Primary Care & Scale. Who should we be collaborating with?

Screen Shot 2018-06-15 at 16.42.13Seduced by Structure

As we wait for the next round of policy papers on the size and shape of GP practice organisations and primary care teams, I’m struck by how much structure always becomes the answer. Many reshaping documents start with the structures of the institutions rather than the identity and capacity of communities, and yet General Practice has grown out of communities, and practices are deeply embedded in their local place.

I thought it would be useful to offer a view about scale from the place of relationships and needs, rather than institutions and structure. But first of all the practices need to be ready to collaborate.

The Fantasy of the Future

The next seduction is to get into a fantasy about how fast the primary care system can move into these new collaborations. The starting place for the policy is well functioning practices with increasingly complex work, whereas many practices aren’t at the starting blocks yet.

Any development of ‘at scale’ collaboration needs the bedrock of efficient practices working with resilient communities.

We know that practices that are adapting effectively to the context of increase complexity [London Primary Care Quality Academy]:

  • Use data to understand presenting needs, review activity and improve the flow through the practice system. Understand their demand and ways of meeting that demand (data capture of patterns of demand and flow in the practice; how they offer appointments and who gets turned away; understanding the profile of their frequent attenders; and their low attenders; number of DNAs.
  • Collaborate with communities to coproduce services locally
  • Use their skill sets more effectively to meet need (diversify their workforce)
  • Use their contact time with patients more effectively (triage, mixed model of appointments)
  • Improve their back office functions to work efficiently (managing the paperwork)
  • Reach out to partner with other providers to manage complex health (care homes, social care, third sector, mental and acute health).
  • Learn fast, looking for examples outside their practice to steal with pride; establishing peer review for their clinical practice across professions. They test and prototype new ideas and check the impact on their demand and capacity using data.

This is the ‘Readiness Stage’, the foundations that practices need before they start collaborating (and merging/ partnering), because this stage exposes the nature of the actual work in practices, and where best the practice teams can intervene, and therefore where they need to collaborate to provide the best option. It also requires practice members to work collaboratively together, beyond the individual roles but as peers to bring all their experience and ideas as a team.

What Scale for What Work?

There are four main resources that secure health and wellbeing in communities –the community itself, the NHS, local government and the third sector. You can’t change General Practice without consideration for the other groups.

In my view there are these ‘scales’ for primary care service development:

Screen Shot 2018-06-14 at 13.48.10

Scale 1: The Community: Up to circa 14,000

General Practices grew out of local communities. These communities have natural boundaries, characteristics, relationships and resources. They have stories they can tell about themselves as a community. They recognise people from their community as being one of them. Most communities are at a scale of up to circa 14,000. This is the size of a small town. Go beyond that and people don’t feel like they are part of something they can identify with – it becomes something others ‘own’ and belong too. They don’t feel that its something they can contribute too.

At 30-50K scale that is the current favourite scale, people don’t identify themselves as a community, there are conflicting stories and different cultures. They see people as ‘other’ and its not a size that generates any sense of belonging.

The NHS could be seen as having generated an over dependence on its services borne out of the founding principles of the NHS ‘Can do, Should do’ and medical model ideology. There are many proponents of the assets that communities can and should bring to their health, and examples emerging beyond social prescribing (language that is laden with the intent that power stays in the hands of the prescriber see Corman Russell’s blog on this),to a partnership model with communities that has the potential to reduce demand.

There are masses of assets in communities or perhaps more accessibly – people who want to volunteer and help each other. We need these people to help with the volume of ‘patients’ presenting with ‘trouble with life’. This has to be the bedrock of any further ‘at scale’ working . See my previous blog on Reducing Demand in Primary Care.

Primary care for the future needs to be built out from resilient communities. So this is the first productive scale for primary care.

Scale Two: The Locality: 30-100K

(a) Across General Practices

At 30-50K

At this scale you can generate efficiencies in the Practice Business Model:

  • Sharing back office functions
  • Sharing workforce (particularly non-GP professionals and receptions staff)
At 30-100K

At this scale collaborating practices can provide some diagnostics, and more specialised rapid assessment and treatment functions to support GP/ Primary care teams work. See the work of Eastleigh Southern Parishes Locality

Aligning with Care Homes

Care homes provide better care where they are aligned with a practice. Within a borough a practice could specialise in this service for a geographical areas akin to Localities.

(b) As part of Integrated Teams

When you have resilient communities working to support local health and wellbeing, and you have functioning general practices, alongside collaborations for business efficiencies, then the next step is how to collaborate with other service providers to manage complex conditions, to prevent people with complex needs becoming unstable.

At the moment there seems to be a focus on growing these collabortions out of primary care, rather than developing integrated MDTs that bring the best of the skills needed from all service providers locally to bear on the complex needs of specific population groups e.g.

  • People who are frail
  • Adults who have both mental and physical health needs
  • Young people with both mental and physical health needs
  • People who have trouble with life which means they can’t cope with their health issues

These different groups require different MDTs with different skills and leadership. The General Practice team are members not neccessarily the leaders. For these teams the GP and practice members are peers in the team. As yet there is some way to go before we have the level of self-organising integrated MDTs that can work collaboratively with devolved authority to support people who have complex needs, preventing them from moving from stable to unstable. The MDT will have different professions taking the lead dependent on the severity of need. So for combined mental and physical health, it could be the community nurse at the lower levels of severity, a GP at a mid level but a Psychiatrist at higher level of severity. People with complex needs do need algorithms that identify the intensity and level of expertise of support required see the Intermountain work (Reiss Brennem at al 2016 ) and Western HealthLinks, Australia.

We at LSBU London Primary Care Quality Academy have the early ideas for the model here:

Screen Shot 2018-06-14 at 14.36.48

Scale 3: The Borough : 150K – 350K

At Borough scale there seem to be two main functions:

  • Some service delivery potential across a larger population.
  • Business Intelligence and Learning : A ‘Collaborative’ where members can work together to spread and share intelligence and to learn. This is in effect a development agency for the NHS in that place. Here Primary Care teams are supported by an Research and Development Team. Every industry needs its R&D function to innovate and adapt. Borough level organisations (e.g. Federations) can partner with AHSNs and CCGs to secure the best business intelligence (data) for quality, and to secure organisational development programmes for member teams, and clinical / professional leadership programmes for leaders.

All of these scales require different relationships:

Screen Shot 2018-06-15 at 15.43.32

And therefore different organisational forms:

Screen Shot 2018-06-15 at 16.06.44

And therefore different personal development capabilities:

  • At the Community and Practice level teams need good management development, team working and coproduction.
  • At the partnership level people leading these need good leadership skills for effective organisational collaboration.
  • At the locality scale people need peer leadership capabilities.
  • For the Borough scale people leading these need good Network Leadership skills.

As you can see a one-size-fits all approach might suit the NHS, but what’s needed is Collaborations borne from an understanding of need, and development to support the nature of that collaboration.

Do get in touch if you are interested in this work.


Frequent Attenders – Breaking the Cycle in Primary Care

People who attend GP practices on a very regular basis are usually seen as one and the same ‘type’ – elderly, multiple conditions (co-morbidities), often including a mental health issue. Despite the person booking multiple 10 minute appointments (sometimes a week), often all they are offered are more 10 minute appointments. Clearly this isn’t working, so what does?

It will be no surprise that there isn’t a magic bullet, but there are a range of approaches that can help, could work and do work.

The first step is to know who these people are, at a level of detail that helps you classify them beyond the number of times they attend (though that’s a good starting point). Morriss et al (2012) found the top 3% of attenders are associated with 15% of all appointments, alongside increased in hospital visits and mental health indicators.

Start with data

So who are people that turn up all the time? Whilst you will have some ideas (and prejudices) sometimes it turns out its not who you think it is. Here is an example from one practice in our London Primary Care Quality Academy, where we found that the top 5% of patients may be using 20% of GP resources at the practice.

The practice may not be thinking in a joined-up way about how it is spending this resource.

Screen Shot 2018-06-05 at 18.23.23

Not only is the practice not meeting these people’s needs, they are spending a lot of money in the process, you can see in the next diagram how this practice is spending 1.5-3K on its Super Attenders, and if you take the people who attend every 3 weeks or more, over 5 years this practice has spent £1million on a service that’s not working.

Screen Shot 2018-06-05 at 18.24.31

Now if you look at who these people are in this practice we had a surprise, as you can see its not all mental health, or old people – it’s a range of people with a range of conditions – so what’s going on?

Screen Shot 2018-06-05 at 18.25.14

The next step is to conduct a deep dive into a range of these people individually. In this case it looks like the problem is ‘trouble with life’ i.e. these people have conditions that other people are coping with, but their context is much more messy and complex and they just can’t cope. Here is a first review of the top 100 in one practice.

Screen Shot 2018-06-06 at 11.36.33Screen Shot 2018-06-06 at 11.38.18

These deep dives are Patient History Maps which are deliberately created manually. By having a summary history map, we can begin to look for insights into how services have interacted with the patient. Where there have been handoffs, where there has been failure demand, where education and other services have been effective. We seek to understand why services and systems act as they do, and the thinking behind it. In summary we seek to understand context and how perhaps sometimes services treat presentations, but not slow the decline in health. They complement the quantitative data captured above. These maps present professionals and senior management with the vital opportunity to do their own enquiry – creating a normative learning loop (Downham 2018).

It is likely in your frequent attender group that you have a range of people, which we have categorised differently as they need different responses:

  1. People with multiple needs
  2. People with increasingly complex physical needs
  3. People with increasingly complex mental and physical needs
  4. People with extremely complex situations which means they can’t cope with their physical/ mental condition.

You can also take a needs based approach to discover the people who are likely to become super attenders – the ones in the low end of the frequency spectrum who have chronic conditions that are relatively stable, but are at risk of becoming unstable (the frailty index for older people would be a good starting point).

Determine the type of need

Clearly going round and round the system isn’t working, and we know that often multiple attendances in general practice also equates to multiple attendance in A&E, which in turn equates to the burden of multimorbidity which is independently associated with social deprivation (Hull, 2018).

The next step is to work through the list and choose from a range of interventions for this cohort of primary care users as follows:

  • People with multiple conditions, which are chronic but stable.

These people (often older in terms of prevalence, but they can be any age) with multiple conditions where a 10 minute appointment only addresses one of their multiple concerns. Here the evidence is that continuity of clinician and longer less frequent appointments scheduled regularly and self-bookable by the person is the best solution[1] (Deeny et al 2017). In terms of best practice, it is considered to be an initial consultation of circa 1 hour[2], and from then on the clinician and person decide how much time the follow-up appointments need, reducing over time as able. Overall this signals that the practice is offering a more tailored consultation approach to meet needs (see log on Reducing Demand in Primary Care part 1 and 2), and so this does need to be part of the change in how the GP system works with triage, e-consult, diversifying the workforce, and offering walk-in and bookable appointments, as this generates the headroom and appointment space to be able to offer the longer appointments.

  • People with mental and physical health problems.

Here the Intermountain approach could work, where the introduction of team based assessment and care reduced hospital admissions, (Reiss-Brennen et al 2016). This approach is to:

  1. Assess the person’s mental health severity, physical health and life situation/social factors to build a holistic picture of their health and complexity of the context they operate in.
  2. Review the assets available within the person’s own context (family / friends) and build the team needed to meet that person’s needs including the person’s own assets.
  3. Develop an integrated team response for this person.
  • People who lack confidence, are undervalued, lack meaning.

These people need to use their skills and make a contribution to feel valued. All the work on coproduction shows that people who are seen to be the most ‘needy’ by the institution, have amazing capabilities to offer, and if used it brings phenomenal self-efficacy (e.g. Coproducing Leeds). Taking a partnership approach with communities to develop new ways of bringing people together (an asset based community type approach), where the practice invites people to work with them to support the health and wellbeing of communities, and co-develops and co-provides a number of activities that regain people’s creativity and fun, and tackles the social determinants of poor health. Healthier Fleetwood is a great example of this, as is the Health Champions approach of Altogether Better. Although as Husk 2017 whilst all good sense tells us that these approaches work, the evidence lags behind practice, and there is no specific evidence for people who attend frequently. We are finding that (and its early days with a very few people so not reliable yet) where the people who are attending regularly become an asset bringing their talents and gifts to the practice and community, their need for appointments reduces.

  • People with complex needs tipping into instability, but who are currently stable.

Here you need to convene the MDT, which will need clear agreements about integrated decision-making and a devolved budget. The job here is to prevent escalation.

Look after your infrequent attenders

All of these expensive 10 minute appointments are of course being paid for from a list that includes people that don’t attend at all or very much. Without them the practice can’t afford to provide services for those that need it. The per-head budget relies on the practice having a mixed list. So what happens when one of your infrequent attenders needs an appointment and there is a wait, or they don’t get through on the phone? Well they could take their business elsewhere. We encourage practices to look after their low attenders, partly to keep them as customers, and partly because prevention starts with these people. Most teenagers and young people don’t have any contact with primary care. Treating complex health needs starts with helping people to reduce their likelihood of developing them in the first place.

You can find out more about the London Primary Care Quality Academy here

Related Blogs Primary Care Reducing Demand Part one  and Part Two

And a good read here from The Health Foundation


Deeny, S., Gardner, T., Al-Zaidy, S., Barker, I., Steventon, A. (2017) Briefing: Reducing Hospital Admission by Improving Continuity of Care in General Practice. The Health Foundation, London.

Downham, N. (2018) Case History Mapping. London South Bank University.

Hull, S., Homer, K., Boomla, K., Robson, J., Ashworth, M. (2018). Population and patient factors affecting emergency department attendance in London: Retrospective cohort analysis of linked primary and secondary care records. British Journal of General Practice, 68.

Husk, K.(2017) Social prescribing offers huge potential but requires a nuanced evidence base. The BMJ Opinion. July 18

Morriss,R., Kai, J., Atha, C., Avery, A., Bayes, S., Franklin, M., George, T., James, M., Malins, S., McDonald, R., Patel, S., Stubley, M., Yang, M. (2012) BMC Family Practice 13:39.

Reiss Brennen, B., Brunisholz, K,D., Dredge, C., Briot, P.,Grazier, K., Wilcox, A., Savitz, L., James, B. (2016) Association of integrated team-based care with health care quality, utilization, and cost, JAMA, 316(8):826-834.

[1] People aged 62-82 had 6% fewer hospital admissions where they saw the same GP more consistently (same GP 2 more times out of every 10).

[2] This is primarily tacit knowledge