Cannabis Conundrum: Demonstrating “Accepted Medical Use” in a Restricted Research Environment


By Delia Ann Deschaine and Megan Robertson

U.S. cannabis businesses operate in an increasingly complex legal and regulatory environment.  Many, if not most, of those complexities stem from marijuana’s classification in Schedule I of the federal Controlled Substances Act.  That classification hinges on the determination that marijuana lacks a currently accepted medical use in treatment in the United States. Thus, the need for research on the potential therapeutic benefits and risks of marijuana use is critical.

Academic medical centers, hospitals, and other researchers have historically faced challenges in conducting research on cannabis and cannabis-derived material.  While the federal government has taken significant steps to remove or reduce some of these obstacles in recent years, many still remain.  To say the least, it takes a degree of sophistication, and some careful maneuvering, to navigate the research marketplace and source cannabis, or cannabinoids, that are of sufficient quality and quantity for use in research.

In the article below, we explore the historic regulation of marijuana for research in the U.S. and highlight the relevant changes that have improved access to cannabis research material. We also examine the considerations that researchers have to make today, and the ways that federal policy can be improved to create better access, and promote research, that is desperately needed to advance sound cannabis policy.

Historical Constraints on Cannabis Research

In 1970, the Controlled Substances Act (“CSA”), among other things, implemented the 1961 Single Convention on Narcotic Drugs (the “Single Convention”), designating cannabis as a Schedule I controlled substance and setting forth the framework for domestic sourcing of cannabis for research. Article 28 of the Single Convention requires countries party to the treaty that allow cultivation of cannabis to apply a system of controls, and to adopt necessary measures to prevent the misuse and illicit traffic of cannabis.  The controls set forth by the Single Convention include designation of “the areas in which, and the plots of land on which” cultivation of cannabis shall be permitted, and that only registered cultivators shall be authorized to engage in the cultivation. These controls were incorporated into the CSA and placed under the authority of the DEA, and these provisions in both the CSA and the Single Convention remain largely unchanged today.

Importantly, the DEA has only issued one registration for marijuana research, awarded to the University of Mississippi, which has been under contract for marijuana cultivation with the National Institutes of Health (“NIH”) National Institute on Drug Abuse (“NIDA”) since 1968.  The University of Mississippi grows crops every few years based on current and expected demand, and stores the marijuana in bulk, both as plants and purified extracts.  As a result, the University of Mississippi is effectively the sole cultivator of marijuana for research in the U.S.

Until recently, challenges to the domestic control of marijuana for research have largely been unfruitful.  In 2001, a single researcher, Lyle E. Craker, sought to obtain a federal DEA registration to cultivate cannabis for privately funded research.  One of Dr. Craker’s assertions, supported by the Multidisciplinary Association for Psychedelic Studies (“MAPS”), was that the marijuana supplied by NIDA was not of sufficient quality or quantity to meet the needs of researchers, and that pharmaceutical developers cannot reasonably rely on NIDA marijuana in connection to a FDA new drug application. Dr. Craker’s application was ultimately published to the Federal Register for public comment in 2003, and in 2004 DEA denied the application. The DEA indicated that approval of the application would be against the public interest and in violation of international treaty obligations. On appeal, despite an Administrative Law Judge in 2007 recommending approval of Dr. Craker’s application, in 2009 the DEA issued a final denial.[1] In 2013, the U.S. Court of Appeals for the First Circuit upheld that denial, finding the DEA Administrator’s interpretation permissible and findings reasonable and supported by evidence.[2]

In addition to obtaining a DEA registration, researchers used to have to receive approval from the Department of Health and Human Services (“HHS”) to purchase marijuana for research from NIDA.  This approval process was initiated by a 1999 HHS Guidance, which required researchers to submit proposals for clinical trials in order for HHS to “determine whether cannabinoid components of marijuana administered through an alternative delivery system can meet the standards enumerated under the Federal Food, Drug, and Cosmetic Act for commercial marketing of a medical product.” HHS largely denied requests under this review process, with one of the few approvals issued in March 2014 relating to a study of marijuana for symptoms of post-traumatic stress disorder.[3]

This multifaceted, multi-agency review process for research using cannabis created a stalemate in advancing federal cannabis policy, leaving states to resort to their own tools to legalize the drug’s use for medical purposes, and in some states adult-use, despite marijuana remaining a federal Schedule I controlled substance.

A Shift in the Landscape

In June of 2015, due to concerns that the HHS review overlapped with FDA’s investigational new drug application process, the HHS review program was eliminated.[4]  However, the DEA registration framework remains.  In 2016, DEA issued a policy to allow for the additional registration of marijuana cultivators for legitimate research and licit commercial purposes, but in June 2018 the Department of Justice Office of Legal Counsel issued an opinion finding that the policy violated certain obligations of the United States’ under applicable treaties, including the Single Convention.  Specifically, one of the requirements of the Single Convention is that the agency overseeing the marijuana cultivation must purchase and take physical possession of the marijuana crops no later than four months after the end of the harvest, and the 2016 policy did not conform with this requirement.

Most recently, in March of 2020, DEA issued a proposed rule to allow for the registration of additional cultivators of cannabis for research purposes.[5]  The proposed rule expressly only applies to persons seeking authorization (a DEA registration) under the CSA to cultivate “marihuana” as that term is currently defined (i.e., excluding hemp). However, the rule has not been finalized, and the University of Mississippi remains the only registered domestic cultivator of marijuana for research.

In addition, in December 2018, Congress enacted the Agricultural Improvement Act of 2018 (the “2018 Farm Bill”) which, among other things, offered a federal definition of hemp and removed it from the list of Schedule I controlled substances. Hemp is defined as cannabis with less than 0.3% delta-9-tetrahydrocannabinol on a dry weight basis, and is expressly carved out from the CSA’s definition of “marihuana” and THC.

By removing hemp from the list of controlled substances, the 2018 Farm Bill theoretically opened the door for researchers to design and run studies using hemp, but did not expressly address, among other things, how researchers may go about sourcing the hemp. Because, the domestic hemp industry is still nascent, researchers may be unable to readily source hemp that is of sufficient quality for use in clinical research.  As FDA has identified, ensuring batch-to-batch consistency and limiting impurities is critical to ensuring research material is suitable for clinical use. See FDA, Draft Guidance, Cannabis and Cannabis-Derived Compounds: Quality Considerations for Clinical Research (July 2020).  Also, the sole fact that a cannabinoid is hemp-derived is not sufficient to justify its status as non-controlled. The derived material itself must contain less than 0.3% delta-9-tetrahydrocannabinol on a dry weight basis; researchers cannot rely on the THC content of the plant source.

The use of synthetic cannabinoids in research provides its own unique challenges. According to a recent DEA interim final rule, published on August 21, 2020, the new definitions in the 2018 Farm Bill do not apply to synthetically derived cannabinoids because they are not derived from the cannabis plant.

Last, while there is an alternative pathway to the procurement of Schedule I research material through importation, it is less well-known, and still requires a DEA registration.

Moving Forward

Industry has spoken out in support of expanding access to cannabis for research, and recent movement on the federal level suggests FDA and DEA are listening to industry concerns. While there have been steps forward, limitations remain.  On September 11, 2020, the Congressional Research Service (“CRS”) issued a report on The Schedule I Status of Marijuana, which identified four existing gaps in marijuana policy based on the discrepancies between federal and state marijuana laws and regulations.  These gaps include the fact that institutions of higher learning may decline to allow research on campus due to the risk of losing federal funding; and that many researchers are unable to meet the “strict requirements” of the CSA or are seeking to study variations of marijuana that are not lawfully available. Indeed, NIDA only supplies marijuana in one form that is unlike the marijuana sold or used under state-authorized marijuana programs. Researchers, therefore, cannot use NIDA-supplied marijuana to study alternative dosage forms of the drug, or, to understand the risks and benefits of marijuana available in the U.S.

Other policy gaps identified by the CRS report include that financial institutions are reluctant to enter into relationships with state-authorized marijuana businesses due to the federal Schedule I status of marijuana; and individuals are wary of violating federal marijuana laws even when acting in compliance with state laws, as the violation of federal law could impact eligibility for student financial aid, housing and food assistance, gun ownership, visas, and employment.  There are also concerns related to the use of smoked cannabis in research by academic medical centers, as such may implicate a university’s anti-smoking policies.

In sum, researchers must navigate a thicket of requirements, which serve as an impediment to advancing sound cannabis policy, informing the decisions of state legislatures and regulators in designing their medical marijuana programs, the research and development of new medical therapies that contain cannabinoids, and ultimately unbridling U.S. cannabis businesses. To demonstrably reduce these obstacles, the federal government should consider the removal of marijuana from Schedule I of the federal CSA, and create a workable scheme for access to cannabis for research in compliance with the U.S.’ international treaty obligations.


[1]74 Fed. Reg. 2101 (Jan. 14, 2009).

[2]Lyle E. Craker v. DEA, 714 F.3d 17 (1st Cir. 2013).

[3]See, e.g.,PRESS RELEASE: Historic Shift in Federal Policy as HHS Approves Medical Marijuana Study for PTSD, MAPS (Mar. 28, 2014), ; Ariana Eunjung Cha, Marijuana research hampered by access from government and politics, scientists say, The Washington Post (Mar. 21, 2014),.

[4]80 Fed. Reg. 35960 (June 23, 2015).

[5]See 85 Fed. Reg. 16292 (Mar. 23, 2020).


About The Co-author

DELIA A. DESCHAINE is a Member of Epstein Becker Green’s Health Care and Life Sciences practice, in the firm’s Washington, DC, office. Named to the Washington DC Rising Stars list (2018-2020) in the areas of Food & Drugs and Cannabis Law, Ms. Deschaine understands the most significant U.S. Food and Drug Administration (FDA) and U.S. Drug Enforcement Administration (DEA) legal and compliance risks that her clients face and knows how to address them. Her advice is informed by a background in defending clients in government enforcement actions and challenging unlawful government conduct.