Kokot and his team analyzed the health records of millions of patients in 64 healthcare organizations across the United States over a 20-year period.

Of the patients included in the study, 116,076 were diagnosed with CUD, while almost 4 million were not.

Patients had no history of head and neck cancer. However, there were some demographic differences between patients with and without CUD.

The CUD group was younger, with a mean age of 46 years, compared to a mean age of 60 years in the other group. There were also slightly fewer women among those diagnosed with CUD (44.5% versus 54.5%).

The researchers then examined the risk of developing any form of head and neck cancer as well as localized cancer in both groups.

To support a possible association, the study considered several time frames: 1 year and 5 years between cancer onset and CUD diagnosis, as well as any time period between the two events.

Depending on the time frame, patients with CUD were 3.5 to 5 times more likely to develop any form of head and neck cancer.

For laryngeal cancer, a form of throat cancer, the association with CUD was strongest, with the risk being more than eight times higher.

Oropharyngeal cancer, a cancer of the throat that affects the throat, showed an almost fivefold increase. These results are similar to previous studies that have shown the risk of Smoking tobacco and the risk of head and neck cancer.

For other cancers, including oral cavity, salivary gland and nasopharyngeal carcinoma, the risk was two to three times higher.

Despite the strength of the results, the study has some significant limitations, which Kokot acknowledged.

Hospital medical records do not contain specific information about a patient’s frequency of cannabis use, dose, or method of administration, only that they have been diagnosed with CUD.

The authors also note that CUD is likely related to tobacco and alcohol use, a potential confounding factor. Although they attempted to control for these variables, they note that “differences in dosing…may persist” between those diagnosed with CUD and those who were not.

John B. Sunwoo, MD, director of head and neck cancer research at Stanford Medicine, told Healthline the study was solid but acknowledged it had limitations. Sunwoo was not involved in the research.

“They tried to control for other factors by taking the two different cohorts and balancing all sorts of variables like smoking and alcohol. But alcohol and smoking consumption were proportionally higher in the cannabis consumption group,” Sunwoo said.

Because the study relied only on medical records indicating that a patient had CUD, additional details about cannabis use, such as frequency, strength, and route of consumption, are not available. As a result, the different ways of consuming cannabis, such as smoking or eating, could not be distinguished.

“As far as actually looking at the type of use, for example edible versus inhaled, or the amount per day or the amount per week and the number of years, we did not have that information in the database we examined,” Kokot said.

Questions about whether smoking cannabis is more harmful than using it remain unanswered. There is also no clear threshold for the frequency of cannabis use and no association with an increased risk of head and neck cancer.

“We cannot draw any conclusions about these factors,” said Kokot. “The next step in the research would require additional information to really analyze the level of risk for different types and amounts of consumption.”

Tobacco is considered biggest risk factor for the development of head and neck cancer. Alcohol consumption is associated with up to 5-fold increased risk for various forms of head and neck cancer.

People who consume both alcohol and tobacco have a significantly higher risk of developing head and neck cancer.

Studies examining the link between cannabis and head and neck cancer have been inconsistent, said Paul Armentano, deputy director of the National Organization for the Reform of Marijuana Laws (NORML). This lack of solid evidence is also acknowledged by the study authors.

For example Study 2015 No link was found between cannabis use and the development of head and neck cancer. Study from 2009found a inverse relationship between the twosuggesting that “moderate” cannabis use was associated with a reduced risk of developing head and neck cancer.

“Given these historically inconsistent results and the potential confounding factor, further research is needed before definitive conclusions can be drawn,” Armentano told Healthline.

“In the meantime, those who regularly use cannabis – such as patients who use cannabis for medicinal purposes – may wish to consider alternative administration methods that either reduce or avoid the ingestion of combustion smoke,” he noted.

Sunwoo was of a similar opinion. “The study has its limitations and I think people should be aware of those limitations,” he said. “But I also think it should encourage people to investigate this further because we just don’t know.”

CUD is recognized as a psychiatric disorder, However, the criteria for diagnosis may be unclear, leaving the diagnosis largely at the discretion of the physician.

To meet the criteria, a patient must meet 2 of 11 criteria. These include:

• Ingestion of large amounts of cannabis
• Cannabis cravings
• Withdrawal symptoms when not consuming cannabis
• persistent desire to reduce with unsuccessful attempts
• continued consumption despite social or interpersonal problems
• tolerance

Frequency or extent of consumption are not objectively defined within the framework of the CUD diagnosis.

According to new research, cannabis use disorder (CUD) is associated with a 3.5 to 5-fold increased risk of developing head and neck cancer.

For certain local cancers, such as those of the throat (larynx), the risk associated with CUD is up to eightfold increased.

Previous studies on the link between cannabis use and these cancers have been inconsistent. More studies are needed to determine whether cannabis use might have the same effect on head and neck cancer risk as smoking.