MWIQ - AI-powered decision support tool for diabetes clinicians
| ISRCTN | ISRCTN17422256 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN17422256 |
| IRAS number | 305267 |
| Secondary identifying numbers | NIHR - AI_AWARD02234, IRAS 305267 |
- Submission date
- 11/01/2022
- Registration date
- 27/01/2022
- Last edited
- 09/02/2024
- Recruitment status
- No longer recruiting
- Overall study status
- Completed
- Condition category
- Nutritional, Metabolic, Endocrine
Plain English Summary
Background and study aims
MyWay IQ (MWIQ) is an electronic software application that provides advice (clinical decision support) around the diagnosis and treatment of patients with diabetes and is delivered as part of the MyWay Clinical (MWC) and MyWay Diabetes (MWD) platforms. MWC is designed for healthcare professional (HCP) use, whilst MWD is designed for people with diabetes. MWIQ can also be used within other systems that are used in patient care (Electronic Health care Records (EHR)). MWIQ predicts the type of diabetes and the risk of complications and provides advice about the best medications to use for individual patients. This study will assess how easy it is to use the system (usability), how useful the system is to users, and how safe the advice it supplies is. This information will help the system developers register the product with the Medicines and Healthcare products Regulatory Agency (MHRA), who are responsible for regulating medical devices, including software applications.
Who can participate?
Healthcare professionals involved in the care of people with diabetes.
What does the study involve?
Participants will be asked to use the system within their normal diabetes clinics, and provide feedback via questionnaires and interviews.
What are the possible benefits and risks of participating?
Participants may find that the system allows them to make a more informed decision about an individual patient's medical management (e.g. the type of medicine to prescribe). If the system results in improved clinical care, then this will benefit patients and the wider population of people with diabetes. Participants may find that the system does not help with their clinical decision-making or is in some way intrusive during the normal clinical consultation. These are the sort of issues that the researchers will learn about through the feedback that they gather. There is a risk that the system does not work as expected, however, there is a process in place to check the system is working well before it is used in the clinic.
Where is the study run from?
University of Dundee (UK)
When is the study starting and how long is it expected to run for?
August 2021 to July 2024
Who is funding the study?
National Institute for Health Research (NIHR) (UK)
Who is the main contact?
Dr Nicholas Conway
nicholas.conway@nhs.scot
Contact information
Principal Investigator
c/o MWDH Ltd
MacKenzie Building
Kirsty Semple Way
Dundee
DD2 4BF
United Kingdom
| Phone | +44 (0)1382 660111 |
|---|---|
| nicholas.conway@nhs.scot |
Study information
| Study design | Open-label single-arm single-site feasibility study |
|---|---|
| Primary study design | Interventional |
| Secondary study design | Feasibility study |
| Study setting(s) | GP practice |
| Study type | Prevention |
| Participant information sheet | 40921_PIS_V1.0_20Jan22_ClinicianExperts.pdf |
| Scientific title | MyWay IQ (MWIQ) - safety and efficacy testing of a diagnosis and precision medicine tool for diabetes management |
| Study acronym | AI Award MWIQ |
| Study hypothesis | Current study hypotheses as of 18/05/2023: Primary hypothesis: MyWay IQ (MWIQ) is safe within both simulated and real-world clinical environments. Secondary hypotheses: MWIQ is usable; MWIQ is useful; MWIQ has an impact on clinician prescribing and diagnosing; MWIQ has an impact on patient metabolic outcomes. Previous study hypotheses: Primary hypothesis: MyDiabetes IQ (MDIQ) is safe within both simulated and real-world clinical environments. Secondary hypotheses: MDIQ is usable; MDIQ is useful; MDIQ has an impact on clinician prescribing and diagnosing; MDIQ has an impact on patient metabolic outcomes. |
| Ethics approval(s) |
Approved 08/01/2024, North of Scotland Research Ethics Committee (Summerfield House, 2 Eday Road, Aberdeen, AB15 6RE, United Kingdom; None available; gram.nosres@nhs.scot), ref: 23/NS/0134 |
| Condition | Diabetes |
| Intervention | Current interventions as of 18/05/2023: This feasibility investigation will determine whether an AI-powered clinical decision support tool for diabetes is (a) safe; (b) usable; (c) alters clinician prescribing or diagnostic behaviour and (d) leads to any changes in metabolic outcomes for patients under the care of clinicians using the tool. This qualitative evaluation will assess usability/safety; examine data relating to clinician behaviour; and analyse routinely-collected clinical patient data. 1. Qualitative evaluation of usability/engagement and safety of MWIQ (online questionnaires, qualitative interviews, system logs, navigation data). Run charts will be used to record usability and safety issues 2. Assess the impact of MWIQ on clinical workflow (descriptive statistics of prescribing/diagnostic data) 3. Measure clinical outcomes (descriptive statistics of metabolic outcomes, including BP and HbA1c, compared with background patient data) Previous interventions: This feasibility investigation will determine whether an AI-powered clinical decision support tool for diabetes is (a) safe; (b) usable; (c) alters clinician prescribing or diagnostic behaviour and (d) leads to any changes in metabolic outcomes for patients under the care of clinicians using the tool. This qualitative evaluation will assess usability/safety; examine data relating to clinician behaviour; and analyse routinely-collected clinical patient data. 1. Qualitative evaluation of usability/engagement and safety of MDIQ (online questionnaires, qualitative interviews, system logs, navigation data). Run charts will be used to record usability and safety issues 2. Assess the impact of MDIQ on clinical workflow (descriptive statistics of prescribing/diagnostic data) 3. Measure clinical outcomes (descriptive statistics of metabolic outcomes, including BP and HbA1c, compared with background patient data) |
| Intervention type | Mixed |
| Primary outcome measure | 1. System safety: frequency of user-flagged safety concerns measured using a combination of patient-level questionnaires, weekly questionnaires and unsolicited issues highlighted by participants. Specific measures include: 1.1. Number of risks identified by patient-level questionnaires (interim and cumulative total), expressed as a percentage of the number of patient records reviewed 1.2. Clinical safety score recorded with a weekly questionnaire for each of the MWIQ features (diabetes sub-type predictor, diabetes drug response predictor, and diabetes complications predictor) as well as overall experience. The score will be recorded on a 5-point Likert scale and the distribution of scores will be expressed as a percentage of responses received. 1.3. Clinical plausibility score recorded with a weekly questionnaire for each of the MWIQ features (diabetes sub-type predictor, diabetes drug response predictor, and diabetes complications predictor) as well as overall experience. The score will be recorded on a 5-point Likert scale and the distribution of scores will express as a percentage of responses received. Free text comments will be analysed for emergent themes All outcomes measured at the end of the study period in addition to monthly interim analysis, as part of the Data and Safety Monitoring Board (DSMB) process |
| Secondary outcome measures | 1. System usability measured using a combination of weekly questionnaires, qualitative data (free text questionnaire comments and semi-structured interviews) and validated questionnaire (system usability scale). Specific measures include: 1.1. Overall ease of use recorded on a 5-point Likert scale. The distribution of scores will be expressed as a percentage of responses received. 1.2. Free text comments and semi-structured interview transcripts will be analysed for emergent themes. 1.3. System Usability Scale to be administered at the end of the study period. 2. System usefulness measured using a combination of weekly questionnaires and qualitative data (free text questionnaire comments and semi-structured interviews). Specific measures include: 2.1. Clinical usefulness score recorded with a weekly questionnaire for each of the MWIQ features (diabetes sub-type predictor, diabetes drug response predictor, and diabetes complications predictor) as well as overall experience. The score will be recorded on a 5-point Likert scale and the distribution of scores will be expressed as a percentage of responses received. 3. System impact measured using a combination of routinely collected clinical data, system navigation data and weekly questionnaire data. Specific measures include: 3.1. Number of patients with the following care processes recorded within the past 12 months: HbA1c, blood pressure (BP), serum cholesterol, serum creatinine, urinary albumin, foot screening, Body Mass Index (BMI) and smoking status. The number of each will be expressed as a percentage of the total number of people with diabetes seen by participants within the study period. 3.2. Number of patients meeting the following treatment targets: HbA1c <58 mmol/mol, blood pressure ≤140/80 mmHg, cholesterol <5 mmol/l, as well as the number meeting all three targets. The number of each will be expressed as a percentage of the total number of people with diabetes seen by participants within the study period. 3.3. Number of patients with reclassified diabetes diagnosis, expressed as a percentage of the total number of people with diabetes seen by participants within the study period. 3.4. Subjective assessment of system impact recorded on a 5-point Likert scale. The distribution of scores will be expressed as a percentage of responses received. 3.5. Free text comments submitted via questionnaire analysed for emergent themes 3.6. Number of participant interactions with the system and time spent within the system (cumulative total and per patient) All outcomes measured at the end of the study period in addition to monthly interim analysis, as part of the Data and Safety Monitoring Board (DSMB) process |
| Overall study start date | 02/08/2021 |
| Overall study end date | 31/07/2024 |
Eligibility
| Participant type(s) | Patient, Health professional |
|---|---|
| Age group | Adult |
| Lower age limit | 18 Years |
| Sex | Both |
| Target number of participants | 2 clinician participants per practice x 8 GP practices = 16 clinicians participants plus 3 validator participants (diabetes specialists) who will validate MDIQ decisions = total 19 participants (all healthcare professionals) |
| Participant inclusion criteria | For healthcare professionals (HCPs), the inclusion criteria are: 1. Male or female 2. 18 years or over 3. GPs involved in the care of people with diabetes within Greater Manchester 4. Able to provide informed consent Patient data from four GP practices will be utilised - data will be from patients who are: 1. Male or female 2. Aged 18 years or over 3. Diagnosis of diabetes (any type) |
| Participant exclusion criteria | Does not meet the inclusion criteria |
| Recruitment start date | 08/01/2024 |
| Recruitment end date | 28/07/2024 |
Locations
Countries of recruitment
- England
- United Kingdom
Study participating centres
Northenden
Manchester
M22 4DH
United Kingdom
Hulme
Manchester
M15 4EA
United Kingdom
Simonsway
Wythenshawe
Manchester
M22 5RX
United Kingdom
Monton Street
Moss Side
Manchester
M14 4GP
United Kingdom
Wythenshawe
Manchester
M23 9AB
United Kingdom
Wythenshawe
Manchester
M23 9JH
United Kingdom
Northern Moor
Manchester
M23 0PH
United Kingdom
Brooklands
Manchester
M23 9RL
United Kingdom
Sponsor information
Industry
MacKenzie Building
Kirsty Semple Way
Dundee
DD2 4BF
Scotland
United Kingdom
| Phone | +44 (0)7904 154101 |
|---|---|
| debbie.wake@mwdh.co.uk | |
| Website | https://www.mywaydigitalhealth.co.uk |
Funders
Funder type
Government
Government organisation / National government
- Alternative name(s)
- National Institute for Health Research, NIHR Research, NIHRresearch, NIHR - National Institute for Health Research, NIHR (The National Institute for Health and Care Research), NIHR
- Location
- United Kingdom
Results and Publications
| Intention to publish date | 31/01/2025 |
|---|---|
| Individual participant data (IPD) Intention to share | No |
| IPD sharing plan summary | Data sharing statement to be made available at a later date |
| Publication and dissemination plan | 1. The Clinical Investigation Plan is drafted but not yet published and will be made available at the appropriate point, given the commercially sensitive nature of the project effort 2. Planned publication in a high-impact peer-reviewed journal |
| IPD sharing plan | The data-sharing plans for the current study are unknown and will be made available at a later date |
Study outputs
| Output type | Details | Date created | Date added | Peer reviewed? | Patient-facing? |
|---|---|---|---|---|---|
| Participant information sheet | Clinician Experts version 1.0 |
20/01/2022 | 20/01/2022 | No | Yes |
| Participant information sheet | Clinician Participants version 1.0 |
20/01/2022 | 20/01/2022 | No | Yes |
Additional files
- 40921_PIS_V1.0_20Jan22_ClinicianParticipants.pdf
- Clinician Participants
- 40921_PIS_V1.0_20Jan22_ClinicianExperts.pdf
- Clinician Experts
Editorial Notes
09/02/2024: Ethics approval added.
16/11/2023: The following changes were made to the study record:
1. The recruitment start date was changed from 01/11/2023 to 08/01/2024.
2. The recruitment end date was changed from 31/03/2024 to 28/07/2024.
3. The overall study end date was changed from 30/04/2024 to 31/07/2024.
17/08/2023: The following changes were made to the study record:
1. The recruitment start date was changed from 01/08/2023 to 01/11/2023.
2. The recruitment end date was changed from 31/12/2023 to 31/03/2024.
3. The overall study end date was changed from 31/01/2024 to 30/04/2024.
18/05/2023: The following updates have been made to the study record and the plain English summary updated accordingly:
1. The public title has been changed from "MyDiabetes IQ - AI-powered decision support tool for diabetes clinicians" to "MWIQ - AI-powered decision support tool for diabetes clinicians".
2. The scientific title has been changed from "MyDiabetes IQ - safety and efficacy testing of a diagnosis and precision medicine tool for diabetes management" to "MyWay IQ (MWIQ) - safety and efficacy testing of a diagnosis and precision medicine tool for diabetes management".
3. The acronym has been changed from "AI Award MDIQ" to "AI Award MWIQ".
4. The study hypotheses have been changed.
5. The overall study end date has been changed from 31/10/2024 to 31/01/2024.
6. The interventions have been changed.
7. The recruitment start date has been changed from 01/02/2023 to 01/08/2023.
8. The recruitment end date has been changed from 30/04/2023 to 31/12/2023.
01/11/2022: A contact was removed.
19/10/2022: The following changes were made to the trial record:
1. The recruitment start date was changed from 01/08/2022 to 01/02/2023.
2. The recruitment end date was changed from 31/10/2022 to 30/04/2023.
3. The overall trial end date was changed from 28/07/2023 to 31/10/2024.
4. The intention to publish date was changed from 28/01/2024 to 31/01/2025.
18/05/2022: The following changes were made to the trial record:
1. The recruitment start date was changed from 01/03/2022 to 01/08/2022.
2. The recruitment end date was changed from 31/05/2022 to 31/10/2022.
3. The target number of participants was changed from '2 clinician participants per practice x 4 GP practices = 8 clinicians; 3 clinician experts (diabetes specialists) who will validate MDIQ decisions = total 11 participants (all healthcare professionals)' to '2 clinician participants per practice x 8 GP practices = 16 clinicians participants plus 3 validator participants (diabetes specialists) who will validate MDIQ decisions = total 19 participants (all healthcare professionals)'.
4. The Park Medical Practice, Brooklands Medical Practice, Northern Moor Medical Practice, and Woodlands Medical Practice were added as trial participating centres.
20/01/2022: Trial's existence confirmed by the NIHR.
