Essential health benefits form a cornerstone of the Affordable Care Act. Our study shows that health plans in California and Massachusetts are not fully compliant with state and federal regulations on essential drug benefits.
To examine health plan compliance with essential drug benefit regulations in California and Massachusetts and at the federal level.
Health plan formulary review and analysis.
We analyzed formularies from the 3 largest small group plans in California and Massachusetts, including each state’s benchmark plan. With respect to both federal and state regulations, for each health plan, we examined whether the drug was covered, the designated patient cost sharing tier of the drug, and which conditions of reimbursement were applied to the drug.
Most drugs included in state and federal mandates are covered by both benchmark and non-benchmark plans. However, health plans are not fully compliant with state and federal regulations. Significant differences among plans relate more to cost sharing and conditions of reimbursement, such as prior authorization, step edits, and quantity limits, than to drug coverage.
Because health plans in California and Massachusetts are not fully compliant with state and federal mandates, they will have to adjust their formularies to meet minimum requirements. State policy makers need to balance competing aims of comprehensiveness of coverage and drug affordability. They must consider: (1) choice of benchmark plan —choice of a more generous benchmark plan implies less leverage for negotiating lower prices; and (2) breadth of state mandates which, if they exceed federal mandates, must be paid for by the states.
Am J Manag Care. 2014;20(2):153-158
States have long been laboratories of healthcare reform.1 For example, in lieu of a universal prescription drug benefit for Medicare beneficiaries, in the late 1990s, dozens of states established prescription drug assistance programs.2,3 These programs provided impetus for passage of the Medicare Prescription Drug, Improvement and Modernization Act in 2003. More recently, establishment of the Patient Protection and Affordable Care Act (ACA) reflected ongoing reforms at the state level, and, in particular, Massachusetts.4,5
Under the ACA, states have primary responsibility for the expansion of Medicaid to reduce the numbers of uninsured, and the creation of health insurance exchanges in 2014. A health insurance exchange is a collection of government-regulated healthcare plans from which individuals may purchase health insurance eligible for federal subsidies. These exchanges will be subject to minimum (ie, essential) requirements regarding health benefits design.
There are 3 fundamental questions associated with regulations on essential health benefit design: (1) How will essential benefits be defined? (2) At what level of government—state versus federal—will benefits be regulated? And finally (3) Which benefits can society afford to provide? 6,7 In this paper, we will address the first 2 questions. Our study examines state-level implementation of ACA as it pertains to design of an essential drug benefit. Specifically, our focus will be on ACA implementation by California and Massachusetts. California has the eighth-highest percentage of uninsured in the nation and the largest total number of uninsured at 6.9 million.8,9 The state has long played a major role in supplying and regulating health insurance. Medicaid expansion may be particularly challenging to California, given the complexity of its budgetary situation. Massachusetts has the lowest uninsured rate at 4%, in part because Massachusetts implemented comprehensive healthcare reform in 2006.5 Eliminating potentially duplicative regulations may pose a special challenge to Massachusetts.
State regulators must identify a benchmark plan with which to anchor an essential benefits package. “The Secretary [of HHS] shall ensure that the scope of the essential health benefits is equal to the scope of benefits provided under a typical employer plan, as determined by the Secretary.” 10 The latest bulletin from the Department of Health and Human Services (HHS) implies broader coverage of prescription drugs, as it would require coverage of either at least 1 drug in each therapeutic class as defined by the United States Pharmacopeia (USP), or the number of drugs that the benchmark plan offers—whichever is more.11,12 Therefore, benefits included in the benchmark health plans become the de facto essential health benefits package. Accordingly, the benchmark chosen by the state directly impacts what must be covered on outpatient formularies in the nongroup and fully insured small-group markets.
California recommended Kaiser Foundation Health Plan Health Maintenance Organization as its benchmark plan.11 Massachusetts recommended Blue Cross and Blue Shield of Massachusetts Health Maintenance Organization Blue.11 Twenty-five states submitted essential health benefit benchmark plans. The other 25 did not make a selection. HHS assigned the state’s “largest small-group plan.” At least 30 states have Blues plans as benchmarks, while 2 have Kaiser plans as benchmarks.
In this paper, we identify differences in state regulations in California and Massachusetts as they relate to essential pharmaceutical benefits (ie, which drugs health plans are required to cover). Subsequently, we assess whether benchmark and other plans are in compliance with specific state regulations or mandates on prescription drug coverage. We then examine state implementation of ACA provisions as they relate to essential pharmaceutical benefits. In this context, we analyze plan coverage of drugs in therapeutic categories and pharmacologic classes included in the USP model formulary guidelines for outpatient drug coverage under Medicare Part D. Currently, the closest thing federal policy makers have to an evidence-based approach to a therapeutic classification system is the set of USP model formulary guidelines. We will determine whether plans meet the 1-per-class minimum requirement. Further, we will assess coverage of drugs in the 6 so-called protected drug classes, in which “all or substantially all” drugs must be offered to Medicare Part D beneficiaries. Protected class regulations are not included in current essential health benefit regulations. However, there is concern among some stakeholders— patient advocacy groups, physicians, the biopharmaceutical industry—that there will be inadequate access to what are perceived of as necessary medications if the protected class provision is not instituted.13
In order to examine health plan compliance with state and federal regulations on essential drug benefits, we analyzed formularies from the 3 largest small-group plans from each state; 1 of each set of 3 is the state’s selected benchmark plan. We retrieved data directly from the health plans. Plans sometimes devise multiple formularies to suit different beneficiary needs and preferences. Per plan, we examined the formulary associated with the largest number of covered lives.
To compile as comprehensive a drug list as possible, we identified therapeutic categories and pharmacologic classes of drugs contained in the USP Model Guidelines.12 The guidelines include generic names for some (not all) drugs in each category and class. We supplemented our tally of drugs by analyzing the Blue Cross Blue Shield’s Federal Employee Plan formulary.14 The total number of drugs we reviewed coverage of was 310, across 10 broad therapeutic categories: 22 anticonvulsants, 24 antidepressants, 123 antineoplastics, 19 antipsychotics, 30 antiretrovirals, 33 immunosuppressants, 34 antidiabetics, 12 contraceptives, 3 smoking cessation products, and 10 calcium regulators.
To determine coverage across the 6 plan formularies, we compared numbers of drugs on the most recently updated formulary from each plan with the Federal Employee Plan formulary as well as the USP Model Guidelines. With respect to both federal and state regulations, for each health plan, we examined whether the drug was covered or not, the designated cost-sharing tier of the drug, and whether the drug was tagged with conditions of reimbursement, which include prior authorization (PA), step edits (SEs), and quantity limits (QLs). If a drug requires prior authorization, a patient’s healthcare provider must request and receive approval from the health plan before the drug may be prescribed and part of the drug cost may be reimbursed. Step edits refer to reimbursement of a more costly medication only after a less costly alternative has been tried. Quantity limits are restrictions on the number of doses per prescription, or the number of prescriptions ordered. We determined coverage under the outpatient pharmacy benefit. If a drug is subsumed under the medical benefit (physicianadministered) we presumed it was covered, as it would be in Medicare Part B, which operates without a formulary. Finally, health plans will list variations of the same drug (eg, brand and generic). For the purpose of this study, we only counted 1 placement of a drug, and we selected the placement in the lowest cost-sharing tier.
First, we determined plan compliance with the 1 drug per class requirement. Second, we examined coverage of drugs in Part D “protected classes.” The protected classes pertain to drugs in 6 broad therapeutic categories: antidepressant, antipsychotic, anticonvulsant, immunosuppressant, antiretroviral, and antineoplastic. See at www.ajmc.com, for more detail on which categories and classes we analyzed. Third, we reviewed compliance with state mandates in California and Massachusetts: California and Massachusetts state formulary requirements are similar with respect to the following therapeutic classes15: a) inclusion of all pharmaceutical treatments for mental diseases, such as bipolar disorder, and schizophrenia; b) inclusion of all diabetes medications; c) inclusion of all oral contraceptives. California also requires: d) inclusion of all smoking cessation products and e) inclusion of all osteoporosis drugs.
Each plan designs formularies using a different therapeutic classification system. Due to the inconsistent therapeutic classification systems adopted by health plans, we had to carefully piece together the data to inform our comparative analysis. While the USP Model Guidelines have separate sections for each of the 10 categories, including pharmacologic classes (except for anticonvulsants), each plan used its own classification system. The systems employed by Harvard Pilgrim and Tufts Health Plan corresponded the most with the USP Model Guidelines, with each category listed separately as well as a designation of classes. Individual drugs within each class were listed in order from lowest to highest cost-sharing tier, and alphabetically within each tier. By contrast, Kaiser and Anthem did not include classes within each category, nor did they organize and list drugs by tier. Drugs were combined and listed alphabetically under a general rubric. For example, in the case of Anthem, anticonvulsants, antidepressants, and antipsychotics were all listed under Autonomic and Central Nervous System Medications. Both Blue Cross and Blue Shield (BCBS) of Massachusetts and Anthem combined antineoplastics and immunosuppressants into 1 category, while the other plans did not. California Public Employees' Retirement System (CalPERS) did not use a therapeutic classification system; drugs were listed alphabetically. In addition to these classification discrepancies, drugs included in one of the USP categories would sometimes be listed in a completely different category in a plan formulary. For these reasons, we had to look up each drug under each formulary to check whether or not it was covered.
RESULTSTable 1shows that across all health plans, coverage is fairly uniform regarding the so-called protected classes. However, except for HIV/AIDS medications (antiretrovirals), coverage is not 100%. Furthermore, 1 Massachusetts plan only covered 86% of anticonvulsants, while 1 California plan only covered 84% of immunosuppressants. There is significant variation with regard to patient cost sharing. With the exceptions of anticonvulsants, antidepressants, and smoking cessation products, for all other categories we observed a range of at least 2 tiers between plans with the lowest and highest median cost-sharing tiers.
Conditions of reimbursement are used as tools to manage, and, in some cases, restrict drug utilization.16 shows the percent of drugs with conditions of reimbursement across the therapeutic categories we analyzed. Of the conditions employed, 75% were prior authorization, 18% quantity limits, and 7% step edits. For oral contraceptives, none of the plans instituted conditions of reimbursement, while for antiretrovirals, only 1 plan employed conditions of reimbursement. At the same time, for immunosuppressants, we observed that 3 plans used conditions of reimbursement for over 50% of drugs in the category.
For physician-administered drugs (Medicare Part B), Medicare beneficiaries pay 20% coinsurance. Non-Medicare plans will often place physician-administered drugs in so-called specialty tiers—with relatively high coinsurance, often higher than 20% of the total cost of the drug. shows the cost-sharing tiers employed by the health plans we analyzed. In the antineoplastic category, the difference between a copayment of between $25 and $50 per prescription and coinsurance of 25% of the cost of a drug is, in most cases, at least several hundred dollars.
All plans consistently covered at least 1 drug per pharmacologic class. There were was 1 notable exception—the class of nonsedating antihistamines. Most of these medications are now available over the counter, which may explain why none of the plans covered nonsedating antihistamines. Looking at compliance with state mandates by therapeutic class we observe in that in California, with the exception of emergency contraceptives, all oral contraceptives are covered by all 3 plans, all smoking cessation products are covered by all 3 plans, and all osteoporosis products (calcium regulators) are covered by all 3 plans. In Massachusetts and California, 5% of antidepressants were not covered by Massachusetts plans and 3% of antipsychotics are not covered. Three percent (3%) of antidepressants were not covered by California plans, and 3% of antipsychotics. Across both Massachusetts and California plans, 6% of antidiabetic medications were not covered. Half of the plans did not comply with 100% coverage of diabetes medications.
We examined 2 of the 50 states in the United States. This limits the generalizability of our findings. Nevertheless, all states must comply with essential health benefits regulations. Each has the autonomy to do so on its own terms. We consciously selected Massachusetts and California. We chose Massachusetts because it provided a blueprint for healthcare reform at the national level. As such, it is important to see how Massachusetts plans handle essential health benefits regulations at both the state and federal levels. We selected California because it is the state with the largest number of uninsured, which implies that its expansion of Medicaid will be greater than that of any other state.
An overarching goal of federal healthcare reform is to provide affordable, comprehensive access to prescription drugs to the currently un- or underinsured. Our focus in this paper is the comprehensiveness of the prescription drug benefit. This objective may be harder to achieve if health plan options have gaps in minimum coverage. Therefore, requiring that health plans cover certain drug benefits may ensure that the coverage available through state exchanges provides sufficiently comprehensive access.
Critics contend mandates drive up health insurance costs, while advocates assert that mandates ensure access to essential healthcare. The key question for policy makers is how well an essential health benefits approach balances consumers’ desire for an affordable and comprehensive benefit package, while offering flexibility with respect to formulary benefit design and hence leverage for competitive pricing. Policy is invariably a balancing act between these competing aims. Therefore, state policy makers must consider: (1) choice of an appropriate benchmark plan—a more generous benchmark implies less leverage to negotiate lower prices and (2) breadth of state mandates which, if they exceed coverage contained in federal mandates, must be paid for by the states.
There may be fear in biopharmaceutical industry circles that plans gravitate toward a minimum level of coverage if minimum standards are established, such as the 1-per-class requirement. Such fear appears to be unfounded as all plans cover multiple drugs per class, the 1 exception being nonsedating antihistamines.
It appears from our findings, however, that there are inconsistencies in coverage, both with respect to federal and state requirements. And, despite not being a federal requirement— except for Medicare Part D beneficiaries—there are disparities in terms of access to pharmaceuticals in the 6 protected classes. Policy makers must carefully consider which drugs and classes are “essential.” Certain drugs may be deemed essential from a health outcomes perspective. For example, because of patient intervariability in terms of responses to different drugs in the same class, offering the broadest possible number of therapeutic options in certain categories like HIV/AIDS and depression may be considered essential. Furthermore, from a fairness perspective, providing access to the broadest possible number of therapeutic options in disease areas such as cancer and other life-threatening conditions for which there are few alternatives available may be considered essential.19
Coverage is 1 access dimension, but equally important are patient cost sharing and imposition of conditions of reimbursement. In some cases, high patient cost sharing and use of restrictions such as prior authorization may be justified—for example, when the drug in question provides little or no additional value when compared with other drugs in its class, or if there is a legitimate safety concern, perhaps in the form of a US Food and Drug Administration black box warning or alert that appears on package inserts for certain drugs. By contrast, in cases of high-value pharmaceuticals, low patient cost sharing may be warranted and few, if any, restrictions such as prior authorization, step edits, and quantity limits, would be imposed.
Here, principles of value-based insurance design would apply. In other words, designing a formulary in which coverage is a function of the value of a pharmaceutical, measured in terms of health outcomes.20 Consequently, this would mitigate the possible adverse health effects of high out-of-pocket costs. In our study, however, for the same drugs and classes of drugs, we observed significant variance across plans in the use of cost sharing as well as conditions of reimbursement, which would suggest that plans are not adopting a coherent strategy to designing formularies. Moreover, there are discrepancies in the use of therapeutic classification systems, which can make it harder to for patients, providers, and policy makers to meaningfully compare coverage between plans.Author Affiliations: Tufts Center for the Study of Drug Development, School of Medicine, Boston, MA (JPC, AF, MV).
Funding Source: None reported.
Author Disclosures: The authors (JPC, AF, MV) 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 (JPC, AF); acquisition of data (JPC, AF); analysis and interpretation of data (JPC, AF); drafting of the manuscript (JPC, AF); critical revision of the manuscript for important intellectual content (JPC, AF); statistical analysis (JPC, AF); supervision (JPC, AF).
Address correspondence to: Joshua Cohen, PhD, Tufts Center for the Center of Drug Development, School of Medicine, 75 Kneeland St, Boston, MA 02111. E-mail: firstname.lastname@example.org. Kaiser Family Foundation. States moving toward comprehensive health care reform. http://www.kff.org/uninsured/upload/State-Health- Reform1.pdf. Published 2009. Accessed April 10, 2013.
2. Antos JR. Ensuring access to affordable drug coverage in Medicare. Health Care Financ Rev. 2005-2006;27(2):103-112.
3. Cohen J. State-based versus federal-based approaches to reducing the Medicare pharmaceutical coverage gap. Int J Health Serv. 2003; 33(2):345-358.
4. Long S, Stockley K. Sustaining health reform in a recession: an update on Massachusetts as of fall 2009. Health Aff (Millwood). 2010; 29(6):1234-1241.
5. Seifert R, Cohen A. Re-forming reform: what the Patient Protection and Affordable Care Act means for Massachusetts. http://masshealthpolicyforum. brandeis.edu/forums/Documents/IssueBrief_ReportFINAL. pdf Published 2010. Accessed April 21, 2013.
6. Bergthold L. Benefit design choices under managed competition. Health Aff (Millwood). 1993;12(suppl):99-109.
7. Department of Health and Human Services (HHS) informational bulletin: essential health benefits. http://www.healthcare.gov/news/ factsheets/2011/12/essential-health-benefits12162011a.html. Published December 2011. Accessed March 1, 2013.
8. California Health Care Foundation. California’s uninsured. http:// www.chcf.org/~/media/MEDIA%20LIBRARY%20Files/PDF/C/PDF%20 CaliforniaUninsured2011.pdf. Published December 2011. Accessed April 13, 2013.
9. Kaiser Family Foundation. Kaiser Fact Sheet: Health Coverage and Uninsured. http://www.statehealthfacts.org/comparecat.jsp?cat=3. Published 2012. Accessed April 19, 2013.
10. Essential Health Benefits: List of the Largest Three Small Group Products by State, Center for Consumer Information and Insurance Oversight Centers for Medicare & Medicaid Services. http://cciio.cms. gov/resources/files/largest-smgroup-products-7-2-2012.pdf. Published July 2012. Accessed November 10, 2012.
11. Essential Health Benefit (EHB) Benchmark Plans, as of January 3, 2013. http://www.statehealthfacts.org/comparereport.jsp?rep=156 &cat=17. Published February 2013. Accessed March 21, 2013.
12. United States Pharmacopeia model formulary guidelines for Medicare Part D. Version 5.0. http://www.usp.org/sites/default/files/usp_pdf/ EN/healthcareProfessionals/2011-03-11uspmgwithexampledrugs.pdf. Published September 2011. Accessed April 11, 2013.
13. Cancer Leadership Council. http://www.cancerleadership.org/ policy/other/pdf/Cancer%20Leadership%20Comments%20on%20Essential% 20Health%20Benefits%20Bulletin%20January%2031,%202012. pdf. Published January 2013. Accessed February 21, 2013.
14. FEP Rx Drug Formulary Standard Option, effective October 2012. http://www.caremark.com/portal/asset/z6500_drug_list607.pdf. Published October 2012. Accessed February 2, 2013.
15. National Council of State Legislatures — state insurance mandates. http://www.ncsl.org/issues-research/health/state-ins-mandates-andaca- essential-benefits.aspx. Published February 20, 2013. Accessed April 30, 2013.
16. Margolis J, Cao Z, Onukwugha E, Sanchez R, Alvir J, Joshi AV, Mullins CD. Healthcare utilization and cost effects of prior authorization for pregabalin in commercial health plans. Am J Manag Care. 2010;16(6):447-456.
17. Institute of Medicine. Essential health benefits: Balancing coverage and cost. http://www.iom.edu/~/media/Files/Report%20Files/2011/ Essential-Health-Benefits-Balancing-Coverage-and-Cost/essentialhealthbenefitsreportbrief4. pdf. Publication October 2011. Accessed March 21, 2013.
18. Essential Health Benefits — White Paper. National Health Council. http://www.nationalhealthcouncil.org/NHC_Files/files/EHB_WhitePaper. pdf. Published September 2010. Accessed April 1, 2013.
19. Daniels N, Sabin J. Setting fair limits: can we learn to share resources for health? Health Aff (Millwood). 2005;24(2):343-344.
20. Chernew M, Rosen A, Fendrick M. Value-based insurance design. Health Aff (Millwood). 2007;26(2):w195-w203. n