WHY INSURANCE SHOULD PAY FOR MEDICAL CANNABIS

Although medical cannabis has been legalized in 33 states and the District of Columbia, it has not yet gained federal acceptance. The Drug Enforcement Administration (DEA) still classifies it with a Schedule I designation, indicating that it has significant risks but no medical benefits. This discrepancy in legality between the state and federal levels has created numerous challenges for patients and physicians, as well as for businesses that must negotiate conflicting rules.

One of the most significant challenges that these conflicting rules create for patients is the lack of insurance coverage for medical cannabis. Payers are understandably reluctant to include benefits for a drug that is still a Schedule I substance at the federal level. Without health insurance coverage, patients must pay out of pocket for treatment that can cost hundreds of dollars per month.

If cannabis is solely a recreational drug with no medical benefits, as the DEA contends, then this policy is logical. Insurance companies should not be compelled to pay for interventions that do not contribute in a meaningful way to health, function, or quality of life. Indeed, asking payers to reimburse for recreational cannabis would be analogous to asking them to cover the costs of tobacco, alcohol, or perhaps heroin, which is another Schedule I drug.

However, there is growing evidence that medical cannabis is indeed beneficial. Several comprehensive metanalyses have found evidence of effectiveness for several indications, including very common symptoms such as chronic pain (Campbell et al. 2001Campbell, F. A., M. R. Tramèr, D.Carroll, D. J. M. Reynolds, R. A.Moore, and H. J. McQuay. 2001. Are cannabinoids an effective and safe treatment option in the management of pain? A qualitative systematic review. BMJ 323(7303): 13. doi: 10.1136/bmj.323.7303.13.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]; Koppel et al. 2014Koppel, B. S., J. C. M. Brust, T. Fife, et al. 2014. Systematic review: Efficacy and safety of medical marijuana in selected neurologic disorders. Neurology 82(17): 1556. doi: 10.1212/WNL.0000000000000363.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]; Lynch and Campbell 2011Lynch, M. E., and F. Campbell. 2011. Cannabinoids for treatment of chronic non-cancer pain; a systematic review of randomized trials. British Journal of Clinical Pharmacology 72(5): 735–744. doi: 10.1111/j.1365-2125.2011.03970.x.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]; Martín-Sánchez et al. 2009Martín-Sánchez, E., T. A. Furukawa, J. Taylor, and J. L. R. Martin. 2009. Systematic review and meta-analysis of cannabis treatment for chronic pain. Pain Medicine 10(8): 1353–1368. doi: 10.1111/j.1526-4637.2009.00703.x.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]; Stockings et al. 2018Stockings, E., G. Campbell, W. D.Hall, et al. 2018. Cannabis and cannabinoids for the treatment of people with chronic noncancer pain conditions: A systematic review and meta-analysis of controlled and observational studies. PAIN 159(10): 1932. doi: 10.1097/j.pain.0000000000001293.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]; Whiting et al. 2015Whiting, P. F., R. F. Wolff, S.Deshpande, et al. 2015. Cannabinoids for medical use: A systematic review and meta-analysis. JAMA 313(24): 2456–2473. doi: 10.1001/jama.2015.6358.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]). More recently, the National Academy of Sciences (2017National Academies of Sciences, Engineering. 2017. The health effects of cannabis and cannabinoids: The current state of evidence and recommendations for research. doi: 10.17226/24625.http://www.nationalacademies.org/hmd/Reports/2017/health-effects-of-cannabis-and-cannabinoids.aspx[Crossref], , [Google Scholar]) has also noted that there is solid evidence supporting the use of medical cannabis in a number of conditions.

This evidence base is not as well developed as it could be, and there are still large gaps in what is known. For instance, some conditions such as posttraumatic stress disorder (PTSD) are qualifying conditions in state laws, but do not have sufficient data yet to support the use of medical cannabis in their treatment. Moreover, many of the studies that have been done are limited by poor quality (Whiting et al. 2015Whiting, P. F., R. F. Wolff, S.Deshpande, et al. 2015. Cannabinoids for medical use: A systematic review and meta-analysis. JAMA 313(24): 2456–2473. doi: 10.1001/jama.2015.6358.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]) or by challenges of adequate blinding (Casarett 2018Casarett, D. 2018. The achilles heel of medical cannabis research—Inadequate blinding of Placebo-Controlled trials. JAMA Internal Medicine 178(1): 9–10. doi: 10.1001/jamainternmed.2017.5308.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]). Nevertheless, medical cannabis now has a substantial body of evidence behind it, and we can no longer dismiss cannabis as simply a recreational drug.

Therefore, we need to think carefully about how to ensure that patients have appropriate access to cannabis, in circumstances for which it may be helpful. A key element in ensuring access is insurance coverage. There are three reasons why payers should begin to cover their beneficiaries’ out-of-pocket expenses for medical cannabis, at least for those indications for which there is evidence to support its use.

First, payers are uniquely well positioned to promote the safe and rational use of medical cannabis by aligning coverage with the evidence base supporting its use. Typical laws permit physicians to recommend medical cannabis across a wide range of conditions, only some of which are supported by evidence. For instance, the California Compassionate Use Act of 1996 allows access for a variety of indications, ranging from those for which there is strong evidence (e.g., nausea) to those for which there is less (California Department of Public Health 1996). Some of the best data support the use of medical cannabis in neuropathic pain (Abrams et al. 2007Abrams, D. I., C. A. Jay, S. B. Shade, et al. 2007. Cannabis in painful HIV-Associated sensory neuropathy: A randomized Placebo-Controlled trial. Neurology 68(7): 515–521. doi: 10.1212/01.wnl.0000253187.66183.9c.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]; Wilsey et al. 2008Wilsey, B., T. Marcotte, A. Tsodikov, et al. 2008. A randomized, Placebo-Controlled, Crossover trial of cannabis cigarettes in neuropathic pain. The Journal of Pain 9(6): 506–521. doi: 10.1016/j.jpain.2007.12.010.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]), which can be very difficult to treat using more traditional therapies such as opioids and nonsteroidals. California law also lists glaucoma as an indication for medical cannabis, even though that use has been discouraged by American Academy of Ophthalmology (2014American Academy of Ophthalmology. 2014. Marijuana in the treatment of glaucoma CTA – 2014. American Academy of Ophthalmology June 26, 2014.https://www.aao.org/complimentary-therapy-assessment/marijuana-in-treatment-of-glaucoma-cta-may-2003. [Google Scholar]). (Although cannabis does lower intraocular pressure, its use for this indication has been largely replaced by newer drugs that are at least as effective and that have fewer side effects.) Insurers can employ coverage decisions to encourage those uses that are evidence-informed and to discourage those that aren’t.

This guidance will be increasingly important as more high-quality studies are conducted of cannabis’s risks and potential benefits. Indeed, it is not realistic to expect that state laws will be updated to reflect the changing evidence. Therefore, insurers are in a unique position to use coverage decisions to guide appropriate use.

Second, by offering medical cannabis coverage, insurers can make medical cannabis use more visible to health care providers. Although patients generally need a recommendation to obtain medical cannabis, they can obtain one from a medical cannabis clinic, often without the knowledge of their primary care provider. But if insurers begin to cover some or all of the costs of medical cannabis, that would create a paper trail of use. That paper trial, in turn, could be used to alert health care providers who might not otherwise be aware of a patient’s medical cannabis use. Those who are knowledgeable about medical cannabis would then be better able to counsel patients about its effectiveness, drug interactions, and safety, including dependence and driving while under the influence.

Third, insurance coverage of medical cannabis can help facilitate a unique source of crowdsourced data regarding medical cannabis’s risks and benefits. Especially if coverage is linked to self-reporting platforms (e.g., Strainprint, MMJ.org, MJBuddy), insurers could encourage reporting in a way that allows patients to learn from the experiences of others. Such reporting would also create a valuable observational database that could advance medical cannabis research.

Insurance coverage of medical cannabis might take two forms. When moderate or strong evidence exists, insurers could cover medical cannabis directly, as they do other medications. Its use for neuropathic pain, for instance, is well supported by existing evidence. There is also good evidence to support its use for insomnia. The cannabinoids in medical cannabis are also effective in the management of anorexia, nausea, and particularly nausea associated with chemotherapy. For these indications, at least, payers could begin to provide coverage just as they do for other medications that are prescribed for these indications.

Granted, most uses will not have this level of evidence to support them. The only legal source of cannabis for research in this country has been the National Institute on Drug Abuse (NIDA), which has a mandate to study cannabis only as a substance of abuse. Therefore, NIDA cannabis can be obtained for clinical research on therapeutic effectiveness only if another funder supports the trial. These hurdles mean that there are few high-quality randomized controlled trials evaluating cannabis’s medical benefits.

For other indications for which the evidence is less robust, insurers could cover medical cannabis use through health savings accounts. That might be more appropriate for chronic nociceptive pain, posttraumatic stress disorder, and anxiety, among other conditions. For these, payers can begin to create the infrastructure that will allow investigation of medical cannabis’s potential benefits.

Insurance companies may object that they cannot provide coverage for medical cannabis until the Food and Drug Administration (FDA) determines that cannabis and its chief ingredients (tetrahydrocannabinol [THC] and cannabidiol [CBD]), are safe and effective. However, THC is already approved for use in the form of dronabinol (Schedule 3), and a predominant CBD product was approved as Epidiolex (Schedule 5). Moreover, insurers commonly provide coverage for products or services that do not have proven health benefits, such as activity tracking devices and health club memberships. So the lack of FDA approval need not be a barrier.

Insurers may also argue that they cannot cover the costs of cannabis because federal law prohibits its use. But this argument, too, is being questioned in the courts. For instance, the New Mexico Court of Appeals ruled that an employee’s insurer must cover medical cannabis (Gregory Vialpando vs. Ben’s Automotive Services and Redwood Fire & Casualty 2014Gregory Vialpando vs. Ben’s Automotive Services and Redwood Fire & Casualty. Filing date May 19, 2014. Case number 32,920. Court of Appeals of the State of New Mexico. Appeal from Terry Kramer, Workers’ Compensation Judge. In:2014. [Google Scholar]). One of that court’s findings was that the employer and insurer could make no convincing argument that federal law would prohibit coverage.

Insurers arguably have an obligation to provide coverage, because they benefit from the current system. Not only do they avoid paying for the costs of medical cannabis, they may currently also avoid paying for other medications when patients substitute cannabis for prescription medications. Indeed, some evidence suggests that medical cannabis may have opioid-sparing effects (Abrams et al. 2011Abrams, D. I., P. Couey, S. B. Shade, M. E. Kelly, and N. L. Benowitz. 2011. Cannabinoid–Opioid interaction in chronic pain. Clinical Pharmacology & Therapeutics 90(6): 844–851. doi: 10.1038/clpt.2011.188.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]). In addition, one large national study found a decrease in rates of fatal opioid overdoses in states in which medical cannabis has been legalized (Bachhuber et al. 2014Bachhuber, M. A., B. Saloner, C. O.Cunningham, and C. L. Barry. 2014. Medical cannabis laws and opioid analgesic overdose mortality in the United States, 1999-2010. JAMA Internal Medicine 174(10): 1668–1673. doi: 10.1001/jamainternmed.2014.4005.[Crossref], [PubMed], [Web of Science ®], , [Google Scholar]). Because the cannabinoids in medical cannabis do not appear to bind to receptors in the brainstem, overdoses of cannabis ingestion may cause tachycardia and hypotension, but have not been reported to be lethal.

Insurance coverage of medical cannabis will almost certainly create problems. For instance, when health insurance is provided by an employer, there is reason for concern that a patient’s use of medical cannabis might adversely affect his or her employment. And some patients may be concerned that the presence of cannabis in their medical record might be used to deny future employment or insurance. However, these and other problems of discrimination are manageable, and should not prevent those patients who can benefit from medical cannabis from gaining access.

Insurers have the opportunity to guide the appropriate use of medical cannabis by offering coverage for those indications that are most supported by data. By doing so, they can improve patient care, and can facilitate the creation of registries that allow research. Moreover, they should arguably bear some of the cost when medical cannabis use is improving the health of their beneficiaries. Therefore, payers should begin to offer coverage for those indications for which there is an appropriate level of evidence.