I went to an impressive InterSystems "Joined Up Health and Care" conference at the Belfry a month or so back. There were lots of interesting use cases, an update on the InterSystems product suite, something on the importance of the NHS Digital Transformation to InterSystems and the need for engagement with clinicians (the lack of which largely explains why Npfit (the NHS National Programme For IT went wrong)) - and, for the building of Trust, InterSystems wants to be its clients' "Trust Partner".
Mark Palmer, InterSystems Country Manager, told us about InterSystems Trakcare, which he described as "the world's leading clinical patient record (CPR) system", getting two 2017 Best in Klas awards, for Global (Non-US) Acute Care EMR, and Patient Administration Systems, from research and insights firm KLAS.
All in all, the conference was very informative. Not least with the "light relief" presentation Ben Goldacre, of Bad Science fame, gave us on the need for more and better analysis of all the data the NHS produces and stores. There is a serious point here, for the NHS. I am interested in NHS Maturity Improvement. A somewhat simplistic definition of Maturity is that an organisation applies "what works" and "what doesn't" post-implementation analyses to any process changes it makes, with metrics so that it can recognise "good practice" - and "good practice" is then migrated throughout the organisation, as part of a "continual improvement" initiative.
There is anecdotal evidence that this doesn't always happen in the NHS. I recently talked to an NHS technology vendor about a department adopting its high-maturity NHS technology solution - a real positive use case for its product. But, the person I was talking to also said that the department just across the corridor was still enmeshed in manual procedures and paper - and simply not keeping up...
The problem is that useful "maturity" is an institutional state-of-mind, not something that applies to specific departments or products. One of Ben Goldacre's stories confirmed that it is probably an NHS-wide issue (perhaps because of the way various Governments have set up the NHS structures). He reported some stats around prescribing proprietary and generic medicines. In a mature NHS organisation, you'd expect prescription of proprietary medicines to fall immediately as soon as patents expire and cheaper generic alternatives become available (there'll always be exceptions if a medicine is genuinely expensive or production is "owned" by a monopoly supplier, but this would be expected in general). In fact, according to Goldacre (using information available at the OpenPrescribing website) the anticipated behaviour is only sometimes what you see, comparing like-with-like (i.e. the same medication). In some NHS Trusts, you see generics taking over as soon as they are available; in others, it takes significant time to notice that generics are now available; and in some, prescription of proprietary medication continues regardless. In a world where the NHS is strapped for cash and generic medications are frequently very much cheaper than the equivalent proprietary version, I think this is strong support for my view that the NHS as a whole has process maturity issues, that are well-worth identifying and removing. If prescription processes/controls, which are at the centre of NHS operations, are immature, it is likely that other processes are too.
The need for process maturity and continual improvement impacts the Digital Transformation effort the NHS is currently putting in. It's a "people thing" as much as anything to do with technology; technology is necessary but not sufficient. That said, the InterSystems suite, TrakCare (an electronic patient record and patient administration system) and HealthShare (for information capture and analytics) could indeed be an enabling technology, as InterSystems seems to be concentrating on enhancing trust between all stakeholders (not least with respect to patient data privacy and GDPR; more on this below).
TrakCare is not sold in the USA, but otherwise, it has a global reach. It provides capabilities for comprehensive clinical, administrative, and departmental management, sharing a single data repository and with a common user interface. Its modular design is said to allow its functionality to be reused across all companies but also to allow full customisation to the healthcare requirements of different countries and even different regions (in the UK, healthcare is different in England and Scotland, for example).
There is also InterSystems HealthShare. The secure sharing of information across the NHS and social care systems is a vital part of the digital transformation of the NHS - patients expect A&E, for example, to have full access to their medical records if they have an accident on a seaside holiday, while expecting their medical details to be entirely private; and if social care is "joined up" with medical care, this could free up a lot of NHS beds. All this is somewhat aspirational at the moment, but InterSystems HealthShare supports information capture from all stakeholders; meaningful sharing of this information amongst all stakeholders in the provision of care; analysis of this information (whether "structured" or "unstructured"; and supports continual improvement in care and efficiency.
We haven't had the chance to assess how effective reuse of functionality across countries/regions is, in practice; but we are impressed with InterSystems' underlying technologies (InterSystems Caché data storage and management and InterSystems Ensemble service bus integration). We note also that InterSystems makes embedded DeepSee analytics technology and an iKnow "bottom up" text exploration capability (which can discover concepts and relations within the text itself) available, both capabilities that we already find impressive.
TrakCare is a mature product, offering EPR and PAS solutions, and has many satisfied customers in the UK. But many of its rivals (Cerner, CMS, EMIS, etc) can say much the same. The NHS is an extremely complex environment and a multi-product approach appears to be institutionalised, so we find it hard to evaluate effective market share and growth opportunities in any meaningful way. In Financial Services, for example, we'd expect to find a convergence journey towards the single product that best satisfies an organisation's requirements. That doesn't seem to be the NHS Way; partly (we presume) because the NHS is much bigger than any financial institution - but also because it can tend to be micro-managed by politicians who see vendor competition as an essential aid to efficiency and effectiveness. We are not sure that we agree - and we can't see that the essential model of NHS care provision can (or should) differ all that much between regions - or even between England and Scotland. But apparently it does, and this allows several competitors to share the NHS cake.
Our impression (from people who know the product and are generally positive about it) is that TrakCare, in particular, is extremely powerful but can get a little "keyboard heavy", especially if it has to integrate with legacy systems; and it may take a few more "click-throughs" to achieve something than alternatives; although both of these might only be "implementation issues". An alternative approach might emphasise agile "intelligent speech recognition" from the end users more and require less clerical input - but TrakCare can integrate with speech recognition, and which approach is preferred could come down to personal taste and the existing organisational culture.
Back to the conference, Siva Anandaciva, Chief Analyst at the Kings Fund, provided a useful introduction to delivering sustainability and transformation plans (which seem to be the coming thing in NHS digital transformation), in the context of the NHS' performance, financial, and workforce constraints. He quoted an NHS Trust CEO saying: "What is the calibration of 'good' when everyone is in deficit and it becomes about 'not being the biggest failure'". This highlights a possible state of the NHS where morale fails and innovation dies, yet (regardless of what politicians are in power), NHS demands will likely grow faster than the resources being made available. He also quoted an NHS ALB (arms-length body) Finance Director: "Someone asked me if the glass is half full or half empty. I told them I don't even have a glass".
According to an NHS Trust NED (non-exec director): "How are things going? Well demand is up to our eyeballs, we are nowhere near our financial control total, and we have a Requires Improvement from the CQC. So we feel we are pretty upper quartile at the moment." This situation is obviously unlikely to be sustainable; but, Anandaciva says, Sustainability and Transformation Plans (STPs) could offer the best hope of a way forward. Although not an easy or a guaranteed one.
There are approaches which will work. Collaboration between all stakeholders will be vital, which implies access to technology supporting information sharing and excellent access control and security. Perhaps we should see STPs not so much in terms of "Plans" but of "Partnerships". Anandaciva quotes Professor Sir David Fish of MD UCL Partners: "There is always a risk that financial pressures will drive rational organisational behaviours that are irrational for the system. But the cake is only so big, and the crisis is not purely local. So if we don't collaborate in partnership, in the end, although we might triumph in the short term, we can't in the long term".
Anandaciva sees the v2 STP as a "basic governance and implementation support chassis" while noting that STPs only add to what came before and aren't new statutory bodies with teeth, nor legal entities that can hold contracts. This won't help. Ultimately, STPs will need beefing up, perhaps with non-exec participation on an STP board and published targets and metrics; perhaps including "accountable care organisation" (ACO) contracts. An ACO contract could cover both Health and Social Care, over many years and multiple billions of pounds. But, always remember that (according to Anandaciva) ACOs are a bit like a marriage: they look great from outside; they have tax implications; they cost more in time and money than you expect; and, they don't magically fix dysfunctional relationships.
Nevertheless, it's a case of finding out what works and what doesn't and doing more of the former. Perhaps a valid approach is to find the keenest people in the organisation and ask them for a 2-minute smartphone video on their pet improvement idea - which will give you lots to think about, and execute on. I like that idea! And, perhaps just one idea that might come up is a Regional CIO network that could harmonise T&Cs across the region; implement job-sharing for peer-based experience gains; and, standardise IT across secondary and social care. Perhaps the mantra must be "We will work together unless we have a reason not to".
One issue that is going to add to the burden on the NHS is patient privacy, particularly with respect to the GDPR - General Data Protection Initiative - which is as much or more about People (culture) and Process as technology. Think Game of Thrones "Winter is coming": see this on Facebook. The UK government is committed to implementing an equivalent to GDPR in the UK; whether the EU will accept this equivalence is still moot. Ken Mortensen, Data Protection Officer for Global Trust & Privacy at InterSystems gave me good reason to think that InterSystems understands the issues; but I wonder how many of its customers do?
At basis, it's all about protecting and securing patient information, while ensuring that access to this, where it is needed, is not impeded; this covers its collection, processing and sharing, as well as its security (confidentiality, integrity and availability). However, as usual, the devil is in the details, some of which are not clear yet, such as:
- When you use information are you sure that you have the patient's explicit consent, for the exact purpose for which you are using it? Could you prove this easily?
- Can the patient remove this consent; and can this removal be retrospective?
- If someone does steal patient data, will you be able to notify the Data Protection authorities within 72 hours of the organisation becoming aware of it (and, I rather suspect, willful refusal to become "aware" will not get you off the hook - read more here.
- What if a third party, which you have a legitimate right to share patient information with (i.e., patient consent), loses personally identifiable information? Did you know that you share responsibility for the breach?
- Can you get business benefit (see here) from addressing patient data privacy and GDPR, rather than just viewing it as a cost of doing business?
Reviewing this InterSystems conference as a whole, I was gratified (and relieved) to meet people who seemed to be getting things right. Nevertheless, it seems to me that most (probably all) NHS initiatives possess all the factors which go to make IT programmes high-risk: high visibility, routine press interest, micromanaging politicians, safety criticality, lots of resource limitations, and so on.
However, there are approaches available for managing this, and plenty of effective technology to help. The prime issues, however, seem to be cultural - the need for process maturity improvements, collaboration and continual improvement without blame - rather than anything much to do with the technology (although some NHS technologies are definitely more "high maturity" than others).