One in Nine Older Adults Who Use Cannabis Meet Criteria for a Disorder. Nobody Is Screening Them.

Two numbers sit at the center of a study published Sept. 28 in the journal Addiction, and neither one sounds like an emergency. Among Americans 65 and older, 8.9 percent used cannabis in the past year. Among those users, 11.4 percent met the criteria for cannabis use disorder.

Put differently: roughly one in eleven older adults uses the drug, and about one in nine of them has a pattern of use that causes problems or interferes with daily life. That second figure is the one the researchers — a team led by the UC San Diego School of Medicine, with co-authors from Duke, NYU and Rutgers Health — want clinicians to stop walking past.

What 21,189 Interviews Can and Cannot Show

The study draws on the National Survey on Drug Use and Health, pooling four survey years from 2021 through 2024 into an analysis of 21,189 adults aged 65 and older. It is a nationally representative sample, and it is also a self-report instrument: respondents describe their own behavior, and researchers apply the DSM-5 checklist afterward. Nothing here was confirmed by a physician or a toxicology screen. A study of this design can measure how many people describe symptoms consistent with a disorder. It cannot measure how many are harmed in ways they do not recognize.

The symptom profile is worth reading line by line, because it separates ordinary recreational use from something stickier. Nearly three-quarters of the group with cannabis use disorder — 74.4 percent — said they spent a great deal of time obtaining, using or recovering from cannabis. About two-thirds, 67.5 percent, reported craving. Just under half, 48.3 percent, described tolerance, meaning they needed more to get the same effect.

Severity skews mild. Of those who met criteria, 76 percent fell into the mild category, 20.1 percent moderate and 3.9 percent severe. That distribution is the study's most actionable finding: most of this is early, which is exactly the stage at which a conversation with a primary care physician can still change the trajectory.

Smoking remains the dominant route of administration, reported by 65.8 percent of past-year users, followed by eating or drinking cannabis products at 40.5 percent and vaping at 15.4 percent. And 19.9 percent said a doctor had recommended some or all of their cannabis use — meaning a sizable share of older users arrived at the habit through the medical system rather than around it.

The 300-Day Signal

One result stands apart from the rest. Older adults who used cannabis on at least 300 days out of the past year were almost four times more likely to meet criteria for cannabis use disorder than those who used it less often. The relationship between frequency and disorder is not a straight line, but the top of the range is where the risk concentrates — and daily, year-round use is exactly the pattern that becomes invisible in a fifteen-minute annual exam.

Why Aging Changes the Arithmetic

Cannabis behaves differently in a 72-year-old body than in a 22-year-old one, and the reasons are mostly physiological rather than behavioral. Body composition shifts, lean mass falls, and drug clearance slows. Chronic conditions accumulate, along with the prescriptions that treat them.

"That's really alarming," said Dr. Scott Hadland, an addiction physician at Mass General Brigham, describing the prevalence figure to GBH News. He pointed to a specific mechanism: ingested cannabis can interact with medications an older patient is already taking and contribute to falls or dizziness, particularly alongside other sedating drugs.

That mechanism is not hypothetical. Prior research has linked cannabis use in older adults to chronic disease, cognitive impairment, falls, cannabis poisoning and cannabis-related emergency department visits. What had been missing was a population-level estimate of how many older users cross from use into disorder. That is the gap this study fills, and the title of the paper — "Cannabis Use and Cannabis Use Disorder among Older Adults in the United States" — is deliberately unglamorous about it.

The Case That the Number Is Inflated

There is a serious argument that 11.4 percent overstates the problem, and it deserves a hearing rather than a dismissal.

Dr. Peter Grinspoon, an addiction specialist at Massachusetts General Hospital, told GBH that cannabis use disorder is overdiagnosed in this population. His reasoning turns on how the DSM-5 counts. A patient needs only two of eleven criteria for a diagnosis. Tolerance and withdrawal supply both of them. For opioids, the manual includes a carve-out: patients on prescribed opioids who show tolerance and withdrawal are not automatically classified as addicted, because those effects follow from the medication itself. Cannabis has no equivalent exclusion.

"If these medical patients have tolerance and withdrawal — that is two out of the 11," Grinspoon said. "That's enough to give them a diagnosis of cannabis use disorder."

Hadland agrees with part of that, and the study itself pushes back on part of it. About two-thirds of the older adults who met criteria reported strong cravings — a symptom that is not a pharmacological side effect but a disruptive experience in its own right. Hadland also put the finding in context: across all U.S. cannabis users, roughly one in three develop cannabis use disorder, so the older-adult rate is actually lower than the general population's. His concern is not the magnitude but the neglect. This is a group, he said, "that often goes overlooked" by clinicians who assume an older patient should simply be left to make his own choices.

Massachusetts Votes on Repeal in November

The study landed about a month before Massachusetts voters weigh Ballot Question 8, which would repeal the state's 2016 legalization of recreational cannabis sales and restrict possession. That timing has pulled the research into a live political argument, though neither specialist treats the study as a verdict on legalization.

Grinspoon opposes repeal. His case is practical: arrests of minorities are down, dispensary products are safer than illicit-market products, legal access gives people a route to the drug without a criminal record, and lower stigma makes it easier to seek help when use becomes a problem. Hadland was candid that he does not know whether Question 8 would change rates of cannabis use disorder either way, and he noted that medical cannabis access in Massachusetts would survive the vote regardless. His attention was on a different question — whether wider availability is producing more problematic use, or whether more people are simply using.

What Screening Would Actually Look Like

The study authors' recommendation is unglamorous: screen older adults for unhealthy cannabis use, and make cannabis part of routine clinical conversation as availability widens and medical use grows. That is a systems problem more than a knowledge problem. Screening tools exist. What is missing is the habit of asking — and the coverage to act on an answer.

The comorbidity data make the case for integrating that question into mental health care as well. Cannabis use disorder was more common among older adults with past-year mental illness, tobacco use or use of drugs other than cannabis. Roughly one in three people with any substance use problem also carries a co-occurring psychiatric condition, which is why dual diagnosis care treats the two together rather than in sequence.

What This Means for Treatment

There is no FDA-approved medication for cannabis use disorder. Treatment rests on behavioral approaches — cognitive behavioral therapy, motivational enhancement, contingency management — and on the same continuity-of-care logic that governs every other substance use disorder.

The optimistic reading of this study is that three-quarters of the cases it found are mild. Mild cases are the ones most likely to respond to a brief intervention, a referral and follow-up, before the pattern hardens. For families trying to work out whether an older parent's daily use has crossed a line, the symptoms to watch are the ones the study measured: time consumed by the drug, craving, escalating dose.

That is a lower bar for intervention than most people assume — and a reminder that marijuana dependence is not a young person's diagnosis. It is a diagnosis that older adults receive less often because fewer people think to look for it.