New York moved last week to join 22 states in legalizing medical marijuana for patients with a diverse array of debilitating ailments, encompassing epilepsy and cancer, Crohn’s disease and Parkinson’s. Yet there is no rigorous scientific evidence that marijuana effectively treats the symptoms of many of the illnesses for which states have authorized its use.

Instead, experts say, lawmakers and the authors of public referendums have acted largely on the basis of animal studies and heart-wrenching anecdotes. The results have sometimes confounded doctors and researchers.

The lists of conditions qualifying patients for marijuana treatment vary considerably from state to state. Like most others, New York’s includes cancer, H.I.V./AIDS and multiple sclerosis. Studies have shown that marijuana can relieve nausea, improve appetite and ease painful spasms in those patients.

But New York’s list also includes Parkinson’s disease, Lou Gehrig’s disease and epilepsy, conditions for which there are no high-quality trials indicating marijuana is useful. In Illinois, more than three dozen conditions qualify for treatment with marijuana, including Alzheimer’s disease, lupus, Sjogren’s syndrome, Tourette’s syndrome, Arnold-Chiari malformation and nail-patella syndrome.

“I just don’t think the evidence is there for these long lists,” said Dr. Molly Cooke, a professor of medicine at the University of California, San Francisco, who helped research a position paper on cannabis for the American College of Physicians. “It’s been so hard to study marijuana. Policy makers are responding to thin data.”

Even some advocates of medical marijuana acknowledge that the state laws legalizing it did not result from careful reviews of the medical literature.

“I wish it were that rational,” said Mitch Earleywine, chairman of the executive board of directors for Norml, a national marijuana advocacy group. Dr. Earleywine said state lawmakers more often ask themselves, “What disease does the person in a wheelchair in my office have?”

Research into marijuana’s effects is thin not because of a lack of scientific interest, but chiefly because the federal government has long classified it as a Schedule 1 drug with “no currently accepted medical use.” Scientists who want to conduct studies must register with the Drug Enforcement Administration, submit an investigational new drug application to the Food and Drug Administration for human trials, and win approval from the Department of Health and Human Services or one of the National Institutes of Health. The National Institute on Drug Abuse is the only supplier of legal, research-grade marijuana.

The legal and administrative hoops make it hard for investigators to start the randomized, placebo-controlled trials that are the gold standard of medical research and the basis for determining which drugs are effective, at what doses, and in which patients.

“It’s one thing to say we need to have more research, and it’s another thing to obstruct the research,” said Dr. Steven A. Jenison, former medical director of New Mexico’s medical cannabis program.

The dearth of data has not prevented legislators and voters across the nation from endorsing marijuana for more than 40 conditions. Patients with rheumatoid arthritis, for instance, qualify for marijuana treatment in at least three states.

Yet there are no published trials of smoked marijuana in rheumatoid arthritis patients, said Dr. Mary-Ann Fitzcharles, a rheumatologist at McGill University who reviewed the evidence of the drug’s efficacy in treating rheumatic diseases. “When we look at herbal cannabis, we have zero evidence for efficacy,” she said. “Unfortunately this is being driven by regulatory authorities, not by sound clinical judgment.”

New York considered including the chronic inflammatory disease on its list, a development that astonished Dr. Mary K. Crow, an arthritis expert at the Hospital for Special Surgery, in Manhattan. People with rheumatoid arthritis have higher rates of certain respiratory problems, she noted.

“Inhaling into your lungs is not a great idea with rheumatoid arthritis, given the substantial number of patients who have lung disease,” Dr. Crow said. (The final version of New York’s law prohibits smoking marijuana and did not end up including rheumatoid arthritis.)

In Arizona and Rhode Island, among other states, people with Alzheimer’s disease may receive medical marijuana to help quell nighttime agitation. But Dr. Gary Small, director of the division of geriatric psychiatry at the University of California, Los Angeles, said he does not recommend cannabis to Alzheimer’s patients: Agitation and increased confusion are possible side effects.

Still, he said he would not discourage a caregiver from providing it if it calmed a family member with dementia.

Parents of children with intractable epilepsy have lobbied hard in several states, including New York, for inclusion in medical marijuana legislation. They want access to an oil called Charlotte’s Web that is rich in CBD, a nonpsychoactive ingredient of marijuana that they say reduces the number of seizures.

This month, Gov. Rick Scott of Florida, a conservative Republican, signed a law allowing epilepsy patients access to the oil, calling it “the best treatment available.”

Scientists have begun randomized, placebo-controlled research to determine whether CBD effectively treats severe forms of childhood epilepsy. But at the moment, high-quality research showing that marijuana is a safe or effective treatment for epilepsy does not exist, experts say.


A green sludge made with marijuana is used for the caramels. Credit Matt Rainey for The New York Times

A green sludge made with marijuana is used for the caramels. Credit Matt Rainey for The New York Times

“As far as data out there, there are great animal models and very provoking anecdotes,” said Dr. Orrin Devinsky, director of the Comprehensive Epilepsy Center at NYU Langone Medical Center. “The human data is not there right now.”

Psoriasis was included in the New York legislation after representatives of Gaia Plant-Based Medicine, a Colorado company operating dispensaries, met with State Senator Diane J. Savino and suggested that cannabis lotions helped people with those red, raised skin plaques. It was dropped from the measure after questions were raised about the lack of supporting evidence — as were other conditions, like diabetes and lupus.

Medical marijuana advocates contend that suffering people should not have to wait for scientific research to catch up to patients’ needs. And why limit marijuana use to only certain conditions, they ask, when doctors routinely prescribe drugs off-label for anything they feel like?

Amanda Hoffman, 35, an information technology specialist in Basking Ridge, N.J., struggles with ulcerative colitis, an inflammatory bowel disease. She has tried steroids and Remicade, an intravenous infusion, but no drug has given her as much relief from frequent daily diarrhea and abdominal pain as her homemade cannabis caramels.

On a recent Sunday, Ms. Hoffman used a green buttery sludge made with marijuana she bought for $500 an ounce from Garden State Dispensary to make a new batch. She is grateful that the state legalized marijuana for patients like her, whatever the scientific evidence.

“Cannabis to a lot of people is a punch line, but it can also be a lifesaver,” she said.

Even if strong medical research regarding marijuana did exist, it is not clear that state lawmakers would be swayed.

“It would be possible to take case studies or anecdotal information from patients or research done from a university, put it in front of a legislator and say, ‘We need to include this disease,’ ” said State Representative Lou Lang, sponsor of the medical marijuana law in Illinois.

“But the legislative mind, be it in D.C. or in Springfield, Illinois, doesn’t always go to public policy,” Mr. Lang said. “The default position is politics.”

Often state legislators have been motivated not just by constituents in distress, but also by the desire to restrict access to limited patient populations so that legal marijuana does not become widely available as a recreational drug in their states.

For example, while there is research suggesting that marijuana alleviates certain kinds of chronic pain, Mr. Lang noted, legislators in Illinois were reluctant to legalize its use in such a broad patient population. The state’s list of qualifying conditions is lengthy partly because lawmakers tried instead to specify a number of diagnoses that result in pain, some quite rare.

“I’ll bet there are hundreds of conditions that cause pain, and now 30 are listed,” Karen O’Keefe, director of state policies at the Marijuana Policy Project, said of Illinois’s legislation.

Medical experts, rarely included in these statehouse discussions, have often been caught off guard by the sudden passage of laws permitting patients to ask them for marijuana.

Since at least 2009, for instance, the American Glaucoma Society has said publicly that marijuana is an impractical way to treat glaucoma. While it does lower intraocular eye pressure, it works only for up to four hours, so patients would need to take it even in the middle of the night to achieve consistent reductions in pressure. Once-a-day eye drops work more predictably.

Yet glaucoma qualifies for treatment with medical marijuana in more than a dozen states, and is included in pending legislation in Ohio and Pennsylvania. At one point, it appeared in New York’s legislation, too.

Dr. Paul N. Orloff, the legislative chairman for the New York State Ophthalmological Society, reached out to Richard N. Gottfried, a Manhattan Democrat and sponsor of the Assembly bill, and succeeded in getting glaucoma removed.

“It’s very illogical to prescribe a medication where it’s not standardized,” Dr. Orloff said, adding, “None of my 60-year-old patients are interested in being stoned to treat their glaucoma.”