Trial Review

Did you know?

The ANZCTR now automatically displays published trial results and simplifies the addition of trial documents such as unpublished protocols and statistical analysis plans.

These enhancements will offer a more comprehensive view of trials, regardless of whether their results are positive, negative, or inconclusive.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12621000345886
Ethics application status
Approved
Date submitted
29/01/2021
Date registered
26/03/2021
Date last updated
26/03/2021
Date data sharing statement initially provided
26/03/2021
Type of registration
Retrospectively registered

Titles & IDs
Public title
Feasibility, Acceptability and Effectiveness of a Machine Learning Based Physical Activity Chatbot
Scientific title
Feasibility, Acceptability and Effectiveness of a Machine Learning Based Chatbot in Modulation of Physical Activity in Inactive Adults
Secondary ID [1] 303289 0
None
Universal Trial Number (UTN)
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Physical inactivity 320496 0
Condition category
Condition code
Public Health 318366 318366 0 0
Health promotion/education

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
The intervention used a chatbot deployed via Facebook Messenger to help participants increase their physical activity. A quasi-experimental design was conducted with outcomes evaluated at two time points: baseline and six weeks after participants started to use the chatbot. Participants provided their time preferences (up to 3 times per day) when they received a push message from the chatbot. Participants were encouraged to self-initiate contact with the chatbot as much as they could. Push notifications included different messages updating participants about their physical activity level at the time of delivery and also encouraging them to add physical activity to meet their daily goal. Participants could ask the chatbot about benefits of physical activity and were provided relevant sources of information. Participants used their own phone but were provided the Fitbit Flex 1 to measure their daily step. Adherence was assessed using questionnaires at post-intervention.

No educational material was developed for use with the chatbot. However, if participants asked for information about physical activity, the chatbot referred participants to relevant source of information from the internet so that participants could self-educate themselves.
Intervention code [1] 319590 0
Lifestyle
Intervention code [2] 319591 0
Treatment: Devices
Comparator / control treatment
No control group
Control group
Uncontrolled

Outcomes
Primary outcome [1] 326340 0
Step counts were measured by Fitbit Flex 1.
Timepoint [1] 326340 0
Baseline and week 6 post-intervention commencement
Primary outcome [2] 326653 0
Self-reported physical activity using the Active Australia Survey
Timepoint [2] 326653 0
Baseline and week 6 post-intervention commencement
Secondary outcome [1] 391086 0
Self-reported BMI
Timepoint [1] 391086 0
Baseline and week 6 post-intervention commencement

Eligibility
Key inclusion criteria
Be inactive (less than 20 minutes/day of moderate to vigorous physical activity), live in Australia, have internet access and a smartphone, be at least 18 years old, motivated to improve physical activity, not already participating in another physical activity program, not already owning and used a physical activity tracking device (e.g., pedometer, Fitbit, Garmin), and able to safely increase their activity levels.
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
Those with health conditions preventing them from increasing physical activity.

Study design
Purpose of the study
Treatment
Allocation to intervention
Non-randomised trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Not applicable
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Not applicable
Masking / blinding
Open (masking not used)
Who is / are masked / blinded?



Intervention assignment
Single group
Other design features
Phase
Not Applicable
Type of endpoint/s
Efficacy
Statistical methods / analysis
Post-hoc power calculation was conducted for Fitbit step counts using the following parameters: difference in means, standard deviations, and correlation between step counts at two time points. The post-hoc power for this study was 81.3%.

Generalized linear mixed models were run to identify changes in the outcomes. Normal distribution and identity link was used for BMI, Fitbit step counts, and total physical activity minutes. However, as total physical activity minutes were highly skewed, Box-Cox transformation analysis was conducted and as a result, a natural logarithm transformation was applied. Estimates for total physical activity minutes were converted back into ratios for the interpretative purposes. Empirical estimators were used to obtain robust standard errors. Binary distribution and logit link was used for the outcome of meeting physical activity guidelines. For each outcome, two models were run to generate crude estimates and estimates adjusted for sample characteristics including age, gender, marital status, years of schooling, ethnicity, household income, living area, work status, and work duration. Differences in BMI, step counts, and total physical activity minutes between follow-up and baseline were reported with 95%CI. Odds ratios (OR) and 95%CI were reported for meeting physical activity guidelines. All p-values were two-sided and considered significant if less than 0.05.

Recruitment
Recruitment status
Completed
Date of first participant enrolment
Anticipated
Actual
22/09/2020
Date of last participant enrolment
Anticipated
Actual
23/10/2020
Date of last data collection
Anticipated
Actual
15/12/2020
Sample size
Target
150
Accrual to date
Final
120
Recruitment in Australia
Recruitment state(s)
ACT,NSW,NT,QLD,SA,TAS,WA

Funding & Sponsors
Funding source category [1] 307706 0
University
Name [1] 307706 0
Central Queensland University
Country [1] 307706 0
Australia
Primary sponsor type
University
Name
Central Queensland University
Address
554-700 Yaamba Rd, Norman Gardens QLD 4701
Country
Australia
Secondary sponsor category [1] 308402 0
None
Name [1] 308402 0
Address [1] 308402 0
Country [1] 308402 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 307734 0
The Human Research Ethics Committee at the Central Queensland University
Ethics committee address [1] 307734 0
554-700 Yaamba Rd, Norman Gardens QLD 4701
Ethics committee country [1] 307734 0
Australia
Date submitted for ethics approval [1] 307734 0
13/01/2020
Approval date [1] 307734 0
04/02/2020
Ethics approval number [1] 307734 0
0000022181

Summary
Brief summary
Behavioural eHealth and mHealth interventions have been moderately successful in increasing physical activity. Therefore, there is still room for further improvement. Chatbots equipped with natural language processing can effectively interact and engage with users. Chatbots can also help continuously self-monitor physical activity levels using data from wearable body sensors and smartphones. However, there is lack of studies evaluating effectiveness of chatbot interventions on physical activity. The aim of this study was to investigate the feasibility, acceptability and effectiveness of an interactive machine learning based chatbot that uses natural language processing and adaptive goal setting to improve physical activity among inactive adults living in Australia.
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 108322 0
Dr Quyen To
Address 108322 0
Central Queensland University
554-700 Yaamba Rd, Norman Gardens QLD 4701
Country 108322 0
Australia
Phone 108322 0
+61 7 4930 6456
Fax 108322 0
Email 108322 0
[email protected]
Contact person for public queries
Name 108323 0
Quyen To
Address 108323 0
Central Queensland University
554-700 Yaamba Rd, Norman Gardens QLD 4701
Country 108323 0
Australia
Phone 108323 0
+61 7 4930 6456
Fax 108323 0
Email 108323 0
[email protected]
Contact person for scientific queries
Name 108324 0
Quyen To
Address 108324 0
Central Queensland University
554-700 Yaamba Rd, Norman Gardens QLD 4701
Country 108324 0
Australia
Phone 108324 0
+61 7 4930 6456
Fax 108324 0
Email 108324 0
[email protected]

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
10808Informed consent form  [email protected]
10809Ethical approval  [email protected]



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.