Primary care networks as a means of supporting primary care: findings from qualitative case study-based evaluation in the English NHS

Community level

One of the key policy goals associated with PCNs in England is to support the development of neighbourhood-based collaborative service provision, working with other providers across their geographical area, with an assumption that this type of integrated working would lead to more care being provided outside hospitals. In practice, we found that in our case study sites this was not generally a priority, in part because incentives associated with PCNs are focused on internal activity, but also because practices felt themselves to be under pressure and therefore unable to engage with additional activity beyond their core service delivery. Within this, there were a number of factors which seemed to influence the extent to which PCNs were able to engage more widely.

First, the extent of pre-existing working relationships in a local area were important. In many sites, pre-existing collaborative arrangements at neighbourhood level were operational before PCNs were announced. These ‘neighbourhood teams’ were often orchestrated in some way by the local commissioning organisation or driven by local collaborations such as federations, and often included representatives from GP practices, the community trust, the mental health service provider and third sector organisations.

The extent to which pre-existing neighbourhood teams and newly constituted PCNs were able to integrate and work together varied between sites. In site A, where neighbourhood teams were well established and resourced, the arrival of PCNs was associated with attempts to neighbourhood teams ‘wrap neighbourhood teams around PCNs’ (N03032, October 2020). One interviewee talked about the PCN being the ‘yolk’ in the neighbourhood team ‘egg’, and emphasised the necessity of a functioning general practice collaborative entity to the productive operation of the neighbourhood team (N720sr, October 2020). However, PCNs are constituted on the basis of practice populations, not geography and this could cause confusion:

…so the [‘neighbourhood teams’] would have been based on a geographic footprint of working together. The PCNs have been on the basis of, well, we own this practice, this practice, this practice and this practice. And that’s caused no end of confusion for some people in terms of how that then kind of comes together. [N3701q]

Furthermore, the contractual requirements of PCNs through the GP contract were perceived as a barrier to more extensive integration of PCNs into neighbourhood teams.

So in some areas, the PCNs are working really well in partnership. And in other areas, I think they’re just not as advanced in their ways of working with other providers round the table. And I think, to be honest, it’s probably been confusing for some of them because [‘neighbourhood teams’] came first. […] Well, neither has precedence, you just work together, that’s the whole purpose of what we’re doing. You know, nobody…it just so happens the PCNs have got the money.

And that has, to be honest, probably caused some friction because actually when that money gets kind of split out, it doesn’t encourage the PCNs to work in partnership, it encourages them to work within their own footprint. [N3701q, July 2021 _Site]

In Site C, a pre-existing neighbourhood team model was in place and PCN arrangements mapped closely to the footprints of these teams. Efforts were underway at the site level to adapt the provision of community services so that it was more coherent with the PCN and neighbourhood team geography. The nature of the dynamic, however, between PCNs and neighbourhood teams was somewhat unclear with different interviewees framing this differently. For example, one interview said this:

Where do PCNs stop and where do neighbourhoods start. The reality is there will be some things that PCNs will lead on in terms of projects and in terms of delivery, there will be some things that neighbourhoods lead on and there will be some things that we will do collectively together. I think partners, how do I say this, what we’ve, primary care networks I think creates a perception that they are led by and owned by primary care and you’re putting primary care right at the centre and obviously you’ve got clinical leadership that is there. But actually in terms of delivering population health needs, that isn’t just primary care that’s delivering on that on their own, it needs to be them alongside their other partners as well. […] So primary care is a really, really important building block, but it has to be partnered as equal working together. [N46026-Site C]

Whereas another talked about it in this way:

Yeah, so Neighbourhood and Networks, I keep saying to the Neighbourhoods, and we’re doing a lot of work around that, we need to stop referring to them as Neighbourhoods and Networks, ‘cause they’re all as one, essentially. So, we call them Networks within Neighbourhoods, all these pilots are taking place within the Networks, on behalf of the wider Neighbourhood if you like. […] Neighbourhoods preceded Networks, but actually, and probably unknowingly at the time, actually they were delivering on the Primary Care Network model. [N520KK-Site C]

These different perceptions of the role of PCNs in Neighbourhood working need to be resolved if cross-sector working is to be effectively established.

Beyond the general question of neighbourhood-working. PCNs were required to participate in multidisciplinary team working to provide additional services to people living in Care Homes. Under the compulsory ‘Enhanced Care in Care Homes Service Specification’, PCNs were required to allocate Care Home Residents to a participating practice and to work with other agencies to deliver comprehensive care.28 In some areas such services had already been established, but in others this was not the case. The specifics of the funding and contractual arrangements in place could render this problematic, as this quote from a representative of a community service provider organisation makes clear:

So that’s where some of the arguments came around care homes and enhanced care in care homes because primary care has now been incentivised to do that through that enhanced care in care homes scheme. […]

What we’re saying to primary care is, yes, of course we’ll do what we can, but we haven’t been given any additional resource and actually we’re trying to do all the other work that would fall by the wayside if we did that. So that’s quite challenging. But yes, if we really want community and primary care to work in an integrated way, we need the same incentives and the same contract. [N1018c, Nov 21, Site A]

In a small number of our sites more ambitious programmes of cross-sector working were being established, but this was very resource intensive, requiring clinical directors to work well-beyond their contracted hours to develop the relationships required to support this type of activity. Adequate and flexible management support was an important factor in allowing this to take place.

Finally, policy guidance suggests that one of the advantages of PCNs is that they will be able to work across their neighbourhoods to established wider programmes of Population Health Management, identifying and targeting support towards high risk individuals in a neighbourhood. In practice we did not find this to be a priority for our case study PCNs, with limited understanding among those we interviewed as to what this approach might involve.

Overall, we found that community-level neighbourhood working, while acknowledged as a potential benefit of PCNs, has yet to be established. Those areas with pre-existing good relationships with other providers were at an advantage, but current incentives and available resources do not fully support this activity. Developing the required trusting local relationships is time-consuming, and requires managers and clinical leaders with the time and skills to invest in this activity.

Network level

At the network level, the key policy objective was around supporting general practices to realise the claimed benefits of at-scale working and to stabilise a care sector under significant pressure. The main mechanism by which this is to be achieved is via funding for additional staff—known as the Additional Roles Reimbursement Scheme, ARRS—and via the softer benefits of working together. The ARRS provides direct reimbursement of 100% for staff employed to work across the PCN. Table 4 sets out the types of staff who could be recruited and the numbers available across all PCNs.

PCNs are not themselves legal entities, and so are unable to employ staff. This means that a variety of contractual mechanisms can be used to provide staff under the ARRS scheme, including:

• Employment by a single ‘lead practice’ on behalf of the PCN.

• Employment by another legal entity such as a legally constituted Federation or other body.

• Contracting for services from another entity such as an NHS Community Trust or a voluntary sector service provider for the provision of a service—may not always be the same individual.

• Contracted from an agency, such that the workers are independent contractors.

Some of our case study PCNs suggested that they would prefer not to be the ‘lead practice’ employing staff because of the implications when there are disciplinary issues, or pensions administration. One suggested that the reluctance of some practices to employ staff directly was underpinned by scepticism over the longevity of PCNs, which led to concerns about liability for redundancy payments should the scheme cease. In one PCN the practices will have nothing to do with the employment of ARRS staff. Our findings mirror those of Baird et al,29 who studied the early employment of ARRS staff and highlighted the crucial importance of managerial and HR support.

Recruitment to these new roles tended to be more straightforward in those PCNs with established trusting relationships, but in many areas there have been problems in filling some roles, associated sometimes with a shortage of particular professionals.

… pharmacists play one PCN off against the other to get a higher banding. So that it’s created that intra—PCN war of attrition, in some places, that people, well, just play each other off to get the highest banding that they possibly can. Which you can’t blame staff for doing, but it’s creating probably some inequities, maybe, within the system. Some people will pay higher and take the risk that they’ve got a shortfall that they’ll make the money up with. Others will say, no, we won’t go higher than this specific banding. N0303t_231020_LWG

Contracting with other organisations for staff did not always alleviate these issues, and ARRS funds were not always spent.

‘In the other one, ARRS fund has not been utilised. It’s been contracted, there are two pharmacists for 17 practices of 90,000 patients . There’s a physiotherapy service that we have again contracted through another private company, but they don’t have enough staff to provide to all the 17 practices so currently they are only providing it to seven practices and they are advertising to recruit more physiotherapists but there aren’t any new recruits that they were able to successfully make. So kind of literally we haven’t utilised our ARRS fund.’ N570mu_090721_DB

Funding for ARRS staff is relatively rigid, with a number of rules in place which could sometimes make things difficult. For example, funds unspent in 1 year cannot be carried forward into the next, and in the first 2 years the types of staff who could be employed were highly specified. Funds unused could not be recycled into other aspects of patient care, and only salaries could be funded. Most of our case study sites were also struggling with accommodation for their new staff, as many buildings were full without spare clinical spaces. There is no additional funding in the scheme for estates, and so this was potentially problematic:

We’re not that lucky, we’ve got nowhere to go. Then there was a bit of…to me, a bit of an issue around funding. So we’ve been forced to make networks in, what, 2019 this all started. We’re getting forced to become a network. We’re getting forced to recruit staff and spend money. No one’s given us anywhere to live. So we’ve got staff on the streets basically. [N011c6_B_Feb22]

Our participants also told us that integrating the new staff into practices could be difficult, particularly if their roles were not understood or if they were moving between practices and only spending short amounts of time in each. Training, development and supervision were all required, and this was not always straightforward for newer roles:

So that now all comes into it, which is a good thing, it’s not a bad thing, because all of these staff need development. But, the training pathway and the support packages are coming out now and they’re too late. For example, the trainee nurse associate package, you know, people have already got them in place, and they’ve got different pathways, and now you’re saying that they have to go on this accredited one, so it’s going backwards to go, and I appreciate you sometimes need to go backwards to go forwards, but it’s unnecessary. [N601jg_SB]

In summary, the ARRS was welcomed, but has proved quite complicated to operationalise, with complexities around many aspects of the scheme. It is too soon to tell whether the anticipated alleviation of GP workloads has materialised, with some respondents telling us that the supervision required for new staff meant that time savings may be limited. PCNs with trusting relationships were at an advantage in employing staff, and in the longer term the experience of working together to employ staff collaboratively in this way may be beneficial in cementing those relationships.

Member level

The COVID-19 pandemic provided an early test of the extent to which PCNs were able to provide the softer benefits associated with working more closely together at scale that were anticipated in the policy. PCNs were deeply involved in the pandemic response, supporting practices by establishing local ‘hot’ hubs to assess patients with COVID-19 and participating in the vaccination programme. Many respondents told us that the experience in the pandemic had acted to accelerate the development of trusting relationships, but it was also true that where interpractice relationships were poor or dysfunctional this could prevent effective collaborative working.

It brought us together in a way that crises can do, so it brought us together as an organisation, and particularly we moved quickly to harmonise quite a few of our processes to get us through COVID, and there’s been a lot of cross-site working in cross cover and help, so it’s really brought us together from that point of view. [N060fj_D_Apr21]

Again, I don’t really know where it all changed and why it changed. I know that we’ve…I feel a lot closer to people since we’ve started doing COVID clinics. ‘Cause I’m very much involved in the COVID clinics and I’ve got a really good relationship, and we never had a relationship with [X] before. And I’ve got a really good relationship with both the doctor and the practice manager there. I don’t know where it all changed, I don’t know if it was when we were starting to talk about clinics, and it was such a massive relief knowing that [X] took on the bulk of that work… [N250wt_A_Mar21]

Initially, it would…I don’t know how to say it really without sounding derogatory, but it was like every man for himself….So, the big practices were like, right we’re sorted bish-bosh, while little practices were, oh my God help me, what about me. And it got quite…we were having lots and lots of Teams meetings trying to sort it all out.[N050oz_A_Mar21]

The latter quote highlights one of the important factors determining how PCNs have been able to work together: their internal configuration and relative size/power of constituent practices. As noted above, PCNs are heterogeneous, with some made up of practices of similar sizes and covering similar populations, while others are much more mixed, for example one very large practice and a number of smaller ones.20 In the longer term, these dynamics are likely to be important, as practices within PCNs are required to work together to deliver services and collaborate to meet incentive targets. Good internal relationships, with mutual trust and reciprocity, will be very important in this. Our study suggests that it remains early days for PCNs, and these dynamics will take time to settle down.