Analyzing Cannabis Compound Effects

De-identified patient data were entered into an Excel spreadsheet and analyzed descriptively. To account for missing data, patient scores for each SAS symptom were grouped into baseline and three-month blocks, averaged, and analyzed by fitting a mixed-effects model [13 (link)]. Post hoc multiple comparisons analysis using Dunnett’s test identified the statistical significance of timepoints compared to baseline. Analysis was stopped at the timepoint which had around 50% of the initial patient cohort for each condition or symptom. Cohorts with low patient numbers were analyzed with a Wilcoxon signed-rank test. Two-tailed t-test compared the highest doses of cannabidiol (CBD) and delta-9-tetrahydrocannabinol (Δ-9-THC). Statistical significance was set at p < 0.05. Statistical tests were performed using GraphPad Prism (v9.1.2, GraphPad Software Inc., San Diego, CA, USA).