Learning health systems promise huge health and care advantages for citizens and clinicians alike. Niels Peek tells Martin Cooper AMBCS about how northern cities are embracing the technology.

Four northern cities are hoping to improve patient care through using data and technology in revolutionary ways. The ground-breaking £20m project is called Connected Health Cities (CHC) and it will see experts from across northern England collaborating closely on a project that hopes to tackle issues like unplanned hospital admissions for patients with chronic diseases.

The CHC project will use existing healthcare data to generate new insights into how practitioners can identify at-risk patients earlier. It's hoped the project will also provide better support for patients who care for themselves and to make better, more targeted, use of community-based care.

Speaking about the project, Rob Finnigan, a kidney disease patient from Manchester says: ‘The benefits are enormous. By merging databases together researchers are able to get a more complete picture of public health and understand how best to plan services or prevent diseases from occurring in the first place.’

'The NHS is unique in the data it collects and it is fantastic to see Health North bringing northern cities together to harness this power to benefit the patients they serve,' said Health and Innovation Minister Nicola Blackwood MP. 'This project has the potential to set an exciting precedent of working more closely together across regions. The ideas being shared in the north have the potential to be developed and used right across the country.'

Going further, digging deeper

Adding detail, Niels Peek - a reader in informatics at the University of Manchester - explains: 'It's about building learning health systems in the north of England. It's a marriage between classical quality improvement and modern data science.'

Peek is Director of the Greater Manchester Connected Health City, and spoke exclusively to BCS about the learning health system project - how it was conceived, what it hopes to achieve and also about some of the biggest challenges CHC faces.
The CHC project was founded to address some familiar challenges. These encompass: the need to reduce costs; ever-increasing demand for healthcare; an aging population; the rising prevalence of long term conditions and also the pressure life style factors are placing on the health and care system.

'To meet these demands you need to link together data from services across health - primary care, secondary care and other health services.’ Achieving this, Peek says, will provide health managers, providers and decision makers with a complete and granular picture.

‘There have been pockets across England where this has happened to some extent. The comprehensiveness with which we're working is new,’ he says.

Looking for outside influences

Learning health systems, such as the CHC project, come from very mixed parenting. Rather than being purebred medical drives they draw inspiration from many different industries - sectors where data has had a profoundly transformative and disruptive effect.

‘Think about the travel industry, retail and music,’ Peek says. ‘There are many examples where our lives have been completely transformed by the internet and ubiquitous computing. I hope we can achieve similar things in healthcare.’

Transforming health and care so it can be consumed like buying products from Amazon is certainly a very compelling vision. It’s also fraught with difficulty. ‘One of the biggest challenges we have to overcome is the NHS's complexity,’ Peek says.

The NHS is also funded with public money. All this means health and social care behave, and is viewed very differently, from businesses in those sectors - where data has wrought the biggest upheaval. ‘Retail is a good example,’ Peek explains. ‘It has been transformed completely by data and computers. The old players have gone and new ones have taken over. That's a trick we can't play in health care and we wouldn't want that.’ Hospitals, unlike yesterday’s high-street stalwarts, can’t fold.

‘That,’ Peek warns, ‘would be very threatening to the sustainability and the safety of our healthcare system.’

Looking forward

Acknowledging transformation’s risks and challenges, Peek is however forthright in his hope that health can learn from these disrupted sectors and that it can achieve similar things. Specifically, he foresees a time when interactions between health services and the public becomes much more flexible and user friendly.

‘Interactions may even happen through my smartphone,’ he says. ‘It'll give me the information I need. We’ll also see a health service that's much better adapted to the needs of the population that it serves.’

Learning health systems also promise to help people with long term conditions. Currently these are generally managed through primary health care system - people visit their GP every three to six months for a check-up. This monitoring, Peek explains, is there to help prevent these conditions deteriorate and also to ensure patients don’t relapse.

The problem is, this process isn’t infallible. ‘If you look at patients with long term conditions like chronic kidney disease, asthma and COPD,’ Peek explains, ‘they all have flare-ups. And the health service isn't very good at preventing those from happening because they typically happen between clinic visits.’

Looking forward, learning health systems can provide a solution to this challenge. Rather than visiting your GP, your health data could be piped regularly to the surgery and your health monitored closely. Any problems or potential relapses could be detected quickly or even, through looking for patterns and changes, predicted. Again, Peek says, much of this - from the patient’s perspective - could be managed through a smartphone.

Learning health systems also promise advantages for clinicians. Primarily they’ll provide clinical decision-makers with much richer and more enveloping data about patients. At the moment important information may be encapsulated in different systems and spread across paper and faxes. Carers working in a connected health city could expect a fully integrated view of a patient - all from within one system.

Big data and grand scale data analysis also promise precision medicine. To illustrate the idea, Peek points to statins. Today he says around 100 million people world-wide take the drugs. ‘But,’ he observes, ‘we know that those statins are only effective in around one in twenty patients. So, in every twenty people nineteen are taking statins for no reason.’ From a wellness perspective, the story gets worse too. It seems that some patients are wise to the likelihood of the drug not working, assume they’re getting no benefit and stop taking the tablets.

‘We can use intelligence tools to target treatments directly at the patients who need them,’ Peek explains. ‘Those treatments will be more effective and we can also prevent a lot more side effects. We can give the patient a much stronger message. We can say "look, this medicine is for you... only for you and you have to take it.” In fact, you are one of the few people to receive this medicine because it works so well for you. And this is very different from what we tell patients now.’

The right to use data

For all these advantages to be achieved, the Connected Health Cities program has a huge need for data. As such, the project has placed a great emphasis on clear communication with the public. ‘You need to build up trusting relationships with the people whose data we're using. And we’re aware that trust isn’t a given,’ Peek explains.

‘One of the project's central pillars is the creation of a "social license" to reuse health data for research. You can't try to build a learning healthcare system without also building up that social license.’

Early on in the project, Citizen Juries were held - expert witness presented the case for anonymous data-sharing to members of the involved communities.

The Juries lasted for three days and have been very successful. At the beginning of the process, many people attending the sessions were defensive and unsure. In part, Peek attributes this to the Care.data debacle. It was an ambitious programme aimed to bring together data from GP practices, hospitals and care homes all with the aim of helping the NHS to provide better care.

Despite its grand and honest intentions, Care.data was hobbled by a disastrous PR campaign. Few people had heard of the drive until they read lurid headlines in the tabloid press about sensitive data being shared with commercial entities - all without the patient’s explicit consent.

The Citizen Juries do appear to be working very well. Attendees often change their position about sharing their data from one of resistance to one of embracement.

Doctors with data diagram