Dynamics of Plan Choice and Prescription Drug Utilization in Medicare Part D. The Medicare Part D prescription drug benefit pioneered the use of multiple private insurance providers to deliver a public insurance product, and as such raises a host of important policy questions. Are consumers capable of making an informed choice across so many private insurance options? Once enrolled in a plan, do consumers respond appropriately to the non-linear schedule of prices they face throughout the year? The key methodological innovation in our work is to construct models where choices reveal information about consumers'preferences, even when those choices are not necessarily optimal. This builds on our earlier work, which uses an incomplete data set to document evidence of inconsistencies in elders'choice of Part D plans. We will replicate and extend this work using a complete 20% sample of Medicare Part D claims provided by CMS. In particular, we will assess whether consumers tend to improve over time as beneficiaries gain experience in the market or whether consumers tend to remain in sub-optimal plans, either because they attempt to choose a better plan and fail, or because they remain in the same plan through inertia. We will develop a structural model which allows us to decompose the demand and supply side factors which drive changes in plan choices over time. Individuals enrolled in Medicare Part D face a complicated dynamic consumption decision: the prices of drugs today depend on out of pocket payments to date. Forward-looking consumers should consider their entire expected path of spending throughout the year in deciding on drug consumption at each point during the year, while time-inconsistent consumers may not do so. We will build a dynamic model of drug consumption with respect to the nonlinear budget sets facing consumers that incorporates uncertainty as well as potential discounting inconsistencies. Changes in drug consumption around the variance in drug plan pricing schedules allow us to identify both whether choices are consistent over time and how individuals react to price uncertainty. We will also extend our analysis to consider the impacts of prescription drug utilization for downstream physician and hospital spending. Our empirical strategy will allow us to estimate both an overall elasticity of hospital and doctor spending with respect to prescription drug expenditures (and coinsurances), and more particular elasticity's for different types of prescription drugs. We can use the results to provide projections of how changes in the time path of prices under Part D (e.g. filling in the donut hole) will impact annual consumption patterns of both prescription drugs and other types of medical consumption.

Public Health Relevance

The Medicare Part D program pioneered the use of multiple private insurers to deliver a public insurance benefit, an approach which is the centerpiece of the Affordable Care Act. Our project will rigorously analyze the ability of individuals to make appropriate choices of plans within Part D, both upon initial enrollment and as they remain in the program. The results will be directly relevant to the design of Part D and will also apply to the design of exchanges and insurance options under health care reform.

Agency
National Institute of Health (NIH)
Institute
National Institute on Aging (NIA)
Type
Research Project (R01)
Project #
2R01AG031270-05
Application #
8372015
Study Section
Special Emphasis Panel (ZRG1-SSPS-H (09))
Program Officer
Baker, Colin S
Project Start
2008-03-15
Project End
2016-07-31
Budget Start
2012-08-01
Budget End
2013-07-31
Support Year
5
Fiscal Year
2012
Total Cost
$309,977
Indirect Cost
$104,977
Name
National Bureau of Economic Research
Department
Type
DUNS #
054552435
City
Cambridge
State
MA
Country
United States
Zip Code
02138