2009
DOI: 10.1377/hlthaff.28.6.1826
|View full text |Cite
|
Sign up to set email alerts
|

Adverse Selection In The Medicare Prescription Drug Program

Abstract: The Medicare Part D drug benefit created choices for beneficiaries among many prescription drug plans with varying levels of coverage. As a result, Medicare enrollees with high prescription drug costs have strong incentives to enroll in Part D, especially in plans with more comprehensive coverage. To measure this potential problem of "adverse selection," which could threaten plans' finances, we compared baseline characteristics among groups of beneficiaries with various drug coverage arrangements in 2006. We f… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
18
0

Year Published

2011
2011
2018
2018

Publication Types

Select...
8
1

Relationship

0
9

Authors

Journals

citations
Cited by 17 publications
(18 citation statements)
references
References 5 publications
0
18
0
Order By: Relevance
“…First, the fee-for-service service beneficiaries for which we have more complete prescription and encounter data may not be representative of the entire US older adult population. 27 Second, pharmacy claims data do not contain information about the condition the drug is intended to treat, so our matching of encounters and prescriptions may have measurement bias or be affected by trends in coding practice. 28,29 In addition, we were not able to classify approximately 18% of antibiotic claims as appropriate or inappropriate because they had no associated diagnoses, similar to a previous study 30 that found that 15% of Medicare antibiotic claims had no associated encounter.…”
Section: Discussionmentioning
confidence: 99%
“…First, the fee-for-service service beneficiaries for which we have more complete prescription and encounter data may not be representative of the entire US older adult population. 27 Second, pharmacy claims data do not contain information about the condition the drug is intended to treat, so our matching of encounters and prescriptions may have measurement bias or be affected by trends in coding practice. 28,29 In addition, we were not able to classify approximately 18% of antibiotic claims as appropriate or inappropriate because they had no associated diagnoses, similar to a previous study 30 that found that 15% of Medicare antibiotic claims had no associated encounter.…”
Section: Discussionmentioning
confidence: 99%
“…Because Part D plan choice is voluntary, it is possible that relatively healthy beneficiaries select into an MA-PD. 27 Propensity score matching partially accounts for this possibility by ensuring similar distributions in observable characteristics between the 2 groups being compared. 28,29 We calculated each beneficiary's propensity score of enrolling in an MA-PD plan based on logistic regression.…”
Section: Discussionmentioning
confidence: 99%
“…Lastly, for PCTs that require data on use of outpatient prescription drugs, another gap in Medicare data that needs to be considered is that only 60% of beneficiaries receive their drug coverage through Medicare Part D [57]. Part D enrollees tend to be poorer and less healthy than their counterparts with alternate drug coverage [58]. …”
Section: Using Administrative Data For More Efficient and Effectivmentioning
confidence: 99%