OBJECTIVES: To describe differences in healthcare resource utilization (HCRU) between chronic post-traumatic stress disorder (cPTSD) patients who use and do not use cannabis and explore moderation by tetrahydrocannabinol (THC) containing products.

METHODS: This retrospective analysis identified patients with cPTSD (ICD-10-CM:F42.12) in CHRONOS, a claims-based real-world data source linked to a point-of-sale cannabis dispensary database. Patients with ≥1 cannabis purchase within one year of their first cPTSD diagnosis (index) were defined as cannabis users (CUs). CUs were matched 1:N to non-CUs by state-of-residence. Patients <18, with acute or undefined PTSD, and without 12-months continuous enrollment before index were excluded. HCRU was defined in the year after index by counting ER, office, outpatient, and inpatient visits. Poisson regression was used to calculate event rate ratios (95% confidence intervals) for each visit type, comparing CUs, stratified by number of purchases [quartile 1 (Q1), Q2, Q3, Q4], to non-CUs.

RESULTS: Of the 61,427 patients with cPTSD, 1,071 were CUs. Most patients were female (73%) and had a mean age of 38.7 (SD:12.9). The mean number of purchases by CUs in Q1, Q2, Q3, and Q4 were 1.0 (SD:0.0), 4.4 (SD:2.0), 18.3 (SD:7.4), and 133.0 (SD:87.7). Compared to non-CUs, CUs in Q3 and Q4 were less likely to have an office visit [0.85(0.75-0.97); 0.85(0.74-0.97)], and CUs in Q1 were less likely to have an ER visit [0.70(0.54-0.90)]. Compared to patients without a THC-containing product, patients with a THC-containing product in Q1 were more likely to have an ER, inpatient, and outpatient visit [1.66(0.96-2.86), 1.73(1.08-2.75), 2.78(1.17-6.61)], respectively.

CONCLUSIONS: To our knowledge, this is the first study to explore cannabis and HCRU among cPTSD patients in a real-world setting. The findings suggest there is less HCRU among CUs with cPTSD, with moderation by THC. A limitation of the study was the use of point-of-sale data to define cannabis use.