Transcript PowerPoint Slides

The Impact of Drug Benefit Caps
Geoffrey Joyce, PhD
Acknowledgements
 Collaborators:
 Dana Goldman
 Pinar Karaca-Mandic
 This research was funded by:
 National Institute on Aging
A6794c-2 6/06
Benefit Cap
• Annual limit on the plan’s contribution
•
In this case, $2,500 benefit cap
• Common in Medicare M+C plans
• Impact of caps on retirees < age 65 and 65+ in
2003-2004
A6794c-3 6/06
Imposing a Spending Cap Creates a
Fundamental Trade-off
Imposing a spending cap decreases the cost to
provide the prescription benefit
 Makes coverage available to more beneficiaries
A spending cap creates a coverage gap (or “donut
hole”) for beneficiaries
 Increases the risk that patients will reduce or
cease drug therapy
A6794c-4 6/06
As Set Up, Medicare Part D Raises Some
Issues
Stop-Loss $5,100
($3,600 in
out-of-pocket)
Catastrophic
Coverage
Catastrophic
Coverage
Insurer
Pays
Insurer Pays
95%
ofCosts
Costs
90% of
Beneficiary
Pays Next
$2,850 in Rx
Spending
5% Cost-Sharing
Above Stop-Loss
Beneficiary Paid
Insurer Paid
Initial
Coverage Limit
$2,250
75%
Paid by
50%
of Costs
Paid
Plan
by Insurer
($1,500)
($2,113)
25% Copay ($500)
$250 Deductible
2006
A6794c-5 6/06
Tseng et al (2004): Surveyed Beneficiaries to
Assess the Effects of Spending Caps
1,300 Medicare+Choice enrollees in one state in
2001:
 Group who exceeded their annual prescription
benefit cap of $750 or $1,200
 Matched controls who did not exceed their
annual cap of $2,000
Those exceeding the cap had resulting coverage
gaps of 75–180 days
A6794c-6 6/06
Beneficiaries Reported Using Several
Strategies When They Exceeded Caps
Switched
Drugs
15 (9)
Used Drugs
Less Often
18 (10)
Used Free
Samples
34 (27)
0
10
20
30
40
Percent of Beneficiaries Using Strategy
A6794c-7 6/06
Hsu et al (2006): Impact of $1,000 Cap on
Utilization, Costs, & Clinical Measures
 Compared clinical and economic outcomes in 2003
among Kaiser M+C members in capped vs. noncapped plans in 2002-2003 (age 65+)
 Employer-supplemental insurance – No cap
 Individual-purchased - $1,000 benefit cap
 About 13% reached the cap in 2003
 Those in capped plan:
 31% lower Rx costs
 No difference in total medical costs
A6794c-8 6/06
Hsu et al (2006)
 But had higher rates of
 ED visits (RR=1.09)
 Nonelective hospitalizations (RR=1.13)
 Mortality rate (1.22)
 Non-adherence (1.2-1.3)
 Capped members had higher odds (1.2 – 1.3)
 Elevated LDL
 Systolic blood pressure
 HbA1c
A6794c-9 6/06
Aims of This Study
 Examine Rx utilization and costs in more detail
 Behavior pre- and post-cap
 Timing of cap
 Stopping, switching, mail-order use, by class
 Do those who stop resume drug therapy in
subsequent year
 Impact on hospitalizations and ED visits
A6794c-10 6/06
Data & Methods
 We linked health care claims to health plan benefits
of 30 large employers (1997-2004)
 Over 50 health plans
 Nearly 8 million person-years
 Analyze 7 plans in 2003-2004 from large employer
 2 plans had an annual Rx benefit cap of $2,500
 Compare Rx and medical use
 Among groups within the same (capped) plan
 Among persons in capped vs. uncapped plans
A6794c-11 6/06
Distribution of Health Plan Spending
in Capped Plans (PPPY)
PPPY Spending by Health Plan
< $2,400
$2,401$2,499
>= $2,500
N
%
6,843
94.1
192
2.6
239
3.3
N
%
25,972
88.6
1,359
4.6
1,981
6.8
Plan 1
Plan 2
A6794c-12 6/06
Classify Members Into 3 Groups
 Group 0: Rx spending by the health plan <= $2,400
 Group 1: Rx spending by the health plan > $2,400
 But no subsequent Rx claims
 Group 2: Rx spending by the health plan > $2,400
 With subsequent Rx claims
A6794c-13 6/06
When Do Members Reach the Cap?
Percentile of Those Reaching the Cap
th
5th
10th
25th
50th
75th
90
Feb
April
June
Sept
Nov
Dec
A6794c-14 6/06
Monthly Rx Spending in Capped vs.
Non-capped Plans (>$2,400)
PMPM Rx Spending
700
600
500
400
300
200
100
0
1
2
3
4
5
6
7
8
9
10
11
12
Month in 2004
A6794c-15 6/06
Monthly Rx Use in Capped vs. Non-capped
Plans (>$2,400)
PMPM N30DE Scripts
12
10
8
6
4
2
0
1
2
3
4
5
6
7
8
9
10
11
12
Month in 2004
A6794c-16 6/06
Percent Switching Medications Post-Cap
(Among Those Reaching the Cap Before November)
Diabetes
Percent Switching Post-Cap
Cap
No Cap
3.3
5.8
Cardiac
9.6
8.1
Hypertension
7.2
6.3
Depression
4.2
6.1
Cholesterol
2.1
2.4
A6794c-17 6/06
Percent Stopping Medications Post-Cap
(Among Those Reaching the Cap Before November)
Diabetes
Percent Stopping Post-Cap
Cap
No Cap
4.9
3.3
Cardiac
5.2
5.4
Hypertension
7.0
6.6
Depression
16.4
8.7
Cholesterol
13.0
4.8
A6794c-18 6/06
Resumption of Medication Use
• Among those who stopped taking a class of
medications in capped plans
 Modest take-up in Q1 of 2004
 May be related to data problem in 2003
A6794c-19 6/06
Preliminary Conclusions
Imposing a spending cap:
 Reduces Rx use overall
 50% - 66% reductions in Nov-December
 Effects vary modestly by therapeutic class
 Increases the risk of adverse health outcomes
 Inconsistent evidence on medical use
A6794c-20 6/06