How doctors built the case for medical marijuana.

Peter Hecht, National Journal

Donald Abrams knew that Mark Henry was going to die. It was 1986 when the two met. Abrams, one of the world's leading AIDS doctors, was delivering a lecture in Maui. The speech had been set up in honor of Henry's former partner, who had been the first person on the island to die of the disease. Henry, who had HIV himself, sought out Abrams at the lecture. Abrams's experiences told him that Henry's deterioration would be rapid and excruciating. Yet that knowledge didn't stop him from developing a relationship with Henry. It didn't stop the two men from vacationing together, from laughing together, or from drawing close to each other.

Abrams was a Stanford University-educated hematologist and oncologist. He'd begun his work on AIDS in 1979 when an unexplainable progression of young gay men with swollen glands began showing up at the Kaiser Foundation Hospital in San Francisco. Lee Wilkinson, the hospital's chief hematologist, summoned Abrams, a 29-year-old openly gay medical resident. "Hey, Duck," Wilkinson said, invoking the Disney character nickname he had bestowed on the promising young doctor. "Can you take a look at this?"

Abrams ordered lab tests on the men's lymph nodes. The results offered nothing from which to draw conclusions. So he found himself cautioning the men on unhealthy lifestyles. "Stop having so many sexual partners," he lectured. "Stop taking drugs. Move out of the fast lane."

At the urging of Wilkinson, Abrams moved on to the hematology and oncology training program at the University of California (San Francisco). There, he started encountering more young gay men with new, more-severe symptoms. By 1981, doctors at the university were seeing the first cases of Kaposi's sarcoma, a haunting, systemic affliction that left patients splotched with lesions. Abrams began focusing on a strange "gay cancer" that would become known as Acquired Immune Deficiency Syndrome. He noticed many of the men had swollen glands. "How long have you had those?" Abrams asked. About two years, they told him, suggesting a troubling progression of the symptoms Abrams had noted in 1979.

Abrams applied for a grant to study lymphadenopathy syndrome, hoping to find clues about the first symptoms of illnesses leading to diagnoses of AIDS. He examined 200 gay men with swollen glands who had yet to develop Kaposi's sarcoma or pneumocystis, the ravaging pneumonia that would signal their rapid decline. He took lymph node specimens to the UC San Francisco laboratory of Harold Varmus, a Nobel Prize-winning researcher. They began studying what would be identified as a retrovirus that attacked cells and altered the body's DNA and its ability to fend off disease.

In 1983, Abrams moved to San Francisco General Hospital. There, oncologist Paul Volberding and infectious-disease specialist Connie Wofsy established the world's first AIDS inpatient ward, Ward 5B, with Abrams as their protégé. For a time, the trio knew every AIDS patient in San Francisco. That year, a French researcher, Luc Montagnier, would be widely credited with discovering the virus that led to AIDS. Abrams later became one of three researchers to name it the Human Immunodeficiency Virus, or HIV. Soon the rate of HIV diagnoses and the number of AIDS patients were skyrocketing. All too soon, Abrams recalled, "we didn't know everybody anymore." San Francisco became a death camp for men wasting away, disfigured, stigmatized with a gay disease. Over two decades, 19,000 city residents would perish from AIDS.

Still, Abrams let himself fall in love with Mark Henry. When the two men met, Henry had just had an episode of AIDS-related pneumonia. Abrams knew nearly everyone with the disease died within 12 months after a pneumocystis outbreak.

By 1986, physicians were prescribing a new antiviral drug called AZT for the treatment of AIDS. Abrams believed the toxicity of AZT outweighed its therapeutic benefits. Henry didn't take the drug. Yet Abrams watched his partner survive through 1986, then 1987, then 1988, and into 1989. Throughout this time, Henry smoked marijuana. Abrams, who as an undergrad at Brown University preferred pot to booze, joined him. Ultimately, the weed made the doctor feel paranoid, so Abrams curtailed his use. But Henry did not. He went on to outlive fellow AIDS patients in one support group, then a second, then a third.

When he died in 1989, Mark Henry had survived three times longer after pneumocystis than the average for AIDS patients. Abrams didn't know for sure if there was a medical benefit to the marijuana Henry had used. The doctor also had no inkling that, years later, he would become California's most renowned researcher on medical marijuana—and that clinical trials in the Golden State would challenge the federal government and health-research orthodoxy by helping establish the medical efficacy of pot. Still, it was not lost on Abrams just how long Henry had managed to live—and he used cannabis "every freakin' day."


Twenty-five years after Henry's death, medical marijuana is legal in 20 states. In 2012, Colorado and Washington became the first states to legalize marijuana for nonmedical reasons. Recently, President Obama said of pot during an interview with The New Yorker: "I don't think it is more dangerous than alcohol."

Yet with the tide of history seemingly on the side of legalization, it's easy to forget that the question of whether pot can be legitimately used for medical reasons remains a fraught political subject. The federal government still classifies marijuana as a Schedule I drug with no accepted medical use. And last year, the government's position on marijuana received a crucial endorsement from the U.S. Court of Appeals for the District of Columbia Circuit. A group called Americans for Safe Access had sued the Drug Enforcement Administration over marijuana's 42-year-long classification as a Schedule I drug, and the court ended up siding with the status quo. "There is serious debate in the United States over the efficacy of marijuana for medical uses," Senior Circuit Judge Harry Edwards wrote in his majority opinion—before concluding that "substantial evidence" supports the government's claim that studies affirming accepted medical use of marijuana "do not exist." In other words, as far as the D.C. Circuit and the federal government were concerned, the work of Donald Abrams—and other doctors who had spent many years exploring the medical benefits of marijuana—never happened.

To be sure, Abrams and his peers in the medical profession were hardly the only people working to change the discussion surrounding marijuana over the last two decades. Throughout the country, but especially in California, there has been no shortage of activists, politicians, intellectuals, and entrepreneurs who pushed society to adopt a different view of pot. Yet of all these groups, it is arguably the doctors—specifically those steeped in the research on medical marijuana—whose work has offered the most direct challenge to the federal government's current stance on cannabis.


By 1990, anguish over the unsolved epidemic of AIDS and HIV was exploding into rage in San Francisco. That year the city hosted the International AIDS Conference, only to have the event disrupted by activists from ACT UP, the AIDS Coalition to Unleash Power. Wearing "Silence=Death" shirts, they blew piercing whistles, threw chairs, and directed shouts of "Shame!" at conference participants, including Donald Abrams and other AIDS doctors and researchers. According to the activists, people were dying because of a conspiracy of government obstruction, because of a lack of funding for AIDS treatments, and because of the continued failure of researchers to find a cure.

By then, another movement was attaching itself to the cause of AIDS and gay rights—marijuana. That same year, narcotics officers raided the apartment of Dennis Peron, a pot dealer who supplied weed to sick people in the gay community. Police roughed up Peron's lover, lesion-wracked and skeletal-thin AIDS sufferer Jonathan West. An enraged Peron would later describe an officer putting a boot on West's neck and taunting him: "Know what AIDS means? Asshole in Deep Shit." In 1991, Peron marshaled his fury to win 80 percent voter support for San Francisco's Measure P, an advisory measure calling on the state to legalize marijuana as medicine.

Peron also found an ally in Mary Jane Rathbun, an eccentric Irish-Catholic septuagenarian with a passion for baking and for pot. She partnered with him to set up the San Francisco Cannabis Buyers Club, which went on to provide marijuana to thousands of gay men with AIDS. They would later cowrite a book, Brownie Mary's Marijuana Cookbook and Dennis Peron's Recipe for Social Change.

It was at San Francisco General Hospital where Donald Abrams and fellow AIDS doctors got to know "Brownie Mary." For years, she was an ever-present volunteer in an outpatient clinic in a converted pediatrics ward on the sixth floor of the hospital's building No. 80. "Ward 86" became America's most famous AIDS clinic. There, Rathbun shuttled sick young men she dubbed her "kids" from the clinic to radiology. She also brought them fresh-baked goods infused with marijuana, becoming a media darling in the process.

In 1992, Abrams arrived in Amsterdam for an International AIDS Conference. He flipped on the television in his hotel room. CNN International was broadcasting a breaking story from the Bay Area: Brownie Mary had been arrested with two and a half pounds of weed, raided by police as she was crafting her confections at a friend's home in Sonoma County. Soon the City of San Francisco—which under the voter-approved Measure P now officially considered marijuana a legitimate source of symptom relief for AIDS, cancer, and other illnesses—staged a "Brownie Mary Day." The Sonoma County district attorney later dropped the charges.

Brownie Mary's message elevated AIDS and pot as an international story. It reached a frustrated medical-marijuana research advocate named Rick Doblin in North Carolina. Doblin was preparing to resume his doctoral studies in public policy at Harvard after completing his Harvard master's degree with a thesis on medical marijuana. He had founded a group called the Multidisciplinary Association for Psychedelic Studies, which advocated research on alternative medicines, including marijuana and the drug known as Ecstasy.

Doblin had no medical degree. But for two years he had shopped around a research protocol for studying marijuana's effectiveness in reversing the nausea and loss of appetite that led to wasting syndrome and starved AIDS patients into hollow-eyed human forms. After Doblin saw Brownie Mary on television, he wrote a letter "to whom it may concern" at San Francisco General Hospital's AIDS program. The letter, urging someone there to take on the research project, was routed to Abrams. The doctor followed up, and Doblin made an instant impression. Doblin insisted that groundbreaking marijuana research "should come from Brownie Mary's institution." Abrams was bemused by the thought of becoming the Brownie Mary Research Institute. But he was inspired by the study idea, which made him think of Henry. It reminded him of how pot had seemed to keep his lover alive and functioning for so long.

By 1992, Abrams was working with San Francisco's Community Consortium, a group of physicians setting up community-based clinical trials to explore treatments with AIDS and HIV patients. Despite being intrigued, Abrams was skeptical of Doblin's plan to investigate the medicinal effectiveness of marijuana brownies. He didn't see a way to standardize cannabis doses in brownies, particularly over a multiweek clinical-research trial. But Doblin insisted that some kind of study of marijuana's effect on wasting syndrome needed to happen. Doblin contacted the Food and Drug Administration about supporting a clinical trial and got a positive response. He arranged with a Dutch medical-marijuana firm to grow cannabis for the research.

The same year, the prescription drug Marinol—containing a synthetically produced version of marijuana's psychoactive delta 9-tetrahydrocannabinol (THC) constituent—had been approved for treatment of people with HIV wasting syndrome. But patients of physicians at Community Consortium reported that swallowing Marinol left them zoned out for hours. Many reported they preferred smoking pot because they could regulate their dose through the number and spacing of hits from a joint. The Community Consortium's board decided to back a study to assess separate groups of wasting patients—who either smoked marijuana or took Marinol—for changes in immune-system levels, body weight, and body composition. Research review boards from UC San Francisco and the state of California signed off on the study.

But in trying to pursue the research, Abrams recalls that he ran into a series of roadblocks in the federal government—at the Drug Enforcement Administration, the National Institute on Drug Abuse, and the National Institutes of Health. NIH, for instance, reviewed the study but effectively killed it by not attaching a research score, an action critical to ranking the trial for potential government funding. Two peer reviewers questioned why researchers wanted to test a "toxic" substance such as pot. Another suggested that wasting patients rediscovering their appetites might develop high cholesterol, clogged arteries, and reduced suppression of tumor cells. Abrams was incredulous. People with wasting syndrome didn't live remotely long enough for any of that to occur.


In November 1996, three months after the NIH helped kill Abrams's study, California voters passed Proposition 215, which made the state the first in the country to legalize medical marijuana. Though it made no reference to medical research, California's historic vote seemed to cry out for studies on pot's effectiveness for those suffering from AIDS/HIV, cancer, anorexia, spasticity, and pain.

After Proposition 215's passage, Abrams attended meetings with the San Francisco Department of Public Health to discuss how to make the law operational for people who needed cannabis. At one meeting, he met an ACT UP member worried about the recent death of an AIDS patient who had been taking Ecstasy. The drug had blocked the liver's metabolism of protease inhibitor drugs, heightening the patient's vulnerability to the disease. By then, an estimated 11,000 gay men in San Francisco were using marijuana for AIDS or HIV. The ACT UP activist wanted to know whether pot, too, could interfere with life-saving antiretroviral drugs.

Abrams had an epiphany. He could study the risk of pot for patients on protease inhibitors. In other words, instead of trying to investigate whether pot was medically useful, he would—as a way of at least moving the science on pot forward—investigate whether it was medically harmful when used with anti-AIDS drugs. He wrote up a research protocol for a clinical trial on the potential interaction between marijuana and the AIDS-fighting medications, arguing that the connection was "worrisome since many HIV-infected patients continue to smoke marijuana as an appetite stimulant or to decrease nausea."

In August 1997, after a peer review by specialists Abrams had suggested—in clinical pharmacology, immunology, virology, endocrinology, and HIV medicine—the National Institute on Drug Abuse and other government agencies signed off on the project. With $1 million in funding awarded for the research, Abrams accepted delivery from NIDA of 1,400 government joints, plus a research supply of the synthetic THC drug Marinol. The pot was stored at San Francisco General Hospital in a locked freezer equipped with a burglar alarm.

On May 12, 1998, the first patients were enrolled in the study. One group daily smoked three marijuana joints that contained 3.95 percent THC and had been prepped in a humidifier. Another group smoked cannabis-fragrant placebo joints with the THC removed. A third group took oral applications of Marinol. The 16-member research team led by Abrams monitored the weight, cell counts, and viral health of the patients.

The study concluded that use of cannabinoids—the natural THC in pot or synthetic THC in Marinol—neither increased the viral load of individuals with HIV infection nor interfered with their protease inhibitor drugs. Anecdotally, the researchers noted that patients given Marinol were more lethargic and spent more time in bed, while those smoking pot were more active. The research, Abrams wrote, failed to demonstrate "clinically significant interactions with cannabinoids that would warrant dose adjustments of protease inhibitors." He delivered the conclusion that marijuana was safe for people with HIV. It didn't interfere with anti-AIDS drugs. And there was something more: Patients using marijuana and Marinol saw increased production of healthy cells, with marginally higher levels for the pot smokers. People in the marijuana group also put on an average of 7.7 pounds in 21 days, compared to 7 pounds for the Marinol group and 2.9 pounds for the placebo group.

Despite the results, no leading medical research journal was eager to publish the study. Abrams was rejected by the prestigious journal The Lancet, by The New England Journal of Medicine, and by The Journal of the American Medical Association. Ultimately, after an initial denial, the Annals of Internal Medicine of the American College of Physicians interviewed Abrams on his study. In 2003, the journal published his research.

Clinical evidence of the efficacy of cannabis with HIV patients was now part of accepted medical literature. Still, in the study and with the media, Abrams was politic about the conclusions. He wrote that the findings meant "placebo-controlled studies of the efficacy of smoked marijuana could be considered in the future." He called for more research.

After the death of Mark Henry, Abrams had found a new life partner in Clint Werner. They began a relationship in 1994 and later married. Years after Abrams's clinical trial on cannabis and patients with HIV infection, Werner revealed the doctor's internal sense of triumph. What Abrams mostly refrained from broadcasting, Werner, a natural-foods chef devoted to the dietary prevention of disease, all but shouted out. "The clinical trial was a Trojan horse, finally allowing researchers to get the data they had been seeking for years," Werner wrote in his book Marijuana Gateway to Health. Abrams's "true purpose" in the study, Werner wrote, "had been to ascertain whether marijuana helped improve the appetite of AIDS patients—and it did." Abrams's husband concluded with a flourish: "Science had spoken. THC really did cause the munchies."


After Donald Abrams's breakthrough clinical trial, state Assemblyman John Vasconcellos sought to make medical cannabis research a fully funded priority of the state. Vasconcellos championed legislation calling for creation of a California Center for Medicinal Cannabis Research. Its mission would be to provide answers—affirmative, negative, or both—to the question, "Does marijuana have therapeutic value?"

Vasconcellos set out to develop the concept in consultation with some of the top medical and research professionals in the University of California system. He also needed political help in order to get his idea through the Legislature. A self-described "old-time liberal," Vasconcellos found an unexpected ally during the 1998 election season in a law-and-order conservative, Dan Lungren, the state's attorney general. Lungren, who had been an ardent opponent of Proposition 215, was running for governor that year. He drew a Republican primary challenge from Dennis Peron, the pro-medical-marijuana activist. With no shot at winning, Peron served as an antagonist to remind people that Lungren opposed the will of voters on medical marijuana. Looking to find middle ground on pot for the November general-election race against Democrat Gray Davis, Lungren endorsed research on the medical use of cannabis. Vasconcellos saw a political opportunity. He reached out to the attorney general. Their staffs began drafting legislation for unprecedented medical-marijuana studies.

In 1999, by the time the legislation—Senate Bill 847—reached the floor, Vasconcellos had moved to the state Senate. Lungren was gone from the attorney general's office and had been defeated in the governor's race. But his endorsement helped pry loose crucial Republican votes. Vasconcellos needed 27 votes—a two-thirds majority of the 40-member Senate—to pass the appropriation. With no votes to spare, Vasconcellos offered procedural cover to three Republican senators—Jim Brulte, Tim Leslie, and Pete Knight—who were skittish about being the 27th vote on a pro-marijuana bill. The GOP trio agreed to shout "aye" in unison so that no individual got the blame as California lawmakers voted to spend $8.7 million in state tax dollars to study weed. In 2000, the Center for Medicinal Cannabis Research was born.

Research proposals began streaming in by 2001. Over the next decade, the center—headquartered at the University of California (San Diego)—would approve and oversee 15 California clinical studies, including seven trials directly testing pot's effect on research subjects. The center established an exhaustive peer-review process and used the legislative clout of the most populous state in the nation to win research approval from multiple federal agencies. The center's director, Igor Grant, a renowned neuropsychiatrist at UC San Diego, didn't conduct any research himself; he saw to it that the center handled all research applications and that his clinicians didn't have to fight the government to do cannabis work. Meanwhile, he demanded rigorous, modern trials that could be published in medical-research literature. He figured there was no point in wasting California tax money on studies "that wouldn't see the light of day."

One of those applying for a research grant, Barth Wilsey—a pain-management physician at the University of California (Davis)—sought funding for a study on whether cannabis provided relief for people with neuropathic pain from spinal cord injuries, diabetes, strokes, and other conditions causing life-disrupting discomfort from nerve damage or injury. Earlier in his career, Wilsey was a fellow in pain management at UC San Francisco, seeing patients in a small clinic in the city. People coming in would tell him they turned to marijuana when nothing else seemed to work. Wilsey, a licensed acupuncturist, was interested in alternative therapies.

Mark Wallace, an anesthesiologist and pain specialist at UC San Diego, proposed a study in which healthy subjects would smoke marijuana and researchers would study the analgesic effects of cannabis after injecting capsaicin, the hot ingredient in chili peppers, into their skin. Jody Corey-Bloom, director of the UC San Diego Multiple Sclerosis Center, set out to see if marijuana could relieve spasticity in MS patients whose use of pharmaceuticals often failed to alleviate their suffering. Donald Abrams and Ron Ellis, a UC San Diego neurologist, worked to see if pot could quell tingling and shooting pains that would start in the feet of AIDS and HIV patients and move to their fingers and hands, disrupting their ability to sleep, to exercise, and, often, to cope.

In separate, overlapping studies between 2002 and 2006, patients with neuropathic pain from HIV smoked pot under the supervision of Abrams's nine-member research team at UC San Francisco's General Clinical Research Center and Ellis's eight-member team at the UC San Diego Medical Center. In San Francisco, 55 patients, mostly men with HIV infection for 14 years, smoked three marijuana cigarettes or placebo joints a day in five-day trials. In surveys for chronic pain during the trials, cannabis was found to reduce the subjects' pain by an average of 34 percent—double the rate of the marijuana placebo. Tests in which brushes were stroked against the subjects' skin showed pot could quell shooting pain sensations in HIV patients, for whom things as benign as pulling a bedsheet over their toes could trigger lightning bolts of agony. However, Abrams's team failed to show similar benefits of marijuana in cases of acute pain such as might be experienced after an injury or surgery.

In San Diego, research subjects were given placebo joints or three potency levels of marijuana. Ellis's team monitored 28 HIV subjects as they toked on marijuana or the pot placebos over two-week periods, with a two-week break with no marijuana use allowed between each new research stage. Notably, Ellis also had all subjects continue taking their pharmaceutical pain medications during the study. He found they still got a boost from cannabis, with the pot group reporting pain relief at two and a half times the frequency of the placebo group.

Ellis also measured impairment. He tested people's ability to connect random letters and numbers on a page, and he had them operate a driving machine, in which they were to avoid simulated traffic obstacles and follow lights and signs. In both tests, patients scored worse after smoking cannabis than before.

Indeed, researchers did not shy away from acknowledging downsides of pot. One of Ellis's original 34 enrolled research subjects had to be excused when he developed an intractable cough from smoking pot. Another subject, who had never before used marijuana, gave researchers a scare. After smoking his first joint, he started staring into space and stopped responding to questions. The man was in a catatonic state, "attending to what was going on in his head," Ellis observed. He came to in a couple of hours with no memory of what had happened. He was dropped from the study.

At UC Davis, Wilsey found something he wasn't counting on. In outpatient sessions of three to 21 days, Wilsey and his seven-member team monitored 32 patients with nerve injuries as they smoked marijuana with 7 percent or 3.5 percent THC or toked on THC-free placebo joints. Subjects in both marijuana groups found significant relief from chronic pain. What surprised Wilsey is that they got the same relief with the lower cannabis dose as with the higher dose, only with less impairment.

Corey-Bloom reported that MS patients given cannabis showed modest cognitive impairment. However, her study on 30 patients also demonstrated that smoking marijuana could reduce painful, often disabling symptoms of spasticity.

At UC San Diego, the results of Wallace's study using the ingredient in hot chili peppers produced compelling evidence of cannabis's effectiveness with pain while also stirring questions about marijuana dosing. Wallace had healthy research subjects smoke marijuana joints with 2 percent, 4 percent, and 8 percent THC, then injected capsaicin into one of each subject's forearms five minutes afterward, and into the other forearm 45 minutes afterward. When the subjects were injected after five minutes, pot had no significant effect in reducing the capsaicin-induced pain. When injected after 45 minutes, subjects taking the 2 percent THC dose reported little or no pain reduction. But people smoking 4 percent THC reported significant relief. And people smoking cannabis with 8 percent THC found their pain actually increased. For Wallace, the results were significant. His clinical trial had revealed exposure levels where marijuana was both effective and not.


There were more studies as well—an attempt to figure out whether patients with painful neuropathy could get relief from marijuana without getting stoned; a trial that tested whether smoking joints or consuming cannabis by vaporizing delivered medicinal effects more efficiently; a study to find out if cannabis could be a suitable cotherapy for patients taking opiate drugs.

The doctors who conducted all these studies were not activists; their goal was not to embrace the political marijuana cause. Nevertheless, in February 2010, when Igor Grant and the University of California researchers gathered at the state Capitol to present their preliminary findings, their answer to the state-mandated question—does marijuana have therapeutic value?—was yes.

"As a result of this program of systematic research, we now have reasonable evidence that cannabis is a promising treatment," Grant announced. At the time, he refrained from pointing out that the results flew in the face of the U.S. government's classification of marijuana as a Schedule I drug with no accepted medical use. But in 2012, with the California clinical trials completed and the last studies being published, Grant let it be known that the Schedule I classification "is completely at odds with existing science" and "seems intellectually dishonest."

Every clinical study by the Center for Medicinal Cannabis Research showed potential medical benefits from cannabis. Yet the California trials didn't include anywhere close to sufficient numbers of patients to meet FDA standards for evaluating marijuana's potential as a prescription medication. Moreover, national drug-abuse and addiction surveys said marijuana—though it presented less perceived danger than other substances—could be addictive, with 9 percent of users developing an unhealthy pot dependency. Within the medical community, there remains a divide between physicians who have studied the potential benefits of marijuana and those who specialize in addiction. The latter group, not surprisingly, is more skeptical about pot.

Grant cautioned that a significant body of medical literature raised legitimate concerns over pot's danger for people susceptible to schizophrenia or other psychotic illnesses and over its potential harms for teenagers, whose brains were still developing. The government pot used in the studies also wasn't even close to the bottom-shelf intensity of designer weed strains sold at many California medical-marijuana dispensaries, meaning researchers didn't test what was actually on the market. They also didn't have the option of looking at new niche marijuana strains bred with low-THC and high concentrations of cannabidiol, or CBD, a cannabis constituent believed to deliver analgesic benefits without impairment.

Still, Grant and his research teams were able to characterize marijuana as a promising therapy that offered relief to people for whom other treatments failed. It was an important juncture in a scientific journey, one that Mark Henry had helped inspire Donald Abrams to go on many years ago.