Payment mechanisms for ICSs: What should the NHS be considering now, next and in future?

Now: Block payment with rewards/ penalties to address waiting times

The current 2023-25 NHS Payment Scheme has another year to run at the time of publication (with some minor midway adjustments). In the immediate run up to and in the wake of the 2025 general election, increasing the volume of secondary elective activity to reduce waiting times and waiting lists for elective acute care is likely to remain a political priority. Additionally, the drive for greater technical productivity in acute services, where most NHS money is still spent, will likely be high on the agenda of the next government. 

The political reality is that any dramatic move from both payment by activity or financial incentives for elective care is unlikely in the short term. We must still recognise that, while intended to reduce elective waits, the present payment scheme is a counter-vailing force against the ICS mission for a left-shift to more preventative care, as it incentivises generally more expensive secondary care activity and undermines allocative efficiency. Within this context, ICSs (both ICBs and provider partners) may seek to innovate using the flexibility within the current payment scheme, although there is no perfect ‘off the shelf’ model ready to be put into practice. 

In the immediate term, local ICSs may want to consider adopting and/or adapting the ‘West Yorkshire’ model, subject to further evaluation (see case study page 25). This model includes a combination of block payment, enabling service change, and fee for service, to reduce waiting times across the provider collaborative. Waiting time targets are agreed between the commissioner and providers, who face a £2,000 penalty for every 52-week waiter above their target number at the end of March 2024. Conversely, trusts that beat their target will receive a £2,000 bonus per patient, and there are no caps for under or over-performance. This can help address long waits and the backlog of elective care without losing the benefits of collaboration and clinical innovation that a level of fixed income had enabled. It also provides a reward for higher value interventions that prevent people from become long waiters, although not necessarily across multiple care settings. In particular, the block payment element of the contract has also enabled innovation in service model delivery and greater productivity that was disincentivised under PbR arrangements. 

We present this option as an attempt at a ‘quick fix’ to address the elective care waiting lists while curbing some of the disadvantages of payment by activity, without an evaluation having yet been conducted. However, to support the leftward/upstream shift towards more preventative care and to incentivise greater integration of services, we believe other options are more desirable in the longer term. We would propose ICS explore opportunities for local experimentation, including approaches such as the West Yorkshire model. Wider use of and learning locally agreed payment mechanisms should be encouraged nationally.
 

 

Next: Outcomes-based payments for specific pathways

Looking beyond short-term political priorities, to deliver on their first core purpose of improving population health and outcomes, ICSs should focus on improving and measuring improvements in specific health outcomes, not just output such as the volume of activity. Shifting focus to ends rather than means could enable greater innovation within systems, driving improvements in technical efficiency. Payments based on outcomes can incentivise both providers and payers to deploy early-stage and more cost-effective interventions, boosting allocative efficiency, provided there is an appropriate ‘gain-share’ mechanism to ensure all system partners see a financial benefit from this. 

This model would involve introducing an outcome-based fee for specific therapy areas, paid in exchange for delivering against agreed specific outcomes for a specified population, rewardingand incentivising preventative interventions which align with care pathways. This would commission services around pathways, not providers, and stretch across primary, community and secondary care providers, encouraging them to work together towards shared goals and incentivising integration. 

To ensure this approach provides a ‘gain-share’ to different providers, it may be beneficial for an ICB to commission a lead provider to sub-commission further interventions – either upstream preventative admissions, if an acute provider, or later secondary care, if a primary or community provider. This would replicate the savings incentive seen in the American ChenMed and German OptiMedis case studies (see above). However, negotiating payments between providers can be a long and cumbersome process. 

Learning from QOF (see payments for quality/outcome, above) this model should therefore focus on measurement and payment for a small number of defined and agreed patient outcomes for a specific population, with payments made at as large a scale as is practical. Providers and commissioners would need to agree the payment for each outcome and ensure there is appropriate outcomes data available on which payments can be informed. Payments could be higher for better outcomes at an earlier stage in pathways, incentivising providers to prevent worsening ill health. 

As with QOF, there will be an administrative burden of implementation, from monitoring outcomes data to making associated payments, although this should be minimal by comparison to the potential benefits. However, the potential to make significant progress on their integrated care strategies will be a powerful incentive for ICSs. The complexity of creating and implementing and outcomes-based fee model will likely be a significant practical challenge. 

A similar approach has been deployed in North Central London ICS, where UCLH is a lead provider for musculoskeletal (MSK) services across the London Borough of Camden, that is at place-level within the system footprint (please see case study below). This approach could also be taken for diabetes care, making payments based on NHS RightCare’s seven outcome targets in its model pathway. [104] We have heard most interest from ICS leaders in taking such an approach to care pathways for frail and elderly patients to help reduce secondary care admissions. 

While place may be an appropriate level for commissioning such an approach and encouraging integration, the learning from the Camden MSK model is that the initial set-up time for the current model was disproportionate to the expected benefit, although this may be mitigated as best practice models are established and can be more easily replicated and rolled out. If upscaled across several different therapy areas, with different ICSs agreeing different outcomes and payments within their footprints and existing payment mechanisms remaining in place elsewhere, this approach does risk adding complexity and a creating a fragmented payment system, which could add cost and dilute incentives. It will also be important for ICSs to be able to exit the costs associated with ‘old’ care pathways/models of care. The time taken to undertake consultation and extensive scrutiny can render beneficial changes uneconomic due to the burdensome regulatory processes, but can reap rewards when agreed. 

While this approach has clear benefits and may be preferable to an outcome-based payment for community care approach, further work is required to develop the proposal in sufficient detail for it to be implemented. In 2024/25, the NHS Confederation will seek to develop an outcomes-based and lead provider pilot programme, in collaboration with local leaders, which could be trialled across willing ICSs in either 2024/25 or 2025/26 to inform longer-term NHS payment policy. This will set out in further detailed proposals for such a model, including metrics for outcomes, how much is paid for delivering those metrics, the balance between fixed and variable costs and the potential benefits, risks and mitigations, with a particular focus on admissions avoidance in elderly and frailty care pathways.
 

 

In future: Capitated risk-based payments

In the longer term, we should look further ahead of these two models for a more comprehensive payment system to underpin ICS developments. Looking abroad, the OptiMedis approach in Germany, ChenMed in the USA and Kaiser Permante in the US point to a way forward, achieving better health outcomes for less money (see case studies below). The crucial learning from these systems is not how money is raised – the debate between public and private insurance funding – but how it is spent through different payment mechanisms and the incentives these create. Elements of this approach have already been tested in the UK, with the deployment of capitated outcome-based and incentivised contracts (COBICs) in Milton Keynes and elsewhere (see case study below). 

This model would involve the introduction of multi-year, capitated payments across systems, with shared savings contracts between ICBs and multiple providers, as a more radical, longer-term approach. These would be risk weighted, reflecting higher healthcare need and likely expense from factors such as age, deprivation and rurality (geographic dispersal) of the population. This approach would seek to incentivise earlier-stage interventions and integrated working by emulating the international best practice approaches, tailored to an ICS environment. 

Significant further work is required to consider the different ways such an approach could be applied more broadly to the English context. One such way may be for ICBs and vertical provider collaboratives to agree expected costs for providing secondary care – any differences with actual healthcare costs (for instance, from using non-secondary care interventions) would be considered savings and split between providers and ICBs through a shared savings contract. Minimum quality standards (aligning with national clinical standards) would be set, with deductions in payments for activity falling below the standard. This could include elements of outcome-based payment, building on the second model. Like the previous model, this would allow providers to retain savings from allocative efficiency, although allowing commissioners to keep some of the revenue. It would also feature some payment relating to outcomes. Compared to GP capitation payments, it would need to cover multiple care settings and ensure gain-share between providers and commissioners, while avoiding the complexity of payments currently associated with the mix and multitude of capitation, QOF and IIF payments for general practice. 

This approach would require significant improvements in how we collate, integrate and monitor data. It would require data on expected and actual costs to be calculated and observed so appropriate shared savings contracts between providers and commissioners, and data on standards to be collected and monitored, with an appropriate discount for missing standards to be agreed. Multi-year contracts would enable systems to invest in more services with a return on investment that spans more than one financial year, but by its nature it would take longer for this approach to help achieve financial balance, a particular challenge when system financial plans are under pressure. Implementing this approach across a wide range of therapy areas would require more time to agree appropriate outcomes metrics and the tools and technology to monitor activity and outcomes. Additionally, determining and allocating gain-share is complex.

We propose that further policy development is undertaken to consider how a capitated risk-based payments could be applied to the English context and ICSs, with a view to local experimentation in 2025 and into the future.