Pharmaceutical companies are now entering a phase where they are launching curative treatments, in addition to drugs which only manage symptoms and disease progression (e.g. statins for cholesterol control). Recent examples include Solvaldi for the treatment of hepatitis C (with a cure rate of >90%) and Glybera for the treatment of lipoprotein lipase deficiency (which was approved in 2012 and has a cost of €1,000,000 per patient). The question for payers is how do we pay for these new curative treatments?
The on-going heated discussions regarding the price of Solvaldi are not related to the effectiveness of the product, but rather towards its up-front budget impact. A three-month course of Solvadi costs $84,000 in the US and ~€41,000 in the EU, with payers across the EU and US struggling to find a suitable approach to pay for such curative treatments as a result of these high up-front costs.
The historic payer approach to managing such high budget impact treatments has been to impose substantial discounts or budget caps, as observed in France where the authorities secured ~25% discount on Sovaldi to enable continued patient access in France. In the UK, NICE considered Solvaldi to be cost effective yet patient access has been delayed due to the NHS needing to develop an infrastructure to support the demand for Solvaldi. As a result, a specific Hepatitis-C fund has been created, similar to the CDF, to fund Solvaldi and other Hepatitis C treatments. The clinical outcome uncertainty is not the main issue with Solvaldi or other curative treatments, it is short-term affordability that poses a problem for payers. How do payers tackle the high upfront costs of a cure that is cost-effective?
A number of potential solutions have been proposed:
- Pay for performance
Where payers (either government or health insurers) would pay pharma a per patient periodic fee for as long as the treatment works. This could result in payments lasting for several years or even a lifetime. At its simplest, this could be achieved with regular health checks to ensure the treatment is still effective.
The main challenge with this approach is for those countries who have multiple health insurers – how do you ensure that the money follows the patient? If a patient switches insurers, the new insurer must agree to continue with the periodic payments, even if the patient is no longer taking the drug treatment (as they have been cured).
Also, how do you measure that the treatment is still effective? For treatments of viral infections, such as Hepatitis C this is relatively straight forward as you can measure the viral load. The question remains, how would this work for other diseases? A clear marker needs to be identified to demonstrate that the treatment is still effective. Finally, what happens to the payments if the patient dies of something else, do the payments stop? - Amortisation
A payment model to spread the high cost of treatment over a defined period of time. For example, the time over which the benefits are realised. This can be considered as similar to leasing whereby, a third party covers the upfront costs and the healthcare payer, or patient, leases the treatment (plus interest) over a pre-determined period of time.
The main challenges here are that the healthcare financial systems cannot currently account for this payment method and it would take considerable effort to change healthcare financial systems to properly account for this payment model. - Licensing model
Similar to the approach used by the IT industry. Here healthcare systems would pay a fixed monthly licensing fee per patient for use of the intellectual property (IP) of the medicine, irrespective of the amount of medicine used.
This approach would make it possible to set the license fee by indication, based on the value the treatment offers for each indication and to change the license fee dependent upon the GDP of the country. The main disadvantages of this approach are that healthcare systems not set up for this approach today and that there is a risk of fraudulent use of medicines to treat additional patients without cost, each patient will have to be closely tracked to ensure medication use can be captured.
Similar to the previous models, the licence agreement would have to have clear criteria highlighting when the license expires e.g. when the treatment is no longer effective, or if a patient dies from something else. - Disease-related group (DRG) pricing
This system is already used by hospitals to costing and paying for non-pharmaceutical healthcare interventions. The system could be expanded to include costs for pharmaceuticals, whereby price is set by indication relative to the value the medicine delivers for that indication.
This will enable payers a greater degree of certainty over their budget impact (as the same fee will be applied to each patient irrespective of the amount of drug utilised by that patient). It also has the added benefit that the majority of healthcare systems currently have, or are adopting, a DRG approach.
The main disadvantages are that there is a risk of fraudulent use of medicines and that it takes considerable time for a new DRG to be created. The other major challenge is that this approach will not, necessarily, remove the initial high budget impact costs for novel curative treatments such as Solvaldi, as payments are likely to be linked to drug usage per indication rather than the length of treatment outcomes.
In the near term, it is likely that payers will continue to use the blunt instruments of budget caps and price cuts to manage their healthcare budgets, as changing the pricing mechanisms to one where a pack or vial does not have intrinsic value would be virtually impossible for current healthcare systems to manage.
However, these new pricing mechanisms can provide benefits to both the healthcare payers and to Pharma companies. To further develop these novel payment solutions pilot schemes will be required to enable the practical challenges to be addressed. The question remains, however, as to who will take the initiative of driving these new drug payment models forward?