Original Investigation | Health Policy
Economic Burden Associated With Extended-Release vs Immediate-Release Drug
Formulations Among Medicare Part D and Medicaid Beneficiaries
Andrew Sumarsono, MD; Nathan Sumarsono, BS; Sandeep R. Das, MD, MPH, MBA; Muthiah Vaduganathan, MD, MPH; Deepak Agrawal, MD; Ambarish Pandey, MD, MSCS
Abstract
IMPORTANCE The United States spends more money on medications than any other country. Most
extended-release drugs have not consistently shown therapeutic or adherence superiority, and
switching these medications to less expensive, generic, immediate-release formulations may offer an
opportunity to reduce health care spending.
OBJECTIVE To evaluate Medicare Part D and Medicaid spending on extended-release drug
Key Points
Question How much do Medicare and
Medicaid spend on extended-release
drug formulations, and what would be
the potential savings associated with
switching to immediate-release
formulations?
formulations and the potential savings associated with switching to generic immediate-release
Findings In this cross-sectional study of
formulations.
20 extended-release drugs, Medicare
Part D and Medicaid spent a combined
DESIGN, SETTING, AND PARTICIPANTS This cross-sectional study used the 2012 to 2017 Medicare
$3.1 billion in 2017. Switching to
Part D Drug Event and Medicaid Spending and Utilization data sets to analyze 20 extended-release
immediate-release drug formulations
drugs with 37 Medicare formulations and 36 Medicaid formulations. Only cardiovascular, diabetes,
was associated with an estimated $2.6
neurologic, and psychiatric extended-release drugs saving at most 1 additional daily dose compared
billion reduction in spending in 2017 and
with their immediate-release counterparts were included. Extended-release drugs with therapeutic
a $13.7 billion reduction from 2012
superiority were excluded. Analyses were conducted from January to December 2019.
to 2017.
Meaning The findings suggest that
MAIN OUTCOMES AND MEASURES Estimated Medicare Part D and Medicaid savings from
switching extended-release to immediate-release drug formulations between 2012 and 2017.
substitution of therapeutically
equivalent extended-release drug
formulations with immediate-release
RESULTS Of the 6252 drugs screened for eligibility from the 2017 Medicaid Drug Utilization
database and the 2017 Medicare Part D database, 67 drugs with extended-release formulations that
were identified in the Medicare data set (20 distinct drugs with 37 formulations [19 brand, 18
formulations represents a possible
option to reduce Medicare and Medicaid
spending.
generic]) were included in the analysis. In 2017, Medicare Part D spent $2.2 billion and Medicaid spent
$952 million (a combined $3.1 billion) on 20 extended-release drugs. Between 2012 and 2017,
Medicare Part D and Medicaid spent $12 billion and $5.9 billion, respectively, on extended-release
formulations. Switching from brand-name to generic extended-release formulations was estimated
to be associated with a $247 million reduction in Medicare spending and $299 million reduction in
+ Supplemental content
Author affiliations and article information are
listed at the end of this article.
Medicaid spending in 2017, whereas switching all brand-name and generic extended-release
formulations to immediate-release formulations in both Medicare and Medicaid was estimated to
reduce spending by $2.6 billion ($1.8 billion for Medicare and $836 million for Medicaid) in 2017.
During the study period, the estimated spending reduction associated with switching all patients
receiving extended-release formulations (brand name extended-release and generic
extended-release) to generic immediate-release formulations was $13.7 billion ($8.5 billion from
Medicare and $5.2 billion from Medicaid).
CONCLUSIONS AND RELEVANCE The findings suggest that switching from extended-release drug
formulations to therapeutically equivalent immediate-release formulations when available
represents a potential option to reduce Medicare and Medicaid spending.
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181
Open Access. This is an open access article distributed under the terms of the CC-BY License.
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181 (Reprinted)
Downloaded From: https://jamanetwork.com/ on 03/30/2020
February 28, 2020
1/10
JAMA Network Open | Health Policy
Extended-Release vs Immediate-Release Drugs For Medicare Part D and Medicaid Recipients
Introduction
Medication costs compose up to 20% of US health care expenditures and are up to 3-fold higher per
capita than in other high-income countries.1 Previous studies have evaluated strategies such as
switching brand-name drugs to their generic counterparts in order to lower costs.2 However, the
economic consequences of switching from extended-release (ER) drug formulations to generic
(therapeutically comparable) immediate-release (IR) drug formulations have not been studied, to our
knowledge. Although ER drug formulations have been traditionally used to lower pill burden and, in
theory, to achieve greater adherence to therapies, data on their overall effectiveness in improving
adherence compared with IR formulations are mixed and vary by drug type. Examples of this include
equal adherence between carvedilol ER and carvedilol IR formulations but improved adherence with
methylphenidate ER vs methylphenidate IR formulations.3,4 A Cochrane review5 that evaluated drug
administration schedule and medication adherence found inadequate evidence to recommend a less
frequent administration schedule but noted the possibility of improved adherence if the regimen
was reduced by 2 or more daily doses. The prospect of significantly increasing the number of daily
administrations to reduce cost may be impractical for many patients. Accordingly, we quantified the
cost of ER formulations saving at most 1 additional daily dose compared with their IR counterparts
and estimated the cost savings from switching to therapeutically comparable generic IR formulations
among Medicare and Medicaid beneficiaries.
Methods
For this cross-sectional study, the primary analysis was performed using the publicly available 20122017 Medicare Part D Prescription Drug Event data set,6 which contains medication expenditures for
approximately 70% of all Medicare beneficiaries, and the Medicaid Spending and Utilization Data
set,7 which contains national-level drug utilization data for outpatient drugs paid for by state
Medicaid agencies. Because all data were deidentified and publicly available, The University of Texas
Southwestern Human Research Protection Program determined this project did not meet criteria
for human subjects’ research and did not require institutional review board approval. This study
followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting
guideline.
The Medicare data set contains 2883 formulations, whereas the Medicaid data set contains
3369 formulations (Figure 1). Drugs with ER formulations in the data sets were identified
independently by 2 of us (A.S and N.S.). The ER formulations with disease-specific indications or wellestablished therapeutic or adverse effect superiority compared with generic IR formulations were
excluded (niacin, nifedipine, metoprolol succinate, venlafaxine hydrochloride, and tolterodine). Only
drugs with cardiovascular, diabetes, neurologic, and psychiatric indications were included. The study
included only oral medications with ER and generic IR formulations with more than 200 claims in
2017 (Figure 1). Only drugs with ER formulations saving at most 1 additional daily dose compared with
their generic IR counterparts were included. We extracted data on drug name, number of claims,
number of units dispensed, and total expenditure incurred (overall and per claim) for the ER and
generic IR formulations. The Medicare data set–reported spending included both beneficiary and
government contributions, whereas the Medicaid data set reported both federal and state
reimbursement to pharmacies. These data include only total costs to both Medicare and Medicaid
and do not account for out-of-pocket costs for the patient. To conservatively estimate the cost, we
assumed that all patients who switched to generic IR formulations would use the maximum number
of daily administrations recommended. We calculated the savings with the following formula:
estimated savings = (daily ER drug dose per unit price − daily IR drug dose per unit price) × total ER
drug units dispensed.
Temporal trends in annual expenditures on all drugs with available ER formulations and the
estimated expenditure substituting generic IR formulations were also assessed between 2012 and
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181 (Reprinted)
Downloaded From: https://jamanetwork.com/ on 03/30/2020
February 28, 2020
2/10
JAMA Network Open | Health Policy
Extended-Release vs Immediate-Release Drugs For Medicare Part D and Medicaid Recipients
2017. Because 100% interchange between ER and generic IR formulations may not be possible, we
performed a sensitivity analysis to estimate overall cost reduction if 25%, 50%, or 75% of all ER
formulations were substituted with generic IR formulations. We additionally performed a second
sensitivity analysis to evaluate whether using a per-claim–level analysis would yield different results
than a per-unit–level analysis.
We estimated the Medicare postrebate spending using the mean 26.3% for all brand-name
cardiovascular drug classes and the mean 13.0% for all brand-name central nervous system drug
classes noted from the 2014 Medicare Part D Rebate Summary.8 Similarly, using the Medicaid Drug
Rebate Program, we estimated a 23.1% per-unit rebate for all brand-name formulations and 13.0%
per-unit rebate for all generic formulations.9
Statistical Analysis
All dollar amounts are described in 2017 US dollars and have been adjusted for inflation. All analyses
were performed in Microsoft Excel, version 16.0 (Microsoft) and Graphpad Prism 7.0 software
(Graphpad).
Results
Medicare
Of the 6252 drugs screened for eligibility from the 2017 Medicaid Drug Utilization database and the
2017 Medicare Part D database, 67 with ER formulations that were identified in the Medicare data set
(20 distinct drugs with 37 formulations [19 brand, 18 generic]) were included in the analysis (Table 1).
In 2017, Medicare spent $2.2 billion ($1.9 billion after correction for brand-name drug rebates) on ER
formulations ($1347 million on brand-name and $806 million on generic formulations) for 1.5 billion
pills at a mean (SD) cost of $1.47 ($2.21) per pill ($13.75 [$7.21] per pill for brand name and $0.59
Figure 1. Flow Diagram of Medicare and Medicaid Drug Selection
6252 Drugs screened for eligibility
3369 From 2017 Medicaid Drug Utilization database
2883 From 2017 Medicare Part D database
6118 Excluded because no oral extended-release brand
3302 Medicaid
2816 Medicare
134 Oral extended-release drugs
67 Medicaid
67 Medicare
94 Excluded
47 Medicaid
23 Immediate-release drugs with >1 extra daily dose
18 No cardiac, diabetes, or neurologic indications
5 Nonequivalent medications for efficacy or safety
(metoprolol succinate, nifedipine, niacin,
tolterodine, venlafaxine)
1 Fewer than 200 claims (lovastatin)
47 Medicare
23 Immediate-release drugs with >1 extra daily dose
18 No cardiac, diabetes, or neurologic indications
5 Nonequivalent medications for efficacy or safety
(metoprolol succinate, nifedipine, niacin,
tolterodine, venlafaxine)
1 Fewer than 200 claims (lovastatin)
40 Extended-release drugs
20 Medicaid with 37 formulations
20 Medicare with 36 formulations
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181 (Reprinted)
Downloaded From: https://jamanetwork.com/ on 03/30/2020
Medicaid-only formulations: Dexedrine
(dextroamphetamine sulfate), Fortamet (metformin
hydrochloride); Medicare-only formulations: Lescol XL
(fluvastatin), Rhythmol SR (propafenone
hydrochloride), Zenzedi (dextroamphetamine sulfate).
February 28, 2020
3/10
JAMA Network Open | Health Policy
Extended-Release vs Immediate-Release Drugs For Medicare Part D and Medicaid Recipients
[$1.23] per pill for generic formulations). Namenda XR (memantine hydrochloride) had highest
overall spending ($891 million for 2.5 million claims and 69 million pills), and brand-name ER
metformin hydrochloride (Glumetza) had the highest per claim spending ($7619.35 per claim or
$74.83 per pill) for Medicare in 2017. The Medicare expenditure for generic IR formulations was $1.2
billion at a mean cost of $14.34 ($15.82) per claim and $0.16 ($0.21) per pill in 2017.
In 2017, the estimated spending reduction for switching from brand-name ER to generic ER
formulations was $247 million ($183 million with rebates). In contrast, switching all (brand and
generic) ER formulations to generic IR formulations in 2017 was associated with a $1.8 billion
Table 1. Per-unit Spending and Potential Savings Associated With Switching ER Drug Formulations to Generic IR Drug Formulations in Medicare Part D, 2017
ER Formulation
Pills Dispensed,
No. in Thousands
Spending,
$US in
Millions
Generic Name
Daily Dose
Frequency,
Pills/d
Mean
Spending
per Pill, $
Estimated
Savings,
$US in Millions
1
7291
8.82
64.3
Carvedilol phosphate
2
0.06
63.2
1
1320
5.00
6.6
Fluvastatin sodium
1-2
3.00
1
275
9.70
2.7
Coreg CR
Fluvastatin ER
Lescol XL
Medication Type, ER Drug Name
IR Formulation
Mean
Spending
per Pill, $
Daily Dose
Frequency,
Pills/d
Cardiometabolic
−1.3
1.0
Glipizide ER
1
122 616
0.27
33.8
Glipizide XL
1
71 750
0.28
20.1
16.6
Glucotrol XL
1
547
1.99
1.1
Isosorbide mononitrate ER
1
312 855
0.32
101.6
Glucophage XR
1
1735
1.04
1.8
Glumetza
1
1049
74.83
73.8
Metformin HCl ER
1
757 381
0.60
320.1
Actoplus Met XR
1
228
12.77
2.7
Propafenone HCl ER
2
8162
5.40
44.3
Rhythmol SR
2
320
12.78
4.1
Dexmethylphenidate HCl ER
1
657
5.25
3.5
Focalin XR
1
101
11.63
1.2
Zenzedi
1-2
70
6.19
0.4
Dextroamphetamine sulfate ER
1-2
3939
2.82
11.1
Fluvoxamine maleate ER
1
1084
7.22
7.8
Fluvoxamine maleate
2
0.51
6.7
Galantamine ER
1
6063
2.86
17.3
Galantamine HBr
2
1.60
−2.1
Lamictal XR
1
1707
24.21
40.3
Lamotrigine ER
1
5897
8.11
46.8
Lamotrigine
2
0.17
Keppra XR
1
3156
8.07
24.9
Levetiracetam ER
1
14 710
0.62
9.0
Levetiracetam
2
0.27
Lithobid
2
234
9.39
2.2
Lithium carbonate ER
2
31 364
0.27
8.3
Namenda XR
1
67 829
13.14
Oxtellar XR
1
1043
Paroxetine CR
1
1550
Paroxetine ER
1
Paxil CR
Quetiapine fumarate ER
Glipizide
1-2
0.07
Isosorbide mononitrate
1-2
0.23
10.0
1.0
−42.1
1.6
Metformin HCl
1-2
0.06
78.4
Pioglitazone HCl plus
metformin HCl
1-2
2.05
2.0
Propafenone HCl
3
0.41
Dexmethylphenidate HCl
2
0.76
356.2
37.4
3.8
Central Nervous System
2.4
1.0
0.2
Dextroamphetamine sulfate
1-3
1.40
0.1
40.8
45.9
23.8
1.2
2.1
Lithium carbonate
3
0.11
891.2
Memantine HCl
1-2
0.94
763.4
11.29
11.7
Oxcarbazepine
2
0.43
10.9
3.67
5.7
4557
3.71
16.8
Paroxetine HCl
1
0.23
14.7
1
472
6.05
2.9
1
11 525
11.02
127.3
Seroquel XR
1
10 039
18.58
188.2
Qudexy XR
1
112
17.41
1.9
Topiramate ER
1
333
11.84
3.9
Trokendi XR
1
1158
20.36
23.6
Ambien CR
1
604
14.28
8.6
Zolpidem tartrate ER
1
12 576
1.75
21.9
1.6
5.0
2.6
118.5
Quetiapine fumarate
2
0.37
Topiramate
1-2
0.16
179.2
1.9
3.8
23.2
8.3
Zolpidem tartrate
1
0.27
15.3
Abbreviations: CR, controlled release; ER, extended release; HBr, hydrobromide; HCl, hydrochloride; IR, immediate release; XR, extended release.
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181 (Reprinted)
Downloaded From: https://jamanetwork.com/ on 03/30/2020
February 28, 2020
4/10
JAMA Network Open | Health Policy
Extended-Release vs Immediate-Release Drugs For Medicare Part D and Medicaid Recipients
reduction ($1.6 billion after accounting for brand-name drug rebates). Between 2012 and 2017,
Medicare Part D spent $12 billion on ER formulations, and switching all patients receiving ER
formulations to generic IR formulations was associated with an $8.5 billion reduction in spending
($7.2 billion after correction for brand-name drug rebates) (Figure 2A).
Medicaid
The same 20 drugs with 36 formulations were included in the Medicaid analysis (Table 2). In 2017,
Medicaid spent $952 million on ER formulations ($598 million on brand-name formulations and $354
million on generic formulations) for 294 million pills at a mean cost of $3.24 ($3.88) per pill ($11.16
[$6.94] per pill for brand formulations and $1.47 [$2.61] for generic formulations). Focalin XR
(dexmethylphenidate hydrochloride) had the highest total expenditure ($307 million for 865 918
claims of 27 million pills), and Glumetza (metformin) had the highest per-claim spending ($4840.33
per claim and $78.71 per pill) for Medicaid. Medicaid spent $429 million for 35 million claims of
generic IR formulations and 2.3 million pills at a mean cost of $12.14 ($10.86) per claim and $0.18
($0.14) per pill in 2017.
In 2017, the estimated reduction for switching from brand-name ER formulations to generic ER
formulations was $299 million. In contrast, switching all ER formulations to generic IR formulations
was associated with a reduction of $836 million. Between 2012 and 2017, Medicaid spent $5.9 billion
on ER formulations. Switching all patients on ER formulations to generic IR formulations was
estimated to be associated with a $5.2 billion spending reduction during the study period ($3.7 billion
with rebates adjustment) (Figure 2B). Between Medicare and Medicaid, the combined spending
reduction by substituting all ER formulations to generic IR formulations was $2.6 billion in 2017 and
$13.7 billion over the study period.
Partial Interchange Sensitivity Analysis
In our sensitivity analysis, the cost reduction to Medicare Part D was $6.4 billion ($5.4 billion with
rebates) for 75% of drugs, $4.25 billion ($3.6 billion with rebates) for 50% of drugs, and $2.1 billion
($1.8 billion with rebates) for 25% of drugs if switching from an ER formulation to a generic IR
formulation occurred (Figure 3A). Similarly, the cumulative cost reduction to Medicaid was $3.9
billion ($2.7 billion with rebates) if a 75% interchange occurred, $2.6 billion ($1.8 billion with rebates)
if a 50% interchange occurred, and $1.3 billion ($0.9 billion with rebates) if a 25% interchange
occurred (Figure 3B).
Figure 2. Medicare Part D and Medicaid Spending and Potential Savings Associated With Switching to Less Costly Formulations
A Medicare Part D formulations for 20 medications
B
Medicaid formulations (n = 36) for 20 medications
3
Medicaid Spending, US$ in Billions
Medicare D Spending, US$ in Billions
3
2
1
0
Spending type
All ER medications
Potential if ER brand substituted to ER generic
2
Potential if all ER substituted to IR generic
1
0
2012
2013
2014
2015
2016
2017
Year
2012
2013
2014
2015
2016
2017
Year
ER indicates extended release; IR, immediate release.
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181 (Reprinted)
Downloaded From: https://jamanetwork.com/ on 03/30/2020
February 28, 2020
5/10
JAMA Network Open | Health Policy
Extended-Release vs Immediate-Release Drugs For Medicare Part D and Medicaid Recipients
Per-Claim Sensitivity Analysis
Medicare Part D
In 2017, the estimated spending reduction for switching from the brand-name ER formulation to the
generic ER formulation was $222 million ($183 million with rebates). In contrast, switching all (brand
and generic) ER formulations to generic IR formulations in 2017 was associated with a $1.8 billion
reduction ($1.6 billion after accounting for brand-name drug rebates) (eTable 1 in the Supplement).
Between 2012 and 2017, Medicare Part D spent $12 billion on ER formulations, and switching all
patients on ER formulations to generic IR formulations was associated with an $8.9 billion reduction
Table 2. Per-Unit Spending and Potential Savings Associated With Switching ER Formulations to Generic IR Formulations in Medicaid, 2017
ER Formulation
IR Formulation
Daily Dose
Frequency,
Pills/d
Estimated
Savings,
Mean Spending $US in
per Dose Unit, $ Millions
Daily Dose
Frequency,
Pills/d
Pills Dispensed,
No. in Thousands
Mean
Spending
per Pill, $US
Spending, $
in Millions
Generic Name
Coreg CR
1
296
9.13
2.7
Carvedilol phosphate
2
0.10
Fluvastatin ER
1
25
5.37
0.1
Fluvastatin sodium
1-2
3.52
Glipizide ER
1
14 504
0.29
4.3
Glipizide XL
1
9651
0.29
2.8
Glucotrol XL
1
12
2.18
0.0
Isosorbide mononitrate ER
1
25 752
0.32
8.1
Fortamet
1
40
33.96
1.4
Glucophage XR
1
84
1.01
0.1
Glumetza
1
115
78.71
8.6
Metformin HCl ER
1
122 806
0.58
53.5
Actoplus Met XR
1
28
15.33
0.4
Pioglitazone HCl plus
metformin HCl
1-2
1.97
0.2
Propafenone HCl ER
2
272
5.21
1.4
Propafenone HCl
3
0.32
2.2
Dexmethylphenidate HCl ER
1
14 546
5.27
76.6
Focalin XR
1
26 780
11.45
306.6
Dexmethylphenidate HCl
2
0.80
Dexedrine
1-2
99
17.30
1.7
Dextroamphetamine sulfate ER
1-2
2605
2.89
7.6
Fluvoxamine maleate ER
1
862
7.17
Galantamine ER
1
50
Lamictal XR
1
Lamotrigine ER
1
Keppra XR
Medication Type, ER Drug Name
Cardiometabolic
2.0
0.0
1.6
Glipizide
1-2
0.09
Isosorbide mononitrate
1-2
0.23
1.0
0.0
-3.3
1.0
0.1
Metformin HCl
1-2
0.08
7.0
44.9
Central Nervous System
46.3
198.3
1.1
Dextroamphetamine sulfate
1-3
1.35
6.2
Fluvoxamine maleate
2
0.36
4.8
2.33
0.1
Galantamine HBr
2
1.34
0.0
1430
22.94
31.9
5441
7.50
39.5
Lamotrigine
2
0.19
1
2365
7.61
17.6
Levetiracetam ER
1
9166
0.58
5.2
Lithium carbonate ER
2
28 091
0.31
8.6
Lithobid
2
180
6.89
1.2
Namenda XR
1
836
12.61
0.4
24.8
33.7
12.9
Levetiracetam
2
0.23
0.9
5.7
Lithium carbonate
3
0.13
10.5
Memantine HCl
1-2
0.64
7.2
Oxcarbazepine
2
0.36
20.4
Paroxetine HCl
1
0.23
4.6
1.8
Oxtellar XR
1
2538
11.24
29.0
Paroxetine CR
1
710
3.84
2.7
Paroxetine ER
1
1486
3.80
5.7
Paxil CR
1
66
6.01
0.4
0.3
Quetiapine Fumarate ER
1
8363
11.26
94.3
76.6
Seroquel XR
1
9420
18.16
171.4
Qudexy XR
1
173
16.27
2.7
Topiramate ER
1
525
11.24
5.7
Trokendi XR
1
1947
19.11
37.2
28.0
Ambien CR
1
73
13.61
1.0
0.7
Zolpidem tartrate ER
1
2824
$1.75
4.9
2.2
Quetiapine fumarate
2
0.37
125.5
2.1
Topiramate
Zolpidem tartrate
1-2
1
0.18
0.20
5.0
3.8
Abbreviations: CR, controlled release; ER, extended release; HBr, hydrobromide; HCl, hydrochloride; IR, immediate release; XR, extemded release.
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181 (Reprinted)
Downloaded From: https://jamanetwork.com/ on 03/30/2020
February 28, 2020
6/10
JAMA Network Open | Health Policy
Extended-Release vs Immediate-Release Drugs For Medicare Part D and Medicaid Recipients
in spending ($8.4 billion after correction for brand-name drug rebates). The absolute spending
reduction difference between Medicare per-claim and per-unit analysis was $24 million (1.3%) in
2017 and $386 million (4.5%) between 2012 and 2017.
Medicaid
In 2017, the estimated reduction for switching from a brand-name ER formulation to a generic ER
formulation was $325 million ($251 million with rebates). In contrast, switching all ER formulations to
generic IR formulations was associated with a reduction of $856 million ($682 million with rebates).
Between 2012 and 2017, Medicaid spent $6.3 billion on ER formulations. Switching all patients
receiving ER formulations to generic IR formulations was estimated to be associated with a $5.5
billion spending reduction during the study period ($4.3 billion with rebate adjustment). The
spending reduction difference between Medicaid per-claim and per-unit analysis was $20 million
(2.4%) in 2017 and $319 million (6.1%) between 2012 and 2017.
Discussion
We quantified Medicare Part D and Medicaid spending associated with prescription of ER
formulations and found a total potential annual cost savings of up $2.6 billion that could be achieved
by switching these drugs to therapeutically comparable generic IR formulations in both Medicare
and Medicaid. These savings were demonstrated using only ER formulations saving at most 1
additional daily dose compared with their generic IR formulation counterpart.
Some ER formulations have superior therapeutic effects (eg, metoprolol succinate for heart
failure, tolterodine for overactive bladder, and venlafaxine for depression) or fewer adverse effects
(eg, niacin for dyslipidemia, nifedipine for patients with heart failure) compared with their generic IR
formulation counterparts. Furthermore, ER formulations may be associated with reduced pill burden
and improved patient convenience and medication adherence.5 However, most of the existing
literature on ER vs IR formulations (eTable 2 in the Supplement) has evaluated therapeutic efficacy
rather than medication adherence and has frequently demonstrated therapeutic equivalence. In
addition, several drugs with both ER and IR formulations available have had no direct ER vs IR
comparison studies performed for adherence or therapeutic efficacy (eTable 2 in the Supplement). A
meta-analysis of 23 studies10 showed no association between administration frequency and
medication adherence. Another meta-analysis of 76 studies11 showed an adherence benefit of ER
formulations only when the administration frequency was reduced by 2 or more daily doses
compared with the corresponding IR formulation. To further address this, we only included ER drugs
with 1 fewer daily doses than the IR drug counterpart. Taken together, the existing evidence suggests
Figure 3. Medicare Part D and Medicaid Cumulative Savings Using Partial Extended-Release to Immediate-Release Interchange With and Without Rebates
A Medicare Part D formulations for 20 medications
B
Medicaid formulations (n = 36) for 20 medications
8
Cumulative Savings, US$ in Billions
Cumulative Savings, US$ in Billions
8
6
4
2
0
75% Interchange
75% Interchange plus rebates
50% Interchange
50% Interchange plus rebates
25% Interchange
25% Interchange plus rebates
6
4
2
0
2012
2013
2014
2015
2016
2017
Year
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181 (Reprinted)
Downloaded From: https://jamanetwork.com/ on 03/30/2020
2012
2013
2014
2015
2016
2017
Year
February 28, 2020
7/10
JAMA Network Open | Health Policy
Extended-Release vs Immediate-Release Drugs For Medicare Part D and Medicaid Recipients
that therapeutic and adherence-related benefits of ER formulations are medication specific and not
consistent across all available ER formulations.
Although the present study focused on costs to the payer attributable to the use of ER
formulations, the financial strain for patients associated with the higher costs for these drugs remains
a significant but underrecognized challenge.12,13 Although fewer pills may be more convenient, the
added cost burden of ER vs generic IR formulations may be associated with negative utility for
patients and perhaps in cost-related nonadherence.14-16 Studies14,15,17 among Medicare patients have
shown that 7% to 16% of patients were nonadherent because of excessive medication costs. Costrelated nonadherence has also been associated with more frequent emergency department visits
and poorer health outcomes.17
However, estimating out-of-pocket costs for Medicare Part D is difficult because plans are
heterogenous and provide varying levels of out-of-pocket benefits to their enrollees. The minimum
Medicare Part D plan requires patients to be responsible for 100% of costs below the $415 threshold
(initial deductible), 25% of costs up to $3820 (initial coverage limit), between 25% and 37% of costs
up to $8140 (coverage gap), and 5% of costs beyond $8140 (catastrophic coverage threshold).18
Although most plans offer a greater level of coverage for medications in the form of lower
copayments or coinsurance, 92% of enrollees did not reach the catastrophic coverage threshold19
and could still find significant out-of-pocket savings when switching to the generic IR formulation.
The out-of-pocket costs were less significant among Medicaid beneficiaries because Medicaid only
requires copayments of $4.00 to $8.00 for beneficiaries below the 150% federal poverty limit.
Among beneficiaries who are above the 150% federal poverty threshold, only nonpreferred drugs
have significant out-of-pocket costs because these medications require a 20% copayment. It is
possible that the only out-of-pocket difference among Medicaid beneficiaries occurs if a
nonpreferred ER formulation is prescribed.
Although out-of-pocket savings are beyond the scope of this analysis, we calculated combined
Medicare and Medicaid cost savings of $13.7 billion during 6 years, with switching from ER
formulations saving at most 1 additional daily dose compared with their generic IR formulations
counterpart. However, achieving these savings would likely take a multipronged effort. First, the use
of blister packs or prepackaged pill boxes has been shown to improve adherence and may offset the
complexity of the added IR pill burden.20 Another option is to permit pharmacist-directed
therapeutic exchange. Legislation passed in Arkansas, Idaho, and Kentucky allows for pharmacists to
substitute a prescribed drug for another in the same therapeutic class.21 This option leverages
pharmacist expertise and provides substantial opportunity to reduce cost. The option similarly
provides patients the ability to make an informed decision because the patient may most readily see
variation in out-of-pocket costs at the pharmacy. In addition, legislative action to reduce the price
disparity between ER and IR formulation prices may be needed. In our study, the daily price of generic
ER formulations was significantly higher than the daily price of generic IR formulations. Reductions
in the difference between generic ER formulations and generic IR formulations through additional
rebates, drug price negotiations, or additional market competition may assist in reducing the
discrepancy between ER and IR medications. Switching from costlier ER formulations to less costly
generic IR formulations, at least for medications that have comparable pill burden with ER vs IR
formulations, warrants discussion with patients beyond the potential cost savings to Medicare.
Limitations
This study has limitations. First, the data sets used in this analysis do not include information on the
out-of-pocket costs experienced by Medicare or Medicaid beneficiaries. Although the system savings
were apparent, it was unclear whether switching to an IR formulation would increase dose burden
while failing to provide any individual financial relief. Second, we could not account for patient-level
characteristics, such as previous IR drug nonadherence or complexity of medication regimen, which
may have contributed to an ER drug prescription. Although underlying patient factors beyond
convenience may be associated with the use of some ER prescriptions, our sensitivity analyses with
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181 (Reprinted)
Downloaded From: https://jamanetwork.com/ on 03/30/2020
February 28, 2020
8/10
JAMA Network Open | Health Policy
Extended-Release vs Immediate-Release Drugs For Medicare Part D and Medicaid Recipients
25% to 75% interchange rates showed a significant cost reduction even if only a small number of
patients switched the type of formulation. In addition, we assumed that patients switching from ER
to generic IR formulations would take the maximum IR pills per day even though some IR
formulations had a range of doses. Thus, our calculations likely underestimated the potential
spending reduction because some patients may receive fewer IR drug daily doses.
Conclusions
This study suggests that there is high spending burden on Medicare and Medicaid associated with the
use of ER formulations and that the potential cost savings may be achieved by substituting
therapeutically comparable, less-expensive generic IR formulations. Future studies are needed to
assess whether this strategy to reduce medication-related expenditures for Medicare and Medicaid
is associated with lower patient out-of-pocket costs and thereby positively affects patient care.
ARTICLE INFORMATION
Accepted for Publication: December 14, 2019.
Published: February 28, 2020. doi:10.1001/jamanetworkopen.2020.0181
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Sumarsono
A et al. JAMA Network Open.
Corresponding Author: Ambarish Pandey, MD, MSCS, Division of Cardiology, Department of Internal Medicine,
The University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390-8830 (ambarish.
pandey@utsouthwestern.edu).
Author Affiliations: Department of Internal Medicine, The University of Texas Southwestern Medical Center,
Dallas (A. Sumarsono); Department of Pediatrics, Stanford University, Palo Alto, California (N. Sumarsono);
Division of Cardiology, Department of Internal Medicine, The University of Texas Southwestern Medical Center,
Dallas (Das, Pandey); Division of Cardiology, Department of Medicine, Brigham and Women’s Hospital, Boston,
Massachusetts (Vaduganathan); Division of Gastroenterology and Hepatology, Department of Internal Medicine,
University of Texas at Austin, Dell Medical School, Austin (Agrawal).
Author Contributions: Dr A. Sumarsono had full access to all of the data in the study and takes responsibility for
the integrity of the data and the accuracy of the data analysis.
Concept and design: A. Sumarsono, Pandey.
Acquisition, analysis, or interpretation of data: A. Sumarsono, N. Sumarsono, Das, Vaduganathan, Agrawal.
Drafting of the manuscript: A. Sumarsono, N. Sumarsono.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: A. Sumarsono, N. Sumarsono, Pandey.
Obtained funding: Pandey.
Administrative, technical, or material support: A. Sumarsono, N. Sumarsono, Pandey.
Conflict of Interest Disclosures: Dr Vaduganathan reported receiving the KL2/Catalyst Medical Research
Investigator Training award from Harvard Catalyst; serving on advisory boards for Amgen Inc, AstraZeneca, Baxter
Healthcare, Bayer AG, Boehringer Ingelheim, and Relypsa Inc; and participating on clinical end point committees
for studies sponsored by Novartis and the National Institutes of Health. Dr Pandey reported receiving a research
grant from the Texas Health Resources Clinical Scholars Program. No other disclosures were reported.
REFERENCES
1. Papanicolas I, Woskie LR, Jha AK. Health care spending in the United States and other high-income countries.
JAMA. 2018;319(10):1024-1039. doi:10.1001/jama.2018.1150
2. Sacks CA, Lee CC, Kesselheim AS, Avorn J. Medicare spending on brand-name combination medications vs their
generic constituents. JAMA. 2018;320(7):650-656. doi:10.1001/jama.2018.11439
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181 (Reprinted)
Downloaded From: https://jamanetwork.com/ on 03/30/2020
February 28, 2020
9/10
JAMA Network Open | Health Policy
Extended-Release vs Immediate-Release Drugs For Medicare Part D and Medicaid Recipients
3. Udelson JE, Pressler SJ, Sackner-Bernstein J, et al. Adherence with once daily versus twice daily carvedilol in
patients with heart failure: the Compliance and Quality of Life Study Comparing Once-Daily Controlled-Release
Carvedilol CR and Twice-Daily Immediate-Release Carvedilol IR in Patients with Heart Failure (CASPER) Trial. J Card
Fail. 2009;15(5):385-393. doi:10.1016/j.cardfail.2008.12.010
4. Spencer TJ, Mick E, Surman CB, et al. A randomized, single-blind, substitution study of OROS methylphenidate
(Concerta) in ADHD adults receiving immediate release methylphenidate. J Atten Disord. 2011;15(4):286-294. doi:
10.1177/1087054710367880
5. Haynes RB, Ackloo E, Sahota N, McDonald HP, Yao X. Interventions for enhancing medication adherence.
Cochrane Database Syst Rev. 2008;(2):CD000011.
6. Centers for Medicare & Medicaid Services. Medicare Part D drug spending dashboard and data. https://www.cms.
gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Information-on-PrescriptionDrugs/MedicarePartD. Accessed September 1, 2019.
7. Centers for Medicare & Medicaid Services. Medicaid drug spending dashboard. https://www.cms.gov/ResearchStatistics-Data-and-Systems/Statistics-Trends-and-Reports/Information-on-Prescription-Drugs/Medicaid. Accessed
September 1, 2019.
8. Centers for Medicare & Medicaid Services. 2014 Part D rebate summary for all brand name drugs. Updated May
15, 2018. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Informationon-Prescription-Drugs/PartD_Rebates.html. Accessed January 17, 2020.
9. Centers for Medicare & Medicaid Services. Medicaid Drug Rebate Program. https://www.medicaid.gov/medicaid/
prescription-drugs/medicaid-drug-rebate-program/index.html. Accessed October 17, 2019.
10. Peterson AM, Takiya L, Finley R. Meta-analysis of trials of interventions to improve medication adherence. Am
J Health Syst Pharm. 2003;60(7):657-665. doi:10.1093/ajhp/60.7.657
11. Claxton AJ, Cramer J, Pierce C. A systematic review of the associations between dose regimens and medication
compliance. Clin Ther. 2001;23(8):1296-1310. doi:10.1016/S0149-2918(01)80109-0
12. Goldman DP, Joyce GF, Zheng Y. Prescription drug cost sharing: associations with medication and medical
utilization and spending and health. JAMA. 2007;298(1):61-69. doi:10.1001/jama.298.1.61
13. Choudhry NK, Fischer MA, Avorn J, et al. At Pitney Bowes, value-based insurance design cut copayments and
increased drug adherence. Health Aff (Millwood). 2010;29(11):1995-2001. doi:10.1377/hlthaff.2010.0336
14. Madden JM, Graves AJ, Zhang F, et al. Cost-related medication nonadherence and spending on basic needs
following implementation of Medicare Part D. JAMA. 2008;299(16):1922-1928. doi:10.1001/jama.299.16.1922
15. Williams J, Steers WN, Ettner SL, Mangione CM, Duru OK. Cost-related nonadherence by medication type
among Medicare Part D beneficiaries with diabetes. Med Care. 2013;51(2):193-198. doi:10.1097/MLR.
0b013e318270dc52
16. Mojtabai R, Olfson M. Medication costs, adherence, and health outcomes among Medicare beneficiaries.
Health Aff (Millwood). 2003;22(4):220-229. doi:10.1377/hlthaff.22.4.220
17. Blanchard J, Madden JM, Ross-Degnan D, Gresenz CR, Soumerai SB. The relationship between emergency
department use and cost-related medication nonadherence among Medicare beneficiaries. Ann Emerg Med. 2013;
62(5):475-485. doi:10.1016/j.annemergmed.2013.04.013
18. Kaiser Family Foundation. An overview of the Medicare Part D prescription drug benefit. https://www.kff.org/
medicare/fact-sheet/an-overview-of-the-medicare-part-d-prescription-drug-benefit/. Published November 13, 2019.
Accessed July 26, 2019.
19. Kaiser Family Foundation. How many Medicare Part D enrollees had high out-of-pocket drug costs in 2017?
https://www.kff.org/medicare/issue-brief/how-many-medicare-part-d-enrollees-had-high-out-of-pocketdrug-costs-in-2017/. Published June 21, 2019. Accessed October 16, 2019.
20. Conn VS, Ruppar TM, Chan KC, Dunbar-Jacob J, Pepper GA, De Geest S. Packaging interventions to increase
medication adherence: systematic review and meta-analysis. Curr Med Res Opin. 2015;31(1):145-160. doi:10.1185/
03007995.2014.978939
21. Vanderholm T, Klepser D, Adams AJ. State approaches to therapeutic interchange in community pharmacy
settings: legislative and regulatory authority. J Manag Care Spec Pharm. 2018;24(12):1260-1263. doi:10.18553/
jmcp.2018.24.12.1260
SUPPLEMENT.
eTable 1. Sensitivity Analysis Using Per Claim Analysis. Spending and Potential Savings of Switching ExtendedRelease Formulations to Generic Immediate-Release Formulations in 2017 in the Medicare Part D
eTable 2. Literature Review of Extended-release Vs Immediate-release of All Included Study Drugs
JAMA Network Open. 2020;3(2):e200181. doi:10.1001/jamanetworkopen.2020.0181 (Reprinted)
Downloaded From: https://jamanetwork.com/ on 03/30/2020
February 28, 2020
10/10