SCAMPI: Self-Care Advice, Monitoring, Planning and Intervention

Lead Research Organisation: City University London
Department Name: Faculty of Management

Abstract

In order to relieve pressure on an increasingly overstretched NHS, there is an ever-growing need to deliver more efficient, effective, patient-centric care in the community. New intelligent healthcare technologies have the potential to deliver this care relieve the pressure, in the form of simple-to-use digital technologies in people's home that support self-care and reduce the need for routine interventions from healthcare professionals.

Therefore, the SCAMPI consortium will develop a new form of computerised toolkit that will allow someone living in their own home with a chronic condition, together with their relatives, carers and healthcare professionals, to self-manage both their care of the condition and life with it. People will interact with the new toolkit through a new form of intelligent visual care plan, called VIZ-CARE. Any care plan is a documented agreement between a patient and healthcare professionals about the patient's care goals and qualities to maintain or work towards, and the desired services, medicines, and activities such as eating, exercising and socialising. SCAMPI's new form of care plan will be visual, natural and simple-to-use, enabling a person living at home with a chronic condition to customise their life and care according to their individual needs and preferences, with pro-active support for thinking about important care goals and qualities, as well as the means to achieve those goals and qualities. The person using VIZ-CARE will also be able to share the plan with named relatives, their carers and targeted healthcare professionals such as specialist nurses and their GPs, and make joint decisions about customising the care plan so that the person's needs can be met more effectively, even when these other people are elsewhere, using web technologies.

Moreover, the visual care plan will update regularly with care-specific feedback from discrete and cost-effective sensor devices placed around the person's home. Using the available data from these sensors as input to different artificial intelligence algorithms, a person's visual care plan in VIZ-CARE will indicate the degree to which the care goals and qualities of the person as specified in the plan are being achieved, and if needed, flag potential risk indicators along with care recommendations when a goal or quality is not being achieved. For example, to monitor planned daily activities such as a 20-minute local walk, VIZ-CARE will collect data not only about walking using a device such as a pedometer, but also about life and care qualities specific to the person and dependent on good hydration associated with walking about weight loss (from scales), movement in the home (to detect disorientation), kitchen temperature (to detect food consumption), loss of energy (from bed sensors) and water usage (from tap meters), and generate risks warnings if needed. This intelligence-led feedback is predicted to support self-care and reduce the need for routine interventions from healthcare professionals in the management of chronic conditions.

To develop and evaluate this new computerised toolkit, leading researchers in computer science, the health sciences and digital business at City University London have joined forces. The team will develop the first version of the toolkit to support with people with two conditions - dementia and Parkinson's disease. And to engage people with these conditions, their families, carers and disease experts in the co-design and evaluation of the toolkit, the researchers will work closely with the Alzheimer's Society and Parkinson's UK.

Moreover, to maximise impact from SCAMPI, the team will work with 6 London-based Care Commissioning Groups (CCGs) - Sutton CCG and the CWHHE Collaborative of 5 CCGs. Digital entrepreneurs Evalucom Consulting will seek to commercialise the research results so that the elements of the toolkit can be made quickly and widely available.

Planned Impact

Computer science research will benefit from new scientific contributions to several of its sub-disciplines - to health informatics, a first computational model of care qualities that will be general rather than disease-specific; to requirements engineering, new uses and insights of advanced goal modelling representations and techniques; and to human-computer interaction, new visual languages for untrained users to interact with intelligent systems. To health service research, the computational model of care qualities will introduce new level of abstraction and precision with which to challenge and evolve existing care indicator frameworks developed in social care. Both the toolkit and standalone elements of it, such as the VIZ-CARE care planning tool, will provide health service researchers with more advanced technologies to evaluate in new clinical trials. It is also anticipated that new care practices, techniques and guidelines can be distilled from the toolkit, and introduced into care practices. Examples of these practices include visual care planning with digital drawing tools, improved care quality indicators framed in questionnaires, and more effective shared decision-making procedures. Business and management research can benefit with new business models to describe, explain and prescribe more effective digital innovation and diffusion, both in health care and social care and more widely. It can also benefit from validation of existing models and taxonomies in a new domain - integrated health and social care.

Future end-user communities can also benefit from the research. People with chronic health conditions such as dementia and Parkinson's disease who use the toolkit and its spin-offs will receive new capabilities to manage their care to their needs and preferences, to contribute to better quality of care received and happier lives. Improved self-care can reduce and/or delay hospital admissions that can be stressful and damaging to health - hospitalised people with Parkinson's have a prolonged stay, poor motor outcomes, infections, prescription errors and increased post-operative mortality, as well as are more likely to have repeat hospital admissions. Informal carers and the relatives of people with chronic health conditions can also benefit from the improved health to the wellbeing of people who are able to manage their own health at home. Their increased participation in shared decision-making about care will improve their understanding of a chronic condition and empathy of life with it.

Organisations that seek to improve the lives of people with chronic healthcare conditions, such as the Alzheimer's Society and Parkinson's UK, can also benefit through improved access to new thinking about dementia and Parkinson's. For example, the use of objective sensor data about daily life with a condition can be used to support others in similar circumstances to manage their own care. More generally, the strategic participation in co-design activities will offer new insights into what advanced digital technologies can offer to chronic care, and spawn new opportunities. Clinical Commissioning Groups that commission patient health and social care, such as the CWHHE, can refine their own care quality indicators and monitoring practices based on the new computational model. Furthermore, over time, these Clinical Commissioning Groups can mandate the use of new frameworks to improve patient safety and quality of services that patients receive from providers, thereby delivering better quality care. And as the CfP argues, the NHS can also benefit from reduction in the pressure of its services that can arise from the widespread uptake.

Digital innovators such as Evalucom Consulting will be able to build on SCAMPI's intelligent technologies, models and algorithms to disrupt the healthcare technologies market and deliver new products and services that will benefit it, similar SMEs and the UK's digital innovation sector.
 
Description Co-design work in year-1 has been used to refine the design of interactive tool that older people and their carers/relatives can use for life planning. The tool is optimised for use on tablet devices.

Research and development work in year-2 has developed a new computational model of quality of life for older people. The descriptive version of the model defines and associates the types of quality of life goals that older people are likely to achieve, and links these with almost 1000 types of regular meaningful activities that, if completed, contribute to the achievement of different quality of life goals. The computational version accepts data about a person's behaviour over time, computes compliance with the person's prioritised quality of life goals, and generates recommendations in terms of changes to goals and/or meaningful activities to undertake.

Research and development work in year-2 has also developed a new domestic sensor toolkit that does not depend on Internet access to function. The toolkit's sensors collect data related a person's prioritised meaningful activities at home and beyond, using Bayes and Deeper Learning models to compute activity descriptions that are input regularly to the computational model of quality of life for older people. The toolkit's robustness and effectiveness has been tested in people's homes, to improve the hardware design and choose between different Bayes and Deeper Learning models.

This research and development work in year-2 has enabled considerable progress towards answers to the following 2 major research questions posed by the project:

RQ1 The computational model of quality of life for older people has the potential to generate relevant quality of life goals and meaningful activities for people with dementia and Parkinson's more quickly than healthcare professionals.
RQ3 The sensor toolkit collect and fuse data related to life goals and meaningful activities.
Exploitation Route A commercial toolset to be exploited in follow-on research
Sectors Communities and Social Services/Policy,Healthcare,Other

URL http://scampi.city.ac.uk
 
Description Response to Lords Select Committee on Artificial Intelligence: Artificial Intelligence to Improve the UK's Health and Social Care
Geographic Reach National 
Policy Influence Type Gave evidence to a government review
 
Description Co-design collaboration with Alzheimer's Society 
Organisation Alzheimer’s Society
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution The research team have contributed leadership, resources and co-designers to work with key people from the Alzheimer's Society to co-design the new computerised toolset.
Collaborator Contribution The Alzheimer's Society have participated actively in the co-design of the quality-of-life artificial intelligence engine that it being researched and developed. Paid care workers have participated in co-design workshops to contribute to then validate the quality-of-life model underpinning the engine. The Alzheimer's Society has also used its networks to locate and engage with people living with dementia and their carers, to research and develop the interactive elements of the planned toolset. The Alzheimer's Society are also providing direction to the project through participation in the project's Advisory Board.
Impact Key outcomes are an improved design of the interface of the planned computerised toolset, and a more complete and accurate quality-of-life model that provides input to the artificial intelligence engine.
Start Year 2017
 
Description Co-design collaboration with Parkinson's UK 
Organisation Parkinson's UK
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution The research team have provided the leadership, co-design expertise, workshop resources and technical knowledge to collaboration with people made available by Parkinson's UK in the research collaboration.
Collaborator Contribution Parkinson's UK have participated actively in the co-design of the quality-of-life artificial intelligence engine that it being researched and developed. Paid care workers identified by Parkinson's UK have participated in co-design workshops to contribute to then validate the quality-of-life model underpinning the engine. Parkinson's UK has also used its networks to locate and engage with people living with dementia and their carers, to research and develop the interactive elements of the planned toolset. Parkinson's UK are also providing direction to the project through participation in the project's Advisory Board.
Impact The outcomes include a more complete and accurate quality-of-life model to inform design of the artificial intelligence algorithm, and effective co-designed interfaces that should be more useful to and usable by people living with Parkinson's UK.
Start Year 2017
 
Title Interactive life planner 
Description This interactive software, optimised for use on tablet devices, has been developed for use by older people with and without cognitive impairments. The planner enables a person and his or her carer and relatives to explore and prioritise quality of life preferences, use these preferences to explore different types of meaningful activity related to achieving these life preferences, access information about concrete instances of these services, and develop a simple visual life plan. This plan can then be verified, critiqued and shared with others. The plan can also be used to monitor progress towards the regular achievement of different quality of life preferences in the plan. 
Type Of Technology Software 
Year Produced 2019 
Impact None yet 
 
Title Quality of Life Computational Model 
Description The software is a computational implementation of a conceptual model of quality of life synthesised from different existing sources. The underpinning model defines 75 different types of quality of life goal of people living with dementia, and how these types of goal are associated. These 75 quality of goal types are associated to almost 900 different types of meaningful activity that, if undertaken effectively, contribute towards the achievement of the quality of life goals. The model can be instantiated for different people living with dementia, with other chronic conditions, and for older people without health problems. Different values can be propagated through the model, so that the model can generate: 1) recommended alternative quality of life goals with which to achieve the same outcomes; 2) recommended meaningful activities to achieve preferred quality of life goals; 3) quality of life goal types that current activities do not impact. 
Type Of Technology Software 
Year Produced 2019 
Impact The software has been implemented as a service, and already new opportunities are available for the model to power different online services. 
 
Title The SCAMPI domestic sensor toolkit 
Description The domestic sensor toolkit is a set of low-cost domestic sensor devices, a local hub built on a Raspberry PI device, and integration with existing machine learning models to generate required outcomes from the sensor data that is collected. The sensors can be configured in the home of an older person to collect data about meaningful activities that the person seeks to undertake according to the person's life plan. The data analysis generates outcomes of whether planning meaningful activities are undertaken or not. 
Type Of Technology Physical Model/Kit 
Year Produced 2018 
Impact None yet - evaluations are taking place 
 
Description Dementia and Tech Workshop at ACM CHI Conference 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact Represented SCAMPI at two-day CHI workshop on Digital and Dementia at ACM CHI Conference, Montreal, April 2018
Year(s) Of Engagement Activity 2018
 
Description Designing the UX for a digital healthcare toolkit 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact In November 2017, Simone Stumpf presented the SCAMPI project at an event as part of World Usability Day: Inclusion through User Experience
Year(s) Of Engagement Activity 2017
 
Description Response to Lords Select Committee on Artificial Intelligence: Artificial Intelligence to Improve the UK's Health and Social Care 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact The project submitted a 6-page response to the House of Lords Select Committee on Artificial Intelligence, with a focus on Health and Social Care. The project made 6 recommendations: 1.Enable and educate the general public to take ownership of their personal health and social care data, as part of their active care and life planning; 2. Ensure that health and social care professionals are equipped to understand, procure and deploy artificial intelligence and machine learning through suitable informatics education and training; 3. To reduce the potential for incorrect decisions, increase the transparency of artificial intelligence algorithms to enable public scrutiny and oversight and intervention by health and care professionals; 4. Determine the mix of regulatory and procurement action necessary to ensure that black-box artificial intelligence does not deny people access to information generated from their own datasets - a risk to the ethical ownership of people's data; 5. Work with social care commissioners and providers to create opportunities for UK-based artificial intelligence research enterprises to support the sector realise the potential of these technologies; and 6. Regulators need to future proof the way they regulate. The changing landscape needs to be mapped against the scope Parliament has determined for each relevant regulator.
Year(s) Of Engagement Activity 2017
 
Description SCAMPI Co-design workshops with experienced Parkinson's nurses working for Parkinson's UK 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Research activity, with external outcome directly
Year(s) Of Engagement Activity 2018
 
Description SCAMPI Co-design workshops with experienced dementia carers working for Alzheimer's Society in South Wales 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Increased engagement with the SCAMPI project and its outcomes
Year(s) Of Engagement Activity 2018
 
Description SCAMPI Co-design workshops with experienced dementia carers working for Alzheimer's Society in South Wales 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Further engagement and excitement with the SCAMPI project
Year(s) Of Engagement Activity 2018
 
Description SCAMPI Presentation to EU-SPRI Early Career Conference, Milano, Italia 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Postgraduate students
Results and Impact Presentation by Luigi Mosca at the EU-SPRI Early Career (Milan, Italy 23 - 24 Nov 2017) Conference on "How to foster innovative entrepreneurship? Trends, challenges, and policy implications"
Year(s) Of Engagement Activity 2017
URL http://www.euspriconference.polimi.it/?p=70
 
Description SCAMPI Presentation to UCL Partners Living Well with Dementia Series 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact In November 2017, Neil Maiden presented SCAMPI project results to dementia care practitioners at the UCL Partners series of dementia, in central London.
Year(s) Of Engagement Activity 2017
URL https://www.eventbrite.co.uk/e/uclpartners-dementia-well-pathway-series-tickets-35691449085
 
Description SCAMPI Project represented at Evidence Generation for Digital Health: A speed-dating event for academics and digital health innovators 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Industry/Business
Results and Impact The project will be represented to explore different possible exploitation channels with entrepreneurs and other impact channels.
Year(s) Of Engagement Activity 2018
URL https://www.eventbrite.co.uk/e/evidence-generation-for-digital-health-a-speed-dating-event-tickets-4...
 
Description SCAMPI email newsletter 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact A focused newsletter, each 2 months, which is sent to people interested in SCAMPI, and sent to them as a formatted email newsletter.
Year(s) Of Engagement Activity 2017
 
Description SCAMPI project launch event, March 2017 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Public launch of the SCAMPI project in London. A half-day event of presentations, demonstrations and inputs from project partners including a person living with Parkinson's Disease.
Year(s) Of Engagement Activity 2017
 
Description The external-facing SCAMPI project website 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact SCAMPI has an external-facing project website, with related social media accounts, which inform people of project outcomes and activities.
Year(s) Of Engagement Activity 2017,2018
URL http://scampi.city.ac.uk
 
Description University Newcastle Seminar 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Postgraduate students
Results and Impact Raise awareness of academic SCAMPI work, to build research collaborations with other relevant research groups and audiences
Year(s) Of Engagement Activity 2018