GP System Change

As the current arrangements for the supply of GP systems under the NPfIT and GPSoC come to an end GPs maybe thinking about a potential system change. In particular some emerging CCGs are promoting system change to achieve a single system on their patch, but my advice would be to wait and I say this for two reasons.

Firstly, because of the pain involved in a system change and secondly because changes in the way IT is delivered as a result of the app revolution, probably removes many of the reasons for a change.

Most GP practices have been computerised long enough to have experienced at least one major system change and will know how painful it can be. I wrote a document in 2004 which included a section about system change and most of what I said then remains as true to day as it was then (click here for an edited extract). A change, even to something clearly better, is likely to result in a period of at least six months plus before a practice gets back to where it started and I would argue that for users of any one of the current GP systems there are not any alternatives on offer that are “clearly better”. Any practice thinking of changing needs to think carefully if the benefit of any change justifies the costs and loss of revenue which inevitable flow from a such a change and what the alternatives to such a bruising experience might be.

The app revolution promises to offer software from third parties that will extend and enhance existing systems without the need to replace the underlying core system. These apps will offer new functionality along with improved user interfaces and an enhanced user experience on top of existing systems.

At the same time as the revolution at the user interface overwhelming Government commitment to open systems and open data, restated in the recently published NHS Information Strategy, along with new technologies to integrate data at the back-end means that the drive towards a single system approach to integration at the level of a care community, that is often cited a the reason for system change, is no longer valid.

Better Integration of care and the more effective sharing of data is central to meeting the challenges faced by the NHS, but the way to achieve this at the level of a care community is no longer through lowest common denominator single system (if it ever was?) Historically many PCTs have pushed single systems solutions and this view remains disturbingly present with some of the emerging CCGs. I suggest that those who still see this approach look again and that individual GP practice resist the pressure to take the pain of a system change for the greater good, when there are actually better ways to make the gain without the pain.

Most GP are reasonably well served by their current system and there are new was to meet GPs IT needs just around the corner so my advice is to sit tight – Now is not the time to change.

Edited extract of relevant section  from my 2004 document is reproduced below. The full document is available here, although much of it is only of historical interest.

Most GPs have experienced a significant upgrade or replacement and know that this involves considerable work and no little pain: re-training staff, re- engineering or re-implementing business process and converting data. They also know that hurried, poorly planned or poorly implemented upgrades can be a disaster.

GPs have large amounts of data relevant to current patient care. Data conversions as part of an upgrade have been problematic in the past, but cannot be avoided. The expertise exists to build data conversion tools and processes that are safe and effective, but individual conversions will require significant work by practices in preparation and data validation that cannot be done by others. GPs will want assurance that a clinically and medico- legally adequate data transfer will be made to new systems, and are particularly concerned that this may not be the case.

Any upgrade will require a GP Practice to make a considerable investment in time and effort if it is to be successful, and it must be accepted that initially a new system will do less and do it less well than the system it replaces.

The impact on “utility” of a system upgrade is illustrated by the graph below.


Established  systems  reach  a  point  (their  “utility  ceiling”)  at which  their original design and technological foundations make it increasing difficult to add new facilities, take advantage of new technologies or implement new ways of working.

Irrespective of how good a new system is at the point of upgrade there is an inevitable loss of utility (L) as users learn to work with a new system, retrain staff and re-implement business processes.  In a highly IT dependent environment like general practice this loss will be material and at worst can result in the practice becoming dysfunctional.

In a well judged and managed upgrade utility recovers and users eventually find themselves back at their starting point. Previous experience tells us that for a well managed upgrade the recovery time (R) is 6-18 months

Finally, as systems mature and users learn to exploit them; utility grows to levels not reachable with the old system until the new system too reaches full maturity and a new utility ceiling. It is this gain (G) that justifies the upgrade.

If GPs are to be persuaded of the need to upgrade they will need to be convinced that the quality of the new system and upgrade process is such that the loss at the point of upgrade and recovery time are minimised and survivable, and that the new system has proven potential to provide an eventually substantial utility gain. An honest approach that sets expectations appropriately will gain support whereas unrealistic promises will result in strife.

GP practices will accept some pain in order to promote the greater good of the NHS and participate in a programme designed to lay foundations for the next generation of NHS IT which does not immediately bring them benefit. However, such altruism has strict limits and GP practices need to be persuaded of the need to upgrade primarily in terms of the direct benefits to themselves.