(A-221) Self-Reported Use of Non-FDA Approved Cannabinoid Products among Hospitalized Patients in a Large US Hospital Network

Background: The growing cannabinoid product (CP) market and its increasing public accessibility have generated unique regulatory challenges.

Objectives: To explore self-reported use of non-FDA approved CP in a large US hospital network.

Methods: We conducted a cross-sectional descriptive study to assess the ability to identify non-FDA approved CP use among hospitalizations at HCA Healthcare, which contributes inpatient electronic health record data for use in the FDA’s Sentinel System. CP use was investigated within ‘Home Medications’ data, which includes medications and other products the patient self-reported taking at home prior to the encounter, and those with which the patient is discharged.
We identified hospitalizations at 165 facilities across 18 states from January 2018 through December 2023. We used ICD-10-CM diagnosis codes (any diagnosis position) to define subcohorts of hospitalizations with epilepsy, pain, or cannabis-related disorders or poisonings (cannabis subcohort). We assessed for evidence of CP in the ‘Home Medications’ data stream using 3- to 5-character search terms.

Results: We identified 12,380,480 hospitalizations among patients with a median age of 57 (IQR: 32-73) years, which mostly occurred among females (56%). Among these, we identified 2,607,210; 489,177; and 386,927 hospitalizations in the pain, epilepsy, and cannabis subcohorts, respectively. CP use was documented in 0.11% (n=13,433) of overall hospitalizations, in 0.16% of the hospitalizations in the epilepsy (n=793) and cannabis (n=633) subcohorts, and in 0.24% (n=6,359) of the hospitalizations in the pain subcohort.

Conclusions: It was feasible to identify non-FDA approved CP documented as self-reported medications during hospitalizations at HCA Healthcare. However, we observed extremely low reporting of CP compared to the prevalence noted in published US surveys. Notably, the reporting found is consistent with results from studies on medication reconciliation in inpatient settings where lack of patient and provider engagement may lead to suboptimal capture. These results suggest the need for further methods development work to better identify CP use in medical records in inpatient settings as well as for engagement of patients in conversation and improved monitoring and documentation of CP use.
Disclaimer
The contents are those of the authors and do not necessarily represent the official views of, nor an endorsement, by FDA/HHS, the US Government, HCA Healthcare or any of its affiliated entities. Some co-authors on this abstract are employed at organizations which conduct work for government and private organizations, including pharmaceutical companies.