Cannabis Mode of Use, Tobacco Co-Use, and Respiratory Risk

TL;DR: Adults who co-use tobacco with cannabis use combustible cannabis more often and in more ways, and bear a disproportionate respiratory and cannabis-use burden compared to those who use cannabis alone.

Abstract

Objective: This study assessed differences in the prevalence and correlates of different cannabis modes of administration (MOAs) between those reporting co-use (past 30-day nicotine/tobacco and cannabis use) and cannabis-only use.

Method: Participants were adults reporting past 30-day cannabis use (n = 3,488) drawn from a larger sample of 12,855 Oklahoma adults recruited into a 7-wave repeated cross-sectional online survey (September 2020 to September 2024). Prevalence and relative rankings of nine cannabis MOAs (smoking, vaping, dabbing, eating, drinking, oil, lotion, dissolvable, blunt use) were examined across cannabis-only users and co-users. ANCOVA tests assessed associations of co-use (vs cannabis-only use) and combustible MOA category (combustible-only, non-combustible-only, both) with respiratory symptoms and past 30-day frequency of medical and recreational cannabis use.

Results: Smoking was the most common MOA among cannabis-only (72%) and co-users (83.1%). Among co-users, blunt use was second (64.8%), followed by vaping (44%) and edibles (36.5%). Among cannabis-only users, edibles ranked second (47.8%), then blunts (36.1%), then vaping (33.4%). Combustible MOA use and co-use were each associated with more severe respiratory symptoms and more frequent past 30-day medical and recreational cannabis use.

Conclusions: Those who co-use nicotine/tobacco with cannabis and those consuming combustible MOAs may be at increased risk of adverse health consequences and more frequent cannabis use.

Study at a Glance

Study Snapshot

Study Facts Table

What Researchers Actually Did

The investigators recruited Oklahoma adults aged 18 and older into a 7-wave repeated cross-sectional online survey between September 2020 and September 2024, using a commercial panel (Lucid) that employs age, sex, and race/ethnicity quotas calibrated to Oklahoma census data. From 12,855 total respondents, 3,488 who reported past 30-day cannabis use were selected for analysis. Participants were categorized as cannabis-only users (23.8%, n = 830) or cannabis-tobacco co-users (76.2%, n = 2,658) based on any past 30-day use of cigarettes, large cigars, little cigars/cigarillos, e-cigarettes, or other tobacco products. Critically, blunt use was treated as a cannabis MOA variable, not as a criterion for co-use classification.

Nine cannabis MOAs were assessed using items adapted from the Behavioral Risk Factor Surveillance System: smoked, ate, drank, vaped, dabbed, dissolved, applied to skin, blunt use, and some other way. Respiratory symptom severity was measured with the 8-item American Thoracic Society Questionnaire (ATSQ), and probable cannabis use disorder (CUD) was assessed using the 3-item CUDIT-SF (scores of 2 or higher defined as probable CUD). Bivariate analyses and hierarchical logistic regression models (with backward selection of covariates in Step 1) examined MOA differences by co-use status. Separate ANCOVA models then evaluated the main effects of co-use status and a 3-category combustible MOA variable (combustible-only, non-combustible-only, or both) on respiratory symptom severity and past 30-day medical and recreational cannabis use frequency.

Key Findings: Primary Outcomes

• Smoking prevalence: Combusted/smoked cannabis was the most common MOA in both groups: 83.1% of co-users and 72.0% of cannabis-only users (AOR for co-use vs cannabis-only: 1.81, 95% CI: 1.44, 2.27, p < .001).
• Blunt use: The second most common MOA among co-users (64.8%) but third among cannabis-only users (36.1%); co-use was associated with more than twice the odds of blunt use (AOR 2.44, 95% CI: 1.97, 3.02, p < .001).
• Vaping: 44.1% among co-users vs 33.4% among cannabis-only users (AOR 1.24, 95% CI: 1.00, 1.53, p < .05).
• Dabbing: 34.0% among co-users vs 19.0% among cannabis-only users (AOR 1.89, 95% CI: 1.45, 2.47, p < .001).
• Edibles: More common among cannabis-only users (47.8%) than co-users (36.5%); co-use was associated with lower odds of edible use (AOR 0.45, 95% CI: 0.36, 0.55, p < .001).
• Dissolvable use: Co-use was associated with lower odds (AOR 0.68, 95% CI: 0.48, 0.95, p < .05).
• Respiratory symptom severity (ANCOVA): Co-users reported significantly worse adjusted mean ATSQ scores than cannabis-only users (17.98 vs 15.20; F(1,3447) = 81.70, p < .001).
• Past 30-day medical cannabis use frequency: Higher among co-users (adjusted mean 18.03 days vs 16.20 days; F(1,3448) = 14.79, p < .001).
• Past 30-day recreational cannabis use frequency: Higher among co-users (adjusted mean 12.57 days vs 10.20 days; F(1,3449) = 16.83, p < .001).

Key Findings: Secondary Outcomes and Subgroup Analyses

• Combustible MOA category and respiratory symptoms: Participants using both combustible and non-combustible MOAs had the highest adjusted ATSQ scores (17.75) compared to combustible-only (16.41) and non-combustible-only users (16.0), with both comparisons significant at p < .001. Combustible-only and non-combustible-only groups did not differ significantly from each other (p > .05).
• Combustible MOA category and cannabis use frequency: Those using both combustible and non-combustible MOAs reported the highest adjusted mean days of medical (19.28 days) and recreational (14.07 days) cannabis use. Non-combustible-only users reported the lowest: 12.25 days medical and 5.45 days recreational. All pairwise differences among the three groups were statistically significant (p < .001) for both medical and recreational use.
• Probable CUD: More prevalent among co-users (53.5%) than cannabis-only users (41.6%; p < .001).
• Number of MOAs used: Co-users used more MOAs per month than cannabis-only users (mean 3.07 vs 2.52, including blunts; F(1,2983) = 55.15, p < .001).
• Medical cannabis license: A greater proportion of cannabis-only users held an Oklahoma medical cannabis license (67.8% vs 55.1%; p < .001).
• Demographics of co-use: Co-users were more likely to be male, ages 25 to 54, employed full- or part-time, and to report lower educational attainment and income below $20,000.

Results: Adverse Events and Safety Profile

This is an observational survey study; no adverse event data were collected in a clinical trial sense. The study measured self-reported respiratory symptom severity as a health outcome. Co-users reported a mean ATSQ score of 18.01 (SD = 8.27) compared to 14.89 (SD = 6.77) for cannabis-only users at the bivariate level (F(1,3486) = 97.55, p < .001). No data on objectively measured pulmonary function, emergency care utilization, or diagnosed respiratory disease were captured.

Statistical Approach and Rigor

The study used ANCOVA to assess main effects of co-use status and combustible MOA category on three outcomes, with Bonferroni-adjusted pairwise comparisons for the 3-level MOA variable. Hierarchical logistic regression with backward covariate selection identified model-specific confounders in Step 1 before co-use status was entered in Step 2. Survey weighting was not applied despite use of demographic quotas during recruitment, which introduces potential residual confounding. The repeated cross-sectional design prevents within-person analysis across waves, and the same participant could contribute data at multiple time points; the paper does not clearly indicate whether repeated participants were handled analytically. The Lucid panel is a non-probability sample; even with quota-based recruitment, non-response bias and self-selection cannot be excluded. Effect sizes are not reported beyond adjusted means and odds ratios.

Clinical Takeaway

For the clinician managing a cannabis-using patient, the findings from this study translate to a clear pattern: tobacco co-users are far more likely to be using cannabis through combusted routes, including blunts and pipes, and this combustible use burden compounds respiratory risk. Roughly three in four adults reporting cannabis use in this Oklahoma sample also reported past 30-day tobacco use. That proportion is high even by national standards. Blunt use, which involves wrapping cannabis in tobacco leaf cigar wraps, was present in 64.8% of co-users, including 36.1% of patients who explicitly denied tobacco use, raising a real clinical documentation problem. Any intake screen that asks about cannabis use must also specifically ask about blunt use and combustion mode, not simply whether a patient uses cannabis.

Clinical Bottom Line: In adults who use both cannabis and tobacco, combustible administration modes dominate and are independently associated with worse respiratory outcomes and higher cannabis use frequency.

Why This Matters Clinically

Cannabis use is no longer a marginal behavior in the US. With 16% of adults reporting past-month use in 2023, and state-level rates exceeding 21% in permissive states like Oklahoma, clinicians encounter cannabis-using patients routinely across virtually every specialty. The available evidence has consistently shown that mode of administration is not a trivial pharmacokinetic detail: combusted cannabis delivers pyrolysis toxicants, irritants, and carbon monoxide in ways that parallel combusted tobacco. This study adds granularity by showing that co-users, the majority of cannabis users, rely predominantly on combusted delivery, amplifying both toxicant exposure and symptom burden. The adjusted mean ATSQ respiratory score difference of approximately 2.8 points between co-users and cannabis-only users, achieved after controlling for wave, cannabis harm perceptions, age, medical and recreational use days, and probable CUD, suggests a clinically meaningful signal, though the observational design cannot confirm causation. The finding that using both combustible and non-combustible MOAs simultaneously is associated with the highest respiratory symptom severity, exceeding combustible-only use, argues against the assumption that adding non-combustible options reduces harm when combusted use continues in parallel.

CED Clinical Relevance

At CED Clinic, where patients routinely present seeking guidance on optimizing cannabis for symptom management, these findings reinforce several counseling priorities. First, patients who report any tobacco or nicotine use warrant a specific conversation about blunt use, which bridges both substance categories and is frequently undercounted in standard intake screens. Second, patients who describe using multiple delivery formats simultaneously, for example, vaping during the day and smoking in the evening, may be exposing themselves to the highest respiratory risk category identified in this study. Third, those using cannabis primarily through non-combustible routes (edibles, tinctures, topicals) and achieving symptom control represent an achievable clinical target, and this population reported the lowest medical and recreational use frequency and the lowest respiratory symptom burden. Mode optimization, specifically guiding patients away from combustion, remains one of the most actionable harm-reduction interventions available to the prescribing clinician.

Fits What We Already Know

This study situates within a consistent literature that the paper itself cites. Nationally representative data from the NSDUH have shown that approximately 77% of past-month cannabis users also report past-month tobacco use (Schauer et al., 2015), closely mirroring the 76.2% co-use rate observed here. Population Assessment of Tobacco and Health (PATH) data have shown that combustible tobacco is the most common administration form among co-users (Cohn & Chen, 2022), and the present findings extend this to demonstrate that combustible cannabis similarly dominates the co-use profile. Prior work has linked blunt use to a greater number of cannabis and alcohol use disorder symptoms (Cohn et al., 2016). Swan et al. (2021) documented that combustible cannabis MOAs are associated with more severe CUD symptoms and worse psychosocial outcomes compared to non-combustible forms, consistent with what this study’s ANCOVA results suggest. The finding that using multiple MOAs simultaneously is associated with higher cannabis use frequency and problems aligns with Bedillion et al. (2022), who showed that multi-mode cannabis use predicts greater hazardous use and consequences than single-mode use. The misclassification problem, where some blunt users deny tobacco use, has been documented by Morean et al. (2023) and Delnevo et al. (2011), and this study replicates that discrepancy.

What This Study Teaches Us

The study teaches three things with reasonable confidence. First, if a patient uses tobacco, they are very likely smoking cannabis too, and almost certainly through multiple combustible routes. Second, the profile of cannabis-only users is meaningfully different: they rely more on edibles and are more likely to hold formal medical cannabis licenses, suggesting a more deliberate and possibly harm-conscious use pattern. Third, using cannabis through more delivery routes simultaneously, rather than replacing one with another, is associated with the heaviest use burden and the worst respiratory symptom profile. The intuitive assumption that switching from smoking to also using edibles reduces harm is not supported here; what matters is whether combusted use continues.

What It Does Not Show

• The cross-sectional design cannot establish whether combusted MOA use or tobacco co-use causes respiratory symptoms, or whether individuals with pre-existing respiratory conditions gravitate toward certain use patterns.
• The study does not include objective pulmonary function measures; all respiratory data are self-reported symptom frequency via the ATSQ.
• Cannabis product potency, THC and CBD content, smoking topography (depth of inhalation, hold time), and daily quantity consumed were not measured.
• Cognitive outcomes, cardiovascular events, and cancer incidence are outside the scope of this study.
• The Oklahoma sample, drawn from a state with unusually high dispensary density, loose licensing requirements, and elevated tobacco use prevalence, may not represent the US cannabis-using population broadly.
• The study cannot determine whether the 36.1% of cannabis-only users who reported blunt use were consuming cannabis-only blunt wraps or tobacco cigar wraps; the blunt use variable is not disaggregated by wrap type.
• Temporal sequencing of co-use initiation relative to MOA adoption cannot be established.

Fits the Broader Conversation

As cannabis legalization expands and product diversity increases, the field has been rightly moving away from treating “cannabis use” as a monolithic exposure. This paper contributes a rare direct comparison of nine specific MOAs across co-users and cannabis-only users in a large, multi-wave sample with real clinical outcome data. That the authors identified independent main effects of both co-use status and combustible MOA type on respiratory symptoms, even after adjustment, adds to the growing evidence base that public health messaging must move beyond drug-level risk communication toward route-specific and co-use-specific framing. The finding that co-users perceive greater cannabis harm yet report better general health than cannabis-only users is a textbook illustration of cognitive dissonance under active substance use, and one that public health communicators and clinicians should understand when designing behavior change interventions. The paper also draws attention to a meaningful surveillance gap: blunt use is often miscounted because consumers may not accurately categorize cigar wraps as tobacco products, a misclassification problem that distorts both co-use prevalence estimates and respiratory risk assessments.

Teaches Us: Families

If someone you care about uses cannabis and also smokes cigarettes or uses other tobacco products, this study suggests

Physician-Led, Whole-Person Care

A doctor who takes the time to truly understand you.

Personal care that starts with listening and is guided by experience and ingenuity.

Health, Longevity, Wellness

One-on-One Cannabis Guidance

Metabolic Balance

Leave a Message
Metabolic Care
Medical Consulting
Cannabis Care