NHS Long Term Plan: Technology helps clinicians monitor the health of dementia patients in their homes
Technology Integrated Health Management (TIHM) for dementia is a pioneering Internet of Things study – devices in the home that can connect to the internet and each other – that enables clinicians to remotely monitor the health of people with dementia living at home.
The award-winning study is led by Surrey and Borders Partnership NHS Foundation Trust and includes partners, the University of Surrey, the Alzheimer’s Society and health technology provider Howz. It is part of the NHS England Test Beds programme, that brings NHS organisations and industry together to test combinations of digital technologies, which could transform the way in which healthcare is delivered. It is the first such programme in the UK or internationally to develop and install an Internet of Things-based intervention for people with dementia in the community.
TIHM uses a network of internet-enabled devices, such as monitors which record vital signs, and sensors, that are installed in the home. GPS trackers that monitor a person’s movements outside of the home are also included in the network. These devices work in combination with data analytics and machine learning (computer programs that can access data and use it learn for themselves) to alert a Clinical Monitoring Team when a participant with dementia needs support. The aim is to identify health problems early on, so that people with dementia can receive prompt treatment and support, helping them to stay well in their homes for longer and reduce pressure on carers.
TIHM is providing the NHS with a new digital and data-driven intervention that is based on real time information indicating when and where support is most needed. This helps to ensure the workforce is used in a targeted and effective way.
The challenge
An estimated 928,655 people are thought to have dementia in the UK with that figure set to jump to more than 1.1 million by 2025. The cost to the UK economy of supporting people with this condition is more than £26 billion and predicted to rise to £55 billion in 2040. The impact on the NHS is significant: one in four hospital beds are occupied by a person with dementia. Supporting a person with dementia also takes its toll on carers with 63% saying they don’t have enough support.
About the study
More than 400 people (204 people with dementia and 204 hundred carers) participated in the first phase of the randomised control study. Half of participants had their homes kitted out with sensors, monitors and gateway devices. The sensors monitored movement in the home and environmental data, such as temperature and light. Monitors recorded vital sign data: blood pressure, body temperature, pulse, oxygen saturation, weight and hydration. Trackers monitored a person’s movements outside the home.
Data streamed by these devices was integrated and analysed using data analytics and machine learning in a back-end system developed and designed by technical experts at the University of Surrey.
Machine learning algorithms were also developed to translate data into actionable clinical information. Over time, the machine learning began to identify individual patterns or personalised norms based on participants’ historical data. This enabled a more nuanced approach, which was important for participants who had specific medical conditions.
As well as vital signs monitoring, the algorithms were developed to integrate environmental and behavioural data to provide a holistic picture of health, activity patterns and wellbeing status. Using sophisticated pattern analysis techniques based on combining physical, environmental and behavioural data, the team were able to generate predictive algorithms for early identification of agitation, irritability and aggression and urinary tract infections, one of the five leading causes of hospitalisation in people with dementia.
The Clinical Monitoring Team responded to alerts by following clinical algorithms. They also exercised clinical judgment. In the first instance, the team was most likely to follow up a concern by contacting the carer. Other courses of action included: asking an Alzheimer’s Society dementia navigator to visit; making a referral to another NHS service; involving social services or the police or, if necessary, contacting the emergency services.
To support TIHM for dementia, a group of 20 people with dementia and their carers were recruited outside of the main trial to provide ongoing feedback about the technological devices being used, monitoring processes and deployment of technology into homes. These Trusted Users played a crucial role in the design of the study and were monitored in the same way as people in the technology arm.
Findings
The evaluation was led by the University of Surrey. The full evaluation report is yet to be published, but early findings are:
Participants with the technology placed in their homes experienced a statistically significant reduction in neuropsychiatric symptoms associated with dementia, such as depression, agitation, anxiety and irritability. It is easy to imagine how a reduction in these symptoms eases the pressure on family carers and makes it possible for them to carry on caring rather than see a loved one admitted to hospital or a care home
There was also positive feedback from carers who said the new layer of support provided by TIHM improved their “peace of mind.” The vast majority said they would recommend TIHM to others
Successful development of machine learning algorithms to detect urinary tract infections and agitation, irritation and aggression.
Feedback from Trusted Users:
“TIHM has reduced our visits to A&E and put our minds at rest”
“It’s like having a doctor’s surgery in your own home so you don’t have to visit the GP so often”
“TIHM helps ease the pressure of caring”
“It’s very reassuring to know someone is always monitoring my husband’s health”
“Being able to share data with the GP meant they could look at the history and prescribe the medication immediately. Exactly what you want”
Professor Helen Rostill, Senior Responsible Officer for the TIHM for dementia study, and Director of Innovation and Development at Surrey and Borders Partnership NHS Foundation Trust said:
“We believe TIHM for dementia, developed in partnership with the University of Surrey, has the potential to transform care for people with dementia and also those with other complex and long term conditions as well as reduce some of the pressures on the NHS. We are delighted that the early findings from the first phase of the study show there was an improvement in the quality of life of both people with dementia and their carers. We are also particularly proud of our pioneering work with the University of Surrey in developing machine learning algorithms to alert clinicians to signs of agitation and urinary tract infections (UTI). UTIs are a top five cause of unplanned hospital admission among people in this group. We will further develop this work in phase two to improve early intervention and help reduce people’s need for hospital admission.”
Next steps
Funding of just over £1 million from NHS England and the Office of Life Sciences for a second phase of the study was announced in May 2018.
The key aim of the second phase is to provide a ‘TIHM in a box’ product that is scalable across the NHS. We also plan to look at how TIHM can be adapted for the private sector.
Up to 150 people (75 people with dementia and 75 carers) will be recruited on to the second phase of the study from across Surrey and NE Hampshire. This will be an agile study that will give the partners the opportunity to make changes as the study develops with a focus on better understanding sleep patterns and their impact on people’s health and also signs of depression. It will not be a randomised control trial.
The second phase is currently expected to start early in 2019, and currently awaiting the outcome of various regulatory approvals.
This will be led by Surrey and Borders Partnership and involves partners, the University of Surrey, Alzheimer’s Society, product provider, Howz, and Kent Surrey Sussex Academic Health Science Network.
For further information visit www.sabp.nhs.uk/tihm or follow us on Twitter @TIHM_dementia
