In one of the first studies to use systematic, self-administered screening for cannabis use among a large primary care clinical population and conducted in a state with legalized medical and recreational cannabis, we found important connections between depression and anxiety diagnoses, prescription treatment, cannabis use, and risk for cannabis use disorder. Cannabis use was common and many patients both with and without diagnosed depression or anxiety reported using cannabis to manage mental health-related symptoms including stress, worry, and sleep disturbances. Among those with diagnosed depression or anxiety, the use of cannabis to manage these symptoms remained high even among patients currently prescribed antidepressant or anxiolytic medications. This suggests that cannabis may be serving as a complementary or alternative approach to conventional treatment for many patients with depression or anxiety. These findings highlight a critical need for routine, informed conversations between patients and providers about cannabis use—particularly among those with depression and anxiety—to ensure satisfactory, safe, and effective care.

Approximately one-fifth of primary care patients with depression or anxiety diagnoses reported cannabis use in the past three months, with one-tenth scoring at moderate-to-high risk of CUD. These prevalences are similar to those identified in prior clinical samples and general population studies assessing depression and anxiety symptoms or diagnoses and cannabis use (Lapham et al. 2017; Kedzior and Laeber 2014; Hunt et al. 2020). Our results also echo prior non-clinical studies that demonstrate that people with depression or anxiety face increased risk of CUD (Onaemo et al. 2021). Our analysis contributes novel findings that compared to other primary care patients, patients with both depression and anxiety diagnoses had the highest prevalence of cannabis use and elevated risk of CUD. Patients with both diagnoses had nearly three times the odds of screening at moderate-to-high risk for CUD compared to those with neither diagnosis, demonstrating a heightened risk that held after adjustment for potential confounders. Altogether, our results suggest that primary care patients diagnosed with depression or anxiety, and particularly patients with comorbid depression and anxiety, may be a priority population for assessing indication of CUD and counseling on cannabis use.

Our study utilized data from a universal cannabis use screener to offer novel insight into the self-reported reasons for cannabis use shared by patients with and without depression and/or anxiety diagnoses. Among cannabis users with these diagnoses, upwards of two-thirds reported using to manage mental health symptoms, which is substantially higher than estimates from a meta-analysis among medical cannabis users (Kosiba et al. 2019). Notably, this pattern did not differ by antidepressant or anxiolytic prescription status. These findings underscore the need for routine discussion of reasons for cannabis use to inform counseling and mental health treatment in primary care. For example, these results may suggest that some patients are using cannabis as a supplement to, or substitute for, prescribed treatments, possibly due to side effects, perceived lack of efficacy, or a desire to reduce medication use altogether (Kvamme et al. 2021; McNabb et al. 2023). The elevated use of cannabis to manage mental health-related symptoms, including among patients already receiving pharmaceutical treatment, may signal undertreated symptoms, gaps in care, or alternative treatment preferences. Importantly, cannabis use alongside psychotropic medications can cause drug-drug interactions, depending on the cannabinoids used and the mode of administration (Vaughn et al. 2021). While this analysis could not account for medication adherence or whether cannabis was medically recommended (Hunt et al. 2020), the findings point to a need for nuanced and proactive clinical engagement about cannabis use that may be better informed by understanding a patient’s reasons for use.

The findings of this study should be interpreted in light of a number of limitations. This was an analysis of adult primary care patients in a large health system in Los Angeles, CA, and results may not apply to other regions or populations. Additionally, the cross-sectional design also means that we cannot distinguish temporality between depression and anxiety diagnoses, symptoms, and cannabis use. It is possible, for example, that cannabis withdrawal may contribute to symptoms of depression and anxiety in some patients, which may contribute to ongoing cannabis use. Since patients may seek mental healthcare from behavioral services externally, some depression and anxiety diagnoses or relevant prescriptions may not be captured in the EHR. We did not have validated mental health symptom screening data, which may have added insight into the patients’ current mental health symptoms. Acknowledging that depression or anxiety may be misclassified or underdiagnosed in a clinical setting, using ICD-10 diagnoses collected from the problem and visit list allowed us to identify clinically relevant cases. Furthermore, using EHR data allowed us to conduct a novel comparison of cannabis use characteristics collected by universal, electronic patient screening by both clinical diagnosis and prescription history, which has yet to be conducted in such a large patient population. The ASSIST is not a diagnostic tool for CUD, so the more conservative cut-off score of 8 points was used to better identify patients who have experienced negative consequences of cannabis use. Patients may not report cannabis use due to stigma and privacy concerns even in a state with legalized access (McNeely et al. 2018). Finally, the presence of a prescription for antidepressant or anxiolytic medication does not necessarily mean that the patient has filled the prescription, is taking the medication, or is adhering to the regimen as prescribed. Despite these limitations, the cannabis screening data provides a more wholistic picture of cannabis use than is available in most large healthcare systems, offering insight into the overlap of cannabis use for mental health symptom management, depression and anxiety diagnoses, and relevant prescription records.