Enable javascript in your browser for better experience. Need to know to enable it? Go here.
nhs ics fresh approach to data architecture

Why NHS England’s new Integrated Care System structure calls for a fresh approach to data architecture

Patient data is the key to fast, personalized, and patient-driven digital healthcare journeys. To properly enable them, data needs to be both easily shareable and interoperable — two things that have historically proven very challenging amidst growing data governance and ownership concerns.


Now, at the beginning of the Integrated Care System (ICS) era, NHS decision-makers are at a crossroad. The system, architecture, and solution decisions they make today have the potential to either set them on a fast-track to data-driven care success, or divert them away from it. 


Better use of data is key to the new Integrated Care Systems


As NHS England adapts to its new ICS model, data has two very important roles to play. Firstly, it’s going to be critical to one of the core purposes of ICSs — tackling local healthcare challenges by enabling stronger and better-informed local care planning.


Secondly, it needs to support a shift towards population health. It needs to help teams build up a true picture of care requirements at scale and use that information to improve the quality and accessibility of care across diverse populations.


However, there is a chance that the way the ICS operating model is designed will limit how effectively data can be utilized to actually achieve those goals. With each ICS making its own digital decisions, there’s a significant risk that empowerment at a local level could come at the cost of national interoperability. This will make it much harder to achieve goals like improving population health at scale.


36% of patients in London will need to move between multiple ICSs throughout their care journeys. For those patients — and hundreds of thousands of others across the country — limited data interoperability means slow journeys, poor patient experiences, and potentially worse health outcomes.


Bringing all patient data together in one place isn’t the answer


The diverse range of systems and solutions currently deployed across NHS organizations and ICSs have created a landscape where data is currently oriented around diverse domains rather than the patient that sits at the core of them. 


The problem with such a structure is that it leads to scenarios where teams have to translate and move data between systems. Every time data is translated, it loses critical meaning. It’s a slow process that creates bottlenecks and ultimately degrades the insights that the data is being used to generate. This leads to more work, increased costs, and greater clinical risks.


On the surface, an obvious and straightforward solution would just be to bring all of that data together in one place. But this poses serious practical challenges: if we bring all patient data together in a single pool, we risk creating even greater bottlenecks. 

It’s not in the NHS’ interest to take control of data away from the domains that know it best. If you create a centralized, homogenized pool of data, there’s a good chance it won’t meet anyone’s exact needs — limiting its use and value.
Scott Davies
Technical Director, UK Public Sector, Thoughtworks

First we need a change of thinking, then a change of architecture


It’s easy to get lost in the complexities and challenges experienced across digital healthcare. The first step to finding the right path forward is zooming out and refocusing on the outcomes that you believe data can help you drive.


Ultimately, all that healthcare providers really need is data that’s timely, accessible, of high quality, and trusted at the point where it’s accessed.

That need for strong, trusted data at the point of care is what causes so much data to have to be recaptured throughout a patient’s journey. But with a decentralized architecture, it doesn’t need to be.
Ashok Subramanian
Head of Technology at Thoughtworks UK

How Data Mesh can support cross-ICS success


Decentralized data architectures like Data Mesh provide a standardized foundation for governance across organizations while putting specific domains in control of the data they need. Each domain can build its own data products to meet its own needs; the mesh then enables those products to be seamlessly used by other domains wherever they are relevant and can make an impact.


For example, a central care planning team might create a data product to help it forecast patient numbers and bed requirements. Then, the procurement domain could pick that data product up and use it to help it model and anticipate upcoming growth in medical supply requirements.


What’s most significant about data mesh is that it is an architecture that can help organizations balance local empowerment — i.e. giving everyone the means to generate the results they need from the data closest to them — while maintaining interoperability, governance, and control across the organization.


Across the mesh, everyone is empowered to achieve their own goals in their own way and at their own pace. Most importantly, the consistent foundation of the mesh ensures every bit of individual effort isn’t locked away and hidden, but is instead pushing the organization as a whole towards its wider collective goals by making the efforts of every domain open and usable for every other team.


To learn more about Data Mesh for healthcare, and explore how it is helping ICSs and NHS England increase interoperability and accessibility, watch our webinar with Scott and Ashok on-demand.   

Find out how Data Mesh can support your data architecture