Development and Feasibility of a Family-Based Health Behavior Intervention Using Intelligent Personal Assistants: Randomized Controlled Trial

Editorial reading guide · Collaboration

Would an assistant’s reminders fit our shared health routine, and would we actually use its health features?

Tested whether families could use an intelligent personal assistant for shared health-behavior support and whether a larger trial was workable. Interaction logs showed a substantial distinction between using the device and using it for the intended health purpose: in phase two, only 11% of interactions were health-related. Family access, reminders, researcher support, and study burden all mattered to feasibility.

Study context

26 Northern Irish families with 34 children across two 12-week feasibility phases; the first phase recruited through a community weight-management program

Age / stage: Children 5–12 years

Read with three questions

1. What did interaction logs reveal that overall device-use counts would conceal?
2. How did recruitment through an existing program differ from recruitment of other families?
3. Which retention, measurement, and usability issues would have to improve before testing health effectiveness?

Keep the limits in view

This small feasibility trial was not powered to establish health effects and did not statistically test effectiveness. Phase-one attrition and low accelerometer completion complicated evaluation. Internet/device access and concurrent program participation limit generalizability; retention and enjoyable use must not be read as proof of healthier behavior.