#35626 1pg 6hrs

profileltdprinwival
Article__2_questions1.pdf

doi: 10.1377/hlthaff.2015.1530 HEALTH AFFAIRS 35, NO. 9 (2016): 1604–1607 ©2016 Project HOPE— The People-to-People Health Foundation, Inc.

Pharmaceutical Spending & Value

By Hefei Wen, Tyrone F. Borders, and Benjamin G. Druss

DATAWATC H

Number Of Medicaid Prescriptions

Grew, Drug Spending Was Steady In Medicaid Expansion States Expansions of eligibility for Medicaid under the Affordable Care Act may have increased the number of Medicaid drug prescriptions. However, the expansions did not drive Medicaid spending on prescription drugs overall in 2014.

In 2014 twenty-six states and the District of Columbia began to expand Medicaid eligibility to almost all residents whose household incomes were at or below 138 percent of the federal poverty level.

By the end of 2014 an estimated nine million Americans had gained insurance coverage through the expansions.

1 In the same year the growth rate of

all prescription drug spending in the United States reached 13.1 percent, its high-est point since 2001.

2

The 2014 growth rate of Medicaid drug spending (24.3 percent) was even higher than that of all prescription drug spend-ing.

3 The concurrent trends

of increasing Medic-aid enrollment and escalating Medicaid drug spending have led people to partially attribute the growth in drug spending to Medicaid expan-sion.

2–4 This may cause concern in states now

contemplating opting into the Medicaid expan-

sion and in those considering whether to contin-ue their existing expansion programs.

We found significant increases in Medicaid drug spending (Exhibit 1) and numbers of pre-scriptions (Exhibit 2) from the preexpansion period (2011–13) to the postexpansion period (2014). For Medicaid drug spending, similar up-ward trends were seen both in states that imple-mented Medicaid expansions under the Afford-able Care Act (ACA) in 2014 and in states that expanded eligibility later or did not expand it (we excluded the District of Columbia and Virginia from the study sample because of incomplete-ness and inconsistency in data reporting). For the number of Medicaid prescriptions, however, the upward trend was not seen in states that expanded eligibility after 2014 or not at all (la-beled “non- or late-expansion states”). The trend in these states held steady during 2014.

Exhibit 1

Trends in quarterly Medicaid spending on all Medicaid-covered outpatient prescription drugs

SOURCE Authors’ analysis of data for 2011–14 from the Medicaid State Drug Utilization Data files of the Centers for Medicare and Medicaid Services. NOTES Dollar amounts were converted to December 2014 values based on the national monthly Consumer Price Index. “Expansion states” are the twenty-six states that began to expand eligibility for Medicaid in 2014. “Non- or late-expansion states” are the four states that began expansion after January 1, 2015, and the nineteen states that have not expanded eligibility (we excluded Virginia, which has not expanded eligibility, and the District of Columbia, which expanded it in 2014, from the study sample because of incompleteness and inconsistency in data reporting).

1 6 0 4 H e a lt h A f fa i r s S e p t e m b e r 2 0 1 6 3 5 : 9

Downloaded from HealthAffairs.org on November 09, 2018. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org.

Exhibit 2 Trends in quarterly Medicaid prescriptions for all Medicaid-covered outpatient prescription drugs SOURCE Authors’ analysis of data for 2011–14 from the Medicaid State Drug Utilization Data files of the Centers for Medicare and Medicaid Services. NOTE “Expansion states” and “non- or late-expansion states” are explained in the Notes to Exhibit 1.

A more rigorous difference-in-differences esti- mation (Exhibit 3) was consistent with the trend comparisons and confirmed that the ACA Med-icaid expansions may have increased the number of Medicaid prescriptions, but the expansions per se are unlikely to be the major driving force behind the growth in drug spending. Study Data And Methods We used sixteen waves of quarterly state-aggregate data, for the period 2011–14, on Med-icaid spending on prescription drugs from the Medicaid State Drug Utilization Data files of the Centers for Medicare and Medicaid Services (CMS).

5

Exhibit 3

Our outcome variables were quarterly Medic-aid spending on all covered outpatient prescrip-tion drugs per state resident and quarterly num-bers of those prescriptions. We also estimated per enrollee Medicaid drug spending and pre-scriptions. Medicaid drug spending was mea-sured as the pre- rebate amount reimbursed by Medicaid only.We converted the nominal spend-ing values to real values as of December 2014 based on the Consumer Price Index.

As noted above, twenty-six states and the Dis-trict of Columbia implemented Medicaid expan-sions under the ACA during 2014. Twenty-two of the states and the District of Columbia imple-mented the expansions in full compliance with the Medicaid state plan amendment provision of

Estimated effects of Affordable Care Act expansions of Medicaid eligibility on Medicaid drug spending and number of prescriptions per state resident

Difference-in-differences

Adjusted for state and Adjusted for state and quarter Difference quarter fixed effects fixed effects and covariates

2011–13 2014 Amount 95% CI Amount 95% CI Amount 95% CI Spending per quarter per resident ($) Non- or late-expansion states 29.86 33.07 3.21*** [1.28, 5.15] Ref —

a Ref —

a Expansion states 32.29 37.03 4.75** [0.02, 9.51] 1.58 [−1.18, 4.33] 0.81 [−2.85, 4.47]

Number of prescriptions per quarter per resident Non- or late-expansion states 0.41 0.41 0.002 [−0.01, 0.02] Ref —a Ref —a Expansion states 0.47 0.53 0.06*** [0.02, 0.11] 0.06*** [0.02, 0.10] 0.07*** [0.03, 0.11]

SOURCE Authors’ analysis of data for 2011–14 from the Medicaid State Drug Utilization Data files of the Centers for Medicare and Medicaid Services. NOTES “Expansion states” and “non- or late-expansion states” are explained in the Notes to Exhibit 1. Covariates are listed in the text. Dollar amounts were converted to December 2014 values based on the national monthly Consumer Price Index. 95% confidence intervals (CIs) were calculated based on state-clustered standard errors. a Not applicable. **p < 0:05 ***p < 0:01

S e p t e m b e r 2 0 1 6 3 5 : 9 H e a lt h A f fa i r s 1 6 0 5

Downloaded from HealthAffairs.org on November 09, 2018.

Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org.

Pharmaceutical Spending & Value

the ACA. The remaining four states used a section 1115 waiver to waive certain statutory require- ments for Medicaid and redirect Medicaid funds into premium assistance programs of qualified health plans in the ACA health insurance Mar-ketplaces.

Our key independent variable was the policy

indicator for expansion of Medicaid in the twen-ty- six states in 2014. We also provide separate estimates for the twenty-two state expansions under the state plan amendment provision and the four state expansions under the section 1115 waiver (online Appendices A1 and A2).

6

The preexpansion period and the states that expanded after 2014 (Alaska, Indiana, Montana, and Pennsylvania) or did not expand served as the comparisons. State-level covariates were the following: unemployment rate; poverty rate; penetration rate of Medicaid managed care, measured by the percentage of Medicaid enroll-ees in comprehensive managed care plans; and an “early adopter” indicator for partial imple-mentation of Medicaid expansions in the period 2011–13.

7

We used a quasi-experimental difference-in-

differences design with state and quarter two-way fixed effects to account for unobserved state heterogeneity and national secular trends in Medicaid drug spending and prescriptions.

8 All

estimates were population-weighted and state- clustered to correct for the heterogeneous policy effect and within-state serial correlation in our difference-in-differences context.

9

We performed sensitivity analyses to exclude sofosbuvir (Sovaldi), a major driver of Medicaid drug spending growth in 2014,

10 and to add the

group-specific linear trends to account for the potential heterogeneous trajectory in Medicaid drug spending and number of prescriptions be-tween the expansion states and the non- or late-expansion states that might have emerged before the expansions (for results of the sensitivity an-alyses, see Appendix A3).

6

Our study had several limitations. One was the fact that the study data included only four quar-ters of postexpansion data. Another was that there may be inconsistency in state reporting of new Medicaid enrollees under the expansions and the increased federal matching rates avail-able to new enrollees. In addition, our analysis was aggregated at the state level, which did not allow us to distinguish the new enrollees after expansion from existing enrollees.

Study Results We found significant increases from the pre- expansion period (2011–13) to the postexpan-sion period (2014) in the amount of Medicaid

1 6 0 6 H e a lt h A f fa i r s S e p t e m b e r 2 0 1 6 3 5 : 9

drug spending per resident in the twenty-three non- or late-expansion states ($3.21 per quarter) and in the twenty-six expansion states ($4.75 per quarter) (Exhibit 3).When we compared the pre-post spending changes between the two groups of states, our difference-in-differences estimates indicated that the difference was not significant. The denominator of the outcome was the num-ber of state residents, which remained stable over the short term. Our estimates thus confirm that state implementation of the ACA Medicaid expansions did not affect total Medicaid drug spending.

We found no discernible change over time in the number of Medicaid prescriptions per resident in the non- or late-expansion states (Exhibit 3). However, there was a significant increase in prescriptions (0.06 per resident per quarter) in the expansion states. This relative increase implies that the new Medicaid enrollees after expansion may have had a considerable level of demand for prescription drugs.

Appendix Exhibit A4 provides additional evi- dence for the effect of the ACA Medicaid expan- sions on Medicaid drug spending and number of prescriptions on a per enrollee basis.

6 The find-ings

suggest that, on average, Medicaid enroll-ees in the expansion states may have been pre-scribed drugs at a rate no different from those in the non- or late- expansion states, but the drugs prescribed for enrollees in the expansion states may have been less expensive than those pre-scribed for enrollees in the other states. Discussion Our study provides some of the first empirical evidence concerning the implications of the ACA Medicaid expansions for prescription drug utili- zation. On one hand, we found that state expan-sions did not affect Medicaid drug spending as a whole or per resident. This finding suggests that Medicaid expansion per se is unlikely to be the primary driver of the record-high drug spending growth in 2014.

On the other hand, we found that implemen-tation of the expansions may have been associ-ated with a relative increase in the numbers of Medicaid prescriptions overall or per resident. We also found a relative decrease in Medicaid drug spending per enrollee that was associated with the implementation of the expansions.

A possible explanation for the lack of signifi-cant impact of the ACA Medicaid expansions on Medicaid drug spending growth in spite of the rising number of prescriptions is that expansion states, facing the potential fiscal impact of ex-pansions and emerging specialty drugs, may have taken proactive approaches to contain costs

Downloaded from HealthAffairs.org on November 09, 2018.

Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org.

for prescription drugs. There are three reasons for this explanation.

First, according to the annual budget survey of Medicaid officials for fiscal years 2014 and 2015 by the Henry J. Kaiser Family Foundation and Health Management Associates, nineteen ex-pansion states but only nine nonexpansion states have implemented pharmacy manage-ment initiatives, such as prior authorization pro-grams, preferred drug lists, pharmacy benefit carve-outs, incentives to use generic drugs, and reduced reimbursements for certain drug ingredients.

11 These cost-containment

strategies may affect Medicaid drug spending not only on the new enrollees after expansion but also on existing enrollees.

Second, many states have also taken actions to increase enrollment in risk-based managed care. These actions include making enrollment in managed care mandatory for new enrollees after expansion, expanding voluntary or mandatory enrollment to additional groups eligible for man-

aged care, and establishing managed care pro-grams in new regions.

Finally, the expansion states generally had a considerable rate of managed care penetration in their Medicaid programs before the expansions (approximately 70 percent of enrollees in these states were in managed care in 2013). This might have helped mitigate the impact of the expan-sions on Medicaid drug spending. Conclusion Our study used timely and comprehensive Med- icaid administrative data and provides some of the first empirical evidence for the impact of the ACA Medicaid expansions on Medicaid drug spending and number of prescriptions. Our findings suggest that state implementation of the expansions may have increased the number of Medicaid drug prescriptions but had no sig-nificant immediate

impact on drug spending growth. ▪

NOTES

1 Centers for Medicare and Medicaid Services. Medicaid and CHIP: De- cember 2014 monthly applications, eligibility determinations, and en- rollment report [Internet]. Balti-more (MD): CMS; 2015 Feb 23 [cited 2016 Jul 14]. Available from: http:// www.medicaid.gov/medicaid-chip- program-information/program- information/downloads/december-2014- enrollment-report.pdf

2 IMS Institute for Healthcare Infor- matics. Medicines use and spending shifts: a review of the use of medi-cines in the U.S. in 2014. Parsippany (NJ): The Institute; 2015.

3 Martin AB, Hartman M, Benson J, Catlin A, National Health Expendi-ture Accounts Team. National health spending in 2014: faster growth driven by coverage expansion and prescription drugs. Health Aff (Millwood). 2016;35(1):150–60.

4 Truffer CJ, Wolfe CJ, Rennie KE. 2014 actuarial report on the finan-cial outlook for Medicaid [Internet].

Baltimore (MD): Centers for Medi-care and Medicaid Services; 2014 [cited 2016 Jul 27]. Available from: https://www.medicaid.gov/ medicaid- chip-program-information/by- topics/financing-and- reimbursement/downloads/ medicaid- actuarial-report-2014.pdf

5 Medicaid.gov. State Drug Utilization Data [Internet]. Baltimore (MD): Centers for Medicare and Medicaid Services; [cited 2016 Jul 14]. Avail-able from: https://www.medicaid .gov/medicaid-chip-program- information/by-topics/benefits/ prescription-drugs/state-drug- utilization-data.html

6 To access the Appendix, click on the Appendix link in the box to the right of the article online.

7 Sommers BD, Kenney GM, Epstein AM. New evidence on the Affordable Care Act: coverage impacts of early Medicaid expansions. Health Aff (Millwood). 2014; 33(1):78–87.

8 Wooldridge JM. Econometric analy-

sis of cross section and panel data. 2nd ed. Cambridge (MA): MIT Press; 2010.

9 Bertrand M, Duflo E, Mullainathan S. How much should we trust dif-ferences- in-differences estimates? Q J Econ. 2004;119(1):249–75.

10 Liao JM, Fischer MA. Early patterns of sofosbuvir utilization by state Medicaid programs. N Engl J Med. 2015;373(13):1279–81.

11 Smith VK, Gifford K, Ellis E, Rudowitz R, Snyder L. Medicaid in an era of health and delivery system reform: results from a 50-state Medicaid budget survey for state fiscal years 2014 and 2015 [Inter- net]. Menlo Park (CA): Henry J. Kaiser Family Foundation; 2014 Oct [cited 2016 Jul 14]. Available from: https://kaiserfamilyfoundation

.files.wordpress.com/2014/10/ 8639- medicaid-in-an-era-of-health-delivery- system-reform3.pdf

S e p t e m b e r 2 0 1 6 3 5 : 9 H e a lt h A f fa i r s 1 6 0 7

Downloaded from HealthAffairs.org on November 09, 2018. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org.