Abstracts

Rationale: The decriminalization of marijuana in the United States has contributed to an increase in tetrahydrocannabinol (THC) positivity rates in patients admitted in hospitals (2). Admissions to the Epilepsy monitoring unit (EMU) are voluntary, they undergo long term EEG monitoring, during this time marijuana use is not permitted per hospital policy. Epilepsy monitoring unit diagnostic yield is dependent on length of stay (LOS), with higher diagnostic yield noted with longer LOS (1). Marijuana use may alter patients’ tolerance of hospital admittance secondary to mood and somatic symptoms (3, 4). Unlike nicotine withdrawal which can be ameliorated with use of nicotine patches, there is no option for marijuana replacement under hospital policy. Without access to THC, withdrawal may present and patients may request to discontinue admission thus reducing their length of stay (LOS). The goal of this study is to determine the correlation between THC positivity rates in patients admitted to the Mayo Clinic Hospital (MCH) epilepsy monitoring unit and the LOS of said patients, in order to determine if THC positive patients have a shorter LOS, thus resulting in reduced diagnostic yield.

Methods: A retrospective statistical analysis was conducted comparing the LOS of patients testing positive for THC and patients testing negative for THC admitted to the epilepsy monitoring unit at MCH over the past 5 years (May 2019 to May 2024). A total of 826 patients were analyzed. Data was collected and sorted via the medical software Epic and THC levels were detected via urine screening. Data was split into two separate categories: less than 3.0 days & 3.0 days or more. Finally, the likelihood ratio was found and statistical significance was determined.

Results: 184 (22.28%) of patients tested positive for THC and the remaining 642 (77.72%) tested negative for THC. Of the patients which tested positive, 49 (26.63%) had a LOS less than 3.0 days and 135 (73.37%) had a LOS of 3.0 days or more. Of the patients which tested negative, 155 (24.14%) had a LOS less than 3.0 days and 487 (75.86%) had a LOS of 3.0 days or more. In comparison 2.49% more THC positive patients stayed less than 3.0 days in comparison to THC negative patients. As a result, the positive likelihood ratio produced a value of 1.11 and the negative likelihood ratio produced a value of 0.97.

Conclusions: There is no indication that a positive THC test result correlates with a shorter LOS. Both the positive (1.11) and negative (0.97) likelihood ratios’ proximity to 1 indicate that THC test results do not change the probability that a patient will have a short or long LOS, indicating no correlation.

Funding: none