The legalization of cannabis and the start of retail sales of the drug in the US are linked to both a rise in its recreational use and concurrent use of tobacco, as well as a fall in sole tobacco use, finds an analysis of health behavioral data, published online in the journal Tobacco Control.

These patterns were evident in higher risk groups, such as those with poor mental health, as well as those with historically lower levels of recreational cannabis use, such as older people and those with higher levels of education, the findings indicate.

Concurrent use of tobacco and cannabis is associated with an increased risk of developing dependence, and emerging evidence suggests that this combo has greater health consequences than use of either substance alone, point out the researchers.

US states have been legalizing cannabis for recreational use since 2012, and as of July 2025, 24 states plus DC had done so for adults aged 21+. Legalization has coincided with increased use of the drug among adults, amid an overall fall in tobacco use, they explain.

But it’s not clear if the legislative change has prompted a rise in concurrent or sole use of tobacco, and to find out, they drew on data from the annual Behavioral Risk Factor Surveillance System, run by the Centers for Disease Control and Prevention and state health departments.

In all, they included information from 874,084 respondents aged 18+ in the 38 states that asked questions about cannabis use from 2016 through 2023 (average annual response rate by state, 44% to 50%). After excluding respondents with missing data, the final analysis included the responses of 854,878 adults.

Respondents were asked how many days during the past 30 they had used marijuana or hashish. They were asked separately about the frequency of current use of tobacco and vapes.

Based on the responses, the researchers created 4 categories of past month use: none; tobacco use only; cannabis use only; and concurrent use of tobacco and cannabis.

They tracked the year of cannabis legalization and the date of the opening of the first licensed recreational cannabis retail store in each of the 38 states.

And they accounted for potentially influential factors: age (5 year groupings); binary sex; educational attainment; race and ethnicity; marital status; health insurance; children under 18 in the household; alcohol use; and poor mental health.

Analysis of the data showed that sole tobacco use varied widely across states from 8% in Massachusetts to 26% in West Virginia. Sole cannabis use ranged from just over 3% in West Virginia to 14% in Vermont and concurrent use of tobacco and cannabis ranged from just over 3% in Utah to 9.5% in Alaska.

From 2016 through 2023, the prevalence of sole tobacco use and sole cannabis use remained relatively stable at 14% and 7%, respectively, while concurrent use increased from just over 4% to just under 5.5%.

Overall, patterns indicated greater use of either substance or concurrent use among higher risk groups: adults who were younger, male, single, had lower levels of education, didn’t have health insurance and who reported poor mental health and binge/heavy drinking.

Compared with no use of either substance, the likelihood of sole cannabis use and concurrent use increased by 88% and 44%, respectively, after legalization, while the likelihood of sole tobacco use fell by 13%.

Sole cannabis use rose across all age groups, with larger increases among 18–24 year olds at just over 6% and those aged 65+ (2.5%.

Other distinct demographic patterns emerged following the legislation. The prevalence of sole tobacco use fell more among women than among men, while cannabis use rose among adults with a high school or higher level of education from just over 3.5% to just under 5%. Similarly, concurrent use of both cannabis and tobacco rose from nearly 1% to around 2%.

Sole cannabis use also increased among adults who identified as White, Black, Hispanic and Other, rising from nearly 3% to 4.5%.

Larger increases in sole cannabis use occurred among adults who reported poor mental health than among those reporting good mental health, while sole tobacco use fell more among adults who reported poor mental health. Larger falls in sole tobacco use also occurred among those who reported heavy/binge drinking.

Further analysis of the impact of retail sales revealed increases in sole cannabis use and concurrent use of tobacco and cannabis.

This is an observational study, and as such no definitive conclusions can be drawn about cause and effect, added to which the researchers acknowledge various limitations to their findings.

For example, the Behavioral Risk Factor Surveillance System doesn’t collect information on the timing, frequency, or method of using tobacco and cannabis together, and it relies on self-report. And as only 38 states participated in it, the findings may not apply to all states.

Nevertheless, the researchers suggest: “Our findings add to growing evidence that adults are shifting product use in response to cannabis legalization.” 

They conclude: “These shifting patterns may be narrowing disparities in use due to rising levels among populations less likely to use cannabis rather than declining use among higher-risk groups.”

And they warn that their findings, and those of other researchers, “suggest that altering the policy environment through legalization, increasing access to cannabis products through retail stores, and increases in public perceptions of cannabis safety could lead to increases in co-use of tobacco and cannabis through multiple pathways even with downward pressure on tobacco use from increased policy restrictions.”