1. Value Based Insurance Design Michael Chernew Oct 10, 2008 Portions of this research were funded by Pfizer and GSK.

2. Issue Problem: Health care costs are high and rising fast –We cannot afford projected expenditure growth –We are not getting enough value for the money we spend Solution: –Supply side reforms payment reform, IT, etc. payment reform, IT, etc. –Demand side reforms Develop more sophisticated benefit packages

3. Saving money is not main objective How do we finance health? –The best way to save money is to provide no care –How much health can we afford? Who decides? –How do we meet that budget target Who pays for that health? What role do patient preferences play?

4. Value Based Insurance Design (1) Focus on the demand side –What we charge patients Basic premise: –Patient demand and preferences should pay a role in the care that is delivered –Patients should share some of the financing burden because it can encourage efficient care –Patients who opt for more conservative treatments should share in the savings –However, standard demand theory should not be blindly applied

5. Consumers do not respond to cost sharing as economists would like Reductions in appropriate use same as for inappropriate use (Sui et al. 1986) –Lack of coverage is associated with worse outcomes Effects concentrated on poor and chronically ill –Copays reduce use of preventive services –Copays reduce use of ‘valuable’ pharmaceuticals

6. Value Based Insurance Design (2) Reduce (or keep low) copays for high value services –For high value patients Different from: Value Based Purchasing –Focus on contracting with providers, payments and provider incentives Areas of overlap: incentives for beneficiaries to select high value plans or providers (tiering) Sources: Fendrick, et. al American Journal of Managed Care, 2001 Chernew. et al. Health Affairs. 2007; Chernew. et al. Health Affairs. 2008

8. Defining Value Value reflects costs AND quality –Similar to the concept of cost effectiveness Existing research identifies high value services and low value services –High value services are promoted in disease management, quality measurement and pay for performance programs. More health services research is needed –Comparative effectiveness –Systems research

10. Will VBID save money Better health reduces adverse events This may fully offset costs of extra utilization if targeted well enough. –Don’t bet on it generally –More hope in Medicare Generally not enough savings to offset costs of extra payment for services that would have been used anyway

11. How do we finance net costs Spread increased costs across all other services Charge more for targeted clinical services –Imaging? –Selected, high variance procedures? –Selected, low incremental value drugs? Better research will support better benefit design

12. Federal VBID Role Benefit design for public programs Health savings account rules Support research Support IT

13. Summary Increasing patient cost sharing saves money Consumers often do not respond as we would like Implication –Charge more in some situations (when you want to change behavior) This saves money –Lower financial barriers for high values services This mitigates adverse health effects Combine with smarter provider payment

14. END