HTN Now: “Making data actionable and relevant to specific settings requires a collaborative approach” – Dr Kathrin Cresswell and Professor Robin Williams on digital integrated care

At HTN Now, we were joined by Dr Kathrin Cresswell, senior lecturer from the Usher Institute at the University of Edinburgh, and Professor Robin Williams, director at the university’s Institute for the study of Science, Technology and Innovation. Kathrin led a discussion on the definitions, concepts and drivers for the integration of care, along with the role of digitisation and the challenges healthcare is facing.

Introductions and background 

Kathrin explained: “We are social scientists from the University of Edinburgh; we evaluated the care record service as part of the national program for IT a while back  and more recently the Global Digital Exemplar Program.

“We also taught technology strategy at the digital academy. Our earlier work largely focused on secondary care providers, but we have recently been looking increasingly at integrating care through digitisation and across organisational boundaries.”

Definitions and concepts 

Kathrin began by outlining some definitions, concepts and drivers for the integration of care: “Definitions and concepts of integrated care diverge significantly and they reflect different conceptions of what constitutes integration and how it should be achieved, and also the complex technical and institutional challenges involved in integration. Attention to these diverse drivers and differing priorities of stakeholders and organisations in the integrated care agenda is really important.”

She shared some wording from the World Health Organisation defining what they call integrated people-centred health services, which according to WHO means “putting people and communities, not diseases, at the centre of health systems, and empowering people to take charge of their own health rather than being passive recipients of services.”

Kathrin commented: “Our overall position here is that the aim of care integration is to improve communication and information flow between settings, to better coordinate the provision of services patient person pathway. The problem is that actually much of the focus has traditionally been on health services and especially hospitals, whilst other settings have been neglected.”

What are the different agendas and drivers for the integration of care? 

Taking into account different perspectives is “very important,” Kathrin noted. She added that drivers can vary between those aimed at reducing cost and harnessing data for secondary uses – such as for service planning and research – and those that focus on quality access and user satisfaction.

“Bringing these varying agendas together is not easy,” she said. “There are significant differences in the level of detail required by each and every one of the stakeholders involved; and there are different levels of codification of information, volumes of data required and also the question of whether information is required in real time or not. For example, health service research needs carefully curated medical records; on the other hand, the ability to make prophylactic interventions may need some more timely information. These different agendas tend to get unhelpfully conflated between big visionary projects and agendas.”

Where does digitisation come into all of this?

“Digitisation is now central to efforts to integrate care internationally,” Kathrin continued. She said that this could be called a ‘technical fix’ to address “fragmented mental health services”, adding: “In this context, information and data are seen as the link between different settings.”

She highlighted a quote from the Integrated Care Foundations, which states: “As each integrated care building block is reliant on information, digital solutions could be seen as the cement that holds the blocks together.”

The challenges of digital integration 

“There has been limited success so far in achieving integration of care facilitated by digitisation, and that is particularly true at scale,” Kathrin acknowledged. “That is partly because the focus on sharing data across organisational silos does not really mean the seamless interaction across institutions.

“The same is true for interoperability; interoperability of data does not necessarily mean interoperation of different stakeholders. So, to promote interoperation, there is a need to make information mutually intelligible across different settings; make information actionable; and to develop shared pathways.”

The problem with existing silos of information across different organisations and communities, she continued, is that each has different languages, different roles, responsibilities and rights. There are also different drivers and variations complexity of data requires for different purposes.

“As I mentioned earlier, care delivery needs up-to-date information or data about an individual, whilst population health research needs normalised population-level data. The problem is that a single health information infrastructure is unlikely to meet these different divergent needs, so we are likely to require a variety of infrastructures. These must somehow be linked in order to become connected; that is complicated by settings having different existing levels of digitisation and separate digital pathways.”

The development of infrastructures that span health and care settings is “lagging”, Kathrin commented. “We don’t actually have products as yet that match inter-organisational ranges of activity. We currently see information in accumulations of separately developed systems, implemented on different proprietary platforms with very limited interoperability. There’s also the continued addition of new forms of data; for example, Fitbits, blood pressure monitors and genomics data. That really complicated attempts to harmonise and standardise interfaces; data that is available on shared care records rarely provides all the information a specific health or social care professional need to deliver good care at the moment.”

Socio-organisational barriers to integration of care 

“Integration of health and care is not only a technological challenge; it doesn’t just require harmonising standards, systems and architectures,” Kathrin said. “We also have significant socio organisational barriers to the integration of care.”

She referenced involvement in a recently-published paper on this topic, and drew attention to the key barriers highlighted within it.

“Firstly, integration of information may involve disrupting existing information flows embedded in current practices

“Secondly, there is the competition and power differentials between settings – as mentioned, there is a traditional focus on hospitals with limited funding for social care in the past. Some professionals are also reluctant to share data if this may be perceived as a threat to professional autonomy.

“Thirdly, we have misaligned incentives; different organisations have different incentives and disincentives for data sharing. There are also issues and costs associated with data capture. For example, increased data capture always means increased workloads for health professionals.”

Kathrin added: “Diverse organisations have different levels of competence in various parts of digitisation, and these can vary significantly. There are big differences in primary care, allied health professionals, secondary and tertiary care. The role of digital information and the way it can be shared varies across these; therefore in some instances there may be no data to send and no data to receive.”

The final barriers, she said, are related to information governance and divisions of responsibility and liability, as they vary across settings and may, at times, clash. “For example, there may be a fear of liability amongst healthcare staff if many other stakeholders can see their records.”

Strategic approached to integrate care 

Kathrin noted that there has been “empirical work helping us to navigate these technological and socio organisational challenges. We looked at one study which was based on 17 case studies exploring the implementation of integrated care for multi-morbidity in eight European countries.”

The authors, she said, found that with regards to service delivery, “successful implementers commonly adopted an incremental growth model rather than a disruptive innovation approach. They found a balance between flexibility and formal structures of integration.

“For leadership and governance, they applied collaborative governance by engaging all stakeholders and distributed leadership throughout all levels of the system. For the workforce, that means these implementers were able to build a multidisciplinary team culture with mutual recognition of each other’s roles; they stimulated the development of new roles and competencies for integrated care.”

With regards to financing, “they found that secure long-term funding and innovative payments were applied as a means to overcome fragmented financing of health and social care. Implementers emphasised the implementation of information and communications technology that was specifically developed to support collaboration and communication – not administrative procedures. They tended to create feedback loops and a continuous monitoring system for organisational monitoring and research.

The overall takeaway, Kathrin concluded, “was that implementers engaged in alignment work across the different components and stakeholders across the health and social care system.”

Key points and examples 

“The integration of care through digitisation also requires wider strategic approaches; it is important to remember that current approaches to building infrastructures are unlikely to be sustainable in the long term. The underlying assumption is that today’s solutions are tomorrow’s legacy problems – meaning technologies get old very quickly and then need to integrated with new systems.”

Kathrin stated: “What we need to develop is longer term strategic foresight and not just create more legacy problems. If we look at Catalonia, a prescribing system succeeded when a Spanish-wide one solution failed. There seems to be a scaled breadth trade-off in the institutionalisation of data sharing; however elements, once established, can be a resource for building broader systems, such as patient registries.”

She shared how scoping work has been undertaken on strategic approaches to building health information infrastructures. “We came up with some options – one strategy may be to build on existing health information infrastructures and connect settings through interfaces; increasing the amount of information exchange between settings gradually,” she said. “It may involve building out from current systems designed for primary healthcare and extending them to social care. We have seen similar development in Finland, where they have developed social care functionality for Epic.”

Alternatively, “a more radical strategy may be to rip and replace current systems. However this is only possible in some settings that don’t have a huge amount of legacy infrastructure. Another example is Estonia, where the foundations for integrated care through digitisation have emerged from their efforts to set up an electronic identification system.

Radical methods of re-conceptualising care 

“In terms of radical solutions, patient-centred architectures may be a viable option. These let people store their data securely in decentralised data stores and users can control which people and applications can access their information,” said Kathrin.

“This approach provides a new way to manage tensions between different requirements of multiple stakeholders which may be neglected in other approaches. The problem is that these architectures have inherent challenges such as the need to develop a viable transition pathway from existing systems. We found in previous work that these architectures are seen by expert groups as very desirable, but very difficult to implement.”

How to make data actionable? 

“Let’s take a step forwards and imagine that all these architectural issues have been resolved,” she suggested. “The problem then is that data availability in itself doesn’t actually mean that data is used. Data includes the raw and unprocessed facts – information is prepared data that has been processed and organised into a more human friendly format. Both data and information set the stage for the discovery of insights that can then influence decisions and drive change.

“Making data actionable and relevant to specific settings requires a collaborative approach from across health and social care systems.”

Kathrin shared: “We are in the process of cultivating some work around learning lessons from previous large-scale digitisation programmes, which includes the national programme and the Global Digital Exemplars, the tech fund and how these may apply to integrated care initiatives. From this work, we have extracted six lessons for local leaders seeking to integrate care regionally.”

One lesson is to “move broadly in the same direction across the region whilst allowing various local visions associated with the integration of care to develop. The second is to keep your strategy and technology open, be realistic and don’t over-estimate benefits; involve, engage and align adopters; share learning and facilitate building communities of practice and finally elate to mitigate risks and demonstrate progress.”

Concluding the session, Kathrin said: “Going forward, it is important to remember that there is no definite recipe for success – so we need to view the integration of care through digitisation as an adaptive challenge. Learning, revising and rethinking our strategy is central when tackling these challenges.

“The key is that we take into account existing technological and socio-organisational infrastructures and barriers to the integration of care; finding ways to develop actionable insights and strategic efforts.”