Skip to content

Exploring the Feasibility of Voice-Controlled Intelligent Personal Assistants for delivery of home-based lifestyle interventions in older adults with type II diabetes

Exploring the Feasibility of Voice-Controlled Intelligent Personal Assistants for delivery of home-based diet and exercise interventions in older adults with type II diabetes

Status
Completed
Phases
Unknown
Study type
Interventional
Source
ANZCTR
Registry ID
ACTRN12621000307808
Enrollment
50
Registered
2021-03-19
Start date
2021-11-01
Completion date
2021-11-08
Last updated
2022-03-07

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

Conditions

None listed

Brief summary

We will conduct a 12-week feasibility and pilot randomised controlled trial of a personalised exercise and dietary intervention delivered via VIPAs in adults aged 50 to 75 years with oral hypoglycaemic agent-controlled type 2 diabetes mellitus (T2DM). Primary Aim: Our primary aim is to assess study processes including recruitment and retention rates, adherence to the exercise and nutrition intervention, acceptability and adverse events. Secondary Aim: Secondary aims are to compare changes in scores for the Diabetes Self-Management Questionnaire (DSMQ), objectively-assessed physical activity, self-reported fruit and vegetable intakes, health-related quality of life (EQ-5D-5L) and productivity, compared to a control group receiving general advice on physical activity and healthy eating via email. Hypotheses: Our primary hypothesis is that a 12-week home-based personalised exercise and dietary intervention delivered by VIPA will be feasible and acceptable, as evidenced by greater than or equal to 70% retention in the study, completion of greater than or equal to 66% of prescribed exercise and dietary advice, recruitment greater than or equal to 20% of eligible participants and overall themes of acceptability expressed. We also hypothesise that the VIPA group will demonstrate improvements in diabetes self-management, objectively-determined physical activity, health-related quality of life, self-reported fruit and vegetable intakes and direct and indirect productivity. We anticipate that these preliminary data will support the need for future studies to explore cost-effectiveness of this telehealth approach in Type 1 or 2 insulin dependent diabetes mellitus and gestational diabetes. This innovative model of service delivery has the potential to markedly reduce health care practitioner workloads and telehealth costs in type 2 diabetes management, while increasing empowerment and accessibility to evidence-based programs for patients most at need

Interventions

Working with an industry partner, we have developed innovative web-based software (Buddy Link) which allows Health professionals to deliver individually tailored lifestyle programs via an Amazon Alexa Echo Show 8 (Alexa) in patients' homes. Health professionals use our software to select various lifestyle behaviour instructions from a wide range of evidence-based options (or can easily create their own program), and these instructions are scheduled for broadcasting (using video and audio) by the

Working with an industry partner, we have developed innovative web-based software (Buddy Link) which allows Health professionals to deliver individually tailored lifestyle programs via an Amazon Alexa Echo Show 8 (Alexa) in patients' homes. Health professionals use our software to select various lifestyle behaviour instructions from a wide range of evidence-based options (or can easily create their own program), and these instructions are scheduled for broadcasting (using video and audio) by the Voice-Controlled Intelligent Personal Assistants (VIPA) at times specified by the health professional. The intervention group will receive, via courier, an Alexa and will also be provided with instructions on connecting the device to their home WiFi and using standard Alexa functions like using voice instructions to set reminders and search for information. They will also receive access to our health professional facing software “Buddy Link” which enables delivery of an individualised lifestyle program, at specified times throughout the day, using video demonstrations, and audio and written instructions. The VIPA will broadcast questions to determine whether the participant completed the activity, self-perceived exertion, self-efficacy, and any concerns. Participants can respond to all questions using natural conversation. The exercise physiologist will prescribe a personalised exercise program delivered by the Alexa to the intervention group for 12-weeks. Exercises will be selected by the exercise physiologist using the health professional interface in Buddy Link, and individually broadcast to participants via Alexa at specified times throughout the day, using video demonstrations, and audio and written instructions based on Exercise and Sport Science Australia guidelines. The prescribed exercise program will use body weight or additional resistance such as weight plates, dumbbells, theraband, weight vests etc if the participant has access to them unless contraindicated. Participants will perform up to 3 sets of 5 repetitions of upper limb and lower limb exercises at a moderate-intensity of approximately 4-6 on the 10-point modified Rating of Perceived Exertion (RPE) scale. Each participant will be encouraged to increase the load of all the prescribed exercises each session while maintaining the desired intensity if able. throughout the intervention duration. Each session will be 20-30 minutes in duration. The protocol will initially deliver two exercise sessions per week for the first four weeks of the intervention, three exercise sessions per week for the second four weeks, and four exercise sessions per week for the final four weeks. Following each exercise, the Alexa will broadcast questions to determine whether the participant completed the exercise, their self-perceived exertion, and any concerns (e.g. Pain, dizziness). Participants’ responses to these questions will be recorded and saved to the Buddlink database, enabling the exercise physiologist to review weekly and modify or progress exercise prescriptions as required. In addition, we have developed a novel automated “decision to exercise tree” based on the current recommendations of American Diabetes Association. VIPA devices will provide Blood glucose level (BGL) monitoring reminders pre and post each exercise session for the first 3 sessions (minimum) to establish exercise response/patterns/control. For example, the decision tree not deliver exercise if participants have signs or symptoms of hypoglycaemia or have self-reported BSL less than 4.0mmol/L or are not feeling well enough to exercise. Instead the decision tree will deliver audio and video instructions on how to self-administer their personalised hypoglycaemia action plan. If a pattern of three sessions in a row of self-reported BSLs less than 4.0mmol/L or greater than 15.0mmol/L occurs the decision tree system will request a referral to their GP or endocrinologist for completion of a Diabetes Management Plan. The study dietician will prescribe a healthy eating program delivered by the Alexa to the intervention group for 12-weeks to increase whole grains, vegetable and fruit daily intake in line with the Australian Dietary Guidelines. In addition, generalised dietary educational videos tailored for type 2 diabetes management will be displayed by Alexa. Examples of videos delivered by Alexa include “meal planning”, “making shopping lists using VIPA” and “healthy snacking”. We will access the Buddy Link platform to quantify the regularity with which participants watch these health eating videos (this will be determined by voice response from participants to questions we deliver via VIPAs, e.g. “Did you find this video useful? Please respond yes, no, or I did not watch the video”). The protocol will deliver two videos per week for the duration of the 12 week intervention.

Sponsors

Deakin University
Lead SponsorUniversity

Study design

Allocation
Randomised controlled trial
Intervention model
Parallel
Primary purpose
Treatment
Masking
Blinded (masking used) (Investigator, Outcomes Assessor)

Eligibility

Sex/Gender
All
Age
50 Years to 75 Years
Healthy volunteers
No

Inclusion criteria

Prospective participants must be aged 50-75 years; English-speaking; have a body mass index greater than or equal to 30 kg/m2, able to walk across a room unaided; with access to a smartphone connected to an internet network; willing to complete a 12-week lifestyle intervention and also be willing to participate should they be randomised to either intervention arm.

Exclusion criteria

Participants are ineligible if they report currently being unable to walk across a room unaided; are non-English speaking or have difficulty communicating with study personnel or a VIPA device due to speech or hearing problems; unwilling to be randomised; planning to be away from the VIPA device for greater than or equal to 4 weeks during the 12-week intervention period; severe knee or hip osteoarthritis (awaiting a joint replacement) that would interfere with ability to complete exercise; recent fracture (past 3 months) limiting exercise; renal disease requiring dialysis; and any other disorder of such severity that life expectancy is less than 12 months, or any cognitive or physical impairment or disability that in the opinion of the study investigators would result in the participant having difficulty interacting with Buddy Link or performing unsupervised exercise safely. Participants who answer ‘yes’ to any of six questions on the Exercise and Sports Science Australia (ESSA) pre-exercise screening tool will also be excluded. This is a validated and recommended pre-exercise screening tool for ensuring safety for exercise at moderate exertion, endorsed by Australia’s peak exercise body (ESSA).

Outcome results

None listed

Source: ANZCTR · Data processed: Feb 5, 2026