In the United States, the number of people who use cannabis almost every day has, for the first time, surpassed the number of people who consume alcohol with the same frequency. This finding reinforces the notion that patterns of substance use are changing in some countries. In Brazil, however, the trend is moving in the opposite direction: heavy alcohol consumption continues to rise, particularly among women and young adults, and alcohol remains by far the psychoactive substance responsible for the greatest burden of disease and mortality.
For decades, alcohol has been the dominant recreational substance in Western culture, while cannabis played a relatively minor role. In some countries, this landscape is changing. Population-based surveys in the United States show that heavy cannabis use has increased rapidly, whereas frequent alcohol consumption has declined. Although these trends are important from a public health perspective, they should not be directly extrapolated to the Brazilian context.
The most influential analysis on this topic was published by Jonathan Caulkins of Carnegie Mellon University in the journal Addiction.¹ Drawing on data from 27 waves of the leading U.S. household survey on substance use, encompassing more than 1.6 million participants between 1979 and 2022, the study identified a remarkable shift. In 1992, there were approximately ten times more people who drank alcohol daily or nearly every day (8.9 million) than people who used cannabis with the same frequency (fewer than one million). By 2022, this pattern had reversed for the first time: 17.7 million individuals reported daily or near-daily cannabis use, compared with 14.7 million who drank alcohol at the same frequency.
Consumption intensity reveals another important distinction. Although far more people drink alcohol than use cannabis, heavy-frequency use is less common among drinkers. Approximately 40% of current cannabis users report daily or near-daily use, a pattern that more closely resembles tobacco use than alcohol consumption. Caulkins also notes that these trends parallel changes in drug policy: cannabis use declined during periods of stricter prohibition and increased during phases of liberalization, particularly following recreational legalization in several U.S. states.
Part of this trend has been interpreted through the lens of substitution—the idea that some individuals may be replacing alcohol with cannabis. This concept has become widely known through the expression California sober, referring to individuals who abstain from alcohol while continuing to use cannabis. Until recently, this hypothesis was supported primarily by observational studies and anecdotal reports.
In 2025, researchers at Brown University published the first experimental evidence on this topic in the American Journal of Psychiatry.² In a randomized, double-blind, placebo-controlled clinical trial, 157 adults with heavy alcohol use who also consumed cannabis at least twice per week participated in a laboratory experiment designed to simulate a bar environment. After smoking cannabis containing active THC, participants consumed between 19% and 27% less alcohol than after receiving placebo cannabis. They also reported reduced craving for alcohol and waited longer before taking their first drink.
The investigators, however, emphasized the need for caution. The observed effect was acute, short-lived, and measured under controlled laboratory conditions. It does not demonstrate that cannabis is a safe treatment for alcohol-related problems. At present, no clinical guidelines recommend cannabis as a strategy for reducing alcohol consumption. Cannabis carries its own health risks, including the potential for dependence, and many individuals with cannabis use disorder also meet diagnostic criteria for alcohol use disorder. Replacing one substance with another is therefore not a risk-free solution.
While frequent alcohol consumption has declined in the United States, Brazil presents a markedly different—and concerning—picture. Although the country has followed the global trend of decreasing per capita alcohol consumption, heavy episodic drinking continues to increase.
According to Vigitel, the Brazilian Ministry of Health's nationwide surveillance system, the prevalence of Heavy Episodic Drinking (HED)—defined as consuming four or more standard drinks on a single occasion for women and five or more for men—increased from 15.7% in 2006 to 20.4% in 2024 across Brazil's state capitals.
This increase has not been uniform. Among men, heavy drinking has remained consistently high, fluctuating between approximately 25% and 27%. The most significant change has occurred among women, whose prevalence nearly doubled over two decades, rising from 7.8% in 2006 to 15.2% in 2023, with particularly marked increases among younger women. Growth has also been concentrated among younger adults and individuals with higher educational attainment, indicating that the problem is not confined to traditionally vulnerable populations.
Comparing these two scenarios helps avoid misleading conclusions. The notion that "the younger generation drinks less," while supported by some recent Brazilian surveys, should be interpreted cautiously. In Brazil, alcohol remains the psychoactive substance responsible for the greatest burden of disease, contributing to chronic illnesses, traffic crashes, violence, and preventable deaths. Moreover, the most harmful pattern of consumption—heavy episodic drinking—is increasing among several population groups despite an overall decline in average alcohol consumption.
Regulatory and cultural contexts also differ substantially. In the United States, the expansion of cannabis use has occurred alongside legalization and an ongoing public debate regarding the relative harms of different substances. In Brazil, alcohol remains deeply embedded in social life and cultural celebrations, while public perception of its risks remains relatively low. This cultural normalization helps explain why harmful drinking patterns persist—and, in some groups, continue to increase—even as overall alcohol consumption declines.
The first step is to avoid viewing alcohol as a public health problem that is "in decline." Brazilian data suggest the opposite and underscore the need for continued vigilance. Monitoring drinking patterns—not merely average alcohol consumption—is essential, as heavy episodic drinking accounts for a substantial proportion of alcohol-related acute harms, even among individuals who do not drink regularly.
Prevention efforts should also expand their focus to populations that have traditionally received less attention, including women and adults with higher educational attainment, among whom heavy drinking has increased significantly. Public health messages should reach these audiences without reinforcing stigma that may discourage individuals from seeking help.
Finally, international findings on the relationship between cannabis and alcohol warrant careful interpretation. They provide valuable insights into behavioral and biological mechanisms and deserve continued scientific investigation. However, they do not support simplistic conclusions or substitution-based strategies. In Brazil, the public health priority remains unchanged: reducing heavy alcohol consumption through evidence-based policies, as alcohol continues to be the substance responsible for the greatest burden on population health.
References: