Insider tales and must-read analysis on how integration is reshaping health and care systems, NHS providers, primary care, and commissioning. This week by deputy editor Dave West.

The biggest row so far over a clinical commissioning group merger has erupted in Staffordshire – a patch with large, long-running problems in the health service, just a tad more serious than commissioning boundaries.

The CCGs have dropped a merger proposal after it became clear it had been voted down by GP members – the local medical committee advised them to vote against it and said in a letter the merger had been “driven by NHS England”.

Their NHS England/Improvement region – the Midlands – feels this is unfair. And, at a guess, it’s likely the merger proposal is driven at least as much by some of the system leaders in Staffs as by NHSE itself.

What NHSE/I is absolutely doing across England is encouraging, incentivising and cajoling CCGs into merger (again, normally with the welcome and backing of some senior leaders in each system).

This has – not irregularly – veered into an exaggeration of the policy Bible on this: the NHS long-term plan rule which says every integrated care system (which everyone must become) will “typically” be covered by only one CCG.

In an unsurprising development, a lot of those who want to see more mergers to bigger CCGs have simply been removing the “typically” when spreading word of the rules.

A letter from the director of another NHSE/I region – East of England – shows all CCGs in that patch have been told there is a “national expectation of having a single CCG per ICS by April 2021”. Note, no “typically”.

Meanwhile, there have been talks about formally tightening up the rule by putting into ICS guidance that, in order to be a good ICS, there must be only one CCG.

The hope of NHSE/I and allies on this is as many as possible will take the hint and merge to match their STP/ICS.

I asked Mr Stevens about the issue at our integrated care summit the other week. The event is held under the Chatham House rule but he agreed to put comments about this on the record:

“In the long-term plan [we said there] will typically be one CCG per ICS. We also said that we are open to people saying they think the current boundaries or geographies of the Sustainability and Transformation Partnerships are wrong in whatever ways, and can we make adjustments to them.

“We’d like to have that discussion with local organisations and systems as part of signing off their local long-term plan implementation plan. So, yes, generally speaking, we will have one CCG per ICS because that will help [prevent] too much complexity and make the governance relationships easier.

“In some places, the ICS geography is not going to be the same as the current 42 STPs/ICS and we’ll have that discussion.

“But… let’s not just focus on organisational structures at the expense of the job we’re trying to get done, which is [develop] joined up services for patients.”

He also noted: “We’re also being required to have a further 20 per cent cost reduction, and the reality is that to fully service lots of small organisations, without shared management infrastructure, is getting increasingly hard.”

A little surprising to me was the clear suggestion of the option of changing STP/ICS boundaries – ie splitting them up. This has not happened at all so far and, in the past, senior NHSE/I figures have been very cool on it – instead wanting systems to be large, if anything. West of London, Berkshire West ICS has only just been persuaded to accept merger with Bucks ICS and Oxfordshire, into a single ICS, for example.

It does present a practical step for a few – splitting North East London STP into the traditional inner and outer NEL might be one. But if it gets out widely that splitting is an option, many would rush to embrace it, probably piling more complexity into the picture.

A most likely and fairly logical route is to allow only a handful of STPs/ICSs over a certain size – well over a million population, probably – to have several CCGs; or, in even fewer cases, to split into several ICSs.

This probably does not solve your Staffordshires (just over one million people), nor systems like Frimley in Surrey – population of only 740,000, but which feels it has been getting on perfectly well despite its fragmented CCG and local government, and would probably choose to avoid the upheaval.

According to Mr Stevens’ comments, NHSE/I will be hoping to thrash this out individually with systems throughout the autumn and winter.

The likely final landing place is that there will be more than 42 CCGs, but substantially fewer than the current 191 – and many fewer too than the 150-odd top-tier local councils.

Around 60-70 seems a likely end point, and appears to hold with the Nigel Edwards rule that no previous number of NHS middle-tier organisations can ever be repeated.