Comparing Patient OOP Spending for Specialty Drugs in Medicare Part D and Employer-Sponsored Insurance

ABSTRACT

Objectives: Per capita spending on specialty drugs increased 55% between 2014 and 2018. Individuals aged 55 to 75 years using specialty drugs make the transition from employer-sponsored insurance (ESI) to Medicare Part D coverage. We compared out-of-pocket (OOP) spending across ESI, Medicare fee-for-service (FFS), and Medicare Advantage (MA) prescription drug plans to examine the impact of benefit design on OOP spending.

Study Design: Analyses consisted of Truven MarketScan and Medicare Part D prescription drug claims from 2013 to 2017 for individuals enrolled in ESI, FFS, and MA drug plans taking at least 1 drug among the top 4 specialty drug classes: rheumatoid arthritis (RA), multiple sclerosis (MS), cancer, and hepatitis C.

Methods: Multivariate regression analyses with fixed effects were used to assess whether there are differences in OOP spending by insurance type and the impact of benefit design differences. A secondary outcome was drug choice within a therapeutic class.

Results: There were small differences in drug choice between Medicare and ESI but significant differences in OOP spending. Monthly OOP spending for ESI relative to FFS was $108 less for RA drugs, $288 less for MS drugs, $504 less for cancer drugs, and $1437 less for hepatitis C drugs. Spending was slightly greater for beneficiaries in MA plans compared with FFS. Higher Medicare spending was driven by gaps in coverage in the Part D benefit phases because beneficiaries pay a percentage of list price.

Conclusions: OOP spending was substantially higher for Medicare enrollees compared with ESI enrollees as a result of the Part D benefit structure.

Am J Manag Care. 2020;26(9):388-394. https://doi.org/10.37765/ajmc.2020.88489

Takeaway Points

Medicare beneficiaries pay substantially higher out-of-pocket (OOP) costs for specialty drugs than employer-sponsored insurance enrollees, primarily because of the lack of an OOP cap and the “donut hole.”

• This study examines differences in OOP spending as individuals age from employer-sponsored insurance into Medicare Part D coverage.
• Based on specialty drug class, average OOP spending for Medicare fee-for-service (FFS) enrollees was $108 to $1437 greater than spending by employer-sponsored insurance enrollees. Medicare Advantage beneficiaries had slightly higher OOP spending than FFS beneficiaries.
• The benefit structure of Part D results in Medicare enrollees paying substantially higher OOP costs for specialty drugs than employer-sponsored insurance enrollees, primarily due to the absence of an OOP cap in Medicare and the donut hole.

The pharmaceutical pipeline is moving toward more high-cost specialty drugs.¹ From 2008 until 2017, CMS defined specialty drugs as those with a monthly cost greater than $600; beginning in 2017, the threshold became $670.^2,3 These drugs are primarily biologics and biosimilars used to treat complex chronic conditions such as rheumatoid arthritis (RA), multiple sclerosis (MS), cancer, and hepatitis C. Patients with these conditions have few other clinical options, forcing them to pay high cost-sharing rates or forgo treatment.^4-6

Care for these diseases can begin around age 40 years and continue throughout life, making health insurance coverage likely to span across employer-sponsored insurance (ESI) and Medicare prescription drug (Part D) coverage.^7-10 Few studies have examined whether the financial burden placed on patients who take specialty drugs differs among those with ESI, Medicare fee-for-service (FFS), and Medicare managed care (Medicare Advantage [MA]) drug coverage.^11-13 Given the policy and clinical concerns that high cost sharing may lead to nonadherence and financial burden, it is important to understand the levels of cost sharing across insurance types and the role that benefit design may play.

ESI and Medicare Part D prescription drug plans are administered via private insurance plans and typically involve the same 3 pharmacy benefit managers that have the greatest market share. The design of the insurance benefit, however, may result in differences in out-of-pocket (OOP) spending.^14-16 We examine if there are differences in specialty drug choice and OOP spending across 3 populations: ESI enrollees, enrollees in stand-alone Medicare Part D plans (for FFS beneficiaries), and enrollees in an MA plan that offers prescription drug coverage. First, we compared Medicare spending with ESI spending, and then we examined Medicare FFS and MA populations separately. We analyze FFS and MA plans separately because MA plans are paid an annual, capitated amount to provide a beneficiary’s inpatient and outpatient medical benefits, whereas FFS drug plans do not have inpatient and outpatient benefits under their auspices. Thus, MA plans have an incentive to minimize total health care spending of their enrollees and encourage greater adherence to drug regimens by having lower cost sharing. Although Part D is outside the MA capitated payment, adherence to prescription drug regimens is often a low-cost substitute for medical spending in the management of common chronic conditions.^17,18

The standard Part D benefit is the same for beneficiaries in FFS and MA drug plans and typically requires beneficiaries to pay a percentage of a drug’s list price. The Part D benefit has 4 phases—deductible, initial coverage period, coverage gap (“donut hole”), and catastrophic coverage—with varying levels of cost sharing. Beneficiaries are responsible for all costs during the deductible period (a maximum of $400 in 2017), 25% of gross drug costs up to the initial coverage limit ($3700 in 2017), and 40% of their branded drug costs and 50% of their generic drug costs in the coverage gap. Once OOP expenses reach $4950 (2017 limit), beneficiaries enter catastrophic coverage, in which they pay 5% of drug costs. There is no OOP maximum. There is wide variation in the benefit structure of ESI plans and no comprehensive data on the benefit structure for drugs across all ESI plans. Anecdotal evidence suggests that ESI plans do not have coverage gaps like Medicare plans (although some have separate tiers for specialty drugs that can utilize coinsurance, rather than co-payments) and most have an OOP cap.¹⁹

Per capita spending on specialty drugs increased 55% from 2014 to 2018, driven primarily by a 57% increase in the price of branded specialty drugs and more drugs entering the market.²⁰ High and rising prices of specialty drugs may have adverse implications for patient financial and clinical outcomes because adherence to specialty drugs is negatively correlated to the level of cost sharing.²¹ This is especially important for a subset of specialty drugs, some of which have list prices above $50,000 and for which the cost sharing can be 25% to 33% of the list price. An improved understanding of how differences in benefit design affect OOP spending is important for patients, physicians, payers, and Medicare policy makers. This study examines differences in OOP costs and drug choice among enrollees with ESI, FFS, and MA.

METHODS

Data and Sample

We analyzed prescription drug claims from 2013 to 2017. We used Medicare Part D claims for a 20% random sample of FFS and MA beneficiaries to capture OOP spending by Medicare beneficiaries. The data included information on beneficiary cost sharing, starting and ending benefit phases for each prescription fill, and the plan’s formulary. Part D claims were linked with age, gender, race, and state and county of residence from the Master Beneficiary Summary File. Individuals receiving low-income subsidies were excluded from the analysis because they have limited cost sharing that does not follow the standard Medicare benefit structure.

To assess drug spending in ESI, we used data from Truven MarketScan, a claims and encounter database encompassing approximately 25% of the commercially insured population (not nationally representative).²² The data come from employers and commercial insurers and include active employees, early retirees, and dependents insured by employer-sponsored plans (including but not limited to Employee Retirement Income Security Act/self-insured plans), but they do not include those insured by individual marketplace plans. We merged prescription drug claims with enrollee characteristics including age, gender, metropolitan statistical area (MSA) of residence, and plan type (eg, health maintenance organization [HMO], preferred provider organization [PPO], FFS, capitated, high deductible).

In both the ESI and Medicare settings, the sample consisted of individuals who filled a prescription for a specialty drug between 2013 and 2017. Based on existing literature, we focused on 16 drugs within the 4 specialty drug classes with the highest per capita annual spending in the United States: those for RA, MS, cancer, and hepatitis C.^6,10,23 Due to the focus on outpatient drug use, physician-administered therapies for RA and cancer were not examined.

The primary outcome of interest was monthly OOP spending. Both databases define OOP spending as the dollar amount paid by the beneficiary, including co-payments, coinsurance, and the deductible, per prescription fill. We divided OOP spending by days supplied and multiplied the daily amount by 30 to create a standardized monthly OOP cost per drug. The main independent variable of interest was whether an individual had coverage through ESI, FFS, or MA.

Statistical Analyses

First, to test for differences in drug choice among enrollees with ESI, FFS, and MA, we conducted t tests for rates of utilization and average OOP costs for each drug within a therapeutic class.

Second, to estimate whether there were differences in OOP spending on specialty drugs among ESI, FFS, and MA enrollees, we conducted multivariate ordinary least squares regressions with fixed effects at the therapeutic class level. We controlled for patient age and sex and included a set of fixed effects: MSA of residence (to account for geographic variation, using a county-MSA crosswalk to identify MSAs for Medicare enrollees), year (to account for unobservable, time-invariant characteristics [ie, changes in drug use because of new clinical studies, the ongoing phase-out of the coverage gap in Medicare, and changes in ESI benefit design]), month (to capture differences in OOP spending over the course of a year because of movement through the benefit phases), and branded drug (to ensure that each drug was being compared with itself across different insurers and results were not affected by differing proportions of drug use across insurers). Robust standard errors clustered at the beneficiary level were used to allow for individual heterogeneity. We conducted several sensitivity analyses to examine the robustness of our results (described later).

RESULTS

Sample Characteristics

The analytic data set included 348,600 individuals and 4,045,564 claims; 80% of the claims were covered by ESI (283,120 enrollees) (Table 1). Among specialty drug users in the Medicare sample, FFS made up the majority of the sample, 63%, compared with 37% enrolled in MA, reflective of national MA enrollment. Additional details regarding the ESI sample are in eAppendix Tables 1 and 2 (eAppendix available at ajmc.com). The gender distribution was similar between ESI and Medicare. The average age difference was 20 years between ESI and Medicare enrollees in the RA, MS, and cancer samples; however, there was only a 13-year age difference for hepatitis C.

Drug-Level Utilization and Spending

There were significant differences (Table 2) in drug choice for RA but not MS, cancer, or hepatitis C. Among RA drugs, the most commonly taken drug by ESI beneficiaries was Humira, by 60% of patients, but the most common drug taken by Medicare beneficiaries was Enbrel, taken by 54% and 53% of FFS and MA beneficiaries, respectively. Despite differences in drug choice, OOP costs were similar within an insurance type. In ESI, OOP costs for the 2 brand name drugs were nearly identical, $103 for Enbrel compared with $102 for Humira, suggesting that OOP cost was not the primary reason for branded drug choice. For FFS patients, average OOP costs were $219 for Enbrel compared with $225 for Humira, whereas for MA patients, average OOP costs for Enbrel were $218 (not statistically different from FFS) compared with $192 for Humira ($40 lower than FFS; P < .01). Compared with Medicare, ESI OOP costs were significantly lower (P < .01) across both drugs.

Among the other 3 disease groups, differences in drug choice were fewer. The majority of MS patients took Copaxone in both ESI and Medicare. For the cancer population, the most commonly used drug was Revlimid (used by 48% of Medicare beneficiaries and 46% of ESI enrollees). Average monthly OOP spending on Revlimid was $111 for ESI patients compared with more than $700 for Medicare ($734 for FFS patients and $757 for MA patients). Finally, for hepatitis C drugs, drug choices were not substantially different between Medicare and ESI, but average OOP costs were higher in Medicare. The most commonly taken drug was Harvoni, with average OOP costs of $448 for ESI patients compared with more than $1900 for Medicare ($1981 for FFS patients and $1995 for MA patients). Given the limited differences in OOP spending among drugs within a given therapeutic class, we conducted subsequent analyses at the therapeutic class level, rather than the drug level.

Therapeutic Class Utilization and Spending

Table 1 compares ESI and Medicare differences in drug utilization. Across all 4 drug classes, Medicare beneficiaries, both FFS and MA, took a similar number of drugs per year (16.1 vs 15.6), about 25% more than the ESI population. Among all classes, a substantial proportion of Medicare enrollees reached catastrophic coverage over the course of the year, with MA patients being more likely to reach the coverage gap and catastrophic coverage than FFS patients (Table 1). Due to the significant difference between FFS and MA beneficiaries in movement through the Part D benefit phases, we continue to separately assess these 2 populations.

The Figure shows average monthly OOP spending by insurance and disease type. There were substantial differences between ESI and Medicare but little variation within ESI (across plan types [eg, PPOs vs HMOs]) or within Medicare (FFS vs MA). Average OOP spending for all Medicare beneficiaries was significantly higher than in ESI, approximately $200 for RA drugs, $400 for MS drugs, $600 for cancer drugs, and $1700 for hepatitis C drugs (Figure), with spending just slightly higher in MA than FFS. OOP spending was consistent across all types of ESI plans for RA, MS, and cancer, about $100 per month, with differences of $30 or less (except for high-deductible health plans [HDHPs]). The 1 exception among ESI plans was for hepatitis C drugs; average spending varied from approximately $200 in point of service (POS) plans to greater than $1000 in HDHPs.

The main finding of this paper can be found in Table 3, which shows adjusted differences in OOP spending by insurance type. The large differences in average OOP spending between Medicare and ESI, seen in the unadjusted data, persist after controlling for beneficiary characteristics. Compared with Medicare FFS (the larger of the 2 Medicare populations), individuals in ESI paid, on average, $108 (P < .01) less per month for RA drugs, $288 (P < .01) less for MS drugs, $504 (P < .01) less for cancer drugs, and $1437 (P < .01) less for hepatitis C drugs. Differences in OOP spending existed between Medicare FFS and MA plans; however, they were smaller in magnitude. Compared with FFS, MA beneficiaries paid, on average, $28 (P < .01) more per month for RA drugs, $36 (P < .01) more for MS drugs, $38 (P < .01) more for cancer drugs, and $75 (P < .01) more for hepatitis C drugs. Full regression results are shown in eAppendix Table 3.

To understand how variation in Medicare OOP spending occurred, we examined differences between FFS and MA by benefit phase. As shown in Table 4, we found large variation in OOP spending during the coverage gap. Relative to FFS, MA beneficiaries paid $117 (P < .01) less for RA drugs, $87 (P < .01) less for MS drugs, $374 (P < .01) less for cancer drugs, and $671 (P < .01) less for hepatitis C drugs. In contrast, MA plans required significantly higher cost sharing than FFS in the initial coverage period (differences ranged from $33 to $106, depending on therapeutic class). During catastrophic coverage, differences were less than $25. It would appear that most of the difference in MA and FFS spending occurs during the donut hole period. Two possible explanations are that MA plans are filling in the donut hole (which has been phased out as of 2020; there is a uniform 25% coinsurance rate in the initial coverage limit and coverage gap phases) and that MA plans have greater use of high, fixed-dollar co-payments instead of coinsurance.²⁴

We also examined variation within ESI plan types. Consistent with the descriptive analysis in the Figure, little variation existed among OOP costs in PPO, HMO, consumer-directed, POS, and comprehensive plans, but there was significantly higher OOP spending in HDHPs (eAppendix Table 4).

Sensitivity Analyses

One concern is that the ESI population may be different from the Medicare population, and so we restricted the sample to enrollees aged 55 to 64 years for ESI and 65 to 75 years for Medicare, allowing for increased homogeneity (eAppendix Table 5). We found that differences in OOP spending by insurance type were consistent with those observed in the full population.

To better understand what was driving observed differences in OOP spending among insurers, we examined annual days supplied to assess whether there were differences in the intensity of drug utilization. As shown in eAppendix Table 6, there were no statistically significant differences in the number of days supplied among ESI, FFS, and MA for MS and hepatitis C drugs and differences of 20 days or less per year (P < .01) for RA and cancer drugs. This suggests minimal differences in utilization.

Finally, we examined gross drug spending, total spending on the prescription claims by all payers, to understand whether insurers were steering patients toward certain drugs. We found that gross drug costs were not statistically different between insurers for MS and cancer drugs and that the differences that existed for RA and hepatitis C drugs were financially small relative to the total cost of the drugs (eAppendix Table 7). Important to note is that these gross costs do not include rebates and so may not be wholly reflective of the actual cost paid by the insurer or employer.

DISCUSSION

Our main finding is that OOP spending for specialty drugs treating RA, MS, cancer, and hepatitis C was significantly higher in Medicare compared with ESI from 2013 to 2017. This was likely due to the standard Medicare Part D benefit structure, which determines OOP spending as a percentage of the drug’s list price rather than a fixed co-payment, which is more common in ESI (although some plans treat specialty drugs differently and can use coinsurance).²⁵ We found that the differences between ESI and Medicare arose primarily during the coverage gap (donut hole) phase. Despite a closing of the coverage gap over the course of our study period, the persistent difference in OOP spending between ESI and Medicare points to fundamental differences in design between the 2 insurance types, with implications for enrollees. Small but statistically significant differences were seen between FFS and MA plans, with the MA plans having slightly higher OOP spending.

Current policy debates in Congress around Medicare Part D reform are focused on limiting OOP spending for Medicare beneficiaries: most importantly, an OOP maximum.^26,27 Although an OOP maximum would not decrease OOP spending per prescription fill, it would slightly reduce spending over the course of a year, reducing OOP differences between Medicare and ESI. This is particularly important because the OOP maximums proposed by Congress occur prior to the coverage gap, so implementing such a policy would eliminate cost sharing for the majority of specialty drug fills. It will be important to continue to monitor ESI plan design because commercial plan changes are reported to be occurring in 2020 and 2021.²⁸

Limitations

This study has several limitations. First, standardizing plan design across plans within FFS, MA, and ESI is difficult given the level of variability in both Medicare and commercial plan design. Second, these insurers may make different trade-offs between medical vs prescription drug spending, which could influence OOP spending. Third, we do not know the prices that FFS, MA, and ESI plans actually pay; some of these effects have been mitigated, but some remain.

A significant limitation is the lack of data on patient assistance programs (PAPs) and coupons. For specialty drugs in particular, PAPs often reimburse some or all of beneficiaries’ OOP costs. Thus, it is possible that individuals have co-payments reported in claims data that are reimbursed outside of their insurance. There are no drug- or individual-level data on these programs, and so no data are available to discern how frequently this is occurring and the amount of OOP burden eased by these programs. The incentive to use PAPs is the same for FFS, MA, and ESI, so it should not bias the results in either direction; however, federal guidance makes the use of PAPs different in Medicare vs ESI, and ESI beneficiaries also have the option to use coupons, which could lower their OOP spending further.

Another limitation is that the MarketScan data disproportionately include claims from large self-insured employers, which tend to offer relatively generous benefits. Thus, estimates of OOP spending may not be representative of the entire employer-sponsored market.

Finally, there is far more clinical variation in cancer than in the other diseases being studied, and these differences could be age-related and thus affect the Medicare-ESI comparisons.

CONCLUSIONS

We compared prescription drug spending between ESI and Medicare and found that OOP spending was substantially higher among Medicare enrollees compared with ESI enrollees, because Medicare is structured for beneficiaries to pay a percentage of a drug’s list price. Higher OOP spending in Medicare is likely to pose a challenge for individuals as they make the transition into Medicare, with potential implications for management of their diseases and, ultimately, health outcomes. Rethinking the design of the Part D benefit given the growth in specialty drugs to move toward fixed costs, rather than a percentage of drug costs, may be necessary to ensure that Medicare beneficiaries receive seamless care as they enter the program.

Author Affiliations: Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health (SP, APS, GFA), Baltimore, MD.

Source of Funding: Supported by grants from the Commonwealth Fund and Arnold Ventures.

Author Disclosures: Dr Sen has received grants from Arnold Ventures. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (SP, APS, GFA); acquisition of data (GFA); analysis and interpretation of data (SP); drafting of the manuscript (SP, APS, GFA); critical revision of the manuscript for important intellectual content (SP, APS, GFA); statistical analysis (SP); obtaining funding (APS, GFA); administrative, technical, or logistic support (APS, GFA); and supervision (APS, GFA).

Address Correspondence to: Sonal Parasrampuria, PhD, MPH, Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, 624 N Broadway, Hampton House 302, Baltimore, MD 21205. Email: sparasr4@jhu.edu.

REFERENCES

1. The Medicare prescription drug program (Part D): status report. In: Report to the Congress: Medicare Payment Policy. Medicare Payment Advisory Commission; 2019:385-425. Accessed March 29, 2019. http://www.medpac.gov/docs/default-source/reports/mar19_medpac_ch14_sec.pdf?sfvrsn=0

2. Medicare Part D specialty tier. CMS. April 7, 2015. Accessed March 29, 2019. https://www.cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovGenIn/Downloads/CY-2016-Specialty-Tier-Methodology.pdf

3. Formulary guidance. CMS. Accessed September 28, 2019. https://www.cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovContra/RxContracting_FormularyGuidance.html

4. IMS Institute for Healthcare Informatics. Medicines use and spending in the US: a review of 2015 and outlook to 2020. Morning Consult. April 2016. Accessed March 29, 2019. https://morningconsult.com/wp-content/uploads/2016/04/IMS-Institute-US-Drug-Spending-2015.pdf

5. 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

6. Schilling B. Specialty drug costs poised to skyrocket but many employers have yet to take note. The Commonwealth Fund. 2012. Accessed April 2, 2019. https://www.commonwealthfund.org/publications/newsletter-article/specialty-drug-costs-poised-skyrocket-many-employers-have-yet-take

7. Freeman J. RA facts: what are the latest statistics on rheumatoid arthritis? Rheumatoid Arthritis Support Network. October 27, 2018. Accessed April 2, 2019. https://www.rheumatoidarthritis.org/ra/facts-and-statistics

8. Multiple sclerosis: overview. Mayo Clinic. Accessed April 2, 2019. https://www.mayoclinic.org/diseases-conditions/multiple-sclerosis/symptoms-causes/syc-20350269

9. Age and cancer risk. National Cancer Institute. Accessed April 2, 2019. https://www.cancer.gov/about-cancer/causes-prevention/risk/age

10. Cubanski J, Koma W, Neuman T. The out-of-pocket cost burden for specialty drugs in Medicare Part D in 2019. Kaiser Family Foundation. February 1, 2019. Accessed April 2, 2019. https://www.kff.org/report-section/the-out-of-pocket-cost-burden-for-specialty-drugs-in-medicare-part-d-in-2019-findings

11. Foulon V, Schöffski P, Wolter P. Patient adherence to oral anticancer drugs: an emerging issue in modern oncology. Acta Clin Belg. 2011;66(2):85-96. doi:10.2143/ACB.66.2.2062525

12. Greer JA, Amoyal N, Nisotel L, et al. A systematic review of adherence to oral antineoplastic therapies. Oncologist. 2016;21(3):354-376. doi:10.1634/theoncologist.2015-0405

13. De Vera MA, Mailman J, Galo JS. Economics of non-adherence to biologic therapies in rheumatoid arthritis. Curr Rheumatol Rep. 2014;16(11):460. doi:10.1007/s11926-014-0460-5

14. Atlas RF. The role of PBMs in implementing the Medicare prescription drug benefit. Health Aff (Millwood). 2004;23(suppl 1):W4-504-W4-515. doi:10.1377/hlthaff.w4.504

15. Specialty drugs and health care costs. Pew Charitable Trusts. November 16, 2015. Updated December 2016. Accessed April 2, 2019. https://www.pewtrusts.org/en/research-and-analysis/fact-sheets/2015/11/specialty-drugs-and-health-care-costs

16. Managing the costs of specialty drugs. Elsevier. 2017. Accessed April 2, 2019. https://www.elsevier.com/clinical-solutions/insights/resources/insights-articles/drug-information/whitepapers/managing-the-costs-of-specialty-drugs

17. Encinosa WE, Bernard D, Dor A. Does prescription drug adherence reduce hospitalizations and costs? the case of diabetes. Adv Health Econ Health Serv Res. 2010;22:151-173. doi:10.1108/s0731-2199(2010)0000022010

18. Brown MT, Bussell JK. Medication adherence: WHO cares? Mayo Clin Proc. 2011;86(4):304-314. doi:10.4065/mcp.2010.0575

19. Claxton G, Rae M, Long M, Damico A, Whitmore H, Foster G. Health benefits in 2016: family premiums rose modestly, and offer rates remained stable. Health Aff (Millwood). 2016;35(10):1908-1917. doi:10.1377/hlthaff.2016.0951

20. Kamal R, Cox C, McDermott D. What are the recent and forecasted trends in prescription drug spending? Peterson-KFF Health System Tracker. February 20, 2019. Accessed March 29, 2019. https://www.healthsystemtracker.org/chart-collection/recent-forecasted-trends-prescription-drug-spending/

21. Burks J, Marshall TS. Ye X. Adherence to disease-modifying therapies and its impact on relapse, health resource utilization, and costs among patients with multiple sclerosis. Clinicoecon Outcomes Res. 2017;9:251-260. doi:10.2147/CEOR.S130334

22. Rae M, McDermott D, Levitt L, Claxton G. Long-term trends in employer-based coverage. Peterson-KFF Health System Tracker. April 3, 2020. Accessed August 7, 2020. https://www.healthsystemtracker.org/brief/long-term-trends-in-employer-based-coverage/#item-start

23. Hoadley J, Cubanski J, Neuman T. It pays to shop: variation in out-of-pocket costs for Medicare Part D enrollees in 2016. Kaiser Family Foundation. December 2015. Accessed May 29, 2019. http://files.kff.org/attachment/Issue-Brief-It-Pays-to-Shop-Variation-in-Out-of-Pocket-Costs-for-Medicare-Part-D-Enrollees-in-2016

24. Cubanski J, Damico A, Neuman T. 10 things to know about Medicare Part D coverage and costs in 2019. Kaiser Family Foundation. June 4, 2019. Accessed February 18, 2020. https://www.kff.org/medicare/issue-brief/10-things-to-know-about-medicare-part-d-coverage-and-costs-in-2019/

25. Claxton G, Rae M, Damico A, Young G, McDermott D, Whitmore H. Health benefits in 2019: premiums inch higher, employers respond to federal policy. Health Aff (Millwood). 2019;38(10):1752-1761. doi:10.1377/hlthaff.2019.01026

26. Elijah E. Cummings Lower Drug Costs Now Act, HR 3, 116th Cong (2019). Accessed August 7, 2020. https://www.congress.gov/bill/116th-congress/house-bill/3

27. Prescription Drug Pricing Reduction Act of 2019, S 2543, 116th Cong (2019). Accessed August 7, 2020. https://www.congress.gov/bill/116th-congress/senate-bill/2543

28. 2019 employer health benefits survey. Kaiser Family Foundation. September 25, 2019. Accessed August 7, 2020. https://www.kff.org/health-costs/report/2019-employer-health-benefits-survey/