For researchers, marijuana is tightly regulated, but some patients can easily buy it at dispensaries like this one.
Kevork Djansezian/Getty Images

Could pot help solve the U.S. opioid epidemic?

In the mid-19th century, some European doctors became fascinated with a plant-derived drug recently imported from India. Cannabis had been used as medicine for millennia in Asia, and physicians were keen to try it with their patients. No less an authority than Sir John Russell Reynolds, the house physician to Queen Victoria and later president of the Royal College of Physicians in London, extolled the medical virtues of cannabis in The Lancet in 1890. “In almost all painful maladies I have found Indian hemp by far the most useful of drugs,” Reynolds wrote.

Like other doctors of his day, Reynolds thought cannabis might help reduce the need for opium-based painkillers, with their potential for abuse and overdose. “The bane of many opiates and sedatives is this, that the relief of the moment, the hour, or the day, is purchased at the expense of to-morrow’s misery,” he wrote. “In no one case to which I have administered Indian hemp, have I witnessed any such results.”

More than 125 years later, the misery caused by opioids is clearer than ever, and there are new hints that cannabis could be a viable alternative. Some clinical studies suggest that the plant may have medical value, especially for difficult-to-treat pain conditions. The liberalization of marijuana laws in the United States has also allowed researchers to compare overdoses from painkiller prescriptions and opioids in states that permit medical marijuana versus those that don’t. Yet following up on those hints isn’t easy. Clinical studies face additional hurdles because the plant is listed on Schedule I, the U.S. Drug Enforcement Administration’s (DEA’s) list of the most dangerous drugs.

Some researchers worry that rigorous research is being outpaced by informal experimentation, as millions of people with access to medical marijuana treat themselves. “It’s clear that the policy has gone way out in front of the science in terms of allowing access to products that haven’t been through the standard clinical trials process,” says Mark Ware, a pain specialist at McGill University in Montreal, Canada.

It’s mind-boggling that we have millions of people in the U.S. using cannabis for medicine and we not only don’t have the proper data to help them take it appropriately, we’re not doing a good job of collecting it.

Ryan Vandrey, behavioral pharmacologist, Johns Hopkins University

Nearly 2 million Americans were addicted to or abusing prescription opioid drugs in 2014, according to the Centers for Disease Control and Prevention, and the Kaiser Family Foundation estimates that more than 21,000 died from overdoses.That same year, a study published in JAMA Internal Medicine hinted that medical marijuana could make a dent in that alarming toll. The researchers, led by Marcus Bachhuber, then at the Philadelphia Veterans Affairs Medical Center in Pennsylvania, examined death certificates in all 50 states between 1999 and 2010. They found that the annual rate of deaths due to overdose on an opioid painkiller was nearly 25% lower in states that permitted medical marijuana. In 2010, that translated into 1729 fewer deaths in those states. The researchers also found that the effect grew stronger in the 5 to 6 years after the states approved medical marijuana.

More recently, David Bradford, a health economist at the University of Georgia in Athens, and his daughter Ashley, a master’s student there, sought to investigate whether marijuana was supplanting conventional drugs in states where it’s legal. Analyzing Medicare drug prescription data from 2010 to 2013, they found a significant difference in the number of prescriptions for several conditions, including anxiety and nausea, in states with medical marijuana. But one condition stood out from the rest: “The effect for pain was three to four times larger than all of the others,” David Bradford says. In medical marijuana states, each physician prescribed an average of 1826 fewer doses of conventional pain medication each year, they reported in the July issue of Health Affairs. That translates into many millions of doses per year in those states.

The Bradfords haven’t yet analyzed how many of those doses were opioid drugs versus other painkillers, but David Bradford suspects it’s a large chunk. “It’s suggestive evidence that medical marijuana might help divert people away from the path where they would start using [an opioid drug], and of course if they don’t start, they’re not on that path to misuse and abuse and potentially death.”

In a follow-up study, the Bradfords analyzed prescription data from Medicaid recipients, a younger population than the Medicare enrollees in their previous study. So far, the reduction in pain prescriptions appears to be even more dramatic in this group,  David Bradford says.

A rising toll

Statistics from the Kaiser Family Foundation show a growing epidemic of opioid abuse in the United States over the last 15 years. In the cartogram below, the size of each state reflects the total number of prescription opioid overdose deaths from 1999–2014. Darker colors indicate more overdose deaths relative to the state’s population in 2014.

(Graphic) J. You/Science; (Data) Kaiser Family Foundation/U.S. Census Bureau

Additional evidence about whether cannabis can reduce opioid use could come from Canada, which legalized medical marijuana in 2001 and might legalize recreational use as soon as next year. In Quebec, researchers established a patient registry in 2015 to collect demographic data on patients who use medical marijuana, the type and dose they take, and the conditions they’re seeking treatment for, along with self-reports on benefits and adverse outcomes. McGill’s Ware, who is leading the effort, says the registry is also collecting data on opioid use. “We’ll certainly be looking at whether patients who manage their pain with cannabis can reduce their opioid doses over time or even wean themselves off opioids entirely,” he says.

Yet in the United States, where 25 states and Washington, D.C., have legalized medical marijuana, there are no state-wide efforts to collect data on how patients are using cannabis or on whether they have been affected for good or ill, in part because marijuana is still illegal at the federal level. That’s a huge missed opportunity, says Ryan Vandrey, a behavioral pharmacologist at Johns Hopkins University in Baltimore, Maryland. “It’s mind-boggling that we have millions of people in the U.S. using cannabis for medicine and we not only don’t have the proper data to help them take it appropriately, we’re not doing a good job of collecting it.”

Researchers have good reason to think marijuana might relieve pain. Tetrahydrocannabinol, or THC, the plant’s main psychoactive ingredient, binds to a class of receptors on neurons that are involved in mediating pain, appetite, and mood, among other things. “It’s working directly on pain pathways in the brain, spinal cord, and periphery,” says Ethan Russo, a neurologist and medical director of Phytecs, a Los Angeles, California–based company developing therapies based on compounds isolated from marijuana. Previously, Russo oversaw international clinical trials for Sativex, an oral spray made by GW Pharmaceuticals in Salisbury, U.K., that has been approved in 27 countries for treating spasticity caused by multiple sclerosis and in Canada for certain types of pain. Sativex combines THC with cannabidiol, another compound in marijuana that may counteract the anxiety and cognitive side effects associated with THC and that appears to have antiinflammatory effects.

But few marijuana-based therapies have gone through clinical trials. A metaanalysis published last year in the Journal of the American Medical Association found just 28 randomized clinical trials investigating cannabis for chronic pain. (Sativex accounted for nearly half of them.) The authors concluded there was “moderate quality evidence” to support its use. Part of the reason for the scarcity of cannabis trials is that whole plants and natural extracts aren’t patentable, giving pharmaceutical companies little incentive to pursue them.

Recently, however, some states that have legalized medical marijuana have begun to fund clinical studies. California, which in 1996 became the first U.S. state to legalize medical marijuana, led the way with its Center for Medicinal Cannabis Research, which has done several placebo-controlled studies on pain. Barth Wilsey, a pain management physician at the University of California in San Diego, led two of them. The first, published in 2008, found that smoking marijuana reduced pain caused by nerve damage in 38 patients, with minimal side effects. The second, published in 2013, found that vaporized cannabis, even in low doses, relieved pain in a similar group of patients who hadn’t responded to traditional medications, including opioid analgesics.

A trial just getting underway at the University of Colorado (CU) Anschutz Medical Campus in Aurora will be the first to directly compare cannabis and opioid painkillers in patients with back and neck pain. “There’s definitely emerging evidence in the literature for [using cannabis to treat] neuropathic pain, but there’s hardly anything for chronic back and neck pain, which is one of the most common reasons people go see their doctor,” says neurobiologist Emily Lindley, who will run the CU study.

Hers is one of nine medical marijuana research grants funded so far by the state of Colorado with a total of $9 million from tax collected on marijuana sales. The impetus for the study was a survey done a few years ago at CU Hospital’s Spine Center. Nearly one-fifth of the 184 patients with chronic back and neck pain who responded to the survey reported using marijuana to treat their pain. Of those, 86% reported that it “moderately” or “very much” relieved their pain, and 77% said marijuana provided as much or more relief than their opioid prescription painkillers. “We expected to see some positive effects regarding pain control but not quite to that extent and not with that many patients,” says Vikas Patel, chief of orthopedic spine&surgery at CU.

Now, Lindley’s study will enroll 50 patients with back and neck pain, who will visit the university three times and receive either vaporized cannabis, the opioid drug oxycodone, or a placebo. (In the case of cannabis, the placebo is marijuana with the THC chemically extracted; for oxycodone, the placebo is a pill.) At each visit, the patients will be given a battery of tests to assess their pain levels and look for side effects like impairments of memory, attention, and concentration.

But such research faces regulatory obstacles, because DEA still classes marijuana as a Schedule I drug: the most dangerous drugs with no known medical benefits. It has taken Lindley nearly 2 years from the time she received her grant to start her study. Getting the required Schedule I license from DEA took about 6 months. Prior to that, the university spent tens of thousands of dollars to install secure narcotics cabinets to meet DEA’s requirements and a new ventilation system to comply with its own no smoking policy.

In August, DEA rejected two petitions to remove marijuana from Schedule I. The decision was made after a scientific review by the Food and Drug Administration (FDA) concluded that the evidence for the medical benefits of marijuana did not meet their standards for new drug approval.

FDA noted that most cannabis studies to date have been fairly small—with a few dozen participants, not hundreds—and they’ve followed patients for a few hours, not the 12 weeks or more that’s typical in the clinical trials pharmaceutical companies conduct. Another complication is the variation in how cannabis is delivered. Patients in many early studies smoked it, and people ingest varying amounts of THC per puff. Newer delivery systems, such as vaporizers and edible products, add still more uncertainty about the doses patients actually receive. Then there’s the natural variation in the concentration of THC and other cannabinoids in different strains of marijuana.

Even scientists who are bullish on the potential for medical marijuana acknowledge that consistent dosing is an issue. Yet many researchers see the situation as a Catch-22: The Schedule I listing and other restrictions on marijuana research hinder the type of studies that are needed to convince regulators to loosen those restrictions.

Two bills introduced in Congress this year aim to lower some of these hurdles. The bills would limit the time that DEA spends reviewing proposed research studies (just as FDA has 30 days to review drug studies). They would also restrict DEA’s role in making sure listed drugs are stored securely. Now, DEA also has to weigh in on changes of scientific protocol, and that can really slow things down, says Vandrey of Johns Hopkins, who is collaborating on a study there to compare the analgesic effects of cannabis and the opioid drug hydrocodone in healthy subjects.

A third bill, introduced in July, aims to ease research with cannabidiol and other chemical components of marijuana. “The current interpretation [of the Controlled Substances Act] is that anything in the plant is Schedule I,” Vandrey says. Even though there is no evidence that cannabidiol is prone to abuse, researchers interested in studying it have to jump through the same hoops as if their study involved whole-leaf marijuana. “That, in my mind, is just silly,” Vandrey says. Research on terpenes, still another group of cannabis compounds that may have analgesic effects, faces the same hurdles.

“With changing attitudes and changing policy, I’m hopeful that research can proceed with fewer barriers,” McGill’s Ware says. He and others hope they’ll soon be able to firm up the case for marijuana as an effective pain treatment. “I’d hate to think we’re still asking the same questions 10 years from now,” Ware says.

Listen to our podcast interview with writer Greg Miller on the impact of legal pot on opioid abuse.