POWER2DM: Predictive model-based decision support system for diabetes patient empowerment

Here, Dr Albert A. de Graaf discusses a novel predictive model-based decision support system to aid diabetes patients in the complex task of self-management

Diabetes increasingly leads to societal challenges in Europe and across the world.

The prevalence of diabetes, a chronic disease that occurs either when the pancreas does not produce enough insulin or when the body cannot effectively use the insulin it produces, has increased markedly over the last 30 years. Suboptimal glycemic regulation potentially leads to serious diabetes-related complications, by causing damage to many of the body’s systems, especially the nerves and blood vessels.

The global diabetes prevalence in 2019 is estimated to be 9.3% (463 million people), with a rising trend. Diabetes-related healthcare costs in Europe were 161.4 billion US$ in 2019.

Adopting and maintaining a healthy lifestyle is key in preventing diabetes complications

Diabetes can be treated and its consequences avoided or delayed with diet, physical activity, medication and regular screening and treatment for complications. However, despite the increasing availablitity of solutions to monitor and control blood glucose levels, considered a key strategy to prevent diabetic complications, only a minority of diabetic patients currently reach their target of glycemic control, with a large percentage of cases arising due to an unhealthy lifestyle.

There is an urgent need to develop cost-effective intervention solutions that support diabetes patients in their daily effort of controlling their blood glucose levels.

Project POWER2DM was conceived to develop, test and validate exactly such a solution, answering to the EU’s Horizon2020 call PHC-28-2015: Self-management of health and disease and decision support systems based on predictive computer modelling used by the patient him or herself.

The POWER2DM (Predictive model-based decision support for diabetes patient empowerment) project proposal was awarded the highest score among 184 submitted proposals and started on the 1st of February 2016 (EU grant is 5M Euro; ref. H2020-PHC-689444). The project will end July 31, 2020. This project involves the research and medical partners TNO (coordinator), IDK, SRDC, LUMC, SAS, and SRFG, and SME’s PrimeData and iHealth.

Key aspects of the POWER2DM solution for both Type-1 and Type-2 diabetes patients

Patient empowerment created by making appropriate support tools available to the patient in the self-control periods;

Personalised feedback and support on a 24/7 basis;

Personalised guidance based on predictive computer models during treatment planning;

Well integrated with care systems to facilitate data sharing with caregivers;

Fully compliant to the EU GDPR (2-factor authentication, software deployment in secure Azure cloud, secure Personal Data Store (PDS), secure HL7 FHIR based RESTful service for all POWER2DM components to access and update the recorded data);

And compliant to new EU medical device regulations (registered as Class I Active Medical Device).

The POWER2DM consortium has built a cloud-based integrated Self-management support system, consisting of user applications and underlying service components, as follows:

POWER2DM Applications and Services:

1. POWER2DM’s web based Shared Decision Making Application (SDMA), supports the healthcare professionals in shared decision making phase to i) register clinical data, ii) analyse the patient’s status and self-management performance via several dashboards, and iii) provide/update the personalised treatment plan for the patient (used by the POWER2DM SMSS as the basis). It offers several data visualisations:

Compass view illustrating Self-Management performance, identified barriers, and Risk Scores for comorbidities and outcomes,

Several dashboard views providing details for glucose management, self-management, progress of diabetes, risk scores and barriers,

Model-based treatment simulation with KADIS®,

Common panels for data entry and goal/treatment plan registration.

SDMA utilises POWER2DM Prediction Services intended for treatment simulation and risk score evaluations:

KADIS®: The KADIS model is a short term analysis and simulation model based on 72 hours of patient data regarding blood glucose values, oral medication, insulin therapy, carbohydrate intake, and exercise. KADIS can visualise 24-h glucose absorption patterns based on aforementioned data, and offers simulation of different therapy options to healthcare professionals.

The ADVANCE risk scores are derived using the ADVANCE Cardiovascular risk engine and the ADVANCE Kidney risk engine which provide the risk of developing cardiac and kidney complications in patients with diabetes.

The Major Outcomes T1D model uses the prognostic factors of age, glycated haemoglobin, waist-hip ratio, albumin/creatinine ratio, and HDL cholesterol to classify people according to risk groups for major outcomes including major coronary heart disease, stroke, end-stage renal failure, amputations, blindness and all-cause death.

The UKPDS risk engine used to calculate a risk score for T2DM patients developing coronary heart disease. The engine uses a combination of demographic information, health information, and clinical lab values to calculate risk for developing coronary heart disease, stroke, or death resulting from either.

2. The POWER2DM SMSS Mobile Application supports and interacts with patients during self-management periods for i) delivering customised Just In Time Adaptive Interventions (JITAI) as push notifications, ii) collection of ODL (Observations of Daily Life) data; guided manual data entry by patient and/or connecting and retrieving measurements from medical devices (glucose sensors) and wearables (activity tracker) transfer them to the POWER2DM HL7 FHIR compliant Personal Data Store (PDS), and iii) customisation of patient profile and settings. Furthermore, patients are continuously supported through mobile application by means of Just-In-Time-Adaptive-Intervention (JITAIs).

For this, the POWER2DM Communication Engine component continuously monitors the patient’s changing behaviour context and helps patients to comply with their goals and action plans, by delivering personalised motivational interventions based on intervention definitions and rules. In POWER2DM, collaborative work has been performed with the domain experts in behaviour change to configure the software for the behaviours, intervention techniques, and intervention content according to scope and requirements of POWER2DM.

3. The POWER2DM SMSS web application serves to complement the mobile version with a wide range of digital interventions; i) specification of personal values and barrier identification, ii) adapting treatment goals to self-management goals, or creating new goals, iii) planning of activities (action plans) periodically in relation to goals, iv) filling questionnaires and collecting ODL data, v) evaluation of performance by providing interventions periodically (basically weekly). Interventions are delivered by the POWER2DM Action Plan Engine which offers support for patients in achieving their self-management goals using motivational messaging, simple psychological exercises to help overcome barriers, psychoeducation related to the patients self-management plans and diabetes status, and the ability to change the self-management plans.

Clinical validation

The POWER2DM system is currently undergoing clinical validation in two existing regional settings in The Netherlands and Spain, using POWER2DM functionalities connected with different local healthcare systems and following the regional guidelines for diabetes care (115 Type-1 and 115 Type-2 diabetes patients randomised to POWER2DM support vs. usual care). The study registry number is NTC03588104.

Expected impact

It is expected that POWER2DM will make an essential step forward in empowering diabetes patients, advancing prevention and potentially decreasing disease burden and costs, specifically:

Intermediate outcomes (0-6 months):

– Improved self-management capabilities, problem-solving and compliance to care plan;

– Obtain a healthy behaviour change and good metabolic control based on the defined care plan and goals by continuous decision support, e-coaching and monitoring.

Post-intermediate outcomes (after 6 months):

– Improved clinical parameters (HbA1c, fasting glucose, etc.), quality of life and patient satisfaction indicators;

– Maintenance of healthy behaviour

Intended long term outcomes:

– Physical and psychological wellness of diabetic patients, resulting from the sensation of empowerment towards their disease.

– Maintenance of improved glycemic control by patients, and as a consequence:

– Prevention or delaying progression of the diabetes-associated complications

– Potentially leading to higher cost-effectiveness in diabetes management by improving the efficacy of self-management.

Perspective

The POWER2DM Consortium partners are looking forward to collaborating with parties who are interested in using or further developing the POWER2DM system or its separate components, in their health practice, e-health products or R&D.

Acknowledgement

POWER2DM has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 689444.

 

Please note: This is a commercial profile