Cannabis Use Was Linked to Lower Odds of Self-Reported Cognitive Decline

A 2023 study found that adults who reported using cannabis for nonmedical reasons had lower odds of reporting worsening confusion or memory loss than people who did not use cannabis. But the finding does not show that marijuana prevents cognitive decline, improves memory, or reduces the risk of Alzheimer’s disease.

The study, published in Current Alzheimer Research, analyzed data from 4,744 U.S. adults aged 45 and older who participated in the 2021 Behavioral Risk Factor Surveillance System. Researchers from SUNY Upstate Medical University examined whether the reason for cannabis use, frequency of use, or consumption method was associated with subjective cognitive decline, or SCD.

SCD is based on a person’s perception that confusion or memory loss has become more frequent or is getting worse. It is not the same as a clinical diagnosis of mild cognitive impairment, dementia, or Alzheimer’s disease. The Centers for Disease Control and Prevention’s BRFSS cognitive-decline module is designed to measure this type of self-reported change for public-health surveillance; it is not a diagnostic test.

What the study found

After adjusting for demographic characteristics, health conditions, mental health, smoking, alcohol use, and other factors, participants who reported nonmedical cannabis use had an adjusted odds ratio of 0.04 for SCD compared with nonusers. The researchers described this as 96% lower odds.

That estimate should be interpreted cautiously. Only about 2.1% of the weighted sample reported using cannabis for nonmedical reasons, and the confidence interval was wide. Cannabis use for medical reasons, or for both medical and nonmedical reasons, was also associated with lower odds in the analysis, but neither result was statistically significant.

The study did not find statistically significant associations between SCD and either cannabis-use frequency or method of consumption. In some unadjusted comparisons, SCD was more common among cannabis users, particularly people who smoked cannabis. The fully adjusted smoking estimate also pointed toward higher odds of SCD, but the result was not statistically significant and was imprecise.

Why the result does not establish a protective effect

This was an observational analysis of survey data, not a randomized trial. The researchers measured cannabis use and self-reported cognitive symptoms at the population level, so the study cannot determine whether cannabis caused the lower odds of SCD.

People who use cannabis recreationally may differ from nonusers in ways that are difficult to measure or fully adjust for. The groups may differ in health, stress, sleep, access to medical care, socioeconomic circumstances, or willingness to report memory concerns. Reverse causation is also possible: people experiencing health problems or cognitive symptoms may choose cannabis for medical reasons, which could help explain why medical users did not show the same pattern as nonmedical users.

The authors proposed that effects on sleep or stress might contribute to the association, but those explanations were not tested directly. The study did not measure sleep improvement, stress reduction, cannabis dose, THC or CBD concentration, or objective performance on memory and thinking tests.

Other research has reported different results. For example, a population-level study of adults over 50 found that past-year cannabis users were more likely to report subjective memory complaints, although the association weakened after adjustment for other factors. Differences in the surveys, definitions of cannabis use, populations, and statistical methods may help explain the conflicting findings.

What the findings mean

The SUNY Upstate study is best viewed as hypothesis-generating evidence that the relationship between cannabis and perceived cognitive health may be more complicated than a simple benefit-or-harm conclusion. Its strongest result concerned nonmedical use and self-reported symptoms—not objectively measured cognition or future dementia risk.

Anyone experiencing persistent or worsening memory problems should discuss them with a health-care professional. As the National Institute on Aging explains, memory changes can have many causes, some of which may be treatable. More rigorous longitudinal studies and clinical trials will be needed to determine whether particular cannabis products, doses, or patterns of use affect cognitive function over time.