New care models explained: How the NHS can successfully integrate care

Multispecialty community providers (or MCPs) and the primary and acute care services (or PACS) are new types of delivery models, integrating hitherto relatively poorly coordinated care providers into new and efficient organisations or alliances capable of delivering on the “triple aim” – a better patient experience, better population health and more efficient use of resources.

While the two models share a common philosophy, they are also marked by some important differences.

The MCP is founded on the registered lists of large scale general practice, either large group practices or networks and federations of GP practices. These practices would swell to incorporate community staff and a range of consultant specialists and could also incorporate social care staff.

As a result, a full range traditional community based services would be coordinated by the MCP but in addition it is envisaged that outpatient and diagnostics would increasingly shift into the community under their control.

‘Should we feel confident that these models will deliver the benefits for which they have been designed?’

The PACs are a more full blooded vision of vertical integration – where a full range of hospital and general practice services are delivered by the same organisation.

PACs have, so far at least, been only lightly described, suggesting that variations are likely to emerge. Two particular variations have so far been highlighted in the forward view: hospitals opening their own GP surgeries and MCPs taking over the running of the local district general hospital.

• Lewis: Care integration is the goose that must lay the golden egg

Radically new approaches needed

The problems facing the NHS have come into sharp relief over the past few months, underlining the need for radically new approaches even with political commitments for future funding growth.

So should we feel confident that these models will deliver the benefits for which they have been designed?

First, the formal evidence on previous examples of care integration tells us that not all of the aims will be easy to achieve.

One important systematic review of integrated care initiatives (Powel Davies, 2006) suggests that only a small minority of attempts to reduce costs through integration succeeded. The EY/Rand evaluation of the national integrated care pilots in 2012 demonstrated modest reductions in costs but not where expected – unplanned hospital admissions did not fall as was intended and may even have increased.

‘NHS England must develop its own insights into how their new models might be developed and implemented’

So as to whether these new models will solve financial problems and raise quality levels, a leap of faith is required.

However, while the existing formal evidence may not make encouraging reading, it should be remembered that the evidence often relates to small scale initiatives, often delivered in circumstances that may be unsupportive.

In contrast, the proposed new models of care allow an opportunity to implement at a far greater scale and are likely to evolve in a far more enabling environment with, one would imagine, strong support from the centre.

Looking to the US, there is some encouraging, albeit early, evidence emerging on the implementation of accountable care organisations (or ACOs) contracts which, of course, have foreshadowed the debate in England.

ACOs are clearly relevant to our own experiments as they focus on provider integration, risk sharing and payments linked to outcomes, all likely characteristics of MCPs and PACS.

One study (McWilliams et al, 2013) found lower spending per person and some quality improvement as a result of early ACO commercial contracts suggesting that, in these circumstances at least, the incentives may deliver the desired outcomes.

NHS England must develop its own insights into how their own new models might be developed and implemented; variety has been deliberately built into the process with a promise of no “top down” enforcement.

Important questions

A number of important questions are raised.

How big should MCPs be – will small groups be more nimble than their larger cousins?

Will MCPs deliver results quickly enough or should we encourage greater hospital involvement early on though PACS?

Again, an analysis of both US independent and hospital based multispecialty group practices may provide some valuable insights into the two proposed NHS models.

Comparing independent physician groups (perhaps a reasonable proxy for MCPs) of different sizes, McWilliams and colleagues found that larger groups had lower spending per beneficiary and better quality than smaller groups when operating in a financial risk sharing environment (ie when paid by capitation rather than fee for service).

‘Will MCPs deliver results quickly enough or should we encourage greater involvement early on though PACS?’

Importantly, among independent physician groups of all sizes (and above the threshold set for participation in ACOs) those that had a stronger primary care orientation delivered lower spending, fewer readmissions and better quality of diabetes care.

Interestingly, larger hospital based physician organisations fared less well in this evaluation, resulting in higher spending per person, higher 30 day readmission rates and similar performance in terms of quality – although these outcomes could be offset by a strong primary care orientation within the physician group.

Of course, caution must be exercised in the interpretation of international evidence, rooted in very different contexts - for example, US specialists are largely paid on a fee for service basis unlike their UK counterparts.

Nevertheless, this evidence may at least hint that the new models of care should be founded on larger scale groups and populations, and that those vanguards that do not comprise of a generous proportion of GPs are likely to fare less well.  

It would be tempting to see the PACS and MCPs are in some form of competition - primary care led versus hospital led integration - or that PACS will result from a “takeover” of the local hospital by GPs.

A more effective strategy is that of collaboration between hospitals and primary care leaders, with joint ventures emerging rather than “winner takes all”.

Primary care understands population health management and hospitals have the size and capability to manage financial as well as clinical risk. This would seem to be a partnership well worth making.

Richard Lewis is partner at EY and a health consulting leader