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.

There are few better illustrations of the varied landscape across integrated care systems (“if you’ve seen one ICS, you’ve seen one ICS”, as it’s often said) than the role of district councils.

Sixty per cent of England’s population doesn’t have one, but in the many areas where they do exist, they’re a big part of the municipal structure, and it’s impossible to see how ICSs could meet their objectives without districts playing a part.

But unlike with upper-tier councils, their involvement in ICSs is not a legal requirement, and, as is pointed out in a new King’s Fund study, it’s very variable whether it is happening. 

The District Councils’ Network-commissioned report sets out the benefits of embracing them, tips for how to do it, and examples of successes where ICSs have done so.

It is well worth a read and I suspect the tips apply to a lot of other partners too (there’s even a brief “dummies guide” especially for ICS bosses…). Reading the report also reminded and reinforced for me some of the important trends and dilemmas looming large in ICS land.

Above all is that the NHS is going to have to stop worrying about organisational boundaries, trying to redraw them or expecting them to line up – and instead get used to ignoring them, working around them, and making informal relationships across them. 

Place

Not so long ago, when speaking about ICSs, both the government and NHS England were at pains to stress the role of “place”, and many systems are still bought into it. More recently it’s dropped off the national agenda, with limited attention from the Hewitt review or at last month’s ConfedExpo conference. Formal delegation to place has all but evaporated, and may be nibbled away further by management cuts, but there are many good “place” leaders in post keen to get on with transformation on their patches, relying largely on partnership and soft levers.

Even at place level, though, it’s rare that places match the boundaries of district councils, even though the latter have a clear (and statutory) role in health promotion and would often see themselves as a natural “place”. In many cases, larger ICSs have chosen upper-tier local authorities as “places”; elsewhere they are below county level, but still bigger than districts. Some single-county ICSs – Gloucestershire and Lincolnshire – have chosen no places; although that doesn’t mean they don’t have local operations and may be working with districts (Lincs is one of the examples praised in the King’s Fund report). 

Rural v urban

District councils cover plenty of towns and cities, but not the biggest cities and urban conurbations, and in many cases are rural. There’s a screaming need in health policy to remember that a single solution won’t work everywhere – especially across the urban/rural divide – and it’s easily forgotten.

One example is that getting acute hospitals to work together on running services (either informally or through the fast-proliferating trust mergers and groups), whether by centralising them or by sharing staff and other “mutual aid”, is proving quite helpful in more urban areas but is much less use in areas where hospitals might be an hour’s drive apart with no public transport. Similarly, there is resistance in more rural areas, especially the South West, to merge ICSs to get them up to subregional patches covering 2 million or so populations.

The race to the ‘neighbourhood NHS’

The urban/rural tension in primary care might be even greater. Arguably the traditional GP model is very broken in London, parts of the South East, inner West Midlands, and other very urban, often deprived and racially diverse areas. But it is holding up a bit better in some more rural patches (see the variation in the GP patient survey or vaccination rates, for example).

The fixes could be quite different too. Choice in primary care has more meaning in London and digital options are different: Babylon GP at Hand has a 100,000+ list across the city, with only five central physical clinics. It’s an anomaly (and how long will it last?) but it shows what’s possible.

Reading the runes, “neighbourhoods” are where the action is going to move next, with Keir Starmer declaring in May that “the NHS must become a Neighbourhood Health Service” (his capitals). He wants a new service offering, “close to communities”, in between the current polar choice of “A&E or the GP”, he said this week – perhaps resonating with a much wider set of direct-access primary services suggested by the Tony Blair Institute’s latest health paper.

That may be a very simplified interpretation of what’s already available and is needed for the future, but there’s little political disagreement on the fundamentals, and they chime pretty well with what many NHS leaders would like to see.

Claire Fuller’s stocktake report has been on ice for a year but has general support, even though there will be big rows about speed, funding, and how hard to push the GP contract, employment and premises models.

The shift will mean GPs and primary care networks – along with a bigger clutch of other services, NHS and non-NHS, spanning several providers – having to transmogrify into integrated neighbourhood teams (and, sometimes, new premises (capital, please?)).

But returning to the urban/rural issue and district councils, an “NHS neighbourhood” of a 30-50,000 population, or perhaps up to 100,000, means something very different in Somerset than it does in Southwark, in all sorts of ways.

In some areas these neighbourhood services could match district councils: About a third of districts are under 100,000 population (but none under 50,000), and a few areas have already begun arranging delivery of community support on these patches. The King’s Fund/DCN report cites work in Leicestershire, for example, where district councils are using their housing functions on “neighbourhood” patches to try to speed up hospital discharge.

More often than not, though, “neighbourhoods” are and will be smaller than districts – reinforcing the feeling that the integration map isn’t going to be getting more straightforward anytime soon.