Marijuana use among American high school students has fallen to the lowest level recorded in the history of the Centers for Disease Control and Prevention’s national Youth Risk Behavior Survey, adding to a growing body of federal data showing that adolescent cannabis use remains historically low even as adult-use legalization has expanded across the United States. According to newly released 2025 YRBS data, 12.9 percent of U.S. high school students reported using marijuana at least once during the previous 30 days, down sharply from 17 percent in 2023. The latest rate is also below the 14.7 percent recorded when the national survey began tracking current marijuana use in 1991.
The decline is especially striking when placed against the evolution of state cannabis laws. In 2011, one year before Colorado and Washington became the first states to legalize marijuana for adults, 23.1 percent of high school students reported past-month use. The new 12.9 percent figure is about 44 percent lower. Twenty-four states, along with the District of Columbia and several territories, now permit adult nonmedical cannabis use. That national comparison does not prove that legalization caused adolescent marijuana use to decline; the relationship between cannabis policy and youth behavior remains actively debated in the scientific literature. What the federal data do establish is narrower but important: the broad expansion of adult legalization has not coincided with the nationwide surge in teen marijuana use that some earlier policy forecasts anticipated.
The 2025 YRBS Shows a Steep Two-Year Drop
CDC defines “current marijuana use” as using marijuana one or more times during the 30 days before completing the survey. The 2025 questionnaire specifically asked students how many times they had used marijuana during that period, while instructing them not to count CBD-only or hemp products. That makes the 12.9 percent figure a measure of relatively recent marijuana consumption, not simply whether a teenager had experimented with cannabis at any point in life.
The fall from 17 percent in 2023 to 12.9 percent in 2025 represents a decline of roughly 24 percent in only two years. Earlier CDC surveys show that current use stood at 21.7 percent in 2019 and dropped to 15.8 percent in 2021 during the COVID-19 period before rising modestly to 17 percent in 2023. The newest data therefore push adolescent marijuana consumption substantially below both the immediate pre-pandemic rate and the previous post-pandemic low.
Looking further back makes the shift even clearer. Current marijuana use climbed from 14.7 percent in 1991 to a peak of 26.7 percent in 1999, then generally declined over the following decades. It measured 20.8 percent in 2009, 23.1 percent in 2011, 23.4 percent in 2013, 21.7 percent in 2015 and 19.8 percent in 2017. The 2025 estimate of 12.9 percent is therefore lower than any comparable national YRBS measurement since the survey began.
The Survey Represents U.S. High School Students Nationwide
The Youth Risk Behavior Survey is one of the federal government’s longest-running systems for monitoring adolescent health behaviors. It surveys students in grades nine through 12 attending public and private schools and uses a complex probability sample designed to produce nationally representative estimates. Participation is anonymous and voluntary, with students completing a self-administered questionnaire.
For the 2025 national survey, 11,097 students at 108 schools completed questionnaires, and 10,710 questionnaires remained usable after CDC data editing. The school response rate was 47 percent and the student response rate was 65.2 percent, resulting in an overall response rate of 30.6 percent. CDC statistically weights the data to account for the sampling design and nonresponse, but the participation rate remains an important limitation when interpreting any individual estimate. Self-reported substance-use surveys can also be affected by inaccurate recall or a willingness to disclose illegal behavior.
The YRBS also represents students who are attending school, which means adolescents who have dropped out are not captured. That matters because youth disconnected from school may have different substance-use patterns. Researchers studying marijuana policy using YRBS data have repeatedly identified this as a limitation.
Two Other Major Federal Surveys Show the Same Overall Direction
The CDC findings do not stand alone. Results from the federal 2025 National Survey on Drug Use and Health, released by the Substance Abuse and Mental Health Services Administration in July, found that past-month marijuana use among adolescents aged 12 to 17 declined from 6.1 percent in 2021 to 5.0 percent in 2025. Past-year marijuana use and marijuana initiation also declined over the same period. SAMHSA estimated that roughly 300,000 fewer adolescents initiated marijuana use in 2025 than in 2021.
Monitoring the Future, another major national survey supported by the National Institute on Drug Abuse, produced similar results. In 2025, 8 percent of eighth graders, 16 percent of 10th graders and 26 percent of 12th graders reported cannabis use during the previous year. NIH said the rates remained around historically low levels reached after 2020. The long-running Monitoring the Future report found that past-year marijuana use among 10th and 12th graders in 2025 was at its lowest level in roughly three decades.
These surveys should not be compared as if they measure exactly the same population or behavior. YRBS surveys students in grades nine through 12 and the headline measure concerns the past 30 days. NSDUH surveys household residents aged 12 to 17, while Monitoring the Future separately samples eighth, 10th and 12th graders and often reports past-year use. Nevertheless, three independently administered national surveillance systems pointing broadly downward strengthens confidence that the decline is not simply an artifact of one particular questionnaire.
Did Legalization Cause Teen Marijuana Use to Fall?
The national trend cannot answer that question by itself. Twenty-four states currently permit adult recreational use, compared with none before Colorado and Washington approved legalization measures in 2012. Seeing adolescent consumption fall over that same period establishes a temporal coexistence, not cause and effect. Many other changes have occurred simultaneously, including shifts in alcohol and nicotine use, social behavior, parental monitoring, school environments, youth attitudes, retail controls and the COVID-19 pandemic.
Peer-reviewed studies of state legalization have produced differing results. A 2024 JAMA Psychiatry analysis using national and state YRBS data from 1993 through 2021 found no evidence that recreational marijuana laws increased either current or frequent teen marijuana use. A separate JAMA Pediatrics analysis of nearly 900,000 students similarly found no net increase associated with recreational legalization or regulated retail sales and reported some modest decreases in adolescent cannabis use after legalization.
Other analyses are less reassuring. A 2024 systematic review and meta-analysis covering 30 quantitative studies found that medical cannabis laws were not significantly associated with youth past-month use, but recreational legalization was associated with a modest increase in the odds of youth cannabis use, with larger estimated effects among young adults than adolescents. The authors emphasized substantial variation between jurisdictions and study designs.
Taken together, the research does not support a simple conclusion that adult legalization either universally increases or reduces teen use. Market regulations, retail density, advertising rules, product availability, age enforcement, prices and local social conditions can differ dramatically from one state to another.
The Pandemic Appears to Have Played a Major Role
One of the clearest clues about the current low rates comes from the timing. Adolescent drug use fell dramatically between 2020 and 2021, when school closures, reduced social interaction and greater time at home disrupted many of the environments where teenagers normally obtained and used substances. NIH researchers have repeatedly highlighted that period as an important turning point. Unlike many predictions that use would rebound quickly when schools reopened, most adolescent substance-use rates stayed near those lower levels through 2025.
A 2026 JAMA Network Open study provides an illustration of how legalization and the pandemic can pull trends in different directions. Researchers examined more than 1.3 million adolescent cannabis screening questionnaires in Northern California from 2011 through 2024. Cannabis use initially rose following California’s recreational legalization and the opening of legal retail sales, but dropped sharply after the pandemic began and continued declining afterward, reaching roughly prelegalization levels by the end of 2024. The investigators concluded that both cannabis policy and the much broader social disruption associated with COVID-19 appeared relevant to adolescent behavior.
That study helps explain why national trends should not be treated as a referendum on legalization alone. A state policy change can have a measurable local association even while a powerful nationwide shift pushes overall use downward.
Marijuana Use Is Down, but Cannabis Products Are Changing
Another complication is that modern youth cannabis use no longer consists solely of conventional marijuana flower. Vape cartridges, concentrated extracts, edibles and intoxicating hemp-derived cannabinoids have expanded the range of available products. The CDC’s 2025 YRBS marijuana questions explicitly told students not to count CBD-only or hemp products, meaning the record-low marijuana figure should not automatically be interpreted as a complete measure of exposure to every intoxicating cannabinoid product.
Monitoring the Future provides a useful warning. A 2026 analysis found that marijuana use among 12th graders declined between 2023 and 2024, while delta-8-THC use remained essentially unchanged. Among adolescents who were already using cannabis products, marijuana and delta-8 co-use was common. In 2025, Monitoring the Future separately reported past-year use of hemp-derived intoxicating cannabis products among 2 percent of eighth graders, 6 percent of 10th graders and 9 percent of 12th graders.
The changing product environment is one reason researchers increasingly argue that youth surveillance needs to track individual cannabinoids and routes of administration rather than relying exclusively on a single question about “marijuana.”
Lower Prevalence Does Not Make Adolescent Cannabis Harmless
The declining national rate is encouraging from a public-health perspective, but roughly one in eight high school students reporting past-month marijuana use still represents a substantial population. CDC notes that adolescent brains continue developing into young adulthood and that cannabis affects functions involving learning, memory, attention, decision-making and coordination. Earlier initiation and heavier use are also associated with greater risk of cannabis use disorder.
Recent research also suggests that frequency matters. A national analysis of more than 160,000 adolescents found progressively stronger associations between increasing cannabis-use frequency and academic problems, poor self-regulation and adverse emotional indicators. That does not mean cannabis alone caused every observed outcome, but it reinforces the importance of distinguishing occasional experimentation from weekly or near-daily use when evaluating youth trends.
This is also why researchers studying legalization increasingly focus not only on how many teenagers use cannabis, but on whether the smaller group that does use it is consuming more frequently or using higher-potency products.
A Record Low With an Important Policy Caveat
The new CDC numbers provide the clearest evidence yet that teen marijuana use has not risen nationally alongside the spread of state adult-use cannabis laws. Past-month use among U.S. high school students has fallen from 23.1 percent immediately before the first recreational legalization measures in 2012 to 12.9 percent in 2025, and independent federal surveys show similarly low or declining adolescent marijuana consumption.
But the data do not demonstrate that legalization itself produced the decline. Studies attempting to isolate legalization’s effects range from finding no increase or modest declines to identifying small increases in youth use in some settings. The pandemic appears to have produced an unusually large nationwide disruption in adolescent substance use, and new hemp-derived intoxicants further complicate surveillance.
The most defensible conclusion from the 2025 federal data is therefore narrower than either side of the legalization debate might prefer: American teenagers are reporting marijuana use at historically low levels during an era in which adult cannabis legalization has become widespread. Determining exactly why that decline occurred—and whether it persists as legal cannabis markets mature—will require continued surveillance of marijuana, vaping, concentrates, edibles and intoxicating hemp products rather than treating adolescent cannabis use as a single static behavior.
