Professor Trisha Greenhalgh OBE is Director of the Healthcare Innovation and Policy Unit at Barts and the London School of Medicine and Dentistry, Queen Mary University of London. She led the independent evaluation of the Summary Care Record Programme.
The evaluation cost taxpayers more than £723,000, and there is no sign that the Department of Health and NHS Connecting for Health will take seriously the concerns the report raises.
The views expressed in the article that follows are Trisha Greenhalgh's:
"Last week, I asked a packed conference hall of general practitioners if any of them had ever seen a Summary Care Record. Not a single hand went up.
"Yet 30 million people in England have received a letter saying that a Summary Care Record will shortly be created for them from their GP-held medical record.
A pantomime-style clash?
"Here is a clash of narratives that would sit comfortably in a pantomime.
"Does the Summary Care Record exist? Oh no it doesn't, say some pundits, except in the dreams, manifestos and business plans of politicians and policymakers.
"Oh yes it does, say others, who point out that contracts are long signed; the technology is built; professional bodies have endorsed it; patients (that's you and me) have been informed; and front-line staff have been trained and issued with all-important smart cards and passwords.
Where does the truth lie?
"In this latter story, it is now a simple matter of installing some extra software and tripping the switch. Where does the truth lie?
"In this latter story, it is now a simple matter of installing some extra software and tripping the switch. Where does the truth lie?
"The Summary Care Record as a proof-of-concept technology most certainly exists. About 1.5 million such records have already been created and are sitting on the NHS Spine waiting to be accessed by authorised staff in the 'emergency and unscheduled care' settings for which they were designed. If 'technology push' were an appropriate model, we would just need to keep pushing.
The Summary Care Record exists - but how much will he technology be used?
"The entity whose existence is contested is the Summary Care Record as a technology-in-use.
"If you live in one of the half-dozen towns which have 'gone live' with Summary Care Records and attend an out-of-hospital clinic having (say) run out of medication or developed chest pain, you will have, on average, a 4% chance that your own Summary Care Record will be viewed by the person treating you.
"If you go to hospital with the same problem, that chance will be below 1%.
"The chance that the clinician will find what they are looking for (an up-to-date and complete list of your medication and allergies) is even lower.
Generous resourcing at the centre
"This is despite generous resourcing at the centre and what have been, by and large, well-planned efforts by competent local implementation teams to get the technology up and running.
"This is despite generous resourcing at the centre and what have been, by and large, well-planned efforts by competent local implementation teams to get the technology up and running.
NPfIT delays and disappointments
"The well-known delays and disappointments with other centrally-driven innovations (such as local detailed records, electronic transfer of prescriptions, GP2GP and Choose and Book) suggest that the significant gap between proof-of-concept and technology-in-use is a systemic problem across the entire National Programme for IT.
"The difference between a technology and a technology-in-use is the complex and often dimly-understood configuration of policies, resources, people, priorities, permissions, upstream tasks, buy-in and bits of infrastructure needed for a particular technology to 'work' in a particular setting at a particular time. And if it does not actually 'work', the technology might as well not exist for that particular use case.
"Complex policy interventions inevitably slow to a crawl. Westminster's stirring mission statements attenuate by the time they reach the provinces. Even in volunteer pilot sites, initial enthusiasm dwindles and real life must go on, despite the succession of away days, briefing breakfasts and promotional mouse mats. Steel and silicon obstinately refuse to appear on time or work as expected.
The elephant in the room
"But it would be a serious mistake for those leading the newly-announced government review of the Summary Care Record programme to slap a could-do-better diagnosis on the current implementation strategy and ask everyone to push harder.
"The well-known delays and disappointments with other centrally-driven innovations (such as local detailed records, electronic transfer of prescriptions, GP2GP and Choose and Book) suggest that the significant gap between proof-of-concept and technology-in-use is a systemic problem across the entire National Programme for IT.
"The difference between a technology and a technology-in-use is the complex and often dimly-understood configuration of policies, resources, people, priorities, permissions, upstream tasks, buy-in and bits of infrastructure needed for a particular technology to 'work' in a particular setting at a particular time. And if it does not actually 'work', the technology might as well not exist for that particular use case.
"Complex policy interventions inevitably slow to a crawl. Westminster's stirring mission statements attenuate by the time they reach the provinces. Even in volunteer pilot sites, initial enthusiasm dwindles and real life must go on, despite the succession of away days, briefing breakfasts and promotional mouse mats. Steel and silicon obstinately refuse to appear on time or work as expected.
The elephant in the room
"But it would be a serious mistake for those leading the newly-announced government review of the Summary Care Record programme to slap a could-do-better diagnosis on the current implementation strategy and ask everyone to push harder.
"Different leadership, better project management, closer surveillance of front-line staff and (perish the thought) tighter central control will not fix the underlying problems, which are not operational but philosophical.
"The National Programme for IT has been built on a mechanical and linear model of change rather than on an ecological and interpretive one. This is the elephant in the room.
The dangers of managing complexity centrally
"It is a mathematical axiom that as a system gets more complex, it becomes less predictable because fixed inputs cease to produce fixed outputs. Different parts of the system adapt locally, insofar as they are able, to local influences and available information, leading to organic rather than linear change.
"It is a mathematical axiom that as a system gets more complex, it becomes less predictable because fixed inputs cease to produce fixed outputs. Different parts of the system adapt locally, insofar as they are able, to local influences and available information, leading to organic rather than linear change.
"With technology-push, implementation stages are linearly coupled (you can't do C until you've done B, but B depends on A and A isn't ready yet). The more complex the system, the more likely efforts to ensure that implementation is "properly managed" (by imposing centrally-driven milestones, inflexible business processes and mandatory reporting of progress) will backfire.
"As predicted by complexity theory, high workload, unforeseen costs and continuing emergence of things that need 'fixing' have been consistent findings from empirical research studies on the National Programme for IT. These problems will be ongoing.
"As predicted by complexity theory, high workload, unforeseen costs and continuing emergence of things that need 'fixing' have been consistent findings from empirical research studies on the National Programme for IT. These problems will be ongoing.
A dangerous illusion to suppose the jigsaw's pieces will fall into place
"The anticipated steady state at which all pieces of the jigsaw fall into place and maintain themselves with minimum or no maintenance effort is a dangerous illusion.
"The anticipated steady state at which all pieces of the jigsaw fall into place and maintain themselves with minimum or no maintenance effort is a dangerous illusion.
"This is why, back in 2006, the British Computer Society produced a report exhorting Connecting for Health (the IT arm of the Department of Health) to drop its technology-push mindset and get its head round the more organic notion of socio-technical change.
There has been progress
"Much progress has been made. Civil servants have embraced the concept, inasmuch as they understand it, and begun to use the rhetoric. The National Programme for IT is described in the corridors of Whitehall as comprising "10% technology, 90% business change".
"Much progress has been made. Civil servants have embraced the concept, inasmuch as they understand it, and begun to use the rhetoric. The National Programme for IT is described in the corridors of Whitehall as comprising "10% technology, 90% business change".
"Senior staff in Connecting for Health wince when reminded of the "ruthless standardisation" policy introduced in 2003. Dialogue between stakeholders and adaptive use of technologies to produce locally-relevant benefits are now strongly encouraged.
New Zealand shuns summary care record - and has successful patient record system
"These significant gains notwithstanding, the National Programme for IT is still largely predicated on the kind of assumptions and business processes that built the Model T Ford rather than the ones that allowed New Zealand to produce a technically robust national electronic patient record system which is being delivered within budget and simultaneously deemed acceptable by the professions, civil liberties campaigners, government, service users and the commercial IT sector.
"These significant gains notwithstanding, the National Programme for IT is still largely predicated on the kind of assumptions and business processes that built the Model T Ford rather than the ones that allowed New Zealand to produce a technically robust national electronic patient record system which is being delivered within budget and simultaneously deemed acceptable by the professions, civil liberties campaigners, government, service users and the commercial IT sector.
"Significantly, New Zealand has eschewed the notion of a centrally stored summary record as rigid, unaffordable, unnecessary and potentially monolithic; it has focussed instead on distributed data storage, agreed interoperability standards, and messaging frameworks.
"There is no tightly-specified over-arching goal or fixed milestones; rather, there is an agreed process whereby goals, milestones, standards and progress metrics are continually negotiated, reflected upon and revised.
Conflicting perspectives
"The NHS is not a homogeneous entity with a single point of contact for negotiators. Every strategic health authority, every primary care trust and every GP surgery up and down the country has its own (implicit or explicit) policy on the different technologies in the National Programme for IT and - crucially - its own Caldicott Guardian on whose job description the protection of patient confidentiality in that organisation lies.
"The NHS is not a homogeneous entity with a single point of contact for negotiators. Every strategic health authority, every primary care trust and every GP surgery up and down the country has its own (implicit or explicit) policy on the different technologies in the National Programme for IT and - crucially - its own Caldicott Guardian on whose job description the protection of patient confidentiality in that organisation lies.
"Professional perspectives conflict with managerial ones and the IT department often sings in a different key entirely. Front line NHS staff are not cogs in a machine but masters and mistresses of the creative workaround that gets the job done despite limitations of time, space, technology and capacity. Some of us owe our lives to those workarounds.
"The first generation of Connecting for Health top brass perhaps failed to appreciate that progress in developing a national electronic record system depends above all else on an ongoing process of sensemaking - that is, on the buzzing, blooming cacophony within the NHS and beyond it on whether a particular technology is, on balance, a "good thing"; on whether it should be implemented right now in its current form and through standard operating procedures, or at some other time in some other form and in idiosyncratic ways; and on whether to rip-and-replace wholescale or retain (in some form or other) a particular devil-we-know legacy system.
NHS Connecting for Health's enduring legacy
"Their successors have a far more nuanced and appreciative understanding of how negotiation, co-design and interpretive flexibility by local teams embeds and strengthens technologies-in-use. But their ability to apply this understanding is severely constrained by what some of them have described as Connecting for Health's enduring legacy: a non-adaptive, reductive, anti-dialogue Gantt chart culture.
"For the past 13 years, the NHS has led the world in a pioneering natural experiment to test the hypothesis that large, densely interconnected and centrally managed information systems would greatly increase accessibility of data (and thereby quality, efficiency, transparency and trust) as well as bringing economies of scale.
"The first generation of Connecting for Health top brass perhaps failed to appreciate that progress in developing a national electronic record system depends above all else on an ongoing process of sensemaking - that is, on the buzzing, blooming cacophony within the NHS and beyond it on whether a particular technology is, on balance, a "good thing"; on whether it should be implemented right now in its current form and through standard operating procedures, or at some other time in some other form and in idiosyncratic ways; and on whether to rip-and-replace wholescale or retain (in some form or other) a particular devil-we-know legacy system.
NHS Connecting for Health's enduring legacy
"Their successors have a far more nuanced and appreciative understanding of how negotiation, co-design and interpretive flexibility by local teams embeds and strengthens technologies-in-use. But their ability to apply this understanding is severely constrained by what some of them have described as Connecting for Health's enduring legacy: a non-adaptive, reductive, anti-dialogue Gantt chart culture.
"For the past 13 years, the NHS has led the world in a pioneering natural experiment to test the hypothesis that large, densely interconnected and centrally managed information systems would greatly increase accessibility of data (and thereby quality, efficiency, transparency and trust) as well as bringing economies of scale.
"On the evidence accumulated to date, we should now carefully lift the baby from the bathwater and move on. Much has been learnt. Many of the components of a flexible, adaptive new partnership between government, industry, the NHS, academia and the third sector could be assembled relatively quickly.
But first, the elephant must be shown the door.
Links:
Full report on evaluation of summary care record - University College London
Do summary care records do more harm than good? - Ross Anderson
A defence of summary care records - Mark Walport [Wellcome Trust]
SCR - time for a rethink - Roz Foad
Clinicians may not access summary care records - IT Projects Blog
Subscribe to blog feed
..and it cost us, what, 16 billion or so to run this pioneering experiment to confirm what any number of us could have told them in the first place.