Faster regulatory approval has not necessarily achieved faster patient access. A new OHE paper sets out how risk-sharing can tackle this, enabling resolution of differences of view between innovators and payers about a technology’s value. The authors find risk-sharing makes sense in 9 out of 16 scenarios, setting out the conditions in which both parties should use them. Innovators need to fund the research in order to align incentives.

There has been increasing emphasis by regulators on approving potentially important treatments rapidly, notably through accelerated access schemes.

As a consequence, there is a challenge to payers. Products are launched with less evidence, creating greater uncertainty as to their relative effectiveness and value for money, the appropriate price to pay and the best use to be made of the drug.  If payers refuse to reimburse new treatments on the grounds of lack of good evidence of incremental effectiveness, there is likely to be political challenge from patient groups and from the innovating pharmaceutical companies. Faster regulatory approval processes are not necessarily achieving faster patient access.

This paper sets out a way forward using conditional reimbursement schemes with risk-sharing. These schemes have costs which, if allocated efficiently, will optimise incentives for uncertainty reduction. One important, but often overlooked, reason for introducing risk-sharing is to resolve differences of opinion between innovators and payers about the value of a technology to the health system. To date there has been no formal attempt to set out the circumstances in which risk sharing can increase the value of the options available to payers and innovators for mutual benefit when perceptions of value-for-money and of the value of undertaking additional research differ between the parties. In this paper the authors explore how a value of information (VOI) framework can help to understand what a performance-based risk-sharing arrangement (PBRSA) can, in principle, add to a reimbursement scheme. They set out the conditions in which both parties should seek such an arrangement.

Risk-sharing allows payers to overcome two important concerns they may have about conditional approval, coverage with evidence development type, schemes:

  • First, it allows prospective (and retrospective) price adjustment which could have the aim of maintaining a constant positive value-for-money, based on available evidence, against the relevant cost-effectiveness threshold or value rule;
  • Second, it reduces the likelihood of early reversal of a decision to adopt.  If there are substantial irrecoverable costs arising from implementing the technology, the payer may need the guarantee of compensation – an insurance policy from the company that irrecoverable costs will be reimbursed.

The authors assume that the innovator will undertake the research or fund the costs of evidence collection in any PBRSA. Thus, the innovator’s commercial calculations, around the price and volume consequences of investment in evidence collection, matter. If the innovator expects to receive a value-based price that reflects the value of the health gain to the payer, then its evidence generation incentives can be aligned with the requirements of payers across jurisdictions.

Payers and innovators may agree, or differ, over one or both of (i) whether the product is expected to provide value for money and (ii) the costs and benefits of conducting further research to reduce uncertainty. This gives 4 options for each party. Mapping this out for the two parties gives a total of 16 combinations.

The analysis shows that risk-sharing provides the optimal solution in 9 out of these 16 combinations.  However, the literature shows very clearly that, although risk sharing is feasible, transaction costs are a barrier to implementation. Payer experience with outcome-based schemes has been mixed. Payers and HTA bodies are put off risk-sharing schemes by the resources involved, in both negotiating and monitoring the schemes, and in the collection and analysis of the data. Innovators are put off by the cost and effort involved in evidence collection and in administering the agreement.  However, there are also issues associated with the alternatives which for payers involve either lower prices or innovators providing additional evidence and for innovators involve the payer adopting the technology, on the basis of the price and evidence offered. Hence, we end up with the case for risk-sharing schemes, albeit with the proviso to take account of transaction costs not included in the assessment of the value of further research.

The key is to align incentives correctly. Implementing risk-sharing does impose costs. If the innovator is getting a value-based price, this price needs to be adjusted for the costs incurred by the payer in implementing a risk-sharing scheme. Although the authors refer to 9 out of 16 combinations, in practice they expect use of PBRSAs may be less frequent than this ratio implies, due to transaction costs and because innovators may choose to find ways to improve evidence generation prior to launch. Regulators, HTA bodies and payers can help with this, for example by seeking greater alignment between regulators and HTA bodies on evidence requirements.

In conclusion, there are steps that can be undertaken to make risk-sharing more practical, both in terms of ensuring that HTA bodies and payers consider risk-sharing as an option and of ensuring that the costs to the health system of implementing risk-sharing fall upon the innovator in an efficient way. The end result should be earlier access to cost-effective treatments for patients, with the innovator getting revenues, and the payer having confidence that it is not wasting money.

Please cite this report as

Towse, A. and Fenwick, E., 2021.  It takes two to tango: when do conditional reimbursement risk-sharing schemes work for both parties? Setting out the conditions in which risk sharing schemes improve value for money. OHE Research Paper, London: Office of Health Economics. Available at: improve value formoney


Related Research

Cole, A., Cubi-Molla, P., Elliott, R., Feast, A., Hocking, L., Lorgelly, P., Payne, K., Peek, N., Sim, D., Sussex, J., Zhang, K and Steuten, L., 2021. Making Outcome-Based Payment a Reality in the NHS. Phase 2: Practical Considerations. OHE, RAND Europe, UCL, University of Manchester and Cancer Research UK Research Paper. Available at

Towse A., and Fenwick E. 2019. Uncertainty and Cures: Discontinuation, Irreversibility, and Outcomes-Based Payments: What Is Different About a One-Off Treatment? Value in Health. 2019; 22(6):677–683.

Garrison LP Jr, Towse A, Briggs A, de Pouvourville G, Grueger J, Mohr PE, et al. 2013. Performance-based risk-sharing arrangements-good practices for design, implementation, and evaluation: report of the ISPOR good practices for performance-based risk-sharing arrangements task force. Value Health;16(5):703–19.