Why Regular Cannabis Users Have Lower Rates of Type 2 Diabetes

Cannabis makes you hungry. This is, by now, one of those facts that barely needs stating. And yet chronic cannabis users are, on average, thinner than non-users. They’re also roughly half as likely to develop type 2 diabetes. The contradiction has been sitting in the epidemiological literature for years, quietly irritating anyone who takes metabolism seriously, and nobody has had a particularly satisfying explanation for it. A study published in The Journal of Physiology this week doesn’t fully resolve the paradox, but it does suggest something genuinely interesting about what’s going on inside fat tissue.

Nicholas DiPatrizio, who directs the cannabinoid research centre at the University of California, Riverside, suspected the answer might lie not in THC itself but in the dozens of other compounds that occur alongside it in the plant. To test this, his team ran an experiment that was straightforward in design if not in implication: take obese mice fed a high-fat, high-sugar western diet, and treat one group with pure delta-9 THC, another with a whole-plant cannabis extract matched for exactly the same THC dose. Then see what happens.

Two Very Different Outcomes

Both treatments caused significant weight loss. Fat mass dropped, body weight fell, and the mice in both groups ended up noticeably leaner than untreated obese controls. So far, so similar. But when the researchers looked at blood glucose, the two groups diverged sharply. Mice given pure THC still had impaired glucose regulation after 30 days of treatment, the kind of dysregulation that, in humans, signals the early stages of type 2 diabetes. The whole-plant extract mice, by contrast, had glucose clearance that had returned to levels seen in lean, healthy animals. Same THC dose, same amount of weight lost, completely different metabolic outcome.

“This suggests that THC alone is not responsible for the metabolic benefits associated with cannabis use,” DiPatrizio said. “Other compounds in the plant appear to play a critical role.”

What those other compounds are doing, specifically, involves a signalling axis that doesn’t get much attention outside specialist circles. Fat tissue isn’t simply a passive storage depot; it’s an endocrine organ. Adipocytes secrete hormones (leptin, adiponectin, adipsin among others) that communicate with the pancreas, helping regulate how much insulin gets released and how sensitive other tissues are to it. In obesity, this conversation becomes garbled. Fat cells enlarge, their secretion patterns shift, and the signals reaching the pancreas start misfiring. The result is progressively worsening insulin resistance, eventually tipping into overt diabetes.

Restoring the Signal

The UCR team found that the whole-plant extract did a substantially better job of restoring these adipokine signals than THC alone. Leptin levels, chronically elevated in obesity, fell more in extract-treated mice. Adipsin, which tends to drop in metabolic disease, recovered more completely. Gene expression in visceral fat tissue showed a pattern that was, in various ways, closer to healthy lean animals in the extract group than in the THC group. The researchers interpret this as evidence that the extract is repairing something specific in the crosstalk between fat and pancreas, the so-called adipoinsular axis, in a way that THC alone cannot manage.

The endocannabinoid system is also implicated. The body produces its own cannabis-like molecules and has receptors for them throughout fat tissue, gut, and brain; in obesity, this system becomes dysregulated, with endocannabinoid levels running chronically high. Chronic cannabinoid exposure in the mice appeared to partially reset this tone. The logic is a bit circular, but it points toward something: the plant’s full chemical portfolio might be doing something more subtle than any single compound can replicate on its own.

The extract used in the study was relatively complex, containing not just THC but tetrahydrocannabivarin, cannabinol, cannabigerol, and small amounts of CBD. Earlier work on rodents and some limited human studies has flagged THCV and CBD as candidates with their own effects on glucose metabolism, so the question of which specific compounds are responsible for the glucose improvement remains genuinely open. DiPatrizio’s lab aims to isolate and test them individually. That work could potentially identify a non-psychoactive compound capable of delivering the metabolic benefit without the high, which would be a rather useful thing to have.

The Munchies Problem, Revisited

There’s also the question of how weight loss happens at all. Appetite suppression in the early days of treatment is plausible; food intake did dip transiently in the first week. But intake normalised and the mice kept losing weight anyway, without any change in physical activity. Something in the way their bodies were partitioning energy must have shifted. The study didn’t measure whole-body energy expenditure directly, which the authors acknowledge, and fat cell bioenergetics in vitro shifted in complex, dose-dependent ways that don’t map neatly onto a simple “more burning” story.

The researchers are careful not to overstate the conclusions. The mice were male, which matters because the endocannabinoid system differs between sexes and so, presumably, might the metabolic response. The animals were also housed at temperatures slightly below their thermoneutral zone, which keeps them burning more energy to stay warm anyway, possibly amplifying metabolic effects that wouldn’t be as pronounced in other conditions. And none of this is in humans yet. “We’re not suggesting people should use cannabis to manage weight or diabetes,” DiPatrizio said, which is probably a statement worth repeating.

What the study does do is give the long-standing epidemiological observation a plausible mechanistic skeleton. The population data showing lower diabetes rates in cannabis users has been sitting there for the better part of two decades; now there’s a biological story, however incomplete, for why that might be. Whether it holds in humans, which compounds are responsible, and what the risks of chronic cannabinoid exposure are alongside any benefits remain entirely open questions. “Clinicians, researchers, and policymakers should stay tuned and pay attention to this space,” DiPatrizio said. “We need evidence-based approaches to fully understand both the risks and potential benefits of cannabis and its components.” Cannabis policy is shifting fast, use rates are climbing, and the clinical picture of long-term exposure is still being written.

https://doi.org/10.1113/JP290431

Frequently Asked Questions

Why do cannabis users tend to weigh less despite the munchies effect?

Population studies consistently show that regular cannabis users have lower rates of obesity and type 2 diabetes than non-users, even though cannabis is known to stimulate appetite acutely. The new research from UC Riverside suggests this paradox may be explained by non-THC compounds in the plant affecting how fat tissue communicates with the pancreas, potentially improving metabolic function in ways that offset increased food intake over time.

What is the adipoinsular axis and why does it matter for diabetes?

The adipoinsular axis refers to the signalling network between fat cells and the pancreas. Fat tissue releases hormones such as leptin, adiponectin, and adipsin that help regulate insulin secretion and sensitivity. In obesity, this communication breaks down, contributing to rising blood sugar and eventually type 2 diabetes. The UC Riverside study found that whole-plant cannabis extract restored this signalling more effectively than pure THC, which may explain why the extract improved glucose tolerance while THC alone did not.

Is this research enough to suggest cannabis as a diabetes treatment?

No, and the researchers are explicit about this. The study was conducted in male mice, not humans, and whole-body energy expenditure was not directly measured. Sex differences in cannabinoid signalling mean the results may not translate equally to female subjects, and clinical trials in humans would be needed before any therapeutic recommendations could be made. The findings are best understood as mechanistic groundwork, not clinical guidance.

Which compounds in cannabis might be responsible for the glucose benefits?

The extract used in the study contained several cannabinoids beyond THC, including tetrahydrocannabivarin (THCV), cannabinol (CBN), cannabigerol (CBG), and small amounts of CBD. Previous research has suggested THCV and CBD may independently influence glucose metabolism in rodent models, but identifying the specific compound or combination responsible for the improved glucose clearance seen here is the next step for DiPatrizio’s lab.

Could this research lead to a non-psychoactive treatment for metabolic disease?

That is precisely what DiPatrizio is working toward. If the metabolic benefits can be traced to non-THC cannabinoids that don’t produce intoxication, it may eventually be possible to develop targeted therapies that improve insulin sensitivity and fat tissue function without the psychoactive effects of THC. That work is still in early preclinical stages, but the principle is sound, and the new findings give it a clearer mechanistic direction to follow.