Assessment of Variation in State Regulation of Generic Drug and Interchangeable Biologic Substitutions

Key Points

Question  How do US state drug product selection laws vary with regard to how pharmacists can substitute prescriptions for brand-name drugs with small-molecule generic drug or interchangeable biologic equivalents?

Findings  In this cross-sectional study of all 50 states and Washington, DC, substantial variation was noted in generic substitution rules for small-molecule drugs, including whether substitution was mandatory or permissive, whether patient consent and notification were required before and after substitution, and whether pharmacists were protected from liability for substitution. Almost all states imposed heightened requirements for interchangeable biologic substitution.

Meaning  The findings of this study suggest a need for optimizing state drug product selection laws to improve medication adherence and reduce drug spending.

Importance  Brand-name drugs, including biologics, have been the primary source of increasing prescription drug spending in the US. Each state has drug product selection laws that regulate whether and how pharmacists can substitute prescriptions for brand-name drugs with more affordable equivalents, either small-molecule generic drugs or interchangeable biologics, but the details of these laws can vary.

Objective  To examine the variation in state drug product selection laws with regard to factors that may affect which version of a drug is dispensed.

Design, Setting, and Participants  A cross-sectional analysis was performed, using a legal database, to obtain information on state laws of all states plus Washington, DC, as they existed on September 1, 2019.

Exposures  Whether substitution was mandatory or permissive, patient consent was needed prior to substitution, patient notification of substitution was required independent of the drug’s packaging, and/or pharmacists were protected from special risk of liability for substitution.

Main Outcomes and Measures  For small-molecule and biologic drugs, descriptive statistics were generated for the 4 exposure variables. In addition, for small-molecule drugs, a generic substitution score with a maximum of 1 point was assigned for each exposure variable (range, 0-4 points), with higher scores indicating regulatory requirements limiting substitution.

Results  This cross-sectional analysis of the generic drug substitution regulations in the 50 US states and Washington, DC, found that for small-molecule drugs, 19 states required pharmacists to perform generic substitution; 7 states and Washington, DC, required patient consent; 31 states and Washington, DC, mandated patient notification independent of the drug’s packaging, and 24 states did not explicitly protect pharmacists from greater liability. Nine states and Washington, DC, had a generic substitution score for small-molecule drugs of 3 or higher, and 45 states had more stringent requirements for interchangeable biologic substitution, most commonly mandatory physician notification.

Conclusions and Relevance  The findings of this study suggest that there is a need for optimizing state drug product selection laws to promote generic and interchangeable biologic substitution, which may help improve medication adherence and reduce drug spending.

The US spends more per capita on prescription drugs than any other country in the world.¹ Brand-name drugs account for most of this spending, making up more than 70% of all prescription drug costs but only about 10% of prescriptions.² Over the past 3 decades, the main counterweight to high brand-name drug spending has been small-molecule generic drugs, which emerge after the brand-name drug’s market exclusivity period and are designated by the US Food and Drug Administration (FDA) as being bioequivalent. Between 2007 and 2016, generic drugs saved the US health care system $1.7 trillion.³ Availability of low-cost generic drugs may also promote medication adherence among patients⁴ and lead to improved clinical outcomes.^5-7

State laws are key to ensuring reductions in spending and improvements in adherence from generic drugs. In the US, pharmacy practices are managed at the state level, and each state has a set of drug product selection laws that regulate how pharmacists dispense drugs.⁸ The operation of drug product selection laws helps to determine the efficiency of the generic drug market. For example, when such laws allow pharmacists to automatically substitute a generic drug for its brand-name version, generic manufacturers are incentivized to earn market share by offering their products at as low a cost as possible and do not have to spend resources on marketing or other efforts to differentiate themselves from other manufacturers. This competition results in drug prices decreasing to close to the cost of production after a sufficient number of interchangeable generic drugs becomes available.⁹

However, state drug product selection laws can vary in important respects, such as whether the substitution is mandatory or permissive and whether patient consent is required. An investigation of Medicaid outpatient dispensing of simvastatin (Zocor) between 2006 and 2007 by Shrank et al¹⁰ found that states requiring patient consent had 25% lower generic substitution rates. By undermining efficient dispensing of generic drugs, such variation in state drug product selection laws lead to unnecessary use of expensive brand-name versions. The investigators estimated that eliminating the patient consent requirement would have led to more than $100 million in savings for just 3 medications—atorvastatin (Lipitor), clopidogrel (Plavix), and olanzapine (Zyprexa)—in the year following loss of market exclusivity.

Despite their widespread use, generic medications have remained underused in many clinical contexts, including cardiovascular disease management,¹¹ combination products,¹² and oral contraceptive medications.¹³ In addition, in the past decade, costly biologic drugs have become more common in the drug market. Today, biologics represent only 2% of dispensed medications but nearly 40% of spending.¹⁴ Biologic drugs had been excluded from generic-style competition until 2010, when the Biologics Price Competition and Innovation Act created a regulatory process for approving follow-on biologic drugs based on the generic drug approval system.¹⁵ Two types of approvals were authorized: biosimilars, which have no clinically meaningful differences in safety, purity, and potency to the originator biologic drug, and interchangeable biologics, which are biosimilar agents that have met additional testing standards demonstrating that they may be safely substituted for the originator biologic without the intervention of the prescribing physician.¹⁶ State laws do not allow pharmacist substitution of biosimilar drugs and vary with their approach to interchangeable biologic substitution.¹⁷ Although the FDA has yet to designate any biosimilar drugs as interchangeable,¹⁸ the first interchangeable biologics may appear soon, with the FDA having begun regulating insulin products as biologics as of March 2020.¹⁹ Efficient state drug product selection laws could then be critical to lowering spending on biologic drugs and ensuring optimal use of products, such as interchangeable insulin.^20,21 We sought to understand the current landscape of drug product selection laws in the US, including how different states approach interchangeable biologic substitution.

We searched the legal database Westlaw Edge (Thomson Reuters)²² for all statutes and regulations in each state and Washington, DC, related to pharmacist dispensing of prescription medications between March 1 and March 30, 2019. We used the search terms drug, generic, biologic, biosimilar, follow-on biologic, product, selection, and substitution and reviewed neighboring sections of retrieved statutes and regulations using the database's table of contents tool to develop the range of each state's drug product selection laws. We reassessed the currency of these laws in October 2019, capturing any updates as of September 1, 2019.

We instituted an iterative process to develop the scheme for data extraction. An investigator (A.S.) first reviewed drug product selection laws in 10 states to identify factors that may affect whether a brand-name or generic/interchangeable biologic version of a drug was dispensed. This review resulted in the identification of the following variables: (1) duty to substitute (whether substitution was mandated or permissive), (2) notification (whether patient notification of the substitution independent of the drug’s packaging was required), (3) consent (whether patient consent for the substitution was required, patient consent was not specifically required but refusal was explicitly allowed, or no such reference was made), and (4) liability (whether reference was made that pharmacists incurred no greater risk of liability for substitution than if the brand-name product were dispensed, specific mention was made of possibly increased liability, or no such reference was made). Two of us (C.A.S. and V.L.W.d.W.) then independently reviewed the laws in 50 states and Washington, DC, and extracted those variables as they applied to generic drugs and interchangeable biologics between April 1, 2019, and November 30, 2019. Disagreements were resolved by consensus.

Variation in these laws was documented using descriptive statistics. For small-molecule drugs, this included developing a generic substitution score for each state and Washington, DC, with a maximum of 1 point assigned for each variable that indicated a regulatory requirement making generic substitution more restrictive (Table 1). For variables with 2 options, a score of 0 or 1 was assigned. For variables with 3 possibilities, a score of 0, 0.5, or 1 was assigned. A lower total score indicates a state regulatory framework more favorable to generic substitution of small-molecule drugs. Differences between requirements for generic and interchangeable biologic substitution within each state were then examined. Institutional review board approval was not required because the study used only public data on state laws and not human participants.

We found substantial variation in drug product selection laws (Figure 1). With regard to small-molecule drugs, less than half of the states (n = 19) mandated generic substitution by pharmacists when generic products were available (noting that pharmacists “shall” substitute). The remaining 31 states and Washington, DC, permitted but did not require substitution (noting that pharmacists “may” substitute); of these jurisdictions, 4 required pharmacies to post notice for patients of the possibility of substitution.

Seven states and Washington, DC, required that patients consent to substitution, while 23 states noted a right of patients to refuse substitution without requiring that they consent. If substitution was performed, 31 states and Washington, DC, mandated that patients be notified of the action independent of the drug’s packaging.

Most states and Washington, DC (n = 27) explicitly protected pharmacists from greater liability for substitution. By contrast, 23 states did not, while 1 state (Connecticut) noted the possibility of increased liability (“Neither the failure to instruct by the purchaser...nor the fact that a sign has been posted...shall be a defense on the part of a pharmacist against a suit brought by any such purchaser”).²³

Figure 2 shows the generic substitution score assigned to each state and Washington, DC. Thirteen states (Arizona, Illinois, Kentucky, Massachusetts, New Jersey, New York, North Carolina, Oklahoma, Rhode Island, Tennessee, Washington, Wisconsin, and Wyoming) had scores between 0 and 1, reflecting few restrictions for generic substitution of small-molecule drugs. By contrast, 9 states (Alaska, Arkansas, Connecticut, Iowa, Louisiana, South Carolina, Texas, Utah, and Virginia) and Washington, DC, had scores of 3 or higher, indicating greater barriers to generic substitution.

Ninety percent of the states (n = 45) had more stringent requirements for substitution of interchangeable biologics than for substitution of generic drugs (Table 2). The most common heightened requirement—enacted by 45 states—was a mandate that physicians be notified of substitution. For example, California stipulated that within 5 days of dispensing, the pharmacist must communicate the substitution through “an entry that can be electronically accessed by the prescriber.”²⁴ Nine states (Alabama, Arizona, Illinois, Massachusetts, Missouri, Nebraska, New Mexico, Oregon, and Rhode Island) did not have patient notification requirements for generic substitution but did have them for interchangeable biologic substitution. Six states (Florida, Indiana, Massachusetts, New Jersey, Pennsylvania, and Tennessee) that mandated generic drug substitution made interchangeable biologic substitution permissive.

Our findings highlight considerable room for optimizing state drug product selection laws to promote generic and interchangeable biologic substitution. Nearly two-thirds of states did not require pharmacists to perform substitution with an FDA-approved generic when physicians prescribe a brand-name small-molecule drug. In addition, 15% of states required patient consent. When interchangeable biosimilar drugs reach the US market, possibly as soon as late 2020, they will encounter important limits on substitution.

Prior analyses suggest that these state policy choices are important. In a national survey conducted to understand beliefs and practices related to narrow therapeutic index drug substitution, the odds of pharmacists in mandatory substitution states with a patient consent requirement not substituting initial prescriptions were nearly twice (odds ratio, 1.9; 95% CI, 1.1-3.3) those of pharmacists in mandatory substitution states without a patient consent requirement.²⁵ In the Medicaid study by Shrank et al¹⁰ of simvastatin, substitution rates in 2006-2007 were nearly 20% higher in states with mandatory vs permissive substitution policies. Our data show that more than one-third of states still require pharmacists to notify patients of substitutions independent of the package labeling, and almost half did not explicitly protect pharmacists from liability for substitution. Research is needed to quantify the effects of these variables on pharmacists’ current practices and patient outcomes.

Overall, the variation we observed highlights an opportunity not only to reduce health system spending, but also to improve population health. Generic drugs are clinically equivalent to their brand-name counterparts and have been associated with improved medication adherence and health outcomes. For example, a study of 6 classes of medications prescribed for long-term use between 2001 and 2003 reported 13% greater adherence among patients who initiated treatment with a generic rather than a brand-name drug.⁵ More recently, researchers found a 6% absolute increase in the proportion of days covered (a measure of adherence) and an 8% relative reduction in all-cause mortality and hospitalization for acute coronary syndrome or stroke among initiators of generic instead of brand-name statins.⁶ Over time, as payers have shifted more drugs onto coinsurance tiering and more costs onto patients through higher deductible plans, the individual burden of higher drug prices has only increased.^26-28

Optimizing state laws to facilitate generic drug substitution as the default option is an important lever to increase medication adherence and reduce excess drug spending. For the rare patients who require a brand-name drug when an interchangeable generic is available, all states allow prescribers to indicate dispense as written on prescriptions, ensuring that a brand-name drug is dispensed.²⁹ In addition, some states have created special substitution rules for some drug classes, such as immunomodulators used in patients who have undergone transplantation or are receiving antiepileptic drugs, based on concerns that some clinicians and patients have about generic interchangeability.²⁴ In the case of antiepileptics, well-controlled observational studies have not shown increased risk of adverse patient outcomes from generic substitution.^30,31

Of special concern are the heightened requirements most states have imposed on interchangeable biosimilar substitution, in particular, the permissive substitution standard. The adoption of these laws reflects in part well-documented efforts of originator biologic manufacturers with state legislators,³² which generally have looser lobbying and transparency standards than the federal government.³³ When interchangeable biosimilar agents enter the market, it will be important to assess the extent to which heightened substitution requirements suppress their uptake.

The requirement for physician authorization for substitution of refills may also have important ramifications. Because most biologics are intended for long-term use, many patients with the indicated condition will be receiving the original biologic when an interchangeable biosimilar reaches the market. Drug therapy in these patients will need to be switched—or at least be capable of being switched—to achieve the full cost-savings potential of interchangeable biosimilar agents.³⁴

In addition, the requirement in nearly all states that pharmacists notify physicians about interchangeable biologic substitution may pose administrative challenges. For example, the logistics required to implement such a notification and acknowledgment process may vary for different electronic medical record systems and may introduce workflow challenges that pose an additional barrier to widespread use. At a minimum, states may need to be flexible about the mode and timing of this communication.

The increased complexity of biologic drugs, such as monoclonal antibodies, clotting factors, and hormones, is often presented as the justification for these heightened requirements. However, the experience of other nations that found higher biosimilar uptake compared with the US offers substantial evidence of the safety of biosimilars³⁵ and their ability to yield cost savings.³⁶

To best facilitate generic and interchangeable biologic use, optimization of state drug product selection laws should be coupled with educational outreach to prescribers and patients. The US Department of Health and Human Services has endorsed such efforts,³⁷ which can help improve public confidence in the safety and efficacy of generic drugs and interchangeable biologics. Academic detailing, which uses an interactive individualized approach to convey to physicians the most current, evidence-based information on optimal care decisions, is one strategy that has been shown in randomized clinical trials to improve prescribing patient outcomes and cost containment.^38-40 Among other benefits, well-designed and deployed educational interventions may decrease the likelihood of physicians writing dispense-as-written prescriptions and increase their likelihood of using nonproprietary names in their clinical practice.

Limitations of our investigation warrant discussion. Although the examined variables were determined after review of literature findings and laws in a subset of states, there may have been relevant points of variation that we did not identify. It is also possible that different pharmacy state boards interpret and enforce similar laws differently. In addition, insurance management tools, such as tiering and prior authorization requirements, play a role in modifying the effect of state drug product selection laws but were not assessed in this investigation.

State laws governing pharmacist dispensing practices represent an important policy lever to contain increasing drug costs and improve health outcomes. The substantial variation in these laws provides opportunities to further evaluate and implement policies that promote generic and interchangeable biologic substitution.

Accepted for Publication: June 19, 2020.

Published Online: August 31, 2020. doi:10.1001/jamainternmed.2020.3588

Corrections: This article was corrected on November 30, 2020, to fix misclassification of some US states caused by coding errors in the data file; and on March 1, 2021, to fix name abbreviations of 2 states in Figure 2.

Corresponding Author: Ameet Sarpatwari, PhD, JD, Program on Regulation, Therapeutics, and Law, Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, 1620 Tremont St, Ste 3030, Boston, MA 02120 (asarpatwari@bwh.harvard.edu).

Author Contributions: Drs Sacks and Sarpatwari had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Sacks, Kesselheim, Sarpatwari.

Acquisition, analysis, or interpretation of data: Sacks, Van de Wiele, Fulchino, Patel, Sarpatwari.

Drafting of the manuscript: Sacks, Van de Wiele, Fulchino, Patel, Sarpatwari.

Critical revision of the manuscript for important intellectual content: Van de Wiele, Kesselheim, Sarpatwari.

Statistical analysis: Van de Wiele, Sarpatwari.

Obtained funding: Kesselheim, Sarpatwari.

Administrative, technical, or material support: Van de Wiele, Patel, Sarpatwari.

Supervision: Kesselheim, Sarpatwari.

Conflict of Interest Disclosures: Dr Sacks reported a grant from the Carney Family Foundation during the conduct of the study. Dr Sarpatwari reported a grant from the Anthem Public Policy Institute outside the submitted work. No other disclosures were reported.

Funding/Support: Funding for the study was provided by the Robert Wood Johnson Foundation, Public Health Law Research Program (72231), Arnold Ventures, and the Harvard-MIT Center for Regulatory Science. The Association for Accessible Medicines provided a grant to Brigham and Women's Hospital that partially supported Mr Van de Wiele's fellowship.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.