Gallup reported on August 24 that 17% of U.S. adults say they smoke marijuana, tying the organization’s 2023 record. Fifteen percent say they consume marijuana edibles, while 11% report smoking cigarettes.

The poll used telephone interviews with 1,200 adults from July 1 through July 19. Gallup reports a plus or minus 4.0 percentage-point sampling margin for the full sample, and notes that wording and practical survey difficulties can introduce additional error.

The national estimate does not define whether a person’s cannabis use is safe, medically appropriate, or problematic. Route, amount, potency, timing, other substances, health conditions, and functional effects matter more for an individual decision.

People who smoke cannabis should discuss cough, wheeze, breathlessness, asthma, tobacco use, and alternative routes with a clinician. Edibles avoid smoke but introduce delayed effects and dosing challenges, so they are not automatically risk free.

A binary cannabis question misses the distinction highlighted by the poll. Ask separately about smoking, vaping, edibles, concentrates, tinctures, dose, frequency, medical purpose, tobacco co-use, and adverse effects.

Screening should remain nonjudgmental and clinically proportionate. Population prevalence does not diagnose cannabis use disorder, but increasing normalization makes it easier to overlook tolerance, withdrawal, impaired control, unsafe driving, psychiatric vulnerability, or respiratory symptoms.

Gallup first asked its current marijuana-smoking question in 2013, when 7% answered yes. The 2026 estimate is more than double that figure, while cigarette smoking has moved in the opposite direction over decades.

The trends reflect changing behavior and social context, but the poll cannot identify why they changed. Legal access, perceived risk, generational replacement, medical use, product availability, tobacco control, and survey disclosure may all contribute.

The poll does not measure dose, potency, frequency, impairment, dependence, respiratory disease, medical benefit, or reasons for use. It does not establish that marijuana smoking caused cigarette use to fall.

Smoking and edible percentages may overlap because respondents could report both. Adding 17% and 15% would therefore overstate the share of adults who use cannabis.

Combustion exposes the respiratory tract to smoke and irritants. Oral products avoid combustion but have slower onset, longer duration, and greater risk of accidental overconsumption when users redose too soon.

Extra caution is appropriate around driving, pregnancy, adolescence, psychosis vulnerability, cardiovascular symptoms, sedating combinations, child access, and jobs that involve safety-sensitive tasks.

Record high is a trend label, not proof of a statistically meaningful increase from every recent year. The 17% result ties 2023 and sits within a survey margin of error.

Telephone self-report can be affected by nonresponse, disclosure, memory, and interpretation of the word smoke. Subgroup percentages have wider uncertainty than the total-sample estimate.

Public-health surveillance should distinguish route, frequency, medical purpose, co-use, product potency, and impairment. A single smoking question is useful for a trend line but too coarse for planning clinical services or evaluating harm.

Policy debates should not use prevalence alone as a proxy for success or failure. Better outcomes include fewer preventable injuries, less youth exposure, accurate labeling, informed adult use, accessible treatment, and reduced smoke-related burden.

Seventeen percent reported smoking marijuana, while 15% reported consuming edibles. These questions may capture overlapping groups, so the percentages should not be added to estimate total cannabis use.

The contrast with cigarettes is culturally important, but clinical comparisons require exposure detail. Combusted cannabis and tobacco have different pharmacology, yet both smoke routes expose lungs to combustion products.

A prevalence number becomes clinically useful only after the next questions are asked. What product is being used, by which route, at what dose and frequency, for what goal, and with what effects on sleep, mood, breathing, cognition, driving, work, and relationships?

The poll should make cannabis conversations more routine, not more moralizing. A calm history can identify helpful use, avoidable smoke exposure, poorly controlled symptoms, escalating patterns, tobacco co-use, or impairment without assuming that every person represented by the 17% has the same experience.

Four distinctions that matter