Prevalence of cannabis consumption methods among people with medically recommended and nonmedical cannabis use in the United States

AIMS: Some cannabis consumption methods (e.g. smoking, vaping, dabbing) are associated with more harms than others (e.g. sublingual, transdermal). We aimed to examine differences in prevalence of cannabis consumption methods by people with medically recommended-only vs. nonmedical-only cannabis use. DESIGN AND SETTING: Cross-sectional, US nationally representative data from the 2022 National Survey on Drug Use and Health (NSDUH). PARTICIPANTS: 14 271 NSDUH participants aged ≥12 with past-year cannabis use. MEASUREMENTS: NSDUH asked respondents with past-year cannabis use about their consumption methods i.e. smoking, vaping, eating or drinking, dabbing (consuming a concentrated form with a dab rig/pen), oral/sublingual, transdermal or taking pills and if healthcare professionals recommended any cannabis use. If 'no', respondents were classified as having 'nonmedical-only cannabis use'. If 'yes', they were asked if all cannabis use was recommended and, if so, they were classified as having 'medically recommended-only cannabis use'; otherwise, they were classified as having 'medical and nonmedical cannabis use'. FINDINGS: Among people reporting cannabis use, 10.7% 95% confidence interval (CI) = 9.7%-11.7% reported medically recommended-only use, and 83.0% (95% CI = 81.7%-84.3%) reported nonmedical-only use. Smoking was the predominant method for nonmedical-only use (79.2%, 95% CI = 77.7%-80.7%) and for medically recommended-only use (73.7%, 95% CI = 68.8%-78.2%). Vaping prevalence (45.6%, 95% CI = 40.9%-50.4%) and eating/drinking prevalence (45.7%, 95% CI = 40.7%-50.7%) were similar among people with medically recommended-only use. Vaping prevalence was 1.3 times higher adjusted prevalence ratio (APR) = 1.3, 95% CI = 1.2-1.4 for medically recommended-only than for nonmedical-only use (38.3%, 95% CI = 32.7%-44.3%). Dabbing prevalence was 1.5 times higher (APR = 1.5, 95% CI = 1.3-1.7) for medically recommended-only than for nonmedical-only use 24.4% (95% CI = 20.5%-28.8%) vs 16.4% (95% CI = 15.3%-17.7%). Higher prevalence of oral/sublingual and transdermal methods was associated with medically recommended-only use. CONCLUSIONS: Among people with past-year medically recommended-only cannabis use in the US, three-fourths reported smoking cannabis and almost half reported vaping and eating/drinking cannabis. Cannabis vaping prevalence and dabbing prevalence were 1.3-1.5 times higher, respectively, among people with medically recommended-only use than people with nonmedical-only use.