Will pharma risk-sharing work better when the patient is the customer?

A Health Affairs article by Peter Neumann et al. (Risk-Sharing Arrangements That Link Payment For Drugs To Health Outcomes Are Proving Hard To Implement) catalogs a set of experiments --mostly in Europe-- where pharmaceutical companies put a portion of their payment at risk contingent on certain outcomes being realized. As the article indicates, these experiments aren't going all that well. The main problems are high transaction costs, the lack of acceptable outcomes measures, difficulty of determining treatment effects and an absence of suitable data capture systems.An example of a difficult program to implement is the UK National Health Services' agreement with makers of beta interferons for multiple sclerosis. The NHS was skeptical of the products' long-term efficacy, so the program calls for price adjustments if the results are 20 percent better or worse than expected over a 10-year period. The long lifecycle, difficult administration and low adherence to the medications makes this one a real nightmare and I can't imagine it's worth the trouble.Meanwhile, a more successful program is an agreement between the NHS and Novartis to limit the number of Lucentis doses that the NHS must pay for to 14. If the patient still needs more injections --based on a measure of visual acuity-- Novartis picks up the tab.As consumers in the US take on an increasing share of medical costs, I expect that drug companies will try various risk sharing arrangements directly with patients, probably administered through PBMs or health plans. The Lucentis example is a good one. Patients will probably be willing to pay a fixed amount to receive all the Lucentis they need --it may be more or less than 14 doses. Certain other drugs, like fertility medications, could also be offered on a risk share basis --although this may require someone to bundle a variety or products to make it work.I can also see the value of long-term contracts for maintenance medications for chronic conditions, e.g., for high cholesterol or high blood pressure. It could actually help patients adhere to their regimens if they had --for example-- a three-year Lipitor contract. They might feel badly enough about not taking a drug they'd already paid for that it would guilt them into keeping up their therapy.