[A] growing body of literature  calls for a renewed focus on  nonmedical determinants of health. Unfortunately, interventions that address them— home visits, education, service-enriched housing, workforce training, healthy eating, exercise, and other nonmedical activities—often exist outside the scope of the traditional health care payment system.
This is largely the result of two related challenges: The existing payment system is designed to pay for treatment after illness has occurred, and the funding necessary to pay for proven nonmedical interventions is tied up remediating the illnesses that the interventions could have prevented. […] Fortunately, there is a new tool—pay-for-success—that could be used to increase spending on the nonmedical determinants of health while maintaining society’s commitment to treating people who are already sick.
All pay-for-success projects, which are also known as social impact bonds, or SIBs, begin with a performance-based contract between a service provider, usually a nonprofit organization, and a payer, usually a government agency. The service provider agrees to administer a program designed to produce a future outcome that is valuable to the payer—which, in turn, commits to pay the service provider when that outcome is delivered.
Once the performance-based contract is in place, the service provider raises money from foundations, banks, and other investors that agree to supply the provider with up-front program funding. In exchange, the investor or investors receive “success payments” in the future based on projected cost savings (usually by a government agency), should the agreed-upon outcome be produced on schedule. […]
Once the outcome has been verified, the payer’s success payment is  passed along to the investor or investors. Depending on the contract, the service provider may also receive a bonus for executing the program successfully. However, if the outcomes are not present, no success payment is disbursed, the investors lose their investment, and the service provider loses credibility as an evidence-based organization.
There have been four such pay-for-success projects. Others are in the works and many more are imaginable. Ian Galloway, quoted above, has more in the latest Health Affairs.