Trial Review

Registering a new trial?

To achieve prospective registration, we recommend submitting your trial for registration at the same time as ethics submission.

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
ACTRN12623000764639
Ethics application status
Approved
Date submitted
28/06/2023
Date registered
13/07/2023
Date last updated
13/07/2023
Date data sharing statement initially provided
13/07/2023
Type of registration
Prospectively registered

Titles & IDs
Public title
Digi-Predict Asthma: Digital predictors of asthma attacks
Scientific title
Physiological, behavioral and environmental predictors of asthma exacerbations: a prospective observational study using digital sensors and artificial intelligence
Secondary ID [1] 309247 0
Nil known
Universal Trial Number (UTN)
U1111-1293-9627
Trial acronym
DIGIPREDICT
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Asthma 330395 0
Condition category
Condition code
Respiratory 327233 327233 0 0
Asthma

Intervention/exposure
Study type
Observational
Patient registry
False
Target follow-up duration
Target follow-up type
Description of intervention(s) / exposure
This study is a prospective observational study to identify risk factors of asthma exacerbations. There is no active intervention. Unlike a traditional observational study, we are not assessing an exposure of interest and assessing outcome, rather we are measuring a diverse range of physiological, behavioural and environmental factors to evaluate their relationship(s) with asthma exacerbations.

These variables include medication data from dispensing records and smart inhalers; peak flow data from a bluetooth smart peak flow meter; nocturnal cough from a cough monitoring app on the participant's phone; and physiological variables (e.g. heart rate) collected by a smart watch. All smart devices will be provided to the study participants. Diet, asthma control and general wellbeing will also be assessed using participant questionnaires.

There will be two study visits in person – one at baseline (enrolment) and one at 6 months. The visits can take place either at The University of Auckland Grafton campus or at the participant's place of residence or other nominated location, whichever suits the participant best. These will be conducted by a member of the research team. At the initial visit at start of the study, participants will be provided with the required smart devices and instructions and a demonstration of how to use these devices. It is expected that these study visits would take up to 2 hours.

In between these visits, participants will have some short questionnaires to complete either self-completed online remotely or researcher-assisted over the phone every 2 weeks for 6 months post-enrolment. These will take up to 5-10 minutes to complete.

For the smart devices - participants will be given: a smart inhaler, smart peak flow meter, and smart watch. Some participants - those who already use a spacer on a regular basis and are familiar with the small volume spacer device - will be offered a smart (digital) spacer to use also instead of a traditional spacer. This measures inspiratory flow and inhalation. These devices will need to be paired to the participant's smart phone and their relevant app downloaded for it to work. Most of the data from the smart devices will be gathered passively i.e., participants won’t need to do anything extra other than use the devices as normal with their inhaler(s) and using the peak flow meter daily instead of your usual peak flow meter. Participants will need to use the cough monitoring app each night.

Participants may need to open the apps a few times a week to ensure the data syncs successfully between the devices and the phone. Participants will need to wear the smartwatch during sleep and during awake hours everyday. Exceptions are made for wearing the smart watch for example when charging the watch and when having a shower or when there is any discomfort. Participants will also receive a quick text or phone call each month (whichever they prefer) to see how they are doing with the technology and devices.

Researchers can access the collected data from the devices from the devices which will sync data with the related apps. To allow transfer of data from the smart devices, Bluetooth will need to be switched on. To send the data (upload), participants need to connect to the internet a few times a week to allow the data to be uploaded.

This study involves collection of a large amount of information using different technologies to help us build a model using artificial intelligence (AI) to predict when an asthma attack might occur. We will use a type of AI called machine learning to analyse the data collected to identify patterns related to asthma attacks. The collected data will be entered into a machine learning component external to participants. No AI data will be fed back to participants at this stage of the study.
Intervention code [1] 326426 0
Early Detection / Screening
Comparator / control treatment
No control group
Control group
Uncontrolled

Outcomes
Primary outcome [1] 335223 0
Primary endpoint will be defined as number of asthma exacerbations per participant whilst in follow-up (over 6 months). This data will be assessed by participant self-report and confirmed with participant's clinical records. An exacerbation will be defined as per the criteria in the American Thoracic Society (ATS) and European Respiratory Society (ERS) definitions for exacerbations, as either severe or moderate.

Note that collected data from the smart devices and questionnaires (e.g. diet) will be variables that will be input into the prediction model for this primary outcome.
Timepoint [1] 335223 0
Six months post-enrolment
Secondary outcome [1] 423543 0
Asthma control will be measured every two weeks between 1 and 6 months using the Asthma Control Test (ACT), which will be self-completed by participants.
Timepoint [1] 423543 0
Every 2 weeks for up to 6 months post-enrolment
Secondary outcome [2] 423544 0
Wellbeing will be assessed every 2 weeks by the WHO-5 index.
Timepoint [2] 423544 0
Every 2 weeks for up to 6 months post-enrolment
Secondary outcome [3] 423545 0
Attendance at the Emergency Department will be obtained by participant self-report and confirmed from the participant's clinical records.
Timepoint [3] 423545 0
6 months post-enrolment
Secondary outcome [4] 423546 0
Unscheduled visits to either the participant’s General Practitioner or to the Accident and Emergency Clinic will be obtained by participant self-report as a composite secondary outcome.
Timepoint [4] 423546 0
6 months post-enrolment
Secondary outcome [5] 423547 0
Productivity loss and activity impairments as a composite outcome.
This will be determined by days absent from school or work for participants and/or caregivers using self-report. For days absent from school, this will be recorded as the number of half days attended each term. The number of days absent from all causes will be used as the end point rather than days absent due to asthma to prevent the risk of misclassification. This may occur, for example, if a child was classified as being absent from school due to a respiratory tract infection, when in fact the cough was secondary to asthma. Information on days absent due to asthma (as per self-report) will also be collected. We will also use the validated Work Productivity and Activity Impairment questionnaire to measure activity impairments.
Timepoint [5] 423547 0
6 months post-enrolment
Secondary outcome [6] 423548 0
Participant acceptability of the devices used in the study will be assessed using an adapted version of the System Usability Scale and questionnaire informed by the Technology Acceptance Model.
Timepoint [6] 423548 0
6 months post-enrolment

Eligibility
Key inclusion criteria
(i) Have a physician diagnosis of asthma;
(ii) Previous history of an asthma attack in the last 12 months as per ATS/ERS definitions(1);
(iii) Currently managing asthma with either preventative or relief medication delivered via either a pressurized metered dose inhaler (pMDI), Turbuhaler™, Ellipta™ or other device compatible with the Hailie® range of digital inhaler sensors;
(iv) Residing in Auckland or Rotorua, or able to travel to either of these places for the study visits;
(v) Able to provide informed consent and be able to follow study procedures or protocols;
(vi) Own or be willing to use a smartphone with Bluetooth capability that is compatible with the study devices and can host the study apps;
(vii) Be available and able to use the technologies for a period of 6 months. Individuals on treatment with other concomitant asthma medication will be eligible for inclusion; and
(viii) Aged 12 to 65 years at the time of presentation to hospital or referral.

(1) Reddel HK, Taylor DR, Bateman ED, Boulet L-P, Boushey HA, Busse WW, et al. An Official American Thoracic Society/European Respiratory Society Statement: Asthma Control and Exacerbations. American Journal of Respiratory and Critical Care Medicine. 2009;180:59-99. doi: 10.1164/rccm.200801-060ST.
Minimum age
12 Years
Maximum age
65 Years
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
(i) they have a diagnosis of lung disease other than asthma e.g. COPD, bronchiectasis, cystic fibrosis, bronchopulmonary dysplasia;
(ii) (ii) smoking history of >10 pack years; or
(iii) (iii) have a significant health condition or disability affecting ability to follow study procedures or protocols.

Study design
Purpose
Natural history
Duration
Longitudinal
Selection
Defined population
Timing
Prospective
Statistical methods / analysis
Traditional statistical regression techniques and ML methods will be applied to examine associations between collected data variables from the smart devices and patient questionnaires, and asthma exacerbations, Results from the two approaches will be compared. To train the algorithm, participants’ data will be used. The dataset will be divided randomly into two: 80% training, 20% testing with 10-folds cross-validation. Embeddings, weight sharing and incorporating invariance will be used to improve model speed and performance for real-life use. Predictors will be validated by comparing variable importance to coefficients from the regression. We will test model performance within three groups: Maori vs Pacific vs non-Maori and non-Pacific samples to ensure equal performance by assessing discrimination, calibration and internal validity. We will also test performance in young persons (<18 years) vs adults.

Analyses will be performed on the intention-to-treat population. In cases where items of data are missing for a particular participant, available data for that participant will be included for the relevant data summaries or analyses and a description of the missing data provided. No estimation or interpolation of missing data will be made. The analyses will be based on the 6 month data.




Recruitment
Recruitment status
Not yet recruiting
Date of first participant enrolment
Anticipated
31/07/2023
Actual
Date of last participant enrolment
Anticipated
1/08/2025
Actual
Date of last data collection
Anticipated
1/02/2026
Actual
Sample size
Target
300
Accrual to date
Final
Recruitment outside Australia
Country [1] 25615 0
New Zealand
State/province [1] 25615 0

Funding & Sponsors
Funding source category [1] 313439 0
Government body
Name [1] 313439 0
Health Research Council
Country [1] 313439 0
New Zealand
Funding source category [2] 314183 0
Charities/Societies/Foundations
Name [2] 314183 0
Auckland Medical Research Foundation
Country [2] 314183 0
New Zealand
Primary sponsor type
University
Name
The University of Auckland
Address
Private Bag 92019
Auckland 1142
New Zealand
Country
New Zealand
Secondary sponsor category [1] 316100 0
None
Name [1] 316100 0
Address [1] 316100 0
Country [1] 316100 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 312645 0
Southern Health and Disability Ethics Committee
Ethics committee address [1] 312645 0
Ministry of Health
Health and Disability Ethics Committees
PO Box 5013
Wellington 6140
Ethics committee country [1] 312645 0
New Zealand
Date submitted for ethics approval [1] 312645 0
Approval date [1] 312645 0
13/02/2023
Ethics approval number [1] 312645 0
2023 FULL 13541

Summary
Brief summary
Asthma rates in New Zealand are among the highest in the world, particularly in Maori and Pacific. Asthma attacks, or exacerbations, are the leading cause of deaths. Attacks are highly preventable if detected early. Unfortunately, there are currently no reliable ways of predicting attacks. Evidence suggests that there are changes that occur physiologically and behaviourally in the days and weeks preceding an attack, but these are not readily identified without technology. This aim of this study is to use a smart technologies to better predict asthma attacks by collecting real-time data from smart technologies such as smart watches and smart inhalers, and combining with NZ health and environmental datasets. We will then use artificial intelligence (AI) to produce a risk prediction model.

The study aims to recruit 300 people, aged 12-65 years, with a previous asthma exacerbation in the last 12 months. Participants will be recruited from primary and secondary care in two regions in New Zealand. Each participant will be given a smart watch, smart peak flow meter, smartinhaler, and cough monitor to use regularly over 6 months with fortnightly questionnaires on asthma control and wellbeing. The occurrence of asthma attacks will be detected by self-report when they occur and from clinical records. The collected data, along with environmental data on weather and air quality, will be analysed to develop a risk prediction model for asthma attacks. We will also measure participant acceptability of the devices at the end of the study.
Trial website
https://www.asthma.org.nz/pages/predicting-asthma-attacks
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 125406 0
Dr Amy Chan
Address 125406 0
School of Pharmacy
Level 3, Building 505
Faculty of Medical and Health Sciences,
The University of Auckland,
85 Park Road, Grafton, Auckland 1023
New Zealand
Country 125406 0
New Zealand
Phone 125406 0
+64 3737599
Fax 125406 0
Email 125406 0
[email protected]
Contact person for public queries
Name 125407 0
Amy Chan
Address 125407 0
School of Pharmacy
Level 3, Building 505
Faculty of Medical and Health Sciences,
The University of Auckland,
85 Park Road, Grafton, Auckland 1023
New Zealand
Country 125407 0
New Zealand
Phone 125407 0
+64 3737599
Fax 125407 0
Email 125407 0
[email protected]
Contact person for scientific queries
Name 125408 0
Amy Chan
Address 125408 0
School of Pharmacy
Level 3, Building 505
Faculty of Medical and Health Sciences,
The University of Auckland,
85 Park Road, Grafton, Auckland 1023
New Zealand
Country 125408 0
New Zealand
Phone 125408 0
+64 3737599
Fax 125408 0
Email 125408 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
Currently no ethics approval has been gained to allow sharing of IPD


What supporting documents are/will be available?

No Supporting Document Provided



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.