Rethinking the architecture of health IT to unlock potential

Tomaz Gornik, CEO of Better, explores why we need to rethink the architecture of health IT in order to unlock the potential of a patient-centric approach to care.  

In just a few years, conversations surrounding the importance of data in healthcare have changed dramatically. And although it took a pandemic for the market to shift, it is encouraging to know that accessible and quality data is now considered paramount to improving health and care.

Providers are finally starting to think about data first and applications second, and we are seeing fundamental changes being made to NHS procurement process, with tenders less likely to give vendors responsibility for data formats. Instead, they are requesting solutions based on open, published, vendor-neutral data standards.

Healthcare systems are evolving too, with patients demanding greater responsibility to manage their own care, and healthcare providers recognising the benefits of making them active participants, particularly in terms of reducing elective care backlogs.

As a result, we are seeing more attention being paid to holistic care models across the sector, which focus on bringing care closer to the citizen and coordinating care activities between different providers. This is a positive step, and a sign that the future of health and care does not revolve around provider organisations, but around the patient.

Myriad of systems

However, for the past 25 years, healthcare providers have been trying to make data follow the patient as they navigate the healthcare system. This means that data has been captured in a myriad of systems across multiple organisations, with only some of it sent to a shared care record.

In most cases, this has resulted in healthcare providers being hamstrung in their ability to share and access information due to the many legal, technical, and business barriers, like information blocking. And even if we overcome these obstacles and manage to get all relevant data into a shared care record, evidence shows that most of it will be worthless due to a lack of common understanding, with only 22% of data ‘understood’ when exchanged across different EHR vendors.

If you have been in the healthcare interoperability space for a while, this will come as no surprise. You will know how difficult standardising data formats and terminologies across a healthcare system is, and how we often have to compromise and make do with what we can get.

That being so, introducing new document types and data to share across a health system means upgrading all participating systems, which will take years — presenting a major roadblock in improving care coordination.

Shared care record as a single source of truth

Recognising the complexities of capturing data locally and sharing it between different systems, some regions are taking a different approach by building applications on top of a shared care record, with centrally-managed components embedded within provider systems using contextual launch and single-sign-on.

Others are using centrally managed portals in low-acuity environments, like care homes, where providers often lack adequate systems. These alternative approaches create a patient-centric care record, with simplified data formats and terminologies, and faster application distribution.

For example, national systems like Slovenia, Malta, Scotland, Finland, Wales, and recently Catalunya, and top institutions like Heidelberg, The Christie, and Karolinska, already support primary care with centralised, cloud-based applications running on top of a shared care record providing a single source of truth.

This approach ensures consistency of data and application functionality, speeds up upgrade cycles, and reduces the likelihood of duplication and cost. Of course, is not possible to deploy this exact model in many regions and countries, but we are seeing it being used to coordinate care for specific population subgroups, usually for managing a chronic disease.

I first came across this idea more than ten years ago, when I was introduced to a solution in the Netherlands that was being used to coordinate care around anticoagulation. It was a centralised, national system which allowed primary and secondary care applications to embed forms into their clinical front-ends. Further to this, clinicians could access information from a portal and patients were given access too, so that everyone involved in the care process were on the same page, accessing the same data.

More recently, several projects in the NHS are following the same example. Somerset Integrated Care System is building a Single Active Medication List to provide a Regional Medications Platform that will consolidate the medication information from different care settings into a single coherent patient record, and the same solution is being adopted by Lancashire and South Cumbria Integrated Care System. Meanwhile both Suffolk and North East Essex Integrated Care System and OneLondon are deploying care planning solutions.

For the most part, current approaches to interoperability provide lacklustre results, which is why we need to rethink the architecture of health IT. Today, providers capture health data in various systems using proprietary data formats, sending some data to regional shared care records. New architectures have a vendor-neutral data layer at the centre and low-code tools to accelerate delivery by pushing application functionality to provider systems. This facilitates care coordination and provides a single source of truth, keeping everyone on the same page, including the patient. After all, care records should serve the patient and provider organisations, in that order.