Aims

This work sought to explore existing views and experiences of managed convergence and concerted adoption.

Methods

We have purposefully sampled national digital leaders based in England in order to contextualise the strategy of managed convergence. These came from a range of geographical locations and care settings, and had experiences working in settings that had implemented (Electronic Health Records) EHRs, with some efforts to integrate systems across care settings.

We asked participants to recommend additional interviewees to develop a robust evidence base drawing upon a diverse range of experience and perceptions.

We conducted semi-structured qualitative interviews to explore contexts and progress, national/regional/local strategies, implementation and adoption experiences over time, challenges encountered and ways these were resolved, reflections, and future plans.

Sample

For the UK focused piece of work, we approached 17 digital leaders from various geographies and settings including acute, mental health and community care.  We interviewed 11 participants. Participant details are given in the table below.

Results

English stakeholders broadly agreed with the levelling-up agenda and bringing less digitally mature settings up to a higher level of digital maturity. They also agreed with the aspiration to improve care pathways and increased data accessibility. An element of convergence was also perceived as the right way forward. Whether this should be measured using HIMSS and achieved through technological convergence, was however contested and some cautioned that a monolithic procurement goes against the NHS constitution. There was a perceived shift from discussing the key problem at hand, towards discussing technological solutions. Many called for taking a step back and considering what problem the convergence policy is seeking to address and how it may achieve these.

A key consideration was perceived to be that the current strategy did not reflect the needs of mental health, community and primary health and social care. Convergence needs are different in different care settings and this needs to be taken into account by current strategy. Convergence is likely to work best where patient flow between organisations creates an immediate need to share information.  These considerations apply to information sharing across settings (e.g. mental health, primary/community and social care) as well as to exchanges within the same setting.

There was a lack of a clear understanding of what the convergence strategy might involve (by some it was seen as a change in strategic direction) and what the consequences for providers could be, as well as a perceived disconnect between existing programmes in NHS England. Some also stated that there had been changing definitions of the concept since it was first announced. There was also a view that there were emerging pockets of convergence by default already (in some cases initiated by the GDE Programme), and some therefore questioned the need for a new policy. A degree of convergence within ICSs is arising and can be expected to develop organically as ICSs come to exercise leadership. This  was for instance observed during the COVID-19 pandemic, where more providers collaborated to address a common problem. Similarly, participants reported a degree of convergence towards PACS and laboratory systems (where different providers consolidated onto common systems for labs and/or PACS), in some cases using the same instance across a variety of settings.

The diverse existing interpretations of the concept were seen as potentially damaging to stakeholder engagement, and there was a perceived lack of clear strategic direction with some disillusionment around a perceived risk of not learning from previous experiences (including NPfIT and GDE). Participants also felt that the policy was disconnected from the relatively successful Wachter strategy and there was a perceived lack of a sense of end-point of investment. As a result, some reported that leaders on the ground were likely to ignore the new managed convergence strategy.

Varying local needs and setups

“There are huge risks with trying to make a monolithic system that meets everybody’s needs”.

The needs and make-up of every ICS vary, and a uniform convergence strategy is likely to be problematic and may indeed be harmful. In some instances, where a large teaching hospital already has a mega-system and smaller district hospitals do have limited functionality that are coming to the end of contract, they may wish to use the existing system. Some of this convergence is already happening (e.g. in London) and there are already established informal partnerships/alignments within ICSs as well as centralised formal efforts to promote sharing and learning in the Frontline Digitisation Programme. Many cautioned against imposing one or two preferred particular solutions that mega-systems were in general not well-suited to the needs of smaller hospitals: “Epic works fine with large academic teaching hospital – it’s not viable for smaller organisations”. And organisations with established information infrastructures are unlikely to want to replace these. An existing degree of organisational competition may, however, hamper this strategy. Some participants reported having worked with provider organisations who intentionally selected a different system from a neighbouring provider as they wanted to stand out. In other cases, implementing the same instance of a system may invite takeover, as organisations effectively become the same.

The position of ICSs was perceived to be emerging and variable.  Some of the better established ones have developed capabilities and vision. However many others were reported not have a digital strategy or clinical leadership in place to facilitate digitalisation of organisations within them. As a result, there was a lack of confidence that ICSs would have structures in place to deliver digital strategy and to ensure that providers will implement systems.

There is also a need to explore where there may be scope to achieve economies of scale – for example where contracts are coming to an end which may present an opportunity to come together and promote joined up working across an ICS. However, policies promoting institutional convergence need to consider issues that might arise where organisations were in competition or wanted to maintain separate identities. This also needs to include considerations surrounding the alignment or misalignment of contract endpoints.

Incremental expansion of information sharing, linked to the development of integrated shared care processes, driven by need, was seen to be the priority, as opposed to implementing systems to deliver large-scale data exchange between some providers but would take a long time to implement. Data exchange within care settings was seen as less beneficial for integrated health and social care delivery than data exchange across care settings. However, convergence within care settings may accelerate learning and help to negotiate better deals with vendors.

Integrated leadership locally was seen as essential. Its importance was highlighted across a number of different settings.

Existing data infrastructures also vary across ICSs. Many already have some data management infrastructure for local population health management and for research and these need to be integrated with national systems.

Participants also stated that in less mature provider organisations digitalisation is not at the forefront of their strategies, as they struggle to provide everyday care. Further burdening them might result in risks to patient safety. These organisations also struggle with consistent leadership: “Trusts who are poor at digital are poor at everything, basket case is basket case.”  In addition, particularly smaller organisations do not have financial resources, leadership and capability to afford and implement EHRs (particularly given the significant implementation challenge of adopting an integrate mega-suite) without national support. The NHS was not perceived to be able to afford deploying costly systems like Epic across provider organisations.

Others worried that managed convergence was just a short-term policy shift and they cautioned that longer-term visions building on previous strategies were needed – while recognising the need to remain flexible to cope with future changes in strategic direction. The Tech Vision from 2018 was generally well received. ^[1]

A wealth of existing experience

There were many examples of attempts to promote integration of care across settings through developing shared records and even instances of deploying the same instance of mega-systems across a small group of provider organisations. This included the National Programme for IT, which sought to deploy centrally procured solutions, which were rejected by local organisations due to a lack of usability and involvement in decision making. Some solutions deployed as part of the programme were only just now reported to be decommissioned at a high cost.

Although approaches varied, they allowed for local ownership and a degree of experimentation, which in turn improved ownership and de-risked national investment.  This wealth of existing experience, nationally and internationally, was not being tapped effectively to guide the formation and implementation of policy.

There is now a need to build on these successful experiences, as well as unsuccessful experiences, and learn from them.

Sharing implementation expertise was seen as crucial, especially with less digitally mature organisations. This may be achieved through sharing expertise across settings.

There were also several examples of health information exchanges across settings, including for analytics and research as well as direct care provision. In some instances, this involved using the same system (we saw examples where community providers used primary care systems), whilst in others it involved an added layer of integration functionality to extract and analyse data from across systems.

Lack of adoption of standards and issues with vendors

Participants further reported a lack of adoption of standards amongst vendors and that current incentives to open up systems did not work. Similar managed convergence strategies^[2] had, for instance created a monopoly in the primary care system, which participants worried would be mirrored in secondary care if the managed convergence agenda was followed. Potential issues were that incumbent vendors were hesitant to open up and change their system, as well as lacking incentives for innovation. This effect on the market was seen as very difficult, if not impossible, to remedy, as systems could not simply be replaced

Potential issues with procurement law

Some also questioned whether procurement law aligns with the aspiration surrounding convergence. Participants cautioned that for ICSs and provider organisations the policy of managed convergence needs to be reflected in considerations surrounding business cases. There need to be clear drivers and rationale around convergence which are currently not clear, and therefore providers will struggle to comply with procurement law (as the commercial strategy when procuring an EHR has to have an element of contestability). Current procurement practices are dependent on approvals at many levels including treasury and national business case approval in addition to internal approvals, ICS level approval and regional and national approval.

Potentially unintended consequences of convergence

Some also cautioned that there may be unintended consequences of managed convergence. Issues may arise particularly with organisations adopting a standard configuration or the same instance of a system. For example, in one case with sites operating various versions of an existing systems found that the decision to converge to the same version, forced those already working on a more advanced version to reduce existing functionality and thereby to effectively use a more limited system than if there was no convergence. There may also be a risk that the strategic focus on convergence will detract from existing digital strategies.

Managed convergence around a small number of large entrenched vendors was also viewed to have potentially adverse consequences on the market, risking smaller vendors to be pushed out (which in some respect was perceived to have already happened through the announcement of managed convergence).

Recommendations for policy

Focus on opportunities to drive provider collaboration with community focus

• There is a need for increased engagement of community and social care
• Convergence within care settings can facilitate learning and reduce variation, and convergence across care settings can facilitate interoperability and patient flow. There is, however, a need to recognise that workflows across settings vary.

Promote proliferation of different systems whilst certifying a group of solutions centrally

• There is a role for the centre to give locales a set of certified systems that conform to national standards, and then allow local decisions as to which system to choose (as has been done successfully previously). These systems need to include a range of functionalities and cost structures. ICSs could have a role in deciding what is the right value for money and technology for the locale.
• The strategic convergence towards open platforms and the separation of data from applications is an important strategic longer-term investment.

Nationally coordinated and locally led

• The needs and make-up of ICSs vary and so do ICS boundaries. ICSs set-ups may need to be re-defined on the basis of historical collaboration and patient flows. There is also a lack of clarity whether the focus for data sharing is the ICS or the region. Some regions may be too big for information sharing – the optimum size appears to be around the 5 million mark.
• The centre may want to shift budgets to ICS’s responsibility as they can drive behaviour in line with local requirements – national funding cannot do this.
• There is a need to allow for and facilitate leapfrogging and innovation through experimentation.
• Finding a couple of ICSs that are brave and creative enough to create exemplars and new digital models may be a good starting point.
• Concerted adoption is likely to be the most effective strategy for levelling up. Work has found that existing informal collaboration between adjacent partners (especially with similar systems and cultures) can facilitate levelling up – here, partnerships can serve as pathways for weaker provider organisations. There is also a need to establish buddying arrangements, user groups, and the provision for allowing experienced staff to be seconded (especially locally to support implementation).

Build on existing national infrastructures

• There may be scope to consider building on existing infrastructures to facilitate data sharing (e.g. the Spine, National Record Locator). These may also may need to include pan-ICS networks e.g. clinical cancer networks, ambulance providers, social care.

Relationship building and engagement

• There is a need for proactive relationship building between regions, implementers, clinicians and policy, in order to increase engagement of stakeholders and maximise chances of policy being supported. This will also help to mitigate the risk of people misinterpreting intentions of policy.
• Policy must allow for a certain level of experimentation and also for risk-taking and thus potentially failure at small scales in order to promote local engagement and better long-term outcomes.

Learn from experience and build on successful initiatives

• There are opportunities to learn from current practice which are not being effectively exploited (e.g. LHCREs, shared care records).
• There is a need to actively promote the exchange of knowledge, encourage reflection and the establishment of communities of practice/learning. This needs to include assessing the impact of new policy through evaluation.
• In order to ensure continuity, there is a need to build on the Tech Vision from 2018.^[3]

Vendor management

• There is a need to mitigate and control for unexpected commercial behaviour of vendors (e.g. putting prices up and exchange rates).

• Market management is a longer term intervention. Sustained engagement between vendor and adopter communities are necessary to develop a common vision and articulate specific priories.
• There is a need to maintain variability in the market and encourage smaller vendors to operate. There are long term concerns within the NHS about the need to maintain variability in the market and encourage smaller vendors to operate. However, market management initiatives to date have made little impact and may have made things worse. The managed convergence strategy is likely to significantly narrow an already oligopolistic and closed market. Indeed its announcement may have already have negatively impacted the market.
• There is a need to sustain scarce expertise in relation to procurement, standardisation – and link it to regions/ICSs where there is purchasing power. Better information sharing about contracts and performance of systems will help to facilitate this.

• There should be a focus on growing and creating solutions in the UK to drive GDP and mitigate the dependence of multi-national vendors and systems tailored to these contexts. There may be scope to initially focus on mental health and social care as these have limited infrastructures and UK systems in place.
• In order to achieve benefits of concerted procurement, there is a need to develop transparent cost/pricing models with vendors.

Develop strategic clarity and allow external input

• There is a need to define convergence (scale, scope, technologies and settings), implications for providers, the strategic approach, and assessment criteria.
• There is a need to define outcomes and not the output or the measures to deliver those outcomes without being too prescriptive e.g. making optimum use of EPRs where they have been developed and do this to support care across organisational boundaries
• Overall a balance between architectural purity and pragmatism needs to be found and the latter is likely to result in more immediate benefits for stakeholders and therefore promote adoption.
• There needs to be a review of strategy, as plans have significantly deviated from the Wachter Review which was overall successful and presented a strategy that stakeholders bought into. A review of strategy is needed (Wachter 2.0?).
• There is a need to integrate programs across NHS England and link these to an overall long-term plan, as there are currently some competing programs and no integrated overall strategy. This should build on the technology vision and strategy developed already.

Build incentives for convergence

• At present there are no incentives to converge and no disincentives to diverge for organisations. There is a need to agree what these will be and how they align with existing incentives.
• There have to be benefits for various stakeholders to converge – these need to be mapped and explicitly stated.
• There is a need to consider how the strategy will fit with procurement law
• There is a need to make national resources available to support the strategy

Developing capability and capacity

• There is a need to promote the development of existing NHS-owned capability and capacity to develop, implement and maintain systems as well as fostering a UK or European vendor base oriented around publicly provided health services.
• Build on existing enthusiasm around informal clustering of experts around a particular topic and communities of practice

[1] https://www.gov.uk/government/publications/the-future-of-healthcare-our-vision-for-digital-data-and-technology-in-health-and-care/the-future-of-healthcare-our-vision-for-digital-data-and-technology-in-health-and-care

[2] e.g. https://digital.nhs.uk/services/gp-systems-of-choice

[3] https://www.gov.uk/government/publications/the-future-of-healthcare-our-vision-for-digital-data-and-technology-in-health-and-care/the-future-of-healthcare-our-vision-for-digital-data-and-technology-in-health-and-care

Exploring the strategy of managed convergence: interviews with national digital leaders / Health IT innovation by blogadmin is licensed under a Creative Commons Attribution CC BY 3.0