NHS IT SHARED SERVICES- IS SHARING ALWAYS A GOOD THING?
In my last post (a bit of a rant about the Carter report confusion between value for money and absolute spend), I noted that the report contains an interesting sentence relating to what NHS Trusts should do if they aren’t meeting the benchmark of turnover being spent on IT. Which is write off to NHS Improvement with a plan of how you’ll get your spend down. But the interesting bit is where the Carter report states such plans “should include plans to commit to national shared service models”.
This is interesting, as there’s a whopper of an assumption in this statement that shared service models lead to lower costs. It’s a solid assumption though. Rationalising services and sharing staff and infrastructure should save you money. That’s a fairly solid model that underpins the concept of outsourcing and economies of scale, and the success of that concept is backed up by the $300bn market that’s grown up around it.
But what about quality? This relates to my criticism in my last blog. It’s not about absolute spend, it’s about value for money. But again, the size and competitiveness of the outsourcing industry would tell you there’s high quality and high value for money services out there. Specifically in the NHS, there’s some excellent CSU’s out there now that the weaker ones have folded.
So far so good, with a little bit of effective market competition, the NHS has at last got some decent shared IT services out there, opening the door for more NHS Trusts to ‘commit to national shared service models’ and lower their IT spend. Other Trusts have successfully outsourced their IT services to private sector IT service providers.
Oh no, hang on, no they haven’t have they? A brief history of NHS IT outsourcing doesn’t make good reading. I can’t think of one large scale outsource that has been successful, and I won’t even mention NPfIT, which people forget was largely a massive outsourcing experiment.
So why can’t the NHS outsource its IT very well? If at all? I think the answer lies in Wardley mapping. Wardley mapping is pretty simple (in concept at least), and is based on the principle that the closer to your user you get (or what your user perceives as valuable), the more flexible and specialist your service needs to be. The further away you get, the more you need standardisation and stability. This standardisation catalyses innovation above it- just look at what happened when electrical current was standardised.
So Wardley mapping tells us that the NHS should look to outsource anything that is well understood, standardised and available as a commodity. The less perceived value to the user the better, in fact a crude rule of thumb could be ‘if staff don’t care about it, outsource it’. I mean, who on the front line of the NHS cares about servers?
But the key to why the NHS doesn’t outsource very well is in the question of if something is well understood. Only if something is well understood should it be a candidate for outsourcing. And therein lies the problem. How many IT outsource companies understand the NHS in all its complexity? How many understand that the difference between the label printer for transfusion bags failing and a label printer for transfusion samples failing? How many understand how ward rounds work? The answer is none. But many do understand how to provide data centres, servers and networks, much better than the NHS does.
So in summary, the issue lies in the NHS trying to outsource areas of IT that rely on specialist knowledge, such as an IT help desk that requires an intimate knowledge of how each area of a hospital may work or device provision that relies on knowing how clinicians interact with devices in each clinical context. There’s no way an IT service provider could or would have that knowledge.
But the NHS is also guilty of not outsourcing in areas where specialist NHS knowledge isn’t required. You don’t need to know how pre-operative assessment works to host a server or provide WiFi.
Outsourced complexity ending in failure is evidenced by the services from the CSU’s and HIS’s that have survived. By and large, they provide IT services to commissioners and primary care. Which are relatively simple in comparison to the complexities of IT in hospitals and the community, which is provided by in-house teams in the vast majority. All the evidence is telling us that outside of the tin and wire commodities of IT, the complexities of acute, community and mental health mean IT service provision in these areas cannot be done at scale.
And this is the danger of the Carter Report stating that Trusts “should include plans to commit to national shared service models”. There is definitely scope for reductions in IT spend through shared services. But it will only work if the NHS recognises the limitations of what can be outsourced, and what should remain in the hands of the experts the NHS already has.