641492; https://clinicaltrials.gov/study/NCT06641492. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/68796.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.031 | 0.029 |
| Meta-epidemiology (narrow) | 0.006 | 0.004 |
| Meta-epidemiology (broad) | 0.012 | 0.006 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.004 | 0.003 |
| Scholarly communication | 0.005 | 0.004 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.008 | 0.009 |
| Insufficient payload (model declined to judge) | 0.087 | 0.013 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".