Cannabis Use Reported by Patients Receiving Primary Care in a Large Health System

Importance Despite the changing legal status of cannabis and the potential impact on health, few health systems routinely screen for cannabis use, and data on the epidemiology of cannabis use, and especially medical cannabis use among primary care patients, are limited. Objective To describe the prevalence of, factors associated with, and reasons for past–3 month cannabis use reported by primary care patients. Design, Setting, and Participants This cross-sectional study used electronic health record data from patients aged 18 years and older who had an annual wellness visit between January 2021 and May 2023 from a primary care clinic within a university-based health system in Los Angeles, California. Exposures Factors of interest included age, race and ethnicity, sex, employment status, and neighborhood Area Deprivation Index (ADI). Main Outcomes and Measures Cannabis use was assessed using the Alcohol Substance Involvement Screening Test (ASSIST). Patients were also asked about reasons for use, symptoms for which they used cannabis, and mode of use. Results Among the 175 734 patients screened, the median (range) age was 47 (18-102) years; 101 657 (58.0%) were female; 25 278 (15.7%) were Asian, 21 971 (13.7%) were Hispanic, and 51 063 (31.7%) were White. Cannabis use was reported by 29 898 (17.0%), with 10 360 (34.7%) having ASSIST scores indicative of moderate to high risk for cannabis use disorder (CUD). Prevalence of cannabis use was higher among male patients than female patients (14 939 20.0% vs 14 916 14.7%) and younger patients (18-29 years, 7592 31.0%; ≥60 years, 4200 8.5%), and lower among those who lived in the most disadvantaged neighborhoods (ADI decile 9-10, 189 13.8%; ADI decile 1-2, 12 431 17.4%). The most common modes of use included edibles (18 201 61.6%), smoking (15 256 51.7%), and vaporizing (8555 29.0%). While 4375 patients who reported using cannabis (15.6%) did so for medical reasons only, 21 986 patients (75.7%) reported using cannabis to manage symptoms including pain (9196 31.7%), stress (14 542 50.2%), and sleep (16 221 56.0%). The median (IQR) number of symptoms managed was 2 (1-4), which was higher among patients who were at moderate to high risk for CUD (4 2-6 symptoms). Conclusions and Relevance In this study, cannabis use and risk of CUD were common, and more than three-quarters of patients who reported any cannabis use reported doing so to manage a health-related symptom. These findings suggest that integration of information regarding cannabis use for symptom management could help provide a crucial point-of-care opportunity for clinicians to understand their patients’ risk for CUD.