Cannabis use is increasing globally with an estimated 219 million users (4.3% of the global adult population) in 2021, but its long-term metabolic effects remain unknown. While some studies have suggested potential anti-inflammatory or weight management properties, others have raised concerns regarding glucose metabolism and insulin resistance, and the magnitude of the risk for developing diabetes hasn’t been clear.

To strengthen the evidence base, Dr Ibrahim Kamel from the Boston Medical Center, Massachusetts, USA and colleagues analyzed electronic health records from 54 healthcare organizations (TriNetX Research Network, with centers from across USA and Europe) to identify 96,795 outpatients (aged between 18 and 50 years, 52.5% female) with cannabis-related diagnoses (ranging from occasional use to dependence, including cases of intoxication and withdrawal) between 2010 and 2018.

They were matched with 4,160,998 healthy individuals (with no record of substance use or major chronic conditions) based on age, sex, and underlying illnesses at the start of the study, and followed for 5 years.

After controlling HDL and LDL cholesterol, uncontrolled high blood pressure, atherosclerotic cardiovascular disease, cocaine use, alcohol use and several other lifestyle risk factors, the researchers found that new cases of diabetes were significantly higher in the cannabis group (1,937; 2.2%) compared to the healthy group (518; 0.6%), with statistical analysis showing cannabis users at nearly four times the risk of developing diabetes compared to non-users.

While the authors note that more research is needed to fully explain the association between cannabis and diabetes, it may come down to insulin resistance and unhealthy dietary behaviours. Nevertheless, the study's results have immediate implications for metabolic monitoring practices and public health messaging.

“As cannabis becomes more widely available and socially accepted, and legalized in various jurisdictions, it is essential to understand its potential health risks,” said lead author Dr Kamel. “These new sights from reliable real-world evidence highlight the importance of integrating diabetes risk awareness into substance use disorder treatment and counseling, as well as the need for healthcare professional to routinely talk to patients about cannabis use so that they can understand their overall diabetes risk and potential need for metabolic monitoring.”

The authors note that more research is needed on the long-term endocrine effects of cannabis use and whether diabetes risks are limited to inhaled products or other forms of cannabis such as edibles.

Despite the important findings, this is a retrospective study and cannot prove that cannabis use causes diabetes, and the authors cannot rule out the possibility that other unmeasured factors may have influenced the results despite efforts to reduce confounding bias via propensity score matching. This study has limitations due to lack of detailed cannabis consumption data and potential misclassification. The authors acknowledge inherent limitations of real-world data often result from inconsistent patient reporting in electronic medical records. They also note that there is a risk of bias because of imprecise measures of cannabis exposure and the reliance on participants to accurately report any cannabis use, even when they lived in places where the drug is illegal.

The study of tens of thousands of people in Denmark found that those with a BMI in the overweight category - and even some of those living with obesity - were no more likely to die during the five years of follow-up than those with a BMI of 22.5-<25.0 kg/m², which is at the top end of the normal weight range.

Individuals with a BMI in the middle and lower parts of the normal weight range 18.5 to <22.5kg/m², were also more likely to die. As were individuals with a BMI in the underweight range.

"Both underweight and obesity are major global health challenges," says Sigrid Bjerge Gribsholt, of the Steno Diabetes Center Aarhus, Aarhus University Hospital, Aarhus, Denmark, who led the research. "Obesity may disrupt the body's metabolism, weaken the immune system and lead to diseases like type 2 diabetes, cardiovascular diseases and up to 15 different cancers, while underweight is tied to malnutrition, weakened immunity and nutrient deficiencies.

"There are conflicting findings about the BMI range linked to lowest mortality. It was once thought to be 20 to 25 but it may be shifting upward over time owing to medical advances and improvements in general health."

To provide some clarity, Dr Gribsholt, Professor Jens Meldgaard Bruun, also of the Steno Diabetes Center Aarhus, and colleagues used health data to examine the relationship between BMI and mortality in 85,761 individuals (81.4% female, median age at baseline 66.4 years).

BMI is a measure of weight to height and a score of 18.5 to <25 kg/m² is generally considered to be of normal weight. A BMI of <18.5 kg/m² is categorized as underweight, 25 to <30 kg/m² is considered overweight and a BMI of 30 kg/m² is described as obesity.

7,555 (8%) of the participants died during follow-up. The analysis found that individuals in the underweight category were almost three times more likely (2.73 times) to have died than individuals with a BMI towards the top of the healthy range (22.5 to <25.0 kg/m², the reference population).

Similarly individuals with BMI of 40 kg/m² and above (categorized as severe obesity) were more than twice as likely (2.1 times) to have died compared with the reference population.

However, higher mortality rates were also found for BMIs that are considered healthy.

Individuals with a BMI of 18.5 to <20.0 kg/m², and so at the lower end of the healthy weight range, were twice as likely to have died as those in the reference population. Similarly, those with a 20.0 to <22.5 kg/m², and so in the middle of the healthy weight range, were 27% more likely to have died than the reference population.

By contrast, individuals with a BMI in the overweight range (25 to <30 kg/m²) and those with a BMI at the lower part of the obese range (30.0 to <35.0 kg/m²) were no more likely to have died than the those in the reference population - a phenomenon sometimes referred to as being metabolically healthy or "fat but fit."

Those with a BMI of 35 to <40.0 kg/m² did have an increased risk of death of 23%.

All of the results were adjusted for sex, comorbidity level and education level.

A similar pattern was obtained when the researchers looked at the relationship between BMI and obesity in participants of different ages, sexes and levels of education.

The researchers were surprised to find that BMI was not associated with a higher mortality up to a BMI of 35 kg/m² and that even a BMI 35 to <40 kg/m² was only associated with a slightly increased risk.

Dr Gribsholt says: "One possible reason for the results is reverse causation: some people may lose weight because of an underlying illness. In those cases, it is the illness, not the low weight itself, that increases the risk of death, which can make it look like having a higher BMI is protective.

"Since our data came from people who were having scans for health reasons, we cannot completely rule this out.

"It is also possible that people with higher BMI who live longer - most of the people we studied were elderly - may have certain protective traits that influence the results.

"Still, in line with earlier research, we found that people who are in the underweight range face a much higher risk of death."

Whatever the explanation, BMI isn't the only indicator that an individual is carrying unhealthy levels of fat, says Professor Bruun.

He explains: "Other important factors include how the fat is distributed. Visceral fat - fat that is very metabolically active and stored deep within the abdomen, wrapped around the organs - secretes compounds that adversely affect metabolic health.

"As a result, an individual who has a BMI of 35 and is apple-shaped - the excess fat is around their abdomen - may have type 2 diabetes or high blood pressure, while another individual with the same BMI may free of these problems because the excess fat is on their hips, buttocks and thighs.

"It is clear that the treatment of obesity should be personalized to take into account factors such as fat distribution and the presence of conditions such as type 2 diabetes when setting a target weight."